Court filing
Amended Consolidated Complaint — In re Carvana Securities
Filed March 29, 2024 in In re Carvana Co Securities Litigation; one of 27 filings from this case.
Record facts
| Court | U.S. District Court for the District of Arizona |
|---|---|
| Filed | 2024-03-29 |
U.S. District Court for the District of Arizona · No. 2:22-cv-02126-MTL · Doc. 71 · 2024-03-29 · Docket on CourtListener
Full text
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
ROBBINS GELLER RUDMAN
& DOWD LLP
DANIEL S. DROSMAN (CA 200643)
RACHEL A. COCALIS (CA 312376)
SARAH A. FALLON (CA 345821)
655 West Broadway, Suite 1900
San Diego, CA 92101
Telephone: 619/231-1058
619/231-7423 (fax)
ddrosman@rgrdlaw.com
rcocalis@rgrdlaw.com
sfallon@rgrdlaw.com
Lead Counsel for Lead Plaintiffs
[Additional counsel appear on signature page.]
UNITED STATES DISTRICT COURT
DISTRICT OF ARIZONA
In re Carvana Co. Securities Litigation
This Document Relates To:
All Actions.
)
)
)
)
)
)
)
)
No. CV-22-2126-PHX-MTL
LEAD PLAINTIFFS’ AMENDED
CONSOLIDATED COMPLAINT FOR
VIOLATIONS OF THE FEDERAL
SECURITIES LAWS
DEMAND FOR JURY TRIAL
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 1 of 332
TABLE OF CONTENTS
Page
- i -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
I.
INTRODUCTION .................................................................................................... 1
II.
JURISDICTION AND VENUE ............................................................................... 8
III.
PARTIES .................................................................................................................. 8
A.
Plaintiffs ........................................................................................................ 8
B.
Exchange Act Defendants ............................................................................. 9
IV.
BACKGROUND .................................................................................................... 11
A.
Title and Registration .................................................................................. 11
B.
Carvana’s Origins ........................................................................................ 12
C.
Carvana’s Key Financial Metrics ................................................................ 13
V.
CONFIDENTIAL WITNESS ACCOUNTS .......................................................... 15
VI.
DEFENDANTS’ PUMP-AND-DUMP SCHEME AND FRAUDULENT
COURSE OF BUSINESS UNDER RULE 10b-5(a) and (c) ................................. 37
A.
Artifice No. 1: Garcia Junior and Garcia Senior Make a Sham
Related-Party Deal Between Carvana and DriveTime to Boost
Carvana’s Retail Sales ................................................................................. 39
B.
Artifice No. 2: Defendants Lower Carvana’s Purchasing and
Verification Standards, Purchasing Lower Quality Cars to Induce
Trade-In Sales ............................................................................................. 41
C.
Artifice No. 3: Garcia Junior and Jenkins Embark on a Rapid and
Unsustainable Nationwide Expansion ......................................................... 43
D.
Artifice No. 4: Defendants Flout Title and Registration Laws ................... 46
E.
Artifice No. 5: Garcia Junior and Jenkins Make Materially
Misleading Statements and Omissions ........................................................ 51
1.
Garcia Junior and Jenkins Make False and Misleading
Statements to Further Conceal the Fraudulent Scheme ................... 52
2.
Garcia Junior and Jenkins Make Their Manufactured Growth
Appear Sustainable ........................................................................... 53
F.
Artifice No. 6: Garcia Senior and Jenkins Manipulate Their 10b5-1
Plans to Capitalize on the Scheme .............................................................. 56
G.
Artifice No. 7: Garcia Senior and Jenkins Dump Their Holdings .............. 57
H.
The Truth Is Revealed ................................................................................. 58
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 2 of 332
Page
- ii -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
VII.
DEFENDANTS’ CLASS PERIOD FALSE AND MISLEADING
STATEMENTS AND OMISSIONS UNDER RULE 10b-5(b) ............................ 60
A.
Defendants’ Materially Misleading Statements and Omissions
Concerning Title and Registration .............................................................. 60
B.
Defendants’ Materially False and Misleading Statements and
Omissions Concerning Buying Cars from Customers ................................ 80
C.
Defendants’ Materially Misleading Statements and Omissions
Concerning Retail Unit Sales Growth ....................................................... 139
D.
Defendants’ Materially False and Misleading Statements Regarding
Expansion and Logistics Infrastructure ..................................................... 146
E.
Defendants’ Materially False and Misleading Statements Regarding
Aged Inventory .......................................................................................... 168
F.
Defendants’ Materially False and Misleading Statements Regarding
Profitability Per Vehicle Sold ................................................................... 183
VIII. ADDITIONAL ALLEGATIONS OF SCIENTER .............................................. 255
A.
Insider Stock Sales Support a Motive to Commit Fraud .......................... 255
1.
Garcia Senior’s Class Period Sales and Insider Trading ............... 255
2.
Jenkins’s Class Period Sales and Insider Trading .......................... 258
B.
Defendants’ 10b5-1 Plans Support a Strong Inference of Scienter .......... 260
C.
Multiple Confidential Witnesses Raised Quality, Indiscriminate
Growth, and Title Problems with Managers, Including Garcia Junior
and Jenkins ................................................................................................ 261
D.
The Fraud Involved the Company’s Core Operations, About Which
Defendants Held Themselves Out as Knowledgeable .............................. 263
E.
Analysts and Investors Frequently Asked Questions About, and
Garcia Junior and Jenkins Frequently Evinced Personal Knowledge
by Commenting on, the Subjects of Their Fraud ...................................... 266
F.
Defendants’ Decision to Cease Reporting Critical Metrics Bolsters
Their Scienter ............................................................................................ 268
1.
Average Days to Sale ..................................................................... 269
2.
Buying Cars from Customers ......................................................... 269
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 3 of 332
Page
- iii -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
G.
Garcia Junior’s and Jenkins’s SOX Certifications and Signing of
SEC Filings Support Scienter .................................................................... 270
H.
Garcia Senior’s Control of and Related-Party Deals with Carvana .......... 271
I.
Garcia Senior’s History of Similar and Fraudulent Misconduct
Supports a Strong Inference of Scienter .................................................... 274
J.
Corporate Scienter ..................................................................................... 275
IX.
LOSS CAUSATION ............................................................................................ 276
X.
APPLICABILITY OF THE PRESUMPTION OF RELIANCE AND THE
FRAUD-ON-THE MARKET DOCTRINE ......................................................... 286
XI.
NO SAFE HARBOR ............................................................................................ 286
XII.
CLASS ACTION ALLEGATIONS ..................................................................... 287
XIII. CLAIMS FOR RELIEF UNDER THE EXCHANGE ACT ................................ 288
XIV. SECURITIES ACT ALLEGATIONS ................................................................. 293
A.
Plaintiffs’ Purchases in the 2022 Public Offering ..................................... 293
B.
Securities Act Defendants ......................................................................... 293
C.
Background of the 2022 Public Offering .................................................. 294
D.
The Registration Statement Contained Materially False and
Misleading Statements and Omitted Material Facts Required to be
Stated Therein ........................................................................................... 296
E.
The Registration Statement Contained Deficient and Inaccurate Risk
Disclosures ................................................................................................ 307
F.
Events Subsequent to the 2022 Public Offering ....................................... 308
G.
The Role of the Underwriter Defendants in Connection with the
2022 Public Offering ................................................................................. 310
H.
Class Action Allegations for Securities Act Claims ................................. 313
XV.
CLAIMS FOR RELIEF UNDER THE SECURITIES ACT ............................... 314
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 4 of 332
- 1 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Lead Plaintiffs United Association National Pension Fund (“UANPF”) and
Saskatchewan Healthcare Employees’ Pension Plan (“SHEPP,” and collectively,
“Plaintiffs”) allege the following against Defendants (defined herein), by and through Lead
Counsel, upon personal knowledge as to themselves and their own acts and upon information
and belief as to all other matters.1
I.
INTRODUCTION
1.
This action concerning a classic pump-and-dump scheme and materially
misleading statements and omissions asserts violations of the federal securities laws against
Carvana, its controlling shareholder, and certain officers on behalf of Plaintiffs and all other
persons and entities that purchased or otherwise acquired Carvana Class A common stock
between May 6, 2020 and February 23, 2023, inclusive (the “Class Period”).2
2.
Throughout the Class Period, Defendants billed Carvana to investors as an e-
commerce company, akin to the Amazon of the used car industry. Unlike traditional car
dealerships, Carvana would be a seemingly limitless growth machine because the
Company’s disruptive model was full of competitive advantages, such as a “capital-light”
expansion model, a scalable business model, and a groundbreaking logistics network that
could readily deliver or pick-up cars nationwide. As Defendants promised investors,
Carvana’s “business gets better as it gets bigger.”
1
Plaintiffs’ information and belief is based on, among other things, the independent
investigation of counsel, which includes, but is not limited to: (a) review of U.S. Securities
and Exchange Commission (“SEC”) filings by Carvana Co. (“Carvana” or the “Company”);
(b) review of transcripts of Carvana’s public conference calls, press releases, and other
publications issued by the Company; (c) review of media reports about the Company; (d)
review of public filings and court orders in other litigation against one or more defendant;
and (e) interviews with third parties conducted by attorneys and/or investigators retained by
attorneys. Many of the facts supporting the allegations contained herein are known only to
Defendants or are exclusively within their custody and control. Plaintiffs’ investigation is
ongoing and they believe that substantial additional evidentiary support will exist for the
allegations set forth herein after a reasonable opportunity for discovery.
2
Plaintiffs assert additional violations of the federal securities laws against Carvana,
certain executives, directors, and underwriters in connection with Carvana’s public offering
on or about April 20, 2022 (the “2022 Public Offering”). See §§XIV.-XV. Unless otherwise
stated, “stock” refers to Carvana Class A common stock and emphasis is added.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 5 of 332
- 2 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
3.
But, unbeknownst to investors, Carvana’s sustainable growth machine was a
lemon, built on a fraudulent pump-and-dump scheme to boost Carvana’s retail sales growth,
and a series of misrepresentations and omissions designed to artificially inflate Carvana’s
share prices for long enough to allow the Company’s founders and executives to sell nearly
$3.76 billion of their personally held stock at artificially inflated prices.
4.
Carvana was founded by Ernest Garcia II (“Garcia Senior”) and his son Ernest
Garcia III (“Garcia Junior”) as a wholly-owned subsidiary of Garcia Senior’s other used car
business, DriveTime Automotive Group, Inc. (“DriveTime”). The father-and-son team spun
Carvana off and took it public, celebrating the culmination of their efforts by ringing the bell
at the New York Stock Exchange (the “NYSE”) side by side. But Garcia Senior had a
problem. His federal felony conviction for fraud barred him from employment at any
NYSE-listed company. Thus, he installed his son as the Company’s Chief Executive Officer
(“CEO”) and Chairman, designed Carvana’s voting structure so that he maintained more
than 80% of the voting power and thus, controlled Carvana and its operations, and appointed
his loyal friends (one of whom previously had been censured by the NYSE for actions he
took on behalf of Garcia Senior) as board members.
5.
Leading up to the Class Period, Defendants faced a dilemma as investors were
concerned that Carvana’s retail sales growth was slowing and Carvana’s bigger is better e-
commerce growth story was in doubt. As a Morgan Stanley analyst noted on February 27,
2020, if Carvana’s “growth starts to slow significantly, and the company is still not
profitable, there will be a transition away from growth investors, and . . . how investors think
about the valuation of the stock will change.”
6.
Defendants could not let this happen; Garcia Junior, Garcia Senior, and Mark
Jenkins (“Jenkins”), Carvana’s Chief Financial Officer (“CFO”), needed the Company to
have a high valuation so they could cash out their holdings in Carvana stock at a top dollar.
Thus, to pump Carvana’s stock price, Defendants planned and executed a scheme to inflate
the Company’s retail vehicle sales growth and convince investors that this growth was
sustainable.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 6 of 332
- 3 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
7.
Accordingly, Defendants employed the following artifices to artificially boost
Carvana’s reported retail sales growth, and make the growth appear profitable and
sustainable, in furtherance of their pump-and-dump scheme: (i) Garcia Senior caused
DriveTime to enter into a sham pass-through sales arrangement with Garcia Junior at
Carvana to boost materially Carvana’s reported retail sales; (ii) Carvana drastically lowered
or disregarded entirely its purchasing and verification standards, and began buying low-
quality cars to induce trade-in sales which artificially boosted Carvana’s reported sales
growth; (iii) Carvana flouted state title and registration laws to secure retail sales before
competitors; (iv) Garcia Junior and Jenkins spearheaded a rapid and unsustainable
nationwide expansion without regard to profitability to boost Carvana’s reported retail sales;
and (v) Garcia Junior and Jenkins made materially misleading statements and omissions
designed to conceal the scheme and convince investors that Carvana’s retail growth was
profitable and sustainable.
8.
Defendants’ limitless growth story produced its desired results. Carvana’s
retail sales growth skyrocketed, propelling Carvana to be named as one of the fastest
companies to ever make the Fortune 500 list based on organic growth alone, along with
Google and Amazon. Naturally, Carvana’s stock price soared, rising as high as $376 per
share during the Class Period. And Defendants took full advantage of the artificially inflated
stock price by employing the final artifice of their pump-and-dump scheme: manipulating
their trading plans and dumping nearly $3.76 billion of their shares on unsuspecting
investors.
9.
Garcia Senior alone reaped more than $3.6 billion in insider sales proceeds.
Daniel Taylor, an accounting professor at the Wharton School of Business and director of the
Wharton Forensic Analytics Lab, explained that Garcia Senior’s insider trades were
nefarious, concluding: “‘What I’m saying is the Garcias knew it was short-lived . . . . The
Garcias knew the music would eventually end.’”3
3
See John Hyatt, Carvana’s ‘Chaotic’ Zoom Firing Caps Company’s Struggles Amid
Market Downturn, Forbes (May 21, 2022).
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 7 of 332
- 4 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
10.
Critically, Garcia Junior and Jenkins furthered their scheme by making
materially misleading statements and omissions designed to convince investors that
Carvana’s growth was profitable and sustainable. After all, if Carvana could pair its
explosive unit growth with purported profitability and a scalable logistics model that could
expand across the country efficiently and cost effectively, the sky was the limit. Thus, in
addition to engaging in a scheme to defraud in violation of Rule 10b-5(a) and (c), Carvana,
Garcia Junior, and Jenkins made misleading statements and omissions regarding the
following subjects in violation of Rule 10b-5(b): (i) title and registration; (ii) buying cars
from customers; (iii) retail unit growth; (iv) expansion and logistics infrastructure; (v) aged
inventory; and (vi) profitability per vehicle sold.
11.
Title and Registration. In each of Carvana’s SEC filings during the Class
Period, Carvana purported to warn of material risks and related harms that could occur if it
failed to comply with state and local laws and regulations, including those related to title and
registration. In fact, beginning at least by the date on which it filed its 2020 10-K (defined
herein), these risks and harms had already materialized because states had begun
investigating and penalizing Carvana for its title and registration noncompliance. When the
news of these investigations and penalties began to leak publicly, Defendants minimized the
regulatory investigations as mere one-offs and trivialized the penalties imposed on the
Company. But, as Defendants later conceded, the reputational damage from just one of these
suspensions was “‘incalculable and irreparable.’”
12.
Buying Cars from Customers. As Defendants reminded investors throughout
the Class Period, a critical part of Carvana’s growth strategy was to “[i]ncrease the purchase
of vehicles from customers.” However, unbeknownst to investors, by the second half of FY
2020, Defendants had drastically lowered or disregarded entirely their purchasing and
verification standards to significantly increase the number of cars Carvana purchased from
customers in order to induce more trade-ins and artificially boost retail sales. While this
action did increase retail unit sales, it adversely impacted Carvana’s bottom line, flooded
Carvana with low-quality cars that had to be sold via the wholesale market at a loss, and
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 8 of 332
- 5 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
crippled its logistics network. Rather than disclose this adverse information, Defendants
touted only the positive “advantages” of buying cars from customers while assuring investors
that Carvana “assess[es] vehicles on the basis of quality.”
13.
Retail Unit Sales Growth. Carvana told investors that “Objective #1” leading
up to and throughout the Class Period was to “Grow Retail Units.” During the Class Period,
Defendants consistently touted Carvana’s retail unit sales growth and legitimate drivers of
such growth. In truth, however, Defendants concealed that Carvana’s sales growth was also
substantially fueled by nefarious tactics, such as: (i) sales in violation of title and registration
laws and regulations; (ii) “less profitable sales . . . in markets with lower profitability due to
long distance from inventory”; (iii) trade-in sales resulting from Carvana’s lowered
purchasing and verification standards; and (iv) “sales that were less profitable in the
immediate period” because of artificially low pricing. By failing to disclose these adverse
facts, Defendants created the materially misleading impression that Carvana’s growth was
organic, profitable, and sustainable.
14.
Expansion and Logistics Infrastructure. Leading up to and throughout the
Class Period, Defendants repeated their mantra: “[W]e want to expand as much as we
possibly can.” Indeed, in every quarterly shareholder letter during the Class Period, Garcia
Junior and Jenkins dedicated an entire section to its expansion, in which they reported the
number of new markets added to tout the purported population coverage that Carvana had
achieved. At the same time, Defendants exalted their “capital-light” expansion model (i.e., a
business model that did not require a substantial number of physical locations). Defendants’
statements were misleading, however, as the vast majority of their new markets were
significant distances from inspection and reconditioning centers (“IRCs”), which,
Defendants later admitted, required Carvana to incur significant additional logistics costs that
rendered sales in distant markets unprofitable. Unsurprisingly, near the end of the Class
Period, Defendants were forced to come clean and take actions to intentionally reduce sales
in distant markets that generated “lower profitability due to long distance from inventory.”
This caused the Company to report its first-ever quarterly sales decline in Q3 2022.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 9 of 332
- 6 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Defendants were also forced to fix Carvana’s deficient logistics infrastructure by spending
over $4.5 billion to acquire ADESA, a wholesale auction company with locations
nationwide. Contrary to Defendants’ Class-Period assertions about Carvana’s capital-light
expansion model, Garcia Junior admitted after the Class Period that “we are without
question . . . a big physical infrastructure business” and that “[a] major factor” behind
Carvana’s ADESA purchase was that a “used car retailer needs a physical footprint . . .
undergirded by a network of massive distribution centers and transport depots.”
15.
Aged Inventory. Average days to sale, which measures the number of days
between when Carvana acquires a car and when it sells the car to a customer, was one of
Carvana’s key internal metrics throughout the Class Period. Defendants acknowledged that
“our business is dependent upon our ability to expeditiously sell inventory” and that an
increase in the average days to sale metric “is unfavorable for retail GPU because if you
hold the car longer, it has more time to experience depreciation before you sell it.” Despite
the fact Defendants clearly viewed average days to sale as a key internal metric, Defendants
abruptly stopped disclosing it to investors, claiming that it was no longer necessary given
“the relative stability of average days to sale over the past three years.” However, by early
2022, average days to sale was anything but “relative[ly] stab[le],” spiking nearly 60% in
just three quarters. Thus, Defendants’ statements were materially misleading because they
created an impression of a state of affairs (relatively stable average days to sale) that differed
in a material way from the one that actually existed (a spiking average days to sale metric).
16.
Profitability per Vehicle Sold. During the Class Period, Carvana
acknowledged that “[t]he Company’s primary business objective is to sell used vehicles to
retail customers and generate a profit doing so.” Thus, “Retail GPU,” which measured the
gross profit Carvana earned per retail vehicle sold (before adding in other ancillary revenue
and profit streams such as warranty, insurance, and financing), was a critical metric to
investors during the Class Period. Throughout the Class Period, Defendants reported
positive Retail GPU in each of Carvana’s earnings calls and shareholder letters and assured
investors that its retail vehicle sales were profitable. However, Defendants concealed
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 10 of 332
- 7 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Carvana’s actual per-vehicle profitability – or lack thereof – by: (i) excluding certain per-
vehicle operations expenses from its calculation of Retail GPU; and (ii) not disclosing or
quantifying these excluded costs separately so investors could decipher Carvana’s “unit
economics” (i.e., overall profitability) of retail sales on their own. These per-vehicle
operations expenses, which were only separately broken out and quantified for the first time
after the Class Period, included material costs associated with completing retail sales, such
as the full cost of shipping cars to a customers and title and registration costs. By
intentionally hiding these significant costs, investors were left to rely on the positive Retail
GPU reported by Defendants as the sole measure of Carvana’s retail sales profitability.
However, had Carvana included these operations expenses in its Retail GPU calculation or
separately disclosed and quantified the costs that were excluded from Retail GPU, it would
have revealed to investors that, on average, Carvana generated negative unit economics on
every retail car sold. This created an impression of a state affairs (positive per-unit
profitability) materially different from the one that actually existed (negative per-unit
profitability).
17.
Through a series of partial disclosures, investors began to learn that Carvana
was a very different company from the one they were sold. Defendants’ bigger is better
Amazon-comparison story was a fiction. Ultimately, Defendants admitted that Carvana’s
retail sales growth was unprofitable and unsustainable – indeed, by the end of the Class
Period, Defendants disclosed that “retail units sold decreased for the first time in our history
this year” and “that going forward, Carvana was intentionally foregoing retail sales growth
for the foreseeable future to instead focus on driving “positive unit economics” (i.e., retail
sales that were actually profitable). Moreover, Carvana was forced to write off over $50
million of aged and impaired vehicle inventory that was worth less than Carvana had paid to
acquire it and prepare it for sale. Defendants further acknowledged that Carvana’s “capital-
light” market expansion model was a farce, and Carvana was forced to issue billions of
dollars of high-interest junk bonds to acquire a nationwide auction house, ADESA, to
backfill its deficient infrastructure.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 11 of 332
- 8 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
18.
By the time the truth about Carvana’s unsustainable growth and business
model was revealed, the price of Carvana’s stock had plummeted from a Class Period high of
$376 per share to $8.01 per share, representing a decline of more than 98%, leaving the
Company on the verge of bankruptcy. While investors suffered billions of dollars in
damages, Defendants’ fraudulent conduct permitted them to walk away with billions.
Plaintiffs seek to recover those damages on behalf of the putative Class through this action.
II.
JURISDICTION AND VENUE
19.
Jurisdiction is conferred by 28 U.S.C. §1331, §22 of the Securities Act of 1933
(the “Securities Act”) (15 U.S.C. §77v), and §27 of the Securities Exchange Act of 1934 (the
“Exchange Act”) (15 U.S.C. §78aa). The claims asserted herein arise under §§11, 12(a)(2),
and 15 of the Securities Act (15 U.S.C. §§77k, 77l(a)(2), and 77o) and §§10(b), 20(a), and
20(A) of the Exchange Act (15 U.S.C. §§78j(b), 78t(a), and 78t-1) and Rule 10b-5
promulgated thereunder (17 C.F.R. §240.10b-5).
20.
Venue is proper in this District pursuant to 28 U.S.C. §1391(b), §27 of the
Exchange Act, and §22 of the Securities Act. Substantial acts in furtherance of the alleged
fraud or the effects of the fraud have occurred in this District. Many of the acts charged
herein, including the dissemination of materially false and/or misleading information,
occurred in substantial part in this District. In addition, the Company’s principal executive
offices are located in this District.
21.
In connection with the acts, transactions, and conduct alleged herein,
Defendants directly and indirectly used the means and instrumentalities of interstate
commerce, including the U.S. mail, interstate telephone communications, and the facilities of
a national securities exchange.
III.
PARTIES
A.
Plaintiffs
22.
Lead Plaintiff UANPF is an Alexandria, Virginia-based multi-employer
defined benefit pension plan. UANPF is one of the nation’s largest Taft-Hartley funds with
approximately $6.5 billion in assets held for the benefit of approximately 150,000
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 12 of 332
- 9 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
participants. ECF 9-5, ¶2. As detailed in UANPF’s previously-filed certification (see ECF
9-3, ECF 9-4, and ECF 9-6) and certification attached hereto, UANPF purchased a
significant number of shares of Carvana stock at artificially inflated prices during the Class
Period and suffered damages as a result of Defendants’ alleged misconduct. As discussed
infra §XIV.A., on April 22, 2022, UANPF also purchased 1,455 shares of stock in the 2022
Public Offering from Citigroup Global Markets Inc. (“Citigroup”) for $80.00 per share.
23.
Lead Plaintiff SHEPP, the largest defined benefit plan in the Saskatchewan
providence of Canada, is a multi-employer defined benefit pension plan serving the
healthcare sector with over 60,000 members and more than $10 billion in assets under
management. ECF 9-5, ¶3. As detailed in SHEPP’s previously-filed certification (see ECF
9-3, ECF 9-4, and ECF 9-6) and certification attached hereto, SHEPP purchased a significant
number of shares of Carvana stock at artificially inflated prices during the Class Period and
suffered damages as a result of Defendants’ alleged misconduct. As discussed infra
§XIV.A., on April 22, 2022, SHEPP also purchased 3,838 shares of stock in the 2022 Public
Offering from Citigroup for $80.00 per share.
B.
Exchange Act Defendants
24.
Defendant Carvana is a Delaware corporation with its principal executive
offices located in Tempe, Arizona. Carvana’s stock trades on the NYSE under the symbol
“CVNA.”
25.
Defendant Garcia Junior is a co-founder of the Company and has served as its
CEO, President, and Chairman since 2012.
26.
Defendant Garcia Senior is a founder of Carvana and was, at all relevant times,
and continues to be, the Company’s controlling shareholder. He maintained 84% of
Carvana’s voting power during the Class Period. Garcia Senior was a real estate developer
until federal investigators began investigating his real estate and stock deals in connection
with the Lincoln Savings & Loan scandal. In 1990, Garcia Senior pled guilty to
“fraudulently obtain[ing] a $30-million line of credit in a series of transactions that also
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 13 of 332
- 10 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
helped Lincoln [Savings & Loan] hide its ownership . . . from regulators.”4 As a result of his
conviction, Garcia Senior was banned for life from serving as an employer, officer, or
director with a company on the NYSE and, therefore, cannot hold a formal position at
Carvana. However, Carvana’s SEC filings leading up to and during the Class Period make
clear: “The Garcia Parties control us”5 and can, among other things, “elect all of the
members of our Board [of Directors] and thereby effectively control our policies and
operations.” Garcia Senior is Garcia Junior’s father and maintains a close relationship with
his son, whom he lived next door to during the Class Period. Garcia Senior is also the largest
owner of Carvana’s former parent company, DriveTime (of which Garcia Junior and his
children own nearly a quarter through a trust). During the Class Period, Defendant Garcia
Senior sold 13,950,000 shares, at artificially inflated prices for proceeds of
$3,676,933,685.13.
27.
Defendant Jenkins was, at all relevant times, and continues to be, Carvana’s
CFO. During the Class Period, Defendant Jenkins sold 336,929 shares, at artificially inflated
prices for proceeds of $79,246,194.95.
28.
The defendants referenced in ¶¶24-27, above, are collectively referred to as the
“Exchange Act Defendants” or “Defendants.”
29.
Defendants Garcia Senior, Garcia Junior, and Jenkins (collectively, the
“Individual Defendants”), because of their voting power and/or positions within the
Company, possessed the power and authority to control the contents of the Company’s
reports to the SEC, press releases, and presentations to securities analysts, money and
portfolio managers, and institutional investors, i.e., the market.
30.
Defendants were provided with copies of the Company’s reports and press
releases alleged herein to be misleading prior to, or shortly after, their issuance and had the
4
James S. Granelli, Lincoln S&L; Figure Pleads Guilty to Fraud: Crime: Ernest C. Garcia
II admits acting to help the thrift hide its ownership of some risky desert land in Arizona,
L.A. Times (Oct. 31, 1990).
5
“Garcia Parties” refers to Garcia Senior, Garcia Junior, and entities controlled by one or
both of them.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 14 of 332
- 11 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
ability and opportunity to prevent their issuance or cause them to be corrected. Because of
their positions and access to material, non-public information (“MNPI”) available to them,
Defendants knew or recklessly disregarded that the adverse facts specified herein had not
been disclosed to, and were being concealed from, the public, and that the representations
that were being made were then materially false and/or misleading. Defendants are liable for
the scheme and Garcia Junior and Jenkins are liable for the false and misleading statements
and omissions pleaded herein.
31.
Each Defendant is liable as a participant in a fraudulent scheme and course of
business that operated as a fraud or deceit on purchasers of Carvana stock by engaging in the
violative conduct alleged herein, which caused Carvana stock to trade at artificially inflated
prices during the Class Period.
IV.
BACKGROUND
A.
Title and Registration
32.
A vehicle’s title is a legal document issued by the state that demonstrates the
vehicle’s ownership. In many states, such as Texas, Pennsylvania, Michigan, and Illinois, it
is illegal to sell a car without holding title. Further, states generally set out the legal
requirements for transferring title when a new owner purchases a vehicle, often including a
time limit within which title must be transferred to the new owner.
33.
A vehicle registration reflects a state’s certification that a car can be driven on
public roads. Each state requires vehicles to be registered with a designated state
government agency. It is illegal in every state to drive an unregistered car or one with a
lapsed registration. License plates and tags are a part of the registration process as they
indicate that a car is properly registered. During the brief window after one has purchased a
car and is waiting for permanent plates, states require drivers to display a temporary plate
provided by the dealer. Many states, like Illinois and North Carolina, require the temporary
license and/or tag to be issued from the state in which the car is to be registered.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 15 of 332
- 12 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
B.
Carvana’s Origins
34.
Following his federal felony conviction for fraud, Garcia Senior bought a
rental car franchise called Ugly Duckling and merged it with his finance company to become
the largest subprime lender and seller of used cars. After a short and litigation-punctuated
stint on the Nasdaq Capital Market (the “NASDAQ”), Garcia Senior bought Ugly Duckling
back to take it private for a pittance, settled a number of shareholder actions, and renamed
the company DriveTime.
35.
In 2012, Garcia Senior and his son founded Carvana Group LLC as a wholly-
owned subsidiary of DriveTime. Garcia Senior installed his son as CEO of the Company.
According to Garcia Junior, Carvana’s reliance on DriveTime cannot be understated as
“‘[w]e were able to build Carvana into a stronger company much quicker than we likely
would have been able to do without the benefits . . . from being able to use the backbone of
[DriveTime].’”6
36.
But shortly thereafter, DriveTime found itself in hot water. In 2014, pursuant
to the Dodd-Frank Wall Street Reform and Consumer Protection Act, the Consumer
Financial Protection Bureau brought an enforcement action against DriveTime and its
finance company for violating federal financial laws. DriveTime was fined $8 million,
subjected to increased oversight, and forced to reform its business practices. Thus, the
father-and-son team decided to spin Carvana Group LLC off of DriveTime at that time.
37.
In 2017, Garcia Junior and Garcia Senior took the Company public on the
NYSE. The father-and-son duo, who live next door to one another, celebrated their
accomplishment by ringing the NYSE bell side by side. Because Garcia Senior was unable
to have a formal role at Carvana, however, upon taking Carvana public, he maintained 97%
of the voting power with his son, established a voting “‘structure [that academics note] has
allowed [the Garcias] to run this $60 billion public company as if it’s a family firm and for
6
Bill Alpert, Used-Car Prices Are Plunging. So Why is Carvana’s Stock Soaring? Time
for a Rethink, Barron’s (May 1, 2020).
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 16 of 332
- 13 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
the family’s benefit,’”7 and installed his former DriveTime employees and Lincoln Savings
& Loan cronies to Carvana’s Board of Directors (the “Board”).8
38.
As Garcia Senior had done with Carvana’s predecessor, Garcia Senior and
Garcia Junior pitched Carvana as a disrupter and innovator in the used car market.
C.
Carvana’s Key Financial Metrics
39.
In Carvana’s Q1 2017 letter to shareholders dated June 6, 2017 (the “Q1 2017
Shareholder Letter”), Defendants told investors that Carvana should be measured by its
primary objectives and corresponding key metrics, including “Objective #1: Grow Retail
Units and Revenue.”
40.
Retail Unit Growth. In Carvana’s SEC Form S-1/A, filed April 27, 2017
(“2017 Registration Statement”) and first-ever shareholder letter, Defendants were
unequivocal that the Company’s number one objective – and metric to watch – was growth
in retail units sold. Indeed, in the Q1 2017 Shareholder Letter – Defendants advised that
“[w]e view retail units sold as the single most important metric in our business.” In the
quarters that followed, Defendants would repeatedly advise investors that its number one
objective was to grow retail units sold because it is “the most important measure of our
growth.”
41.
Critically, Defendants repeatedly told investors in the Company’s SEC filings
that Carvana’s retail sales were essential to the Company’s profitability because they “drive[]
the majority of our revenue and allow[] us to capture additional revenue streams associated
with financing, vehicle service contracts (‘VSCs’) and trade-in vehicles.” Moreover, these
7
Carvana’s dual-class ownership structure provides that any share owned by one of the
Garcia Parties is entitled to ten votes per share, compared to the single vote per share on each
share held by the public.
8
Garcia Senior was thick as thieves with Ira Platt (“Platt”) and Greg Sullivan (“Sullivan”),
and they were, in fact, independent neither from him nor his son. In re Carvana Co.
S’holders Litig., 2022 WL 2352457, at *4 (Del. Ch. June 30, 2022). For example, director
Sullivan “was censured by the NYSE due to actions he took on behalf of Garcia Senior” in
connection with Garcia Senior’s fraud. Id. at *9. But, as the court noted, while “[t]he NYSE
censure could have been career-ending for Sullivan, . . . it was not” as Garcia Senior hired
him at DriveTime and eventually installed him as DriveTime’s CEO. Id. at *9-*13. This led
to the court’s “inference that Sullivan and Platt are beholden to the Garcias.” Id. at *18.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 17 of 332
- 14 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
sales “allow [Carvana] to benefit from economies of scale due to our centralized online sales
model.” Accordingly, throughout the Class Period, Defendants highlighted Carvana’s
exponential growth in retail units sold in all of Carvana’s investor presentations, earnings
calls, SEC filings, and shareholder letters.
42.
Retail GPU. During the Class Period, Carvana acknowledged, in an
August 23, 2022 letter to the SEC Division of Corporation Finance, that its “primary
business objective is to sell used vehicles to retail customers and generate a profit doing so.”
Thus, Retail GPU, which measured “the difference between the retail selling price of the
vehicle and [the] cost of sales associated with acquiring the vehicle and preparing it for sale”
was a critical metric to investors during the Class Period.9 Accordingly, Used Vehicle Gross
Profit (i.e., Retail GPU) was included in Carvana’s S-1 Registration Statement at the time of
its IPO and in each of the Company’s Form 10-K’s filed with the SEC during the Class
Period. In addition, Defendants reported Carvana’s Retail GPU in each quarterly
shareholder letter and on each of Carvana’s quarterly earnings calls.
43.
Average Days to Sale. As Defendants explained in their 2017 Registration
Statement, average days to sale was one of Carvana’s six “Key Operating Metrics” because
Carvana’s “business is dependent upon our ability to expeditiously sell inventory.” Indeed,
Carvana’s failure to do so, reflected in an increased average days to sale, “could have a
material adverse effect on our business, sales and results of operations” because “[a]n over-
supply of used vehicle inventory will generally cause downward pressure on our product
sales prices and margins.” Accordingly, during much of the Class Period, Defendants
reported average days to sale in its SEC filings as a key metric to evaluating Carvana’s gross
profit per unit sold. Further, throughout the entire Class Period, Defendants continued to
9
Defendants separately disclosed a “Total GPU” metric, which included Retail GPU,
wholesale gross profit, and “Other Gross Profit” that Carvana earned on ancillary revenue
and profit streams, such as commissions from sales of extended warranties and insurance,
and gains on sales of auto loans.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 18 of 332
- 15 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
explain that reducing average days to sale was essential to “increase our total gross profit per
unit.”
44.
Market Expansion. Carvana also reported the number of markets in which it
sold vehicles as a key metric. Carvana’s 2017 Registration Statement defined a market “as a
metropolitan area in which we have commenced local advertising and offer free home
delivery to customers with a Carvana employee and branded delivery truck.” Defendants
explained that market expansion drove retail sales growth because, as the number of markets
grow, “the population of consumers who have access to our fully integrated customer
experience increases.” Thus, Carvana also reported the percentage of the U.S. population it
served through these markets along with the number of IRCs as key metrics.
V.
CONFIDENTIAL WITNESS ACCOUNTS
45.
Several former Company employees have provided information demonstrating
that Defendants’ Class Period statements were false and misleading, that Defendants knew or
recklessly disregarded the falsity or misleading nature of their statements, and that
Defendants engaged in a scheme to defraud investors. The confidential witnesses (“CWs”)
include individuals formerly employed at the Company during the Class Period whose
accounts corroborate one another, other sources set forth herein, and facts now admitted by
Carvana. The CWs provided information to Plaintiffs’ counsel and Plaintiffs’ counsel’s
investigators on a confidential basis and are particularly described by job description and/or
responsibility and duration of employment, thereby providing sufficient detail to establish
their reliability and personal knowledge. As set forth below, the information provided by the
CWs supports a strong inference that the Exchange Act Defendants acted with scienter.
46.
Confidential Witness No. 1 (“CW-1”) was employed by Carvana in a
supervisory role on the Wholesale Team at a large Carvana facility in the Northeast U.S. for
more than six months during the Class Period. CW-1’s role involved managing the
“lifecycle” of vehicles that were going to be sold wholesale. This included overseeing a
team of inspectors who inspected vehicles before they were put up for auction and ensuring
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 19 of 332
- 16 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
that vehicles were picked up by wholesale buyers’ haulers after they were purchased at
auction.
47.
CW-1 explained that the wholesale facility where he/she worked was
essentially an overflow for the IRCs because the IRCs did not have the capacity to store and
process all the vehicles that were going to be sold wholesale. CW-1’s facility was basically
a giant parking lot and thousands of vehicles could be stored there.
48.
CW-1 explained that vehicles that had been purchased with the intent to be
sold wholesale were shipped directly to CW-1’s facility, but CW-1’s facility also received
many vehicles from IRCs after they were “flipped” to wholesale. CW-1 explained that
inspectors at the IRCs inspected vehicles that were supposed to be sold retail. Sometimes the
inspectors at the IRCs decided to flip vehicles to wholesale, typically because it was going to
cost too much money or require too much labor to repair the vehicles well enough to sell
them as retail. When this occurred, CW-1 would do an evaluation and approve the transfer
of the vehicles to wholesale, and they would be shipped to CW-1’s facility. CW-1 said that a
lot of the vehicles that CW-1’s facility received were trade-ins. CW-1 said that so many
vehicles had to be sold wholesale because Carvana had been trying to purchase “everything,”
regardless of the quality of their vehicles.
49.
CW-1 said that when an individual seller submitted information about a
vehicle, Carvana did very little or nothing to verify the accuracy of the information. This
meant that Carvana would buy vehicles even if the condition had been misrepresented by the
seller.
50.
CW-1 also explained that sometimes Carvana used third-party contractors to
pick up vehicles from sellers when the market operations team was too busy. CW-1 believed
that the third-party contractors just picked up the vehicles without looking at the quality at
all. CW-1 provided the example of a vehicle that Carvana purchased that had bullet holes in
it, but the seller had rated it in fair condition. CW-1 said that this vehicle should never have
been purchased. CW-1 also recalled an instance when a vehicle that was supposedly in good
condition had missing seats. CW-1 knew that sellers were rating poor quality vehicles in fair
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 20 of 332
- 17 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
or good condition because CW-1 saw what the sellers had reported about vehicles when
he/she approved them to be flipped from retail to wholesale based on data CW-1 pulled from
Carvana’s Saleforce.com application. CW-1 wondered how these poor-quality vehicles had
been purchased at retail price and why Carvana did not seem to do anything to prevent
sellers from misrepresenting the quality of their cars.
51.
CW-1 participated in Carvana all-hands meetings that were held by Zoom in
which Garcia Junior addressed employees. In these calls, Garcia Junior would discuss the
Company’s performance. The tenor of the calls was always “upbeat.” CW-1 is pretty sure
that his/her bosses, and “probably Garcia” had “talked about high growth” including how the
Company had grown and how the Company was expected to do.
52.
CW-1 said that it was pretty much the case that Carvana lowered its standards
when it came to trade-ins as long as the customer was also buying a car from Carvana. CW-
1 explained that Carvana “wouldn’t care” about the poor quality. CW-1 said that as long as
someone was also “buying from us,” then Carvana would take the trade-ins. But this
practice resulted in, for instance, a car that had been missing its backseat being accepted,
which CW-1 is pretty sure had been a trade-in. CW-1 was also pretty sure the car with bullet
holes had been a trade-in because, with trade-ins, Carvana was not doing any verifications of
what the customer was representing (whereas with wholesale, usually what the seller was
representing about the vehicle’s condition was in fact what was being sold).
53.
CW-1 confirmed that the parties picking up a customer’s trade-in were
supposed to assess the vehicle and not accept the vehicle if it was inoperable or did not meet
the description provided by the customer. CW-1 said that at the time he/she began working
at Carvana, his/her “partner” at the wholesale facility had previously been a Customer
Advocate and then transferred to wholesale. CW-1 believed that a lot of the cars that were
flipped to wholesale had been trade-ins because he/she heard this from CW-1’s “counterpart”
who had been with Carvana “for years.” According to CW-1, vehicles that ended up being
flipped to wholesale included vehicles that had also been procured with the intention of
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 21 of 332
- 18 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
being sold retail, but the quality of those vehicles ended up being too poor. CW-1 said this
reflected that Carvana “wanted to purchase as much as was humanly possible.”
54.
CW-1 also said that Carvana did not make any efforts to stop purchasing
vehicles from sellers who had misrepresented the condition of vehicles to Carvana in
previous sales. CW-1 recalled an instance where he/she recognized a particular seller’s
name when he/she was reviewing vehicles that were supposed to have been sold retail but
had to be flipped to wholesale. CW-1 determined that this seller had sold five vehicles to
Carvana at retail, or “full value” price, and all five of them had to be sold wholesale because
the seller had misrepresented the quality of the vehicles.
55.
CW-1 stated that the issues he/she observed at Carvana were “obvious to
anyone with eyes.”
56.
Confidential Witness No. 2 (“CW-2”) was employed by Carvana as a Sell to
Carvana (“STC”) Advisor for more than six months during the Class Period at the
Company’s facility in a Southeastern state. CW-2’s duties entailed physically acquiring cars
from individuals who had sold their vehicles to Carvana. To that end, CW-2 travelled to
wherever the selling individual was located throughout a large portion of that Southeastern
state. CW-2 appraised the vehicle to validate that it was in the condition represented by the
customer at the time of the online sale to Carvana, obtain the title and registration
documentation, and arrange for a towing company to transport the vehicle to the Carvana
facility in which he/she worked, and sometimes to other Carvana facilities.
57.
CW-2’s appraisal included a “functionality test” of the vehicle’s interior and
exterior. CW-2 said that in the event his/her appraisal identified issues and concerns that the
seller had not identified at the time of the online sale, then he/she entered those matters into a
laptop computer he/she carried with him/her. There were times CW-2 was overruled by
his/her superiors and was directed to procure a car regardless of his/her concerns. And
CW-2 said it was definitely the case that Carvana had been overpaying for vehicles.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 22 of 332
- 19 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
58.
CW-2 knew how much Carvana had agreed to pay for the vehicles because the
amount was included in the appraisal. CW-2 reiterated that he/she had been surprised at how
much Carvana spent for some vehicles.
59.
Transporting the vehicles from the place of purchase to the Carvana
distribution center (“DC”) initially entailed coordinating with towing companies. CW-2 said
there were rumors that Carvana was incurring thousands of dollars of costs by engaging with
a “wrecker” company. CW-2 went on to say that he/she and the other STC Advisors used to
drive to the seller’s location using a Carvana-issued car. However, in the last three months
of CW-2’s employment, a change was made whereby the STC Advisors instead took Uber
rides to the seller’s location and then drove the purchased vehicle back to the DC. CW-2
was not completely comfortable with this change, in part because the vehicles that had been
purchased were not always safe to drive. To that point, two vehicles he/she had to drive died
while CW-2 was driving them and he/she had to be picked up by an Uber. While CW-2 had
not thought it was efficient when he/she was driving a company vehicle to, say, drive two
hours one way, handle a vehicle purchase and pick-up, only to then drive three hours to the
next vehicle, he/she also did not think switching to Uber was much better. Overall, CW-2
believed that Carvana had been “overscheduling” the STC Advisors because Carvana was
focused more on the quantity of used vehicles it was procuring, as opposed to the quality of
those vehicles and the efficiency of its operations.
60.
CW-2 was aware of “numerous instances” of Carvana buying vehicles very
inexpensively (he/she used the amount of $400) that were “not safe to drive” and on which
the “check engine light” came on right after they were purchased. If it had been up to
him/her, CW-2 would never have purchased these vehicles, but CW-2 was often overruled.
As CW-2 put it: “I was more cautious about quality.” At least a couple of times CW-2
reported his/her concerns and that he/she did not think a given vehicle should be purchased
but “leadership would say take it” anyway.
61.
Confidential Witness No. 3 (“CW-3”) was employed by Carvana in Phoenix,
Arizona before the Class Period began until the middle of the Class Period. During the Class
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 23 of 332
- 20 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Period, CW-3 was employed as a Dealer Success Advocate on Carvana’s Wholesale Team,
where CW-3 assisted dealers that had purchased vehicles from Carvana through auction and
had issues that made the vehicles unsellable to retail customers.
62.
CW-3 said that the vehicles Carvana sold at auction had been purchased from
auctions and from individuals. CW-3 believes that most of them were purchased from
individuals. CW-3 explained that these were vehicles that did not meet certain standards set
by Carvana to sell retail. CW-3 used Salesforce to document his/her communications with
dealers. CW-3 believes that more vehicles (both those that had been procured wholesale as
well as from STC) were being sold wholesale because Carvana had lowered its standards for
buying vehicles in 2019.
63.
CW-3 explained that whenever vehicles were sold through auction, there was
an arbitration process where Carvana had to buy back vehicles when buyers wanted to return
them because there were issues that had not been noted by Carvana at the time of the sale.
CW-3 said that buybacks were common and missing titles were the main reason vehicles
were bought back. However, some of the buybacks also occurred because of quality issues.
CW-3 commented that it was well known internally that the vehicles Carvana was buying
from both wholesale and STC were “not the best.”
64.
Regarding the missing titles, CW-3 said that Carvana had around 60 days to
provide buyers with the titles for the vehicles they had purchased from Carvana and the
buyers could request buybacks because of that issue during this period of time. CW-3
estimated that Carvana generally had around half of the titles for the vehicles they sold
wholesale at the time of sale. One of the reasons why CW-3 quit was because so many
vehicles were missing titles and he/she frequently was not able to get adequate answers from
the Titles Team as to why. CW-3 would reach out to the Titles Team and they would tell
him/her to check back in a week, so CW-3 would, but sometimes a month would pass and it
turned out that nothing had been done. CW-3 said that one of the supposed reasons that it
was difficult for the Titles Team to get the titles was because Departments of Motor Vehicles
(“DMVs”) were falling behind in their workload during the pandemic. However, sometimes
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 24 of 332
- 21 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
CW-3 called the DMVs to try to get the titles himself/herself and sometimes CW-3 was
provided with different information than what he/she had been told by the Titles Team. CW-
3 recalled that there was one dealer who had purchased over 100 vehicles from Carvana that
were missing titles and around half of them were past the arbitration deadline. Prior to CW-
3’s departure, CW-3 said there were other dealers who were so angry they told him/her they
would never work with Carvana again.
65.
In early 2020, CW-3 only received around 10 to 15 calls a week from dealers
regarding title issues. CW-3 described early 2020 through May 2020 as a “slow time.” As
CW-3 recounted, after May 2020, it was as if a “fire started,” with around 100 calls per week
from angry dealers regarding title issues. In fact, additional employees had to be hired on
CW-3’s team to deal with the increased call volume.
66.
CW-3 discussed the title issues with his/her manager, but nothing was done
and he/she was just given excuses like the DMVs were backed up, or the Titles Team had
been restructured. The title issues also were discussed during weekly meetings with the
Wholesale Team that were led by the Director of Wholesale.
67.
CW-3 said that the process to sell vehicles to Carvana became “way too easy.”
CW-3 explained that when he/she first started working at Carvana more than a year before
the beginning of the Class Period, Carvana refused to complete purchases when it was
discovered that sellers misrepresented information they provided to Carvana regarding
vehicle history and condition. In those instances, CW-3 said customers may have been
required to re-submit the questionnaires with accurate information prior to purchase.
However, sometime in 2019, when CW-3 was a member of the Trades Team which is part of
the STC Team, Carvana started accepting “all vehicles” and did not inspect them before
purchasing them anymore. CW-3 said that Carvana purchased a lot of “trash vehicles” after
this change, which had to be sold at auction.
68.
CW-3 said that these relaxed standards also led to Carvana accepting more
stolen vehicles and vehicles with fraudulent paperwork, which CW-3 observed when he/she
was a part of the High Risk Team near the end of 2019. CW-3 said that one of the most
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 25 of 332
- 22 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
common problems he/she saw were sellers who did not disclose that they had a loan on their
vehicles. CW-3 said that sometimes the team was able to catch the person, but if they could
not then Carvana would just pay off the loan after they had purchased the vehicles, which
could mean that they were out upwards of $10,000 on that sale.
69.
Confidential Witness No. 4 (“CW-4”) was employed by Carvana as an
Operations Specialist in the Wholesale Department at an IRC in the Midwest from before the
Class Period began until early 2022. CW-4 worked in a supervisory role from the middle of
2021 until he/she left Carvana. The IRC at which CW-4 worked had over 600 employees
and housed over 10,000 vehicles when CW-4 left Carvana. CW-4’s duties and
responsibilities at Carvana included (but were not limited to) tracking inventory using
spreadsheets and various software programs, working with the Reconditioning Managers and
Department Leads to transition vehicles to wholesale, working with the Logistics and
Inventory Teams to transport wholesale inventory to other locations, and working with hubs
and auctions to manage inventory and arbitration claims.
70.
The team CW-4 oversaw at the IRC processed vehicles that were going to be
sold to the wholesale market, which meant that they would be sold through an auction,
usually to used car lots. CW-4 explained that if a car met any of the following criteria that
car would automatically be transferred to the wholesale team: (i) vehicles that had over a
certain mileage; (ii) were over a certain year; or (iii) had been in a reported accident. CW-4
also said that there was a separate retail team at the IRC. Accordingly, if the retail team’s
inspections of a vehicle discovered certain types of damage to the vehicles, those vehicles
would be sent to the wholesale team since they did not meet Carvana’s criteria to sell to the
retail market. CW-4 said that rust damage, including rusted out frames, was a common
reason why vehicles in the IRC were sent to the wholesale team. CW-4 also said that any
inoperable vehicles were sent to the wholesale team.
71.
CW-4 explained that, beginning in the first quarter of 2021, he/she was told in
a meeting that there would be a “a huge increase” in the volume of cars to be sold wholesale
and was asked how much room he/she had at the IRC to store cars. The meeting, which took
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 26 of 332
- 23 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
place over Zoom, included CW-4, an Area Manager, various Wholesale Managers,
Directors, and Associate Directors. As CW-4 put it, Carvana was “looking to buy what cars
they could.” Carvana’s initiative resulted in it buying “any car” that may have resulted in an
increase in trade-ins. CW-4 said that the volume of cars at the IRC increased from just over
100 in the beginning of 2021 to almost 1,000 over the course of the year, a 10-fold increase
that CW-4 called “pretty epic,” causing the IRC to run “out of room.”
72.
CW-4 said that Carvana was “picking up anything” regardless of the condition,
including cars that literally had “no rear ends,” even though the cars that were picked up
were supposed to be operable and drivable. CW-4 reiterated that Carvana was not supposed
to be receiving these kinds of vehicles but that “we were getting them on a regular basis.”
73.
CW-4 characterized Carvana as having a huge mess when it came to titles.
CW-4 recalled that when he/she left, there were over 70 wholesale vehicles at his/her IRC
alone for which Carvana lacked the title and were just sitting around. CW-4 also recalled
that there was a Google spreadsheet that tracked vehicles that needed titles for the entire IRC
(retail and wholesale), which CW-4 believes existed since CW-4 first started working at
Carvana. CW-4 emphasized that there were “a ridiculous amount of cars on the list.”
74.
CW-4 said that it was well known internally that title issues were a very
common problem at Carvana, and everyone complained about it. CW-4 recalled that the
Title Department was revamped three times, but they still had not gotten the title process
right, and CW-4’s understanding was that the title team had “tons” of titles they needed to
process and sometimes titles were lost. CW-4 recalled that the team was working on
acquiring some titles from years earlier. CW-4 also recalled that retail titles were “Priority
1” and wholesale titles were “Priority 2.” CW-4 also said that sometimes there were
arbitrations after vehicles were sold by Carvana. According to CW-4, title issues were the
most common reason for arbitrations.
75.
CW-4 said that the titles were supposed to be obtained at the time the vehicles
were being picked up (unless the vehicles were still being financed.) CW-4 said that
Carvana had four levels regarding title status: A, B, C, and D. The D category meant
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 27 of 332
- 24 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Carvana did not have the title. The C category meant that Carvana did not have the title, but
was working on getting it. The B category meant that the necessary steps to get the title had
been done, but Carvana was still waiting to receive the title. And the A category meant
Carvana had the title actually in-hand. According to CW-4, there were “tons of D” category
cars. As CW-4 explained, “everyone knew we have title issues.”
76.
In February 2022, one week before the ADESA acquisition was announced
publicly, CW-4 attended a meeting with the wholesale team, including director-level
employees. Carvana’s Chief Operating Officer, Huston, also attended this meeting. During
this meeting, Huston mentioned that ADESA’s title department was good and Carvana
would adopt their processes and improve the title process. These comments made CW-4
believe that Huston must have known about the title issues.
77.
CW-4 said that Carvana’s acquisition of ADESA was part of its efforts to
expand into new markets that might have been far from existing Carvana IRCs. According
to CW-4, when he/she was hired, “they were saying” that Carvana was not operating in
certain areas like the Pacific Northwest because of logistics. Carvana’s range for picking up
and delivering vehicles was only five hours out and five hours back (although third-party
haulers could go further). As such, for example, a customer in Ohio purchasing a car from
Carvana that was in Oregon would exceed this range. Although Carvana had operations in
some of the western states like California and Nevada, there was essentially not a hub or IRC
between Indiana and Oregon that would fall within the five-hour transportation limit. The
sheer amount of money it would cost to pay third-party haulers to cover such distances was
cost prohibitive. To that point, the “biggest reason” for acquiring ADESA had been its
locations which reflected that ADESA operated nationwide and had something like 51 or 52
locations. This had been discussed at the meeting.
78.
CW-4 said Carvana had a logistics team that was “dysfunctional.” CW-4 also
said there were vehicles sitting “everywhere,” and sometimes it seemed as though no one
was planning routes to pick them up. CW-4 also recalled an incident where he/she found a
vehicle at the IRC that had been missing in the system for two years. CW-4 found it just by
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 28 of 332
- 25 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
doing simple inventory tasks. CW-4 thought the logistics and inventory problems were “out
of this world,” and did not understand why there were so many problems at Carvana,
especially since it was not a new company. CW-4 said that it was difficult to describe how
disorganized the logistics team was. CW-4 tried to give them advice, but they were “all over
the place,” and logistics and inventory issues remained a constant problem throughout CW-
4’s employment at Carvana.
79.
According to CW-4, cars delivered to retail were not being added to the retail
inventory on the day they were received. Instead, the retail group might wait a week before
adding the cars to inventory. The reason for this was that, as soon as a car was added to the
retail inventory, it began the aging process. Therefore, the aging report would “look bad” if
there were cars sitting there for 20 days. As CW-4 put it, Carvana’s business was “all about
get-in/get-out” when it came to inventory, so the retail business would wait a week before
scanning vehicles into inventory, even if the vehicles were actually there. CW-4 said his/her
cars would “get caught in that mess.” The General Managers and Logistics Managers made
the decision not to enter these vehicles into the retail inventory. The General Managers did
“not want anything aging,” so they worked with the Logistics Managers to delay scanning in
vehicles.
80.
Confidential Witness No. 5 (“CW-5”) was employed by Carvana as a Market
Operations Manager in a West Coast hub from shortly before the Class Period began until
the middle of 2022. Employees at the hub at which CW-5 worked delivered vehicles to
customers that they had purchased from Carvana via the Company’s website and picked up
vehicles that customers had sold to Carvana. CW-5 was a high-level employee at the hub
and managed the hub’s day-to-day operations, including ensuring that vehicles were properly
delivered to customers. CW-5 also managed the employees at the hub. Additionally, CW-5
oversaw key performance indicators, budgets, profit and loss, and inventory management for
the hub. The inventory management function included reviewing which vehicles were in the
lot at the hub and the vending machine, and being aware of how many vehicles were
“coming and leaving.”
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 29 of 332
- 26 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
81.
CW-5 explained that the hub received vehicles from a Carvana IRC and many
of the vehicles had issues even though they supposedly had been thoroughly inspected at the
IRC. Sometimes the vehicles were damaged during transit. However, other times, the
vehicles clearly had damage prior to being sent from the IRC to the hub (and therefore must
have had this damage when they were at the IRC) but had not been noted on Carvana’s
website when the vehicle was sold, so the depiction of the vehicle on the website was not
“100% honest.”
82.
CW-5 said that because “so many cars” that came to the hub were damaged, it
became, and continued to be, “way too much.” CW-5 heard from his/her predecessor at the
hub that the issues with the quality of the vehicles had gotten worse “all of the sudden”
(apparently prior to when CW-5 joined Carvana), and the quality problems persisted
throughout CW-5’s own employment at Carvana. CW-5 explained that Carvana “overspent”
on repairs for vehicles because there were so many quality issues and the third-party
mechanics he/she hired for the repairs were very expensive. Overall, CW-5 felt that he/she
just “threw money” at vehicles that should have already been fixed at the IRC. CW-5
mentioned there were a lot of returns since Carvana had a seven-day money back guarantee
and there were so many issues with the vehicles. CW-5 said that returns were not too
common when he/she first started working at the hub, but got worse after spring 2022.
83.
CW-5 explained that when an individual wanted to sell their vehicle to
Carvana, that person had to fill out a survey regarding the quality of the vehicle and Carvana
essentially “took their word” for it when an offer was made. When an Advocate from the
hub went to pick up the car, the Advocate would try to confirm if the quality of the car
matched what the seller had represented. If it did not match, the Advocate was supposed to
call an “inside team” to get a reappraisal. However, there was only about a 50/50 chance that
the inside team would do their job and reappraise the vehicle. Sometimes Carvana’s inside
team would tell the Advocate to do the reappraisal, or Carvana’s inside team would just
agree to pay thousands of dollars for “super crappy” cars.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 30 of 332
- 27 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
84.
CW-5 said that the employees at the hub had to check the paperwork for the
vehicles they were picking up from the sellers or that were dropped off at the purchaser’s
premises. They also had to check IDs to make sure they were not fake. CW-5 said that there
were a lot of fake, or at least “sketchy,” buyers and sellers, including some accounts that
were obviously fraudulent. CW-5 also said that sometimes customers did not have insurance
or did not show up for their appointments. CW-5 said that money was wasted when the hub
employees hauled a vehicle out to a customer that had intended to purchase the vehicle, but
the customer did not show up for the appointment.
85.
CW-5 said that inventory for his/her hub was low when CW-5 first started
working at the hub so it arguably made sense to overpay for poor quality vehicles, but
Carvana continued to pay “crazy amounts” of money for vehicles after the inventory level at
the hub was no longer low. CW-5 recalled that demand really slowed down around June or
July 2022 because of increasing interest rates, but Carvana did not reduce the number of
vehicles Carvana bought, so Carvana ended up with too much inventory. CW-5 said that
demand started to slow down before June 2022, but it became a bigger problem around
summer 2022. CW-5 heard this was a trend at other hubs from other Operations Managers.
CW-5 said that there were weekly meetings for the Operations Managers, and CW-5
occasionally spoke with them outside of these meetings. CW-5 also saw communications
from other managers on Slack messaging software.
86.
The cost of the third-party mechanics was the main key performance indicator
CW-5 was concerned about for his/her hub, and CW-5 discussed it with his/her manager
during meetings.
87.
CW-5 attended a leadership summit with all Market Operations Managers and
senior leadership in April 2022 at Carvana’s headquarters in Phoenix, Arizona. This summit
included a Q&A session with Garcia Junior. During this session, the executives answered
questions that had been submitted ahead of time and they also took additional questions
afterward. CW-5 recalled that someone asked about what was going to be done regarding
the quality issues with the vehicles that the hubs were receiving from the IRCs. CW-5
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 31 of 332
- 28 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
believes this individual may have asked something along the lines of whether the 150-point
inspections were actually being performed since there were so many issues with the vehicles
the hubs were receiving.
88.
Confidential Witness No. 6 (“CW-6”) was employed by Carvana as a
Centralized Planning Analyst for more than six months during the Class Period and was also
employed as a Finance Analyst for more than six months during the Class Period. CW-6’s
role in Centralized Planning involved managing vehicle pick-ups and deliveries to
customers.
89.
In the Finance Analyst role, CW-6 heard from other employees (possibly
during a staff meeting) that there were inefficiencies with logistics. For instance, it was the
practice of the Logistics Group to initiate delivery of a vehicle as soon as a customer placed
an order for a vehicle, but before any payment had been made. If a customer subsequently
canceled the order, which customers did, then the delivery that was en route had to be
canceled, and Carvana incurred the cost unnecessarily. CW-6 pointed to other logistical
inefficiencies. For instance, there might be two hubs fairly close to each other and a vehicle
was scheduled to be delivered to one of the hubs, only for the delivery to be shifted to the
other hub for one reason or another. Carvana’s processes were such that sometimes, instead
of simply routing the delivery en route to the nearby hub, the delivery would go back to the
point of origin and then be shipped to the intended hub.
90.
CW-6 said he/she “absolutely” learned that Carvana was incurring excessive
costs in regards to transporting vehicles from the hubs and IRCs to customers, especially
with the geographic expansion of Carvana. CW-6 said that picking up a customer’s trade-in
and delivering another car to the same customer who had provided the trade-in required two
separate trips. CW-6 said “they were talking about consolidating” the trips into a single one,
but he/she does not know if this was ever implemented.
91.
Confidential Witness No. 7 (“CW-7”) was employed by Carvana as an
Associate Buyer in Tempe, Arizona for more than a year during the Class Period. As an
Associate Buyer, CW-7 reviewed more than 15,000 digital vehicle condition reports and
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 32 of 332
- 29 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
purchased more than 2,000 vehicles valued at $65 million from various vehicle auction
platforms. In addition, CW-7 appraised vehicles using standard vehicle appraisal tools and
audited more than 2,000 listings for errors and features that fell below inventory quality
standards.
92.
CW-7 said that Carvana used a Tableau server which distilled various financial
details, including “the total costs” Carvana incurred to acquire, recondition, and otherwise
make the cars available for sale, as well as what the vehicles had actually sold for, from
which was derived the gross profit per unit, i.e., profit for each vehicle sold. CW-7 said that
as of spring 2022, the Tableau server had shown the average GPU for the vehicles to which
CW-7 had access had been about $100, but in fall 2022, the GPU for the vehicles CW-7
could see in Tableau had declined to $23, meaning a near 80% decline in less than a year.
93.
CW-7 said he/she believed the Tableau server data was updated daily. While
the majority of vehicles in the Tableau system to which CW-7 had access had been procured
by way of online wholesale auctions, some of the cars that had been procured by way of the
STC program may have been included.
94.
CW-7 said that another important metric available in the Tableau server was
the average number of days on-site, which was the number of days from when Carvana had
purchased a vehicle to when the vehicle had been sold. In the same timeframe as when the
GPU was declining, the average number of days on-site for the same vehicles had been
increasing.
95.
CW-7 confirmed that Carvana definitely overpaid for vehicles. In CW-7’s
opinion, the reason Carvana ended up overpaying for vehicles reflected the Company’s
“mission statement” to become the biggest online car company. CW-7 noted that Carvana
“loved to compare” itself to CarMax and Amazon. CW-7 believed Garcia Junior wanted to
make Carvana as big as possible regardless of the cost or internal differences in opinion in
regards to strategy.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 33 of 332
- 30 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
96.
CW-7 said that he/she and other employees were told that Carvana acquired
ADESA so that Carvana could get within 100 miles of the most populated areas of the
country and thereby extend its reach.
97.
Confidential Witness No. 8 (“CW-8”) was employed by Carvana as a Buyer
and Category Analyst in Carvana’s Tempe, Arizona, Corporate Headquarters for more than
18 months during Class Period. As a Category Analyst, CW-8 oversaw all aspects of
inventory management, pricing, inventory levels, margins, and analytics for certain vehicle
segments. CW-8 also managed pricing operations, including executing initial price targets
and markdowns in order to move inventory. As a Buyer, CW-8 worked to acquire vehicles
for Carvana from auctions using a system called vAuto, which was owned by Manheim.
98.
CW-8 and his/her team used Tableau software, from which numerous reports
could be derived, and Microsoft SQL, which connected to a computer server where data was
stored. Using this software enabled CW-8’s team to see inventory data and various metrics,
including how many vehicles Carvana had in inventory, how long these vehicles had been
sitting around without being sold, Carvana’s website traffic, and supply/demand ratios.
99.
Category Analysts like CW-8 set the final prices based upon the ranges from
the algorithm and other factors, such as the current inventory Carvana had on hand. CW-8
also worked with the data scientists on “markdowns” (i.e., lowering prices on previously
priced vehicles) for vehicles that were not selling. This process involved lowering the price
of a vehicle by $500 to $1,000 and then potentially lowering it again if the vehicle still did
not sell after 7 days.
100. According to CW-8, sometimes Carvana’s Buyers paid “outrageous” prices for
vehicles, which lowered margins because the vehicles could not support higher selling
prices. CW-8 explained that vehicles were rated Tier 1 to Tier 5, and sometimes the Buyers
paid up to $5,000 over the market price to acquire vehicles in certain Tiers. CW-8 also said
that around the summer/fall 2021 timeframe there was a lot of pressure on CW-8’s team, the
Buyers, and the STC team to get every vehicle possible, no matter the price. As a reflection
of the efforts to procure as many vehicles as possible, CW-8 added that the expectation was
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 34 of 332
- 31 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
that the titles for the vehicles could be acquired later if they were not available at the time of
purchase.
101. CW-8 said that, as a result of this pressure, Carvana essentially “overbought”
vehicles during 2021, and CW-8’s team started to become concerned as they saw the
inventory levels rising. Around fall and winter of 2021, CW-8’s team began to raise
concerns about the high level of inventory to director-level employees because it was
becoming difficult to sell certain types of vehicles, CW-8’s team had to lower prices, and
margins were decreasing. CW-8’s team had weekly meetings every Thursday with an
Associate Director where they would go over each segment and provide inventory updates.
CW-8’s teammates raised concerns about the excess inventory in a couple of these meetings
and one of his/her coworkers, who was a Financial Analyst, made a report about this topic.
However, his/her team’s concerns were ignored and the Buyers were told to continue buying
vehicles. By March 2022, “everything was on fire” because Carvana was “totally
overstocked” and the forecast had to be lowered.
102. Confidential Witness No. 9 (“CW-9”) was employed by Carvana as a Lead
for Customer Experience in Carvana’s Tempe, Arizona headquarters for more than six
months during the Class Period. As a Team Lead, CW-9 supervised more than a dozen
Customer Experience Advocates. Prior to Q1 2022, CW-9’s team had handled only post-
sales calls from customers, primarily pertaining to issues with vehicle delivery delays,
registration and title delays, and vehicle quality and condition complaints. Beginning around
Q1 2022, Carvana merged the pre-sales and post-sales Customer Experience teams, so that
CW-9’s team began handling both types of calls. Pre-sales calls primarily pertained to
questions about the purchasing process, including the STC program.
103. CW-9 observed a lack of action taken by other Carvana departments in regard
to issues such as vehicle quality, including delivering vehicles without the vehicle
inspections being completed, vehicle delivery delays, registration delays, the expiration of
temporary plates, and Carvana encouraging these customers to continue driving the cars
while reassuring them that Carvana would reimburse them if they were to be cited.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 35 of 332
- 32 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
104. CW-9 said the Customer Advocates on his/her team handled anywhere from 10
to 30 calls a day (the number varied depending on call duration). The calls were from all
over the country. Registration delays, which primarily meant that a title had not been
provided to the customer, amounted to roughly half of the calls the team had handled before
the team began handling both pre- and post-sale calls. CW-9 said that he/she noticed an
increase in the number of calls from customers experiencing title delays beginning in Q1
2022.
105. CW-9 attended Carvana all-hands calls with Garcia Junior. According to CW-
9, the title and registration issues Carvana was experiencing were raised by employees on
these calls. CW-9 said that the questions to Garcia Junior from employees were along the
lines that they were getting inundated with calls regarding registration problems and they
asked Garcia Junior to elaborate on steps the Company was taking to mitigate the problems
and provide better solutions. Garcia Junior’s response was always along the lines that
Carvana was working on a solution. CW-9 particularly remembered a call in January 2022
in which questions about registration were posed to Garcia Junior.
106. Vehicle quality issues also represented a large portion of the calls that CW-9’s
team handled. CW-9 said those issues ranged from check engine lights coming on, to
mismatched tires, faulty emissions, damage to the vehicle (including at the time of delivery),
accidents the vehicle had been in that had not been reported to CarFax, and vehicle recalls
that had not been processed correctly – “or at all.”
107. When the pre-sales and post-sales teams were merged in Q1 2022, CW-9
estimated that probably 30% of calls pertained to pre-sales, 35% to 40% pertained to
registration and title issues, and 25% to 30% pertained to quality. Prior to the merger, the
calls had been fairly evenly split between quality issues and registration and title issues.
CW-9 dealt with customers in other states who were experiencing quality problems.
According to CW-9, the quality shortcomings and problems seemed to CW-9 to be pervasive
everywhere Carvana operated, not just in specific areas.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 36 of 332
- 33 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
108. CW-9 said that, as Carvana grew, it began accepting substandard cars and that
the STC program took vehicles that were often converted for sale through wholesale
channels.
109. Confidential Witness No. 10 (“CW-10”) was employed by Carvana as a
Registration Specialist in Carvana’s Tempe, Arizona headquarters for more than six months
during the Class Period. As a Registration Specialist, CW-10 reviewed documentation from
customers registering vehicles in multiple states, made sure that certain documents (i.e.,
proof of insurance, driver’s license) were valid, completed vehicle title work with dealer and
customer information, worked with multiple vendors to ensure that all deal and customer
information had been entered correctly and that all necessary documentation had been
received before submitting to the DMV, and created temporary operating plates for
customers. In mid-2022, CW-10 became a Specialist Point of Contact (“SPOC”), where
CW-10 helped Customer Advocates answer customer questions regarding registration. As a
SPOC, CW-10 also worked with his/her team and others in the Company to make sure that a
customer’s registration was completed correctly.
110. When CW-10 was on the SPOC team, he/she recalled one instance where a
vehicle had been purchased from a seller in Texas, but the seller never provided a copy of the
title. Carvana then sold that car to a buyer in Illinois. However, Carvana was unable to get a
duplicate copy of the title, CW-10 believes, because there was an issue with the vehicle’s
Carfax report. CW-10 also said that there were other issues with getting copies of the titles
for new customers, including when sellers had not signed the paperwork that allowed
Carvana to get a duplicate title, or when sellers had not signed a power of attorney form that
allowed Carvana to transfer the title. CW-10 said that sometimes getting the title could be a
“big mess.”
111. CW-10 recounted that there had been two vehicle sales when he/she had been
on the SPOC team for which the Customer Advocates had been trying to get the titles for
over a year, which meant the buyers had been unable to drive their vehicles. CW-10 said
there had been a lot of research notes and in regards to one of them, there had been a major
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 37 of 332
- 34 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
discrepancy between the mileage in the Carfax report and the vehicle’s odometer. CW-10
reported this, and it was apparently found to have happened a “multitude” of other times.
112. CW-10 recounted another incident in which the buyer had been unable to drive
the vehicle he/she had purchased from Carvana because Carvana was waiting to obtain the
title, but all CW-10 could say to that customer was “we’re working on it.” CW-10 learned
about the other incidents when he/she had made a report to his/her Team Captain (who was
not CW-10’s direct supervisor, but someone from whom he/she sought answers to questions
he/she had), and CW-10 brought it up to the Supervisor, and then to a manager. In this
discussion, the Team Captain apparently referenced another team that was involved with
issues and discrepancies between the Carfax reports and the actual condition of the vehicles
Carvana had sold to customers, including the odometer readings not matching the Carfax
reports. CW-10 said there was “a huge hiccup” with the Carfax report for another vehicle
that had a salvage title, which apparently set forth who had repaired the vehicle and when,
but the mileage on the title ended up being off by 100,000 miles. This discrepancy had not
been picked up by Carvana before the erroneous mileage was entered into Carvana’s system
and Carvana sold the vehicle.
113. In CW-10’s opinion, these kinds of problems transpired because Carvana
primarily just wanted to “get money.” Specifically, CW-10 thinks it is the case that mistakes
like these occurred because Carvana was so anxious to get and sell cars that they did not
adequately check everything. CW-10 confirmed and said it was a common opinion amongst
personnel that Carvana was buying cars at a breakneck speed and that it was not adequately
checking them. This common sentiment was also shared by CW-10’s former supervisor. On
the basis of customer questions and complaints, it had seemed to him/her that these kinds of
problems with poor vehicle quality had been happening a lot.
114. Confidential Witness No. 11 (“CW-11”) was employed by Carvana as a
Market Operations Team Lead at a hub in the Southeastern U.S. for over a year during the
Class Period. CW-11’s role involved running the team that delivered vehicles to customers
that customers had purchased online from Carvana. CW-11’s team also brought vehicles to
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 38 of 332
- 35 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
the Company’s local vending machine, and they also picked up vehicles that customers had
sold to Carvana online. CW-11 oversaw multiple Managers, Customer Advocates, and Lot
Attendants.
115. CW-11 said that Carvana was essentially purchasing vehicles “sight unseen,”
and initially his/her team was not expected to do much to verify the condition of the vehicles.
However, he/she recalled attending a conference call with the buying department from the
corporate office, and was told that CW-11’s team should call the buying department when
they went to pick up a vehicle if the vehicle was not in the condition that had been
represented by the seller. CW-11 recalled an incident where a seller dropped off a Jeep
Wrangler that was in “awful” condition, including missing an entire panel, but the seller had
been offered “a ton of money” for the vehicle by Carvana’s website because he/she had listed
the vehicle in perfect condition. CW-11 called the buying department and sent them pictures
of the vehicle, but was told “it’s fine” and that Carvana would purchase the vehicle for the
quote that had been provided by the website.
116. CW-11 recounted that the same day as the Jeep Wrangler was picked up, three
other vehicles with similar problems were dropped off at the hub. Two of these were
handled by CW-11’s subordinates and one of them – a Ford Explorer – CW-11 had handled
himself/herself. The Ford Explorer was completely lacking two A-frame pillars, even
though the seller had not identified any such deficiencies and had rated the car as in good
condition. As with the Jeep Wrangler incident, when CW-11 contacted the buying
department, he/she was told to “do the deal.” The same thing happened to members of CW-
11’s team, so they stopped calling the buying department when vehicles were in bad
condition since doing so did not seem to accomplish anything. CW-11 said it was definitely
the case that Carvana’s practices in such situations meant Carvana was purchasing vehicles
at inflated prices. According to CW-11, the buying department did not actually care about
the condition of the vehicles and just wanted to “get as many cars as it could.”
117. CW-11 said the majority of reconditioned vehicles were stored at the IRCs.
However, as Carvana’s accumulation of vehicles increased, this also necessitated his/her hub
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 39 of 332
- 36 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
to begin storing vehicles at an adjacent lot and then also at another lot located perhaps half-a-
mile away. Altogether, there ended up being 300 vehicles stored at the 3 lots. As CW-11
put it, Carvana was “always looking for places” to store vehicles.
118. Employees at CW-11’s level and market team leaders attended Zoom calls
with the Company’s C-level executives (including Garcia Junior and Huston) probably every
quarter. On one of these Zoom calls in early 2021 attended by Garcia Junior, a meeting
attendee expressed concern about the poor quality of the cars Carvana was purchasing.
According to CW-11, one of the C-level executives responded: Carvana was “just worried
about growth and not procedural” operations. CW-11 recalled that this same C-level
executive spoke about the importance of growth on that call and suggested that employees
read a certain book (he/she could not remember which one) because it would help them
understand that growth is all that matters. CW-11 remembers he/she recalled thinking:
“Wow, they don’t care about operations, just growth.” There was time for questions and
answers at the end of these calls, and during every call someone in a market operations role
complained about the poor quality of the vehicles they received from the reconditioning
centers. CW-11 said that this was the number one complaint from the operations side of the
business because they felt guilty delivering poor quality vehicles to customers. CW-11 was
pretty sure that these C-level Zoom calls were recorded because he recalled listening to a
recording of one of the Zoom calls that he had missed.
119. CW-11 eventually learned that selling vehicles without the title and issuing
temporary plates from other states was a common practice at Carvana. CW-11 started to
learn the truth because he/she spoke with employees at the hub who worked on registering
vehicles. They told CW-11 that there was a drawer of documentation for vehicles that had
been sold by Carvana but which they could not get registered since the titles were missing.
120. CW-11 recalled reviewing a Google document spreadsheet of the cars for
which registration could not be completed since the titles were missing and there were notes
made by Carvana’s Arizona personnel along the lines that they were working on obtaining
the title.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 40 of 332
- 37 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
121. Confidential Witness No. 12 (“CW-12”) was employed by ADESA before
the Class Period as a Director responsible for working with large dealer consortiums to buy
and sell wholesale vehicles. After Carvana purchased ADESA, CW-12 was promoted to
Executive Director and continued to work with large dealer consortiums to buy and sell
wholesale vehicles. CW-12 remained in that position until CW-12 left Carvana in the early
part of 2023.
122. According to CW-12, a major reason that Carvana purchased ADESA was to
have access to numerous additional centers across the country to recondition the cars. CW-
12 met with Garcia Junior, who informed CW-12 that Carvana’s management team needed
ADESA’s reconditioning centers, which were profitable and located nationwide. CW-12
spoke to Garcia Junior a few times, and he continuously inquired about how ADESA
operated and functioned. CW-12 also attended quarterly town hall meetings conducted over
Zoom and attended by Garcia Junior.
123. CW-12 observed that Carvana was setting up new marketing towers and
adding new sites across the nation. Accordingly, CW-12 said Carvana needed to “get their
lots filled and find cars.” CW-12 continued that Carvana needed to show that they were
growing and needed to buy cars to fill its lots. This resulted in Carvana aggressively buying
vehicles.
124. CW-12 said that the aging of inventory items (also called “low turn”) is
something that everyone in the industry monitors. CW-12 continued that if a vehicle is on
the list after 60-90-100 days, it does not look good.
VI.
DEFENDANTS’ PUMP-AND-DUMP SCHEME AND FRAUDULENT
COURSE OF BUSINESS UNDER RULE 10b-5(a) and (c)
125. As set forth below, Defendants engaged in a fraudulent pump-and-dump
scheme and course of business to boost Carvana’s retail sales – its most important metric –
and thus, its stock price for the purpose of selling shares at an artificially inflated price.
Unbeknownst to investors, Defendants carried out their pump-and-dump scheme through the
following concealed methods and means:
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 41 of 332
- 38 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
(a)
Garcia Senior causes DriveTime to enter into a sham pass-through sales
arrangement with Garcia Junior at Carvana to boost materially Carvana’s reported retail
sales;
(b)
Garcia Junior and Jenkins drastically lower Carvana’s purchasing and
verification standards, buying low-quality cars to induce trade-in sales;
(c)
Garcia Junior and Jenkins flout state title and registration laws to secure
retail sales before competitors;
(d)
Garcia Junior and Jenkins spearhead a rapid and unsustainable
nationwide expansion without regard to profitability to boost Carvana’s reported retail sales;
(e)
Garcia Junior and Jenkins make materially misleading statements and
omissions, including statements and omissions designed to conceal the scheme and convince
investors that Carvana’s retail growth was sustainable;
(f)
Garcia Senior and Jenkins enter into and/or modify their 10b5-1 plans
while in possession of MNPI to capitalize on the artificial inflation in Carvana’s stock price;
and
(g)
Garcia Senior and Jenkins sell nearly 14.3 million Carvana shares –
while in possession of MNPI – for proceeds of nearly $3.76 billion.
THE PUMP
126. Immediately preceding the Class Period, two events combined to drive down
Carvana’s stock price. First, COVID-19 created a global pandemic, driving down stock
markets. Second, and most importantly, analysts expressed alarm that Carvana’s growth was
slowing. On February 27, 2020, a Morgan Stanley analyst stated: “If the growth starts to
slow significantly, and the company is still not profitable, there will be a transition away
from growth investors, and we believe that how investors think about the valuation of the
stock will change.” That same day, an analyst from CFRA stated: “We think investors will
be less willing to look past CVNA’s slowing top line growth and persistent lack of
profitability than they were in 2019.” In response, Carvana’s stock price fell to a multi-year
low. Defendants could not let this happen; they needed the Company to have a high
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 42 of 332
- 39 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
valuation so they could cash out their personal holdings in Carvana stock at a top dollar.
Thus, to pump Carvana’s stock price, Defendants planned and executed a scheme to inflate
the Company’s retail vehicle sales growth and convince investors that this growth was
sustainable.
A.
Artifice No. 1: Garcia Junior and Garcia Senior Make a Sham
Related-Party Deal Between Carvana and DriveTime to Boost
Carvana’s Retail Sales
127. To increase Carvana’s retail unit sales, Garcia Senior caused his company,
DriveTime, to enter into a series of revenue pass-through arrangements with Carvana by
September 2021. There was no legitimate business purpose for these sham deals as they
lacked any economic substance for Carvana, and served only to inflate Carvana’s reported
retail sales. Under this father-and-son deal, Carvana would acquire cars from DriveTime,
sell the car on its website, and then pass the entire proceeds from the sale back to
DriveTime. While Carvana merely acted as DriveTime’s middle man or agent, it still
recorded the revenue and retail unit sales on its books to pump up Carvana’s stock.10
128. Defendants belatedly buried the fact that Carvana “purchase[d] reconditioned
vehicles from DriveTime . . . [and Carvana’s] purchase price [paid to DriveTime] is the list
price on Carvana.com after a customer has placed an order on the vehicle” (i.e., Carvana’s
sales price) in its March 2022 Form DEF 14-A Proxy Statement (the “2022 Proxy
Statement”) signed by Garcia Junior. Further, Defendants concealed the full extent of these
arrangements from investors, offering only piecemeal and inconsistent disclosures
concerning the pass-through revenue arrangements. For example, Carvana did not report
10 The SEC is currently investigating Carvana for its related-party deals. Critically, Garcia
Senior’s privately-held companies sell VSCs and GAP insurance coverage to Carvana’s
customers, generating an even greater revenue stream for him personally (and for Garcia
Junior via his DriveTime trust). As a law professor at Washburn University stated, Garcia
Junior and Garcia Senior “‘have a convoluted tangle of interrelated companies and related-
party transactions, and it’s very difficult to understand or pull apart.’” Jane Hahn, Family
Business Deals Help Fuel Carvana’s Explosive Growth, Wall St. J. (Dec. 17, 2021) (the
“December 17, 2021 The Wall Street Journal article”). Moreover, “[t]he problem with
related-party transactions with large shareholders . . . is ‘they don’t need to make money
from this entity because they own the other entities.’” Simply put, Carvana “‘doesn’t have to
make money for [the Garcias] to make money.’”
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 43 of 332
- 40 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
actual retail unit sales sold pursuant to these arrangements. Rather, they disclosed only
information regarding the expense incurred to purchase vehicles from DriveTime. Carvana
also disclosed inconsistent facts about these arrangements across its SEC filings, which
included SOX certifications signed by Garcia Junior and Jenkins. For example, in Carvana’s
August 5, 2021 quarterly report on Form 10-Q for the period ended June 30, 2021 (the “Q2
2021 10-Q”), Defendants disclosed only the “cost of goods sold” incurred by Carvana.
Then, Carvana’s 2022 Proxy Statement belatedly revealed that these transactions lacked
economic value and provided the “total expenses” incurred by Carvana for the first time.
Tellingly, the Company’s February 24, 2022 annual report on Form 10-K (the “2021 10-K”),
for the fiscal year ended December 31, 2021 (“FY 2021”) issued just a month before, fails to
mention these terms, despite the fact that these deals had been occurring for months.
129. Garcia Junior and Garcia Senior’s sham DriveTime pass-through arrangements
had a material impact on Carvana’s reported retail sales volume and growth. Although the
full details regarding them were concealed, certain expense figures can be used to estimate
the number of cars involved in these arrangements. As shown below, Carvana’s estimated
unit sales from these arrangements were material as they made up over 169% of Carvana’s
reported sequential growth in Q4 2021 and 19% of its sequential growth in Q3 2021.11
11 Given Defendants’ inconsistent and piecemeal disclosures on these sham deals, the chart
depicts a conservative calculation of the unit sales pursuant to these arrangements.
Alternatively, if one utilized: (i) the estimated retail units sold pursuant to the Wholesale
Vehicle Purchase Agreements that Carvana and DriveTime entered into on September 29,
2022; and (ii) the estimated retail units sold pursuant to other “Retail Vehicle Acquisition”
agreements that Carvana and DriveTime entered into during Q2 2021, which Plaintiffs
believe is an accurate way to calculate the estimated retail units sold pursuant to the sham
arrangements, these sales would have actually accounted for 660% of Carvana’s sequential
growth in Q4 2021, 19% of its sequential growth in Q3 2021, and 17% of its reported annual
growth.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 44 of 332
- 41 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
B.
Artifice No. 2: Defendants Lower Carvana’s Purchasing and
Verification Standards, Purchasing Lower Quality Cars to
Induce Trade-In Sales
130. As the Class Period progressed, Defendants drastically increased the number of
cars Carvana purchased from customers to induce trade-ins and increase inventory, which
increased retail sales and, thus, Carvana’s stock price. To carry out this surge of purchases,
unbeknownst to investors, Defendants abruptly and drastically lowered Carvana’s purchasing
and verification standards for the cars that it purchased following Q2 2020.
131. For example, according to CW-3, before the beginning of the Class Period,
Carvana was strict about not completing purchases when it discovered that sellers had not
accurately described the condition of their vehicle, even making them re-submit
questionnaires with accurate information. Immediately before and during the Class Period,
however, CW-3 stated that Carvana relaxed its standards to accept “all vehicles” and did not
genuinely inspect cars before purchasing them anymore. CW-3 observed that Carvana
purchased a lot of “trash vehicles” after lowering its standards. CW-5 heard from his/her
predecessor at the hub that the issues with the quality of the cars had gotten worse “all of the
sudden.” Likewise, CW-1 stated that Carvana lowered its standards when it came to
accepting trade-ins as long as the customer bought a car from Carvana. In fact, CW-4 was
told at a meeting of Carvana’s initiative to buy “any car” to secure trade-ins.
132. Indeed, CWs-1, 3, 4, 5, 8, 10, and 11 report that, during the Class Period, it was
Carvana’s general practice to purchase cars sight unseen and do little to nothing to verify the
accuracy of sellers’ representations of their vehicles. CW-11 reported that it appeared that
the buying department did not actually care about the condition of the vehicles and just
wanted to “get as many cars as they could.” CW-5 stated that there was only a 50/50 chance
that a vehicle would be reappraised if it did not match the seller’s description. Likewise,
CW-2 confirmed that he/she was overruled when voicing concerns about the quality of
vehicles as “leadership would say take it.” As CW-7 reported, he/she understood that the
reason Carvana did this was the Company’s “mission statement” to become the biggest
online car company.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 45 of 332
- 42 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
133. CW-1 explained that Carvana “wouldn’t care” about the poor quality. CW-1
said that as long as someone was also “buying from us,” then Carvana would take the trade-
ins. According to CW-1, this practice resulted in, for instance, Carvana purchasing cars as a
trade-ins that were missing a backseat or were riddled with bullet holes.
134. By lowering Carvana’s stated purchasing and verification standards as multiple
CWs describe, Carvana was able to increase the number of cars bought from customers in
Q3 2020 by a staggering 261% from the prior quarter. Similarly, the following quarter,
Carvana bought 100% more cars from customers than it had the prior year.
135. Unbeknownst to investors, Defendants’ concealed course of conduct caused
Carvana to be flooded with “trash cars” that were unfit to be sold retail and had to be sold
wholesale. Indeed, Carvana’s wholesale volume spiked immediately after Defendants
lowered their purchasing and verification standards, with wholesale volume growing over
100% year-over-year in Q4 2020 and Q1 2021 before surging 500% in Q2 2021. While
investors were aware that buying cars from customers could potentially result in some
increase in wholesale vehicles, investors did not know that Carvana was intentionally
ignoring its purchasing standards to ensure a flood of wholesale cars or that Carvana was
losing money on wholesale vehicle sales.
136. Indeed, as alleged herein, Defendants concealed that Carvana lost money on its
wholesale vehicle sales by obscuring one of its “two key drivers of variable unit economics”
(i.e., the amount of actual profit Carvana earned on each wholesale vehicle it sold):
operations expenses. They did so by: (i) excluding certain per-vehicle operations expenses
from its calculation of wholesale GPU; and (ii) not disclosing or quantifying these excluded
costs separately so investors could calculate overall profitability of wholesale sales on their
own. These per-vehicle operations expenses, which were only separately broken out and
quantified for the first time after the Class Period, included material costs associated with
completing a wholesale sale, such as the full cost to ship a car to a wholesale auction site
and title and registration costs. See, e.g., ¶192(c)(ii) (further detailing how Defendants
obscured that Carvana sold wholesale vehicles at a loss).
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 46 of 332
- 43 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
137. Further, unbeknownst to investors, this increase in wholesale volume created a
costly, logistical nightmare as “wholesale units acquired from customers have typically been
transported to the nearest Carvana IRC, generating additional vehicles [sic] moves and
increased complexity in our multi-car logistics network.”12 Defendants later admitted that,
on or about August 2021, the increased wholesale volume also led to “[m]ore vehicle moves,
more miles traveled, . . . constrained routes,” a “higher . . . backlog on constrained routes, . . .
[l]ower performance on operational metrics,” and “constraints in our logistics network.” At
bottom, the problems Defendants created to maintain their growth story came at a steep
price. Indeed, in addition to the above, growth in wholesale volume from buying cars from
customers spiked Carvana’s logistical expenses, and eventually required Defendants to issue
billions of dollars of high-interest junk bonds to acquire ADESA, a nationwide auction house
(i.e., a wholesale business), to stay afloat and address this strain.
138. Garcia Junior and Jenkins were critical to this artifice. In fact, as discussed in
more detail below, they made materially false and misleading statements and omissions to
further conceal this fraudulent conduct. See ¶151, see also ¶¶190-210.
C.
Artifice No. 3: Garcia Junior and Jenkins Embark on a Rapid
and Unsustainable Nationwide Expansion
139. To boost sales, Garcia Junior and Jenkins also spearheaded a rapid and
unsustainable nationwide expansion plan to more than double the Company’s footprint in
just six quarters. Critically, however, because Carvana had already achieved market
coverage near its existing IRCs, the vast majority of new markets added were necessarily a
significant distance from existing IRCs as demonstrated by the map below. Thus,
unbeknownst to investors, Garcia Junior and Jenkins were increasing Carvana’s retail sales
with unprofitable and less profitable sales in far-flung markets.
12 Update on Carvana Operating Plan Slide Deck (May 2022) (the “May 2022 Operating
Plan”). Defendants’ concealed conduct also created overwhelming inventory growth
generally, as well, which similarly caused unsustainable “parking constraints,” “load
imbalances,” and “increased the number of vehicles moved and total vehicle miles traveled,
adding to costs and logistics network complexity.”
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 47 of 332
- 44 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
140. Garcia Junior and Jenkins knew or recklessly disregarded that these new
markets were less profitable due to their greater distance from Carvana’s existing IRCs. As
stated in a September 13, 2022 Vehicle Marketing article, Garcia Junior knew all along,
“‘[y]ou need to warehouse and recon[dition] those cars to get them retail ready . . . . You
need a place to stage the logistics . . . .’” In fact, Garcia Junior acknowledged he knew “how
powerful proximity can be,” when he later revealed that Carvana incurred at least $750 more
in costs per retail car sold in markets 200 miles or more from a Carvana IRC. Even markets
that were 100 miles from an existing IRC materially increased costs because “SHORTER
DISTANCE = SAVINGS . . . Increased customer conversion; Faster inventory turn times;
Lower inbound transport & logistics costs; [and] Lower shipping costs to customers.”
Nevertheless, a full 90% of the new markets opened during the Class Period were more than
100 miles away from an existing IRC and 65% were more than 200 miles from an existing
IRC.
141. In addition, unbeknownst to investors, Defendants’ expansion ravaged
Carvana’s purportedly efficient nationwide logistics and reconditioning network and
significantly increased Carvana’s expenses. Indeed, Defendants later admitted that,
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 48 of 332
- 45 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
beginning on or around August 2021, it created “[m]ore vehicle moves, more miles
traveled,” a “[h]igher number of constrained routes,” and a “[h]igher degree of backlog on
constrained routes.”
142. Moreover, by the second half of FY 2021, Defendants were forced to begin
spending millions of dollars on costly third-party logistics and reconditioning providers to
make up for Carvana’s internal logistics and reconditioning deficiencies. After the Class
Period, Defendants disclosed “[r]econditioning and [i]nbound [t]ransport [c]osts [p]er [r]etail
[u]nit” for the very first time, which revealed that, by Q3 2021, these costs had spiked $201
per vehicle and were continuing to steadily increase throughout the Class Period. In fact,
between mid-2021 and Q4 2021, these costs had spiked $369 per vehicle. And by Q1 2022,
they had spiked more than $600 per vehicle. Defendants further admitted the need to
“[r]educe third-party inbound transport share,” “[i]n-sourc[e] third-party pickups” for
Carvana’s “[l]ast-mile delivery,” and “[r]educe third-party shipping” to curtail these excess
logistics expenses.
143. Garcia Junior and Jenkins were able to conceal this fraudulent course of
conduct, in part, by: (i) excluding significant per-vehicle “operations expenses” (i.e.,
“expenses associated with completing retail and wholesale vehicle sales”), which included
outbound logistics and title and registration, from its calculation of wholesale and Retail
GPU; and (ii) not adequately disclosing or quantifying these excluded costs separately so
investors could calculate overall profitability of vehicle sales on their own. This
concealment was meant to deceive investors as Garcia Junior included these operations
expenses in assessing the profitability of each car sold internally, and Carvana later admitted
that operations expenses (i.e., “expenses associated with completing retail and wholesale
vehicle sales”) were one of the “two key drivers” of profitability per vehicle sold.13 If
13 Indeed, during the Company’s Q3 2022 earnings call held November 3, 2022, Garcia
Junior told analysts:
When we think about trying to aim for more profitable sales, what does that
mean? There’s certainly variability in the kind of gross profit associated with
different types of sales. . . .
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 49 of 332
- 46 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Defendants had not obscured these expenses, investors would have known that the limitless
growth story was a farce as the operations expenses skyrocketed with the scheme.
D.
Artifice No. 4: Defendants Flout Title and Registration Laws
144. Because complying with title and registration laws slowed sales growth,
Defendants flouted these laws, pushing through retail sales to inflate Carvana’s stock.
Indeed, as a June 24, 2022 Barron’s article, entitled “Carvana Sought to Disrupt Auto Sales.
It Delivered Undriveable Cars” (the “Barron’s exposé”) later revealed, in Carvana’s “haste
to seize market share from competitors” and report additional units sold, Carvana sold cars to
customers before it held title to those cars and before it could get them registered to their
new owners.
145. Unbeknownst to investors, this practice was systematic and nationwide. For
example, CW-3 (Phoenix, Arizona) estimated that Carvana only had around half of the titles
for the vehicles it sold wholesale at the time of sale. Further, CW-3 reported that, after the
Class Period, a “fire started” and he/she started receiving 10 times the number of calls per
week regarding title issues than before the Class Period. Corroborating CW-3’s account,
CW-9 (Tempe, Arizona) said the Customer Advocates on his/her team handled calls from all
over the country. Registration delays, which primarily meant that a title had not been
provided to the customer, amounted to roughly half of the calls the team had handled before
the team began handling both pre- and post-sale calls. CW-9 said that he/she noticed an
increase in the number of calls from customers experiencing title delays beginning in Q1
2022. CW-10 (Tempe, Arizona) described instances where titles on vehicles Carvana sold
could not be obtained for over a year, which meant the buyers were unable to drive their cars.
CW-8 (Corporate Headquarters) said that Carvana’s practice was simply to assume that titles
for vehicles could be acquired later if they were not available at the time of purchase. CW-
11 (Southeastern U.S.) reported that there was a drawer of documentation for vehicles that
had been sold by Carvana but which could not be registered since the titles were missing.
. . . There’s also kind of variation in the underlying costs of completing a
sale.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 50 of 332
- 47 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
CW-11 also described reviewing a Google spreadsheet of the vehicles for which registration
could not be completed since the titles were missing. Similarly, CW-4 (Midwestern U.S.)
said there was a Google spreadsheet that tracked vehicles without titles for the entire IRC,
and “there were a ridiculous amount of cars on the list.” According to CW-4, it was well
known internally that title issues were a common problem.
146. Worse still, because Carvana could not register the cars it sold absent title, the
Company issued countless temporary plates, sometimes as many as ten times to a single
customer. And these temporary plates were often issued by states with more liberal
registration requirements than the state in which the car was purchased, in violation of state
laws and regulations. For example, CW-11 explained that the issuance of these temporary
plates was a common practice at Carvana. Corroborating the confidential witness accounts,
an October 22, 2021 The Wall Street Journal article, entitled “Carvana Faces Government
Scrutiny and Fines Following Consumer Complaints” (the “WSJ exposé”) revealed that
Carvana “provided [a Michigan buyer] with temporary registrations from Georgia,
Tennessee and Arizona in lieu of registering his car in Michigan.” Similarly, the Barron’s
exposé revealed that “[i]nterviews with state officials and former Carvana employees show
the [registration delay] issue is wide[]-reaching,” and described instances where Carvana
had sent at least ten different license plates from different states to a customer.
147. Defendants circumventing title and registration laws and regulations drastically
increased Carvana’s SG&A expenses. For example, in May 2022, Carvana belatedly
admitted that it relied on costly third parties for “a significant portion of [its] title and
registration processing.” In addition, unbeknownst to investors, Carvana had to: (i) create a
new department to deal with the title and registration problems in early 2021; (ii) hire even
more new staffers, or “paperwork specialists,” by fall 2021; and (iii) create the “undriveable-
car task force” in 2022. Notably, unlike Carvana’s competitors, Carvana excluded title and
registration expenses from its Total and Retail GPU calculations. Instead, Garcia Junior and
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 51 of 332
- 48 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Jenkins buried these material costs in Carvana’s “Other” SG&A category, where the costs
were unquantified.14
148. Worse, Defendants’ undisclosed decision to boost sales by sidestepping title
and registration laws and regulations exposed Carvana to devastating financial, regulatory,
legal, and reputational loss. Indeed, as set forth below, multiple states revoked or suspended
Carvana’s dealer license:
(a)
North Carolina. Following an investigation and hearing on July 21,
2021, Carvana entered into a settlement agreement with the North Carolina Division of
Motor Vehicles. Pursuant to the settlement, North Carolina suspended Carvana’s dealer
license for six months because Carvana failed to deliver titles, sold motor vehicles without a
state inspection, and illegally issued out-of-state temporary tags and plates for vehicles sold
to customers in North Carolina.
(b)
Michigan. Unbeknownst to investors, by February 2021, the Michigan
Department of State (the “MDOS”) was investigating Carvana and met with Company
executives on March 23, 2021. On May 7, 2021, unbeknownst to investors, Michigan fined
Carvana thousands of dollars and placed Carvana on an 18-month probation. Carvana
repeatedly violated the terms of its probation agreement. On February 7, 2022, following
additional meetings between Carvana and MDOS, Carvana “admi[tted] . . . several more
violations of the Code,” paid thousands of dollars in additional fines, and agreed to extend its
probation. Nevertheless, Carvana continued violating title and registration laws.
Accordingly, on October 7, 2022, Michigan suspended Carvana’s license for: (i) “failing to
make application for title and registration within 15 days of delivery for 112 customers since
agreeing to an earlier probation extension”; (ii) “committing fraudulent acts in connection
with selling or otherwise dealing in vehicles where Carvana employees admitted to
destroying title applications and all applicable documents pertaining to the sale of [certain]
14 Defendants broke out “Other” SG&A for their first time in May 2022, revealing that
transaction expenses, which primarily consisted of title, registration, and related expenses,
totaled $410 dollars per unit in FY 2021.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 52 of 332
- 49 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
vehicles” that it later took back from customers; (iii) “improperly issuing temporary
registrations”; and (iv) “violating [the] terms of a probation agreement 127 times.” In
response, Carvana sued Michigan’s Secretary of State, arguing that it was not given a
hearing on the license matter, and that “‘the damage to Carvana’s reputation and goodwill
posed by the suspension order is incalculable and irreparable.’” Thereafter, on January 11,
2023, Carvana paid thousands of additional dollars in fines and signed a plea deal with
Michigan, in which Carvana “‘agreed it had violated the law and to have its dealer license
revoked and be barred from reapplying for a new license for three years.’”
(c)
Illinois. In February 2022, unbeknownst to investors, Illinois began
investigating Carvana for issuing out-of-state temporary registration permits and for failing
to transfer titles in a timely manner as required by Illinois’s regulations. On May 12, 2022,
Illinois suspended Carvana’s dealer license. According to the Illinois Secretary of State
Office, Carvana was failing to timely transfer title and improperly supplying customers with
out-of-state temporary plates, causing some consumers to receive police citations. Illinois
officials initially lifted the suspension. As reported by Barron’s mere months later, however,
Illinois reinstated the suspension because “‘Carvana continued to conduct business in a
manner that violates Illinois state law’ . . . . by again issuing out-of-state temporary plates
to Illinois customers, which is illegal in Illinois, and by missing required deadlines to
process title and registration paperwork.” Defendants, nevertheless, continued to fight the
suspension in court and were granted a restrictive temporary restraining order. Garcia Junior
referred to the suspensions as a “miscommunication” to inquiring analysts on August 4, 2022
during the Q2 2022 earnings call. But, in January 2023, Carvana entered into a settlement
with Illinois’s Secretary of State, in which Carvana admitted to violating Illinois law,
agreed to abide by new restrictions and increased oversight, and surrendered its $250,000
bond. Moreover, the settlement agreement allows Illinois summarily to suspend and revoke
Carvana’s dealership license once again if it fails to comply with either the agreement or the
law.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 53 of 332
- 50 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
(d)
Arizona. As first reported in the Barron’s exposé on June 24, 2022, the
Arizona Department of Transportation (the “ADOT”) canceled a program that allowed
Carvana to issue temporary plates for sales that occurred in other states after receiving more
than 80 complaints around “‘such issues as missing, delayed, and altered title and
registration documents, contract discrepancies, [and] significant odometer discrepancies.’”
Further, while ADOT’s spokesperson declined to provide further details due to open
investigations into Carvana’s registration practices, he confirmed that while “‘ADOT tries to
accommodate businesses like Carvana with their unique online business model as best we
can . . . . for consumer protection, we need to make sure that all car dealers operating in
Arizona are following state laws.’”
(e)
Pennsylvania. As a local Pennsylvania media outlet revealed mere days
before the end of the Class Period, Pennsylvania had suspended the dealer license of one of
Carvana’s dealership locations in Pennsylvania. Initially, Carvana denied the reports until
Pennsylvania officials provided the suspension letters to the media. It was later revealed that
Pennsylvania had suspended both of Carvana’s locations in Philadelphia for title and
registration issues.
(f)
Florida. As first reported in the WSJ exposé, unbeknownst to investors,
by October 22, 2021, Carvana had already settled a complaint with the Florida DMV for
failing to provide title paperwork for a number of customers for up to eight months. In
December 2021, Florida’s Department of Highway Safety and Motor Vehicles threatened the
Company with a statewide license suspension over Carvana’s failure to timely register cars,
identifying hundreds of vehicles with issues. While Florida withdrew that threat in February
after the Company had demonstrated progress in registering the hundreds of delinquent
vehicles, by August 2022, media outlets reported that Florida had filed two additional
complaints identifying numerous cases in which consumers waited more than 100 days and,
in one case, 253 days, to get the titles for their cars.
(g)
Texas. As first reported in the WSJ exposé, by October 2021, Texas
had fined Carvana thousands of dollars for violations related to title and registration.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 54 of 332
- 51 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
(h)
Maryland. On October 11, 2022, media outlets reported that Maryland
had joined the mounting number of states fining Carvana. Further, according to the
Maryland Motor Vehicle Administration, Carvana failed to timely register approximately
10% of its sales from June 2021 to July 2022.
(i)
Ohio. On November 22, 2022, Fox 28 reported that Ohio had
suspended Carvana’s privileges to issue temporary tags nearly two years earlier.
Specifically, the Ohio Bureau of Motor Vehicles (“BMV”) told Fox28:
“In December of 2020, the Ohio BMV became aware of an inordinately high
number of recent temporary tag issuances by Carvana. After investigating the
matter, it was determined that the issuances did not comply with Ohio law.
The Ohio BMV subsequently suspended temporary tag issuance privileges for
all applicable Carvana locations in Ohio.”
Although the Ohio BMV has since reinstated Carvana’s privileges “‘[a]fter a number of
meetings with Carvana representatives and written assurance that prior practices have been
halted, . . . . [t]he Ohio BMV continues to monitor Carvana.’”15
149. Garcia Junior and Jenkins were essential to this fraudulent conduct, discussing
it in countless Board meetings since at least July 2020. See, e.g., Exs. 1-2. And, critically,
once the truth about Carvana’s widespread title and registration violations began to leak out,
they minimized and downplayed the violations, misleading investors. See ¶¶148(c), 152,
180-181. Their misstatements were intended to conceal the scheme and maintain the
artificial inflation in Carvana’s stock.
E.
Artifice No. 5: Garcia Junior and Jenkins Make Materially
Misleading Statements and Omissions
150. In addition to the above methods and means, Garcia Junior and Jenkins also
executed the pump-and-dump scheme through a series of materially false and misleading
public statements and omissions, concerning the following subjects: (i) buying cars from
customers; (ii) title and registration; (iii) average days to sale metric; (iv) gross profit earned
15 Over Defendants’ vehement objections, on September 30, 2022, U.S. District Judge
Edward G. Smith denied defendants’ motion to dismiss a January 13, 2022 class action
consumer complaint alleging that Carvana violated common law and the Pennsylvania
Unfair and Deceptive Trade Practices Act by failing to timely transfer title.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 55 of 332
- 52 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
per retail vehicle sold (“Retail GPU”); and (v) Carvana’s “scalable” business model. Garcia
Junior and Jenkins’s public statements helped facilitate the fraudulent scheme and course of
business by concealing the scheme and convincing investors that Carvana’s growth was
sustainable.
1.
Garcia Junior and Jenkins Make False and Misleading
Statements to Further Conceal the Fraudulent Scheme
151. First, Garcia Senior and Jenkins misleadingly assured investors that Carvana
“assess[es] vehicles on the basis of quality” when, in fact, they had substantially lowered
their purchasing and verification standards and were “intentionally buying lower quality
cars.” See ¶¶191-192, 195-196, 199-200, 203-204, 207-210. In addition, they publicly
touted the benefits of buying cars from customers (the good), while omitting that buying cars
from customers was ravaging Carvana’s bottom line by: (i) flooding Carvana with wholesale
cars that had to be sold at a loss; and (ii) crippling its logistics network (the bad). See ¶¶193-
194, 197-198, 201-202, 205-206.
152. Second, Garcia Junior and Jenkins signed Forms 10-K and 10-Q purporting to
warn investors of risks and harms, such as “administrative, civil, or criminal penalties,” –
that “could result” if Carvana failed to comply with state title and registration laws and
regulations when those risks had already come to fruition. See ¶¶174-179, 182-187. Further,
when news regarding Defendants’ title and registration violations began to leak out, Garcia
Junior and Jenkins provided a misleading impression of the violations when they assured
investors that Carvana’s violations were mere “miscommunication[s]” and “North Carolina
specific.” See ¶¶148(c), 180-181.
153. Third, Garcia Junior and Jenkins made misleading statements and omissions
regarding Carvana’s average days to sale metric to conceal the negative consequences of
their scheme (growing inventory) and prop up Carvana’s stock price long enough for Garcia
Senior and Jenkins to cash out at top dollar. Average days to sale was one of Carvana’s six
“Key Operating Metrics” since becoming a public company. As Defendants emphasized,
Carvana’s “business is dependent upon our ability to expeditiously sell inventory.” Simply
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 56 of 332
- 53 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
put, an “increase [in] our average days to sale” served as a warning to investors that Carvana
was failing to rapidly sell its inventory, which “could have a material adverse effect” on
Carvana’s margins because of depreciation. Although Defendants continued to view average
days to sale as a key metric during and after the Class Period, Defendants suddenly stopped
publicly reporting the metric to investors. Instead, Garcia Junior and Jenkins replaced it with
a metric showing the number of IRCs (which has nothing to do with average days to sale).
Defendants repeatedly claimed that, somehow, Carvana’s “number of IRCs is a more
important metric than average days to sale due to the impact of IRC capacity on retail units
sold and the relative stability of average days to sale over the past three years.” ¶¶238, 240,
242. In truth, by the second half of the Class Period, Carvana was facing a glut of aging and
rapidly depreciating inventory that was spiking the average days to sale metric far beyond
historical norms as a result of Defendants’ fraudulent scheme. ¶¶239, 241, 243. Thus,
Garcia Junior and Jenkins’s abrupt decision to hide this metric from the public ensured that
investors were not alerted to the scheme.
2.
Garcia Junior and Jenkins Make Their Manufactured
Growth Appear Sustainable
154. Critical to the success of Defendants’ fraudulent scheme was their fiction of
sustainable hyper growth based on a scalable model where Carvana claimed: “[O]ur
business gets better as it gets bigger.” After all, if Carvana could pair its explosive unit
growth with purported profitability and a scalable logistics model that could cost effectively
expand across the country, the sky was the limit. At bottom, to secure a high valuation so
Garcia Senior and Jenkins could cash out at top dollar, Garcia Junior and Jenkins made the
following misleading statements and omissions to persuade investors that Carvana’s inflated
retail sales growth was sustainable.
(a)
First, throughout the Class Period, Garcia Junior and Jenkins touted
Carvana’s retail sales growth while simultaneously touting positive Retail GPU in all of
Carvana’s investor presentations, earnings calls, and shareholder letters. ¶¶247, 249, 251,
253, 255, 257, 259. For example, the February 24, 2021 Q4 2021 letter to shareholders (the
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 57 of 332
- 54 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
“Q4 2021 Shareholder Letter”) stated “Retail GPU was $1,495,” an increase of almost 20%
over the prior year, “driven by an increase in the percentage of retail units sourced from
customers.” Investors viewed Retail GPU as a proxy for Carvana’s retail “unit economics”
(i.e., profitability per-vehicle). However, as alleged herein, Defendants concealed Carvana’s
actual per-vehicle profitability – or lack thereof. ¶¶248, 250, 252, 254, 256, 258, 260. They
did so by: (i) excluding certain per-vehicle operations expenses from its calculation of Retail
GPU; and (ii) not adequately disclosing or quantifying these excluded costs separately so
investors could decipher overall profitability of retail sales on their own. These excluded
operations expenses included material “costs associated with completing retail sales,” such
as the full cost to ship cars to customers and title and registration costs. Defendants’
concealment of “operations expenses” when describing Carvana’s retail profitability was
especially misleading in light of the fact that Defendants internally viewed “the underlying
costs of completing a sale” as a critical component of Carvana’s actual retail vehicle
profitability. Indeed, as Garcia Junior later acknowledged:
When we think about trying to aim for more profitable sales, what does that
mean? There’s certainly variability in the kind of gross profit associated with
different types of sales. . . .
And then I think there’s a number of other dynamics kind of across car
type, et cetera. There’s also kind of variation in the underlying costs of
completing a sale.
155. By intentionally obscuring these significant operations expenses, investors
were left to rely on the positive Retail GPU reported by Defendants as the sole measure of
Carvana’s per-vehicle retail profitability. Critically, had Carvana included these operations
expenses in its Retail GPU calculation or separately disclosed and quantified the costs that
were excluded from Retail GPU, it would have revealed to investors that, on average,
Carvana generated negative unit economics on every retail car sold during the Class
Period. ¶¶248, 250, 252, 254, 256, 258, 260. For example, in Q4 2021, rather than the
$1,495 of positive Retail GPU reported by Defendants, Carvana actually generated negative
unit economics of $1,248 per retail unit sold. This created an impression of a state affairs
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 58 of 332
- 55 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
(positive per-unit profitability) that was materially different from the one that actually
existed (negative per-unit profitability). See, e.g., ¶260(a).
156. Second, Garcia Junior and Jenkins exalted Carvana’s purportedly “capital-
light” expansion model, and “logistics capabilities” in Carvana’s 2020 and 2021 10-Ks.
¶¶224, 234. For example, the 2021 10-K touted that Carvana has a “lower variable cost
structure at scale [as] . . . . [w]e do not require a network of brick-and-mortar dealerships . . .
instead, we utilize . . . an in-house logistics network.” They did so to persuade investors that
“[a]s we grow – by expanding geographically . . . – our business becomes stronger[,] . . . .
[g]rowth . . . enables economies of scale, . . . [and] these dynamics create a flywheel that
drives even more growth.” As discussed herein, Carvana neither had a capital-light
expansion model nor groundbreaking logistics. ¶¶225, 235.
157. When Carvana announced that it was “named to the Fortune 500 list for the
first time, becoming one of the four fastest companies to ever make the list with organic
growth, along with Amazon,” Garcia Junior and Jenkins stated that this purportedly
demonstrated that Carvana’s business “model has uniquely powerful unit economics and
scalability.” Likewise, in a “paid” interview/advertisement between Garcia Junior and
Fortune, Defendants touted “rapid ascension [to the Fortune 500] marks the fastest organic
growth of any automotive retailer in U.S. history and is a testament to the strength of its . . .
vertical integration, and innovative technology. Carvana boasts its own efficient internal
logistics network that allows the company to purchase, inspect, and house its inventory.”16
158. To be sure, Garcia Junior and Jenkins’ public statements facilitated
Defendants’ fraudulent scheme and course of business by deceiving investors regarding
Carvana’s purportedly disruptive business model and initiatives, persuading investors that
Carvana was demonstrating growth and profitability, reassuring investors that Carvana could
continue to do so because of its disruptive business model, and concealing the consequences
of Defendants’ scheme.
16 Fortune made clear that its “editorial staff was not involved in its creation or production”
of the paid article.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 59 of 332
- 56 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
159. Defendants’ fraudulent scheme worked. The market rewarded Defendants for
their purported “scalable” and sustainable growth, allowing Carvana’s stock to reach its
record-high price of $376 per share. In June 2021, Fortune noted: “How did Carvana make it
onto the Fortune 500? . . . The business is a hub-and-spoke setup . . . . [I]t turns out, the
business is built on a foundation of legitimate scalability . . . .”
THE DUMP
F.
Artifice No. 6: Garcia Senior and Jenkins Manipulate Their
10b5-1 Plans to Capitalize on the Scheme
160. After pumping up Carvana’s stock price by inflating retail sales and
profitability, Garcia Senior and Jenkins took action to enjoy the fruits of the Defendants’
fraudulent labor.17 As such, during the Class Period, Garcia Senior – while in possession of
MNPI – entered into a 10b5-1 plan on June 15, 2020. On November 4, 2020 – just four days
into trading and mere months into the new plan – Garcia Senior modified his trading plan
while in possession of MNPI to accelerate the rate per day at which he could sell his shares
to unsuspecting investors. Indeed, on November 4, 2020, Garcia Senior: (i) controlled
Carvana; (ii) held approximately 84% of Carvana’s voting power; (iii) was engaging in
hundreds of millions of dollars in related-party deals that were material to both his privately-
held companies and Carvana; and (iv) was participating in and aware of the fraudulent
scheme.
161. Worse, on May 7, 2021 – less than six months after his first modification of his
trading plan and weeks after Carvana was placed on probation in Michigan after admitting to
violating state title and register laws (unbeknownst to investors), Carvana’s shares continued
to climb to record highs. Once again, Garcia Senior modified his trading plan to accelerate
the number of shares he could sell per day while in possession of MNPI.
162. Professor Taylor of the Wharton School of Business noted: “‘I’ve studied
20,000 10b5-1 plans [and] I can’t recall another of this size where there are modifications
17 Garcia Junior, who is a beneficiary of his father’s trust, benefited by Garcia Senior’s
insider trading.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 60 of 332
- 57 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
every six months.’”18 Unsurprisingly, he concluded: “‘The Garcias knew it was short-lived
. . . . The Garcias knew the music would eventually end.’”
G.
Artifice No. 7: Garcia Senior and Jenkins Dump Their Holdings
163. Before the truth regarding Defendants’ so-called “sustainable” growth was
revealed, Jenkins and “‘the Garcias kn[owing] it was short-lived’” and that “‘the music
would eventually end’” sold nearly 14.3 million shares of their personal holdings of Carvana
stock for proceeds of nearly $3.76 billion before the bottom fell out.
164. Jenkins’ and Garcia Senior’s suspiciously timed sales, made while each
possessed MNPI, are summarized in the chart below, in which the vertical axis on the left
shows Carvana’s stock price and the vertical axis on the right depicts the dollar amount of
stock sold. A detailed description of the insider trading is set forth in Appendix A, attached
hereto.
18 Ben Foldy, CEO’s Father Gets a $3.6 Billion Stock Windfall at Carvana, Wall St. J.
(Sept. 17, 2021) (the “September 17, 2021 The Wall Street Journal article”).
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 61 of 332
- 58 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
165. Simply put, as Professor Taylor concluded in a June 2022 The Wall Street
Journal article: “‘[T]he actions of [Carvana’s] executives signal to me that the executives
knew what was coming, and consequently took advantage of the lofty valuations to . . . sell
their own equity . . . .’”
H.
The Truth Is Revealed
166. After Defendants sold nearly $3.76 billion shares at artificially inflated prices,
Defendants’ fraudulent scheme unraveled over several months during which Carvana
acknowledged in a series of corrective disclosures that bigger was not better. See §IX. In
truth, Carvana’s growth model was not “scalable.” Its expansion model was not “capital-
light.” It had not seen “economies of scale” or “uniquely powerful unit economics” as it
grew. And, its logistics model was neither efficient nor groundbreaking. Indeed, as Carvana
fell apart, the Company would be forced to do an about-face and change its entire ethos of
being the largest auto retailer. Now, for the first time, retail unit sales growth was not the
Company’s priority.
167. For example, because Carvana needed critical infrastructure to support its new
disparate markets spanning the country and to stem the increased costs and network
complexity from the “wholesale [vehicles] acquired from customers,” Carvana disclosed,
over a few days, that it would be forced to take on more than $3.4 billion in high interest rate
debt to purchase ADESA to “improve our logistics network,” by placing the Company
“within 200 miles of 94%” of the population, which “will have the benefit of reducing
shipping distances, times, and costs.” Carvana also announced disappointing earnings that
were caused by Defendants’ fraudulent scheme and course of conduct, including that: (i)
retail sales growth was hampered that quarter by significant “logistics network constraints,”
reconditioning expenses, and “higher wholesale volume from buying more cars from
customers”; and (ii) Carvana no longer expected to sell 550,000 retail cars in 2022 as
previously stated.
168. Mere weeks later, as a result of Defendants’ fraudulent scheme, Carvana
disclosed that it was laying off 2,500 employees, transitioning operations away from some
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 62 of 332
- 59 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
logistical hubs and an IRC “in connection with these right-sizing initiatives,” and that it
would release an updated operating plan later that week. Carvana’s stock plunged as
investors took notice and awaited the new operating plan. For example, an article published
by Forbes on May 21, 2022 reported that Carvana’s disclosure “was a sign of much bigger
problems at the company, according to 10 former employees . . . and several industry
analysts. They describe a spendthrift business, whose growth-at-all-costs mentality
undermined business operations” and led to its current poor condition.
169. Then, on November 3, 2022, Defendants issued disappointing Q3 2022
financial results that were caused by Defendants’ fraudulent scheme and course of conduct
and revealed, in part, the unraveling of their scheme. Indeed, Defendants revealed that
Carvana suffered an 8% decline in retail unit sales from the prior year. They further revealed
that their singular focus would be on slashing expenses, eliminating “less profitable sales,”
“reduc[ing] advertising” aimed at acquiring “less profitable sales” in its new markets, and
“take other actions to improve profitability, such as increasing long-distance shipping fees.”
As an analyst at Needham & Company noted, “‘[t]he path forward for Carvana is to sell as
many cars as possible, but to do so on a profitable basis, versus prior it was more about
selling as many cars as possible.’” A February 2, 2023 Motley Fool article entitled, “3 Signs
Carvana Hit the Wall” (the “Hit the Wall Article”), told investors to “[s]tep back and think
about that [disclosure] for a moment” because “[i]t means that the company had historically
grown in these markets even though they weren’t all that attractive. It was growth for
growth’s sake and not necessarily growth to improve profitability.”
170. By the end of the Class Period, Defendants revealed the unraveling of their
fraudulent scheme and course of conduct. Indeed, Defendants disclosed that “retail units
sold decreased [23%] for the first time in our history this year,” and “we stepped back on the
key metrics of retail units sold” “for the first time.” In fact, Carvana revealed that it was
intentionally foregoing retail sales growth for the foreseeable future to instead focus on
driving “positive unit economics” (i.e., retail sales that were actually profitable). Moreover,
Carvana admitted gross profits had declined by 63% year-over-year. Further, Carvana
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 63 of 332
- 60 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
“significantly accelerated . . . reductions in inventory,” “significantly reduc[ed] retail vehicle
acquisitions,” and wrote off over $50 million of aged and impaired vehicle inventory that
was worth less than Carvana had paid to acquire it and prepare it for sale.
171. At the same time, over a series of months, the media and multiple states
revealed what Defendants knew all along – Carvana routinely violated state title and
registration laws and regulations. Critically, it was eventually revealed that Carvana’s
violations were not one-offs or miscommunications, as Defendants claimed, but a
widespread course of conduct to increase sales.
172. In sum, by the end of the Class Period, Carvana was merely a shell of the
limitless, e-commerce growth machine that Defendants’ fraudulent scheme and course of
conduct had portrayed. Instead, it needed to stop growing altogether and spend billions of
dollars to fix the Company. The unsustainable growth-at-all-costs scheme – which pumped
up Carvana’s stock price to historic highs and enabled Garcia Senior and Jenkins to pocket
billions of dollars in proceeds – proved very costly to investors. By the end of the Class
Period, Carvana’s stock price had declined by 98%, from its Class Period high, and it still
has not recovered.
VII. DEFENDANTS’ CLASS PERIOD FALSE AND MISLEADING
STATEMENTS AND OMISSIONS UNDER RULE 10b-5(b)
A.
Defendants’ Materially Misleading Statements and Omissions
Concerning Title and Registration
173. As required by the SEC, Defendants disclosed Carvana’s purported material
risks, and any changes thereto, in every Form 10-K and Form 10-Q. In each of these SEC
filings during the Class Period, Carvana purported to warn of material risks and related
harms that could occur if it failed to comply with state and local laws and regulations,
including those related to title and registration. While Defendants characterized the risks and
related harms as mere hypotheticals, they failed to disclose that, beginning at least by the
date on which Carvana filed its 2020 10-K, these risks and harms had already materialized
because states had begun investigating and penalizing Carvana for its title and registration
noncompliance. Worse yet, as these violations became known to investors, Defendants
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 64 of 332
- 61 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
continued to conceal the extent of Carvana’s title and registration issues, including then-
ongoing investigations and penalties – minimizing the regulatory investigations as mere one-
offs and trivializing the penalties imposed on the Company. But, as Defendants later
conceded, the reputational damage from just one of these suspensions was “‘incalculable and
irreparable.’”
174. Statement No. 1: On February 25, 2021, Carvana filed its Form 2020 10-K
(the “2020 10-K”) with the SEC, signed by Garcia Junior and Jenkins and attaching
Sarbanes-Oxley Act (“SOX”) certifications signed by Garcia Junior and Jenkins. The 2020
10-K contained the following purported risk disclosure:
We operate in several highly regulated industries and are subject to a
wide range of federal, state, and local laws and regulations. . . . [O]ur failure
to comply [with these laws and regulations], could have a material adverse
effect on our business, results of operations, and financial condition.
We are subject to a wide range of evolving federal, state, and local laws
and regulations, many of which may have limited to no interpretation
precedent as it relates to our business model. Our sale and purchase of used
vehicles and related activities, including the sale of complementary products
and services, are subject to state and local licensing requirements, state laws,
regulations, and systems and process requirements related to title and
registration . . . .
*
*
*
The violation of any of these laws or regulations could result in
administrative, civil, or criminal penalties or in a cease-and-desist order
against some or all of our business activities, any of which could damage
our reputation and have a material adverse effect on our business, sales, and
results of operations. Additionally, even an allegation that we violated these
laws, by regulators, competitors, individuals, or consumers, could result in
costly litigation with uncertain results.
175. As detailed below, Statement No. 1 by Carvana, Jenkins, and Garcia Junior
was materially misleading when made because Defendants omitted to state material facts
necessary in order to make the statement not misleading.
(a)
Defendants’ February 25, 2021 statement that the violation of any state
regulations and laws “could result in administrative, civil, or criminal penalties or in a cease-
and-desist order against some or all of our business activities, any of which could damage
our reputation and have a material adverse effect on our business, sales, and results of
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 65 of 332
- 62 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
operations,” was misleading because it omitted that the risks it was warning investors about
had already come to fruition. In fact: (i) the Ohio BMV had suspended Carvana’s temporary
tag issuance privileges for all Ohio locations and subjected the Company to increased
oversight in December 2020; and (ii) by February 2021, the MDOS was investigating
Carvana and, shortly thereafter, would fine Carvana and place it on an 18-month probation.
Defendants also concealed the inevitably of the foregoing risks and identified harms
materializing.
(b)
Defendants presented Carvana’s “failure to comply” with title and
registration laws and others’ “allegation[s] that we violated these laws,” as hypothetical risks
that could materially harm Carvana if they materialized when, in truth, these risks had
already come to fruition. Indeed, Carvana routinely and repeatedly: (i) sold cars before
holding title to the vehicles; (ii) failed to register cars within the legally required timeframe;
and (iii) issued out-of-state temporary tags and/or license plates in violation of numerous
state laws. §VI.D. For example, CWs-3, 4, 8, 9, 10, and 11 universally reported that
Carvana would sell these vehicles before it received title. Specifically, CW-3 estimated that
Carvana only had around half of the titles for the vehicles it sold wholesale at the time of
sale. Further, he/she noted that, after the Class Period began, a “fire started” because he/she
was flooded with calls regarding title issues. CW-11 reported that there was a drawer of
documentation for vehicles that had been sold by Carvana, but which Carvana could not
register since the titles were missing. CW-4 said it was well known internally that title
issues were a problem. CW-4 also said there was a Google spreadsheet that tracked vehicles
without titles for his/her entire IRC and there were “a ridiculous amount of cars on the list.”
Further, the Barron’s exposé revealed that: (i) “[i]nterviews with state officials and former
Carvana employees show the [registration delay] issue is wide[]-reaching”; and (ii) Carvana
admitted the violations had primarily occurred in 2020 and 2021. In addition, a presentation
from a July 2020 Board meeting that Jenkins and Garcia Junior attended reveals that Carvana
had undertaken an initiative “to reduce registration delays,” as part of its “Key Focus Areas
and Risks.” Ex. 1. Worse, title and registration issues and penalties continued to be a
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 66 of 332
- 63 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
prominent topic in every Board meeting thereafter through at least December 2021. See,
e.g., Exs. 1-2; see also Schertz v. Garcia II, No. 2023-0600-KSJM, Exhibits 10-23 to
Transmittal Affidavit of R. Garrett Rice in Connection with Defendants’ Opening Brief in
Support of Their Motion to Dismiss or Stay Plaintiff’s Verified Amended Stockholder
Derivative Complaint (Del. Ch. Feb. 26, 2024) (ECF 23). Lastly, Defendants also concealed
the inevitably of the foregoing risks and identified harms materializing because of the above
omitted facts.
(c)
A reasonable investor would have viewed the above omitted facts as
important for the reasons stated herein. Indeed, Defendants admitted in an October 2022
court filing that “‘the damage to Carvana’s reputation and goodwill posed by’” even one
“‘suspension order [wa]s incalculable and irreparable.’”
176. Statement No. 2: On May 6, 2021, Carvana filed its quarterly report on Form
10-Q for the period ended March 31, 2021 (“Q1 2021 10-Q”) with the SEC, signed by
Jenkins and attaching SOX certifications signed by Garcia Junior and Jenkins. The Q1 2021
10-Q stated: “There have been no material changes to the risk factors disclosed under the
heading ‘Risk Factors’ in our most recent Annual Report on Form 10-K, filed on
February 25, 2021.” The risk factors in the 2020 10-K, stated:
We operate in several highly regulated industries and are subject to a
wide range of federal, state, and local laws and regulations. . . . [O]ur failure
to comply [with these laws and regulations], could have a material adverse
effect on our business, results of operations, and financial condition.
We are subject to a wide range of evolving federal, state, and local laws
and regulations, many of which may have limited to no interpretation
precedent as it relates to our business model. Our sale and purchase of used
vehicles and related activities, including the sale of complementary products
and services, are subject to state and local licensing requirements, state laws,
regulations, and systems and process requirements related to title and
registration . . . .
*
*
*
The violation of any of these laws or regulations could result in
administrative, civil, or criminal penalties or in a cease-and-desist order
against some or all of our business activities, any of which could damage
our reputation and have a material adverse effect on our business, sales, and
results of operations. Additionally, even an allegation that we violated these
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 67 of 332
- 64 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
laws, by regulators, competitors, individuals, or consumers, could result in
costly litigation with uncertain results.
177. As detailed below, Statement No. 2 by Carvana, Jenkins, and Garcia Junior
was materially misleading when made because Defendants omitted to state material facts
necessary in order to make it not misleading.
(a)
Defendants’ May 6, 2021 statement that “[t]here have been no material
changes to the risk factors disclosed under the heading ‘Risk Factors’ in” the 2020 10-K, was
misleading because it omitted that the material risks previously warned of – “administrative,
civil, or criminal penalties or in a cease-and-desist order against some or all of our business
activities” – had already come to fruition and materially changed since February 25, 2021.
In fact: (i) the Ohio BMV had investigated Carvana, suspended its temporary tag issuance
privileges for all Ohio locations, and subjected the Company to increased oversight in
December 2020; and (ii) on March 23, 2021, MDOS had met with Carvana executives
following its investigation of the Company’s violations, and, on May 7, 2021, MDOS fined
Carvana thousands of dollars and placed the Company on an 18-month probation.
Defendants also concealed the inevitably of the foregoing risks and identified harms
materializing because of the above omitted facts.
(b)
Defendants presented Carvana’s “failure to comply” with title and
registration laws and others’ “allegation[s] that we violated these laws,” as hypothetical risks
that could materially harm Carvana if they materialized when, in truth, these risks had
already come to fruition. Indeed, Carvana routinely and repeatedly: (i) sold cars before
holding title to the vehicles; (ii) failed to register cars within the legally required timeframe;
and (iii) issued out-of-state temporary tags and/or license plates in violation of numerous
state laws. §VI.D. For example, CWs-3, 4, 8, 9, 10, and 11 universally reported that
Carvana would sell these vehicles before it received title. Specifically, CW-3 estimated that
Carvana only had around half of the titles for the vehicles it sold wholesale at the time of
sale. Further, he/she noted that, after the Class Period began, a “fire started” because he/she
was flooded with calls regarding title issues. CW-11 reported that there was a drawer of
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 68 of 332
- 65 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
documentation for vehicles that had been sold by Carvana, but which Carvana could not
register since the titles were missing. CW-8 said that Carvana’s practice was simply to
assume that titles for vehicles could be acquired later if they were not available at the time of
purchase. CW-4 said it was well known internally that title issues were a problem. CW-4
also said there was a Google spreadsheet that tracked vehicles without titles for his/her entire
IRC and there were “a ridiculous amount of cars on the list.” Further, the Barron’s exposé
revealed that: (i) “[i]nterviews with state officials and former Carvana employees show the
[registration delay] issue is wide[]-reaching”; and (ii) Carvana admitted the violations had
primarily occurred in 2020 and 2021. In addition, a presentation from a July 2020 Board
meeting that Jenkins and Garcia Junior attended reveals that Carvana had undertaken an
initiative “to reduce registration delays,” as part of its “Key Focus Areas and Risks.” Ex. 1.
Worse, title and registration issues and penalties continued to be a prominent topic in every
Board meeting thereafter until at least December 2021. See, e.g., Exs. 1-2; see also Schertz,
ECF 23. Lastly, Defendants also concealed the inevitably of the foregoing risks and
identified harms materializing because of the above omitted facts.
(c)
A reasonable investor would have viewed the above omitted facts as
important for the reasons stated herein. Indeed, Defendants admitted in an October 2022
court filing that “‘the damage to Carvana’s reputation and goodwill posed by’” even one
“‘suspension order [wa]s incalculable and irreparable.’”
178. Statement No. 3: On August 5, 2021, Carvana filed its Q2 2021 10-Q with the
SEC, signed by Jenkins and attaching SOX certifications signed by Garcia Junior and
Jenkins. The Q2 2021 10-Q stated: “There have been no material changes to the risk
factors disclosed under the heading ‘Risk Factors’ in our most recent Annual Report on
Form 10-K, filed on February 25, 2021.” The risk factors disclosed in the 2020 10-K,
stated:
We operate in several highly regulated industries and are subject to a
wide range of federal, state, and local laws and regulations. . . . [O]ur failure
to comply [with these laws and regulations], could have a material adverse
effect on our business, results of operations, and financial condition.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 69 of 332
- 66 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
We are subject to a wide range of evolving federal, state, and local laws
and regulations, many of which may have limited to no interpretation
precedent as it relates to our business model. Our sale and purchase of used
vehicles and related activities, including the sale of complementary products
and services, are subject to state and local licensing requirements, state laws,
regulations, and systems and process requirements related to title and
registration . . . .
*
*
*
The violation of any of these laws or regulations could result in
administrative, civil, or criminal penalties or in a cease-and-desist order
against some or all of our business activities, any of which could damage
our reputation and have a material adverse effect on our business, sales, and
results of operations. Additionally, even an allegation that we violated these
laws, by regulators, competitors, individuals, or consumers, could result in
costly litigation with uncertain results.
179. As detailed below, Statement No. 3 by Carvana, Jenkins, and Garcia Junior
was materially misleading when made because Defendants omitted to state material facts
necessary in order to make it not misleading.
(a)
Defendants’ August 5, 2021 statement that “[t]here have been no
material changes to the risk factors” identified in the 2020 10-K was misleading because it
omitted that the material risks warned of – “administrative, civil, or criminal penalties or in a
cease-and-desist order against some or all of our business activities” – had already come to
fruition and materially changed since February 25, 2021. In fact: (i) the Ohio BMV had
investigated Carvana, suspended its temporary tag issuance privileges for all Ohio locations,
and subjected the Company to increased oversight in December 2020; (ii) MDOS had fined
Carvana thousands of dollars and placed the Company on an 18-month probation – a
probation they would repeatedly violate – following months of investigation and a meeting
with Carvana executives in March of 2021; and (iii) North Carolina had fined and suspended
Carvana’s license for six months following an investigation and a July 2021 hearing. In
addition, Carvana had settled a complaint with Florida for title violations, and Texas had
fined Carvana thousands of dollars for violations related to title and registration no later than
October 22, 2021. Defendants also concealed the inevitably of the foregoing risks and
identified harms materializing because of the above omitted facts.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 70 of 332
- 67 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
(b)
Defendants presented Carvana’s “failure to comply” with title and
registration laws and others’ “allegation[s] that we violated these laws,” as hypothetical risks
that could materially harm Carvana if they materialized when, in truth, these risks had
already come to fruition. Indeed, Carvana routinely and repeatedly: (i) sold cars before
holding title to the vehicles; (ii) failed to register cars within the legally required timeframe;
and (iii) issued out-of-state temporary tags and/or license plates in violation of numerous
state laws. §VI.D. For example, CWs-3, 4, 8, 9, 10, and 11 universally reported that
Carvana would sell these vehicles before it received title. Specifically, CW-3 estimated that
Carvana only had around half of the titles for the vehicles it sold wholesale at the time of
sale. Further, he/she noted that, after the Class Period began, a “fire started” because he/she
was flooded with calls regarding title issues. CW-11 reported that there was a drawer of
documentation for vehicles that had been sold by Carvana, but which Carvana could not
register since the titles were missing. CW-8 said that Carvana’s practice was simply to
assume that titles for vehicles could be acquired later if they were not available at the time of
purchase. CW-4 said it was well known internally that title issues were a problem. CW-4
also said there was a Google spreadsheet that tracked vehicles without titles for his/her entire
IRC and there were “a ridiculous amount of cars on the list.” Further, the Barron’s exposé
revealed that: (i) “[i]nterviews with state officials and former Carvana employees show the
[registration delay] issue is wide[]-reaching”; and (ii) Carvana admitted the violations had
primarily occurred in 2020 and 2021. In addition, a presentation from a July 2020 Board
meeting that Jenkins and Garcia Junior attended reveals that Carvana had undertaken an
initiative “to reduce registration delays,” as part of its “Key Focus Areas and Risks.” Ex. 1.
Worse, title and registration issues and penalties continued to be a prominent topic in every
Board meeting thereafter until at least December 2021. See, e.g., Exs. 1-2; see also Schertz,
ECF 23. Further, from June 2021 to July 2022, Carvana failed to timely register
approximately 10% of its sales in Maryland, which would later result in additional fines.
Defendants also concealed the inevitably of the foregoing risks and identified harms
materializing because of the above omitted facts.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 71 of 332
- 68 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
(c)
A reasonable investor would have viewed the above omitted facts as
important for the reasons stated herein. Indeed, Defendants admitted that, should the risks
occur, the effects could be material. Further, Defendants admitted in an October 2022 court
filing that “‘the damage to Carvana’s reputation and goodwill posed by’” even one
“‘suspension order [wa]s incalculable and irreparable.’”
180. Statement No. 4: On August 11, 2021, Carvana was asked about the title and
licensing issue in North Carolina at the J.P. Morgan Auto Conference presentation attended
by investors and analysts. Carvana’s VP of IR, Mike Levin (“Levin”) and Jenkins gave the
following responses:
[J.P. Morgan Analyst:] Great. Thanks. So maybe let’s just get right
into it. I wasn’t expecting to ask this question, but since the news hit the
headlines around the North Carolina DMV in the suspension – in the Raleigh
region.
Just curious to know, any update you could provide in up front, like
what exactly happened. What’s going on? Like is it just restricted to certain
ZIP codes? Is it the whole state? Can you buy cars from consumers? Just if
you could give us a quick rundown, just start would be helpful.
[Jenkins:] Yes, sure. So the brief summary there is following COVID,
I think there are a number of things that led to title and registration getting
backed up. And I think since some of the delays associated with COVID,
we’ve continued to be somewhat backed up in title and registration, and that’s
impacted the customer experience on title and registration in certain instances.
In this particular case, I think we had quite a small fraction of
customers that were impacted by title and registration delays, but it did
happen in the state of North Carolina. And so I think the DMV there thought
that with having a set of customers experienced title and registration delays
that they wanted to do something to reflect the fact that they didn’t like
customers in that state experiencing these delays.
So what they did was they asked us to stop delivering cars from the
vending machine in Raleigh. We can still deliver cars in the metro area of
Raleigh, just not from the specific vending machine location. And we can also
still do work in the vending team, but we just can’t deliver cars from that
particular location. And so that’s what they elected to do. Based on our
research, that’s a relatively sort of unusual approach for something like this
and having a set of customers have title and registration delays, but that’s
where we are today. And yes, we announced it, and that’s the background.
[J.P. Morgan Analyst:] Got it. And are you comfortable that this is a
very North Carolina specific issue and not something you might need to
investigate in like other states, just to make sure it’s covered? Or just curious
like has this led you to like maybe like just check what’s happening and make
sure that this is not more of a widespread issue?
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 72 of 332
- 69 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
[Jenkins:] Sure. Well, I would make 2 points on that. One, our
understanding is that this is quite unprecedented. But having said that, our
goal is to provide the best customer experiences possible and make sure that
we’re – our processes are dialed . . . .
*
*
*
[Levin:] Yes. Just I think that there is always a lot of room for us to
improve, and the #1 focus here is on customer experience, and we work with
DMVs in every single state around the country to achieve that goal. And I
think this, based on the pandemic, what was an area where we saw delays.
And this was a relatively unusual action, but is also pretty small in scope,
relatively speaking.
181. As detailed below, Statement No. 4 by Carvana, Levin, and Jenkins omitted
material facts that affirmatively created an impression of a state of affairs that differed in a
material way from the one that actually existed.
(a)
Contrary to Defendants’ statement that North Carolina’s action was
“unusual,” “unprecedented,” and “specific” to North Carolina: (i) the Ohio BMV had
investigated Carvana, suspended its temporary tag issuance privileges for all Ohio locations,
and subjected the Company to increased oversight in December 2020; and (ii) MDOS had
fined Carvana thousands of dollars and placed the Company on an 18-month probation on
May 7, 2021 after meeting with its executives in March of 2021. Further, the Board had
been discussing title and registration issues at every Board meeting for over a year. See, e.g.,
Exs. 1-2; see also Schertz, ECF 23.
(b)
Contrary to the impression that the violations and penalties were a
“North Carolina specific issue,” the Barron’s exposé later revealed that “[i]nterviews with
state officials and former Carvana employees show the [registration delay] issue is wide[]-
reaching.” §VI.D. CWs-3, 4, 8, 9, 10, and 11 universally reported that title issues were
pervasive at the Company. Specifically, CW-3 estimated that Carvana only had around half
of the titles for the vehicles it sold wholesale at the time of sale. Further, he/she noted that,
after the Class Period began, a “fire started” because he/she was flooded with calls regarding
title issues. CW-11 reported that there was a drawer of documentation for vehicles that had
been sold by Carvana, but which Carvana could not register since the titles were missing.
CW-8 said that Carvana’s practice was simply to assume that titles for vehicles could be
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 73 of 332
- 70 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
acquired later if they were not available at the time of purchase. CW-4 said it was well
known internally that title issues were a problem. CW-4 also said there was a Google
spreadsheet that tracked vehicles without titles for his/her entire IRC and there were “a
ridiculous amount of cars on the list.” Further, the Barron’s exposé revealed that Carvana
had admitted the violations had primarily occurred in 2020 and 2021. In addition, a
presentation from a July 2020 Board meeting that Jenkins and Garcia Junior attended reveals
that Carvana had undertaken an initiative “to reduce registration delays,” as part of its “Key
Focus Areas and Risks.” Ex. 1. Worse, title and registration issues and penalties continued
to be a prominent topic in every Board meeting thereafter until at least December 2021. See
Schertz, ECF 23. Indeed, the Company’s October 18, 2021 Board minutes reveal that
internally Garcia Junior and Jenkins were discussing how to “explain the NC situation,
amongst others.” Ex. 2. Further, as alleged herein, subsequent government investigations,
plea agreements, suspensions, and media reports confirm that, during the Class Period,
Defendants also violated laws and regulations in Texas, Florida, Michigan, Maryland, Ohio,
Oklahoma, Illinois, Pennsylvania, and Georgia.
(c)
Notably, analysts bought Defendants’ misleading assurances. For
example, a William Blair analyst noted on August 11, 2021, that “[a]fter speaking with
management, . . . we are optimistic that this will be an isolated incident.”
(d)
The omitted facts would have been viewed by a reasonable investor as
significant for the reasons stated herein, and because Defendants stated that, as to just one of
the state actions, “‘the damage to Carvana’s reputation and goodwill posed by the suspension
order is incalculable and irreparable.’”
182. Statement No. 5: On November 4, 2021, Carvana filed its quarterly report
Form 10-Q for the period ended September 30, 2021 (“Q3 2021 10-Q”), with the SEC,
signed by Jenkins and attaching SOX certifications signed by Garcia Junior and Jenkins.
The Q3 2021 10-Q stated: “There have been no material changes to the risk factors
disclosed under the heading ‘Risk Factors’ in our most recent Annual Report on Form 10-
K, filed on February 25, 2021.” The risk factors disclosed in the 2020 10-K, stated:
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 74 of 332
- 71 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
We operate in several highly regulated industries and are subject to a
wide range of federal, state, and local laws and regulations. . . . [O]ur failure
to comply [with these laws and regulations], could have a material adverse
effect on our business, results of operations, and financial condition.
We are subject to a wide range of evolving federal, state, and local laws
and regulations, many of which may have limited to no interpretation
precedent as it relates to our business model. Our sale and purchase of used
vehicles and related activities, including the sale of complementary products
and services, are subject to state and local licensing requirements, state laws,
regulations, and systems and process requirements related to title and
registration . . . .
*
*
*
The violation of any of these laws or regulations could result in
administrative, civil, or criminal penalties or in a cease-and-desist order
against some or all of our business activities, any of which could damage
our reputation and have a material adverse effect on our business, sales, and
results of operations. Additionally, even an allegation that we violated these
laws, by regulators, competitors, individuals, or consumers, could result in
costly litigation with uncertain results.
183. As detailed below, Statement No. 5 by Carvana, Garcia Junior, and Jenkins
was materially misleading when made because Defendants omitted to state material facts
necessary in order to make it not misleading.
(a)
Specifically, Defendants’ November 4, 2021 statement that “[t]here
have been no material changes to the risk factors disclosed under the heading ‘Risk Factors’”
identified in the 2020 10-K was misleading because it omitted that the material risks
previously warned of – “administrative, civil, or criminal penalties or in a cease-and-desist
order against some or all of our business activities” – had already come to fruition and
materially changed since February 25, 2021. In fact: (i) the Ohio BMV had investigated
Carvana, suspended its temporary tag issuance privileges for all Ohio locations, and
subjected the Company to increased oversight in December 2020; and (ii) Carvana was
repeatedly violating its probation in Michigan and would be forced to extend it in three
months after multiple meetings between Carvana and MDOS. At the same time, Defendants
had portrayed the North Carolina fine and suspension as “unprecedented” and “North
Carolina specific.” Defendants also concealed the inevitably of the foregoing risks and
identified harms materializing because of the above omitted facts.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 75 of 332
- 72 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
(b)
Defendants presented Carvana’s “failure to comply” with title and
registration laws and others’ “allegation[s] that we violated these laws,” as hypothetical risks
that could materially harm Carvana if they materialized when, in truth, these risks had
already come to fruition. Indeed, Carvana routinely and repeatedly: (i) sold cars before
holding title to the vehicles; (ii) failed to register cars within the legally required timeframe;
and (iii) issued out-of-state temporary tags and/or license plates in violation of numerous
state laws. §VI.D. For example, CWs-3, 4, 8, 9, 10, and 11 universally reported that
Carvana would sell these vehicles before they received title. Specifically, CW-10 described
instances where titles on vehicles Carvana sold could not be obtained for over a year, which
meant the buyers were unable to drive their cars. CW-11 reported that there was a drawer of
documentation for vehicles that had been sold by Carvana, but which Carvana could not
register since the titles were missing. CW-8 said that Carvana’s practice was simply to
assume that titles for vehicles could be acquired later if they were not available at the time of
purchase. CW-4 said it was well known internally that title issues were a problem. CW-4
also said there was a Google spreadsheet that tracked vehicles without titles for his/her entire
IRC and there were “a ridiculous amount of cars on the list.” As the Barron’s exposé later
revealed, “[i]nterviews with state officials and former Carvana employees show the
[registration delay] issue is wide[]-reaching.” Further, from June 2021 to July 2022, Carvana
failed to timely register approximately 10% of its sales in Maryland, which would later result
in additional fines. In addition, a presentation from a July 2020 Board meeting that Jenkins
and Garcia Junior attended reveals that Carvana had undertaken an initiative “to reduce
registration delays,” as part of its “Key Focus Areas and Risks.” Ex. 1. Worse, title and
registration issues and penalties continued to be a prominent topic in every Board meeting
thereafter until at least December 2021. See, e.g., Exs. 1-2; see also Schertz, ECF 23.
Defendants also concealed the inevitably of the foregoing risks and identified harms
materializing because of the above omitted facts.
(c)
A reasonable investor would have viewed the above omitted facts as
important for the reasons stated herein. Indeed, Defendants admitted that, should the risks
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 76 of 332
- 73 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
occur, the effects could be material. Further, Defendants admitted in an October 2022 court
filing that “‘the damage to Carvana’s reputation and goodwill posed by’” even one
“‘suspension order [wa]s incalculable and irreparable.’”
184. Statement No. 6: On February 24, 2022, Carvana filed its 2021 10-K with the
SEC, signed by Garcia Junior and Jenkins and attaching SOX certifications signed by Garcia
Junior and Jenkins. The 2021 10-K contained the following purported risk disclosure:
We operate in several highly regulated industries and are subject to a
wide range of federal, state, and local laws and regulations. . . . [O]ur
failure to comply [with these laws and regulations], could have a material
adverse effect on our business, results of operations, and financial
condition.
We are subject to a wide range of evolving federal, state, and local laws
and regulations, many of which may have limited to no interpretation
precedent as it relates to our business model. Our sale and purchase of used
vehicles and related activities, including the sale of complementary products
and services, are subject to state and local licensing requirements, state laws,
regulations, and systems and process requirements related to title and
registration . . . .
*
*
*
The violation of any of these laws or regulations could result in
administrative, civil, or criminal penalties or in a cease-and-desist order
against some or all of our business activities, any of which could damage
our reputation and have a material adverse effect on our business, sales, and
results of operations. Additionally, even an allegation that we violated these
laws, by regulators, competitors, individuals, or consumers, could result in
costly litigation with uncertain results.
185. As detailed below, Statement No. 6 by Carvana, Jenkins, and Garcia Junior
was materially misleading when made because Defendants omitted to state material facts
necessary in order to make the statement made not misleading.
(a)
Defendants’ February 24, 2022 statement that the violation of any state
regulations and laws “could result in administrative, civil, or criminal penalties or in a cease-
and-desist order against some or all of our business activities, any of which could damage
our reputation and have a material adverse effect on our business, sales, and results of
operations,” was misleading because it omitted that the risks it was warning investors about
had already come to fruition. In fact: (i) the Ohio BMV had investigated Carvana,
suspended its temporary tag issuance privileges for all Ohio locations, and subjected the
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 77 of 332
- 74 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Company to increased oversight in December 2020; (ii) following additional meetings
between MDOS and Carvana, Carvana “admi[tted] . . . several more violations of
[Michigan’s] code,” faced thousands of dollars in additional fines, and had its 18-month
probation extended on February 7, 2022; and (iii) Illinois had begun investigating Carvana in
February 2022 for issuing out-of-state temporary registration permits and for failing to
transfer titles in a timely manner in violation of its laws. Defendants also concealed the
inevitably of the foregoing risks and identified harms materializing because of the above
omitted facts.
(b)
Defendants presented Carvana’s “failure to comply” with title and
registration laws and others’ “allegation[s] that we violated these laws,” as hypothetical risks
that could materially harm Carvana if they materialized when, in truth, these risks had
already come to fruition. Indeed, Carvana routinely and repeatedly: (i) sold cars before
holding title to the vehicles; (ii) failed to register cars within the legally required timeframe;
and (iii) issued out-of-state temporary tags and/or license plates in violation of numerous
state laws. §VI.D. For example, CWs-3, 4, 8, 9, 10, and 11 universally reported that
Carvana would sell these vehicles before they received title. Specifically, CW-10 described
instances where titles on vehicles Carvana sold could not be obtained for over a year, which
meant the buyers were unable to drive their cars. CW-8 said that Carvana’s practice was
simply to assume that titles for vehicles could be acquired later if they were not available at
the time of purchase. CW-4 said it was well known internally that title issues were a
problem. CW-4 also said there was a Google spreadsheet that tracked vehicles without titles
for his/her entire IRC and there were “a ridiculous amount of cars on the list.” CW-9 said,
beginning in Q1 2022, registration delays amounted to roughly half of the calls the team had
handled and that employees informed Garcia Junior of this in January 2022. Further, from
June 2021 to July 2022, Carvana failed to timely register approximately 10% of its sales in
Maryland, which would later result in additional fines. In addition, the Barron’s exposé
would later reveal that Carvana’s violations were systematic and wide-reaching. Defendants
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 78 of 332
- 75 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
also concealed the inevitably of the foregoing risks and identified harms materializing
because of the above omitted facts.
(c)
A reasonable investor would have viewed the above omitted facts as
important for the reasons stated herein. Indeed, Defendants admitted in an October 2022
court filing that “‘the damage to Carvana’s reputation and goodwill posed by’” even one
“‘suspension order [wa]s incalculable and irreparable.’”
186. Statement No. 7: On May 10, 2022, Carvana filed its quarterly report on Form
10-Q for the period ended March 31, 2022 (the “Q1 2022 10-Q”) with the SEC, signed by
Jenkins and attaching SOX certifications signed by Garcia Junior and Jenkins. The Q1 2022
10-Q stated: “There have been no material changes to the risk factors disclosed under the
heading ‘Risk Factors’ in our most recent Annual Report on Form 10-K, filed on
February 24, 2022, except” as related to risks concerning a “larger automotive ecosystem,
including consumer demand, global supply chain challenges, and other macroeconomic
issues.” The risk factors disclosed in the 2021 10-K, stated:
We operate in several highly regulated industries and are subject to a
wide range of federal, state, and local laws and regulations. . . . [O]ur failure
to comply [with these laws and regulations], could have a material adverse
effect on our business, results of operations, and financial condition.
We are subject to a wide range of evolving federal, state, and local laws
and regulations, many of which may have limited to no interpretation
precedent as it relates to our business model. Our sale and purchase of used
vehicles and related activities, including the sale of complementary products
and services, are subject to state and local licensing requirements, state laws,
regulations, and systems and process requirements related to title and
registration . . . .
*
*
*
The violation of any of these laws or regulations could result in
administrative, civil, or criminal penalties or in a cease-and-desist order
against some or all of our business activities, any of which could damage
our reputation and have a material adverse effect on our business, sales, and
results of operations. Additionally, even an allegation that we violated these
laws, by regulators, competitors, individuals, or consumers, could result in
costly litigation with uncertain results.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 79 of 332
- 76 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
187. As detailed below, Statement No. 7 made by Carvana, Garcia Junior, and
Jenkins was materially misleading when made because Defendants omitted to state material
facts necessary in order to make it not misleading.
(a)
Defendants’ May 10, 2022 statement that “[t]here have been no material
changes to the risk factors disclosed under the heading ‘Risk Factors’ in” the 2021 10-K was
misleading because it omitted that the material risks and harms previously warned of –
“administrative, civil, or criminal penalties or in a cease-and-desist order against some or all
of our business activities” – had already come to fruition and materially changed since
February 24, 2022. In fact: (i) the Ohio BMV had investigated Carvana, suspended its
temporary tag issuance privileges for all Ohio locations, and subjected the Company to
increased oversight in December 2020; (ii) following additional meetings between MDOS
and Carvana, Carvana “admi[tted] . . . several more violations of [Michigan’s] code,” faced
thousands of dollars in additional fines, and had its 18-month probation extended on
February 7, 2022; (iii) Carvana was continuing to violate the terms of its Michigan probation
and would later have its license revoked for doing so 127 times; and (iv) Illinois had begun
investigating Carvana in February 2022 for issuing out-of-state temporary registration
permits and for failing to transfer titles in a timely manner. Defendants also concealed the
inevitably of the foregoing risks materializing because of the above facts.
(b)
Defendants’ statement that “[t]here have been no material changes to
the risk factors disclosed under the heading ‘Risk Factors’ in” the 2021 10-K, omitted that
the material risks previously warned of – an “adverse effect on our . . . sales” – had already
occurred as Defendants admitted on August 4, 2022 that “over the last several months,”
Carvana had to implement “buffers in many states to provide cushion for their title and
registration teams to work with the various states to complete necessary registration
paperwork,” which materially slowed growth.
(c)
Defendants presented Carvana’s “failure to comply” with title and
registration laws and others’ “allegation[s] that we violated these laws,” as hypothetical risks
that could materially harm Carvana if they materialized when, in truth, these risks had
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 80 of 332
- 77 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
already come to fruition. Carvana routinely and repeatedly: (i) sold cars before holding title
to the vehicles; (ii) failed to register cars within the legally required timeframe; and (iii)
issued out-of-state temporary tags and/or license plates in violation of numerous state laws.
§VI.D. Indeed, just two days after this statement, Illinois suspended Carvana’s dealer
license for failing to timely transfer title and improperly supplying customers with out-of-
state temporary plates, causing some consumers to receive police citations. Even following
its suspension “‘Carvana continued to conduct business in a manner that violates Illinois
state law.’” In addition, CWs-3, 4, 8, 9, 10, and 11 universally reported that Carvana would
sell these vehicles before they received title. Specifically, CW-10 described instances where
titles on vehicles Carvana sold could not be obtained for over a year, which meant the buyers
were unable to drive their cars. CW-8 said that Carvana’s practice was simply to assume
that titles for vehicles could be acquired later if they were not available at the time of
purchase. CW-4 said it was well known internally that title issues were a problem. CW-4
also said there was a Google spreadsheet that tracked vehicles without titles for his/her entire
IRC and there were “a ridiculous amount of cars on the list.” CW-9 said, beginning in Q1
2022, registration delays amounted to roughly half of the calls the team had handled and that
employees informed Garcia Junior of this in January 2022. Further, from June 2021 to July
2022, Carvana failed to timely register approximately 10% of its sales in Maryland, which
would later result in additional fines. In addition, the Barron’s exposé would later reveal
that Carvana’s violations were systematic and wide-reaching. Defendants also concealed the
inevitably of the foregoing risks and identified harms materializing because of the above
omitted facts.
(d)
A reasonable investor would have viewed the above omitted facts as
important for the reasons stated herein. Indeed, Defendants admitted in an October 2022
court filing that “‘the damage to Carvana’s reputation and goodwill posed by’” even one
“‘suspension order [wa]s incalculable and irreparable.’”
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 81 of 332
- 78 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
188. Statement No. 8: On June 24, 2022, a Carvana spokesman minimized
Carvana’s title and registration problems and the nature of its disagreements with regulators,
telling Barron’s:
“Carvana, like many dealers over the past two years, has in limited
instances encountered challenges processing title and registration paperwork
for its customers after the sale,” the company says. “In a very small
percentage of a very small percentage of instances, customers did not receive
permanent license plates or transferred title within the time frame set forth
by the respective states.”
*
*
*
Carvana says the regulatory issues focus on a relatively small number
of sales from 2020 and 2021. “We’ve had productive conversations with
regulators in all of those states and feel very confident about our operations
going forward,” it says.
189. As detailed below, Statement No. 8 made by Carvana omitted material facts
that affirmatively created an impression of a state of affairs that differed in a material way
from the one that actually existed.
(a)
Contrary to Carvana’s statement that these challenges were occurring
“[i]n a very small percentage of a very small percentage of instances”: (i) the Ohio BMV had
investigated Carvana, suspended its temporary tag issuance privileges for all Ohio locations,
and subjected the Company to increased oversight in December 2020; (ii) following
additional meetings between MDOS and Carvana, Carvana “admi[tted] . . . several more
violations of [Michigan’s] code,” faced thousands of dollars in additional fines, and had its
18-month probation extended on February 7, 2022; and (iii) Carvana was continuing to
violate the terms of its Michigan probation and would later have its license revoked for doing
so 127 times. Further, far from being a rare or isolated occurrence, CWs-3, 4, 8, 9, 10, and
11 universally reported that Carvana would sell these vehicles before they received title.
Specifically, CW-10 described instances where titles on vehicles Carvana sold could not be
obtained for over a year, which meant the buyers were unable to drive their cars. CW-8 said
that Carvana’s practice was simply to assume that titles for vehicles could be acquired later if
they were not available at the time of purchase. CW-4 said it was well known internally that
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 82 of 332
- 79 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
title issues were a problem. CW-4 also said there was a Google spreadsheet that tracked
vehicles without titles for his/her entire IRC and there were “a ridiculous amount of cars on
the list.” CW-9 said, beginning in Q1 2022, registration delays amounted to roughly half of
the calls the team had handled and that employees informed Garcia Junior of this in January
2022. Further, from June 2021 to July 2022, Carvana failed to timely register approximately
10% of its sales in Maryland, which would later result in additional fines. Not surprisingly,
title and registration issues and penalties had been discussed in every Board meeting since
July of 2020 until at least December 2021. See, e.g., Exs. 1-2; see also Schertz, ECF 23.
Moreover, as alleged herein, subsequent government investigations, plea agreements,
suspensions, and media reports confirm that, during the Class Period, Defendants violated
laws and regulations in Texas, Florida, Michigan, Maryland, Oklahoma, Illinois,
Pennsylvania, Ohio, and Georgia.
(b)
Contrary to Carvana’s statement that it was having “‘productive
conversations with regulators’” and “‘fe[lt] very confident . . . going forward,’” mere weeks
later, Michigan suspended Carvana’s license for: (i) “[f]ailing to make application for title
and registration within 15 days of delivery for 112 customers since agreeing to an earlier
probation extension”; (ii) “[c]ommitting fraud in connection with selling or otherwise
dealing in vehicles where Carvana employees admitted to destroying title applications and all
applicable documents pertaining to the sale of [certain vehicles]” that it later took back from
customers; (iii) “[i]mproperly issuing temporary registrations”; and (4) “[v]iolating [the]
terms of a probation agreement 127 times.” Further, shortly after Carvana’s statement, the
Company would see its licenses suspended or revoked in Pennsylvania and Illinois.
(c)
The omitted facts would have been viewed by a reasonable investor as
significant for the reasons stated herein, and because Defendants stated that, as to just one of
the state actions, “‘the damage to Carvana’s reputation and goodwill posed by the suspension
order is incalculable and irreparable.’”
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 83 of 332
- 80 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
B.
Defendants’ Materially False and Misleading Statements and
Omissions Concerning Buying Cars from Customers
190. As Jenkins reminded investors on the first day of the Class Period, buying cars
from customers was purportedly “a major innovation and a major add to the business.”
Thus, during the Class Period, Carvana emphasized in its SEC filings that a critical part of its
growth strategy was to “[i]ncrease the purchase of vehicles from customers.” Accordingly,
leading up to and during much of the Class Period, Garcia Junior and Jenkins dedicated an
entire section of their shareholder letters to touting the number of customer-sourced cars and
describing the advantages of their strategy. However, unbeknownst to investors, by the
second half of fiscal year ended December 31, 2020 (“FY 2020”), Defendants had drastically
lowered (or disregarded) their purchasing and verification standards to significantly increase
the number of cars they were purchasing from customers in order to induce more trade-ins
and increase retail sales. While this action did increase retail unit sales, it adversely
impacted Carvana’s bottom line, flooded Carvana with wholesale cars that had to be sold at a
loss, and crippled its logistics network. At the same time, Defendants touted only the
positive “advantages” of buying cars from customers while assuring investors that Carvana
“assess[es] vehicles on the basis of quality.” In short, Defendants’ statements below were
misleading by omission because Defendants touted positive information to the market but
failed to disclose material adverse information that cut against that positive information.
191. Statement No. 9: On October 29, 2020, after the market closed, Carvana filed
its quarterly report on Form 10-Q for the period ended September 30, 2020 (the “Q3 2020
10-Q”), signed by Jenkins and attaching SOX certifications signed by Jenkins and Garcia
Junior. The Q3 2020 10-Q stated:
Vehicle sourcing and acquisition. . . . For vehicles sold to us through
our website, we use proprietary algorithms to determine an appropriate offer.
We assess vehicles on the basis of quality, inventory fit, consumer
desirability, relative value, expected reconditioning costs, and vehicle
location to identify what we believe represent the most in-demand and
profitable vehicles to acquire for inventory. We utilize a broad range of data
sources, including proprietary site data, and a variety of external data sources
to support our assessments.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 84 of 332
- 81 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
192. As detailed below, Statement No. 9 by Carvana, Jenkins, and Garcia Junior
omitted material facts that affirmatively created an impression of a state of affairs that
differed in a material way from the one that actually existed.
(a)
Contrary to Defendants’ statement that they “assess vehicles on the
basis of quality,” Defendants concealed from investors that Carvana “was intentionally
buying lower quality cars.” (Emphasis in original.)19
(b)
Defendants omitted that Carvana abruptly and drastically lowered or
disregarded entirely the Company’s purchasing and verification standards when buying cars
from customers in order to induce trade-ins and increase retail sales. §VI.B. Indeed, by
lowering or disregarding the Company’s purchasing and verification standards, Carvana
increased cars sourced from customers in Q3 2020 by 261% over the prior quarter, as shown
in the chart below. CWs-1 and 4 confirmed that, to boost retail sales through trade-ins,
Carvana bought cars from customers that clearly did not meet its purchasing and verification
standards. For example, CW-1 reported that Carvana had purchased a bullet-ridden car and a
car missing its back seat. CW-3 stated that Carvana had relaxed its standards to accept “all
vehicles” and did not genuinely inspect vehicles before purchasing them anymore, resulting
in a lot of “trash vehicles.” CW-11 said that Carvana was buying vehicles “sight unseen”
and his/her team was not expected to do much to verify the condition of the vehicles. CW-
11 observed that Carvana did not actually care about the condition of the vehicles and just
wanted to “get as many cars as it could.” In short, contrary to the statement that Carvana
was “assess[ing] vehicles on the basis of quality, inventory fit, consumer desirability, relative
value, expected reconditioning costs, and vehicle location to identify what we believe
represent the most in-demand and profitable vehicles to acquire for inventory,” Carvana
largely disregarded the condition of the vehicles they purchased as CWs-1, 3, 4, and 11 all
reported.
19 Defendants judicially admitted this omitted fact by including it in a brief in this action.
ECF 58 at 12.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 85 of 332
- 82 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
(c)
As detailed below, Defendants omitted that buying low-quality cars
from customers adversely impacted Carvana’s bottom line by flooding Carvana with
wholesale cars that had to be sold at a loss and crippling its logistics network.
(i)
Because Carvana was “intentionally buying lower quality cars,”
it was flooded with cars that did not meet its retail standards and, thus, had to be sold
wholesale. Indeed, upon lowering or disregarding its purchasing and verification standards,
Carvana’s wholesale growth began to far outpace its retail sales growth. As shown in the
charts below, immediately following Carvana’s 128% spike in the number of cars purchased
from customers during Q3 2020, its wholesale sales spiked by over 100% during Q4 2020
and Q1 2021. Meanwhile, wholesale sales made up 20% of the Company’s total sales in Q3
2020 compared to just 12% one quarter earlier.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 86 of 332
- 83 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
(ii)
This massive increase in wholesale vehicles adversely impacted
Carvana’s bottom line because, unbeknownst to investors, Carvana lost money on wholesale
vehicle sales, as detailed below:
(1)
Carvana sold cars at wholesale at significantly lower
prices than those sold at retail. Carvana, however, still incurred many of the same costs
involved in selling retail vehicles, including inbound transportation, reconditioning, and
other logistics. Thus, the lower average sales price combined with customary selling costs
resulted in significantly lower reported GPU on wholesale cars. In Q3 2020, wholesale GPU
was $1,113, 40% lower than Retail GPU of $1,857.20
(2)
Although Carvana disclosed that it generated a modest
gross profit on wholesale vehicle sales, Defendants masked the actual profitability – or lack
thereof – of wholesale sales by: (i) excluding certain per-vehicle operations expenses from its
calculation of wholesale GPU; and (ii) not disclosing or quantifying these excluded costs
separately so investors could decipher overall profitability of wholesale sales on their own.
These per-vehicle operations expenses, which were only separately broken out and
quantified for the first time after the Class Period, included material costs associated with
20 Carvana also did not earn any ancillary revenue or profit on wholesale vehicle sales, such
as warranty, insurance, or financing-related revenue. These “other” revenue streams
represented the vast majority of Carvana’s reported total GPU on its retail sales, but
generated $0 on wholesale sales.
0%
100%
200%
300%
400%
500%
Q1 20Q2 20Q3 20Q4 20Q1 21Q2 21Q3 21Q4 21Q1 22
Wholesale Growth Exceeded Retail Growth
y/y growth (retail cars sold)
y/y growth (wholesale cars sold)
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 87 of 332
- 84 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
completing a wholesale sale, such as the full cost to ship a car to a wholesale auction site
and title and registration costs.21 Tellingly, Carvana admitted after the Class Period in its Q3
2023 letter to shareholders dated November 2, 2023 (the “Q3 2023 Shareholder Letter”) and
the November 2, 2023 Cost Structure Details presentation, that there are “two key drivers” of
“unit economics” (i.e., per-vehicle profitability): (i) cost of sales, which were the expenses
Carvana had always included in its reported GPU calculations; and (ii) “operations
expenses” per vehicle sold, which consisted of per-vehicle selling expenses that were not
included in Carvana’s reported GPU and had never been separately broken out to investors.22
Instead, during the Class Period, these expenses that drove Carvana’s actual profit per car
sold, were lumped into and comingled with other SG&A expenses.23 Defendants later
admitted, however, that these “operations expenses” were not typical fixed or corporate
overhead SG&A costs, conceding: “Operations expenses include the fulfillment, customer
service, and transaction expenses associated with completing retail and wholesale vehicle
sales.”24 Indeed, these vehicle selling expenses that Carvana had arbitrarily excluded from
21 In its November 2, 2023 Cost Structure Details presentation, Carvana disclosed and
quantified retail and wholesale vehicle operations expense.
22 After disclosing and quantifying operations expenses for the first time in November
2023, Jenkins acknowledged that the additional disclosure “allows a much deeper dive into
the economics of Carvana.” Jenkins added: “we’ve really been focused on . . . improving the
unit economics of the business. In other words, driving down variable costs per car,” which
he confirmed included “operations expenses, which are the more variable component of our
selling, general and administrative expenses.” Jenkins continued: “And the reason that we’re
focused on that is obviously, the more gains we make on reducing variable cost per car
today, the more profitable our growth can be in the future . . . .” Jenkins also described
“driving down operational expenses per unit” as one of Carvana’s “profitability initiatives,
and in particular, . . . our unit economics initiatives.” Thus, according to Jenkins, operations
expenses were a critical component of assessing Carvana’s profitability.
23 While Defendants included per-vehicle operations expenses in SG&A expenses (rather
than GPU), they did not separately disclose or quantify these expenses in a manner that
would have allowed investors to decipher the overall per-vehicle profitability on their own.
In fact, as detailed in ¶252(d), the disclosures that Defendants did make regarding SG&A
expenses during Q3 2020 did more to mislead investors than to provide transparency around
Carvana’s true profitability per vehicle sold.
24 During the Class Period, Defendants described Carvana’s SG&A costs as “expenses
associated with advertising and providing customer service to customers, operating our
vending machines and hubs, operating our logistics and fulfillment network and other
corporate overhead expenses, including expenses associated with information technology,
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 88 of 332
- 85 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
its reported GPU included: (i) “customer care, multi-car logistics, last-mile delivery, and
other operations payroll” “associated with completing . . . wholesale vehicle sales”; (ii) “non-
payroll logistics expenses, including fuel, repairs and maintenance, and third-party transport
services” “associated with completing . . . wholesale vehicle sales”; and (iii) “[t]ransaction
and other expenses, including limited warranty, title and registration, and finance platform
expenses” “associated with completing . . . wholesale vehicle sales.” Defendants’
concealment of “operations expenses” when describing Carvana’s wholesale profitability
was especially misleading in light of the fact that Defendants internally viewed “the
underlying costs of completing a sale” as a critical component of Carvana’s actual
wholesale vehicle profitability. Indeed, as Garcia Junior later acknowledged:
When we think about trying to aim for more profitable sales, what does that
mean? There’s certainly variability in the kind of gross profit associated with
different types of sales. . . .
And then I think there’s a number of other dynamics kind of across car
type, et cetera. There’s also kind of variation in the underlying costs of
completing a sale.
Because Defendants did not specifically break out operations expenses (i.e., the underlying
costs of completing a sale) and instead buried and comingled them among SG&A expenses,
investors could not calculate the actual overall profitability of Carvana’s wholesale vehicle
sales, like Garcia Junior did internally. As demonstrated in the chart below, upon
incorporating all of the estimated vehicle operations expenses (i.e., the actual costs Carvana
incurred in completing wholesale vehicle sales), Carvana lost money on wholesale vehicle
sales every quarter during the Class Period.25 For example, in Q3 2020, Carvana’s average
product development, engineering, legal, accounting, finance, and business development.”
Moreover, while Defendants made vague Class Period disclosures indicating that certain of
expenses were recorded in SG&A, until the end of the Class Period Carvana never: (i) broke
out “operations expenses” to investors; (ii) described that these vehicle “operations
expenses” were direct per-vehicle costs “associated with completing . . . wholesale vehicle
sales”; nor (iii) described that, in addition to the selling costs included in Carvana’s GPU
metric, these additional per-vehicle “operations expenses,” were one of “two key drivers” of
per-unit profitability.
25 Estimated operations expense per wholesale vehicle calculated as follows: total retail and
wholesale vehicle operations expenses in Q3 2020 divided by total retail and wholesale
vehicles sold in Q3 2020.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 89 of 332
- 86 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
selling price per wholesale vehicle was $8,450 while its total costs (cost of sales in addition
to estimated undisclosed operations expense) totaled $8,831, for a total loss of $381 per
vehicle.
(iii)
Defendants omitted that the surplus of wholesale cars resulting
from Carvana “intentionally buying lower quality cars” also led to significant logistics
constraints. For example, in Carvana’s May 2022 Operating Plan, Defendants admitted that
“[b]uying cars from customers growth” led to “additional wholesale volume” which, in turn,
led to “constraints in our nationwide logistics network.” Carvana further admitted that
“wholesale units acquired from customers have typically been transported to the nearest
Carvana IRC, generating additional vehicles [sic] moves and increased complexity in our
multi-car logistics network.”
(iv)
Defendants omitted that Carvana lacked critical infrastructure
necessary to process the huge surplus of wholesale vehicles. Consequently, Carvana was
forced to spend $3 billion to purchase ADESA, a nationwide wholesale auction house. In
describing his rationale for the purchase, Garcia Junior pointed to ADESA’s 56 brick-and-
mortar locations, consisting of 6.5 million square feet of buildings on more than 4,000 acres
of land, and as reported in a September 13, 2022 Vehicle Remarketing article, explained
“‘[y]ou need a place to stage the logistics, and there’s no better infrastructure than auctions
to serve that purpose.’” Indeed, Defendants later revealed that, prior to acquiring ADESA,
Carvana “frequently ha[d] our last mile delivery advocates pick these cars up from
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 90 of 332
- 87 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
customer’s homes and transport them to a market hub and then transport these cars long
distances from their origin to our nearest IRC where they are either reconditioned or held
until they are wholesaled.” Further, they admitted that “[w]ith this acquisition [of ADESA],
we can now position advocates at the auction locations and transport cars from our
customer’s driveways directly to an ADESA U.S. location without any additional moves.”
193. Statement No. 10: On February 25, 2021, Carvana held a conference call for
analysts and investors to discuss the Q4 2020 results. During the Q4 2020 earnings call,
Garcia Junior responded to an analyst’s question regarding the excess demand for consumers
to sell to Carvana:
[W]e’ve got a model where we can sell cars across geographies and across
makes and models and across the price spectrum. And we’ve got very
efficient ways to sell even wholesale cars. It puts us in a position where we
can be a very high-quality buyer. And so we’ve seen a lot of success in that
business.
So I think that we’re excited by where we are. I think going forward,
it’s – there’s no – until we get extremely large, there isn’t necessarily a
fundamental ratio that we would shoot for with respect to cars bought relative
to cars sold. I think we will just seek to grow both businesses as quickly as we
possibly can because we’ve set up the business to do so. If we’re buying
significantly more cars than we’re selling, then we have the capacity to sell
the excess wholesale. If we’re not, then we have the capacity to buy cars from
other channels. And so we’re just going to look to grow both businesses as
quickly as we can independently.
194. As detailed below, Statement No. 10 made by Carvana and Garcia Junior
omitted material facts that affirmatively created an impression of a state of affairs that
differed in a material way from the one that actually existed.
(a)
Contrary to Defendants’ statement that Carvana has “very efficient ways
to sell even wholesale cars,” and it readily “ha[s] the capacity to sell the excess [cars from
customers] wholesale,” Carvana lacked critical infrastructure necessary to process the huge
surplus of wholesale vehicles resulting from Carvana “intentionally buying lower quality
cars.” (Emphasis in original.) In Carvana’s May 2022 Operating Plan, Defendants admitted
that “[b]uying cars from customers growth” led to “additional wholesale volume” which, in
turn, led to “constraints in our nationwide logistics network.” Carvana also admitted that
“wholesale units acquired from customers have typically been transported to the nearest
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 91 of 332
- 88 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Carvana IRC, generating additional vehicles [sic] moves and increased complexity in our
multi-car logistics network.” Consequently, Carvana was forced to spend $3 billion to
purchase ADESA, a nationwide wholesale auction house. In describing his rationale for the
purchase, Garcia Junior pointed to ADESA’s 56 brick-and-mortar locations, consisting of 6.5
million square feet of buildings on more than 4,000 acres of land, and as reported in a
September 13, 2022 Vehicle Remarketing article, explained “‘[y]ou need a place to stage the
logistics, and there’s no better infrastructure than auctions to serve that purpose.’” Indeed,
Defendants later revealed that, prior to acquiring ADESA, Carvana “frequently ha[d] our last
mile delivery advocates pick these cars up from customer’s homes and transport them to a
market hub and then transport these cars long distances from their origin to our nearest IRC
where they are either reconditioned or held until they are wholesaled.” Further, they
admitted that “[w]ith this acquisition [of ADESA], we can now position advocates at the
auction locations and transport cars from our customer’s driveways directly to an ADESA
U.S. location without any additional moves.”
(b)
In addition, it was also materially misleading to tout Carvana’s “very
efficient ways to sell” wholesale cars and “capacity to sell the excess [cars from customers]
wholesale,” while omitting that the flood of wholesale cars was adversely impacting
Carvana’s bottom line because the wholesale vehicles were sold at a loss and crippled
Carvana’s logistic network, as detailed below.
(i)
Carvana sold cars at wholesale at significantly lower prices than
those sold at retail. Carvana, however, still incurred many of the same costs involved in
selling retail vehicles, including inbound transportation, reconditioning, and other logistics.
Thus, the lower average sales price combined with customary selling costs resulted in
significantly lower reported GPU on wholesale cars. In Q4 2020, wholesale GPU was $358,
over 70% lower than Retail GPU of $1,265.26
26 Carvana also did not earn any ancillary revenue or profit on wholesale vehicle sales, such
as warranty, insurance, or financing-related revenue. These “other” revenue streams
represented the vast majority of Carvana’s reported total GPU on its retail sales, but
generated $0 on wholesale sales.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 92 of 332
- 89 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
(ii)
Although Carvana disclosed that it generated a modest gross
profit on wholesale vehicle sales, Defendants masked the actual profitability – or lack thereof
– of wholesale sales by: (1) excluding certain per-vehicle operations expenses from its
calculation of wholesale GPU; and (2) not disclosing or quantifying these excluded costs
separately so investors could decipher overall profitability of wholesale sales on their own.
These per-vehicle operations expenses, which were only separately broken out and
quantified for the first time after the Class Period, included material costs associated with
completing a wholesale sale, such as the full cost to ship a car to a wholesale auction site
and title and registration costs.27 Tellingly, Carvana admitted after the Class Period in its Q3
2023 Shareholder Letter and November 2, 2023 Cost Structure Details presentation that there
are “two key drivers” of “unit economics” (i.e., per-vehicle profitability): (i) cost of sales,
which were the expenses Carvana had always included in its reported GPU calculations; and
(ii) “operations expenses” per vehicle sold, which consisted of per-vehicle selling expenses
that were not included in Carvana’s reported GPU and had never been separately broken out
to investors.28 Instead, during the Class Period, these expenses that drove Carvana’s actual
profit per car sold, were lumped into and comingled with other SG&A expenses.29
27 In its November 2, 2023 Cost Structure Details presentation, Carvana disclosed and
quantified retail and wholesale vehicle operations expense.
28 After disclosing and quantifying operations expenses for the first time in November
2023, Jenkins acknowledged that the additional disclosure “allows a much deeper dive into
the economics of Carvana.” Jenkins added: “we’ve really been focused on . . . improving the
unit economics of the business. In other words, driving down variable costs per car,” which
he confirmed included “operations expenses, which are the more variable component of our
selling, general and administrative expenses.” Jenkins continued: “And the reason that we’re
focused on that is obviously, the more gains we make on reducing variable cost per car
today, the more profitable our growth can be in the future . . . .” Jenkins also described
“driving down operational expenses per unit” as one of Carvana’s “profitability initiatives,
and in particular, . . . our unit economics initiatives.” Thus, according to Jenkins, operations
expenses were a critical component of assessing Carvana’s profitability.
29 While Defendants included per-vehicle operations expenses in SG&A expenses (rather
than GPU), they did not separately disclose or quantify these expenses in a manner that
would have allowed investors to decipher the overall per-vehicle profitability on their own.
In fact, as detailed in ¶252(d), the disclosures that Defendants did make regarding SG&A
expenses during the Class Period did more to mislead investors than to provide transparency
around Carvana’s true profitability per vehicle sold.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 93 of 332
- 90 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Defendants later admitted, however, that these “operations expenses” were not typical fixed
or corporate overhead SG&A costs, conceding: “Operations expenses include the fulfillment,
customer service, and transaction expenses associated with completing retail and wholesale
vehicle sales.”30 Indeed, these vehicle selling expenses that Carvana had arbitrarily excluded
from its reported GPU included: (1) “customer care, multi-car logistics, last-mile delivery,
and other operations payroll” “associated with completing . . . wholesale vehicle sales”; (2)
“non-payroll logistics expenses, including fuel, repairs and maintenance, and third-party
transport services” “associated with completing . . . wholesale vehicle sales”; and (3)
“[t]ransaction and other expenses, including limited warranty, title and registration, and
finance platform expenses” “associated with completing . . . wholesale vehicle sales.”
Defendants’ concealment of “operations expenses” when describing Carvana’s wholesale
profitability was especially misleading in light of the fact that Defendants internally viewed
“the underlying costs of completing a sale” as a critical component of Carvana’s actual
wholesale vehicle profitability. Indeed, as Garcia Junior later acknowledged:
When we think about trying to aim for more profitable sales, what does that
mean? There’s certainly variability in the kind of gross profit associated with
different types of sales. . . .
And then I think there’s a number of other dynamics kind of across car
type, et cetera. There’s also kind of variation in the underlying costs of
completing a sale.
Because Defendants did not specifically break out operations expenses (i.e., the underlying
costs of completing a sale) and instead buried and comingled them among SG&A expenses,
investors could not calculate the actual overall profitability of Carvana’s wholesale vehicle
30 During the Class Period, Defendants described Carvana’s SG&A costs as “expenses
associated with advertising and providing customer service to customers, operating our
vending machines and hubs, operating our logistics and fulfillment network and other
corporate overhead expenses, including expenses associated with information technology,
product development, engineering, legal, accounting, finance, and business development.”
Moreover, while Defendants made vague Class Period disclosures indicating that certain of
expenses were recorded in SG&A, until the end of the Class Period Carvana never: (i) broke
out “operations expenses” to investors; (ii) described that these vehicle “operations
expenses” were direct per-vehicle costs “associated with completing . . . wholesale vehicle
sales”; nor (iii) described that, in addition to the selling costs included in Carvana’s GPU
metric, these additional per-vehicle “operations expenses,” were one of “two key drivers” of
per-unit profitability.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 94 of 332
- 91 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
sales, like Garcia Junior did internally. As demonstrated in the chart below, upon
incorporating all of the estimated vehicle operations expenses (i.e., the actual costs Carvana
incurred in completing wholesale vehicle sales), Carvana lost money on wholesale vehicle
sales every quarter during the Class Period.31 For example, in Q4 2020, Carvana’s average
selling price per wholesale vehicle was $8,545, while its total costs (cost of sales in addition
to estimated undisclosed operations expense) totaled $9,877, for a total loss of $1,332 per
wholesale vehicle sold.
(c)
Analysts were misled. On February 26, 2021, a BofA Securities analyst
stated, “[r]evenue upside seems to be at least mostly driven by non-core wholesale unit.”
195. Statement No. 11: On February 25, 2021, Carvana filed its 2020 10-K with the
SEC, signed by Garcia Junior and Jenkins and attaching SOX certifications signed by Garcia
Junior and Jenkins. The 2020 10-K stated:
Vehicle acquisition. . . . For vehicles sold to us through our website,
we use proprietary algorithms to determine an appropriate offer. We assess
vehicles on the basis of quality, inventory fit, consumer desirability, relative
value, expected reconditioning costs, and vehicle location to identify what we
believe represent the most in-demand and profitable vehicles to acquire for
inventory. We utilize a broad range of data sources, including proprietary site
data, and a variety of external data sources to support our assessments.
31 Estimated operations expense per wholesale vehicle calculated as follows: total retail and
wholesale vehicle operations expenses in Q4 2020 divided by total retail and wholesale
vehicles sold in Q4 2020.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 95 of 332
- 92 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
196. As detailed below, Statement No. 11 made by Carvana, Garcia Junior, and
Jenkins omitted material facts that affirmatively created an impression of a state of affairs
that differed in a material way from the one that actually existed.
(a)
Contrary to Defendants’ statement that they “assess vehicles on the
basis of quality,” Defendants concealed from investors that Carvana “was intentionally
buying lower quality cars.” (Emphasis in original.)32
(b)
Defendants omitted that Carvana abruptly and drastically lowered or
disregarded entirely the Company’s purchasing and verification standards when buying cars
from customers in order to induce trade-ins and increase retail sales. §VI.B. Indeed, by
lowering or disregarding the Company’s purchasing and verification standards, Carvana
increased cars sourced from customers during Q4 2020 by 110% year-over-year. CWs-1 and
4 confirmed that, to boost retail sales through trade-ins, Carvana bought cars from customers
that clearly did not meet its purchasing and verification standards. For example, CW-1
reported that Carvana had purchased a car missing its back seat and one with bullet holes in
it which the customer had rated in fair condition. CW-3 stated that Carvana had relaxed its
standards to accept “all vehicles” and did not genuinely inspect vehicles before purchasing
them anymore, resulting in a lot of “trash vehicles.” CW-11 said that Carvana was buying
vehicles “sight unseen” and his/her team was not expected to do much to verify the condition
of the vehicles. CW-11 observed that Carvana did not actually care about the condition of
the vehicles and just wanted to “get as many cars as it could.” In short, contrary to the
statement that Carvana was “assess[ing] vehicles on the basis of quality, inventory fit,
consumer desirability, relative value, expected reconditioning costs, and vehicle location to
identify what we believe represent the most in-demand and profitable vehicles to acquire for
inventory,” Carvana largely disregarded the condition of the vehicles they purchased as
CWs-1, 3, 4, and 11 all reported.
32 Defendants judicially admitted this omitted fact by including it in a brief in this action.
ECF 58 at 12.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 96 of 332
- 93 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
(c)
As detailed below, Defendants omitted that buying low-quality cars
from customers adversely impacted the Company’s bottom line by flooding Carvana with
less profitable wholesale cars that had to be sold at a loss and crippling its logistics network.
(i)
Because Carvana was “intentionally buying lower quality cars,”
it was flooded with cars that did not meet its retail standards and, thus, had to be sold
wholesale. Indeed, upon lowering or disregarding its purchasing and verification standards,
Carvana’s wholesale growth began to far outpace its retail sales growth. As shown in the
charts below, immediately following Carvana’s 128% spike in the number of cars purchased
from customers during Q3 2020, its wholesale sales spiked by over 100% during Q4 2020
and Q1 2021. Meanwhile, wholesale sales made up over 20% of the Company’s total sales
in Q4 2020 as compared to just 12% two quarters earlier.
(ii)
This massive increase in wholesale vehicles adversely impacted
Carvana’s bottom line because, unbeknownst to investors, Carvana lost money on wholesale
vehicle sales, as detailed below:
(1)
Carvana sold cars at wholesale at significantly lower
prices than those sold at retail. Carvana, however, still incurred many of the same costs
involved in selling retail vehicles, including inbound transportation, reconditioning, and
other logistics. Thus, the lower average sales price combined with customary selling costs
0%
100%
200%
300%
400%
500%
Q1 20Q2 20Q3 20Q4 20Q1 21Q2 21Q3 21Q4 21Q1 22
Wholesale Growth Exceeded Retail Growth
y/y growth (retail cars sold)
y/y growth (wholesale cars sold)
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 97 of 332
- 94 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
resulted in significantly lower reported GPU on wholesale cars. In Q4 2020, wholesale GPU
was $358, over 70% lower than Retail GPU of $1,265.33
(2)
Although Carvana disclosed that it generated a modest
gross profit on wholesale vehicle sales, Defendants masked the actual profitability – or lack
thereof – of wholesale sales by: (i) excluding certain per-vehicle operations expenses from its
calculation of wholesale GPU; and (ii) not disclosing or quantifying these excluded costs
separately so investors could decipher overall profitability of wholesale sales on their own.
These per-vehicle operations expenses, which were only separately broken out and
quantified for the first time after the Class Period, included material costs associated with
completing a wholesale sale, such as the full cost to ship a car to a wholesale auction site
and title and registration costs.34 Tellingly, Carvana admitted after the Class Period in its Q3
2023 Shareholder Letter and November 2, 2023 Cost Structure Details presentation that there
are “two key drivers” of “unit economics” (i.e., per-vehicle profitability): (i) cost of sales,
which were the expenses Carvana had always included in its reported GPU calculations; and
(ii) “operations expenses” per vehicle sold, which consisted of per-vehicle selling expenses
that were not included in Carvana’s reported GPU and had never been separately broken out
to investors.35 Instead, during the Class Period, these expenses that drove Carvana’s actual
33 Carvana also did not earn any ancillary revenue or profit on wholesale vehicle sales, such
as warranty, insurance, or financing-related revenue. These “other” revenue streams
represented the vast majority of Carvana’s reported total GPU on its retail sales, but
generated $0 on wholesale sales.
34 In its November 2, 2023 Cost Structure Details presentation, Carvana disclosed and
quantified retail and wholesale vehicle operations expense.
35 After disclosing and quantifying operations expenses for the first time in November
2023, Jenkins acknowledged that the additional disclosure “allows a much deeper dive into
the economics of Carvana.” Jenkins added: “we’ve really been focused on . . . improving the
unit economics of the business. In other words, driving down variable costs per car,” which
he confirmed included “operations expenses, which are the more variable component of our
selling, general and administrative expenses.” Jenkins continued: “And the reason that we’re
focused on that is obviously, the more gains we make on reducing variable cost per car
today, the more profitable our growth can be in the future . . . .” Jenkins also described
“driving down operational expenses per unit” as one of Carvana’s “profitability initiatives,
and in particular, . . . our unit economics initiatives.” Thus, according to Jenkins, operations
expenses were a critical component of assessing Carvana’s profitability.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 98 of 332
- 95 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
profit per car sold, were lumped into and comingled with other SG&A expenses.36
Defendants later admitted, however, that these “operations expenses” were not typical fixed
or corporate overhead SG&A costs, conceding: “Operations expenses include the fulfillment,
customer service, and transaction expenses associated with completing retail and wholesale
vehicle sales.”37 Indeed, these vehicle selling expenses that Carvana had arbitrarily excluded
from its reported GPU included: (i) “customer care, multi-car logistics, last-mile delivery,
and other operations payroll” “associated with completing . . . wholesale vehicle sales”; (ii)
“non-payroll logistics expenses, including fuel, repairs and maintenance, and third-party
transport services” “associated with completing . . . wholesale vehicle sales”; and (iii)
“[t]ransaction and other expenses, including limited warranty, title and registration, and
finance platform expenses” “associated with completing . . . wholesale vehicle sales.”
Defendants’ concealment of “operations expenses” when describing Carvana’s wholesale
profitability was especially misleading in light of the fact that Defendants internally viewed
“the underlying costs of completing a sale” as a critical component of Carvana’s actual
wholesale vehicle profitability. Indeed, as Garcia Junior later acknowledged:
When we think about trying to aim for more profitable sales, what does that
mean? There’s certainly variability in the kind of gross profit associated with
different types of sales. . . .
36 While Defendants included per-vehicle operations expenses in SG&A expenses (rather
than GPU), they did not separately disclose or quantify these expenses in a manner that
would have allowed investors to decipher the overall per-vehicle profitability on their own.
In fact, as detailed in ¶252(d), the disclosures that Defendants did make regarding SG&A
expenses during the Class Period did more to mislead investors than to provide transparency
around Carvana’s true profitability per vehicle sold.
37 During the Class Period, Defendants described Carvana’s SG&A costs as “expenses
associated with advertising and providing customer service to customers, operating our
vending machines and hubs, operating our logistics and fulfillment network and other
corporate overhead expenses, including expenses associated with information technology,
product development, engineering, legal, accounting, finance, and business development.”
Moreover, while Defendants made vague Class Period disclosures indicating that certain of
expenses were recorded in SG&A, until the end of the Class Period Carvana never: (i) broke
out “operations expenses” to investors; (ii) described that these vehicle “operations
expenses” were direct per-vehicle costs “associated with completing . . . wholesale vehicle
sales”; nor (iii) described that, in addition to the selling costs included in Carvana’s GPU
metric, these additional per-vehicle “operations expenses,” were one of “two key drivers” of
per-unit profitability.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 99 of 332
- 96 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
And then I think there’s a number of other dynamics kind of across car
type, et cetera. There’s also kind of variation in the underlying costs of
completing a sale.
Because Defendants did not specifically break out operations expenses (i.e., the underlying
costs of completing a sale) and instead buried and comingled them among SG&A expenses,
investors could not calculate the actual overall profitability of Carvana’s wholesale vehicle
sales, like Garcia Junior did internally. As demonstrated in the chart below, upon
incorporating all of the estimated vehicle operations expenses (i.e., the actual costs Carvana
incurred in completing wholesale vehicle sales), Carvana lost money on wholesale vehicle
sales every quarter during the Class Period.38 For example, in Q4 2020, Carvana’s average
selling price per wholesale vehicle was $8,545, while its total costs (cost of sales in addition
to estimated undisclosed operations expense) totaled $9,877, for a total loss of $1,332 per
wholesale vehicle sold.
(iii)
Defendants omitted that the surplus of wholesale cars resulting
from Carvana “intentionally buying lower quality cars” also led to significant logistics
constraints. For example, in Carvana’s May 2022 Operating Plan, Defendants admitted that
“[b]uying cars from customers growth” led to “additional wholesale volume” which, in turn,
led to “constraints in our nationwide logistics network.” Carvana further admitted that
38 Estimated operations expense per wholesale vehicle calculated as follows: total retail and
wholesale vehicle operations expenses in Q4 2020 divided by total retail and wholesale
vehicles sold in Q4 2020.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 100 of 332
- 97 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
“wholesale units acquired from customers have typically been transported to the nearest
Carvana IRC, generating additional vehicles [sic] moves and increased complexity in our
multi-car logistics network.”
(iv)
Defendants omitted that Carvana lacked critical infrastructure
necessary to process the huge surplus of wholesale vehicles. Consequently, Carvana was
forced to spend $3 billion to purchase ADESA, a nationwide wholesale auction house. In
describing his rationale for the purchase, Garcia Junior pointed to ADESA’s 56 brick-and-
mortar locations, consisting of 6.5 million square feet of buildings on more than 4,000 acres
of land, and as reported in a September 13, 2022 Vehicle Remarketing article, explained
“‘[y]ou need a place to stage the logistics, and there’s no better infrastructure than auctions
to serve that purpose.’” Indeed, Defendants later revealed that, prior to acquiring ADESA,
Carvana “frequently ha[d] our last mile delivery advocates pick these cars up from
customer’s homes and transport them to a market hub and then transport these cars long
distances from their origin to our nearest IRC where they are either reconditioned or held
until they are wholesaled.” Further, they admitted that “[w]ith this acquisition [of ADESA],
we can now position advocates at the auction locations and transport cars from our
customer’s driveways directly to an ADESA U.S. location without any additional moves.”
(d)
Analysts – unaware of the havoc Defendants’ lowering of standards
wreaked on Carvana’s operations and financials – were deceived. A February 26, 2021 JMP
Securities analyst approved of Carvana’s vehicle sourcing and noted, “customer inventory
remains a key source of inventory while also widening selection (more high-end options on
vehicles) and lowering vehicle acquisition prices.”
197. Statement No. 12: On May 6, 2021, Carvana held a conference call for
analysts and investors to discuss the Q1 2021 results. During the Q1 2021 earnings call,
Jenkins responded to an analyst question regarding buying cars from customers:
Sure. So I can take that one. So I mean, I think our #1 channel for
sourcing inventory has now become sourcing cars directly from customers,
which has several advantages that we’ve talked about historically. I think
we’re certainly going to continue to focus on that. That’s a business that we
want to build and continue to grow over the long term. And so I think that’s
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 101 of 332
- 98 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
our #1 area of focus and would be the – sort of the first answer to that
question.
198. As detailed below, Statement No. 12 by Carvana and Jenkins omitted material
facts that affirmatively created an impression of a state of affairs that differed in a material
way from the one that actually existed. While there were “advantages” of sourcing cars from
customers, Defendants omitted that Carvana had lowered or disregarded its purchasing and
verification standards and was, thus, purchasing thousands of cars that could only be sold via
the wholesale market at a loss, as alleged herein.
(a)
CWs-1, 3, 4, and 11 confirm that Carvana omitted it was intentionally
acquiring low-quality vehicles. CWs-1 and 4 confirmed that, to boost retail sales through
trade-ins, Carvana bought cars from customers that clearly did not meet its purchasing and
verification standards. For example, CW-1 reported that Carvana had purchased a bullet-
ridden car and a car missing its back seat. CW-3 stated that Carvana had relaxed its
standards to accept “all vehicles” and did not genuinely inspect vehicles before purchasing
them anymore, resulting in a lot of “trash vehicles.” CW-11 said that Carvana was buying
vehicles “sight unseen” and his/her team was not expected to do much to verify the condition
of the vehicles. CW-11 observed that Carvana did not actually care about the condition of
the vehicles and just wanted to “get as many cars as it could.”
(b)
As detailed below, Defendants omitted that buying low-quality cars
from customers adversely impacted the Company’s bottom line by flooding Carvana with
less profitable wholesale cars and crippling its logistics network.
(i)
Because Carvana was “intentionally buying lower quality cars,”
it was flooded with cars that did not meet its retail standards and, thus, had to be sold
wholesale. Indeed, upon lowering or disregarding its purchasing and verification standards,
Carvana’s wholesale growth began to far outpace its retail sales growth. As shown in the
charts below, immediately following Carvana’s 119% spike in the number of cars purchased
from customers during the last two quarters of FY 2020, its wholesale sales spiked by over
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 102 of 332
- 99 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
100% in Q1 2021, and over 500% in Q2 2021. Meanwhile, wholesale sales made up over
20% of the Company’s total sales in Q1 2021 as compared to just 12% in Q2 2020.
(ii)
This massive increase in wholesale vehicles adversely impacted
Carvana’s bottom line because, unbeknownst to investors, Carvana lost money on wholesale
vehicle sales, as detailed below:
(1)
Carvana sold cars at wholesale at significantly lower
prices than those sold at retail. Carvana, however, still incurred many of the same costs
involved in selling retail vehicles, including inbound transportation, reconditioning, and
other logistics. Thus, the lower average sales price combined with customary selling costs
resulted in significantly lower reported GPU on wholesale cars. In Q1 2021, wholesale GPU
was $806, 33% less than Retail GPU of $1,211.39
(2)
Although Carvana disclosed that it generated a modest
gross profit on wholesale vehicle sales, Defendants masked the actual profitability – or lack
thereof – of wholesale sales by: (i) excluding certain per-vehicle operations expenses from its
calculation of wholesale GPU; and (ii) not disclosing or quantifying these excluded costs
separately so investors could decipher overall profitability of wholesale sales on their own.
39 Carvana also did not earn any ancillary revenue or profit on wholesale vehicle sales, such
as warranty, insurance, or financing-related revenue. These “other” revenue streams
represented the vast majority of Carvana’s reported total GPU on its retail sales, but
generated $0 on wholesale sales.
0%
100%
200%
300%
400%
500%
Q1 20Q2 20Q3 20Q4 20Q1 21Q2 21Q3 21Q4 21Q1 22
Wholesale Growth Exceeded Retail Growth
y/y growth (retail cars sold)
y/y growth (wholesale cars sold)
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 103 of 332
- 100 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
These per-vehicle operations expenses, which were only separately broken out and
quantified for the first time after the Class Period, included material costs associated with
completing a wholesale sale, such as the full cost to ship a car to a wholesale auction site
and title and registration costs.40 Tellingly, Carvana admitted after the Class Period in its Q3
2023 Shareholder Letter and November 2, 2023 Cost Structure Details presentation that there
are “two key drivers” of “unit economics” (i.e., per-vehicle profitability): (i) cost of sales,
which were the expenses Carvana had always included in its reported GPU calculations; and
(ii) “operations expenses” per vehicle sold, which consisted of per-vehicle selling expenses
that were not included in Carvana’s reported GPU and had never been separately broken out
to investors.41 Instead, during the Class Period, these expenses that drove Carvana’s actual
profit per car sold, were lumped into and comingled with other SG&A expenses.42
Defendants later admitted, however, that these “operations expenses” were not typical fixed
or corporate overhead SG&A costs, conceding: “Operations expenses include the fulfillment,
customer service, and transaction expenses associated with completing retail and wholesale
vehicle sales.”43 Indeed, these vehicle selling expenses that Carvana had arbitrarily excluded
40 In its November 2, 2023 Cost Structure Details presentation, Carvana disclosed and
quantified retail and wholesale vehicle operations expense.
41 After disclosing and quantifying operations expenses for the first time in November
2023, Jenkins acknowledged that the additional disclosure “allows a much deeper dive into
the economics of Carvana.” Jenkins added: “we’ve really been focused on . . . improving the
unit economics of the business. In other words, driving down variable costs per car,” which
he confirmed included “operations expenses, which are the more variable component of our
selling, general and administrative expenses.” Jenkins continued: “And the reason that we’re
focused on that is obviously, the more gains we make on reducing variable cost per car
today, the more profitable our growth can be in the future . . . .” Jenkins also described
“driving down operational expenses per unit” as one of Carvana’s “profitability initiatives,
and in particular, . . . our unit economics initiatives.” Thus, according to Jenkins, operations
expenses were a critical component of assessing Carvana’s profitability.
42 While Defendants included per-vehicle operations expenses in SG&A expenses (rather
than GPU), they did not separately disclose or quantify these expenses in a manner that
would have allowed investors to decipher the overall per-vehicle profitability on their own.
In fact, as detailed in ¶254(d), the disclosures that Defendants did make regarding SG&A
expenses during Q1 2021 did more to mislead investors than to provide transparency around
Carvana’s true profitability per vehicle sold.
43 During the Class Period, Defendants described Carvana’s SG&A costs as “expenses
associated with advertising and providing customer service to customers, operating our
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 104 of 332
- 101 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
from its reported GPU included: (i) “customer care, multi-car logistics, last-mile delivery,
and other operations payroll” “associated with completing . . . wholesale vehicle sales”; (ii)
“non-payroll logistics expenses, including fuel, repairs and maintenance, and third-party
transport services” “associated with completing . . . wholesale vehicle sales”; and (iii)
“[t]ransaction and other expenses, including limited warranty, title and registration, and
finance platform expenses” “associated with completing . . . wholesale vehicle sales.”
Defendants’ concealment of “operations expenses” when describing Carvana’s wholesale
profitability was especially misleading in light of the fact that Defendants internally viewed
“the underlying costs of completing a sale” as a critical component of Carvana’s actual
wholesale vehicle profitability. Indeed, as Garcia Junior later acknowledged:
When we think about trying to aim for more profitable sales, what does that
mean? There’s certainly variability in the kind of gross profit associated with
different types of sales. . . .
And then I think there’s a number of other dynamics kind of across car
type, et cetera. There’s also kind of variation in the underlying costs of
completing a sale.
Because Defendants did not specifically break out operations expenses (i.e., the underlying
costs of completing a sale) and instead buried and comingled them among SG&A expenses,
investors could not calculate the actual overall profitability of Carvana’s wholesale vehicle
sales, like Garcia Junior did internally. As demonstrated in the chart below, upon
incorporating all of the estimated vehicle operations expenses (i.e., the actual costs Carvana
incurred in completing wholesale vehicle sales), Carvana lost money on wholesale vehicle
vending machines and hubs, operating our logistics and fulfillment network and other
corporate overhead expenses, including expenses associated with information technology,
product development, engineering, legal, accounting, finance, and business development.”
Moreover, while Defendants made vague Class Period disclosures indicating that certain of
expenses were recorded in SG&A, until the end of the Class Period Carvana never: (i) broke
out “operations expenses” to investors; (ii) described that these vehicle “operations
expenses” were direct per-vehicle costs “associated with completing . . . wholesale vehicle
sales”; nor (iii) described that, in addition to the selling costs included in Carvana’s GPU
metric, these additional per-vehicle “operations expenses,” were one of “two key drivers” of
per-unit profitability.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 105 of 332
- 102 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
sales every quarter during the Class Period.44 For example, in Q1 2021, Carvana’s average
selling price per wholesale vehicle was $9,217, while its total costs (cost of sales in addition
to estimated undisclosed operations expense) totaled $10,014, for a total loss of $797 per
vehicle.
(iii)
Defendants omitted that the surplus of wholesale cars resulting
from Carvana “intentionally buying lower quality cars” also led to significant logistics
constraints. For example, in Carvana’s May 2022 Operating Plan, Defendants admitted that
“[b]uying cars from customers growth” led to “additional wholesale volume” which, in turn,
led to “constraints in our nationwide logistics network.” Carvana further admitted that
“wholesale units acquired from customers have typically been transported to the nearest
Carvana IRC, generating additional vehicles [sic] moves and increased complexity in our
multi-car logistics network.”
(iv)
Defendants omitted that Carvana lacked critical infrastructure
necessary to process the huge surplus of wholesale vehicles. Consequently, Carvana was
forced to spend $3 billion to purchase ADESA, a nationwide wholesale auction house. In
describing his rationale for the purchase, Garcia Junior pointed to ADESA’s 56 brick-and-
mortar locations, consisting of 6.5 million square feet of buildings on more than 4,000 acres
44 Estimated operations expense per wholesale vehicle calculated as follows: total retail and
wholesale vehicle operations expenses in Q1 2021 divided by total retail and wholesale
vehicles sold in Q1 2021.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 106 of 332
- 103 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
of land, and as reported in a September 13, 2022 Vehicle Remarketing article, explained
“‘[y]ou need a place to stage the logistics, and there’s no better infrastructure than auctions
to serve that purpose.’” Indeed, Defendants later revealed that, prior to acquiring ADESA,
Carvana “frequently ha[d] our last mile delivery advocates pick these cars up from
customer’s homes and transport them to a market hub and then transport these cars long
distances from their origin to our nearest IRC where they are either reconditioned or held
until they are wholesaled.” Further, they admitted that “[w]ith this acquisition [of ADESA],
we can now position advocates at the auction locations and transport cars from our
customer’s driveways directly to an ADESA U.S. location without any additional moves.”
(c)
Analysts, who consistently focused on Carvana’s vehicle sourcing, were
duped by Carvana’s and Jenkins’s materially misleading misstatements and omissions. For
example, on May 7, 2021, Truist Securities raised its price target because the increase in
“[c]onsumer-sourced vehicles . . . attest[ed] to the success of this program. We believe the
company has honed its efforts to efficiently acquire vehicles from the consumer and
believe this represents a GPU growth opportunity over time.” On the same day, a DA
Davidson analyst report stated that “[t]otal GPU was $3,656, . . . the upside came . . . mostly
from wholesale profits.” Additionally, a Wedbush analyst report stated “the company’s
consolidated GPU still beat expectations because of . . . more wholesale units sold than
anticipated.”
199. Statement No. 13: Also on May 6, 2021, Carvana filed its Q1 2021 10-Q with
the SEC, signed by Jenkins and attaching SOX certifications signed by Garcia Junior and
Jenkins. The Q1 2021 10-Q stated:
Vehicle acquisition. . . . For vehicles sold to us through our website,
we use proprietary algorithms to determine an appropriate offer. We assess
vehicles on the basis of quality, inventory fit, consumer desirability, relative
value, expected reconditioning costs, and vehicle location to identify what we
believe represent the most in-demand and profitable vehicles to acquire for
inventory. We utilize a broad range of data sources, including proprietary site
data, and a variety of external data sources to support our assessments.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 107 of 332
- 104 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
200. As detailed below, Statement No. 13 made by Carvana, Garcia Junior, and
Jenkins omitted material facts that affirmatively created an impression of a state of affairs
that differed in a material way from the one that actually existed.
(a)
Contrary to Defendants’ statement that they “assess vehicles on the
basis of quality,” Defendants concealed from investors that Carvana “was intentionally
buying lower quality cars.” (Emphasis in original.)45
(b)
Defendants omitted that Carvana abruptly and drastically lowered or
disregarded entirely the Company’s purchasing and verification standards when buying cars
from customers to induce trade-ins and increase inventory, and thus, boost retail sales.
§VI.B. Indeed, by lowering or disregarding the Company’s purchasing and verification
standards, Carvana achieved a customer source ratio over 60% that quarter. CWs-1 and 4
confirmed that, to boost retail sales through trade-ins, Carvana bought cars from customers
that clearly did not meet its purchasing and verification standards. For example, CW-1
reported that Carvana had purchased a bullet-ridden car and a car missing its back seat. CW-
3 stated that Carvana had relaxed its standards to accept “all vehicles” and did not genuinely
inspect vehicles before purchasing them anymore, resulting in a lot of “trash vehicles.” CW-
11 said that Carvana was buying vehicles “sight unseen” and his/her team was not expected
to do much to verify the condition of the vehicles. CW-11 observed that Carvana did not
actually care about the condition of the vehicles and just wanted to “get as many cars as it
could.” In short, contrary to the statement that Carvana was “assess[ing] vehicles on the
basis of quality, inventory fit, consumer desirability, relative value, expected reconditioning
costs, and vehicle location to identify what we believe represent the most in-demand and
profitable vehicles to acquire for inventory,” Carvana largely disregarded the condition of the
vehicles they purchased as CWs-1, 3, 4, and 11 all reported.
45 Defendants judicially admitted this omitted fact by including it in a brief in this action.
ECF 58 at 12.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 108 of 332
- 105 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
(c)
As detailed below, Defendants omitted that buying low-quality cars
from customers adversely impacted the Company’s bottom line by flooding Carvana with
less profitable wholesale cars that had to be sold at a loss and crippling its logistics network.
(i)
Because Carvana was “intentionally buying lower quality cars,”
it was flooded with cars that did not meet its retail standards and, thus, had to be sold
wholesale. Indeed, upon lowering or disregarding its purchasing and verification standards,
Carvana’s wholesale growth began to far outpace its retail sales growth. As shown in the
charts below, immediately following Carvana’s 119% spike in the number of cars purchased
from customers during the last two quarters of FY 2020, its wholesale sales spiked by over
100% during Q1 2021 and by over 500% in Q2 2021. Meanwhile, wholesale sales made up
over 20% of the Company’s total sales in Q1 2021 compared to just 12% in Q2 2020.
(ii)
This massive increase in wholesale vehicles adversely impacted
Carvana’s bottom line because, unbeknownst to investors, Carvana lost money on wholesale
vehicle sales, as detailed below:
(1)
Carvana sold cars at wholesale at significantly lower
prices than those sold at retail. Carvana, however, still incurred many of the same costs
involved in selling retail vehicles, including inbound transportation, reconditioning, and
other logistics. Thus, the lower average sales price combined with customary selling costs
0%
100%
200%
300%
400%
500%
Q1 20Q2 20Q3 20Q4 20Q1 21Q2 21Q3 21Q4 21Q1 22
Wholesale Growth Exceeded Retail Growth
y/y growth (retail cars sold)
y/y growth (wholesale cars sold)
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 109 of 332
- 106 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
resulted in significantly lower reported GPU on wholesale cars. In Q1 2021, wholesale GPU
was $806, 33% less than Retail GPU of $1,211.46
(2)
Although Carvana disclosed that it generated a modest
gross profit on wholesale vehicle sales, Defendants masked the actual profitability – or lack
thereof – of wholesale sales by: (i) excluding certain per-vehicle operations expenses from its
calculation of wholesale GPU; and (ii) not disclosing or quantifying these excluded costs
separately so investors could decipher overall profitability of wholesale sales on their own.
These per-vehicle operations expenses, which were only separately broken out and
quantified for the first time after the Class Period, included material costs associated with
completing a wholesale sale, such as the full cost to ship a car to a wholesale auction site
and title and registration costs.47 Tellingly, Carvana admitted after the Class Period in its Q3
2023 Shareholder Letter and November 2, 2023 Cost Structure Details presentation that there
are “two key drivers” of “unit economics” (i.e., per-vehicle profitability): (i) cost of sales,
which were the expenses Carvana had always included in its reported GPU calculations; and
(ii) “operations expenses” per vehicle sold, which consisted of per-vehicle selling expenses
that were not included in Carvana’s reported GPU and had never been separately broken out
to investors.48 Instead, during the Class Period, these expenses that drove Carvana’s actual
46 Carvana also did not earn any ancillary revenue or profit on wholesale vehicle sales, such
as warranty, insurance, or financing-related revenue. These “other” revenue streams
represented the vast majority of Carvana’s reported total GPU on its retail sales, but
generated $0 on wholesale sales.
47 In its November 2, 2023 Cost Structure Details presentation, Carvana disclosed and
quantified retail and wholesale vehicle operations expense.
48 After disclosing and quantifying operations expenses for the first time in November
2023, Jenkins acknowledged that the additional disclosure “allows a much deeper dive into
the economics of Carvana.” Jenkins added: “we’ve really been focused on . . . improving the
unit economics of the business. In other words, driving down variable costs per car,” which
he confirmed included “operations expenses, which are the more variable component of our
selling, general and administrative expenses.” Jenkins continued: “And the reason that we’re
focused on that is obviously, the more gains we make on reducing variable cost per car
today, the more profitable our growth can be in the future . . . .” Jenkins also described
“driving down operational expenses per unit” as one of Carvana’s “profitability initiatives,
and in particular, . . . our unit economics initiatives.” Thus, according to Jenkins, operations
expenses were a critical component of assessing Carvana’s profitability.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 110 of 332
- 107 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
profit per car sold, were lumped into and comingled with other SG&A expenses.49
Defendants later admitted, however, that these “operations expenses” were not typical fixed
or corporate overhead SG&A costs, conceding: “Operations expenses include the fulfillment,
customer service, and transaction expenses associated with completing retail and wholesale
vehicle sales.”50 Indeed, these vehicle selling expenses that Carvana had arbitrarily excluded
from its reported GPU included: (i) “customer care, multi-car logistics, last-mile delivery,
and other operations payroll” “associated with completing . . . wholesale vehicle sales”; (ii)
“non-payroll logistics expenses, including fuel, repairs and maintenance, and third-party
transport services” “associated with completing . . . wholesale vehicle sales”; and (iii)
“[t]ransaction and other expenses, including limited warranty, title and registration, and
finance platform expenses” “associated with completing . . . wholesale vehicle sales.”
Defendants’ concealment of “operations expenses” when describing Carvana’s wholesale
profitability was especially misleading in light of the fact that Defendants internally viewed
“the underlying costs of completing a sale” as a critical component of Carvana’s actual
wholesale vehicle profitability. Indeed, as Garcia Junior later acknowledged:
When we think about trying to aim for more profitable sales, what does that
mean? There’s certainly variability in the kind of gross profit associated with
different types of sales. . . .
49 While Defendants included per-vehicle operations expenses in SG&A expenses (rather
than GPU), they did not separately disclose or quantify these expenses in a manner that
would have allowed investors to decipher the overall per-vehicle profitability on their own.
In fact, as detailed in ¶254(d), the disclosures that Defendants did make regarding SG&A
expenses during Q1 2021 did more to mislead investors than to provide transparency around
Carvana’s true profitability per vehicle sold.
50 During the Class Period, Defendants described Carvana’s SG&A costs as “expenses
associated with advertising and providing customer service to customers, operating our
vending machines and hubs, operating our logistics and fulfillment network and other
corporate overhead expenses, including expenses associated with information technology,
product development, engineering, legal, accounting, finance, and business development.”
Moreover, while Defendants made vague Class Period disclosures indicating that certain of
expenses were recorded in SG&A, until the end of the Class Period Carvana never: (i) broke
out “operations expenses” to investors; (ii) described that these vehicle “operations
expenses” were direct per-vehicle costs “associated with completing . . . wholesale vehicle
sales”; nor (iii) described that, in addition to the selling costs included in Carvana’s GPU
metric, these additional per-vehicle “operations expenses,” were one of “two key drivers” of
per-unit profitability.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 111 of 332
- 108 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
And then I think there’s a number of other dynamics kind of across car
type, et cetera. There’s also kind of variation in the underlying costs of
completing a sale.
Because Defendants did not specifically break out operations expenses (i.e., the underlying
costs of completing a sale) and instead buried and comingled them among SG&A expenses,
investors could not calculate the actual overall profitability of Carvana’s wholesale vehicle
sales, like Garcia Junior did internally. As demonstrated in the chart below, upon
incorporating all of the estimated vehicle operations expenses (i.e., the actual costs Carvana
incurred in completing wholesale vehicle sales), Carvana lost money on wholesale vehicle
sales every quarter during the Class Period.51 For example, in Q1 2021, Carvana’s average
selling price per wholesale vehicle was $9,217 while its total costs (cost of sales in addition
to estimated undisclosed operations expense) totaled $10,014, for a total loss of $797 per
vehicle.
(iii)
Defendants omitted that the surplus of wholesale cars resulting
from Carvana “intentionally buying lower quality cars” also led to significant logistics
constraints. For example, in Carvana’s May 2022 Operating Plan, Defendants admitted that
“[b]uying cars from customers growth” led to “additional wholesale volume” which, in turn,
led to “constraints in our nationwide logistics network.” Carvana further admitted that
51 Estimated operations expense per wholesale vehicle calculated as follows: total retail and
wholesale vehicle operations expenses in Q1 2021 divided by total retail and wholesale
vehicles sold in Q1 2021.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 112 of 332
- 109 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
“wholesale units acquired from customers have typically been transported to the nearest
Carvana IRC, generating additional vehicles [sic] moves and increased complexity in our
multi-car logistics network.”
(iv)
Defendants omitted that Carvana lacked critical infrastructure
necessary to process the huge surplus of wholesale vehicles. Consequently, Carvana was
forced to spend $3 billion to purchase ADESA, a nationwide wholesale auction house. In
describing his rationale for the purchase, Garcia Junior pointed to ADESA’s 56 brick-and-
mortar locations, consisting of 6.5 million square feet of buildings on more than 4,000 acres
of land, and as reported in a September 13, 2022 Vehicle Remarketing article, explained
“‘[y]ou need a place to stage the logistics, and there’s no better infrastructure than auctions
to serve that purpose.’” Indeed, Defendants later revealed that, prior to acquiring ADESA,
Carvana “frequently ha[d] our last mile delivery advocates pick these cars up from
customer’s homes and transport them to a market hub and then transport these cars long
distances from their origin to our nearest IRC where they are either reconditioned or held
until they are wholesaled.” Further, they admitted that “[w]ith this acquisition [of ADESA],
we can now position advocates at the auction locations and transport cars from our
customer’s driveways directly to an ADESA U.S. location without any additional moves.”
(d)
Analysts, focused on Carvana’s vehicle sourcing metric because of its
importance to Carvana’s purported profitability, were misled. For example, a May 7, 2021
Truist Securities analyst report shows it raised its price target because the increase in
“[c]onsumer-sourced vehicles . . . attest[ed] to the success of this program. We believe the
company has honed its efforts to efficiently acquire vehicles from the consumer and
believe this represents a GPU growth opportunity over time.”
201. Statement No. 14: On August 5, 2021, Carvana held a conference call for
analysts and investors to discuss the Q2 2021 results. On the Q2 2021 earnings call, Garcia
Junior responded to an analysts’ questions regarding Carvana’s pricing and profitability:
[W]e’ve built this channel of buying cars from customers that has performed
very well for us over a long period of time. We’ve made continual high-
quality progress in both the number of cars we’re buying, the cars we’re
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 113 of 332
- 110 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
buying relative to retail sales and the profits that we’re making on cars that
we buy from customers.
*
*
*
I think the thing that we would point to is that is the biggest by a long, long
way, is just buying cars from customers.
We just saw a massive, massive increase in the number of cars we’re
buying from customers and in the profitability of buying cars from
customers.
202. As detailed below, Statement No. 14 by Carvana and Garcia Junior omitted
material facts that affirmatively created an impression of a state of affairs that differed in a
material way from the one that actually existed.
(a)
While Defendants touted the “high-quality progress in both the number
of cars were buying” and “the profits that we’re making on cars that we buy from
customers,” they omitted that Carvana was “intentionally buying lower quality cars.”
Further, Defendants omitted that Carvana abruptly and drastically lowered or disregarded
entirely the Company’s purchasing and verification standards when buying cars from
customers in order to induce trade-ins and increase retail sales. §VI.B. CWs-1, 2, 3, 4, and
11 confirm that Carvana omitted that it had lowered or disregarded its purchasing and
verification standards and was intentionally acquiring low-quality vehicles. CW-4
confirmed that, to boost retail sales through trade-ins, Carvana bought cars from customers
that plainly did not meet its purchasing and verification standards. CW-3 stated that Carvana
had relaxed its standards to accept “all vehicles” and did not genuinely inspect vehicles
before purchasing them anymore, resulting in a lot of “trash vehicles.” CW-11 said that
Carvana was buying vehicles “sight unseen” and his/her team was not expected to do much
to verify the condition of the vehicles. CW-11 observed that Carvana did not actually care
about the condition of the vehicles and just wanted to “get as many cars as it could.” In
addition, CW-2 was overruled when voicing concerns about the quality of vehicles as
“leadership would say take it.”
(b)
As detailed below, while Defendants touted “the profitability of buying
cars from customers,” Defendants omitted that buying low-quality cars from customers
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 114 of 332
- 111 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
adversely impacted the Company’s bottom line by flooding Carvana with less profitable
wholesale cars and crippling its logistics network.
(i)
Because Carvana was “intentionally buying lower quality cars,”
it was flooded with cars that did not meet its retail standards and, thus, had to be sold
wholesale. Indeed, upon lowering or disregarding its purchasing and verification standards,
Carvana’s wholesale growth began to far outpace its retail sales growth. As shown in the
charts below, immediately following Carvana’s 119% spike in the number of cars purchased
from customers during the last two quarters of FY 2020, its wholesale sales spiked by over
100% during the next three quarters, including over 500% in Q2 2021. Meanwhile,
wholesale sales made up 30% of the Company’s total sales in Q2 2021, nearly double the
ratio at the beginning of the Class Period.
(ii)
This massive increase in wholesale vehicles adversely impacted
Carvana’s bottom line because, unbeknownst to investors, Carvana lost money on wholesale
vehicle sales, as detailed below:
(1)
Carvana sold cars at wholesale at significantly lower
prices than those sold at retail. Carvana, however, still incurred many of the same costs
involved in selling retail vehicles, including inbound transportation, reconditioning, and
other logistics. Thus, the lower average sales price combined with customary selling costs
0%
100%
200%
300%
400%
500%
Q1 20Q2 20Q3 20Q4 20Q1 21Q2 21Q3 21Q4 21Q1 22
Wholesale Growth Exceeded Retail Growth
y/y growth (retail cars sold)
y/y growth (wholesale cars sold)
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 115 of 332
- 112 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
resulted in significantly lower reported GPU on wholesale cars. In Q2 2021, wholesale GPU
was $1,254, almost 40% less than Retail GPU of $2,022.52
(2)
Although Carvana disclosed that it generated a modest
gross profit on wholesale vehicle sales, Defendants masked the actual profitability – or lack
thereof – of wholesale sales by: (i) excluding certain per-vehicle operations expenses from its
calculation of wholesale GPU; and (ii) not disclosing or quantifying these excluded costs
separately so investors could decipher overall profitability of wholesale sales on their own.
These per-vehicle operations expenses, which were only separately broken out and
quantified for the first time after the Class Period, included material costs associated with
completing a wholesale sale, such as the full cost to ship a car to a wholesale auction site
and title and registration costs.53 Tellingly, Carvana admitted after the Class Period in its Q3
2023 Shareholder Letter and November 2, 2023 Cost Structure Details presentation that there
are “two key drivers” of “unit economics” (i.e., per-vehicle profitability): (i) cost of sales,
which were the expenses Carvana had always included in its reported GPU calculations; and
(ii) “operations expenses” per vehicle sold, which consisted of per-vehicle selling expenses
that were not included in Carvana’s reported GPU and had never been separately broken out
to investors.54 Instead, during the Class Period, these expenses that drove Carvana’s actual
52 Carvana also did not earn any ancillary revenue or profit on wholesale vehicle sales, such
as warranty, insurance, or financing-related revenue. These “other” revenue streams
represented the vast majority of Carvana’s reported total GPU on its retail sales, but
generated $0 on wholesale sales.
53 In its November 2, 2023 Cost Structure Details presentation, Carvana disclosed and
quantified retail and wholesale vehicle operations expense.
54 After disclosing and quantifying operations expenses for the first time in November
2023, Jenkins acknowledged that the additional disclosure “allows a much deeper dive into
the economics of Carvana.” Jenkins added: “we’ve really been focused on . . . improving the
unit economics of the business. In other words, driving down variable costs per car,” which
he confirmed included “operations expenses, which are the more variable component of our
selling, general and administrative expenses.” Jenkins continued: “And the reason that we’re
focused on that is obviously, the more gains we make on reducing variable cost per car
today, the more profitable our growth can be in the future . . . .” Jenkins also described
“driving down operational expenses per unit” as one of Carvana’s “profitability initiatives,
and in particular, . . . our unit economics initiatives.” Thus, according to Jenkins, operations
expenses were a critical component of assessing Carvana’s profitability.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 116 of 332
- 113 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
profit per car sold, were lumped into and comingled with other SG&A expenses.55
Defendants later admitted, however, that these “operations expenses” were not typical fixed
or corporate overhead SG&A costs, conceding: “Operations expenses include the fulfillment,
customer service, and transaction expenses associated with completing retail and wholesale
vehicle sales.”56 Indeed, these vehicle selling expenses that Carvana had arbitrarily excluded
from its reported GPU included: (i) “customer care, multi-car logistics, last-mile delivery,
and other operations payroll” “associated with completing . . . wholesale vehicle sales”; (ii)
“non-payroll logistics expenses, including fuel, repairs and maintenance, and third-party
transport services” “associated with completing . . . wholesale vehicle sales”; and (iii)
“[t]ransaction and other expenses, including limited warranty, title and registration, and
finance platform expenses” “associated with completing . . . wholesale vehicle sales.”
Defendants’ concealment of “operations expenses” when describing Carvana’s wholesale
profitability was especially misleading in light of the fact that Defendants internally viewed
“the underlying costs of completing a sale” as a critical component of Carvana’s actual
wholesale vehicle profitability. Indeed, as Garcia Junior later acknowledged:
When we think about trying to aim for more profitable sales, what does that
mean? There’s certainly variability in the kind of gross profit associated with
different types of sales. . . .
55 While Defendants included per-vehicle operations expenses in SG&A expenses (rather
than GPU), they did not separately disclose or quantify these expenses in a manner that
would have allowed investors to decipher the overall per-vehicle profitability on their own.
In fact, as detailed in ¶256(d), the disclosures that Defendants did make regarding SG&A
expenses during Q2 2021 did more to mislead investors than to provide transparency around
Carvana’s true profitability per vehicle sold.
56 During the Class Period, Defendants described Carvana’s SG&A costs as “expenses
associated with advertising and providing customer service to customers, operating our
vending machines and hubs, operating our logistics and fulfillment network and other
corporate overhead expenses, including expenses associated with information technology,
product development, engineering, legal, accounting, finance, and business development.”
Moreover, while Defendants made vague Class Period disclosures indicating that certain of
expenses were recorded in SG&A, until the end of the Class Period Carvana never: (i) broke
out “operations expenses” to investors; (ii) described that these vehicle “operations
expenses” were direct per-vehicle costs “associated with completing . . . wholesale vehicle
sales”; nor (iii) described that, in addition to the selling costs included in Carvana’s GPU
metric, these additional per-vehicle “operations expenses,” were one of “two key drivers” of
per-unit profitability.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 117 of 332
- 114 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
And then I think there’s a number of other dynamics kind of across car
type, et cetera. There’s also kind of variation in the underlying costs of
completing a sale.
Because Defendants did not specifically break out operations expenses (i.e., the underlying
costs of completing a sale) and instead buried and comingled them among SG&A expenses,
investors could not calculate the actual overall profitability of Carvana’s wholesale vehicle
sales, like Garcia Junior did internally. As demonstrated in the chart below, upon
incorporating all of the estimated vehicle operations expenses (i.e., the actual costs Carvana
incurred in completing wholesale vehicle sales), Carvana lost money on wholesale vehicle
sales every quarter during the Class Period.57 For example, in Q2 2021, Carvana’s average
selling price per wholesale vehicle was $11,838 while its total costs (cost of sales in addition
to estimated undisclosed operations expense) totaled $12,082, for a total loss of $244 per
vehicle.
(iii)
Defendants omitted that the surplus of wholesale cars resulting
from Carvana “intentionally buying lower quality cars” also led to significant logistics
constraints. For example, in Carvana’s May 2022 Operating Plan, Defendants admitted that
“[b]uying cars from customers growth” led to “additional wholesale volume” which, in turn,
led to “constraints in our nationwide logistics network.” Carvana further admitted that
57 Estimated operations expense per wholesale vehicle calculated as follows: total retail and
wholesale vehicle operations expenses in Q2 2021 divided by total retail and wholesale
vehicles sold in Q2 2021.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 118 of 332
- 115 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
“wholesale units acquired from customers have typically been transported to the nearest
Carvana IRC, generating additional vehicles [sic] moves and increased complexity in our
multi-car logistics network.”
(iv)
Defendants omitted that Carvana lacked critical infrastructure
necessary to process the huge surplus of wholesale vehicles. Consequently, Carvana was
forced to spend $3 billion to purchase ADESA, a nationwide wholesale auction house. In
describing his rationale for the purchase, Garcia Junior pointed to ADESA’s 56 brick-and-
mortar locations, consisting of 6.5 million square feet of buildings on more than 4,000 acres
of land, and as reported in a September 13, 2022 Vehicle Remarketing article, explained
“‘[y]ou need a place to stage the logistics, and there’s no better infrastructure than auctions
to serve that purpose.’” Indeed, Defendants later revealed that, prior to acquiring ADESA,
Carvana “frequently ha[d] our last mile delivery advocates pick these cars up from
customer’s homes and transport them to a market hub and then transport these cars long
distances from their origin to our nearest IRC where they are either reconditioned or held
until they are wholesaled.” Further, they admitted that “[w]ith this acquisition [of ADESA],
we can now position advocates at the auction locations and transport cars from our
customer’s driveways directly to an ADESA U.S. location without any additional moves.”
(c)
Defendants omitted that as a result of seeing “a massive increase in the
number of cars we’re buying from customers,” Carvana’s wholesale vehicles sales far
outpaced the growth in retail sales between late 2020 and the end of the Class Period, as
described above.
(d)
Defendants omitted that at this time, Carvana’s glut of low-quality
inventory was causing “parking constraints,” “load imbalances,” and “increased the number
of vehicles moved and total vehicle miles traveled, adding to costs and logistics network
complexity.” Indeed, Defendants would later admit that two of three primary drivers of the
Company’s logistics costs and constraints were Carvana’s increased inventory and increased
wholesale volume from buying cars from customers.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 119 of 332
- 116 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
203. Statement No. 15: Also on August 5, 2021, Carvana filed its Q2 2021 10-Q
with the SEC, signed by Jenkins and attaching SOX certifications signed by Garcia Junior
and Jenkins. The Q2 2021 10-Q stated:
Vehicle acquisition. . . . For vehicles sold to us through our website,
we use proprietary algorithms to determine an appropriate offer. We assess
vehicles on the basis of quality, inventory fit, consumer desirability, relative
value, expected reconditioning costs, and vehicle location to identify what we
believe represent the most in-demand and profitable vehicles to acquire for
inventory. We utilize a broad range of data sources, including proprietary site
data, and a variety of external data sources to support our assessments.
204. As detailed below, Statement No. 15 by Carvana, Garcia Junior, and Jenkins
omitted material facts that affirmatively created an impression of a state of affairs that
differed in a material way from the one that actually existed.
(a)
Contrary to Defendants’ statement that they “assess vehicles on the
basis of quality,” Defendants concealed from investors that Carvana “was intentionally
buying lower quality cars.” (Emphasis in original.)58
(b)
Defendants omitted that Carvana had drastically lowered or disregarded
entirely the Company’s purchasing and verification standards when buying cars from
customers to induce trade-ins and increase inventory, and thus, boost retail sales. §VI.B.
CW-4 confirmed that, to boost retail sales through trade-ins, Carvana bought cars from
customers that plainly did not meet its purchasing and verification standards. CW-3 stated
that Carvana had relaxed its standards to accept “all vehicles” and did not genuinely inspect
vehicles before purchasing them anymore, resulting in a lot of “trash vehicles.” CW-11 said
that Carvana was buying vehicles “sight unseen” and his/her team was not expected to do
much to verify the condition of the vehicles. CW-11 observed that Carvana did not actually
care about the condition of the vehicles and just wanted to “get as many cars as it could.” In
addition, CW-2 was overruled when voicing concerns about the quality of vehicles as
“leadership would say take it.” In short, contrary to the statement that Carvana was
“assess[ing] vehicles on the basis of quality, inventory fit, consumer desirability, relative
58 Defendants judicially admitted this omitted fact by including it in a brief in this action.
ECF 58 at 12.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 120 of 332
- 117 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
value, expected reconditioning costs, and vehicle location to identify what we believe
represent the most in-demand and profitable vehicles to acquire for inventory,” Carvana
largely disregarded the condition of the vehicles they purchased as CWs-1, 2, 3, 4, and 11
reported.
(c)
As detailed below, Defendants omitted that buying low-quality cars
from customers adversely impacted the Company’s bottom line by flooding Carvana with
less profitable wholesale cars that had to be sold at a loss and crippling its logistics network.
(i)
Because Carvana was “intentionally buying lower quality cars,”
it was flooded with cars that did not meet its retail standards and, thus, had to be sold
wholesale. Indeed, upon lowering or disregarding its purchasing and verification standards,
Carvana’s wholesale growth began to far outpace its retail sales growth. As shown in the
charts below, immediately following Carvana’s 119% spike in the number of cars purchased
from customers during the last two quarters of FY 2020, its wholesale sales spiked by over
100% over the next three quarters, including over 500% in Q2 2021. Meanwhile, wholesale
sales made up 30% of the Company’s total sales in Q2 2021, nearly double the ratio at the
beginning of the Class Period.
(ii)
This massive increase in wholesale vehicles adversely impacted
Carvana’s bottom line because, unbeknownst to investors, Carvana lost money on wholesale
vehicle sales, as detailed below:
0%
100%
200%
300%
400%
500%
Q1 20Q2 20Q3 20Q4 20Q1 21Q2 21Q3 21Q4 21Q1 22
Wholesale Growth Exceeded Retail Growth
y/y growth (retail cars sold)
y/y growth (wholesale cars sold)
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 121 of 332
- 118 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
(1)
Carvana sold cars at wholesale at significantly lower
prices than those sold at retail. Carvana, however, still incurred many of the same costs
involved in selling retail vehicles, including inbound transportation, reconditioning, and
other logistics. Thus, the lower average sales price combined with customary selling costs
resulted in significantly lower reported GPU on wholesale cars. In Q2 2021, wholesale GPU
was $1,254, nearly 40% less than Retail GPU of $2,022.59
(2)
Although Carvana disclosed that it generated a modest
gross profit on wholesale vehicle sales, Defendants masked the actual profitability – or lack
thereof – of wholesale sales by: (i) excluding certain per-vehicle operations expenses from its
calculation of wholesale GPU; and (ii) not disclosing or quantifying these excluded costs
separately so investors could decipher overall profitability of wholesale sales on their own.
These per-vehicle operations expenses, which were only separately broken out and
quantified for the first time after the Class Period, included material costs associated with
completing a wholesale sale, such as the full cost to ship a car to a wholesale auction site
and title and registration costs.60 Tellingly, Carvana admitted after the Class Period in its Q3
2003 Shareholder Letter and November 2, 2023 Cost Structure Details presentation that there
are “two key drivers” of “unit economics” (i.e., per-vehicle profitability): (i) cost of sales,
which were the expenses Carvana had always included in its reported GPU calculations; and
(ii) “operations expenses” per vehicle sold, which consisted of per-vehicle selling expenses
that were not included in Carvana’s reported GPU and had never been separately broken out
to investors.61 Instead, during the Class Period, these expenses that drove Carvana’s actual
59 Carvana also did not earn any ancillary revenue or profit on wholesale vehicle sales, such
as warranty, insurance, or financing-related revenue. These “other” revenue streams
represented the vast majority of Carvana’s reported total GPU on its retail sales, but
generated $0 on wholesale sales.
60 In its November 2, 2023 Cost Structure Details presentation, Carvana disclosed and
quantified retail and wholesale vehicle operations expense.
61 After disclosing and quantifying operations expenses for the first time in November
2023, Jenkins acknowledged that the additional disclosure “allows a much deeper dive into
the economics of Carvana.” Jenkins added: “we’ve really been focused on . . . improving the
unit economics of the business. In other words, driving down variable costs per car,” which
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 122 of 332
- 119 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
profit per car sold, were lumped into and comingled with other SG&A expenses.62
Defendants later admitted, however, that these “operations expenses” were not typical fixed
or corporate overhead SG&A costs, conceding: “Operations expenses include the fulfillment,
customer service, and transaction expenses associated with completing retail and wholesale
vehicle sales.”63 Indeed, these vehicle selling expenses that Carvana had arbitrarily excluded
from its reported GPU included: (i) “customer care, multi-car logistics, last-mile delivery,
and other operations payroll” “associated with completing . . . wholesale vehicle sales”; (ii)
“non-payroll logistics expenses, including fuel, repairs and maintenance, and third-party
transport services” “associated with completing . . . wholesale vehicle sales”; and (iii)
“[t]ransaction and other expenses, including limited warranty, title and registration, and
finance platform expenses” “associated with completing . . . wholesale vehicle sales.”
Defendants’ concealment of “operations expenses” when describing Carvana’s wholesale
profitability was especially misleading in light of the fact that Defendants internally viewed
he confirmed included “operations expenses, which are the more variable component of our
selling, general and administrative expenses.” Jenkins continued: “And the reason that we’re
focused on that is obviously, the more gains we make on reducing variable cost per car
today, the more profitable our growth can be in the future . . . .” Jenkins also described
“driving down operational expenses per unit” as one of Carvana’s “profitability initiatives,
and in particular, . . . our unit economics initiatives.” Thus, according to Jenkins, operations
expenses were a critical component of assessing Carvana’s profitability.
62 While Defendants included per-vehicle operations expenses in SG&A expenses (rather
than GPU), they did not separately disclose or quantify these expenses in a manner that
would have allowed investors to decipher the overall per-vehicle profitability on their own.
In fact, as detailed in ¶256(d), the disclosures that Defendants did make regarding SG&A
expenses during Q2 2021 did more to mislead investors than to provide transparency around
Carvana’s true profitability per vehicle sold.
63 During the Class Period, Defendants described Carvana’s SG&A costs as “expenses
associated with advertising and providing customer service to customers, operating our
vending machines and hubs, operating our logistics and fulfillment network and other
corporate overhead expenses, including expenses associated with information technology,
product development, engineering, legal, accounting, finance, and business development.”
Moreover, while Defendants made vague Class Period disclosures indicating that certain of
expenses were recorded in SG&A, until the end of the Class Period Carvana never: (i) broke
out “operations expenses” to investors; (ii) described that these vehicle “operations
expenses” were direct per-vehicle costs “associated with completing . . . wholesale vehicle
sales”; nor (iii) described that, in addition to the selling costs included in Carvana’s GPU
metric, these additional per-vehicle “operations expenses,” were one of “two key drivers” of
per-unit profitability.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 123 of 332
- 120 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
“the underlying costs of completing a sale” as a critical component of Carvana’s actual
wholesale vehicle profitability. Indeed, as Garcia Junior later acknowledged:
When we think about trying to aim for more profitable sales, what does that
mean? There’s certainly variability in the kind of gross profit associated with
different types of sales. . . .
And then I think there’s a number of other dynamics kind of across car
type, et cetera. There’s also kind of variation in the underlying costs of
completing a sale.
Because Defendants did not specifically break out operations expenses (i.e., the underlying
costs of completing a sale) and instead buried and comingled them among SG&A expenses,
investors could not calculate the actual overall profitability of Carvana’s wholesale vehicle
sales, like Garcia Junior did internally. As demonstrated in the chart below, upon
incorporating all of the estimated vehicle operations expenses (i.e., the actual costs Carvana
incurred in completing wholesale vehicle sales), Carvana lost money on wholesale vehicle
sales every quarter during the Class Period.64 For example, in Q2 2021, Carvana’s average
selling price per wholesale vehicle was $11,838 while its total costs (cost of sales in addition
to estimated undisclosed operations expense) totaled $12,081, for a total loss of $244 per
vehicle.
64 Estimated operations expense per wholesale vehicle calculated as follows: total retail and
wholesale vehicle operations expenses in Q2 2021 divided by total retail and wholesale
vehicles sold in Q2 2021.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 124 of 332
- 121 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
(iii)
Defendants omitted that the surplus of wholesale cars resulting
from Carvana “intentionally buying lower quality cars” also led to significant logistics
constraints. For example, in Carvana’s May 2022 Operating Plan, Defendants admitted that
“[b]uying cars from customers growth” led to “additional wholesale volume” which, in turn,
led to “constraints in our nationwide logistics network.” Carvana further admitted that
“wholesale units acquired from customers have typically been transported to the nearest
Carvana IRC, generating additional vehicles [sic] moves and increased complexity in our
multi-car logistics network.”
(iv)
Defendants omitted that Carvana lacked critical infrastructure
necessary to process the huge surplus of wholesale vehicles. Consequently, Carvana was
forced to spend $3 billion to purchase ADESA, a nationwide wholesale auction house. In
describing his rationale for the purchase, Garcia Junior pointed to ADESA’s 56 brick-and-
mortar locations, consisting of 6.5 million square feet of buildings on more than 4,000 acres
of land, and as reported in a September 13, 2022 Vehicle Remarketing article, explained
“‘[y]ou need a place to stage the logistics, and there’s no better infrastructure than auctions
to serve that purpose.’” Indeed, Defendants later revealed that, prior to acquiring ADESA,
Carvana “frequently ha[d] our last mile delivery advocates pick these cars up from
customer’s homes and transport them to a market hub and then transport these cars long
distances from their origin to our nearest IRC where they are either reconditioned or held
until they are wholesaled.” Further, they admitted that “[w]ith this acquisition [of ADESA],
we can now position advocates at the auction locations and transport cars from our
customer’s driveways directly to an ADESA U.S. location without any additional moves.”
205. Statement No. 16: On November 4, 2021, Carvana held a conference call for
analysts and investors to discuss the Q3 2021 results. On the Q3 2021 earnings call, Garcia
Junior responded to an analyst’s question regarding Carvana purchasing cars from
customers:
So I think first and foremost, on buying cars from customers, I think that’s –
necessity would not be the way that I would describe it. I think the way that I
would describe it is it’s better. It allows us to provide a high-quality
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 125 of 332
- 122 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
experience to a customer that we’re buying a car from, and then it gets us
access to a higher-quality pool of inventory that is, on average, more
profitable. And so we want to do as much of that as we possibly can. We
want to build the business capacity to be able to handle as much of that as
we possibly can.
206. As detailed below, Statement No. 16 by Carvana and Garcia Junior omitted
material facts that affirmatively created an impression of a state of affairs that differed in a
material way from the one that actually existed.
(a)
While Defendants touted positive information concerning buying cars
from customers, including that “it gets us access to a higher-quality pool of inventory,”
Defendants omitted that Carvana had drastically lowered or disregarded entirely its
purchasing and verification standards when buying cars from customers to induce trade-ins
and increase inventory. §VI.B.
(b)
While Defendants touted positive information about buying from
customers, including that “it gets us access to a higher-quality pool of inventory,”
Defendants omitted that Carvana “was intentionally buying lower quality cars.” (Emphasis
in original.)65
(c)
As detailed below, while Defendants touted the benefits of buying cars
from customers, including that “it gets us access to a higher-quality pool of inventory,”
Defendants omitted that buying low-quality cars from customers adversely impacted
Carvana’s bottom line by flooding Carvana with less profitable wholesale cars that had to be
sold at a loss and crippling its logistics network.
(i)
Because Carvana was “intentionally buying lower quality cars,”
it was flooded with cars that did not meet its retail standards and, thus, had to be sold
wholesale. Indeed, upon lowering or disregarding its purchasing and verification standards,
Carvana’s wholesale growth began to far outpace its retail sales growth. As shown in the
charts below, immediately following Carvana’s 119% spike in the number of cars purchased
from customers during the last two quarters of FY 2020, its wholesale sales spiked by over
65 Defendants judicially admitted this omitted fact by including it in a brief in this action.
ECF 58 at 12.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 126 of 332
- 123 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
100% during the next four quarters, including over 225% in Q3 2021. Meanwhile,
wholesale sales made up over 30% of the Company’s total sales in Q3 2021, nearly double
the ratio at the beginning of the Class Period.
(ii)
This massive increase in wholesale vehicles adversely impacted
Carvana’s bottom line because, unbeknownst to investors, Carvana lost money on wholesale
vehicle sales, as detailed below:
(1)
Carvana sold cars at wholesale at significantly lower
prices than those sold at retail. Carvana, however, still incurred many of the same costs
involved in selling retail vehicles, including inbound transportation, reconditioning, and
other logistics. Thus, the lower average sales price combined with customary selling costs
resulted in significantly lower reported GPU on wholesale cars. In Q3 2021, wholesale GPU
was $936, nearly 50% less than Retail GPU of $1,769.66
(2)
Although Carvana disclosed that it generated a modest
gross profit on wholesale vehicle sales, Defendants masked the actual profitability – or lack
thereof – of wholesale sales by: (i) excluding certain per-vehicle operations expenses from its
calculation of wholesale GPU; and (ii) not disclosing or quantifying these excluded costs
66 Carvana also did not earn any ancillary revenue or profit on wholesale vehicle sales, such
as warranty, insurance, or financing-related revenue. These “other” revenue streams
represented the vast majority of Carvana’s reported total GPU on its retail sales, but
generated $0 on wholesale sales.
0%
100%
200%
300%
400%
500%
Q1 20Q2 20Q3 20Q4 20Q1 21Q2 21Q3 21Q4 21Q1 22
Wholesale Growth Exceeded Retail Growth
y/y growth (retail cars sold)
y/y growth (wholesale cars sold)
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 127 of 332
- 124 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
separately so investors could decipher overall profitability of wholesale sales on their own.
These per-vehicle operations expenses, which were only separately broken out and
quantified for the first time after the Class Period, included material costs associated with
completing a wholesale sale, such as the full cost to ship a car to a wholesale auction site
and title and registration costs.67 Tellingly, Carvana admitted after the Class Period in its Q3
2023 Shareholder Letter and November 2, 2023 Cost Structure Details presentation that there
are “two key drivers” of “unit economics” (i.e., per-vehicle profitability): (i) cost of sales,
which were the expenses Carvana had always included in its reported GPU calculations; and
(ii) “operations expenses” per vehicle sold, which consisted of per-vehicle selling expenses
that were not included in Carvana’s reported GPU and had never been separately broken out
to investors.68 Instead, during the Class Period, these expenses that drove Carvana’s actual
profit per car sold, were lumped into and comingled with other SG&A expenses.69
Defendants later admitted, however, that these “operations expenses” were not typical fixed
or corporate overhead SG&A costs, conceding: “Operations expenses include the fulfillment,
customer service, and transaction expenses associated with completing retail and wholesale
67 In its November 2, 2023 Cost Structure Details presentation, Carvana disclosed and
quantified retail and wholesale vehicle operations expense.
68 After disclosing and quantifying operations expenses for the first time in November
2023, Jenkins acknowledged that the additional disclosure “allows a much deeper dive into
the economics of Carvana.” Jenkins added: “we’ve really been focused on . . . improving the
unit economics of the business. In other words, driving down variable costs per car,” which
he confirmed included “operations expenses, which are the more variable component of our
selling, general and administrative expenses.” Jenkins continued: “And the reason that we’re
focused on that is obviously, the more gains we make on reducing variable cost per car
today, the more profitable our growth can be in the future . . . .” Jenkins also described
“driving down operational expenses per unit” as one of Carvana’s “profitability initiatives,
and in particular, . . . our unit economics initiatives.” Thus, according to Jenkins, operations
expenses were a critical component of assessing Carvana’s profitability.
69 While Defendants included per-vehicle operations expenses in SG&A expenses (rather
than GPU), they did not separately disclose or quantify these expenses in a manner that
would have allowed investors to decipher the overall per-vehicle profitability on their own.
In fact, as detailed in ¶258(e), the disclosures that Defendants did make regarding SG&A
expenses during Q3 2021 did more to mislead investors than to provide transparency around
Carvana’s true profitability per vehicle sold.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 128 of 332
- 125 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
vehicle sales.”70 Indeed, these vehicle selling expenses that Carvana had arbitrarily excluded
from its reported GPU included: (i) “customer care, multi-car logistics, last-mile delivery,
and other operations payroll” “associated with completing . . . wholesale vehicle sales”; (ii)
“non-payroll logistics expenses, including fuel, repairs and maintenance, and third-party
transport services” “associated with completing . . . wholesale vehicle sales”; and (iii)
“[t]ransaction and other expenses, including limited warranty, title and registration, and
finance platform expenses” “associated with completing . . . wholesale vehicle sales.”
Defendants’ concealment of “operations expenses” when describing Carvana’s wholesale
profitability was especially misleading in light of the fact that Defendants internally viewed
“the underlying costs of completing a sale” as a critical component of Carvana’s actual
wholesale vehicle profitability. Indeed, as Garcia Junior later acknowledged:
When we think about trying to aim for more profitable sales, what does that
mean? There’s certainly variability in the kind of gross profit associated with
different types of sales. . . .
And then I think there’s a number of other dynamics kind of across car
type, et cetera. There’s also kind of variation in the underlying costs of
completing a sale.
Because Defendants did not specifically break out operations expenses (i.e., the underlying
costs of completing a sale) and instead buried and comingled them among SG&A expenses,
investors could not calculate the actual overall profitability of Carvana’s wholesale vehicle
sales, like Garcia Junior did internally. As demonstrated in the chart below, upon
incorporating all of the estimated vehicle operations expenses (i.e., the actual costs Carvana
incurred in completing wholesale vehicle sales), Carvana lost money on wholesale vehicle
70 During the Class Period, Defendants described Carvana’s SG&A costs as “expenses
associated with advertising and providing customer service to customers, operating our
vending machines and hubs, operating our logistics and fulfillment network and other
corporate overhead expenses, including expenses associated with information technology,
product development, engineering, legal, accounting, finance, and business development.”
Moreover, while Defendants made vague Class Period disclosures indicating that certain of
expenses were recorded in SG&A, until the end of the Class Period Carvana never: (i) broke
out “operations expenses” to investors; (ii) described that these vehicle “operations
expenses” were direct per-vehicle costs “associated with completing . . . wholesale vehicle
sales”; nor (iii) described that, in addition to the selling costs included in Carvana’s GPU
metric, these additional per-vehicle “operations expenses,” were one of “two key drivers” of
per-unit profitability.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 129 of 332
- 126 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
sales every quarter during the Class Period.71 For example, in Q3 2021, Carvana’s average
selling price per wholesale vehicle was $10,995 while its total costs (cost of sales in addition
to estimated undisclosed operations expense) totaled $11,750, for a total loss of $755 per
vehicle.
(iii)
Defendants omitted that the surplus of wholesale cars resulting
from Carvana “intentionally buying lower quality cars” also led to significant logistics
constraints. For example, in Carvana’s May 2022 Operating Plan, Defendants admitted that
“[b]uying cars from customers growth” led to “additional wholesale volume” which, in turn,
led to “constraints in our nationwide logistics network.” Carvana also admitted that
“wholesale units acquired from customers have typically been transported to the nearest
Carvana IRC, generating additional vehicles [sic] moves and increased complexity in our
multi-car logistics network.”
(iv)
Defendants omitted that Carvana lacked critical infrastructure
necessary to process the huge surplus of wholesale vehicles. Consequently, Carvana was
forced to spend $3 billion to purchase ADESA, a nationwide wholesale auction house. In
describing his rationale for the purchase, Garcia Junior pointed to ADESA’s 56 brick-and-
mortar locations, consisting of 6.5 million square feet of buildings on more than 4,000 acres
71 Estimated operations expense per wholesale vehicle calculated as follows: total retail and
wholesale vehicle operations expenses in Q3 2021 divided by total retail and wholesale
vehicles sold in Q3 2021.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 130 of 332
- 127 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
of land, and as reported in a September 13, 2022 Vehicle Remarketing article, explained
“‘[y]ou need a place to stage the logistics, and there’s no better infrastructure than auctions
to serve that purpose.’” Indeed, Defendants later revealed that, prior to acquiring ADESA,
Carvana “frequently ha[d] our last mile delivery advocates pick these cars up from
customer’s homes and transport them to a market hub and then transport these cars long
distances from their origin to our nearest IRC where they are either reconditioned or held
until they are wholesaled.” Further, they admitted that “[w]ith this acquisition [of ADESA],
we can now position advocates at the auction locations and transport cars from our
customer’s driveways directly to an ADESA U.S. location without any additional moves.”
207. Statement No. 17: Also on November 4, 2021, Carvana filed its Q3 2021 10-Q
with the SEC, signed by Jenkins and attaching SOX certifications signed by Garcia Junior
and Jenkins. The Q3 2021 10-Q stated:
Vehicle acquisition. . . . For vehicles sold to us through our website, we
use proprietary algorithms to determine an appropriate offer. We assess
vehicles on the basis of quality, inventory fit, consumer desirability, relative
value, expected reconditioning costs, and vehicle location to identify what we
believe represent the most in-demand and profitable vehicles to acquire for
inventory. We utilize a broad range of data sources, including proprietary site
data, and a variety of external data sources to support our assessments.
208. As detailed below, Statement No. 17 by Carvana, Garcia Junior, and Jenkins
omitted material facts that affirmatively created an impression of a state of affairs that
differed in a material way from the one that actually existed.
(a)
Contrary to Defendants’ statement that they “assess vehicles on the
basis of quality,” Defendants concealed from investors that Carvana “was intentionally
buying lower quality cars.” (Emphasis in original.)72
(b)
Defendants omitted that Carvana had drastically lowered or disregarded
entirely the Company’s purchasing and verification standards when buying cars from
customers to induce trade-ins and increase inventory, and thus, boost retail sales. §VI.B.
CW-4 confirmed that, to boost retail sales through trade-ins, Carvana bought cars from
72 Defendants judicially admitted this omitted fact by including it in a brief in this action.
ECF 58 at 12.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 131 of 332
- 128 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
customers that plainly did not meet its purchasing and verification standards. CW-11 said
that Carvana was buying vehicles “sight unseen” and his/her team was not expected to do
much to verify the condition of the vehicles. CW-11 observed that Carvana did not actually
care about the condition of the vehicles and just wanted to “get as many cars as it could.”
CW-10 corroborated this account, explaining that Carvana was buying cars at breakneck
speed and not adequately inspecting them. In addition, CW-2 was overruled when voicing
concerns about the quality of vehicles as “leadership would say take it.” In short, contrary to
the statement that Carvana was “assess[ing] vehicles on the basis of quality, inventory fit,
consumer desirability, relative value, expected reconditioning costs, and vehicle location to
identify what we believe represent the most in-demand and profitable vehicles to acquire for
inventory,” Carvana largely disregarded the condition of the vehicles they purchased as
CWs-1, 2, 3, 4, and 11 reported.
(c)
As detailed below, Defendants omitted that buying low-quality cars
from customers adversely impacted the Company’s bottom line by flooding Carvana with
less profitable wholesale cars that had to be sold at a loss and crippling its logistics network.
(i)
Because Carvana was “intentionally buying lower quality cars,”
it was flooded with cars that did not meet its retail standards and, thus, had to be sold
wholesale. Indeed, upon lowering or disregarding its purchasing and verification standards,
Carvana’s wholesale growth began to far outpace its retail sales growth. As shown in the
charts below, immediately following Carvana’s 119% spike in the number of cars purchased
from customers in late 2020, its wholesale sales spiked by over 100% in each of the next four
quarters, including over 225% in Q3 2021. Meanwhile, wholesale sales made up over 30%
of the Company’s total sales in Q3 2021, nearly double the ratio at the beginning of the Class
Period.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 132 of 332
- 129 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
(ii)
This massive increase in wholesale vehicles adversely impacted
Carvana’s bottom line because, unbeknownst to investors, Carvana lost money on wholesale
vehicle sales, as detailed below:
(1)
Carvana sold cars at wholesale at significantly lower
prices than those sold at retail. Carvana, however, still incurred many of the same costs
involved in selling retail vehicles, including inbound transportation, reconditioning, and
other logistics. Thus, the lower average sales price combined with customary selling costs
resulted in significantly lower reported GPU on wholesale cars. In Q3 2021, wholesale GPU
was $936, almost 50% lower than Retail GPU of $1,769.73
(2)
Although Carvana disclosed that it generated a modest
gross profit on wholesale vehicle sales, Defendants masked the actual profitability – or lack
thereof – of wholesale sales by: (i) excluding certain per-vehicle operations expenses from its
calculation of wholesale GPU; and (ii) not disclosing or quantifying these excluded costs
separately so investors could decipher overall profitability of wholesale sales on their own.
These per-vehicle operations expenses, which were only separately broken out and
quantified for the first time after the Class Period, included material costs associated with
73 Carvana also did not earn any ancillary revenue or profit on wholesale vehicle sales, such
as warranty, insurance, or financing-related revenue. These “other” revenue streams
represented the vast majority of Carvana’s reported total GPU on its retail sales, but
generated $0 on wholesale sales.
0%
100%
200%
300%
400%
500%
Q1 20Q2 20Q3 20Q4 20Q1 21Q2 21Q3 21Q4 21Q1 22
Wholesale Growth Exceeded Retail Growth
y/y growth (retail cars sold)
y/y growth (wholesale cars sold)
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 133 of 332
- 130 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
completing a wholesale sale, such as the full cost to ship a car to a wholesale auction site
and title and registration costs.74 Tellingly, Carvana admitted after the Class Period in its Q3
2023 Shareholder Letter and November 2, 2023 Cost Structure Details presentation that there
are “two key drivers” of “unit economics” (i.e., per-vehicle profitability): (i) cost of sales,
which were the expenses Carvana had always included in its reported GPU calculations; and
(ii) “operations expenses” per vehicle sold, which consisted of per-vehicle selling expenses
that were not included in Carvana’s reported GPU and had never been separately broken out
to investors.75 Instead, during the Class Period, these expenses that drove Carvana’s actual
profit per car sold, were lumped into and comingled with other SG&A expenses.76
Defendants later admitted, however, that these “operations expenses” were not typical fixed
or corporate overhead SG&A costs, conceding: “Operations expenses include the fulfillment,
customer service, and transaction expenses associated with completing retail and wholesale
vehicle sales.”77 Indeed, these vehicle selling expenses that Carvana had arbitrarily excluded
74 In its November 2, 2023 Cost Structure Details presentation, Carvana disclosed and
quantified retail and wholesale vehicle operations expense.
75 After disclosing and quantifying operations expenses for the first time in November
2023, Jenkins acknowledged that the additional disclosure “allows a much deeper dive into
the economics of Carvana.” Jenkins added: “we’ve really been focused on . . . improving the
unit economics of the business. In other words, driving down variable costs per car,” which
he confirmed included “operations expenses, which are the more variable component of our
selling, general and administrative expenses.” Jenkins continued: “And the reason that we’re
focused on that is obviously, the more gains we make on reducing variable cost per car
today, the more profitable our growth can be in the future . . . .” Jenkins also described
“driving down operational expenses per unit” as one of Carvana’s “profitability initiatives,
and in particular, . . . our unit economics initiatives.” Thus, according to Jenkins, operations
expenses were a critical component of assessing Carvana’s profitability.
76 While Defendants included per-vehicle operations expenses in SG&A expenses (rather
than GPU), they did not separately disclose or quantify these expenses in a manner that
would have allowed investors to decipher the overall per-vehicle profitability on their own.
In fact, as detailed in ¶258(e), the disclosures that Defendants did make regarding SG&A
expenses during Q3 2021 did more to mislead investors than to provide transparency around
Carvana’s true profitability per vehicle sold.
77 During the Class Period, Defendants described Carvana’s SG&A costs as “expenses
associated with advertising and providing customer service to customers, operating our
vending machines and hubs, operating our logistics and fulfillment network and other
corporate overhead expenses, including expenses associated with information technology,
product development, engineering, legal, accounting, finance, and business development.”
Moreover, while Defendants made vague Class Period disclosures indicating that certain of
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 134 of 332
- 131 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
from its reported GPU included: (i) “customer care, multi-car logistics, last-mile delivery,
and other operations payroll” “associated with completing . . . wholesale vehicle sales”; (ii)
“non-payroll logistics expenses, including fuel, repairs and maintenance, and third-party
transport services” “associated with completing . . . wholesale vehicle sales”; and (iii)
“[t]ransaction and other expenses, including limited warranty, title and registration, and
finance platform expenses” “associated with completing . . . wholesale vehicle sales.”
Defendants’ concealment of “operations expenses” when describing Carvana’s wholesale
profitability was especially misleading in light of the fact that Defendants internally viewed
“the underlying costs of completing a sale” as a critical component of Carvana’s actual
wholesale vehicle profitability. Indeed, as Garcia Junior later acknowledged:
When we think about trying to aim for more profitable sales, what does that
mean? There’s certainly variability in the kind of gross profit associated with
different types of sales. . . .
And then I think there’s a number of other dynamics kind of across car
type, et cetera. There’s also kind of variation in the underlying costs of
completing a sale.
Because Defendants did not specifically break out operations expenses (i.e., the underlying
costs of completing a sale) and instead buried and comingled them among SG&A expenses,
investors could not calculate the actual overall profitability of Carvana’s wholesale vehicle
sales, like Garcia Junior did internally. As demonstrated in the chart below, upon
incorporating all of the estimated vehicle operations expenses (i.e., the actual costs Carvana
incurred in completing wholesale vehicle sales), Carvana lost money on wholesale vehicle
sales every quarter during the Class Period.78 For example, in Q3 2021, Carvana’s average
selling price per wholesale vehicle was $10,995 while its total costs (cost of sales in addition
expenses were recorded in SG&A, until the end of the Class Period Carvana never: (i) broke
out “operations expenses” to investors; (ii) described that these vehicle “operations
expenses” were direct per-vehicle costs “associated with completing . . . wholesale vehicle
sales”; nor (iii) described that, in addition to the selling costs included in Carvana’s GPU
metric, these additional per-vehicle “operations expenses,” were one of “two key drivers” of
per-unit profitability.
78 Estimated operations expense per wholesale vehicle calculated as follows: total retail and
wholesale vehicle operations expenses in Q3 2021 divided by total retail and wholesale
vehicles sold in Q3 2021.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 135 of 332
- 132 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
to estimated undisclosed operations expense) totaled $11,750, for a total loss of $755 per
vehicle.
(iii)
Defendants omitted that the surplus of wholesale cars resulting
from Carvana “intentionally buying lower quality cars” also led to significant logistics
constraints. For example, in Carvana’s May 2022 Operating Plan, Defendants admitted that
“[b]uying cars from customers growth” led to “additional wholesale volume” which, in turn,
led to “constraints in our nationwide logistics network.” Carvana further admitted that
“wholesale units acquired from customers have typically been transported to the nearest
Carvana IRC, generating additional vehicles [sic] moves and increased complexity in our
multi-car logistics network.”
(iv)
Defendants omitted that Carvana lacked critical infrastructure
necessary to process the huge surplus of wholesale vehicles. Consequently, Carvana was
forced to spend $3 billion to purchase ADESA, a nationwide wholesale auction house. In
describing his rationale for the purchase, Garcia Junior pointed to ADESA’s 56 brick-and-
mortar locations, consisting of 6.5 million square feet of buildings on more than 4,000 acres
of land, and as reported in a September 13, 2022 Vehicle Remarketing article, explained
“‘[y]ou need a place to stage the logistics, and there’s no better infrastructure than auctions
to serve that purpose.’” Indeed, Defendants later revealed that, prior to acquiring ADESA,
Carvana “frequently ha[d] our last mile delivery advocates pick these cars up from
customer’s homes and transport them to a market hub and then transport these cars long
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 136 of 332
- 133 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
distances from their origin to our nearest IRC where they are either reconditioned or held
until they are wholesaled.” Further, they admitted that “[w]ith this acquisition [of ADESA],
we can now position advocates at the auction locations and transport cars from our
customer’s driveways directly to an ADESA U.S. location without any additional moves.”
209. Statement No. 18: On February 24, 2022, Carvana filed its 2021 10-K with the
SEC, signed by Garcia Junior and Jenkins and attaching SOX certifications signed by Garcia
Junior and Jenkins. The 2021 10-K stated:
Vehicle acquisition. . . . For vehicles sold to us through our website,
we use proprietary algorithms to determine an appropriate offer. We assess
vehicles on the basis of quality, inventory fit, consumer desirability, relative
value, expected reconditioning costs, and vehicle location to identify what we
believe represent the most in-demand and profitable vehicles to acquire for
inventory. We utilize a broad range of data sources, including proprietary site
data, and a variety of external data sources to support our assessments.
210. As detailed below, Statement No. 18 made by Carvana, Garcia Junior, and
Jenkins omitted material facts that affirmatively created an impression of a state of affairs
that differed in a material way from the one that actually existed.
(a)
Contrary to Defendants’ statement that they “assess vehicles on the
basis of quality,” Defendants concealed from investors that Carvana “was intentionally
buying lower quality cars.” (Emphasis in original.)79
(b)
Defendants omitted that Carvana had drastically lowered or disregarded
entirely the Company’s purchasing and verification standards when buying cars from
customers in order to induce trade-ins and increase inventory, and thus, boost retail sales.
§VI.B. CW-4 confirmed that, to boost retail sales through trade-ins, Carvana bought cars
from customers that plainly did not meet its purchasing and verification standards. CW-11
said that Carvana was buying vehicles “sight unseen” and his/her team was not expected to
do much to verify the condition of the vehicles. CW-11 observed that Carvana did not
actually care about the condition of the vehicles and just wanted to “get as many cars as it
could.” CW-10 corroborated this account, explaining that Carvana was buying cars at
79 Defendants judicially admitted this omitted fact by including it in a brief in this action.
ECF 58 at 12.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 137 of 332
- 134 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
breakneck speed and not adequately inspecting them. In addition, CW-2 was overruled
when voicing concerns about the quality of vehicles as “leadership would say take it.”
Similarly, CW-5 stated that there was only a 50/50 chance that a vehicle would be
reappraised if it did not match the seller’s description. In short, contrary to the statement that
Carvana was “assess[ing] vehicles on the basis of quality, inventory fit, consumer
desirability, relative value, expected reconditioning costs, and vehicle location to identify
what we believe represent the most in-demand and profitable vehicles to acquire for
inventory,” Carvana largely disregarded the condition of the vehicles they purchased as
CWs-1, 2, 3, 4, 5, 10, and 11 reported.
(c)
As detailed below, Defendants omitted that buying low-quality cars
from customers adversely impacted the Company’s bottom line by flooding Carvana with
less profitable wholesale cars that had to be sold at a loss and crippling its logistics network.
(i)
Because Carvana was “intentionally buying lower quality cars,”
it was flooded with cars that did not meet its retail standards and, thus, had to be sold
wholesale. Indeed, upon lowering or disregarding its purchasing and verification standards,
Carvana’s wholesale growth began to far outpace its retail sales growth. As shown in the
charts below, immediately following Carvana’s spike in the number of cars purchased from
customers during late 2020 and early 2021, its wholesale sales spiked by over 100% in each
of the next five quarters, including 115% in Q4 2021. Meanwhile, wholesale sales made up
roughly 30% of the Company’s total sales in Q4 2021, nearly double the ratio at the
beginning of the Class Period.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 138 of 332
- 135 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
(ii)
This massive increase in wholesale vehicles adversely impacted
Carvana’s bottom line because, unbeknownst to investors, Carvana lost money on wholesale
vehicle sales, as detailed below:
(1)
Carvana sold cars at wholesale at significantly lower
prices than those sold at retail. Carvana, however, still incurred many of the same costs
involved in selling retail vehicles, including inbound transportation, reconditioning, and
other logistics. Thus, the lower average sales price combined with customary selling costs
resulted in significantly lower reported GPU on wholesale cars. In Q4 2021, wholesale GPU
was $1,326 while Retail GPU was $1,495.80
(2)
Although Carvana disclosed that it generated a modest
gross profit on wholesale vehicle sales, Defendants masked the actual profitability – or lack
thereof – of wholesale sales by: (i) excluding certain per-vehicle operations expenses from its
calculation of wholesale GPU; and (ii) not disclosing or quantifying these excluded costs
separately so investors could decipher overall profitability of wholesale sales on their own.
These per-vehicle operations expenses, which were only separately broken out and
quantified for the first time after the Class Period, included material costs associated with
80 Carvana also did not earn any ancillary revenue or profit on wholesale vehicle sales, such
as warranty, insurance, or financing-related revenue. These “other” revenue streams
represented the vast majority of Carvana’s reported total GPU on its retail sales, but
generated $0 on wholesale sales.
0%
100%
200%
300%
400%
500%
Q1 20Q2 20Q3 20Q4 20Q1 21Q2 21Q3 21Q4 21Q1 22
Wholesale Growth Exceeded Retail Growth
y/y growth (retail cars sold)
y/y growth (wholesale cars sold)
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 139 of 332
- 136 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
completing a wholesale sale, such as the full cost to ship a car to a wholesale auction site
and title and registration costs.81 Tellingly, Carvana admitted after the Class Period in its Q3
2023 Shareholder Letter and November 2, 2023 Cost Structure Details presentation that there
are “two key drivers” of “unit economics” (i.e., per-vehicle profitability): (i) cost of sales,
which were the expenses Carvana had always included in its reported GPU calculations; and
(ii) “operations expenses” per vehicle sold, which consisted of per-vehicle selling expenses
that were not included in Carvana’s reported GPU and had never been separately broken out
to investors.82 Instead, during the Class Period, these expenses that drove Carvana’s actual
profit per car sold, were lumped into and comingled with other SG&A expenses.83
Defendants later admitted, however, that these “operations expenses” were not typical fixed
or corporate overhead SG&A costs, conceding: “Operations expenses include the fulfillment,
customer service, and transaction expenses associated with completing retail and wholesale
vehicle sales.”84 Indeed, these vehicle selling expenses that Carvana had arbitrarily excluded
81 In its November 2, 2023 Cost Structure Details presentation, Carvana disclosed and
quantified retail and wholesale vehicle operations expense.
82 After disclosing and quantifying operations expenses for the first time in November
2023, Jenkins acknowledged that the additional disclosure “allows a much deeper dive into
the economics of Carvana.” Jenkins added: “we’ve really been focused on . . . improving the
unit economics of the business. In other words, driving down variable costs per car,” which
he confirmed included “operations expenses, which are the more variable component of our
selling, general and administrative expenses.” Jenkins continued: “And the reason that we’re
focused on that is obviously, the more gains we make on reducing variable cost per car
today, the more profitable our growth can be in the future . . . .” Jenkins also described
“driving down operational expenses per unit” as one of Carvana’s “profitability initiatives,
and in particular, . . . our unit economics initiatives.” Thus, according to Jenkins, operations
expenses were a critical component of assessing Carvana’s profitability.
83 While Defendants included per-vehicle operations expenses in SG&A expenses (rather
than GPU), they did not separately disclose or quantify these expenses in a manner that
would have allowed investors to decipher the overall per-vehicle profitability on their own.
In fact, as detailed in ¶260(e), the disclosures that Defendants did make regarding SG&A
expenses during Q4 2021 did more to mislead investors than to provide transparency around
Carvana’s true profitability per vehicle sold.
84 During the Class Period, Defendants described Carvana’s SG&A costs as “expenses
associated with advertising and providing customer service to customers, operating our
vending machines and hubs, operating our logistics and fulfillment network and other
corporate overhead expenses, including expenses associated with information technology,
product development, engineering, legal, accounting, finance, and business development.”
Moreover, while Defendants made vague Class Period disclosures indicating that certain of
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 140 of 332
- 137 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
from its reported GPU included: (i) “customer care, multi-car logistics, last-mile delivery,
and other operations payroll” “associated with completing . . . wholesale vehicle sales”; (ii)
“non-payroll logistics expenses, including fuel, repairs and maintenance, and third-party
transport services” “associated with completing . . . wholesale vehicle sales”; and (iii)
“[t]ransaction and other expenses, including limited warranty, title and registration, and
finance platform expenses” “associated with completing . . . wholesale vehicle sales.”
Defendants’ concealment of “operations expenses” when describing Carvana’s wholesale
profitability was especially misleading in light of the fact that Defendants internally viewed
“the underlying costs of completing a sale” as a critical component of Carvana’s actual
wholesale vehicle profitability. Indeed, as Garcia Junior later acknowledged:
When we think about trying to aim for more profitable sales, what does that
mean? There’s certainly variability in the kind of gross profit associated with
different types of sales. . . .
And then I think there’s a number of other dynamics kind of across car
type, et cetera. There’s also kind of variation in the underlying costs of
completing a sale.
Because Defendants did not specifically break out operations expenses (i.e., the underlying
costs of completing a sale) and instead buried and comingled them among SG&A expenses,
investors could not calculate the actual overall profitability of Carvana’s wholesale vehicle
sales, like Garcia Junior did internally. As demonstrated in the chart below, upon
incorporating all of the estimated vehicle operations expenses (i.e., the actual costs Carvana
incurred in completing wholesale vehicle sales), Carvana lost money on wholesale vehicle
sales every quarter during the Class Period.85 For example, in Q4 2021, Carvana’s average
selling price per wholesale vehicle was $12,211 while its total costs (cost of sales in addition
expenses were recorded in SG&A, until the end of the Class Period Carvana never: (i) broke
out “operations expenses” to investors; (ii) described that these vehicle “operations
expenses” were direct per-vehicle costs “associated with completing . . . wholesale vehicle
sales”; nor (iii) described that, in addition to the selling costs included in Carvana’s GPU
metric, these additional per-vehicle “operations expenses,” were one of “two key drivers” of
per-unit profitability.
85 Estimated operations expense per wholesale vehicle calculated as follows: total retail and
wholesale vehicle operations expenses in Q4 2021 divided by total retail and wholesale
vehicles sold in Q4 2021.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 141 of 332
- 138 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
to estimated undisclosed operations expense) totaled $12,830, for a total loss of $619 per
vehicle.
(iii)
Defendants omitted that the surplus of wholesale cars resulting
from Carvana “intentionally buying lower quality cars” also led to significant logistics
constraints. For example, in Carvana’s May 2022 Operating Plan, Defendants admitted that
“[b]uying cars from customers growth” led to “additional wholesale volume” which, in turn,
led to “constraints in our nationwide logistics network.” Carvana further admitted that
“wholesale units acquired from customers have typically been transported to the nearest
Carvana IRC, generating additional vehicles [sic] moves and increased complexity in our
multi-car logistics network.”
(iv)
Defendants omitted that Carvana lacked critical infrastructure
necessary to process the huge surplus of wholesale vehicles. Consequently, Carvana was
forced to spend $3 billion to purchase ADESA, a nationwide wholesale auction house. In
describing his rationale for the purchase, Garcia Junior pointed to ADESA’s 56 brick-and-
mortar locations, consisting of 6.5 million square feet of buildings on more than 4,000 acres
of land, and as reported in a September 13, 2022 Vehicle Remarketing article, as reported in a
September 13, 2022 Vehicle Remarketing article, explained “‘[y]ou need a place to stage the
logistics, and there’s no better infrastructure than auctions to serve that purpose.’” Indeed,
Defendants later revealed that, prior to acquiring ADESA, Carvana “frequently ha[d] our last
mile delivery advocates pick these cars up from customer’s homes and transport them to a
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 142 of 332
- 139 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
market hub and then transport these cars long distances from their origin to our nearest IRC
where they are either reconditioned or held until they are wholesaled.” Further, they
admitted that “[w]ith this acquisition [of ADESA], we can now position advocates at the
auction locations and transport cars from our customer’s driveways directly to an ADESA
U.S. location without any additional moves.”
C.
Defendants’ Materially Misleading Statements and Omissions
Concerning Retail Unit Sales Growth
211. Carvana’s “Objective #1” leading up to and throughout the Class Period was to
“Grow Retail Units.” Indeed, Defendants reminded investors that retail sales growth was
“Objective #1” in every SEC filing and shareholder letter. Accordingly, during the Class
Period, Defendants consistently touted Carvana’s retail unit sales growth and the non-
fraudulent drivers of such growth. In truth, however, Carvana’s sales growth was also
substantially fueled by the following artifices: (i) sales in violation of title and registration
laws and regulations; (ii) “less profitable sales” in “markets with lower profitability due to
long distance from inventory”; (iii) trade-in sales resulting from Carvana’s lowered
purchasing and verification standards; and (iv) “sales that were less profitable in the
immediate period.” Thus, by omitting that Carvana’s retail sales growth was fueled in part
by Defendants’ machinations, Defendants’ statements created the materially misleading
impression that Carvana’s growth was organic and sustainable.
212. Statement No. 19: On August 5, 2021, after the market closed, Carvana
published its Q2 2021 Shareholder Letter, signed by Garcia Junior and Jenkins. The Q2
2021 Shareholder Letter touted Carvana’s retail unit sales growth and its drivers:
[R]etail units sold totaled 107,815 growing 96% YoY vs. 55,098 in Q2 2020,
and up 145% vs. Q2 2019. Q2 revenue grew to $3.336 billion, up 198% YoY
from $1.118 billion, and up 238% vs. Q2 2019. Year-over-year retail unit
and revenue growth were . . . primarily driven by strong demand for our
offering this year.
213. As detailed below, Statement No. 19 by Carvana, Garcia Junior, and Jenkins
omitted material facts that affirmatively created an impression of a state of affairs that
differed in a material way from the one that actually existed.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 143 of 332
- 140 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
(a)
Although Defendants chose to speak about the drivers of retail unit
growth, they did not disclose that this growth was primarily driven by a number of
unsustainable artifices. Specifically, and discussed in more detail below, the stated retail unit
sales growth was primarily fueled by: (i) sales to customers awaiting proper title and
registration processing; (ii) “less profitable sales” in “markets with lower profitability due to
long distance from inventory”; (iii) trade-in sales resulting from Carvana’s lowered
purchasing and verification standards; and (iv) “sales that were less profitable in the
immediate period.” For example:
(i)
Defendants omitted that Carvana routinely sold cars to customers
before it held title to those cars and faster than it could get them registered to their new
owners. §VI.D. Indeed, as of July 2020, the Board discussed this issue at every one of its
meetings that Garcia Junior and Jenkins attended until at least December 2021. See, e.g.,
Exs. 1-2. Further, CWs-3, 4, 8, 9, 10, and 11 universally reported that Carvana would sell
these vehicles before it received title. Specifically, CW-3 estimated that Carvana only had
around half of the titles for the vehicles it sold wholesale at the time of sale. Further, he/she
noted that, after the Class Period began, a “fire started” because he/she was flooded with
calls regarding title issues. CW-11 reported that there was a drawer of documentation for
vehicles that had been sold by Carvana, but which Carvana could not register since the titles
were missing. CW-8 said that Carvana’s practice was simply to assume that titles for
vehicles could be acquired later if they were not available at the time of purchase. CW-4
said it was well known internally that title issues were a problem. CW-4 also said there was
a Google spreadsheet that tracked vehicles without titles for his/her entire IRC and “there
were a ridiculous amount of cars on the list.” The CWs’ accounts are corroborated by state
investigations, findings, and suspensions of Carvana’s dealership licenses in North Carolina,
Michigan, Illinois, Arizona, Pennsylvania, Florida, Texas, Maryland, Georgia, and Ohio. In
fact: (1) the Ohio BMV had investigated Carvana, suspended its temporary tag issuance
privileges for all Ohio locations, and subjected the Company to increased oversight in
December 2020; (2) MDOS had fined Carvana thousands of dollars and placed the Company
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 144 of 332
- 141 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
on an 18-month probation – a probation they would repeatedly violate – following months of
investigation and a meeting with Carvana executives in March of 2021; and (3) North
Carolina had fined and suspended Carvana’s license for six months following an
investigation and a July 2021 hearing. In addition, Carvana had settled a complaint with
Florida for title violations, and Texas had fined Carvana thousands of dollars for violations
related to title and registration no later than October 22, 2021. These omitted facts were
material as Defendants admitted that Carvana’s retail sales growth was significantly hindered
by Defendants’ belated enactment of “buffer” periods near the end of the Class Period to
allow for sufficient time to secure title.
(ii)
Defendants omitted that Carvana’s retail sales were comprised of
less profitable, and thus, unsustainable sales in far-flung markets in connection with
Carvana’s nationwide expansion. §VI.C. For example, in May 2021, Carvana opened its
first five new markets in the Pacific Northwest, which were unprofitable, as Garcia Junior
acknowledged: “[W]e didn’t have an inspection center. . . . So we would just kind of incur
all the cost associated with shipping the cars out there . . . .” These costs were material.
Defendants would later admit that sales in markets more than 200 miles from an IRC
incurred at least $750 more per car in logistics and related expenses. Indeed, near the end of
the Class Period, Defendants were forced to reduce sales in these markets. As one misled
commentator realized near the end of the Class Period, “[Carvana’s expansion] was growth
for growth’s sake and not necessarily growth to improve profitability.”
(iii)
Defendants omitted that Carvana’s retail sales growth was
supported by its unsustainable action to lower the Company’s purchasing and verification
standards when buying cars from customers. §VI.B. Indeed, CW-4 confirmed that, to boost
retail sales through trade-ins, Carvana bought cars from customers that plainly did not meet
its purchasing and verification standards. CW-3 stated that Carvana had relaxed its standards
to accept “all vehicles” and did not genuinely inspect vehicles before purchasing them
anymore, resulting in a lot of “trash vehicles.” CW-11 said that Carvana was buying
vehicles “sight unseen” and his/her team was not expected to do much to verify the condition
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 145 of 332
- 142 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
of the vehicles. CW-11 observed that Carvana did not actually care about the condition of
the vehicles and just wanted to “get as many cars as it could.” In addition, CW-2 was
overruled when voicing concerns about the quality of vehicles as “leadership would say take
it.” This omitted fact was material as it adversely impacted Carvana’s bottom line, flooded
Carvana with low-quality cars that had to be sold via the wholesale market at a loss, and
crippled its logistics network. In fact, Carvana was forced to meter (i.e., not sell) some of its
purchases of cars from customers because of the strain it placed on its logistics network,
which would hinder Carvana’s retail growth.
(iv)
Defendants omitted that Carvana’s retail sales growth was
unsustainable because Carvana “frequently acquired sales that were less profitable in the
immediate period.” Unbeknownst to investors, Defendants’ sales’ calculation “incorporated
the value of future sales” because Carvana singularly focused on growth without regard to
profitability. These sales were not sustainable, however, as Defendants would be forced to
forego these critical sales near the end of the Class Period to survive.
(b)
The omitted facts above also would have been viewed by the reasonable
investor as significant for the reasons stated herein, and because Carvana’s “Objective #1: [Is
to] Grow Retail Units and Revenue” and “Objective #2: [Is to] Increase Total Gross Profit
Per Unit.” Thus, a reasonable investor would have wanted to know that Carvana’s growth
was inorganic and unsustainable, and the Company would have to begin foregoing growth to
survive. As a Needham & Company analyst noted near the end of the Class Period, “‘[t]he
[only] path forward for Carvana is to sell as many cars as possible, but to do so on a
profitable basis, versus prior it was more about selling as many cars as possible.’”
214. Statement No. 20: On February 24, 2022, Carvana held a conference call for
analysts and investors to discuss the Q4 2021 results. On the Q4 2021 earnings call, Jenkins
stated: “[R]etail units sold totaled 425,237, an increase of 74%, making us the fastest used
automotive retailer to sell over 400,000 vehicles in 1 year. . . . Our exceptional growth in
2021 was driven by rapid growth within our market cohorts.”
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 146 of 332
- 143 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
215. As detailed below, Statement No. 20 by Carvana and Jenkins omitted material
facts that affirmatively created an impression of a state of affairs that differed in a material
way from the one that actually existed.
(a)
Although Defendants chose to speak about the drivers of retail unit
growth, they did not disclose that this growth was primarily driven by a number of
unsustainable artifices. Specifically, and discussed in more detail below, the stated retail unit
sales growth was primarily fueled by: (i) sales to customers awaiting proper title and
registration processing; (ii) “less profitable sales” in “markets with lower profitability due to
long distance from inventory”; (iii) trade-in sales resulting from Carvana’s lowered
purchasing and verification standards; (iv) sales, pursuant to a pass-through sales agreement
with Garcia Senior’s DriveTime, that lacked economic substance for Carvana; and (iv) “sales
that were less profitable in the immediate period.” For example:
(i)
Defendants omitted that Carvana routinely sold cars to customers
before it held title to those cars and faster than it could get them registered to their new
owners. §VI.D. Indeed, as of July 2020, the Board discussed this issue at every one of its
meetings that Garcia Junior and Jenkins attended until at least December 2021. See, e.g.,
Exs. 1-2. Further, CWs-3, 4, 8, 9, 10, and 11 universally reported that Carvana would sell
these vehicles before they received title. Specifically, CW-10 described instances where
titles on vehicles Carvana sold could not be obtained for over a year, which meant the buyers
were unable to drive their cars. CW-8 said that Carvana’s practice was simply to assume
that titles for vehicles could be acquired later if they were not available at the time of
purchase. CW-4 said it was well known internally that title issues were a problem. CW-4
also said there was a Google spreadsheet that tracked vehicles without titles for his/her entire
IRC and there were “a ridiculous amount of cars on the list.” CW-9 said, beginning in Q1
2022, registration delays amounted to roughly half of the calls the team had handled and that
employees informed Garcia Junior of this in January 2022. The CWs’ accounts are
corroborated by state investigations, findings, and suspensions of Carvana’s dealership
licenses in North Carolina, Michigan, Illinois, Arizona, Pennsylvania, Florida, Texas,
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 147 of 332
- 144 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Maryland, Georgia, and Ohio. In fact: (1) the Ohio BMV had investigated Carvana,
suspended its temporary tag issuance privileges for all Ohio locations, and subjected the
Company to increased oversight in December 2020; (2) following additional meetings
between MDOS and Carvana, Carvana “admi[tted] . . . several more violations of
[Michigan’s] code,” faced thousands of dollars in additional fines, and had its 18-month
probation extended on February 7, 2022; and (3) Illinois had begun investigating Carvana in
February 2022 for issuing out-of-state temporary registration permits and for failing to
transfer titles in a timely manner in violation of its laws. These omitted facts were material
as Defendants admitted that Carvana’s retail sales growth was significantly hindered by
Defendants’ belated enactment of “buffer” periods near the end of the Class Period to allow
for sufficient time to secure title.
(ii)
Defendants omitted that Carvana’s retail sales growth was largely
driven by less profitable sales in far-flung markets in connection with Carvana’s nationwide
expansion. §VI.C. In fact, as alleged herein, in FY 2021, 100% of the new markets that
Defendants entered were over 200 miles from an existing IRC, which cost Carvana an
additional $750 more per retail car sold. Further, these sales in far-flung markets stretched
Carvana’s logistics and reconditioning network to the brink, and thus, forced Defendants to
spend millions of dollars on costly third-party logistics and reconditioning providers to make
up for Carvana’s internal deficiencies in these critical functions. After the Class Period, in
November 2023, Defendants separately broke out “[r]econditioning and [i]nbound
[t]ransport [c]ost [p]er [r]etail [u]nit” for the first time, revealing that, between mid-2021 and
Q4 2021, these costs had spiked $369 per vehicle. For example, less than three months after
this statement was made, Defendants admitted that, beginning on or around August 2021,
Carvana had begun outsourcing logistics, including increasing “third-party reconditioning
locations” and “add[ing] many third-party reconditioning locations.” Then, near the end of
the Class Period, Defendants emphasized that, to curtail logistics expenses, Carvana needed
to quickly “[r]educe third-party production [i.e., reconditioning] volume,” “[r]educe third-
party inbound transport share,” “[i]n-sourc[e] third-party pickups” for Carvana’s “[l]ast-
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 148 of 332
- 145 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
mile delivery,” and “[r]educe third-party shipping.” Carvana also acknowledged that
“location growth” had led to “the addition of many third-party reconditioning locations.”
These costs were material. Indeed, near the end of the Class Period, Defendants were forced
to reduce sales in these markets. As one misled commentator realized near the end of the
Class Period, “[Carvana’s expansion] was growth for growth’s sake and not necessarily
growth to improve profitability.”
(iii)
Defendants omitted that Carvana’s retail sales growth was
supported by its unsustainable action to lower the Company’s purchasing and verification
standards when buying cars from customers. §VI.B. Indeed, CW-4 confirmed that, to boost
retail sales through trade-ins, Carvana bought cars from customers that plainly did not meet
its purchasing and verification standards. CW-11 said that Carvana was buying vehicles
“sight unseen” and his/her team was not expected to do much to verify the condition of the
vehicles. CW-11 observed that Carvana did not actually care about the condition of the
vehicles and just wanted to “get as many cars as it could.” CW-10 corroborated this account,
explaining that Carvana was buying cars at breakneck speed and not adequately inspecting
them. In addition, CW-2 was overruled when voicing concerns about the quality of vehicles
as “leadership would say take it.” Similarly, CW-5 stated that there was only a 50/50 chance
that a vehicle would be reappraised if it did not match the seller’s description. This omitted
fact was material as it adversely impacted Carvana’s bottom line, flooded Carvana with low-
quality cars that had to be sold via the wholesale market at a loss, and crippled its logistics
network. In fact, Carvana was forced to meter (i.e., not sell) some of its purchases of cars
from customers because of the strain it placed on its logistics network, which would hinder
Carvana’s retail growth.
(iv)
Defendants omitted that Carvana “frequently acquired sales that
were less profitable in the immediate period.” Defendants omitted that Carvana’s internal
sales’ calculation “incorporated the value of future sales” as Carvana singularly focused on
growth without regard to profitability. These sales were material and, near the end of the
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 149 of 332
- 146 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Class Period, Defendants would be forced to forego these sales to survive, which adversely
impacted Carvana’s retail sales growth.
(v)
Defendants omitted that they entered a sham pass-through
arrangement with DriveTime. §VI.A. Carvana later admitted in inconsistent, piecemeal, and
buried disclosures that, for many of these sales, Carvana kept none of the proceeds, but still
recorded the revenue and retail unit sale on its books. These vacuous sales were material as
they made up 169% of Carvana’s reported sequential growth in Q4 2021.
(b)
The omitted facts above also would have been viewed by the reasonable
investor as significant for the reasons stated herein, and because Carvana’s “Objective #1: [Is
to] Grow Retail Units and Revenue” and “Objective #2: [Is to] Increase Total Gross Profit
Per [Retail] Unit.” Thus, a reasonable investor would have wanted to know that Carvana’s
growth was inorganic and unsustainable, and the Company would have to begin foregoing
growth to survive. As a Needham & Company analyst noted near the end of the Class
Period, “‘[t]he [only] path forward for Carvana is to sell as many cars as possible, but to do
so on a profitable basis, versus prior it was more about selling as many cars as possible.’”
D.
Defendants’ Materially False and Misleading Statements
Regarding Expansion and Logistics Infrastructure
216. Leading up to and throughout the Class Period, Defendants repeated their
mantra: “[W]e want to expand as much as we possibly can.” Indeed, market expansion was
one of Carvana’s few key operating metrics. Thus, in every quarterly shareholder letter
during the Class Period, Garcia Junior and Jenkins dedicated an entire section to Carvana’s
expansion, in which they reported the number of new markets added to tout the purported
population coverage that Carvana had achieved. At the same time, Defendants exalted their
“capital-light” expansion model (i.e., a model that did not require a substantial number of
physical locations). Defendants’ statements, including those touting Carvana’s purported
population coverage, were misleading, however, as the vast majority of new markets were
significant distances from IRCs. The distance between IRCs and markets was material as
Carvana later admitted that it incurred $750 in additional per-vehicle cost in markets located
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 150 of 332
- 147 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
200 miles or more from an IRC. Unsurprisingly, near the end of the Class Period,
Defendants were forced to come clean that Carvana’s sales in its new distant markets
generated “lower profitability due to long distance from inventory.” Near the end of the
Class Period, Defendants took action to reduce sales in these far-flung markets to avoid these
less profitable sales. This caused the Company to report its first-ever quarterly sales decline
in Q3 2022.
217. In sum, because Defendants chose to tout positive information to the market
(its capital-light expansion model which enabled Carvana to launch new markets purportedly
covering the vast majority of the U.S. population) but omitted adverse facts that cut against
that information (Carvana’s expansion was not capital-light, would require the last-minute
purchase of ADESA’s 56 locations around the country, and could not support sales in new
markets), Defendants’ statements below were materially misleading.
218. Statement No. 21: On May 6, 2020, after the market closed, Carvana
announced its financial results for the fiscal quarter ended March 31, 2020 by issuing a Form
8-K that included a letter to shareholders (the “Q1 2020 Shareholder Letter”), which was
signed by Garcia Junior and Jenkins. The Q1 2020 Shareholder Letter touted Carvana’s
purported population coverage, stating: “[W]e launched 15 new markets and 1 vending
machine in Q1, bringing our total markets to 161, our total U.S. population coverage to
68.7%, and our total vending machines to 24 as of March 31, 2020.”
219. As detailed below, Statement No. 21 by Carvana, Garcia Junior, and Jenkins
omitted material facts that affirmatively created an impression of a state of affairs that
differed in a material way from the one that actually existed.
(a)
While Defendants touted the addition of 15 new markets during Q1
2020, they omitted that all of the markets were more than 100 miles from an existing IRC
and all but one of the new markets, 93%, were more than 200 miles from an existing IRC.
Critically, Defendants would later admit that sales in markets more than 200 miles from an
IRC incurred at least $750 more per car in logistics and related expenses and that, thus, these
distant markets generated “lower profitability due to long distance from inventory.”
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 151 of 332
- 148 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
(b)
While Defendants touted Carvana’s increase in “population coverage to
68.7%,” they omitted that the vast majority of that purported “population coverage”
consisted of markets far from existing IRCs. In fact, in Q1 2020, less than 32% of the total
U.S. population was located within 100 miles of an existing IRC and less than 54% was
located within 200 miles of an existing IRC.86 Indeed, as demonstrated below, the vast
majority of new markets opened during the Class Period, including the 15 new markets
added during Q1 2020, were greater than 100 miles from an existing IRC.
(c)
The omission of this information was material. Defendants would later
admit that sales in markets more than 200 miles from an IRC incurred at least $750 more per
car in logistics and related expenses and that, thus, these distant markets generated “lower
86 When Carvana purchased ADESA in early 2022, it announced that 32% of the total U.S.
population was located within 100 miles of an existing IRC and 54% was located within 200
miles of an existing IRC. Because Carvana added nine IRCs between the time of this
statement and Q1 2022, the percent of the population located within 100 or 200 miles from
an IRC was even lower.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 152 of 332
- 149 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
profitability due to long distance from inventory.” Indeed, Garcia Junior later acknowledged
Carvana expanded to places like the “Pacific Northwest” where “we didn’t have an
inspection center. . . . So we would just kind of incur all the cost associated with shipping
the cars out there . . . .” Further, he acknowledged that the Pacific Northwest was only one
example of an unprofitable distant market, stating, “there really are many, many, I mean,
probably dozens of versions of the same story.”
(d)
While Defendants touted Carvana’s expansion, they omitted the fact
that Carvana was lacking billions of dollars of critical brick-and-mortar infrastructure,
including reconditioning centers and logistics capabilities, to operate its nationwide footprint.
In fact, it would later be revealed that they would have to spend $3 billion to purchase
ADESA’s 56 brick-and-mortar locations, consisting of 6.5 million square feet of buildings
on more than 4,000 acres of land. As reported in a September 13, 2022 Vehicle Remarketing
article, Garcia Junior knew from the outset, “‘[y]ou need a place to stage the logistics, and
there’s no better infrastructure than auctions to serve that purpose.’” In fact, he later
admitted that “[a] major factor motivating the deal is the fact that an e-commerce used car
retailer needs a physical footprint like Amazon, which is undergirded by a network of
massive distribution centers and transport depots.” After the Class Period, Garcia Junior
went as far as to exalt this capital-heavy approach saying that, with ADESA, “we now have
6,500 acres and 500,000 parking spots connected by a logistics network.” He also clarified
an analyst’s characterization of Carvana as a technology business by stating, “I don’t know
exactly what kind of business we are . . . [b]ut part of what we are without question is a big
physical infrastructure business. Like that’s part of what we have to be.”
(e)
Analysts were misled. On May 6, 2020, a William Blair analyst stated,
Carvana’s “expansion will shift to opening smaller markets that can be served by Carvana’s
existing logistics network with little incremental investment.”
220. Statement No. 22: On August 5, 2020, after the market closed, Carvana
announced its financial results for the fiscal quarter ended June 30, 2020 by issuing a Form
8-K that included a letter to shareholders (the “Q2 2020 Shareholder Letter”), which was
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 153 of 332
- 150 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
signed by Garcia Junior and Jenkins. The Q2 2020 Shareholder Letter touted Carvana’s
purported population coverage, stating: “We opened a record 100 new markets in Q2 2020,
increasing the total percentage of the U.S. population we serve in our 261 markets to
73.2%, up from 68.7% at the end of Q1 2020.”
221. As detailed below, Statement No. 22 by Carvana, Garcia Junior, and Jenkins
omitted material facts that affirmatively created an impression of a state of affairs that
differed in a material way from the one that actually existed.
(a)
While Defendants touted the addition of 100 new markets, they omitted
that 83% of these markets were more than 100 miles from an existing IRC and 46% were
more than 200 miles from an existing IRC. Critically, Defendants would later admit that
sales in markets more than 200 miles from an IRC incurred at least $750 more per car in
logistics and related expenses and that, thus, these distant markets generated “lower
profitability due to long distance from inventory.”
(b)
While Defendants touted Carvana’s increase in the “total percentage of
the U.S. population we serve in our 261 markets to 73.2%” they omitted that the vast
majority of that purported “population coverage” consisted of markets far from existing
IRCs. In fact, in Q2 2020, less than 32% of the total U.S. population was located within 100
miles of an existing IRC and less than 54% was located within 200 miles of an existing
IRC.87 These omissions were material. Defendants would later admit that sales in markets
more than 200 miles from an IRC incurred at least $750 more per car in logistics and related
expenses and that, thus, these distant markets generated “lower profitability due to long
distance from inventory.” Garcia Junior later acknowledged Carvana expanded to “many,
many, I mean, probably dozens” of places like the “Pacific Northwest” where “we didn’t
have an inspection center. . . . So we would just kind of incur all the cost associated with
87 When Carvana purchased ADESA in early 2022, it announced that 32% of the total U.S.
population was located within 100 miles of an existing IRC and 54% was located within 200
miles of an existing IRC. Because Carvana added eight IRCs between the time of this
statement and Q1 2022, the percent of the population located within 100 or 200 miles from
an IRC was even lower.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 154 of 332
- 151 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
shipping the cars out there . . . .” Indeed, near the end of the Class Period, Defendants were
forced to reduce advertising in these markets and increase long-distance shipping fees to stay
afloat.
(c)
While Defendants touted Carvana’s expansion, they omitted the fact
that Carvana was lacking billions of dollars of critical brick-and-mortar infrastructure,
including reconditioning centers and logistics capabilities, to operate its nationwide footprint.
In fact, it would later be revealed that they would have to spend $3 billion to purchase
ADESA’s 56 brick-and-mortar locations, consisting of 6.5 million square feet of buildings
on more than 4,000 acres of land. As reported in a September 13, 2022 Vehicle Remarketing
article, Garcia Junior knew from the outset, “‘[y]ou need a place to stage the logistics, and
there’s no better infrastructure than auctions to serve that purpose.’” In fact, he later
admitted that “[a] major factor motivating the deal is the fact that an e-commerce used car
retailer needs a physical footprint like Amazon, which is undergirded by a network of
massive distribution centers and transport depots.” After the Class Period, Garcia Junior
went as far as to exalt this capital-heavy approach saying that, with ADESA, “we now have
6,500 acres and 500,000 parking spots connected by a logistics network.” He also clarified
an analyst’s characterization of Carvana as a technology business by stating, “I don’t know
exactly what kind of business we are . . . [b]ut part of what we are without question is a big
physical infrastructure business. Like that’s part of what we have to be.”
(d)
Analysts were misled. On August 6, 2020, a Cowen analyst stated,
“CVNA launched in a record 100 new markets in 2Q20, bringing coverage to 73.2% of the
US population . . . with minimal incremental cost by expanding into smaller adjacent
markets that could be supported via their existing logistics and delivery network.” As one
misled commentator realized near the end of the Class Period when Defendants disclosed
that they would reduce advertising in these new far-flung markets, “[Carvana’s expansion]
was growth for growth’s sake and not necessarily growth to improve profitability.”
222. Statement No. 23: On February 25, 2021, after the market closed, Carvana
announced its financial results for the fiscal quarter and year ended December 31, 2020, by
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 155 of 332
- 152 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
issuing a Form 8-K that included a letter to shareholders (the “Q4 2020 Shareholder Letter”),
which was signed by Garcia Junior and Jenkins. The Q4 2020 Shareholder Letter stated:
The Company “[a]dded five new markets, bringing our end-of-year total to 266 covering
73.7% of the population.”
223. As detailed below, Statement No. 23 made by Carvana, Garcia Junior, and
Jenkins omitted material facts that affirmatively created an impression of a state of affairs
that differed in a material way from the one that actually existed.
(a)
While Defendants touted the addition of five new markets, they omitted
that 100% of these markets were more than 100 miles from an existing IRC and 60% – three
markets – were more than 200 miles from an existing IRC. Defendants also omitted that, of
the 120 markets they added in FY 2020, 86% were more than 100 miles from an existing
IRC and 53% were more than 200 miles from an existing IRC. Critically, Defendants would
later admit that sales in markets more than 200 miles from an IRC incurred at least $750
more per car in logistics and related expenses and that, thus, these distant markets generated
“lower profitability due to long distance from inventory.”
(b)
While Defendants touted Carvana’s increase in “population coverage”
to 73.7%, they omitted that the vast majority of that purported “population coverage”
consisted of markets far from existing IRCs. In fact, in Q4 2020, less than 32% of the total
U.S. population was located within 100 miles of an existing IRC and less than 54% was
located within 200 miles of an existing IRC.88 Indeed, the vast majority of new markets
opened during the Class Period, including the five new markets added during Q4 2020, were
greater than 100 miles from an existing IRC. The omission of this information was material.
Defendants would later admit that sales in markets more than 200 miles from an IRC
incurred at least $750 more per car in logistics and related expenses and that, thus, these
88 When Carvana purchased ADESA in early 2022, it announced that 32% of the total U.S.
population was located within 100 miles of an existing IRC and 54% was located within 200
miles of an existing IRC. Because Carvana added six IRCs between the time of this
statement and Q1 2022, the percent of the population located within 100 or 200 miles from
an IRC was even lower.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 156 of 332
- 153 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
distant markets generated “lower profitability due to long distance from inventory.” Garcia
Junior later acknowledged Carvana expanded to “many, many, I mean, probably dozens” of
places like the “Pacific Northwest” where “we didn’t have an inspection center. . . . So we
would just kind of incur all the cost associated with shipping the cars out there . . . .” Indeed,
near the end of the Class Period, Defendants were forced to reduce sales in these markets.
(c)
While Defendants touted Carvana’s expansion, they omitted the fact
that Carvana was lacking billions of dollars of critical brick-and-mortar infrastructure,
including reconditioning centers and logistics capabilities, to operate its nationwide footprint.
In fact, it would later be revealed that they would have to spend $3 billion to purchase
ADESA’s 56 brick-and-mortar locations, consisting of 6.5 million square feet of buildings
on more than 4,000 acres of land. As reported in a September 13, 2022 Vehicle Remarketing
article, Garcia Junior knew from the outset, “‘[y]ou need a place to stage the logistics, and
there’s no better infrastructure than auctions to serve that purpose.’” In fact, he later
admitted that “[a] major factor motivating the deal is the fact that an e-commerce used car
retailer needs a physical footprint like Amazon, which is undergirded by a network of
massive distribution centers and transport depots.” After the Class Period, Garcia Junior
went as far as to exalt this capital-heavy approach saying that, with ADESA, “we now have
6,500 acres and 500,000 parking spots connected by a logistics network.” He also clarified
an analyst’s characterization of Carvana as a technology business by stating, “I don’t know
exactly what kind of business we are . . . [b]ut part of what we are without question is a big
physical infrastructure business. Like that’s part of what we have to be.”
(d)
The market was misled. On February 25, 2021, a Benchmark analyst
stated it viewed Carvana as “[p]ositioned for growth: . . . The company opened 120 new
markets in 2020, reaching 74% of the US market . . . .” As one misled commentator realized
near the end of the Class Period when Defendants disclosed that they would reduce
advertising in these new far-flung markets, “[Carvana’s expansion] was growth for growth’s
sake and not necessarily growth to improve profitability.”
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 157 of 332
- 154 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
224. Statement No. 24: Also on February 25, 2021, Carvana filed its 2020 10-K
with the SEC, signed by Garcia Junior and Jenkins and attaching SOX certifications signed
by Garcia Junior and Jenkins. The 2020 10-K exalted the strengths and advantages of
Carvana’s logistics capabilities and “capital-light expansion model,” stating:
Our business benefits from powerful network effects. Our logistics
capabilities allow us to offer every car in our inventory to customers across
all of our markets.
*
*
*
We believe there is a substantial opportunity to utilize our capital-light
expansion model and proven go-to-market strategy to enter additional markets
by expanding our existing logistics network and advertising in those markets.
225. As detailed below, Statement No. 24 made by Carvana, Garcia Junior, and
Jenkins omitted material facts that affirmatively created an impression of a state of affairs
that differed in a material way from the one that actually existed.
(a)
Contrary to Defendants’ statement that Carvana had a “capital-light
expansion model,” Carvana’s expansion actually required billions of dollars of critical brick-
and-mortar infrastructure, including reconditioning centers and logistics capabilities, to
operate its nationwide footprint. In fact, it would later be revealed that Carvana would have
to spend $3 billion to purchase ADESA’s 56 brick-and-mortar locations, consisting of 6.5
million square feet of buildings on more than 4,000 acres of land. As reported in a
September 13, 2022 Vehicle Remarketing article, Garcia Junior knew from the outset,
“‘[y]ou need a place to stage the logistics, and there’s no better infrastructure than auctions
to serve that purpose.’” In fact, he later admitted that “[a] major factor motivating the
[ADESA] deal is the fact that an e-commerce used car retailer needs a physical footprint like
Amazon, which is undergirded by a network of massive distribution centers and transport
depots.” Moreover, CW-12 met with Garcia Junior, who informed CW-12 that Carvana’s
management team needed ADESA’s reconditioning centers located nationwide. CW-4 and
CW-7 also stated that Carvana’s acquisition of ADESA was part of Carvana’s expansion
efforts. After the Class Period, Garcia Junior went as far as to exalt this capital-heavy
approach saying that, with ADESA, “we now have 6,500 acres and 500,000 parking spots
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 158 of 332
- 155 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
connected by a logistics network.” He also clarified an analyst’s characterization of Carvana
as a technology business by stating, “I don’t know exactly what kind of business we are . . .
[b]ut part of what we are without question is a big physical infrastructure business. Like
that’s part of what we have to be.”
(b)
In fact, prior to its purchase of ADESA, far from relying on “[o]ur
logistics network,” Carvana was forced to rely on costly third-party providers for vital
logistics and reconditioning functions in its new far-flung markets. CW-4 explained that
Carvana’s range for picking up and delivering vehicles was only five hours out and five
hours back (although third-party haulers could go further). Although Carvana had operations
in some of the western states like California and Nevada, there was not a hub or IRC
between Indiana and Oregon that would fall within the five-hour transportation limit. As
such, a customer in Ohio purchasing a car that was in Oregon would exceed this
range. However, the sheer amount of money it would cost to pay third-party haulers to cover
such distances was prohibitive. Indeed, it would inevitably cause logistic expenses to spike
300% from the start of the Class Period to Q1 2022. Defendants were able to largely conceal
these costs from investors until late in the Class Period when Defendants admitted the need
to “[r]educe third-party inbound transport share,” “[i]n-sourc[e] third-party pickups” for
Carvana’s “[l]ast-mile delivery,” and “[r]educe third-party shipping” to curtail logistics
expenses. Further, Defendants were able to substantially conceal these costs until they
disclosed operations expenses and “[r]econditioning and [i]nbound [t]ransport [c]ost [p]er
[r]etail [u]nit” for the first time after the Class Period. Carvana also acknowledged near the
end of the Class Period that “location growth” had led to “the addition of many third-party
reconditioning locations.” In sum, far from having a “capital-light expansion model,”
Carvana’s costs increased as its geographic footprint expanded.
(c)
Contrary to Defendants’ statement that “[o]ur logistics capabilities
allow us to offer every car in our inventory to customers across all of our markets,”
Defendants knew that it was unprofitable to pick up and deliver cars over “all of [their]
markets.” Indeed, near the end of the Class Period, Defendants acknowledged the need to
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 159 of 332
- 156 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
“meter” (i.e., stop) selling retail cars to reduce “less profitable sales” in “markets with lower
profitability due to long distance from inventory.”
226. Statement No. 25: On May 6, 2021, after the market closed, Carvana held a
conference call for analysts and investors to discuss the fiscal results for Q1 2021, in which
Jenkins spoke about expansion:
So far in Q1, we have opened 22 new markets, bringing our total to
288 and increasing our population coverage to more than 77% of the U.S.
population. In May, we also launched our first 5 markets in the Pacific
Northwest, adding the last major region to our nationwide footprint.
227. As detailed below, Statement No. 25 made by Carvana and Jenkins omitted
material facts that affirmatively created an impression of a state of affairs that differed in a
material way from the one that actually existed.
(a)
While Jenkins touted the addition of 22 new markets and five new
markets in the Pacific Northwest, he omitted that 100% of these markets were more than 200
miles from an existing IRC. Critically, Defendants would later admit that sales in markets
more than 200 miles from an IRC incurred at least $750 more per car in logistics and related
expenses and that, thus, these distant markets generated “lower profitability due to long
distance from inventory.”
(b)
While Jenkins touted Carvana’s increase in population coverage to
“more than 77%,” he omitted that the vast majority of that purported “population coverage”
consisted of markets far from existing IRCs. In fact, in Q1 2021 less than 32% of the total
U.S. population was located within 100 miles of an existing IRC and less than 54% was
located within 200 miles of an existing IRC.89 Indeed, the vast majority of new markets
opened during the Class Period, including the five new markets added during Q1 2021, were
greater than 100 miles from an existing IRC. The omission of this information was material.
Defendants would later admit that sales in markets more than 200 miles from an IRC
89 When Carvana purchased ADESA in early 2022, it announced that 32% of the total U.S.
population was located within 100 miles of an existing IRC and 54% was located within 200
miles of an existing IRC. Because Carvana added five IRCs between the time of this
statement and Q1 2022, the percent of the population located within 100 or 200 miles from
an IRC was even lower.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 160 of 332
- 157 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
incurred at least $750 more per car in logistics and related expenses and that, thus, these
distant markets generated “lower profitability due to long distance from inventory.” Indeed,
Garcia Junior later acknowledged Carvana expanded to “many, many, I mean, probably
dozens” of places like the “Pacific Northwest” where “we didn’t have an inspection
center. . . . So we would just kind of incur all the cost associated with shipping the cars out
there . . . .” In fact, near the end of the Class Period, Defendants were forced to reduce
advertising in these markets and increase long-distance shipping fees to stay afloat.
(c)
When Jenkins touted Carvana’s expansion, he omitted the fact that
Carvana was lacking billions of dollars of critical brick-and-mortar infrastructure, including
reconditioning centers and logistics capabilities, to operate its nationwide footprint. In fact,
it would later be revealed that they would have to spend $3 billion to purchase ADESA’s 56
brick-and-mortar locations, consisting of 6.5 million square feet of buildings on more than
4,000 acres of land. As reported in a September 13, 2022 Vehicle Remarketing article,
Garcia Junior later admitted, “[a] major factor motivating the deal is the fact that an e-
commerce used car retailer needs a physical footprint like Amazon, which is undergirded by
a network of massive distribution centers and transport depots.” After the Class Period,
Garcia Junior went as far as to exalt this capital-heavy approach saying that, with ADESA,
“we now have 6,500 acres and 500,000 parking spots connected by a logistics network.” He
also clarified an analyst’s characterization of Carvana as a technology business by stating, “I
don’t know exactly what kind of business we are . . . [b]ut part of what we are without
question is a big physical infrastructure business. Like that’s part of what we have to be.”
(d)
Unbeknownst to investors, Defendants were disregarding these
significant costs and sacrificing profitability for unit sales growth and “population coverage”
growth. Near the end of the Class Period, Defendants admitted that they had chased growth
with “less profitable sales” in “markets with lower profitability due to long distance from
inventory.” Garcia Junior cited one of the “many, many, I mean, probably dozens” of
examples, explaining “[w]e used to sell cars in the Pacific Northwest, even though we didn’t
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 161 of 332
- 158 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
have an inspection center there. . . . So we would just kind of incur all the cost associated
with shipping the cars out there because that was part of our long-term plan . . . .”
(e)
The market was misled. On May 7, 2021, a JMP Securities analyst
stated, “[p]erhaps our biggest takeaway from the quarter was Carvana’s continued market
expansion with the service now live in the Pacific Northwest with plans to cover 78-80% of
the U.S. population by year end.” As one misled commentator realized near the end of the
Class Period when Defendants disclosed that they would reduce advertising in these new far-
flung markets, “[Carvana’s expansion] was growth for growth’s sake and not necessarily
growth to improve profitability.”
228. Statement No. 26: Also during the Q1 2021 earnings call, Garcia Junior spoke
about the expansion:
[W]e’re excited about launching the Pacific Northwest. That was kind of
the last region in our footprint that we weren’t yet in. So I think opening
that up is exciting because it basically means from here to roughly our 95%
population coverage goal, in the long run, we basically have markets to fill
in. So that’s very exciting.
229. As detailed below, Statement No. 26 made by Carvana and Garcia Junior
omitted material facts that affirmatively created an impression of a state of affairs that
differed in a material way from the one that actually existed.
(a)
While Garcia Junior touted the launch of the first markets in the Pacific
Northwest, he omitted that these markets were far from existing IRCs. The omission of this
information was material. Defendants would later admit that sales in markets more than 200
miles from an IRC incurred at least $750 more per car in logistics and related expenses and
that, thus, these distant markets generated “lower profitability due to long distance from
inventory.” Garcia Junior later acknowledged they expanded to “many, many, I mean,
probably dozens” of places like the “Pacific Northwest” where “we didn’t have an inspection
center. . . . So we would just kind of incur all the cost associated with shipping the cars out
there . . . .” Indeed, near the end of the Class Period, Defendants were forced to reduce sales
in these markets.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 162 of 332
- 159 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
(b)
When Garcia Junior touted Carvana’s expansion to the Pacific
Northwest, he omitted the fact that Carvana was lacking billions of dollars of critical brick-
and-mortar infrastructure, including reconditioning centers and logistics capabilities, to
operate its nationwide footprint and these markets in particular. In fact, it would later be
revealed that they would have to spend $3 billion to purchase ADESA’s 56 brick-and-mortar
locations, consisting of 6.5 million square feet of buildings on more than 4,000 acres of land.
As reported in a September 13, 2022 Vehicle Remarketing article, Garcia Junior later
admitted in the ADESA Details Interview, “[a] major factor motivating the deal is the fact
that an e-commerce used car retailer needs a physical footprint like Amazon, which is
undergirded by a network of massive distribution centers and transport depots.” After the
Class Period, Garcia Junior went as far as to exalt this capital-heavy approach saying that,
with ADESA, “we now have 6,500 acres and 500,000 parking spots connected by a logistics
network.” He also clarified an analyst’s characterization of Carvana as a technology
business by stating, “I don’t know exactly what kind of business we are . . . [b]ut part of
what we are without question is a big physical infrastructure business. Like that’s part of
what we have to be.”
(c)
Unbeknownst to investors, Defendants were disregarding these
significant costs and sacrificing profitability for unit sales growth and “population coverage”
growth. Near the end of the Class Period, Defendants admitted that they had chased growth
with “less profitable sales” in “markets with lower profitability due to long distance from
inventory.” Garcia Junior cited one of the “many, many, I mean, probably dozens” of
examples, explaining “[w]e used to sell cars in the Pacific Northwest, even though we didn’t
have an inspection center there. . . . So we would just kind of incur all the cost associated
with shipping the cars out there because that was part of our long-term plan . . . .”
(d)
The market was misled. On May 7, 2021, a JMP Securities analyst
stated, “[p]erhaps our biggest takeaway from the quarter was Carvana’s continued market
expansion with the service now live in the Pacific Northwest with plans to cover 78-80% of
the U.S. population by year end.” As one misled commentator realized near the end of the
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 163 of 332
- 160 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Class Period when Defendants disclosed that they would reduce advertising in these new far-
flung markets, “[Carvana’s expansion] was growth for growth’s sake and not necessarily
growth to improve profitability.”
230. Statement No. 27: On August 5, 2021, Carvana issued its Q2 2021
Shareholder Letter signed by Garcia Junior and Jenkins, which exalted the Company’s
population coverage, stating: “In Q2 2021 we expanded the total percentage of the U.S.
population we serve to 79.4%, up from 74.5% at the end of Q1 2021 through the addition
of 27 new markets.”
231. As detailed below, Statement No. 27 made by Carvana, Garcia Junior, and
Jenkins omitted material facts that affirmatively created an impression of a state of affairs
that differed in a material way from the one that actually existed.
(a)
While Defendants touted the addition of 27 new markets, they omitted
that 100% of these markets were more than 200 miles from an existing IRC. Critically,
Defendants would later admit that sales in markets more than 200 miles from an IRC
incurred at least $750 more per car in logistics and related expenses and that, thus, these
distant markets generated “lower profitability due to long distance from inventory.”
(b)
While Defendants touted Carvana’s increase in population coverage to
79.4%, they omitted that the vast majority of that purported “population coverage” consisted
of markets far from existing IRCs. In fact, in Q2 2021, less than 32% of the total U.S.
population was located within 100 miles of an existing IRC and less than 54% was located
within 200 miles of an existing IRC.90 Indeed, the vast majority of new markets opened
during the Class Period, including the 27 new markets added during Q2 2021, were greater
than 100 miles from an existing IRC. The omission of this information was material.
Defendants would later admit that sales in markets more than 200 miles from an IRC
90 When Carvana purchased ADESA in early 2022, it announced that 32% of the total U.S.
population was located within 100 miles of an existing IRC and 54% was located within 200
miles of an existing IRC. Because Carvana added four IRCs between the time of this
statement and Q1 2022, the percent of the population located within 100 or 200 miles from
an IRC was even lower.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 164 of 332
- 161 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
incurred at least $750 more per car in logistics and related expenses and that, thus, these
distant markets generated “lower profitability due to long distance from inventory.” Indeed,
Garcia Junior later acknowledged Carvana expanded to places like the “Pacific Northwest”
where “we didn’t have an inspection center. . . . So we would just kind of incur all the cost
associated with shipping the cars out there . . . .” Further, he acknowledged that the Pacific
Northwest was only one example of an unprofitable distant market, stating, “there really are
many, many, I mean, probably dozens of versions of the same story.” Indeed, near the end
of the Class Period, Defendants were forced to reduce sales in these markets.
(c)
While Defendants touted Carvana’s expansion, they omitted the fact
that Carvana was lacking billions of dollars of critical brick-and-mortar infrastructure,
including reconditioning centers and logistics capabilities, to operate its nationwide footprint.
In fact, it would later be revealed that they would have to spend $3 billion to purchase
ADESA’s 56 brick-and-mortar locations, consisting of 6.5 million square feet of buildings
on more than 4,000 acres of land. According to Garcia Junior, a major factor for purchasing
ADESA was “the fact that an e-commerce used car retailer needs a physical footprint like
Amazon, which is undergirded by a network of massive distribution centers and transport
depots,” and “‘[y]ou need to warehouse and recon those cars to get them retail ready. . .
[y]ou need a place to stage the logistics.’” After the Class Period, Garcia Junior went as far
as to exalt this capital-heavy approach saying that, with ADESA, “we now have 6,500 acres
and 500,000 parking spots connected by a logistics network.” He also clarified an analyst’s
characterization of Carvana as a technology business by stating, “I don’t know exactly what
kind of business we are . . . [b]ut part of what we are without question is a big physical
infrastructure business. Like that’s part of what we have to be.”
(d)
As one misled commentator realized near the end of the Class Period,
when Defendants disclosed that they would reduce advertising in these new far-flung
markets, “[Carvana’s expansion] was growth for growth’s sake and not necessarily growth to
improve profitability.”
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 165 of 332
- 162 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
232. Statement No. 28: On November 4, 2021, Defendants issued a shareholder
letter (“Q3 2021 Shareholder Letter”), which was signed by Garcia Junior and Jenkins. The
Q3 2021 Shareholder Letter touted Carvana’s purported population coverage: “In Q3 2021
we expanded the total percentage of the U.S. population we serve to 80.6%, up from 79.4%
at the end of Q2 2021 through the addition of 9 new markets, taking another step toward
our goal of 95% population coverage in the U.S.”
233. As detailed below, Statement No. 28 made by Carvana, Garcia Junior, and
Jenkins omitted material facts that affirmatively created an impression of a state of affairs
that differed in a material way from the one that actually existed.
(a)
While Defendants touted the addition of nine new markets, they omitted
that 100% of these markets were more than 200 miles from an existing IRC. Critically,
Defendants would later admit that sales in markets more than 200 miles from an IRC
incurred at least $750 more per car in logistics and related expenses and that, thus, these
distant markets generated “lower profitability due to long distance from inventory.”
(b)
While Defendants touted Carvana’s increase in the “total percentage of
the U.S. population we serve to 80.6%” they omitted that less than 32% of the total U.S.
population was located within 100 miles of an existing IRC and less than 54% was located
within 200 miles of an existing IRC.91 These omissions were material. Defendants would
later admit that sales in markets more than 200 miles from an IRC incurred at least $750
more per car in logistics and related expenses and that, thus, these distant markets generated
“lower profitability due to long distance from inventory.” Defendants would also later admit
that Carvana had spent billions of dollars on ADESA in order to bring its markets within 100
miles of a reconditioning facility. Indeed, Garcia Junior later admitted that Carvana
expanded to “many, many, I mean, probably dozens” of places like the “Pacific Northwest”
91 When Carvana purchased ADESA in early 2022, it announced that 32% of the total U.S.
population was located within 100 miles of an existing IRC and 54% was located within 200
miles of an existing IRC. Because Carvana added four IRCs between the time of this
statement and Q1 2022, the percent of the population located within 100 or 200 miles from
an IRC was even lower.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 166 of 332
- 163 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
where “we didn’t have an inspection center. . . . So we would just kind of incur all the cost
associated with shipping the cars out there . . . .” Indeed, near the end of the Class Period,
Defendants were forced to reduce sales in these markets.
(c)
While Defendants touted Carvana’s expansion, they omitted the fact
that Carvana was lacking billions of dollars of critical brick-and-mortar infrastructure,
including reconditioning centers and logistics capabilities, to operate its nationwide footprint.
In fact, it would later be revealed that they would have to spend $3 billion to purchase
ADESA’s 56 brick-and-mortar locations, consisting of 6.5 million square feet of buildings
on more than 4,000 acres of land. According to Garcia Junior, a major factor for purchasing
ADESA was “the fact that an e-commerce used car retailer needs a physical footprint like
Amazon, which is undergirded by a network of massive distribution centers and transport
depots,” and “‘[y]ou need to warehouse and recon those cars to get them retail ready . . .
[y]ou need a place to stage the logistics.’” After the Class Period, Garcia Junior went as far
as to exalt this capital-heavy approach saying that, with ADESA, “we now have 6,500 acres
and 500,000 parking spots connected by a logistics network.” He also clarified an analyst’s
characterization of Carvana as a technology business by stating, “I don’t know exactly what
kind of business we are . . . [b]ut part of what we are without question is a big physical
infrastructure business. Like that’s part of what we have to be.”
(d)
By the second half of FY 2021, Carvana’s unsustainable nationwide
expansion had stretched its logistics and reconditioning network to the brink. As would be
later revealed, Defendants were forced to spend millions of dollars on costly third-party
logistics and reconditioning providers to make up for Carvana’s internal deficiencies in these
critical functions. After the Class Period, Defendants disclosed “[r]econditioning and
[i]nbound [t]ransport [c]ost [p]er [r]etail [u]nit” for the first time, which revealed that, during
Q3 2021, these costs had spiked $201 per vehicle and were continuing to steadily increase,
as depicted in the graphic below.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 167 of 332
- 164 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
These excess costs were highly material to Carvana’s reported results. In fact, when Carvana
reported Retail GPU of $1,769 in Q3 2021, a $253 decrease from the prior quarter,
Defendants concealed that 79% of the decline was due to the excess reconditioning and
inbound transportation costs. Defendants later conceded that the excess reconditioning and
inbound transportation costs during late 2021 and early 2022 were ultimately the result of
using costly third-party providers to perform these critical functions. For example, near the
end of the Class Period, Defendants admitted that, beginning on or around August 2021,
Carvana had begun outsourcing logistics, including increasing “third-party reconditioning
locations.” Further, Defendants emphasized that, to curtail logistics, Carvana needed to
quickly “[r]educe third-party production [i.e., reconditioning] volume,” “[r]educe third-
party inbound transport share,” “[i]n-sourc[e] third-party pickups” for Carvana’s “[l]ast-
mile delivery,” and “[r]educe third-party shipping.” Moreover, Defendants later attributed
significant post-Class-Period cost reductions to the “in-sourcing” of these same logistics and
reconditioning functions that Carvana had out-sourced to costly third-party providers during
the Class Period. In an August 9, 2023 investor presentation, Jenkins stated “[t]he single
biggest source of the [cost-saving] gains over [the last three fiscal quarters] ha[s] been in-
sourcing.” Tellingly, following the Class Period, Garcia Junior explained that, in bringing
down reconditioning costs, “[i]n-sourcing was probably the single biggest driver. So
basically taking services that had previously been provided by third parties in the
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 168 of 332
- 165 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
inspection and reconditioning centers and taking them in-house and doing them ourselves.
And that obviously saves costs and allow us to have better control of the process.”
(e)
As one misled commentator realized near the end of the Class Period
when Defendants disclosed that they would reduce advertising in these new far-flung
markets, “[Carvana’s expansion] was growth for growth’s sake and not necessarily growth to
improve profitability.”
234. Statement No. 29: On February 24, 2022, Carvana filed its 2021 10-K, signed
by Garcia Junior and Jenkins and attaching SOX certifications signed by Garcia Junior and
Jenkins attesting to the accuracy of financial reporting in the 10-K. The 2021 10-K exalted
the strengths and advantages of Carvana’s logistics capabilities and “capital-light expansion
model,” stating:
Our business benefits from powerful network effects. Our logistics
capabilities allow us to offer every car in our inventory to customers across
all of our markets.
*
*
*
We believe there is a substantial opportunity to utilize our capital-light
expansion model and proven go-to-market strategy to enter additional markets
by expanding our existing logistics network and advertising in those markets.
235. As detailed below, Statement No. 29 made by Carvana, Garcia Junior, and
Jenkins omitted material facts that affirmatively created an impression of a state of affairs
that differed in a material way from the one that actually existed.
(a)
While Defendants touted Carvana’s expansion model as “capital-light,”
its expansion required the buildout of costly IRCs to support the new markets, including the
intake, storage, processing, and reconditioning of vehicle inventory. However, in an effort to
appear “capital-light,” Carvana delayed the buildout of new IRCs. New IRC growth
significantly lagged behind Carvana’s expansion into new markets. Specifically, during
2021, Carvana added 45 new markets, but it added only 6 new IRCs. As a result, the costs
per IRC ballooned as Carvana “frequently moved vehicles between IRCs in different markets
due to parking constraints” causing “total vehicle miles traveled” to increase and “adding to
costs and logistics network complexity.” Eventually, the lack of critical infrastructure
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 169 of 332
- 166 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
crippled the Company’s operations, and Defendants were forced to spend $3 billion to
acquire ADESA and its 56 locations across the country to bridge the gap. CW-4 and CW-7
said that Carvana’s acquisition of ADESA was part of Carvana’s efforts to expand into new
markets that might have been far from existing Carvana IRCs. After the Class Period,
Garcia Junior went as far as to exalt this capital-heavy approach saying that, with ADESA,
“we now have 6,500 acres and 500,000 parking spots connected by a logistics network.” He
also clarified an analyst’s characterization of Carvana as a technology business by stating, “I
don’t know exactly what kind of business we are . . . [b]ut part of what we are without
question is a big physical infrastructure business. Like that’s part of what we have to be.”
(b)
Defendants omitted that, by the second half of FY 2021, Carvana’s
unsustainable “capital-light” nationwide “expansion model” had stretched its logistics and
reconditioning network to the brink. As a result, Defendants were forced to spend millions
of dollars on costly third-party logistics and reconditioning providers to make up for
Carvana’s internal deficiencies in these critical functions. After the Class Period, in
November 2023, Defendants made an important admission – they provided additional details
regarding Class Period expenses and separately broke out “[r]econditioning and [i]nbound
[t]ransport [c]ost [p]er [r]etail [u]nit” for the first time. The disclosures revealed that,
between mid-2021 and Q4 2021, these costs had spiked $369 per vehicle, as depicted in the
graphic below. Simply put, Carvana’s purportedly “capital-light” expansion model produced
the opposite results with costs increasing as its geographic footprint expanded.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 170 of 332
- 167 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
(c)
These excess costs were highly material to Carvana’s reported results.
In fact, of the $527 decrease in Carvana’s reported Retail GPU between Q2 2021 and Q4
2021, 70% was due to excess reconditioning and inbound transportation costs. Defendants
later conceded that the excess reconditioning and inbound transportation costs during late
2021 and early 2022 were ultimately the result of using costly third-party providers to
perform these critical functions. For example, less than three months after this statement was
made, Defendants admitted that, beginning on or around August 2021, Carvana had begun
outsourcing logistics, including increasing “third-party reconditioning locations” and
“add[ing] many third-party reconditioning locations.” Then, near the end of the Class
Period, Defendants emphasized that, to curtail logistics expenses, Carvana needed to quickly
“[r]educe third-party production [i.e., reconditioning] volume,” “[r]educe third-party
inbound transport share,” “[i]n-sourc[e] third-party pickups” for Carvana’s “[l]ast-mile
delivery,” and “[r]educe third-party shipping.” Carvana also acknowledged that “location
growth” had led to “the addition of many third-party reconditioning locations.” Moreover,
after the Class Period, Defendants attributed significant post-Class Period cost reductions to
the “in-sourcing” of these same logistics and reconditioning functions that Carvana had out-
sourced to costly third-party providers during the Class Period. In an August 9, 2023
investor presentation, Jenkins stated “[t]he single biggest source of the gains over [the last
three fiscal quarters] ha[s] been in-sourcing. . . . [W]e’ve really focused on . . . in-sourcing
services at more and more of IRCs as the single biggest driver [of cost savings].” Tellingly,
following the Class Period, Garcia Junior explained that, in bringing down reconditioning
costs, “[i]n-sourcing was probably the single biggest driver. So basically taking services
that had previously been provided by third parties in the inspection and reconditioning
centers and taking them in-house and doing them ourselves. And that obviously saves
costs and allow us to have better control of the process.”
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 171 of 332
- 168 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
E.
Defendants’ Materially False and Misleading Statements
Regarding Aged Inventory
236. Average days to sale is the number of days between when Carvana acquires a
car either from a customer or in the wholesale market and when they sell the car to a
customer in a retail transaction. Defendants described that an increase in the average days to
sale metric “is unfavorable for retail GPU because if you hold the car longer, it has more
time to experience depreciation before you sell it, then thereby selling at a lower retail
GPU.” As such, Carvana has always considered average days to sale a key internal metric.
For example, in its S-1 registration statement, at the time of its IPO, Carvana defined average
days to sale as one of just six “Key Operating Metrics.” Carvana also described:
Our business is dependent upon our ability to expeditiously sell
inventory. Failure to expeditiously sell our inventory could have a material
adverse effect on our business, sales and results of operations.
. . . An over-supply of used vehicle inventory will generally cause downward
pressure on our product sales prices and margins and increase our average
days to sale.
237. Average days to sale remained a key internal metric throughout the Class
Period. For example, Carvana repeated the above disclosure in its 2021 10-K and identified
the “[r]educ[tion of] average days to sale” as one of its primary “strategies designed to
increase our total gross profit per unit.” Even after the Class Period, Carvana highlighted the
importance of monitoring average days to sale. For example, in an August 2023 investor
presentation, Defendants highlighted average days to sale “impacts retail GPU through its
impact on the cumulative depreciation on vehicles before they sell.” During the same
presentation, Jenkins described average days to sale as “one of the useful metrics” to help
investors understand changes in Retail GPU. Despite the fact Defendants clearly viewed
average days to sale as a key internal metric before, during, and after the Class Period,
Defendants abruptly stopped disclosing it to investors in Q1 2021. At that time, Defendants
claimed “our number of IRCs is a more important metric than average days to sale due to the
impact of IRC capacity on retail units sold and the relative stability of average days to sale
over the past three years.” Defendants repeated this statement every quarter, even as average
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 172 of 332
- 169 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
days to sale was spiking as a result of Defendants’ scheme. Thus, Defendants’ statements
were materially misleading because they created an impression of a state of affairs (relatively
stable average days to sale) that differed in a material way from the one that actually existed
(a spiking average days to sale metric, which reflected Carvana’s glut of aging and
depreciating inventory, that would eventually cause Carvana to reduce inventory and forego
retail growth).
238. Statement No. 30: On February 24, 2022, Carvana filed its 2021 10-K with the
SEC, signed by Garcia Junior and Jenkins and attaching SOX certifications signed by Garcia
Junior and Jenkins. The 2021 10-K disclosed that it would no longer be reporting the
average days to sale metric. Further, Defendants stated: “As we continue to grow, our
number of IRCs is a more important metric than average days to sale due to the impact of
IRC capacity on retail units sold and the relative stability of average days to sale in recent
years.”
239. As detailed below, Statement No. 30 made by Carvana, Garcia Junior, and
Jenkins omitted material facts that affirmatively created an impression of a state of affairs
that differed in a material way from the one that actually existed.
(a)
As demonstrated in the first chart below, contrary to Defendants’
statement that average days to sale was “relative[ly] stab[le] . . . in recent years,” during Q4
2021, average days to sale had spiked to 71 days, increasing almost 20% above Q3 2021 and
the recent historical range in the low-60-days range.92 At the time that Statement No. 30 was
made on February 24, 2022 (i.e., 55 days into Q1 2022), Carvana’s average days to sale had
spiked even further above its recent historical norms. Defendants’ statements were
materially misleading because they created an impression of a state of affairs (relatively
stable average days to sale) that differed in a material way from the one that actually existed
92 Defendants’ statement conveyed “relative stability” of the average days to sale metric in
the context of recent historical ranges, which was consistently in the low-60-days range. The
average days to sale in recent years were as follows: 64 in FY 2018; 62 in FY 2019; 58 in Q3
2020 (67 in FY 2020 including the impact of a temporary spike in Q1 and Q2 2020 caused
by the COVID-19 disruption); 60 in Q1 2021; 54 in Q2 2021; and 60 in Q3 2021.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 173 of 332
- 170 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
(a rapidly spiking average days to sale metric which had significant negative impact on
Carvana’s Retail GPU).
(b)
Defendants’ omission of average days to sale was material. Average
days to sale is the number of days between when Carvana acquires a car either from a
customer or in the wholesale market and when they sell the car to a customer in a retail
transaction. Defendants described that an increase in the average days to sale metric “is
unfavorable for retail GPU because if you hold the car longer, it has more time to
experience depreciation before you sell it, then thereby selling at a lower retail GPU.” As
such, Carvana has always considered average days to sale a key internal metric. For
example, at the time of its IPO, Carvana defined average days to sale as one of just six “Key
Operating Metrics” in its S-1 registration statement. Carvana also stated:
Our business is dependent upon our ability to expeditiously sell
inventory. Failure to expeditiously sell our inventory could have a material
adverse effect on our business, sales and results of operations.
. . . An over-supply of used vehicle inventory will generally cause
downward pressure on our product sales prices and margins and increase our
average days to sale.
Average days to sale remained a key internal metric throughout the Class Period. For
example, despite not disclosing the actual metric in its 2021 10-K, Carvana repeated the
above disclosure and identified the “[r]educ[tion] average days to sale” as one of its primary
“strategies designed to increase our total gross profit per unit” in the 2021 10-K. After the
Class Period, Carvana continued to highlight the importance of monitoring this key metric.
For example, in an August 2023 investor presentation, Defendants highlighted that average
2021 10-K
February 24, 2022
(55 days into Q1
2022)
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 174 of 332
- 171 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
days to sale “impacts retail GPU through its impact on the cumulative depreciation on
vehicles before they sell.” Jenkins also stated in that presentation that average days to sale is
“one of the useful metrics” to help investors understand changes in Retail GPU. Thus,
Defendants clearly viewed average days to sale as a key internal metric before, during, and
after the Class Period.
(c)
As detailed below, Defendants further concealed the fact that the spike
in average days to sale by the second half of FY 2021, was the direct result of several factors
associated with their fraudulent scheme and course of conduct.
(i)
First, Defendants’ unsustainable increase in purchasing cars from
customers, including low-quality cars, left Carvana with a glut of aging inventory. In Q3
2021, Defendants described “explosive growth in buying cars from customers over the last
two quarters.” This surplus of cars had a direct impact on average days to sale by Q4 2021.
At the end of Class Period, Defendants would admit that Carvana needed to “significantly
reduc[e] retail vehicle acquisitions.”
(ii)
Second, Defendants were forced to significantly “meter,” or
intentionally not list inventory for sale in certain markets as a result of its unsustainable
nationwide expansion. Indeed, Defendants later admitted that it metered sales to reduce sales
in “markets with lower profitability due to long distance from inventory (e.g., the Pacific
Northwest).” This had a direct impact on average days to sale by Q4 2021.
(iii)
Third, Defendants admitted that “over the last several months,”
preceding August 2022, Carvana had to implement “buffers in many states to provide
cushion for their title and registration teams to work with the various states to complete
necessary registration paperwork,” which materially slowed retail sales growth and adversely
impacted average days to sale.
(d)
Near the end of the Class Period, Defendants admitted that, far from
being at a “relative[ly] stab[le]” level, Carvana had to “reduce[] inventory,” which was one
of the “largest drivers” of Carvana’s disappointing retail unit sales decline in Q3 2022.
Carvana further acknowledged that it would be “continuing to normalize our inventory size
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 175 of 332
- 172 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
and expect to further reduce inventory in Q4 [2022].” In addition, at the end of the Class
Period, Defendants were forced to write off millions of dollars of aged and impaired vehicle
inventory that was worth less than Carvana had paid to acquire it and prepare it for sale. In
fact, Carvana recorded a $52 million retail inventory allowance adjustment, which reduced
Q4 2022 Retail GPU by $598 per retail unit – the equivalent per-unit cost of holding a
vehicle more than 90 days. Carvana also recorded a $5 million wholesale inventory
allowance adjustment and elected to sell certain impaired retail vehicles in the wholesale
market, resulting in a loss of $4 million, which reduced Q4 2022 GPU by an additional $103
per retail unit. Moreover, Defendants later reported a “substantial reduction in our inventory
size,” admitting “[f]rom the end of Q1 to the end of Q4, we reduced our inventory balance by
43%.” Defendants also belatedly acknowledged that “[r]educing inventory size positively
impacts Retail GPU.” Finally, on August 9, 2023, Defendants acknowledged in an investor
presentation:
[T]he first key driver of . . . our recent success on retail GPU, is what I’ll call
normalizing inventory size or really it’s more about normalizing inventory
turn times. . . . Average days of sale is particularly – it’s the number of days
between when we acquire a car either from a customer or in the wholesale
market and when we sell a car to a customer on a completed retail transaction.
And so you can see just as a benchmark, 2021, we were running on
average in the low 60s average across the 4 quarters. We really increased
this metric, which is unfavorable for retail GPU because if you hold the car
longer, it has more time to experience depreciation before you sell it, then
thereby selling at a lower retail GPU. But it stayed really elevated through
2022 when we overbuilt the business for the ultimate sales environment. You
can see it sort of in the mid- to high 90s there through most of 2022.
(e)
Defendants conceded in the August 9, 2023 investor presentation that an
increase in average days to sale over Carvana’s historical benchmark “in the low 60s” was
“unfavorable for retail GPU because if you hold the car longer, it has more time to
experience depreciation before you sell it, then thereby selling at a lower retail GPU.” For
example, because daily depreciation in the used car market could exceed $20/day at this
time, a ten-day spike in average days to sale could negatively impact GPU by more than
$200 per vehicle. At the end of the Class Period, Carvana characterized the impact of an
increase in average days to sale, explaining that retail vehicles that were not sold within 90
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 176 of 332
- 173 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
days of acquisition realized ~$600 of lower Retail GPU compared to retail units sold within
90 days. To put it in perspective, in Q4 2021, $600 represented 40% of Carvana’s retail total
GPU.
(f)
Defendants claim that “our number of IRCs is a more important metric
than average days to sale due to the impact of IRC capacity on retail units sold” was also
false and misleading. Average days to sale is the number of days between when Carvana
acquires a car either from a customer or in the wholesale market and when they sell the car to
a customer in a retail transaction. The number of IRCs, a purportedly more important metric,
had almost no relation to average days of sale as it was not a measure of how quickly
Carvana sold its vehicles. Defendants viewed average days to sale as a key internal metric
before, during, and after the Class Period. Indeed, in its S-1 registration statement, at the
time of its IPO, Carvana defined average days to sale as one of just six “Key Operating
Metrics” because “[o]ur business is dependent upon our ability to expeditiously sell
inventory.” During the Class Period, despite concealing the metric in the 2021 10-K,
Carvana still identified “[r]educe average days to sale” among its primary “strategies
designed to increase our total gross profit per unit.” And even after the Class Period,
Defendants highlighted average days to sale in an August 2023 investor presentation, noting
that it “impacts retail GPU through its impact on the cumulative depreciation on vehicles
before they sell.” In addition, Jenkins described average days to sale as “one of the useful
metrics” to help investors understand changes in Retail GPU. The number of IRCs,
however, was not a “useful metric” to understanding changes in Retail GPU.
240. Statement No. 31: On May 10, 2022, Carvana filed its Q1 2022 10-Q with the
SEC, signed by Jenkins and attaching SOX certifications by Garcia Junior and Jenkins. In
the Q1 2022 10-Q, Defendants disclosed: “As we continue to grow, our number of IRCs is a
more important metric than average days to sale due to the impact of IRC capacity on
retail units sold and the relative stability of average days to sale in recent years.”
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 177 of 332
- 174 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
241. As detailed below, Statement No. 31 made by Carvana, Garcia Junior, and
Jenkins omitted material facts that affirmatively created an impression of a state of affairs
that differed in a material way from the one that actually existed.
(a)
As demonstrated in the first chart below, contrary to Defendants’
statement that average days to sale was “relative[ly] stab[le] . . . in recent years,” during Q1
2022, average days to sale had spiked to 82 days, increasing almost 38% above Q3 2021 and
the recent historical range in the low-60-days range.93 At the time that Statement No. 31 was
made on May 10, 2022, 40 days into Q2 2022, Carvana’s average days to sale had spiked
even further above its recent historical norms. Defendants’ statements were materially
misleading because they created an impression of a state of affairs (relatively stable average
days to sale) that differed in a material way from the one that actually existed (a rapidly
spiking average days to sale metric which had significant negative impact on Carvana’s
Retail GPU).
(b)
Defendants’ omission of average days to sale was material. Average
days to sale is the number of days between when Carvana acquires a car either from a
customer or in the wholesale market and when they sell the car to a customer in a retail
93 Defendants’ statement conveyed “relative stability” of the average days to sale metric in
the context of recent historical ranges, which was consistently in the low-60-days range. The
average days to sale in recent years were as follows: 64 in FY 2018; 62 in FY 2019; 58 in Q3
2020 (67 in FY 2020 including the impact of a temporary spike in Q1 and Q2 2020 caused
by the COVID-19 disruption); 60 in Q1 2021; 54 in Q2 2021; and 60 in Q3 2021.
Q1 2022 10-Q May
10, 2022 (40 days
into Q2 2022)
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 178 of 332
- 175 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
transaction. Defendants described that an increase in the average days to sale metric “is
unfavorable for retail GPU because if you hold the car longer, it has more time to
experience depreciation before you sell it, then thereby selling at a lower retail GPU.” As
such, Carvana has always considered average days to sale a key internal metric. For
example, at the time of its IPO, Carvana defined average days to sale as one of just six “Key
Operating Metrics” in its S-1 registration statement. Carvana also described:
Our business is dependent upon our ability to expeditiously sell
inventory. Failure to expeditiously sell our inventory could have a material
adverse effect on our business, sales and results of operations.
. . . An over-supply of used vehicle inventory will generally cause
downward pressure on our product sales prices and margins and increase our
average days to sale.
Average days to sale remained a key internal metric throughout the Class Period. For
example, despite not disclosing the actual metric in the 2021 10-K, Carvana repeated the
above disclosure and identified the “[r]educ[tion of] average days to sale” as one of its
primary “strategies designed to increase our total gross profit per unit” in the 2021 10-K.
After the Class Period, Carvana continued to highlight the importance of monitoring this
metric. For example, in an August 2023 investor presentation, Defendants highlighted that
average days to sale “impacts retail GPU through its impact on the cumulative depreciation
on vehicles before they sell.” Jenkins also stated during the presentation that average days to
sale is “one of the useful metrics” to help investors understand changes in Retail GPU.
Thus, Defendants clearly viewed average days to sale as a key internal metric before, during,
and after the Class Period.
(c)
As detailed below, Defendants further concealed the fact that the spike
in average days to sale by the second half of FY 2021 was the direct result of several factors
associated with their fraudulent scheme and course of conduct.
(i)
First, Defendants’ unsustainable increase in purchasing cars from
customers, including low-quality cars, left Carvana with a glut of aging inventory. In Q3
2021, Defendants described “explosive growth in buying cars from customers over the last
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 179 of 332
- 176 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
two quarters.” This surplus of cars had a direct impact on average days to sale by Q1 2022.
At the end of the Class Period, Defendants disclosed that Carvana needed to “significantly
reduc[e] retail vehicle acquisitions.”
(ii)
Second, Defendants were forced to significantly “meter,” or
intentionally not list inventory for sale in certain markets as a result of its unsustainable
nationwide expansion. Indeed, Defendants later admitted that it metered sales to reduce sales
in “markets with lower profitability due to long distance from inventory (e.g., the Pacific
Northwest).” This had a direct impact on average days to sale by Q1 2022.
(iii)
Third, Defendants admitted that “over the last several months,”
preceding August 2022, Carvana had to implement “buffers in many states to provide
cushion for their title and registration teams to work with the various states to complete
necessary registration paperwork,” which materially slowed retail sales growth and adversely
impacted average days to sale.
(d)
Near the end of the Class Period, Defendants admitted that, far from
being at a “relative[ly] stab[le]” level, Carvana had to “reduce[] inventory,” which was one
of the “largest drivers” of Carvana’s disappointing retail unit sales decline in Q3 2022.
Carvana further acknowledged that it would be “continuing to normalize our inventory size
and expect to further reduce inventory in Q4 [2022].” In addition, at the end of the Class
Period, Defendants were forced to write off millions of dollars of aged and impaired vehicle
inventory that was worth less than Carvana had paid to acquire it and prepare it for sale. In
fact, Carvana recorded a $52 million retail inventory allowance adjustment, which reduced
Q4 2022 Retail GPU by $598 per retail unit – the equivalent per-unit cost of holding a
vehicle more than 90 days. Carvana also recorded a $5 million wholesale inventory
allowance adjustment and elected to sell certain impaired retail vehicles in the wholesale
market, resulting in a loss of $4 million, which reduced Q4 2022 GPU by an additional $103
per retail unit. Moreover, Defendants later reported a “substantial reduction in our inventory
size,” admitting “[f]rom the end of Q1 to the end of Q4, we reduced our inventory balance by
43%.” Defendants also belatedly acknowledged that “[r]educing inventory size positively
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 180 of 332
- 177 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
impacts Retail GPU.” Finally, on August 29, 2023, Defendants acknowledged in an investor
presentation:
[T]he first key driver of . . . our recent success on retail GPU, is what I’ll call
normalizing inventory size or really it’s more about normalizing inventory
turn times. . . . Average days of sale is particularly – it’s the number of days
between when we acquire a car either from a customer or in the wholesale
market and when we sell a car to a customer on a completed retail transaction.
And so you can see just as a benchmark, 2021, we were running on
average in the low 60s average across the 4 quarters. We really increased
this metric, which is unfavorable for retail GPU because if you hold the car
longer, it has more time to experience depreciation before you sell it, then
thereby selling at a lower retail GPU. But it stayed really elevated through
2022 when we overbuilt the business for the ultimate sales environment. You
can see it sort of in the mid- to high 90s there through most of 2022.
(e)
Defendants conceded in the August 9, 2023 investor presentation that an
increase in average days to sale over Carvana’s historical benchmark “in the low 60s” was
“unfavorable for retail GPU because if you hold the car longer, it has more time to
experience depreciation before you sell it, then thereby selling at a lower retail GPU.” For
example, because daily depreciation in the used car market could exceed $20/day at this
time, a ten-day spike in average days to sale could negatively impact GPU by more than
$200 per vehicle. At the end of the Class Period, Carvana characterized the impact of an
increase in average days to sale, explaining that retail vehicles that were not sold within 90
days of acquisition realized ~$600 of lower Retail GPU compared to retail units sold within
90 days. To put it in perspective, in Q4 2021, $600 represented 40% of Carvana’s retail total
GPU.
(f)
Defendants claim that “our number of IRCs is a more important metric
than average days to sale due to the impact of IRC capacity on retail units sold” was also
false and misleading. Average days to sale is the number of days between when Carvana
acquires a car either from a customer or in the wholesale market and when they sell the car to
a customer in a retail transaction. The number of IRCs, a purportedly more important metric,
had almost no relation to average days of sale as it was not a measure of how quickly
Carvana sold its vehicles. Defendants viewed average days to sale as a key internal metric
before, during, and after the Class Period. Indeed, in its S-1 registration statement, at the
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 181 of 332
- 178 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
time of its IPO, Carvana defined average days to sale as one of just six “Key Operating
Metrics” because “[o]ur business is dependent upon our ability to expeditiously sell
inventory.” During the Class Period, despite concealing the metric in the 2021 10-K,
Carvana still identified “[r]educe average days to sale” among its primary “strategies
designed to increase our total gross profit per unit.” And even after the Class Period,
Defendants highlighted average days to sale in an August 2023 investor presentation, noting
that it “impacts retail GPU through its impact on the cumulative depreciation on vehicles
before they sell.” In addition, Jenkins described average days to sale is “one of the useful
metrics” to help investors understand changes in Retail GPU. The number of IRCs,
however, was not a “useful metric” to understanding changes in Retail GPU.
242. Statement No. 32: On August 4, 2022, Carvana filed its quarterly report on
Form 10-Q for the period ended June 30, 2022 (the “Q2 2022 10-Q”) with the SEC, signed
by Jenkins and attaching SOX certifications signed by Jenkins and Garcia Junior. In the Q2
2022 10-Q Defendants stated: “As we continue to grow, our number of IRCs is a more
important metric than average days to sale due to the impact of IRC capacity on retail
units sold and the relative stability of average days to sale over the past three years.”
243. As detailed below, Statement No. 32 made by Carvana, Garcia Junior, and
Jenkins omitted material facts that affirmatively created an impression of a state of affairs
that differed in a material way from the one that actually existed.
(a)
As demonstrated in the chart below, contrary to Defendants’ statement
that average days to sale was “relative[ly] stab[le] . . . in recent years,” during Q2 2022,
average days to sale had spiked to 95 days, increasing almost 58% above Q3 2021 and the
recent historical range in the low-60-days range.94 At the time that Statement No. 32 was
94 Defendants’ statement conveyed “relative stability” of the average days to sale metric in
the context of recent historical ranges, which was consistently in the low-60-days range. The
average days to sale in recent years were as follows: 64 in FY 2018; 62 in FY 2019; 58 in Q3
2020 (67 in FY 2020 including the impact of a temporary spike in Q1 and Q2 2020 caused
by the COVID-19 disruption); 60 in Q1 2021; 54 in Q2 2021; and 60 in Q3 2021.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 182 of 332
- 179 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
made on August 4, 2022, 35 days into Q3 2022, Carvana’s average days to sale had spiked
even further above its recent historical norms.
(b)
Defendants’ omission of average days to sale was material. Average
days to sale is the number of days between when Carvana acquires a car either from a
customer or in the wholesale market and when they sell the car to a customer in a retail
transaction. Defendants described that an increase in the average days to sale metric “is
unfavorable for retail GPU because if you hold the car longer, it has more time to
experience depreciation before you sell it, then thereby selling at a lower retail GPU.” As
such, Carvana has always considered average days to sale a key internal metric. For
example, at the time of its IPO, Carvana defined average days to sale as one of just six “Key
Operating Metrics” in its S-1 registration statement. Carvana also described:
Our business is dependent upon our ability to expeditiously sell
inventory. Failure to expeditiously sell our inventory could have a material
adverse effect on our business, sales and results of operations.
. . . An over-supply of used vehicle inventory will generally cause
downward pressure on our product sales prices and margins and increase our
average days to sale.
Average days to sale remained a key internal metric throughout the Class Period. For
example, despite not disclosing the actual metric in the 2021 10-K, Carvana repeated the
Q2 2022 10-Q
August 4, 2022 (35
days into Q3 2022)
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 183 of 332
- 180 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
above disclosure and identified the “[r]educ[tion] average days to sale” as one of its primary
“strategies designed to increase our total gross profit per unit” in the 2021 Form 10-K. After
the Class Period, Carvana continued to highlight the importance of monitoring this metric.
For example, in an August 2023 investor presentation, Defendants highlighted that average
days to sale “impacts retail GPU through its impact on the cumulative depreciation on
vehicles before they sell.” Jenkins also stated during the presentation that average days to
sale is “one of the useful metrics” to help investors understand changes in Retail GPU.
Thus, Defendants clearly viewed average days to sale as a key internal metric before, during,
and after the Class Period.
(c)
As detailed below, Defendants further concealed the fact that the spike
in average days to sale in the second half of FY 2021 and accelerating in early 2022 was the
direct result of Defendants’ fraudulent scheme and course of conduct.
(i)
First, Defendants’ unsustainable increase in purchasing cars from
customers, including low-quality cars, left Carvana with a glut of aging inventory. In Q3
2021, Defendants described “explosive growth in buying cars from customers over the last
two quarters.” This surplus of cars had a direct impact on average days to sale by Q2 2022.
At the end of the Class Period, Defendants disclosed that Carvana needed to “significantly
reduc[e] retail vehicle acquisitions.”
(ii)
Second, Defendants were forced to significantly “meter,” or
intentionally not list inventory for sale in certain markets as a result of its unsustainable
nationwide expansion. Indeed, Defendants later admitted that it metered sales to reduce sales
in “markets with lower profitability due to long distance from inventory (e.g., the Pacific
Northwest).” This had a direct impact on average days to sale by Q4 2021.
(iii)
Third, Defendants admitted that “over the last several months,”
preceding August 2022, Carvana had to implement “buffers in many states to provide
cushion for their title and registration teams to work with the various states to complete
necessary registration paperwork,” which materially slowed retail sales growth and adversely
impacted average days to sale.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 184 of 332
- 181 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
(d)
Near the end of the Class Period, Defendants admitted that, far from
being at a “relative[ly] stab[le]” level, Carvana had to “reduce[] inventory,” which was one
of the “largest drivers” of Carvana’s disappointing retail unit sales decline in Q3 2022.
Carvana further acknowledged that it would be “continuing to normalize our inventory size
and expect to further reduce inventory in Q4 [2022].” In addition, at the end of the Class
Period, Defendants were forced to write off millions of dollars of aged and impaired vehicle
inventory that was worth less than Carvana had paid to acquire it and prepare it for sale. In
fact, Carvana recorded a $52 million retail inventory allowance adjustment, which reduced
Q4 2022 Retail GPU by $598 per retail unit – the equivalent per-unit cost of holding a
vehicle more than 90 days. Carvana also recorded a $5 million wholesale inventory
allowance adjustment and elected to sell certain impaired retail vehicles in the wholesale
market, resulting in a loss of $4 million, which reduced Q4 2022 GPU by an additional $103
per retail unit. Moreover, Defendants later reported a “substantial reduction in our inventory
size,” admitting “[f]rom the end of Q1 to the end of Q4, we reduced our inventory balance by
43%.” Defendants also belatedly acknowledged that “[r]educing inventory size positively
impacts Retail GPU.” Finally, on August 9, 2023, Defendants acknowledged in an investor
presentation:
[T]he first key driver of . . . our recent success on retail GPU, is what I’ll call
normalizing inventory size or really it’s more about normalizing inventory
turn times. . . . Average days of sale is particularly – it’s the number of days
between when we acquire a car either from a customer or in the wholesale
market and when we sell a car to a customer on a completed retail transaction.
And so you can see just as a benchmark, 2021, we were running on
average in the low 60s average across the 4 quarters. We really increased
this metric, which is unfavorable for retail GPU because if you hold the car
longer, it has more time to experience depreciation before you sell it, then
thereby selling at a lower retail GPU. But it stayed really elevated through
2022 when we overbuilt the business for the ultimate sales environment. You
can see it sort of in the mid- to high 90s there through most of 2022.
(e)
Defendants conceded in the August 9, 2023 investor presentation that an
increase in average days to sale over Carvana’s historical benchmark “in the low 60s” was
“unfavorable for retail GPU because if you hold the car longer, it has more time to
experience depreciation before you sell it, then thereby selling at a lower retail GPU.” For
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 185 of 332
- 182 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
example, because daily depreciation in the used car market could exceed $20/day at this
time, a ten-day spike in average days to sale could negatively impact GPU by more than
$200 per vehicle. At the end of the Class Period, Carvana characterized the impact of an
increase in average days to sale, explaining that retail vehicles that were not sold within 90
days of acquisition realized ~$600 of lower Retail GPU compared to retail units sold within
90 days. To put it in perspective, in Q4 2021, $600 represented 40% of Carvana’s retail total
GPU.
(f)
Defendants claim that “our number of IRCs is a more important metric
than average days to sale due to the impact of IRC capacity on retail units sold” was also
false and misleading. Average days to sale is the number of days between when Carvana
acquires a car either from a customer or in the wholesale market and when they sell the car to
a customer in a retail transaction. The number of IRCs, a purportedly more important metric,
had almost no relation to average days of sale as it was not a measure of how quickly
Carvana sold its vehicles. Defendants viewed average days to sale as a key internal metric
before, during, and after the Class Period. Indeed, in its S-1 registration statement, at the
time of its IPO, Carvana defined average days to sale as one of just six “Key Operating
Metrics” because “[o]ur business is dependent upon our ability to expeditiously sell
inventory.” During the Class Period, despite concealing the metric in the 2021 10-K,
Carvana still identified “[r]educe average days to sale” among its primary “strategies
designed to increase our total gross profit per unit.” And even after the Class Period,
Defendants highlighted average days to sale in an August 2023 investor presentation, noting
that it “impacts retail GPU through its impact on the cumulative depreciation on vehicles
before they sell.” In addition, Jenkins described average days to sale as “one of the useful
metrics” to help investors understand changes in Retail GPU. The number of IRCs,
however, was not a “useful metric” to understanding changes in Retail GPU.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 186 of 332
- 183 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
F.
Defendants’ Materially False and Misleading Statements
Regarding Profitability Per Vehicle Sold
244. During the Class Period, Carvana acknowledged that “[t]he Company’s
primary business objective is to sell used vehicles to retail customers and generate a profit
doing so.” Thus, Retail GPU, which measured the gross profit Carvana earned per retail
vehicle sold (before adding in other ancillary revenue and profit streams such as warranty,
insurance, and financing), was a critical metric to investors during the Class Period.
Accordingly, throughout the Class Period, Defendants reported Carvana’s Retail GPU in
each of Carvana’s earnings calls and shareholder letters.
245. Defendants reported positive Retail GPU throughout the Class Period and
assured investors that its retail vehicle sales were profitable. However, as alleged herein,
Defendants concealed Carvana’s actual per-vehicle profitability – or lack thereof. Carvana
did so by: (i) excluding certain per-vehicle operations expenses from its calculation of Retail
GPU; and (ii) not disclosing or quantifying these excluded costs separately so investors
could decipher Carvana’s “unit economics” (i.e., overall profitability) of retail sales on their
own. These per-vehicle operations expenses, which were only separately broken out and
quantified for the first time after the Class Period, included material costs associated with
completing a retail sale, such as the full cost to ship a car to a customer and title and
registration costs. By intentionally hiding these significant costs, investors were left to rely
on the positive Retail GPU reported by Defendants as the sole measure of Carvana’s retail
sales profitability. However, had Carvana included these operations expenses in its Retail
GPU calculation or separately disclosed and quantified the costs that were excluded from
Retail GPU, it would have revealed to investors that, on average, Carvana lost money on
every retail car sold. This created an impression of a state affairs (positive per-unit
profitability) materially different from the one that actually existed (negative per-unit
profitability).
246. To be sure, Defendants could choose how to calculate Carvana’s Retail GPU
metric and whether to categorize certain expenses as SG&A, so as to exclude them from the
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 187 of 332
- 184 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Retail GPU metric it touted to investors. However, touting positive Retail GPU while failing
to provide any meaningful disclosures or quantification of the per-vehicle operations
expenses (i.e., the costs to complete a retail sale, which Defendants themselves viewed as a
key variable in measuring retail “unit economics” and profitability) that Defendants excluded
from Carvana’s Retail GPU metric created a misleading impression of the state of affairs that
existed at the time. The omission of material information regarding Carvana’s operations
expenses rendered numerous Class Period statements materially false and misleading, as
described below.
247. Statement No. 33: On May 6, 2020, Jenkins stated on the Carvana Q1 2020
earnings call: “Our growth in GPU was driven by strong retail GPU, which increased $299
to $1,581, our highest ever, driven primarily by buying cars from customers.”
248. As detailed below, Statement No. 33 made by Carvana and Jenkins omitted
material facts that affirmatively created an impression of a state of affairs that differed in a
material way from the one that actually existed.
(a)
During the Class Period, Carvana acknowledged that “[t]he Company’s
primary business objective is to sell used vehicles to retail customers and generate a profit
doing so.” Thus, Retail GPU, which measured the gross profit Carvana earned per retail
vehicle sold (before adding in other ancillary revenue and profit streams such as warranty,
insurance, and financing), was a critical metric to investors during the Class Period.
Accordingly, on the Q1 2020 earnings call, Defendants reported a “strong retail GPU” of
$1,581, which investors viewed as a proxy for Carvana’s retail “unit economics” (i.e., per-
vehicle profitability). However, as alleged herein, Defendants concealed Carvana’s actual
per-vehicle profitability – or lack thereof. Carvana did so by: (i) excluding certain per-
vehicle operations expenses from its calculation of Retail GPU; and (ii) not disclosing or
quantifying these excluded costs separately so investors could decipher Carvana’s overall
profitability of retail sales on their own. These per-vehicle operations expenses, which were
only separately broken out and quantified for the first time after the Class Period, included
material costs associated with completing a retail sale, such as the full cost to ship a car to a
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 188 of 332
- 185 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
customer and title and registration costs. By intentionally hiding these significant costs,
investors were left to rely on the Retail GPU reported by Defendants (a positive Retail GPU
of $1,581) as the sole measure of Carvana’s retail “unit economics” (i.e., per-vehicle
profitability). However, had Carvana included these operations expenses in its Retail GPU
calculation or separately disclosed and quantified the per-vehicle costs associated with
completing retail sales that had been excluded from reported Retail GPU, Carvana would
have revealed to investors that, on average, it generated negative unit economics of more
than $840 per retail unit sold in Q1 2020, as shown in the chart below. This created an
impression of a state affairs (positive per-vehicle profitability) materially different from the
one that actually existed (negative per-vehicle profitability).
(b)
After the Class Period, Carvana admitted in its Q3 2023 Shareholder
Letter and November 2, 2023 Cost Structure Details presentation that the operations
95 Actual total expense per retail unit sold reflects the cost of sales per retail unit sold (as
disclosed) in addition to the undisclosed operations expense per retail unit sold.
96 Actual loss per retail unit sold reflects the average vehicle selling price per retail unit sold
(as disclosed) less the undisclosed actual total expense per retail unit sold.
Q1 2020
Disclosed Average selling
price per retail unit sold
$18,393
Disclosed Cost of sales per
retail unit sold
$16,812
Disclosed Retail GPU
$1,581
Undisclosed
Operations
expense per retail unit sold
~$2,425
Undisclosed Actual total
expense per retail unit sold95
~$19,237
Undisclosed Actual negative
“unit economics” per retail
unit sold 96
~($844)
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 189 of 332
- 186 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
expenses it had concealed from investors during the Class Period were a “key driver[]” of
“unit economics” (i.e., per-vehicle profitability).97 Specifically, Defendants admitted that,
internally, they viewed “two key drivers” of Carvana’s “unit economics” (i.e., profitability
per vehicle): (i) retail cost of sales (costs which Carvana included in Retail GPU during the
Class Period); and (ii) “operations expenses per retail unit sold” (costs which Carvana
excluded from Retail GPU during the Class Period and which had never previously been
separately disclosed or quantified for investors). Defendants further admitted that, although
these vehicle operations expenses had previously been buried and comingled in “SG&A”
expenses and excluded from Retail GPU, they were not fixed or corporate overhead expenses
like typical SG&A costs. Rather, Defendants admitted that vehicle operations costs were the
direct per-vehicle costs of completing a vehicle sale (“[o]perations expenses include the
fulfillment, customer service, and transaction expenses associated with completing retail and
wholesale vehicle sales”). Defendants acknowledged that these previously undisclosed and
unquantified vehicle “operations expenses” included:
“customer care, multi-car logistics, last-mile delivery, and other
operations payroll” associated with completing retail vehicle sales;
“[n]on-payroll logistics expenses, including fuel, repairs and
maintenance, and third-party transport services” associated with
completing retail vehicle sales; and
“[t]ransaction and other expenses, including limited warranty, title and
registration, and finance platform expenses” associated with
completing retail vehicle sales.
97 After disclosing and quantifying operations expenses for the first time in November
2023, Jenkins acknowledged that the additional disclosure “allows a much deeper dive into
the economics of Carvana.” Jenkins added: “we’ve really been focused on . . . improving the
unit economics of the business. In other words, driving down variable costs per car,” which
he confirmed included “operations expenses, which are the more variable component of our
selling, general and administrative expenses.” Jenkins continued: “And the reason that we’re
focused on that is obviously, the more gains we make on reducing variable cost per car
today, the more profitable our growth can be in the future . . . .” Jenkins also described
“driving down operational expenses per unit” as one of Carvana’s “profitability initiatives,
and in particular, . . . our unit economics initiatives.” Thus, according to Jenkins, operations
expenses were a critical component of assessing Carvana’s profitability.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 190 of 332
- 187 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
(c)
Defendants’ concealment of “operations expenses” when describing
Carvana’s retail profitability was especially misleading in light of the fact that Defendants
internally viewed “the underlying costs of completing a sale” as a critical component of
Carvana’s actual retail vehicle profitability. Indeed, as Garcia Junior later acknowledged:
When we think about trying to aim for more profitable sales, what does that
mean? There’s certainly variability in the kind of gross profit associated with
different types of sales. . . .
And then I think there’s a number of other dynamics kind of across car
type, et cetera. There’s also kind of variation in the underlying costs of
completing a sale.
Because Defendants did not specifically break out operations expenses (i.e., the underlying
costs of completing a sale) and instead buried and comingled them among SG&A expenses,
investors could not calculate the actual overall profitability of Carvana’s retail vehicle sales,
like Garcia Junior did internally. After the Class Period, in November 2023, Defendants
further admitted that, internally, they viewed operations expenses (i.e., costs associated with
completing a retail vehicle sale) as a “key driver[]” of “unit economics” (i.e., per-vehicle
profitability). Garcia Junior also described Carvana’s disclosure of operations expenses in
November 2023 as “clear evidence” of Carvana’s potential profitability.
(d)
During Q1 2020, Defendants included per-vehicle operations expenses
in SG&A expenses (rather than Retail GPU), but they did not separately disclose or quantify
these expenses in a manner that would have allowed investors to decipher the overall
profitability (i.e., “unit economics”) of Carvana’s retail sales on their own. In fact, as
detailed below, the disclosures Defendants did make regarding SG&A expenses during Q1
2020 did more to mislead investors than to provide transparency around Carvana’s true
profitability per retail vehicle sold (i.e., “unit economics”).
(i)
First, Defendants’ Class Period description of SG&A expenses
was so vague that investors could not have discerned that material operations expenses
pertaining to Carvana’s actual per-vehicle profitability (i.e., “unit economics”) were included
in SG&A and excluded from Carvana’s Retail GPU calculation. For example, during the
Class Period, Defendants emphasized that SG&A costs consisted primarily of corporate and
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 191 of 332
- 188 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
overhead expenses, as is common across all companies.98 Defendants, however, omitted the
fact that 46% of its total reported SG&A costs were actually not corporate and overhead
expenses but variable “[o]perations expenses” which, as they later revealed, “include[d] the
fulfillment, customer service, and transaction expenses associated with completing retail
and wholesale vehicle sales.”99 Defendants’ intentionally opaque SG&A disclosures kept
analysts and investors in the dark. In fact, numerous analysts referred to Carvana’s SG&A
disclosures as a “black box.” For example, on April 27, 2020, a Morgan Stanley analyst
noted “[w]e (and the investors we spoke with) continue to look for more disclosure on
Other SG&A . . . [it] is a bit of a ‘black box.’” Likewise, on February 26, 2021, a J.P.
Morgan analyst referred to Carvana’s disclosure of SG&A expenses as “[t]he big ‘black
box.’”
(ii)
Second, beyond Defendants’ vague general description of SG&A
expenses, any references to certain specific expenses that were included in SG&A during the
Class Period also painted a materially incomplete picture of which operations expenses had
been excluded from Carvana’s Retail GPU calculation. For example, Defendants broke out
SG&A into five high-level categories during the Class Period, with general overhead
categories like compensation (“all payroll and related costs, including benefits, payroll taxes,
and equity-based compensation”), advertising, and “other” (including “general and
administrative expenses such as IT expenses, corporate occupancy, professional services and
insurance”), making up over 90% of the total reported SG&A costs in Q1 2020.100 The only
98 According to Carvana’s Q1 2020 10-Q: “[S]elling, general and administrative (‘SG&A’)
expenses include expenses associated with advertising and providing customer service to
customers, operating our vending machines and hubs, operating our logistics and
fulfillment network and other corporate overhead expenses, including expenses associated
with information technology, product development, engineering, legal, accounting,
finance, and business development.”
99 46% represents Q1 2020 total SG&A of $275.7 million compared to Q1 2020
approximate total operations expense of $127.1 million.
100 Q1 2020 10-Q: Compensation and benefits, advertising, market occupancy, logistics, and
other.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 192 of 332
- 189 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
specific SG&A category disclosed by Defendants that was loosely tied to Carvana’s post-
Class Period disclosure of per-vehicle operations expenses (i.e., expenses that were directly
“associated with completing retail and wholesale vehicle sales”) was “logistics.” But even
the logistics disclosure was inconsistent with Carvana’s post-Class Period disclosures
revealing actual per-vehicle operations expenses.101
(iii)
Third, the quantification of any specifically identified operations
expenses within SG&A during Q4 2021 would have certainly left investors with a false
impression of which expenses had been excluded from Carvana’s Retail GPU calculation
and, thus, what Carvana’s actual per-vehicle profitability (i.e., “unit economics”) looked like
during Q4 2021. For example, the only specific retail vehicle operations expense Defendants
obliquely quantified during the Class Period was “logistics” costs, which Defendants vaguely
described as encompassing outbound shipping costs. Carvana disclosed logistics costs of
$18.9 million in Q1 2020, or $360 per retail vehicle sold. Meanwhile, operations expense
per retail vehicle sold totaled approximately $127.1 million in Q1 2020, or $2,425 per
vehicle sold. Defendants did not specifically quantify any other specific vehicle operations
expenses in Carvana’s Class Period SG&A disclosures.102 Thus, at most, investors would
have been specifically aware of just $360, less than 15%, of the $2,425 in actual per-vehicle
operations expenses (i.e., costs “associated with completing retail . . . vehicle sales”) that
materially impacted Carvana’s retail vehicles sales profitability (“unit economics”).
101 Carvana disclosed that the logistics SG&A category, also disclosed as shipping and
handling, included “fuel, maintenance and depreciation related to operating our own
transportation fleet, and third party transportation fees.” This disclosure, however, was
inconsistent with Carvana’s post-Class Period disclosures revealing operations expenses
because: (i) Carvana included depreciation and amortization expenses in the SG&A logistics
category – these were overhead costs rather than per-vehicle shipping costs and, thus, were
specifically excluded from operations expenses; and (ii) Carvana excluded compensation
associated with outbound shipping from the SG&A logistics category but included these
expenses in the per-vehicle shipping costs it incurred and, thus, in operations expenses
disclosed after the Class Period.
102 Title and registration expenses were listed among expenses in the “Other” SG&A
category, but were not separately quantified.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 193 of 332
- 190 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
(iv)
Fourth, Defendants’ Class Period SG&A disclosures did not
break out fixed versus variable SG&A expenses. Like most companies, Carvana incurred
variable expenses (incremental expenses on each unit sold) and overhead expenses (fixed
expenses that did not significantly change based on the number of units sold). Operations
expenses, i.e., variable costs incurred in completing retail sales, increased with each
incremental vehicle sold and, thus, had a direct impact on per-vehicle profitability.
Meanwhile, fixed SG&A expenses, such as corporate overhead expenses, did not directly
impact the amount of profit Carvana earned on each vehicle it sold. During the Class Period,
analysts asked Defendants for more clarity on per-vehicle operations expenses (i.e., variable
versus fixed SG&A costs). For example, on the Q4 2020 earnings call, a Wells Fargo
analyst asked Jenkins: “[C]an you talk about incremental SG&A per unit, and walk me
through how to think about the SG&A . . .? And what is expected to be a fixed cost, . . . and
then how much would be a variable cost?” Tellingly, Jenkins declined to provide a clear
response to the question. On February 26, 2021, a J.P. Morgan analyst expressed concern
over the lack of disclosure, stating “we are concerned if this bucket of SG&A is starting to
become more variable vs fixed in nature.” When Carvana finally disclosed operations
expense per retail unit in November 2023, a J.P. Morgan analyst commented that “the
company finally provided disclosure on fixed vs. variable SG&A expenses.” The same
analyst revealed that, due to the opaque nature of Carvana’s Class Period SG&A disclosures,
his estimate of fixed versus variable costs was significantly off: “[T]he implied fixed cost
mix of ~55-60% . . . was lower than our prior assumption of ~70% that was driving our
[earnings model].” In other words, the analyst had viewed Carvana’s SG&A expenses as
predominantly fixed and was surprised to learn that, in reality, a significantly higher
proportion of Carvana’s SG&A expenses were variable operations expenses, which
increased with each incremental vehicle sold and negatively impacted Carvana’s retail
profitability (i.e., “unit economics”).
(e)
The operations expenses that were excluded from Retail GPU (and not
separately disclosed or quantified to allow investors to decipher Carvana’s overall
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 194 of 332
- 191 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
profitability of retail sales on their own) were directly tied to problems stemming from
Carvana’s undisclosed growth-at-any-cost scheme, as described herein. For example, by
concealing a complete picture of “multi-car logistics, last-mile delivery . . . [n]on-payroll
logistics expenses . . . third-party transport services . . . [t]ransaction . . . expenses, [and]
title and registration . . . expenses” that “were associated with completing retail . . . vehicle
sales,” Defendants were able to conceal numerous problems from investors, including that:
(i) Carvana was boosting retail sales with unprofitable sales in distant markets; (ii) Carvana’s
rapid increase in buying cars from customers left it with a glut of low-quality cars that were
causing massive logistics constraints and costs; and (iii) Carvana was experiencing
significant title and registration issues.
(f)
While Defendants stated that the increase in Q1 2020 Retail GPU was
“driven primarily by buying cars from customers,” they omitted the fact that buying cars
from customers was adversely impacting operations expenses and unit economics – another
key measure of actual retail profitability. As alleged herein, the rapid increase in buying
low-quality cars from customers caused a significant spike in wholesale vehicle sales. In Q1
2020, wholesale sales growth had spiked 61% year-over-year, far outpacing retail sales
growth. By Q4 2020, wholesale growth reached over 100%. Carvana later admitted that it
experienced significant logistics constraints and costs as a result of buying cars from
customers and its corresponding surplus of wholesale vehicle sales. For example, in
Carvana’s May 2022 Operating Plan, Defendants admitted the need to “[r]apidly reduce
SG&A expense per retail unit sold” (i.e., operations expenses). Defendants admitted that
“[b]uying cars from customers growth” led to “additional wholesale volume” which, in turn,
led to “constraints in our nationwide logistics network” and a spike in SG&A logistics costs.
Defendants further disclosed that “wholesale units acquired from customers have . . .
increased complexity in our multi-car logistics network.” Nevertheless, the majority of the
negative cost impacts of these logistics constraints were not reflected in Carvana’s Q1 2020
Retail GPU because Defendants categorized many of these logistics costs as operations
expenses, which were buried in SG&A and not separately broken out. Meanwhile,
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 195 of 332
- 192 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
unbeknownst to investors, these costs did contribute to the $2,425 in undisclosed “operations
expense per retail unit sold” in Q1 2020 and did negatively impact Carvana’s retail unit
economics. Thus, it was materially misleading to describe buying cars from customers as the
main “driver” of Carvana’s purported improvement in Retail GPU – the primary metric
Defendants provided for retail profitability – while simultaneously concealing that buying
cars from customers was also adversely impacting another key measure of profitability: unit
economics per retail vehicle sold.
(g)
The fact that Defendants created a misleading impression as to
Carvana’s actual retail sales profitability, or lack thereof, during the Class Period is further
supported by a series of revelations, and corresponding analyst reactions, that began during
2022. First, in April 2022, Carvana disclosed a quarterly loss of $506 million and admitted it
needed to purchase ADESA to “improve our logistics network,” and bring down “shipping
distances, times, and costs.” Analysts commented on these revelations. For example, a J.P.
Morgan analyst labeled the news “‘confidence shattering’” and commented on Carvana’s
opaque SG&A disclosures, which encompassed operations expenses, stating: “‘we do not
believe Carvana (CVNA) is likely to get a free pass on SG&A any longer.’” Then, in May
2022, Carvana announced that it was laying off approximately 2,500 employees “primarily
in operational groups. . .” and issued an updated operating plan that emphasized its need to
“[r]apidly reduce SG&A expense per retail unit sold” (i.e., operations expenses) and make
profitability a priority. Analysts commented on this surprising news. For example, on
May 21, 2022, Forbes cited to several industry analysts who described Carvana’s
“spendthrift business, whose growth-at-all-costs mentality undermined business operations
and sowed the seeds of its recent layoffs.” In November 2022, Carvana issued disappointing
Q3 2022 financial results and revealed that, unbeknownst to investors, Carvana had been
“frequently acquir[ing]” retail sales “that were less profitable in the immediate period,”
including “less profitable sales” in “markets with lower profitability due to long distance
from inventory.” Carvana further admitted it was now taking actions “to improve
profitability.” Analysts and the media reported on the news regarding Carvana’s retail sales
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 196 of 332
- 193 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
profitability. For example, a Needham & Company analyst noted, “‘[t]he path forward for
Carvana is to sell as many cars as possible, but to do so on a profitable basis, versus prior it
was more about selling as many cars as possible.’” Another media outlet told investors to
“[s]tep back and think about [Carvana’s disclosure] for a moment” because “[i]t means that
the company had historically grown in these markets even though they weren’t all that
attractive. It was growth for growth’s sake and not necessarily growth to improve
profitability.” Finally, on the last day of the Class Period, Defendants revealed that Carvana
was intentionally foregoing retail sales growth for the foreseeable future to instead focus on
driving “positive unit economics” (i.e., retail sales that were actually profitable). Analysts
commented on this disclosure. For example, a Morgan Stanley analyst, on February 24,
2023, explained that “4Q was a sizeable miss,” and commented on Carvana’s “[s]hift in
strategy away from growth and towards EBITDA profitability.” The same day, a RBC
Capital Markets analyst also noted that Carvana’s GPU “miss[ed] Street expectations by
32% . . . due in large part to . . . lower unit economics from cars.” Additionally, a Stephens
analyst noted “we do not foresee the company achieving its targeting unit economics this
year.”
249. Statement No. 34: On August 5, 2020, Jenkins stated on the Carvana Q2 2020
earnings call:
Retail GPU was $1,190 . . . .
*
*
*
[T]here are definitely some exciting trends in retail GPU. I think the
most exciting is that, as I mentioned earlier, in July, we bought more than
100% as many cars from customers as we sold to customers. . . .
I think that’s obviously a positive, that drives wholesale GPU, that
drives incremental retail GPU, because cars that you acquire from
customers are typically more profitable than cars that you acquire at auction.
250. As detailed below, Statement No. 34 made by Carvana and Jenkins omitted
material facts that affirmatively created an impression of a state of affairs that differed in a
material way from the one that actually existed.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 197 of 332
- 194 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
(a)
During the Class Period, Carvana acknowledged that “[t]he Company’s
primary business objective is to sell used vehicles to retail customers and generate a profit
doing so.” Thus, Retail GPU, which measured the gross profit Carvana earned per retail
vehicle sold (before adding in other ancillary revenue and profit streams such as warranty,
insurance, and financing), was a critical metric to investors during the Class Period.
Accordingly, on the Q2 2020 earnings call, Defendants reported positive “Retail GPU of
$1,190” and described “exciting trends in retail GPU,” which investors viewed as a proxy
for Carvana’s “unit economics” (i.e., per-vehicle profitability). However, as alleged herein,
Defendants concealed Carvana’s actual per-vehicle profitability – or lack thereof. Carvana
did so by: (i) excluding certain per-vehicle operations expenses from its calculation of Retail
GPU; and (ii) not disclosing or quantifying these excluded costs separately so investors
could decipher Carvana’s overall profitability of retail sales on their own. These per-vehicle
operations expenses, which were only separately broken out and quantified for the first time
after the Class Period, included material costs associated with completing a retail sale, such
as the full cost to ship a car to a customer and title and registration costs. By intentionally
hiding these significant costs, investors were left to rely on the Retail GPU reported by
Defendants (a positive Retail GPU of $1,190) as the sole measure of Carvana’s retail “unit
economics” (i.e., per-vehicle profitability). However, had Carvana included these operations
expenses in its Retail GPU calculation or separately disclosed and quantified the per-vehicle
costs associated with completing retail sales that had been excluded from reported Retail
GPU, Carvana would have revealed to investors that, on average, it generated negative
unit economics of more than $680 per retail unit sold in Q2 2020, as shown in the chart
below. This created an impression of a state affairs (positive per-vehicle profitability)
materially different from the one that actually existed (negative per-vehicle profitability).
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 198 of 332
- 195 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
(b)
After the Class Period, Carvana admitted in its Q3 2023 Shareholder
Letter and November 2, 2023 Cost Structure Details presentation that the operations
expenses it had concealed from investors during the Class Period were a “key driver[]” of
“unit economics” (i.e., per-vehicle profitability).105 Specifically, Defendants admitted that,
internally, they viewed “two key drivers” of Carvana’s “unit economics” (i.e., profitability
per retail vehicle sold): (i) retail cost of sales (costs which Carvana included in Retail GPU
103 Actual total expense per retail unit sold reflects the cost of sales per retail unit sold (as
disclosed) in addition to the undisclosed operations expense per retail unit sold.
104 Actual loss per retail unit sold reflects the average vehicle selling price per retail unit sold
(as disclosed) less the undisclosed actual total expense per retail unit sold.
105 After disclosing and quantifying operations expenses for the first time in November
2023, Jenkins acknowledged that the additional disclosure “allows a much deeper dive into
the economics of Carvana.” Jenkins added: “we’ve really been focused on . . . improving the
unit economics of the business. In other words, driving down variable costs per car,” which
he confirmed included “operations expenses, which are the more variable component of our
selling, general and administrative expenses.” Jenkins continued: “And the reason that we’re
focused on that is obviously, the more gains we make on reducing variable cost per car
today, the more profitable our growth can be in the future . . . .” Jenkins also described
“driving down operational expenses per unit” as one of Carvana’s “profitability initiatives,
and in particular, . . . our unit economics initiatives.” Thus, according to Jenkins, operations
expenses were a critical component of assessing Carvana’s profitability.
Q2 2020
Disclosed Average selling
price per retail unit sold
$18,001
Disclosed Cost of sales per
retail unit sold
$16,811
Disclosed Retail GPU
$1,190
Undisclosed Operations
expense per retail unit sold
~$1,875
Undisclosed Actual total
expense per retail unit
sold103
~$18,686
Undisclosed Actual
negative “unit economics”
per retail unit sold104
~($685)
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 199 of 332
- 196 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
during the Class Period); and (ii) “operations expenses per retail unit sold” (costs which
Carvana excluded from Retail GPU during the Class Period and which had never previously
been separately disclosed or quantified for investors). Defendants further admitted that,
although these vehicle operations expenses had previously been buried and comingled in
“SG&A” expenses and excluded from Retail GPU, they were not fixed or corporate
overhead expenses like typical SG&A costs. Rather, Defendants admitted that vehicle
operations costs were the direct per-vehicle costs of completing a vehicle sale (“[o]perations
expenses include the fulfillment, customer service, and transaction expenses associated with
completing retail and wholesale vehicle sales”). Defendants acknowledged that these
previously undisclosed and unquantified vehicle “operations expenses” included:
“customer care, multi-car logistics, last-mile delivery, and other
operations payroll” associated with completing retail vehicle sales;
“non-payroll logistics expenses, including fuel, repairs and
maintenance, and third-party transport services” associated with
completing retail vehicle sales; and
“[t]ransaction and other expenses, including limited warranty, title and
registration, and finance platform expenses” associated with
completing retail vehicle sales.
(c)
Defendants’ concealment of “operations expenses” when describing
Carvana’s retail profitability was especially misleading in light of the fact that Defendants
internally viewed “the underlying costs of completing a sale” as a critical component of
Carvana’s actual retail vehicle profitability. Indeed, as Garcia Junior later acknowledged:
When we think about trying to aim for more profitable sales, what does that
mean? There’s certainly variability in the kind of gross profit associated with
different types of sales. . . .
And then I think there’s a number of other dynamics kind of across car
type, et cetera. There’s also kind of variation in the underlying costs of
completing a sale.
Because Defendants did not specifically break out operations expenses (i.e., the underlying
costs of completing a sale) and instead buried and comingled them among SG&A expenses,
investors could not calculate the actual overall profitability of Carvana’s retail vehicle sales,
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 200 of 332
- 197 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
like Garcia Junior did internally. After the Class Period, in November 2023, Defendants
further admitted that, internally, they viewed operations expenses (i.e., costs associated with
completing a retail vehicle sale) as a “key driver[]” of “unit economics” (i.e., per-vehicle
profitability). Garcia Junior also described Carvana’s disclosure of operations expenses in
November 2023 as “clear evidence” of Carvana’s potential profitability.
(d)
During Q2 2020, Defendants included per-vehicle operations expenses
in SG&A expenses (rather than Retail GPU), but they did not separately disclose or quantify
these expenses in a manner that would have allowed investors to decipher the overall
profitability (i.e., “unit economics”) of Carvana’s retail sales on their own. In fact, as
detailed below, the disclosures Defendants did make regarding SG&A expenses during Q2
2020 did more to mislead investors than to provide transparency around Carvana’s true
profitability per retail vehicle sold (i.e., “unit economics”).
(i)
First, Defendants’ Class Period description of SG&A expenses
was so vague that investors could not have discerned that material operations expenses
pertaining to Carvana’s actual per-vehicle profitability (i.e., “unit economics”) were included
in SG&A and excluded from Carvana’s Retail GPU calculation. For example, during the
Class Period, Defendants emphasized that SG&A costs consisted primarily of corporate and
overhead expenses, as is common across all companies.106 Defendants, however, omitted the
fact that 43% of its total reported SG&A costs were actually not corporate and overhead
expenses but variable “[o]perations expenses” which, as they later revealed, “include[d] the
fulfillment, customer service, and transaction expenses associated with completing retail
and wholesale vehicle sales.”107 Defendants’ intentionally opaque SG&A disclosures kept
106 According to Carvana’s Q2 2020 10-Q: “[S]elling, general and administrative (‘SG&A’)
expenses include expenses associated with advertising and providing customer service to
customers, operating our vending machines and hubs, operating our logistics and
fulfillment network and other corporate overhead expenses, including expenses associated
with information technology, product development, engineering, legal, accounting,
finance, and business development.”
107 43% represents Q2 2020 total SG&A of $239.9 million compared to Q2 2020
approximate total operations expense of $103.3 million.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 201 of 332
- 198 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
analysts and investors in the dark. In fact, numerous analysts referred to Carvana’s SG&A
disclosures as a “black box.” For example, on April 27, 2020, a Morgan Stanley analyst
noted “[w]e (and the investors we spoke with) continue to look for more disclosure on
Other SG&A . . . [it] is a bit of a ‘black box.’” Likewise, on February 26, 2021, a J.P.
Morgan analyst referred to Carvana’s disclosure of SG&A expenses as “[t]he big ‘black
box.’”
(ii)
Second, beyond Defendants’ vague general description of SG&A
expenses, any references to certain specific expenses that were included in SG&A during the
Class Period also painted a materially incomplete picture of which operations expenses had
been excluded from Carvana’s Retail GPU calculation. For example, Defendants broke out
SG&A into five high-level categories during the Class Period, with general overhead
categories like compensation (“all payroll and related costs, including benefits, payroll taxes,
and equity-based compensation”), advertising, and “other” (including “general and
administrative expenses such as IT expenses, corporate occupancy, professional services and
insurance”), making up over 90% of the total reported SG&A costs in Q2 2020.108 The only
specific SG&A category disclosed by Defendants that was loosely tied to Carvana’s post-
Class Period disclosure of per-vehicle operations expenses (i.e., expenses that were directly
“associated with completing retail and wholesale vehicle sales”) was “logistics.” But even
the logistics disclosure was inconsistent with Carvana’s post-Class Period disclosures
revealing actual per-vehicle operations expenses.109
108 Q2 2020 10-Q: Compensation and benefits, advertising, market occupancy, logistics, and
other.
109 Carvana disclosed that the logistics SG&A category, also disclosed as shipping and
handling, included “fuel, maintenance and depreciation related to operating our own
transportation fleet, and third party transportation fees.” This disclosure, however, was
inconsistent with Carvana’s post-Class Period disclosures revealing operations expenses
because: (i) Carvana included depreciation and amortization expenses in the SG&A logistics
category – these were overhead costs rather than per-vehicle shipping costs and, thus, were
specifically excluded from operations expenses; and (ii) Carvana excluded compensation
associated with outbound shipping from the SG&A logistics category but included these
expenses in the per-vehicle shipping costs it incurred and, thus, in operations expenses
disclosed after the Class Period.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 202 of 332
- 199 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
(iii)
Third, the quantification of any specifically identified operations
expenses within SG&A during the Class Period would have certainly left investors with a
false impression of which expenses had been excluded from Carvana’s Retail GPU
calculation and, thus, what Carvana’s actual per-vehicle profitability (i.e., unit economics)
looked like during the Class Period. For example, the only specific retail vehicle operations
expense Defendants obliquely quantified during the Class Period was “logistics” costs,
which Defendants vaguely described as encompassing outbound shipping costs. Carvana
disclosed logistics costs of $16.7 million in Q2 2020, or $303 per retail vehicle sold.
Meanwhile, operations expense per retail vehicle sold totaled approximately $103.3 million
in Q2 2020, or $1,875 per vehicle sold. Defendants did not specifically quantify any other
specific vehicle operations expenses in Carvana’s Class Period SG&A disclosures.110 Thus,
at most, investors would have been specifically aware of just $303, approximately 16%, of
the $1,875 in actual per-vehicle operations expenses (i.e., costs “associated with completing
retail . . . vehicle sales”) that materially impacted Carvana’s retail vehicles sales profitability
(“unit economics”).
(iv)
Fourth, Defendants’ Class Period SG&A disclosures did not
break out fixed versus variable SG&A expenses. Like most companies, Carvana incurred
variable expenses (incremental expenses on each unit sold) and overhead expenses (fixed
expenses that did not significantly change based on the number of units sold). Operations
expenses, i.e., variable costs incurred in completing retail sales, increased with each
incremental vehicle sold and, thus, had a direct impact on per-vehicle profitability.
Meanwhile, fixed SG&A expenses, such as corporate overhead expenses, did not directly
impact the amount of profit Carvana earned on each vehicle it sold. During the Class Period,
analysts asked Defendants for more clarity on per-vehicle operations expenses (i.e., variable
versus fixed SG&A costs). For example, on the Q4 2020 earnings call, a Wells Fargo
analyst asked Jenkins: “[C]an you talk about incremental SG&A per unit, and walk me
110 Title and registration expenses were listed among expenses in the “Other” SG&A
category, but were not separately quantified.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 203 of 332
- 200 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
through how to think about the SG&A . . . ? And what is expected to be a fixed cost, . . . and
then how much would be a variable cost?” Tellingly, Jenkins declined to provide a clear
response to the question. On February 26, 2021, a J.P. Morgan analyst expressed concern
over the lack of disclosure, stating “we are concerned if this bucket of SG&A is starting to
become more variable vs fixed in nature.” When Carvana finally disclosed operations
expense per retail unit in November 2023, a J.P. Morgan analyst commented that “the
company finally provided disclosure on fixed vs. variable SG&A expenses.” The same
analyst revealed that, due to the opaque nature of Carvana’s Class Period SG&A disclosures,
his estimate of fixed versus variable costs was significantly off: “[T]he implied fixed cost
mix of ~55-60% . . . was lower than our prior assumption of ~70% that was driving our
[earnings model].” In other words, the analyst had viewed Carvana’s SG&A expenses as
predominantly fixed and was surprised to learn that, in reality, a significantly higher
proportion of Carvana’s SG&A expenses were variable operations expenses, which
increased with each incremental vehicle sold and negatively impacted Carvana’s retail
profitability (i.e., unit economics).
(e)
The operations expenses that were excluded from Retail GPU (and not
separately disclosed or quantified to allow investors to decipher Carvana’s overall
profitability of retail sales on their own) were directly tied to problems stemming from
Carvana’s undisclosed growth-at-any-cost scheme, as described herein. For example, by
concealing a complete picture of “multi-car logistics, last-mile delivery . . . [n]on-payroll
logistics expenses . . . third-party transport services . . . [t]ransaction . . . expenses, [and]
title and registration . . . expenses” that “were associated with completing retail . . . vehicle
sales,” Defendants were able to conceal numerous problems from investors, including that:
(i) Carvana was boosting retail sales with unprofitable sales in distant markets; (ii) Carvana’s
rapid increase in buying cars from customers left it with a glut of low-quality cars that were
causing massive logistics constraints and costs; and (iii) Carvana was experiencing
significant title and registration issues.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 204 of 332
- 201 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
(f)
In describing what was driving Carvana’s “exciting trends in retail
GPU,” Defendants cited buying cars from customers, claiming “that [it] drives incremental
retail GPU, because cars that you acquire from customers are typically more profitable.”
While Defendants touted the beneficial impact of buying cars from customers, they omitted
the fact that it was also adversely impacting operations expense and unit economics, which
Defendants later admitted was another key measure of actual retail profitability. As alleged
herein, the rapid increase in buying low-quality cars from customers caused a significant
spike in wholesale vehicle sales. Following a temporary pause in buying cars from
customers due to the pandemic, Carvana had resumed purchases as of August 5, 2020, when
Jenkins described the GPU benefits of buying cars from customers. In fact, Carvana
increased its cars bought from customers by 261% during Q3 2020 (when the statement was
made). Carvana later admitted that it experienced significant logistics constraints and costs
stemming from buying cars from customers and the resulting surplus of wholesale vehicle
sales. For example, in Carvana’s May 2022 Operating Plan, Defendants admitted the need to
“[r]apidly reduce SG&A expense per retail unit sold” (i.e., operations expenses). Defendants
admitted that “[b]uying cars from customers growth” led to “additional wholesale volume”
which, in turn, led to “constraints in our nationwide logistics network” and a spike in SG&A
logistics costs. Defendants further disclosed that “wholesale units acquired from customers
have . . . increased complexity in our multi-car logistics network.” Nevertheless, the
majority of the negative cost impacts of these logistics constraints were not reflected in
Carvana’s Q2 2020 Retail GPU because Defendants categorized many of these logistics
costs as operations expenses, which were buried in SG&A and not separately broken out.
Meanwhile, unbeknownst to investors, these costs did contribute to the $1,875 in undisclosed
“operations expense per retail unit sold” in Q2 2020 did negatively impact Carvana’s retail
unit economics. Thus, it was materially misleading to describe buying cars from customers
as the main “driver” of Carvana’s purported improvement in Retail GPU – the primary
metric Defendants provided for retail profitability – while simultaneously concealing that
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 205 of 332
- 202 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
buying cars from customers was also adversely impacting another key measure of
profitability: unit economics per retail vehicle sold.
(g)
The fact that Defendants created a misleading impression as to
Carvana’s actual retail sales profitability, or lack thereof, during the Class Period is further
supported by a series of revelations, and corresponding analyst reactions, that began during
2022. First, in April 2022, Carvana issued disappointing Q1 2022 financial results, which
partially revealed the truth about Carvana’s actual profitability, including the operations
expenses it had concealed from investors. For example, Defendants disclosed a quarterly
loss of $506 million and explained that they needed to purchase ADESA to “improve our
logistics network,” and bring down “shipping distances, times, and costs.” Analysts
commented on these revelations. For example, a J.P. Morgan analyst labeled the news
“‘confidence shattering’” and commented on Carvana’s opaque SG&A disclosures, which
encompassed operations expenses, stating: “‘we do not believe Carvana (CVNA) is likely to
get a free pass on SG&A any longer.’” Then, in May 2022, Carvana announced that it was
laying off approximately 2,500 employees “primarily in operational groups” and issued an
updated operating plan that emphasized the need to “[r]apidly reduce SG&A expense per
retail unit sold” (i.e., operations expenses) and make profitability a priority. Analysts
commented on this surprising news. For example, on May 21, 2022, Forbes cited to several
industry analysts who described Carvana’s “spendthrift business, whose growth-at-all-costs
mentality undermined business operations and sowed the seeds of its recent layoffs.” In
November 2022, Carvana issued disappointing Q3 2022 financial results and revealed that,
unbeknownst to investors, Carvana had been “frequently acquir[ing]” retail sales “that were
less profitable in the immediate period,” including “less profitable sales” in “markets with
lower profitability due to long distance from inventory.” Carvana further admitted it was
now taking actions “to improve profitability.” Analysts and the media reported on the news
regarding Carvana’s retail sales profitability. For example, a Needham & Company analyst
noted, “‘[t]he path forward for Carvana is to sell as many cars as possible, but to do so on a
profitable basis, versus prior it was more about selling as many cars as possible.’” Another
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 206 of 332
- 203 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
media outlet told investors to “[s]tep back and think about [Carvana’s disclosure] for a
moment” because “[i]t means that the company had historically grown in these markets
even though they weren’t all that attractive. It was growth for growth’s sake and not
necessarily growth to improve profitability.” Finally, on the last day of the Class Period,
Defendants revealed that Carvana was intentionally foregoing retail sales growth for the
foreseeable future to instead focus on driving “positive unit economics” (i.e., retail sales that
were actually profitable). Analysts commented on this disclosure. For example, a Morgan
Stanley analyst, on February 24, 2023, explained that “4Q was a sizeable miss,” and
commented on Carvana’s “[s]hift in strategy away from growth and towards EBITDA
profitability.” The same day, a RBC Capital Markets analyst also noted that Carvana’s GPU
“miss[ed] Street expectations by 32% . . . due in large part to . . . lower unit economics from
cars.” Additionally, a Stephens analyst noted “we do not foresee the company achieving its
targeting unit economics this year.”
251. Statement No. 35: On October 29, 2020, Jenkins stated on the Carvana Q3
2020 earnings call:
Retail GPU was $1,857 in Q3, an increase of $552. Growth in retail GPU
was driven by a significant increase in the share of our vehicles sourced
from customers.
252. As detailed below, Statement No. 35 made by Carvana and Jenkins omitted
material facts that affirmatively created an impression of a state of affairs that differed in a
material way from the one that actually existed.
(a)
During the Class Period, Carvana acknowledged that “[t]he Company’s
primary business objective is to sell used vehicles to retail customers and generate a profit
doing so.” Thus, Retail GPU, which measured the gross profit Carvana earned per retail
vehicle sold (before adding in other ancillary revenue and profit streams such as warranty,
insurance, and financing), was a critical metric to investors during the Class Period.
Accordingly, on the Q3 2020 earnings call, Defendants reported positive “Retail GPU [of]
$1,857,” which investors viewed as a proxy for Carvana’s “unit economics” (i.e., per-vehicle
profitability). However, as alleged herein, Defendants concealed Carvana’s actual per-
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 207 of 332
- 204 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
vehicle profitability – or lack thereof. Carvana did so by: (i) excluding certain per-vehicle
operations expenses from its calculation of Retail GPU; and (ii) not disclosing or quantifying
these excluded costs separately so investors could decipher Carvana’s overall profitability of
retail sales on their own. These per-vehicle operations expenses, which were only separately
broken out and quantified for the first time after the Class Period, included material costs
associated with completing a retail sale, such as the full cost to ship a car to a customer and
title and registration costs. By intentionally hiding these significant costs, investors were left
to rely on the Retail GPU reported by Defendants (a positive Retail GPU of $1,857) as the
sole measure of Carvana’s retail “unit economics” (i.e., per-vehicle profitability). However,
had Carvana included these operations expenses in its Retail GPU calculation or separately
disclosed and quantified the per-vehicle costs associated with completing retail sales that had
been excluded from reported Retail GPU, Carvana would have revealed to investors that,
on average, it generated negative unit economics on retail vehicles sales Q3 2020, as
shown in the chart below. This created an impression of a state affairs (positive per-vehicle
profitability) materially different from the one that actually existed (negative per-vehicle
profitability).
Q3 2020
Disclosed Average selling
price per retail unit sold
$20,013
Disclosed Cost of sales per
retail unit sold
$18,156
Disclosed Retail GPU
$1,857
Undisclosed Operations
expense per retail unit sold
~$1,875
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 208 of 332
- 205 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
(b)
After the Class Period, Carvana admitted in its Q3 2023 Shareholder
Letter and November 2, 2023 Cost Structure Details presentation that the operations
expenses it had concealed from investors during the Class Period were a “key driver[]” of
“unit economics” (i.e., per-vehicle profitability).113 Specifically, Defendants admitted that,
internally, they viewed “two key drivers” of Carvana’s “unit economics” (i.e., profitability
per retail vehicle sold): (i) retail cost of sales (costs which Carvana included in Retail GPU
during the Class Period); and (ii) “operations expenses per retail unit sold” (costs which
Carvana excluded from Retail GPU during the Class Period and which had never previously
been separately disclosed or quantified for investors). Defendants further admitted that,
although these vehicle operations expenses had previously been buried and comingled in
“SG&A” expenses and excluded from Retail GPU, they were not fixed or corporate
overhead expenses like typical SG&A costs. Rather, Defendants admitted that vehicle
operations costs were the direct per-vehicle costs of completing a vehicle sale (“[o]perations
expenses include the fulfillment, customer service, and transaction expenses associated with
111 Actual total expense per retail unit sold reflects the cost of sales per retail unit sold (as
disclosed) in addition to the undisclosed operations expense per retail unit sold.
112 Actual loss per retail unit sold reflects the average vehicle selling price per retail unit sold
(as disclosed) less the undisclosed actual total expense per retail unit sold.
113 After disclosing and quantifying operations expenses for the first time in November
2023, Jenkins acknowledged that the additional disclosure “allows a much deeper dive into
the economics of Carvana.” Jenkins added: “we’ve really been focused on . . . improving the
unit economics of the business. In other words, driving down variable costs per car,” which
he confirmed included “operations expenses, which are the more variable component of our
selling, general and administrative expenses.” Jenkins continued: “And the reason that we’re
focused on that is obviously, the more gains we make on reducing variable cost per car
today, the more profitable our growth can be in the future . . . .” Jenkins also described
“driving down operational expenses per unit” as one of Carvana’s “profitability initiatives,
and in particular, . . . our unit economics initiatives.” Thus, according to Jenkins, operations
expenses were a critical component of assessing Carvana’s profitability.
Undisclosed Actual total
expense per retail unit
sold111
~$20,031
Undisclosed Actual
negative “unit economics”
per retail unit sold 112
~($18)
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 209 of 332
- 206 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
completing retail and wholesale vehicle sales”). Defendants acknowledged that these
previously undisclosed and unquantified vehicle “operations expenses” included:
“customer care, multi-car logistics, last-mile delivery, and other
operations payroll” associated with completing retail vehicle sales;
“non-payroll logistics expenses, including fuel, repairs and
maintenance, and third-party transport services” associated with
completing retail vehicle sales; and
“[t]ransaction and other expenses, including limited warranty, title and
registration, and finance platform expenses” associated with
completing retail vehicle sales.
(c)
Defendants’ concealment of “operations expenses” when describing
Carvana’s retail profitability was especially misleading in light of the fact that Defendants
internally viewed “the underlying costs of completing a sale” as a critical component of
Carvana’s actual retail vehicle profitability. Indeed, as Garcia Junior later acknowledged:
When we think about trying to aim for more profitable sales, what does that
mean? There’s certainly variability in the kind of gross profit associated with
different types of sales. . . .
And then I think there’s a number of other dynamics kind of across car
type, et cetera. There’s also kind of variation in the underlying costs of
completing a sale.
Because Defendants did not specifically break out operations expenses (i.e., the underlying
costs of completing a sale) and instead buried and comingled them among SG&A expenses,
investors could not calculate the actual overall profitability of Carvana’s retail vehicle sales,
like Garcia Junior did internally. After the Class Period, in November 2023, Defendants
further admitted that, internally, they viewed operations expenses (i.e., costs associated with
completing a retail vehicle sale) as a “key driver[]” of “unit economics” (i.e., per-vehicle
profitability). Garcia Junior also described Carvana’s disclosure of operations expenses in
November 2023 as “clear evidence” of Carvana’s potential profitability.
(d)
During Q3 2020, Defendants included per-vehicle operations expenses
in SG&A expenses (rather than Retail GPU), but they did not separately disclose or quantify
these expenses in a manner that would have allowed investors to decipher the overall
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 210 of 332
- 207 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
profitability (i.e., “unit economics”) of Carvana’s retail sales on their own. In fact, as
detailed below, the disclosures Defendants did make regarding SG&A expenses during Q3
2020 did more to mislead investors than to provide transparency around Carvana’s true
profitability per retail vehicle sold (i.e., “unit economics”).
(i)
First, Defendants’ Class Period description of SG&A expenses
was so vague that investors could not have discerned that material operations expenses
pertaining to Carvana’s actual per-vehicle profitability (i.e., “unit economics”) were included
in SG&A and excluded from Carvana’s Retail GPU calculation. For example, during the
Class Period, Defendants emphasized that SG&A costs consisted primarily of corporate and
overhead expenses, as is common across all companies.114 Defendants, however, omitted the
fact that 45% of its total reported SG&A costs were actually not corporate and overhead
expenses but variable “[o]perations expenses” which, as they later revealed, “include[d] the
fulfillment, customer service, and transaction expenses associated with completing retail
and wholesale vehicle sales.”115 Defendants’ intentionally opaque SG&A disclosures kept
analysts and investors in the dark. In fact, numerous analysts referred to Carvana’s SG&A
disclosures as a “black box.” For example, on April 27, 2020, a Morgan Stanley analyst
noted “[w]e (and the investors we spoke with) continue to look for more disclosure on
Other SG&A . . . [it] is a bit of a ‘black box.’” Likewise, on February 26, 2021, a J.P.
Morgan analyst referred to Carvana’s disclosure of SG&A expenses as “[t]he big ‘black
box.’”
(ii)
Second, beyond Defendants’ vague general description of SG&A
expenses, any references to certain specific expenses that were included in SG&A during the
114 According to Carvana’s Q2 2020 10-Q: “Selling, general and administrative (‘SG&A’)
expenses include expenses associated with advertising and providing customer service to
customers, operating our vending machines and hubs, operating our logistics and
fulfillment network and other corporate overhead expenses, including expenses associated
with information technology, product development, engineering, legal, accounting,
finance, and business development.”
115 45% represents Q3 2020 total SG&A of $267.8 million compared to Q3 2020
approximate total operations expense of $120.8 million.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 211 of 332
- 208 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Class Period also painted a materially incomplete picture of which operations expenses had
been excluded from Carvana’s Retail GPU calculation. For example, Defendants broke out
SG&A into five high-level categories during the Class Period, with general overhead
categories like compensation (“all payroll and related costs, including benefits, payroll taxes,
and equity-based compensation”), advertising, and “other” (including “general and
administrative expenses such as IT expenses, corporate occupancy, professional services and
insurance”), making up over 89% of the total reported SG&A costs in Q3 2020.116 The only
specific SG&A category disclosed by Defendants that was loosely tied to Carvana’s post-
Class Period disclosure of per-vehicle operations expenses (i.e., expenses that were directly
“associated with completing retail and wholesale vehicle sales”) was “logistics.” But even
the logistics disclosure was inconsistent with Carvana’s post-Class Period disclosures
revealing actual per-vehicle operations expenses.117
(iii)
Third, the quantification of any specifically identified operations
expenses within SG&A during the Class Period would have certainly left investors with a
false impression of which expenses had been excluded from Carvana’s Retail GPU
calculation and, thus, what Carvana’s actual per-vehicle profitability (i.e., unit economics)
looked like during the Class Period. For example, the only specific retail vehicle operations
expense Defendants obliquely quantified during the Class Period was “logistics” costs,
which Defendants vaguely described as encompassing outbound shipping costs. Carvana
disclosed logistics costs of $18.0 million in Q3 2020, or $280 per retail vehicle sold.
116 Q3 2020 10-Q: Compensation and benefits, advertising, market occupancy, logistics, and
other.
117 Carvana disclosed that the logistics SG&A category, also disclosed as shipping and
handling, included “fuel, maintenance and depreciation related to operating our own
transportation fleet, and third party transportation fees.” This disclosure, however, was
inconsistent with Carvana’s post-Class Period disclosures revealing operations expenses
because: (i) Carvana included depreciation and amortization expenses in the SG&A logistics
category – these were overhead costs rather than per-vehicle shipping costs and, thus, were
specifically excluded from operations expenses; and (ii) Carvana excluded compensation
associated with outbound shipping from the SG&A logistics category but included these
expenses in the per-vehicle shipping costs it incurred and, thus, in operations expenses
disclosed after the Class Period.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 212 of 332
- 209 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Meanwhile, operations expense per retail vehicle sold totaled approximately $120.8 million
in Q3 2020, or $1,875 per vehicle sold. Defendants did not specifically quantify any other
specific vehicle operations expenses in Carvana’s Class Period SG&A disclosures.118 Thus,
at most, investors would have been specifically aware of just $280, less than 15%, of the
$1,875 in actual per-vehicle operations expenses (i.e., costs “associated with completing
retail . . . vehicle sales”) that materially impacted Carvana’s retail vehicles sales profitability
(“unit economics”).
(iv)
Fourth, Defendants’ Class Period SG&A disclosures did not
break out fixed versus variable SG&A expenses. Like most companies, Carvana incurred
variable expenses (incremental expenses on each unit sold) and overhead expenses (fixed
expenses that did not significantly change based on the number of units sold). Operations
expenses, i.e., variable costs incurred in completing retail sales, increased with each
incremental vehicle sold and, thus, had a direct impact on per-vehicle profitability.
Meanwhile, fixed SG&A expenses, such as corporate overhead expenses, did not directly
impact the amount of profit Carvana earned on each vehicle it sold. During the Class Period,
analysts asked Defendants for more clarity on per-vehicle operations expenses (i.e., variable
versus fixed SG&A costs). For example, on the Q4 2020 earnings call, a Wells Fargo
analyst asked Jenkins: “[C]an you talk about incremental SG&A per unit, and walk me
through how to think about the SG&A . . .? And what is expected to be a fixed cost, . . . and
then how much would be a variable cost?” Tellingly, Jenkins declined to provide a clear
response to the question. On February 26, 2021, a J.P. Morgan analyst expressed concern
over the lack of disclosure, stating “we are concerned if this bucket of SG&A is starting to
become more variable vs fixed in nature.” When Carvana finally disclosed operations
expense per retail unit in November 2023, a J.P. Morgan analyst commented that “the
company finally provided disclosure on fixed vs. variable SG&A expenses.” The same
analyst revealed that, due to the opaque nature of Carvana’s Class Period SG&A disclosures,
118 Title and registration expenses were listed among expenses in the “Other” SG&A
category, but were not separately quantified.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 213 of 332
- 210 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
his estimate of fixed versus variable costs was significantly off: “[T]he implied fixed cost
mix of ~55-60% . . . was lower than our prior assumption of ~70% that was driving our
[earnings model].” In other words, the analyst had viewed Carvana’s SG&A expenses as
predominantly fixed and was surprised to learn that, in reality, a significantly higher
proportion of Carvana’s SG&A expenses were variable operations expenses, which
increased with each incremental vehicle sold and negatively impacted Carvana’s retail
profitability (i.e., unit economics).
(e)
The operations expenses that were excluded from Retail GPU (and not
separately disclosed or quantified to allow investors to decipher Carvana’s overall
profitability of retail sales on their own) were directly tied to problems stemming from
Carvana’s undisclosed growth-at-any-cost scheme, as described herein. For example, by
concealing a complete picture of “multi-car logistics, last-mile delivery . . . [n]on-payroll
logistics expenses . . . third-party transport services . . . [t]ransaction . . . expenses, [and]
title and registration . . . expenses” that “were associated with completing retail . . . vehicle
sales,” Defendants were able to conceal numerous problems from investors, including that:
(i) Carvana was boosting retail sales with unprofitable sales in distant markets; (ii) Carvana’s
rapid increase in buying cars from customers left it with a glut of low-quality cars that were
causing massive logistics constraints and costs; and (iii) Carvana was experiencing
significant title and registration issues.
(f)
Defendants described that “[g]rowth in retail GPU was driven by a
significant increase in the share of our vehicles sourced from customers.” Defendants,
however, omitted the fact that buying cars from customers was also adversely impacting
operations expense and unit economics – another key measure of actual retail profitability.
As alleged herein, the rapid increase in buying low-quality cars from customers caused a
significant spike in wholesale vehicle sales. In Q3 2020, Carvana increased its cars bought
from customers by 261%. Meanwhile, wholesale sales made up 20% of the Company’s total
sales in Q3 2020, as shown in the chart below. By the following quarter, Q4 2020,
Carvana’s wholesale sales spiked over 100% year-over-year.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 214 of 332
- 211 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Carvana later admitted that it experienced significant logistics constraints and costs
stemming from buying cars from customers and the resulting surplus of wholesale vehicle
sales. For example, in Carvana’s May 2022 Operating Plan, Defendants admitted the need to
“[r]apidly reduce SG&A expense per retail unit sold” (i.e., operations expenses). Defendants
admitted that “[b]uying cars from customers growth” led to “additional wholesale volume”
which, in turn, led to “constraints in our nationwide logistics network” and a spike in SG&A
logistics costs. Defendants further disclosed that “wholesale units acquired from customers
have . . . increased complexity in our multi-car logistics network.” Nevertheless, the
majority of the negative cost impacts of these logistics constraints were not reflected in
Carvana’s Q3 2020 Retail GPU because Defendants categorized many of these logistics
costs as operations expenses, which were buried in SG&A and not separately broken out.
Meanwhile, unbeknownst to investors, these costs did contribute to the $1,875 in undisclosed
“operations expense per retail unit sold” in Q3 2020 and did negatively impact Carvana’s
retail unit economics. Thus, it was materially misleading to describe buying cars from
customers as the main “driver” of the purported improvement in Retail GPU – the primary
metric Defendants provided for retail profitability – while simultaneously concealing that
buying cars from customers was also adversely impacting another key measure of
profitability: unit economics per retail vehicle sold.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 215 of 332
- 212 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
(g)
The fact that Defendants created a misleading impression as to
Carvana’s actual retail sales profitability, or lack thereof, during the Class Period is further
supported by a series of revelations, and corresponding analyst reactions that began during
2022. First, in April 2022, Carvana issued disappointing Q1 2022 financial results, which
partially revealed the truth about Carvana’s actual profitability, including the operations
expenses it had concealed from investors. For example, Defendants disclosed a quarterly
loss of $506 million and explained that they needed to purchase ADESA to “improve our
logistics network,” and bring down “shipping distances, times, and costs.” Analysts
commented on these revelations. For example, a J.P. Morgan analyst labeled the news
“‘confidence shattering’” and commented on Carvana’s opaque SG&A disclosures, which
encompassed operations expenses, stating “‘we do not believe Carvana (CVNA) is likely to
get a free pass on SG&A any longer.’” Then, in May 2022, Carvana announced that it was
laying off approximately 2,500 employees “primarily in operational groups” and issued an
updated operating plan that emphasized the need to “[r]apidly reduce SG&A expense per
retail unit sold” and make profitability a priority. Analysts commented on this surprising
news. For example, on May 21, 2022, Forbes cited to several industry analysts who
described Carvana’s “spendthrift business, whose growth-at-all-costs mentality undermined
business operations and sowed the seeds of its recent layoffs.” In November 2022, Carvana
issued disappointing Q3 2022 financial results and revealed that, unbeknownst to investors,
Carvana had been “frequently acquir[ing]” retail sales “that were less profitable in the
immediate period,” including “less profitable sales” in “markets with lower profitability due
to long distance from inventory.” Carvana further admitted it was now taking actions “to
improve profitability.” Analysts and the media reported on the news regarding Carvana’s
retail sales profitability. For example, a Needham & Company analyst noted, “‘[t]he path
forward for Carvana is to sell as many cars as possible, but to do so on a profitable basis,
versus prior it was more about selling as many cars as possible.’” Another media outlet
told investors to “[s]tep back and think about [Carvana’s disclosure] for a moment” because
“[i]t means that the company had historically grown in these markets even though they
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 216 of 332
- 213 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
weren’t all that attractive. It was growth for growth’s sake and not necessarily growth to
improve profitability.” Finally, on the last day of the Class Period, Defendants revealed that
Carvana was intentionally foregoing retail sales growth for the foreseeable future to instead
focus on driving “positive unit economics” (i.e., retail sales that were actually profitable).
Analysts commented on this disclosure. For example, a Morgan Stanley analyst, on
February 24, 2023, explained that “4Q was a sizeable miss,” and commented on Carvana’s
“[s]hift in strategy away from growth and towards EBITDA profitability.” The same day, a
RBC Capital Markets analyst also noted that Carvana’s GPU “miss[ed] Street expectations
by 32% . . . due in large part to . . . lower unit economics from cars.” Additionally, a
Stephens analyst noted “we do not foresee the company achieving its targeting unit
economics this year.”
253. Statement No. 36: On May 6, 2021, Jenkins stated on the Carvana Q1 2021
earnings call:
Retail GPU declined slightly to $1,211 from $1,265 in Q4, reflecting a
continuation of approximately $200 per unit of transitory costs . . . .
*
*
*
Q1 retail GPU was largely in line with our expectations when taking into
account the transitory costs that we talked about in Q4. . . . I think when we
look at where we’re headed from here, we did call out that we expect the
majority of those transitory costs that impacted Q4 and Q1 this year to be gone
by Q2. And so that’s a tailwind to retail GPU as we move towards 2Q, other
things being equal. Obviously, overall, we feel really good about our
progress there . . . .
254. As detailed below, Statement No. 36 made by Carvana and Jenkins omitted
material facts that affirmatively created an impression of a state of affairs that differed in a
material way from the one that actually existed.
(a)
During the Class Period, Carvana acknowledged that “[t]he Company’s
primary business objective is to sell used vehicles to retail customers and generate a profit
doing so.” Thus, Retail GPU, which measured the gross profit Carvana earned per retail
vehicle sold (before adding in other ancillary revenue and profit streams such as warranty,
insurance, and financing), was a critical metric to investors during the Class Period.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 217 of 332
- 214 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Accordingly, on the Q1 2021 earnings call, Defendants reported positive Retail GPU of
$1,211 and described “a tailwind to retail GPU,” which investors viewed as a proxy for
Carvana’s “unit economics” (i.e., per-vehicle profitability). However, as alleged herein,
Defendants concealed Carvana’s actual per-vehicle profitability – or lack thereof. Carvana
did so by: (i) excluding certain per-vehicle operations expenses from its calculation of Retail
GPU; and (ii) not disclosing or quantifying these excluded costs separately so investors
could decipher Carvana’s overall profitability of retail sales on their own. These per-vehicle
operations expenses, which were only separately broken out and quantified for the first time
after the Class Period, included material costs associated with completing a retail sale, such
as the full cost to ship a car to a customer and title and registration costs. By intentionally
hiding these significant costs, investors were left to rely on the Retail GPU reported by
Defendants (a positive Retail GPU of $1,211) as the sole measure of Carvana’s retail “unit
economics” (i.e., per-vehicle profitability). However, had Carvana included these operations
expenses in its Retail GPU calculation or separately disclosed and quantified the per-vehicle
costs associated with completing retail sales that had been excluded from reported Retail
GPU, Carvana would have revealed to investors that, on average, it generated negative
unit economics on retail vehicles sales in Q1 2021, as shown in the chart below. This
created an impression of a state affairs (positive per-vehicle profitability) materially different
from the one that actually existed (negative per-vehicle profitability).
Q1 2021
Disclosed Average selling
price per retail unit sold
$19,469
Disclosed Cost of sales per
retail unit sold
$18,258
Disclosed Retail GPU
$1,211
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 218 of 332
- 215 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
(b)
After the Class Period, Carvana admitted in its Q3 2023 Shareholder
Letter and November 2, 2023 Cost Structure Details presentation that the operations
expenses it had concealed from investors during the Class Period were a “key driver[]” of
“unit economics” (i.e., per-vehicle profitability).121 Specifically, Defendants admitted that,
internally, they viewed “two key drivers” of Carvana’s “unit economics” (i.e., profitability
per retail vehicle sold): (i) retail cost of sales (costs which Carvana included in Retail GPU
during the Class Period); and (ii) “operations expenses per retail unit sold” (costs which
Carvana excluded from Retail GPU during the Class Period and which had never previously
been separately disclosed or quantified for investors). Defendants further admitted that,
although these vehicle operations expenses had previously been buried and comingled in
“SG&A” expenses and excluded from Retail GPU, they were not fixed or corporate
119 Actual total expense per retail unit sold reflects the cost of sales per retail unit sold (as
disclosed) in addition to the undisclosed operations expense per retail unit sold.
120 Actual loss per retail unit sold reflects the average vehicle selling price per retail unit sold
(as disclosed) less the undisclosed actual total expense per retail unit sold.
121 After disclosing and quantifying operations expenses for the first time in November
2023, Jenkins acknowledged that the additional disclosure “allows a much deeper dive into
the economics of Carvana.” Jenkins added: “we’ve really been focused on . . . improving the
unit economics of the business. In other words, driving down variable costs per car,” which
he confirmed included “operations expenses, which are the more variable component of our
selling, general and administrative expenses.” Jenkins continued: “And the reason that we’re
focused on that is obviously, the more gains we make on reducing variable cost per car
today, the more profitable our growth can be in the future . . . .” Jenkins also described
“driving down operational expenses per unit” as one of Carvana’s “profitability initiatives,
and in particular, . . . our unit economics initiatives.” Thus, according to Jenkins, operations
expenses were a critical component of assessing Carvana’s profitability.
Undisclosed Operations
expense per retail unit sold
$2,055
Undisclosed Actual total
expense per retail unit
sold119
$20,313
Undisclosed Actual
negative “unit economics”
per retail unit sold120
~($844)
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 219 of 332
- 216 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
overhead expenses like typical SG&A costs. Rather, Defendants admitted that vehicle
operations costs were the direct per-vehicle costs of completing a vehicle sale (“[o]perations
expenses include the fulfillment, customer service, and transaction expenses associated with
completing retail and wholesale vehicle sales”). Defendants acknowledged that these
previously undisclosed and unquantified vehicle “operations expenses” included:
“customer care, multi-car logistics, last-mile delivery, and other
operations payroll” associated with completing retail vehicle sales;
“non-payroll logistics expenses, including fuel, repairs and
maintenance, and third-party transport services” associated with
completing retail vehicle sales; and
“[t]ransaction and other expenses, including limited warranty, title and
registration, and finance platform expenses” associated with
completing retail vehicle sales.
(c)
Defendants’ concealment of “operations expenses” when describing
Carvana’s retail profitability was especially misleading in light of the fact that Defendants
internally viewed “the underlying costs of completing a sale” as a critical component of
Carvana’s actual retail vehicle profitability. Indeed, as Garcia Junior later acknowledged:
When we think about trying to aim for more profitable sales, what does that
mean? There’s certainly variability in the kind of gross profit associated with
different types of sales. . . .
And then I think there’s a number of other dynamics kind of across car
type, et cetera. There’s also kind of variation in the underlying costs of
completing a sale.
Because Defendants did not specifically break out operations expenses (i.e., the underlying
costs of completing a sale) and instead buried and comingled them among SG&A expenses,
investors could not calculate the actual overall profitability of Carvana’s retail vehicle sales,
like Garcia Junior did internally. After the Class Period, in November 2023, Defendants
further admitted that, internally, they viewed operations expenses (i.e., costs associated with
completing a retail vehicle sale) as a “key driver[]” of “unit economics” (i.e., per-vehicle
profitability). Garcia Junior also described Carvana’s disclosure of operations expenses in
November 2023 as “clear evidence” of Carvana’s potential profitability.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 220 of 332
- 217 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
(d)
During Q1 2021, Defendants included per-vehicle operations expenses
in SG&A expenses (rather than Retail GPU), but they did not separately disclose or quantify
these expenses in a manner that would have allowed investors to decipher the overall
profitability (i.e., “unit economics”) of Carvana’s retail sales on their own. In fact, as
detailed below, the disclosures Defendants did make regarding SG&A expenses during Q1
2021 did more to mislead investors than to provide transparency around Carvana’s true
profitability per retail vehicle sold (i.e., “unit economics”).
(i)
First, Defendants’ Class Period description of SG&A expenses
was so vague that investors could not have discerned that material operations expenses
pertaining to Carvana’s actual per-vehicle profitability (i.e., “unit economics”) were included
in SG&A and excluded from Carvana’s Retail GPU calculation. For example, during the
Class Period, Defendants emphasized that SG&A costs consisted primarily of corporate and
overhead expenses, as is common across all companies.122 Defendants, however, omitted the
fact that 48% of its total reported SG&A costs were actually not corporate and overhead
expenses but variable “[o]perations expenses” which, as they later revealed, “include[d] the
fulfillment, customer service, and transaction expenses associated with completing retail
and wholesale vehicle sales.”123 Defendants’ intentionally opaque SG&A disclosures kept
analysts and investors in the dark. In fact, numerous analysts referred to Carvana’s SG&A
disclosures as a “black box.” For example, on April 27, 2020, a Morgan Stanley analyst
noted “[w]e (and the investors we spoke with) continue to look for more disclosure on
Other SG&A . . . [it] is a bit of a ‘black box.’” Likewise, on February 26, 2021, a J.P.
122 According to Carvana’s Q2 2020 10-Q: “Selling, general and administrative (‘SG&A’)
expenses include expenses associated with advertising and providing customer service to
customers, operating our vending machines and hubs, operating our logistics and
fulfillment network and other corporate overhead expenses, including expenses associated
with information technology, product development, engineering, legal, accounting,
finance, and business development.”
123 48% represents Q1 2021 total SG&A costs of $397 million compared to Q1 2021 total
operations expense of $190 million.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 221 of 332
- 218 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Morgan analyst referred to Carvana’s disclosure of SG&A expenses as “[t]he big ‘black
box.’”
(ii)
Second, beyond Defendants’ vague general description of SG&A
expenses, any references to certain specific expenses that were included in SG&A during the
Class Period also painted a materially incomplete picture of which operations expenses had
been excluded from Carvana’s Retail GPU calculation. For example, Defendants broke out
SG&A into five high-level categories during the Class Period, with general overhead
categories like compensation (“all payroll and related costs, including benefits, payroll taxes,
and equity-based compensation”), advertising, and “other” (including “general and
administrative expenses such as IT expenses, corporate occupancy, professional services and
insurance”), making up over 89.2% of the total reported SG&A costs in Q1 2021.124 The
only specific SG&A category disclosed by Defendants that was loosely tied to Carvana’s
post-Class Period disclosure of per-vehicle operations expenses (i.e., expenses that were
directly “associated with completing retail and wholesale vehicle sales”) was “logistics.”
But even the logistics disclosure was inconsistent with Carvana’s post-Class Period
disclosures revealing actual per-vehicle operations expenses.125
(iii)
Third, the quantification of any specifically identified operations
expenses within SG&A during the Class Period would have certainly left investors with a
false impression of which expenses had been excluded from Carvana’s Retail GPU
calculation and, thus, what Carvana’s actual per-vehicle profitability (i.e., unit economics)
124 Q1 2021 10-Q: Compensation and benefits, advertising, market occupancy, logistics, and
other.
125 Carvana disclosed that the logistics SG&A category, also disclosed as shipping and
handling, included “fuel, maintenance and depreciation related to operating our own
transportation fleet, and third party transportation fees.” This disclosure, however, was
inconsistent with Carvana’s post-Class Period disclosures revealing operations expenses
because: (i) Carvana included depreciation and amortization expenses in the SG&A logistics
category – these were overhead costs rather than per-vehicle shipping costs and, thus, were
specifically excluded from operations expenses; and (ii) Carvana excluded compensation
associated with outbound shipping from the SG&A logistics category but included these
expenses in the per-vehicle shipping costs it incurred and, thus, in operations expenses
disclosed after the Class Period.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 222 of 332
- 219 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
looked like during the Class Period. For example, the only specific retail vehicle operations
expense Defendants obliquely quantified during the Class Period was “logistics” costs,
which Defendants vaguely described as encompassing outbound shipping costs. Carvana
disclosed logistics costs of $30 million in Q1 2021, or $324 per retail vehicle sold.
Meanwhile, operations expense per retail vehicle sold totaled approximately $190 million in
Q1 2021, or $2,055 per vehicle sold. Defendants did not specifically quantify any other
specific vehicle operations expenses in Carvana’s Class Period SG&A disclosures.126 Thus,
at most, investors would have been specifically aware of just $324, approximately 16%, of
the $2,055 in actual per-vehicle operations expenses (i.e., costs “associated with completing
retail . . . vehicle sales”) that materially impacted Carvana’s retail vehicles sales profitability
(“unit economics”).
(iv)
Fourth, Defendants’ Class Period SG&A disclosures did not
break out fixed versus variable SG&A expenses. Like most companies, Carvana incurred
variable expenses (incremental expenses on each unit sold) and overhead expenses (fixed
expenses that did not significantly change based on the number of units sold). Operations
expenses, i.e., variable costs incurred in completing retail sales, increased with each
incremental vehicle sold and, thus, had a direct impact on per-vehicle profitability.
Meanwhile, fixed SG&A expenses, such as corporate overhead expenses, did not directly
impact the amount of profit Carvana earned on each vehicle it sold. During the Class Period,
analysts asked Defendants for more clarity on per-vehicle operations expenses (i.e., variable
versus fixed SG&A costs). For example, on the Q4 2020 earnings call, a Wells Fargo
analyst asked Jenkins: “[C]an you talk about incremental SG&A per unit, and walk me
through how to think about the SG&A . . .? And what is expected to be a fixed cost, . . . and
then how much would be a variable cost?” Tellingly, Jenkins declined to provide a clear
response to the question. On February 26, 2021, a J.P. Morgan analyst expressed concern
over the lack of disclosure, stating “we are concerned if this bucket of SG&A is starting to
126 Title and registration expenses were listed among expenses in the “Other” SG&A
category, but were not separately quantified.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 223 of 332
- 220 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
become more variable vs fixed in nature.” When Carvana finally disclosed operations
expense per retail unit in November 2023, a J.P. Morgan analyst commented that “the
company finally provided disclosure on fixed vs. variable SG&A expenses.” The same
analyst revealed that, due to the opaque nature of Carvana’s Class Period SG&A disclosures,
his estimate of fixed versus variable costs was significantly off: “[T]he implied fixed cost
mix of ~55-60% . . . was lower than our prior assumption of ~70% that was driving our
[earnings model].” In other words, the analyst had viewed Carvana’s SG&A expenses as
predominantly fixed and was surprised to learn that, in reality, a significantly higher
proportion of Carvana’s SG&A expenses were variable operations expenses, which
increased with each incremental vehicle sold and negatively impacted Carvana’s retail
profitability (i.e., unit economics).
(e)
The operations expenses that were excluded from Retail GPU (and not
separately disclosed or quantified to allow investors to decipher Carvana’s overall
profitability of retail sales on their own) were directly tied to problems stemming from
Carvana’s undisclosed growth-at-any-cost scheme, as described herein. For example, by
concealing a complete picture of “multi-car logistics, last-mile delivery . . . [n]on-payroll
logistics expenses . . . third-party transport services . . . [t]ransaction . . . expenses, [and]
title and registration . . . expenses” that “were associated with completing retail . . . vehicle
sales,” Defendants were able to conceal numerous problems from investors, including that:
(i) Carvana was boosting retail sales with unprofitable sales in distant markets; (ii) Carvana’s
rapid increase in buying cars from customers left it with a glut of low-quality cars that were
causing massive logistics constraints and costs; and (iii) Carvana was experiencing
significant title and registration issues.
(f)
The fact that Defendants created a misleading impression as to
Carvana’s actual retail sales profitability, or lack thereof, during the Class Period is further
supported by a series of revelations, and corresponding analyst reactions, that began during
2022. First, in April 2022, Carvana issued disappointing Q1 2022 financial results, which
partially revealed the truth about Carvana’s actual profitability, including the operations
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 224 of 332
- 221 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
expenses it had concealed from investors. For example, Defendants disclosed a quarterly
loss of $506 million and admitted they needed to purchase ADESA to “improve our logistics
network,” and bring down “shipping distances, times, and costs.” Analysts commented on
these revelations. For example, a J.P. Morgan analyst labeled the news “‘confidence
shattering’” and commented on Carvana’s opaque SG&A disclosures, which encompassed
operations expenses, stating “‘we do not believe Carvana (CVNA) is likely to get a free pass
on SG&A any longer.’” Then, in May 2022, Carvana announced that it was laying off
approximately 2,500 employees “primarily in operational groups” and issued an updated
operating plan that emphasized the need to “[r]apidly reduce SG&A expense per retail unit
sold” and make profitability a priority. Analysts commented on this surprising news. For
example, on May 21, 2022, Forbes cited to several industry analysts who described
Carvana’s “spendthrift business, whose growth-at-all-costs mentality undermined business
operations and sowed the seeds of its recent layoffs.” In November 2022, Carvana issued
disappointing Q3 2022 financial results and revealed that, unbeknownst to investors,
Carvana had been “frequently acquir[ing]” retail sales “that were less profitable in the
immediate period,” including “less profitable sales” in “markets with lower profitability due
to long distance from inventory.” Carvana further admitted it was now taking actions “to
improve profitability.” Analysts and the media reported on the news regarding Carvana’s
retail sales profitability. For example, a Needham & Company analyst noted, “‘[t]he path
forward for Carvana is to sell as many cars as possible, but to do so on a profitable basis,
versus prior it was more about selling as many cars as possible.’” Another media outlet
told investors to “[s]tep back and think about [Carvana’s disclosure] for a moment” because
“[i]t means that the company had historically grown in these markets even though they
weren’t all that attractive. It was growth for growth’s sake and not necessarily growth to
improve profitability.” Finally, on the last day of the Class Period, Defendants revealed that
Carvana was intentionally foregoing retail sales growth for the foreseeable future to instead
focus on driving “positive unit economics” (i.e., retail sales that were actually profitable).
Analysts commented on this disclosure. For example, a Morgan Stanley analyst on
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 225 of 332
- 222 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
February 24, 2023, explained that “4Q was a sizeable miss,” and commented on Carvana’s
“[s]hift in strategy away from growth and towards EBITDA profitability.” The same day, a
RBC Capital Markets analyst also noted that Carvana’s GPU “miss[ed] Street expectations
by 32% . . . due in large part to . . . lower unit economics from cars.” Additionally, a
Stephens analyst noted “we do not foresee the company achieving its targeting unit
economics this year.”
255. Statement No. 37: On August 5, 2021, Jenkins stated on the Carvana Q2 2021
earnings call:
[O]n retail GPU, we had our first quarter over $2,000 retail GPU. We’ve
come close a couple of times before. I think we’ve done $1,850 before and
$1,700, not including adjustments also in 2020. But the $2,000 is a record for
us. I think if we think through the drivers, I think maybe it’s helpful to think
through it sequentially from Q1. The biggest driver there was certainly the
performance [of] buying cars from customers.
*
*
*
[W]e feel really great about our retail GPU progress. Buying cars from
customers continues to be a strong driver of the improvements in the
business, it’s something we’re going to look to continue to do going forward.
256. As detailed below, Statement No. 37 made by Carvana and Jenkins omitted
material facts that affirmatively created an impression of a state of affairs that differed in a
material way from the one that actually existed.
(a)
During the Class Period, Carvana acknowledged that “[t]he Company’s
primary business objective is to sell used vehicles to retail customers and generate a profit
doing so.” Thus, Retail GPU, which measured the gross profit Carvana earned per retail
vehicle sold (before adding in other ancillary revenue and profit streams such as warranty,
insurance, and financing), was a critical metric to investors during the Class Period.
Accordingly, on the Q2 2021 earnings call, Jenkins touted a “record” Retail GPU of $2,022,
which investors viewed as a proxy for Carvana’s “unit economics (i.e., per-vehicle
profitability). However, as alleged herein, Defendants concealed Carvana’s actual per-
vehicle profitability – or lack thereof. Carvana did so by: (i) excluding certain per-vehicle
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 226 of 332
- 223 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
operations expenses from its calculation of Retail GPU; and (ii) not disclosing or quantifying
these excluded costs separately so investors could decipher Carvana’s overall profitability of
retail sales on their own. These per-vehicle operations expenses, which were only separately
broken out and quantified for the first time after the Class Period, included material costs
associated with completing a retail sale, such as the full cost to ship a car to a customer and
title and registration costs. By intentionally hiding these significant costs, investors were left
to rely on the Retail GPU reported by Defendants (a positive Retail GPU of $2,022) as the
sole measure of Carvana’s retail “unit economics” (i.e., per-vehicle profitability). However,
had Carvana included these operations expenses in its Retail GPU calculation or separately
disclosed and quantified the per-vehicle costs associated with completing retail sales that had
been excluded from reported Retail GPU, Carvana would have revealed to investors that,
on average, it generated negative unit economics of $130 per retail unit sold in Q2 2021, as
shown in the chart below. This created an impression of a state affairs (positive per-unit
profitability) materially different from the one that actually existed (negative per-unit
profitability).
Q2 2021
Disclosed Average selling
price per retail unit sold
$23,225
Disclosed Cost of sales per
retail unit sold
$21,203
Disclosed Retail GPU
$2,022
Undisclosed Operations
expense per retail unit sold
$2,152
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 227 of 332
- 224 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
(b)
After the Class Period, Carvana admitted in its Q3 2023 Shareholder
Letter and November 2, 2023 Cost Structure Details presentation that the operations
expenses it had concealed from investors during the Class Period were a “key driver[]” of
“unit economics” (i.e., per-vehicle profitability).129 Specifically, Defendants admitted that,
internally, they viewed “two key drivers” of Carvana’s “unit economics” (i.e., profitability
per retail vehicle sold): (i) retail cost of sales (costs which Carvana included in Retail GPU
during the Class Period); and (ii) “operations expenses per retail unit sold” (costs which
Carvana excluded from Retail GPU during the Class Period and which had never previously
been separately disclosed or quantified for investors). Defendants further admitted that,
although these vehicle operations expenses had previously been buried and comingled in
“SG&A” expenses and excluded from Retail GPU, they were not fixed or corporate
overhead expenses like typical SG&A costs. Rather, Defendants admitted that vehicle
operations costs were the direct per-vehicle costs of completing a vehicle sale (“[o]perations
127 Actual total expense per retail unit sold reflects the cost of sales per retail unit sold (as
disclosed) in addition to the undisclosed operations expense per retail unit sold.
128 Actual loss per retail unit sold reflects the average vehicle selling price per retail unit sold
(as disclosed) less the undisclosed actual total expense per retail unit sold.
129 After disclosing and quantifying operations expenses for the first time in November
2023, Jenkins acknowledged that the additional disclosure “allows a much deeper dive into
the economics of Carvana.” Jenkins added: “we’ve really been focused on . . . improving the
unit economics of the business. In other words, driving down variable costs per car,” which
he confirmed included “operations expenses, which are the more variable component of our
selling, general and administrative expenses.” Jenkins continued: “And the reason that we’re
focused on that is obviously, the more gains we make on reducing variable cost per car
today, the more profitable our growth can be in the future . . . .” Jenkins also described
“driving down operational expenses per unit” as one of Carvana’s “profitability initiatives,
and in particular, . . . our unit economics initiatives.” Thus, according to Jenkins, operations
expenses were a critical component of assessing Carvana’s profitability.
Undisclosed Actual total
expense per retail unit
sold127
$23,355
Undisclosed Actual
negative “unit economics”
per retail unit sold128
($130)
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 228 of 332
- 225 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
expenses include the fulfillment, customer service, and transaction expenses associated with
completing retail and wholesale vehicle sales”). Defendants acknowledged that these
previously undisclosed and unquantified vehicle “operations expenses” included:
“customer care, multi-car logistics, last-mile delivery, and other
operations payroll” associated with completing retail vehicle sales;
“non-payroll logistics expenses, including fuel, repairs and
maintenance, and third-party transport services” associated with
completing retail vehicle sales; and
“[t]ransaction and other expenses, including limited warranty, title and
registration, and finance platform expenses” associated with
completing retail vehicle sales.
(c)
Defendants’ concealment of “operations expenses” when describing
Carvana’s retail profitability was especially misleading in light of the fact that Defendants
internally viewed “the underlying costs of completing a sale” as a critical component of
Carvana’s actual retail vehicle profitability. Indeed, as Garcia Junior later acknowledged:
When we think about trying to aim for more profitable sales, what does that
mean? There’s certainly variability in the kind of gross profit associated with
different types of sales. . . .
And then I think there’s a number of other dynamics kind of across car
type, et cetera. There’s also kind of variation in the underlying costs of
completing a sale.
Because Defendants did not specifically break out operations expenses (i.e., the underlying
costs of completing a sale) and instead buried and comingled them among SG&A expenses,
investors could not calculate the actual overall profitability of Carvana’s retail vehicle sales,
like Garcia Junior did internally. After the Class Period, in November 2023, Defendants
further admitted that, internally, they viewed operations expenses (i.e., costs associated with
completing a retail vehicle sale) as a “key driver[]” of “unit economics” (i.e., per-vehicle
profitability). Garcia Junior also described Carvana’s disclosure of operations expenses in
November 2023 as “clear evidence” of Carvana’s potential profitability.
(d)
During Q2 2021, Defendants included per-vehicle operations expenses
in SG&A expenses (rather than Retail GPU), but they did not separately disclose or quantify
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 229 of 332
- 226 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
these expenses in a manner that would have allowed investors to decipher the overall
profitability (i.e., “unit economics”) of Carvana’s retail sales on their own. In fact, as
detailed below, the disclosures Defendants did make regarding SG&A expenses during Q2
2021 did more to mislead investors than to provide transparency around Carvana’s true
profitability per retail vehicle sold (i.e., “unit economics”).
(i)
First, Defendants’ Class Period description of SG&A expenses
was so vague that investors could not have discerned that material operations expenses
pertaining to Carvana’s actual per-vehicle profitability (i.e., “unit economics”) were included
in SG&A and excluded from Carvana’s Retail GPU calculation. For example, during the
Class Period, Defendants emphasized that SG&A costs consisted primarily of corporate and
overhead expenses, as is common across all companies.130 Defendants, however, omitted the
fact that 49% of its total reported SG&A costs were actually not corporate and overhead
expenses but variable “[o]perations expenses” which, as they later revealed, “include[d] the
fulfillment, customer service, and transaction expenses associated with completing retail
and wholesale vehicle sales.”131 Defendants’ intentionally opaque SG&A disclosures kept
analysts and investors in the dark. In fact, numerous analysts referred to Carvana’s SG&A
disclosures as a “black box.” For example, on April 27, 2020, a Morgan Stanley analyst
noted “[w]e (and the investors we spoke with) continue to look for more disclosure on
Other SG&A . . . [it] is a bit of a ‘black box.’” Likewise, on February 26, 2021, a J.P.
Morgan analyst referred to Carvana’s disclosure of SG&A expenses as “[t]he big ‘black
box.’”
130 According to Carvana’s Q2 2020 10-Q: “Selling, general and administrative (‘SG&A’)
expenses include expenses associated with advertising and providing customer service to
customers, operating our vending machines and hubs, operating our logistics and
fulfillment network and other corporate overhead expenses, including expenses associated
with information technology, product development, engineering, legal, accounting,
finance, and business development.”
131 49% represents Q2 2021 total SG&A costs of $470 million compared to Q2 2021 total
operations expense of $232 million.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 230 of 332
- 227 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
(ii)
Second, beyond Defendants’ vague general description of SG&A
expenses, any references to certain specific expenses that were included in SG&A during the
Class Period also painted a materially incomplete picture of which operations expenses had
been excluded from Carvana’s Retail GPU calculation. For example, Defendants broke out
SG&A into five high-level categories during the Class Period, with general overhead
categories like compensation (“all payroll and related costs, including benefits, payroll taxes,
and equity-based compensation”), advertising, and “other” (including “general and
administrative expenses such as IT expenses, corporate occupancy, professional services and
insurance”), making up over 85% of the total reported SG&A costs in Q2 2021.132 The only
specific SG&A category disclosed by Defendants that was loosely tied to Carvana’s post-
Class Period disclosure of per-vehicle operations expenses (i.e., expenses that were directly
“associated with completing retail and wholesale vehicle sales”) was “logistics.” But even
the logistics disclosure was inconsistent with Carvana’s post-Class Period disclosures
revealing actual per-vehicle operations expenses.133
(iii)
Third, the quantification of any specifically identified operations
expenses within SG&A during Q2 2021 would have certainly left investors with a false
impression of which expenses had been excluded from Carvana’s Retail GPU calculation
and, thus, what Carvana’s actual per-vehicle profitability (i.e., unit economics) looked like
during the Class Period. For example, the only specific retail vehicle operations expense
Defendants obliquely quantified during the Class Period was “logistics” costs, which
132 Q2 2021 10-Q: Compensation and benefits, advertising, market occupancy, logistics, and
other.
133 Carvana disclosed that the logistics SG&A category, also disclosed as shipping and
handling, included “fuel, maintenance and depreciation related to operating our own
transportation fleet, and third party transportation fees.” This disclosure, however, was
inconsistent with Carvana’s post-Class Period disclosures revealing operations expenses
because: (i) Carvana included depreciation and amortization expenses in the SG&A logistics
category – these were overhead costs rather than per-vehicle shipping costs and, thus, were
specifically excluded from operations expenses; and (ii) Carvana excluded compensation
associated with outbound shipping from the SG&A logistics category but included these
expenses in the per-vehicle shipping costs it incurred and, thus, in operations expenses
disclosed after the Class Period.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 231 of 332
- 228 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Defendants vaguely described as encompassing outbound shipping costs. Carvana disclosed
logistics costs of $34 million in Q2 2021, or $315 per retail vehicle sold. Meanwhile,
operations expense per retail vehicle sold totaled approximately $232 million in Q2 2021, or
$2,152 per vehicle sold. Defendants did not specifically quantify any other specific vehicle
operations expenses in Carvana’s Class Period SG&A disclosures.134 Thus, at most,
investors would have been specifically aware of just $315, less than 15%, of the $2,152 in
actual per-vehicle operations expenses (i.e., costs “associated with completing retail . . .
vehicle sales”) that materially impacted Carvana’s retail vehicles sales profitability (“unit
economics”).
(iv)
Fourth, Defendants’ Class Period SG&A disclosures did not
break out fixed versus variable SG&A expenses. Like most companies, Carvana incurred
variable expenses (incremental expenses on each unit sold) and overhead expenses (fixed
expenses that did not significantly change based on the number of units sold). Operations
expenses, i.e., variable costs incurred in completing retail sales, increased with each
incremental vehicle sold and, thus, had a direct impact on per-vehicle profitability.
Meanwhile, fixed SG&A expenses, such as corporate overhead expenses, did not directly
impact the amount of profit Carvana earned on each vehicle it sold. During the Class Period,
analysts asked Defendants for more clarity on per-vehicle operations expenses (i.e., variable
versus fixed SG&A costs). For example, on the Q4 2020 earnings call, a Wells Fargo
analyst asked Jenkins: “[C]an you talk about incremental SG&A per unit, and walk me
through how to think about the SG&A . . .? And what is expected to be a fixed cost, . . . and
then how much would be a variable cost?” Tellingly, Jenkins declined to provide a clear
response to the question. On February 26, 2021, a J.P. Morgan analyst expressed concern
over the lack of disclosure, stating “we are concerned if this bucket of SG&A is starting to
become more variable vs fixed in nature.” When Carvana finally disclosed operations
expense per retail unit in November 2023, a J.P. Morgan analyst commented that “the
134 Title and registration expenses were listed among expenses in the “Other” SG&A
category, but were not separately quantified.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 232 of 332
- 229 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
company finally provided disclosure on fixed vs. variable SG&A expenses.” The same
analyst revealed that, due to the opaque nature of Carvana’s Class Period SG&A disclosures,
his estimate of fixed versus variable costs was significantly off: “[T]he implied fixed cost
mix of ~55-60% . . . was lower than our prior assumption of ~70% that was driving our
[earnings model].” In other words, the analyst had viewed Carvana’s SG&A expenses as
predominantly fixed and was surprised to learn that, in reality, a significantly higher
proportion of Carvana’s SG&A expenses were variable operations expenses, which
increased with each incremental vehicle sold and negatively impacted Carvana’s retail
profitability (i.e., unit economics).
(e)
The operations expenses that were excluded from Retail GPU (and not
separately disclosed or quantified to allow investors to decipher Carvana’s overall
profitability of retail sales on their own) were directly tied to problems stemming from
Carvana’s undisclosed growth-at-any-cost scheme, as described herein. For example, by
concealing a complete picture of “multi-car logistics, last-mile delivery . . . [n]on-payroll
logistics expenses . . . third-party transport services . . . [t]ransaction . . . expenses, [and]
title and registration . . . expenses” that “were associated with completing retail . . . vehicle
sales,” Defendants were able to conceal numerous problems from investors, including that:
(i) Carvana was boosting retail sales with unprofitable sales in distant markets; (ii) Carvana’s
rapid increase in buying cars from customers left it with a glut of low-quality cars that were
causing massive logistics constraints and costs; and (iii) Carvana was experiencing
significant title and registration issues.
(f)
Defendants touted that “[t]he biggest driver [of the purported
improvement in retail sales profitability] was certainly. . . buying cars from customers” as
“[b]uying cars from customers continues to be a strong driver of the improvements in the
business.” Defendants, however, omitted the fact that buying cars from customers was also
adversely impacting operations expense and unit economics – another key measure of actual
retail profitability. As alleged herein, the rapid increase in buying low-quality cars from
customers caused a significant spike in wholesale vehicle sales. In Q2 2021, wholesale sales
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 233 of 332
- 230 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
growth had spiked to over 500% year-over-year, far outpacing retail sales growth, as shown
in the first chart below. Meanwhile, wholesale sales made up made up 30% of the
Company’s total sales in Q2 2021, as shown in the second chart below.
Carvana later admitted that it experienced significant logistics constraints and costs as a
result of buying cars from customers and its corresponding surplus of wholesale vehicle
sales. For example, in Carvana’s May 2022 Operating Plan, Defendants admitted the need to
“[r]apidly reduce SG&A expense per retail unit sold” (i.e., operations expenses). Defendants
admitted that “[b]uying cars from customers growth” led to “additional wholesale volume”
which, in turn, led to “constraints in our nationwide logistics network” and a spike in SG&A
logistics costs. Defendants further disclosed that “wholesale units acquired from customers
have . . . increased complexity in our multi-car logistics network.” Nevertheless, the
majority of the negative cost impacts of these logistics constraints were not reflected in
Carvana’s Q2 2021 Retail GPU because Defendants categorized many of the logistics costs
as operations expenses, which were buried in SG&A and not separately broken out.
Meanwhile, unbeknownst to investors, these costs did contribute to the $2,152 in undisclosed
“operations expense per retail unit sold” in Q2 2021 and did negatively impact Carvana’s
retail unit economics. Thus, it was materially misleading to describe buying cars from
customers as the main “driver” of the purported improvement in Retail GPU, the primary
metric Defendants provided for retail profitability, while simultaneously concealing that
0%
100%
200%
300%
400%
500%
Q1 20Q2 20Q3 20Q4 20Q1 21Q2 21Q3 21Q4 21Q1 22
Wholesale Growth Exceeded Retail Growth
y/y growth (retail cars sold)
y/y growth (wholesale cars sold)
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 234 of 332
- 231 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
buying cars from customers was also adversely impacting another key measure of
profitability: unit economics per retail vehicle sold.
(g)
The fact that Defendants created a misleading impression as to
Carvana’s actual retail sales profitability, or lack thereof, during the Class Period is further
supported by a series of revelations, and corresponding analyst reactions, that began during
2022. First, in April 2022, Carvana issued disappointing Q1 2022 financial results, which
partially revealed the truth about Carvana’s actual profitability, including the operations
expenses it had concealed from investors. For example, Defendants disclosed a quarterly
loss of $506 million and explained that they needed to purchase ADESA to “improve our
logistics network,” and bring down “shipping distances, times, and costs.” Analysts
commented on these revelations. For example, a J.P. Morgan analyst labeled the news
“‘confidence shattering’” and commented on Carvana’s opaque SG&A disclosures, which
encompassed operations expenses, stating: “‘we do not believe Carvana (CVNA) is likely to
get a free pass on SG&A any longer.’” Then, in May 2022, Carvana announced that it was
laying off approximately 2,500 employees “primarily in operational groups” and issued an
updated operating plan that emphasized its need to “[r]apidly reduce SG&A expense per
retail unit sold” (i.e., operations expenses) and make profitability a priority. Analysts
commented on this surprising news. For example, on May 21, 2022, Forbes cited to several
industry analysts who described Carvana’s “spendthrift business, whose growth-at-all-costs
mentality undermined business operations and sowed the seeds of its recent layoffs.” In
November 2022, Carvana issued disappointing Q3 2022 financial results and revealed that,
unbeknownst to investors, Carvana had been “frequently acquir[ing]” retail sales “that were
less profitable in the immediate period,” including “less profitable sales” in “markets with
lower profitability due to long distance from inventory.” Carvana further admitted it was
now taking actions “to improve profitability.” Analysts and the media reported on the news
regarding Carvana’s retail sales profitability. For example, a Needham & Company analyst
noted, “‘[t]he path forward for Carvana is to sell as many cars as possible, but to do so on a
profitable basis, versus prior it was more about selling as many cars as possible.’” Another
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 235 of 332
- 232 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
media outlet told investors to “[s]tep back and think about [Carvana’s disclosure] for a
moment” because “[i]t means that the company had historically grown in these markets
even though they weren’t all that attractive. It was growth for growth’s sake and not
necessarily growth to improve profitability.” Finally, on the last day of the Class Period,
Defendants revealed that Carvana was intentionally foregoing retail sales growth for the
foreseeable future to instead focus on driving “positive unit economics” (i.e., retail sales that
were actually profitable). Analysts commented on this disclosure. For example, a Morgan
Stanley analyst, on February 24, 2023, explained that “4Q was a sizeable miss,” and
commented on Carvana’s “[s]hift in strategy away from growth and towards EBITDA
profitability.” The same day, a RBC Capital Markets analyst also noted that Carvana’s GPU
“miss[ed] Street expectations by 32% . . . due in large part to . . . lower unit economics from
cars.” Additionally, a Stephens analyst noted “we do not foresee the company achieving its
targeting unit economics this year.”
257. Statement No. 38: On November 4, 2021, Jenkins stated on the Carvana Q3
2021 earnings call:
Retail GPU was $1,769, a decrease of $88. The change in retail GPU was
primarily driven by higher reconditioning costs, in part resulting from the
impact of the Delta variant on production throughput, and higher wholesale
acquisition prices, partially offset by higher customer-sourced ratio.
*
*
*
So we had a strong quarter on retail GPU in Q3, came in at $1,769.
258. As detailed below, Statement No. 38 made by Carvana and Jenkins omitted
material facts that affirmatively created an impression of a state of affairs that differed in a
material way from the one that actually existed.
(a)
During the Class Period, Carvana acknowledged that “[t]he Company’s
primary business objective is to sell used vehicles to retail customers and generate a profit
doing so.” Thus, Retail GPU, which measured the gross profit Carvana earned per retail
vehicle sold (before adding in other ancillary revenue and profit streams such as warranty,
insurance, and financing), was a critical metric to investors during the Class Period.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 236 of 332
- 233 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Accordingly, on the Q3 2021 earnings call, Defendants disclosed Retail GPU of $1,769 and
described that it was a “strong quarter on retail GPU,” which investors viewed as a proxy
for Carvana’s retail “unit economics” (i.e., per-vehicle profitability). However, as alleged
herein, Defendants concealed Carvana’s actual per-vehicle profitability – or lack thereof.
Carvana did so by: (i) excluding certain per-vehicle operations expenses from its calculation
of Retail GPU; and (ii) not disclosing or quantifying these excluded costs separately so
investors could decipher Carvana’s overall profitability of retail sales on their own. These
per-vehicle operations expenses, which were only separately broken out and quantified for
the first time after the Class Period, included material costs associated with completing a
retail sale, such as the full cost to ship a car to a customer and title and registration costs. By
intentionally hiding these significant costs, investors were left to rely on the Retail GPU
reported by Defendants (a positive Retail GPU of $1,769) as the sole measure of Carvana’s
retail “unit economics” (i.e., per-vehicle profitability). However, had Carvana included these
operations expenses in its Retail GPU calculation or separately disclosed and quantified
these per-vehicle costs associated with completing retail sales that had been excluded from
reported Retail GPU, Carvana would have revealed to investors that, on average, it
generated negative unit economics of $696 per retail unit sold in Q3 2021, as shown in the
chart below. This created an impression of a state affairs (positive per-unit profitability)
materially different from the one that actually existed (negative per-unit profitability).
Q3 2021
Disclosed Average selling
price per retail unit sold
$23,671
Disclosed Cost of sales per
retail unit sold
$21,902
Disclosed Retail GPU
$1,769
Undisclosed Operations
expense per retail unit sold
$2,465
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 237 of 332
- 234 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
(b)
After the Class Period, Carvana admitted in its Q3 2023 Shareholder
Letter and November 2, 2023 Cost Structure Details presentation that the operations
expenses it had concealed from investors during the Class Period were a “key driver[]” of
“unit economics” (i.e., per-vehicle profitability).137 Specifically, Defendants admitted that,
internally, they viewed “two key drivers” of Carvana’s “unit economics” (i.e., profitability
per retail vehicle sold): (i) retail cost of sales (costs which Carvana included in Retail GPU
during the Class Period); and (ii) “operations expenses per retail unit sold” (costs which
Carvana excluded from Retail GPU during the Class Period and which had never previously
been separately disclosed or quantified for investors). Defendants further admitted that,
although these vehicle operations expenses had previously been buried and comingled in
“SG&A” expenses and excluded from Retail GPU, they were not fixed or corporate
overhead expenses like typical SG&A costs. Rather, Defendants admitted that vehicle
operations costs were the direct per-vehicle costs of completing a vehicle sale (“[o]perations
expenses include the fulfillment, customer service, and transaction expenses associated with
135 Actual total expense per retail unit sold reflects the cost of sales per retail unit sold (as
disclosed) in addition to the undisclosed operations expense per retail unit sold.
136 Actual loss per retail unit sold reflects the average vehicle selling price per retail unit sold
(as disclosed) less the undisclosed actual total expense per retail unit sold.
137 After disclosing and quantifying operations expenses for the first time in November
2023, Jenkins acknowledged that the additional disclosure “allows a much deeper dive into
the economics of Carvana.” Jenkins added: “we’ve really been focused on . . . improving the
unit economics of the business. In other words, driving down variable costs per car,” which
he confirmed included “operations expenses, which are the more variable component of our
selling, general and administrative expenses.” Jenkins continued: “And the reason that we’re
focused on that is obviously, the more gains we make on reducing variable cost per car
today, the more profitable our growth can be in the future . . . .” Jenkins also described
“driving down operational expenses per unit” as one of Carvana’s “profitability initiatives,
and in particular, . . . our unit economics initiatives.” Thus, according to Jenkins, operations
expenses were a critical component of assessing Carvana’s profitability.
Undisclosed Actual total
expense per retail unit
sold135
$24,367
Undisclosed Actual
negative “unit economics”
per retail unit sold136
($696)
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 238 of 332
- 235 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
completing retail and wholesale vehicle sales”). Defendants acknowledged that these
previously undisclosed and unquantified vehicle “operations expenses” included:
“customer care, multi-car logistics, last-mile delivery, and other operations
payroll” associated with completing retail vehicle sales;
“non-payroll logistics expenses, including fuel, repairs and maintenance, and
third-party transport services” associated with completing retail vehicle sales;
and
“[t]ransaction and other expenses, including limited warranty, title and
registration, and finance platform expenses” associated with completing retail
vehicle sales.
(c)
Carvana’s concealment of “operations expenses” when describing
Carvana’s retail profitability was especially misleading in light of the fact that Defendants
internally viewed “the underlying costs of completing a sale” as a critical component of
Carvana’s actual retail vehicle profitability. Indeed, as Garcia Junior later acknowledged:
When we think about trying to aim for more profitable sales, what does that
mean? There’s certainly variability in the kind of gross profit associated with
different types of sales. . . .
And then I think there’s a number of other dynamics kind of across car
type, et cetera. There’s also kind of variation in the underlying costs of
completing a sale.
Because Defendants did not specifically break out operations expenses (i.e., the underlying
costs of completing a retail sale) and instead buried and comingled them among SG&A
expenses, investors could not calculate the actual overall profitability of Carvana’s retail
vehicle sales, like Garcia Junior did internally. After the Class Period, in November 2023,
Defendants further admitted that, internally, they viewed operations expenses (i.e., costs
“associated with completing a retail vehicle sale) as a “key driver[]” of “unit economics”
(i.e., per-vehicle profitability). Garcia Junior also described Carvana’s disclosure of
operations expenses in November 2023 as “clear evidence” of Carvana’s potential
profitability.
(d)
Defendants’ concealment of operations expenses when describing
Carvana’s retail profitability in Q3 2021 was also misleading because Defendants omitted
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 239 of 332
- 236 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
that operations expenses were spiking significantly during the second half of FY 2021. In
fact, when Defendants disclosed and quantified operations expenses for the first time in
November 2023, they revealed that these previously concealed expenses had spiked 75%
during the Class Period, as shown in the first chart below.138 By Q3 2021, the costs had
spiked 32% year-over-year as depicted in the second chart below.
(e)
During Q3 2021, Defendants included per-vehicle operations expenses
in SG&A expenses (rather than Retail GPU), but they did not separately disclose or quantify
these expenses in a manner that would have allowed investors to decipher the overall
138 Source: November 2, 2023 Cost Structure Details presentation. (Emphasis in red added.)
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 240 of 332
- 237 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
profitability (i.e., “unit economics”) of Carvana’s retail sales on their own. In fact, as
detailed below, the disclosures Defendants did make regarding SG&A expenses during Q3
2021 did more to mislead investors than to provide transparency around Carvana’s true
profitability per retail vehicle sold (i.e., “unit economics”).
(i)
First, Defendants’ Class Period description of SG&A expenses
was so vague that investors could not have discerned that material operations expenses
pertaining to Carvana’s actual per-vehicle profitability (i.e., “unit economics”) were included
in SG&A and excluded from Carvana’s Retail GPU calculation. For example, during the
Class Period, Defendants emphasized that SG&A costs consisted primarily of corporate and
overhead expenses, as is common across all companies.139 Defendants, however, omitted the
fact that 51% of its total reported SG&A costs were actually not corporate and overhead
expenses but variable “[o]perations expenses” which, as they later revealed, “include[d] the
fulfillment, customer service, and transaction expenses associated with completing retail
and wholesale vehicle sales.”140 Defendants’ intentionally opaque SG&A disclosures kept
analysts and investors in the dark. In fact, numerous analysts referred to Carvana’s SG&A
disclosures as a “black box.” For example, on April 27, 2020, a Morgan Stanley analyst
noted “[w]e (and the investors we spoke with) continue to look for more disclosure on
Other SG&A . . . [it] is a bit of a ‘black box.’” Likewise, on February 26, 2021, a J.P.
Morgan analyst referred to Carvana’s disclosure of SG&A expenses as “[t]he big ‘black
box.’”
(ii)
Second, beyond Defendants’ vague general description of SG&A
expenses, any references to certain specific expenses that were included in SG&A during the
139 According to Carvana’s Q2 2020 10-Q: “Selling, general and administrative (‘SG&A’)
expenses include expenses associated with advertising and providing customer service to
customers, operating our vending machines and hubs, operating our logistics and
fulfillment network and other corporate overhead expenses, including expenses associated
with information technology, product development, engineering, legal, accounting,
finance, and business development.”
140 51% represents Q3 2021 total SG&A costs of $546 million compared to Q3 2021 total
operations expense of $276 million.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 241 of 332
- 238 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Class Period also painted a materially incomplete picture of which operations expenses had
been excluded from Carvana’s Retail GPU calculation. For example, Defendants broke out
SG&A into five high-level categories during the Class Period, with general overhead
categories like compensation (“all payroll and related costs, including benefits, payroll taxes,
and equity-based compensation”), advertising, and “other” (including “general and
administrative expenses such as IT expenses, corporate occupancy, professional services and
insurance”), making up over 89.3% of the total reported SG&A costs in Q3 2021.141 The
only specific SG&A category disclosed by Defendants that was loosely tied to Carvana’s
post-Class Period disclosure of per-vehicle operations expenses (i.e., expenses that were
directly “associated with completing retail and wholesale vehicle sales”) was “logistics.”
But even the logistics disclosure was inconsistent with Carvana’s post-Class Period
disclosures revealing actual per-vehicle operations expenses.142
(iii)
Third, the quantification of any specifically identified operations
expenses within SG&A during the Class Period would have certainly left investors with a
false impression of which expenses had been excluded from Carvana’s Retail GPU
calculation and, thus, what Carvana’s actual per-vehicle profitability (i.e., unit economics)
looked like during the Class Period. For example, the only specific retail vehicle operations
expense Defendants obliquely quantified during the Class Period was “logistics” costs,
which Defendants vaguely described as encompassing outbound shipping costs. Carvana
disclosed logistics costs of $40 million in Q3 2021, or $357 per retail vehicle sold.
141 Q3 2021 10-Q: Compensation and benefits, advertising, market occupancy, logistics, and
other.
142 Carvana disclosed that the logistics SG&A category, also disclosed as shipping and
handling, included “fuel, maintenance and depreciation related to operating our own
transportation fleet, and third party transportation fees.” This disclosure, however, was
inconsistent with Carvana’s post-Class Period disclosures revealing operations expenses
because: (i) Carvana included depreciation and amortization expenses in the SG&A logistics
category – these were overhead costs rather than per-vehicle shipping costs and, thus, were
specifically excluded from operations expenses; and (ii) Carvana excluded compensation
associated with outbound shipping from the SG&A logistics category but included these
expenses in the per-vehicle shipping costs it incurred and, thus, in operations expenses
disclosed after the Class Period.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 242 of 332
- 239 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Meanwhile, operations expense per retail vehicle sold totaled approximately $276 million in
Q3 2021, or $2,465 per vehicle sold. Defendants did not specifically quantify any other
specific vehicle operations expenses in Carvana’s Class Period SG&A disclosures.143 Thus,
at most, investors would have been specifically aware of just $357, less than 15%, of the
$2,465 in actual per-vehicle operations expenses (i.e., costs “associated with completing
retail . . . vehicle sales”) that materially impacted Carvana’s retail vehicles sales profitability
(“unit economics”) in Q3 2021.
(iv)
Fourth, Defendants’ Class Period SG&A disclosures did not
break out fixed versus variable SG&A expenses. Like most companies, Carvana incurred
variable expenses (incremental expenses on each unit sold) and overhead expenses (fixed
expenses that did not significantly change based on the number of units sold). Operations
expenses, i.e., variable costs incurred in completing retail sales, increased with each
incremental vehicle sold and, thus, had a direct impact on per-vehicle profitability.
Meanwhile, fixed SG&A expenses, such as corporate overhead expenses, did not directly
impact the amount of profit Carvana earned on each vehicle it sold. During the Class Period,
analysts asked Defendants for more clarity on per-vehicle operations expenses (i.e., variable
versus fixed SG&A costs). For example, on the Q4 2020 earnings call, a Wells Fargo
analyst asked Jenkins: “[C]an you talk about incremental SG&A per unit, and walk me
through how to think about the SG&A . . .? And what is expected to be a fixed cost, . . . and
then how much would be a variable cost?” Tellingly, Jenkins declined to provide a clear
response to the question. On February 26, 2021, a J.P. Morgan analyst expressed concern
over the lack of disclosure, stating “we are concerned if this bucket of SG&A is starting to
become more variable vs fixed in nature.” When Carvana finally disclosed operations
expense per retail unit in November 2023, a J.P. Morgan analyst commented that “the
company finally provided disclosure on fixed vs. variable SG&A expenses.” The same
analyst revealed that, due to the opaque nature of Carvana’s Class Period SG&A disclosures,
143 Title and registration expenses were listed among expenses in the “Other” SG&A
category, but were not separately quantified.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 243 of 332
- 240 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
his estimate of fixed versus variable costs was significantly off: “[T]he implied fixed cost
mix of ~55-60% . . . was lower than our prior assumption of ~70% that was driving our
[earnings model].” In other words, the analyst had viewed Carvana’s SG&A expenses as
predominantly fixed and was surprised to learn that, in reality, a significantly higher
proportion of Carvana’s SG&A expenses were variable operations expenses, which
increased with each incremental vehicle sold and negatively impacted Carvana’s retail
profitability (i.e., unit economics).
(f)
The operations expenses that were excluded from Retail GPU (and not
separately disclosed or quantified to allow investors to decipher Carvana’s overall
profitability of retail sales on their own) were directly tied to problems stemming from
Carvana’s undisclosed growth-at-any-cost scheme, as described herein. For example, by
concealing a complete picture of “multi-car logistics, last-mile delivery . . . [n]on-payroll
logistics expenses . . . third-party transport services . . . [t]transaction . . . expenses, [and]
title and registration . . . expenses” that “were associated with completing retail . . . vehicle
sales,” Defendants were able to conceal numerous problems from investors, including that:
(i) Carvana was boosting retail sales with unprofitable sales in distant markets; (ii) Carvana’s
rapid increase in buying cars from customers left it with a glut of low-quality cars that were
causing massive logistics constraints and costs; and (iii) Carvana was experiencing
significant title and registration issues.
(g)
In describing the drivers of retail sales profitability, Defendants pointed
to a positive impact from buying cars from customers, claiming the decline in Retail GPU
“partially offset by higher customer-sourced ratio.” Meanwhile, Defendants omitted the
fact that buying cars from customers was also adversely impacting operations expense and
unit economics – another key measure of actual retail profitability. As alleged herein, the
rapid increase in buying low-quality cars from customers caused a significant spike in
wholesale vehicle sales. In Q3 2021, wholesale sales growth spiked 261% year-over-year,
far outpacing retail sales growth, as shown in the first chart below. Meanwhile, wholesale
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 244 of 332
- 241 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
sales made up over 30% of the Company’s total sales in Q3 2021, as shown in the second
chart below.
Carvana later admitted that it experienced significant logistics constraints and costs as a
result of buying cars from customers and its corresponding surplus of wholesale vehicle
sales. For example, in Carvana’s May 2022 Operating Plan, Defendants admitted the need to
“[r]apidly reduce SG&A expense per retail unit sold” (i.e., operations expenses). Defendants
admitted that “[b]uying cars from customers growth” led to “additional wholesale volume”
which, in turn, led to “constraints in our nationwide logistics network” and a spike in SG&A
logistics costs. Defendants further disclosed that “wholesale units acquired from customers
have . . . increased complexity in our multi-car logistics network.” Nevertheless, the
majority of the negative cost impacts of these logistics constraints were not reflected in
Carvana’s Q3 2021 Retail GPU because Defendants categorized many of these logistics
costs as operations expenses, which were buried in SG&A and not separately broken out.
Meanwhile, unbeknownst to investors, these costs did contribute to the $2,465 in undisclosed
“operations expense per retail unit sold” in Q3 2021 and did negatively impact Carvana’s
retail unit economics. Thus, it was materially misleading to describe buying cars from
customers as the main “driver” of the purported improvement in Retail GPU, the primary
metric Defendants provided for retail profitability, while simultaneously concealing that
0%
100%
200%
300%
400%
500%
Q1 20Q2 20Q3 20Q4 20Q1 21Q2 21Q3 21Q4 21Q1 22
Wholesale Growth Exceeded Retail Growth
y/y growth (retail cars sold)
y/y growth (wholesale cars sold)
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 245 of 332
- 242 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
buying cars from customers was also adversely impacting another key measure of
profitability: unit economics per retail vehicle sold.
(h)
The fact that Defendants created a misleading impression as to
Carvana’s actual retail sales profitability, or lack thereof, during the Class Period is further
supported by a series of revelations, and corresponding analyst reactions, that began during
2022. First, in April 2022, Carvana issued disappointing Q1 2022 financial results, which
partially revealed the truth about Carvana’s actual profitability, including the operations
expenses it had concealed from investors. For example, Defendants disclosed a quarterly
loss of $506 million and explained that they needed to purchase ADESA to “improve our
logistics network,” and bring down “shipping distances, times, and costs.” Analysts
commented on these revelations. For example, a J.P. Morgan analyst labeled the news
“‘confidence shattering’” and commented on Carvana’s opaque SG&A disclosures, which
encompassed operations expenses, stating: “‘we do not believe Carvana (CVNA) is likely to
get a free pass on SG&A any longer.’” Then, in May 2022, Carvana announced that it was
laying off approximately 2,500 employees “primarily in operational groups” and issued an
updated operating plan that emphasized its need to “[r]apidly reduce SG&A expense per
retail unit sold” (i.e., operations expenses) and make profitability a priority. Analysts
commented on this surprising news. For example, on May 21, 2022, Forbes cited to several
industry analysts who described Carvana’s “spendthrift business, whose growth-at-all-costs
mentality undermined business operations and sowed the seeds of its recent layoffs.” In
November 2022, Carvana issued disappointing Q3 2022 financial results and revealed that,
unbeknownst to investors, Carvana had been “frequently acquir[ing]” retail sales “that were
less profitable in the immediate period,” including “less profitable sales” in “markets with
lower profitability due to long distance from inventory.” Carvana further admitted it was
now taking actions “to improve profitability.” Analysts and the media reported on the news
regarding Carvana’s retail sales profitability. For example, a Needham & Company analyst
noted, “‘[t]he path forward for Carvana is to sell as many cars as possible, but to do so on a
profitable basis, versus prior it was more about selling as many cars as possible.’” Another
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 246 of 332
- 243 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
media outlet told investors to “[s]tep back and think about [Carvana’s disclosure] for a
moment” because “[i]t means that the company had historically grown in these markets
even though they weren’t all that attractive. It was growth for growth’s sake and not
necessarily growth to improve profitability.” Finally, on the last day of the Class Period,
Defendants revealed that Carvana was intentionally foregoing retail sales growth for the
foreseeable future to instead focus on driving “positive unit economics” (i.e., retail sales that
were actually profitable). Analysts commented on this disclosure. For example, a Morgan
Stanley analyst, on February 24, 2023, explained that “4Q was a sizeable miss,” and
commented on Carvana’s “[s]hift in strategy away from growth and towards EBITDA
profitability.” The same day, a RBC Capital Markets analyst also noted that Carvana’s GPU
“miss[ed] Street expectations by 32% . . . due in large part to . . . lower unit economics from
cars.” Additionally, a Stephens analyst noted “we do not foresee the company achieving its
targeting unit economics this year.”
259. Statement No. 39: On February 24, 2022 Jenkins stated on the Carvana Q4
2021 earnings call:
Retail GPU increased by $230 in Q4, primarily driven by an increase in the
percentage of retail vehicles sourced from customers.
260. As detailed below, Statement No. 39 made by Carvana and Jenkins omitted
facts that affirmatively created an impression of a state of affairs that differed in a material
way from the one that actually existed.
(a)
During the Class Period, Carvana acknowledged that “[t]he Company’s
primary business objective is to sell used vehicles to retail customers and generate a profit
doing so.” Thus, Retail GPU, which measured the gross profit Carvana earned per retail
vehicle sold (before adding in other ancillary revenue and profit streams such as warranty,
insurance, and financing), was a critical metric for Carvana during the Class Period.
Investors tracked Retail GPU to evaluate the profitability of Carvana’s retail vehicle sales.
Accordingly, in Q4 2021, Defendants disclosed “Retail GPU increased by $230 in Q4” to
$1,495, which investors viewed as a proxy for Carvana’s “unit economics” (i.e., per-vehicle
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 247 of 332
- 244 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
profitability). However, as alleged herein, Defendants concealed Carvana’s actual per-
vehicle profitability – or lack thereof. Carvana did so by: (i) excluding certain per-vehicle
operations expenses from its calculation of Retail GPU; and (ii) not disclosing or quantifying
these excluded costs separately so investors could decipher Carvana’s overall profitability of
retail sales on their own. These per-vehicle operations expenses, which were only separately
broken out and quantified for the first time after the Class Period, included material costs
associated with completing a retail sale, such as the full cost to ship a car to a customer and
title and registration costs. By intentionally hiding these significant costs, investors were left
to rely on the Retail GPU reported by Defendants – a positive GPU of $1,495 per retail
vehicle sold in Q4 2021, and an 18% improvement versus Q4 2020. However, had Carvana
included these operations expenses in its Retail GPU calculation or separately disclosed and
quantified these per-vehicle costs associated with completing retail sales, it would have
revealed to investors that, on average, Carvana generated negative unit economics of more
than $1,248 per retail unit sold and retail profitability actually decreased from Q4 2021, as
shown in the chart below. This created an impression of a state affairs (positive per-unit
profitability and profitability increased) materially different from the one that actually
existed (negative per-unit profitability and profitability decreased).
Q4 2020
Q4 2021
Disclosed Average selling
price per retail unit sold
$20,720
$25,364
Disclosed Cost of sales per
retail unit sold
$19,455
$23,869
Disclosed Retail GPU
$1,265
$1,495
Disclosed Improvement in
profitability per retail unit sold
--
+18%
Undisclosed
Operations
expense per retail unit sold
$2,200
$2,743
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 248 of 332
- 245 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Undisclosed Actual total
expense per retail unit sold144
$21,655
$26,612
Undisclosed Actual negative
“unit economics” per retail unit
sold 145
($935)
($1,248)
Undisclosed Actual decline in
profitability per vehicle sold
--
(34%)
(b)
After the Class Period, Carvana admitted in its Q3 2023 Shareholder
Letter and November 2, 2023 Cost Structure Details presentation that the operations
expenses it had concealed from investors during the Class Period were a “key driver[]” of
“unit economics” (i.e., per-vehicle profitability).146 Specifically, Defendants admitted that,
internally, they viewed “two key drivers” of Carvana’s “unit economics” (i.e., profitability
per retail vehicle sold): (i) retail cost of sales (costs which Carvana included in Retail GPU
during the Class Period); and (ii) “operations expenses per retail unit sold” (costs which
Carvana excluded from Retail GPU during the Class Period and which had never previously
been separately disclosed or quantified for investors). Defendants further admitted that,
although these vehicle operations expenses had previously been buried and comingled in
“SG&A” expenses and excluded from Retail GPU, they were not fixed or corporate
overhead expenses like typical SG&A costs. Rather, Defendants admitted that vehicle
operations costs were the direct per-vehicle costs of completing a vehicle sale (“[o]perations
144 Actual total expense per retail unit sold reflects the cost of sales per retail unit sold (as
disclosed) in addition to the undisclosed operations expense per retail unit sold.
145 Actual loss per retail unit sold reflects the average vehicle selling price per retail unit sold
(as disclosed) less the undisclosed actual total expense per retail unit sold.
146 After disclosing and quantifying operations expenses for the first time in November
2023, Jenkins acknowledged that the additional disclosure “allows a much deeper dive into
the economics of Carvana.” Jenkins added: “we’ve really been focused on . . . improving the
unit economics of the business. In other words, driving down variable costs per car,” which
he confirmed included “operations expenses, which are the more variable component of our
selling, general and administrative expenses.” Jenkins continued: “And the reason that we’re
focused on that is obviously, the more gains we make on reducing variable cost per car
today, the more profitable our growth can be in the future . . . .” Jenkins also described
“driving down operational expenses per unit” as one of Carvana’s “profitability initiatives,
and in particular, . . . our unit economics initiatives.” Thus, according to Jenkins, operations
expenses were a critical component of assessing Carvana’s profitability.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 249 of 332
- 246 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
expenses include the fulfillment, customer service, and transaction expenses associated with
completing retail and wholesale vehicle sales”). Defendants acknowledged that these
previously undisclosed and unquantified vehicle “operations expenses” included:
“customer care, multi-car logistics, last-mile delivery, and other operations
payroll” associated with completing retail vehicle sales;
“non-payroll logistics expenses, including fuel, repairs and maintenance, and
third-party transport services” associated with completing retail vehicle sales;
and
“[t]ransaction and other expenses, including limited warranty, title and
registration, and finance platform expenses” associated with completing retail
vehicle sales.
(c)
Carvana’s concealment of “operations expenses” when describing
Carvana’s retail profitability was especially misleading in light of the fact that Defendants
internally viewed “the underlying costs of completing a sale” as a critical component of
Carvana’s actual retail vehicle profitability. Indeed, as Garcia Junior later acknowledged:
When we think about trying to aim for more profitable sales, what does that
mean? There’s certainly variability in the kind of gross profit associated with
different types of sales. . . .
And then I think there’s a number of other dynamics kind of across car
type, et cetera. There’s also kind of variation in the underlying costs of
completing a sale.
Because Defendants did not specifically break out operations expenses (i.e., the underlying
costs of completing a retail sale), and instead buried and comingled them among SG&A
expenses, investors could not calculate the actual overall profitability of Carvana’s retail
vehicle sales, like Garcia Junior did internally. After the Class Period, in November 2023,
Defendants further admitted that, internally, they viewed operations expenses (i.e., costs
“associated with completing retail and wholesale vehicle sales” as a “key driver[]” of “unit
economics” (i.e., per-vehicle profitability). Garcia Junior also described Carvana’s
disclosure of operations expenses in November 2023 as “clear evidence” of Carvana’s
potential profitability.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 250 of 332
- 247 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
(d)
Defendants touted the increase in Retail GPU in Q4 2021 as a purported
improvement in Carvana’s per-vehicle profitability, but concealed that another “key driver[]”
of per-vehicle profitability, “operations expenses per retail unit sold,” had spiked in Q4 2021.
In fact, when Defendants disclosed and quantified operations expenses for the first time in
November 2023, they revealed that these previously concealed expenses had spiked 75%
during the Class Period, as shown in the first chart below.147 By Q4 2021, these costs had
spiked 49% since Q2 2020, as depicted in the second chart below.
147 See Source: Slide from November 2023 Cost Structure Details investor presentation
(emphasis in red added).
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 251 of 332
- 248 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
(e)
During Q4 2021, Defendants included per-vehicle operations expenses
in SG&A expenses (rather than Retail GPU), but they did not separately disclose or quantify
these expenses in a manner that would have allowed investors to decipher the overall
profitability (i.e., “unit economics”) of Carvana’s retail sales on their own. In fact, as
detailed below, the disclosures Defendants did make regarding SG&A expenses during Q4
2021 did more to mislead investors than to provide transparency around Carvana’s true
profitability per retail vehicle sold (i.e., “unit economics”).
(i)
First, Defendants’ Class Period description of SG&A expenses
was so vague that investors could not have discerned that material operations expenses
pertaining to Carvana’s actual per-vehicle profitability (i.e., “unit economics”) were included
in SG&A and excluded from Carvana’s Retail GPU calculation. For example, during the
Class Period, Defendants emphasized that SG&A costs consisted primarily of corporate and
overhead expenses, as is common across all companies.148 Defendants, however, omitted the
148 According to Carvana’s Q2 2020 10-Q: “Selling, general and administrative (‘SG&A’)
expenses include expenses associated with advertising and providing customer service to
customers, operating our vending machines and hubs, operating our logistics and
fulfillment network and other corporate overhead expenses, including expenses associated
with information technology, product development, engineering, legal, accounting,
finance, and business development.”
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 252 of 332
- 249 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
fact that 50% of its total reported SG&A costs were actually not corporate and overhead
expenses but variable “[o]perations expenses” which, as they later revealed, “include[d] the
fulfillment, customer service, and transaction expenses associated with completing retail
and wholesale vehicle sales.”149 Defendants’ intentionally opaque SG&A disclosures kept
analysts and investors in the dark. In fact, numerous analysts referred to Carvana’s SG&A
disclosures as a “black box.” For example, on April 27, 2020, a Morgan Stanley analyst
noted “[w]e (and the investors we spoke with) continue to look for more disclosure on
Other SG&A . . . [it] is a bit of a ‘black box.’” Likewise, on February 26, 2021, a J.P.
Morgan analyst referred to Carvana’s disclosure of SG&A expenses as “[t]he big ‘black
box.’”
(ii)
Second, beyond Defendants’ vague general description of SG&A
expenses, any references to certain specific expenses that were included in SG&A during Q4
2021 also painted a materially incomplete picture of which operations expenses had been
excluded from Carvana’s Retail GPU calculation. For example, Defendants broke out
SG&A into five high-level categories during the Class Period, with general overhead
categories like compensation (“all payroll and related costs, including benefits, payroll taxes,
and equity-based compensation”), advertising, and “other” (including “general and
administrative expenses such as IT expenses, corporate occupancy, professional services and
insurance”), making up over 89% of the total reported SG&A costs in Q4 2021.150 The only
specific SG&A category disclosed by Defendants that was loosely tied to Carvana’s post-
Class Period disclosure of per-vehicle operations expenses (i.e., expenses that were directly
“associated with completing retail and wholesale vehicle sales”) was “logistics.” But even
149 50% represents FY 2021 total SG&A costs of $2.03 billion compared to FY 2021 total
operations expense of $1.01 billion.
150 FY 2021 10-K: Compensation and benefits, advertising, market occupancy, logistics, and
other.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 253 of 332
- 250 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
the logistics disclosure was inconsistent with Carvana’s post-Class Period disclosures
revealing actual per-vehicle operations expenses.151
(iii)
Third, the quantification of any specifically identified operations
expenses within SG&A during the Class Period would have certainly left investors with a
false impression of which expenses had been excluded from Carvana’s Retail GPU
calculation and, thus, what Carvana’s actual per-vehicle profitability (i.e., unit economics)
looked like during the Class Period. For example, the only specific retail vehicle operations
expense Defendants obliquely quantified during the Class Period was “logistics” costs,
which Defendants vaguely described as encompassing outbound shipping costs. Carvana
disclosed logistics costs of $44 million in Q4 2021, or $389 per retail vehicle sold.
Meanwhile, operations expense per retail vehicle sold totaled approximately $310 million in
Q4 2021, or $2,743 per vehicle sold. Defendants did not specifically quantify any other
specific vehicle operations expenses in Carvana’s Q4 2021 SG&A disclosures.152 Thus, at
most, investors would have been specifically aware of just $389, less than 15%, of the
$2,743 in actual per-vehicle operations expenses (i.e., costs “associated with completing
retail . . . vehicle sales”) that materially impacted Carvana’s retail vehicles sales profitability
(“unit economics”) in Q4 2021.
(iv)
Fourth, Defendants’ Class Period SG&A disclosures did not
break out fixed versus variable SG&A expenses. Like most companies, Carvana incurred
variable expenses (incremental expenses on each unit sold) and overhead expenses (fixed
151 Carvana disclosed that the logistics SG&A category, also disclosed as shipping and
handling, included “fuel, maintenance and depreciation related to operating our own
transportation fleet, and third party transportation fees.” This disclosure, however, was
inconsistent with Carvana’s post-Class Period disclosures revealing operations expenses
because: (i) Carvana included depreciation and amortization expenses in the SG&A logistics
category – these were overhead costs rather than per-vehicle shipping costs and, thus, were
specifically excluded from operations expenses; and (ii) Carvana excluded compensation
associated with outbound shipping from the SG&A logistics category but included these
expenses in the per-vehicle shipping costs it incurred and, thus, in operations expenses
disclosed after the Class Period.
152 Title and registration expenses were listed among expenses in the “Other” SG&A
category, but were not separately quantified.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 254 of 332
- 251 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
expenses that did not significantly change based on the number of units sold). Operations
expenses, i.e., variable costs incurred in completing retail sales, increased with each
incremental vehicle sold and, thus, had a direct impact on per-vehicle profitability.
Meanwhile, fixed SG&A expenses, such as corporate overhead expenses, did not directly
impact the amount of profit Carvana earned on each vehicle it sold. During the Class Period,
analysts asked Defendants for more clarity on per-vehicle operations expenses (i.e., variable
versus fixed SG&A costs). For example, on the Q4 2020 earnings call, a Wells Fargo
analyst asked Jenkins: “[C]an you talk about incremental SG&A per unit, and walk me
through how to think about the SG&A . . .? And what is expected to be a fixed cost, . . . and
then how much would be a variable cost?” Tellingly, Jenkins declined to provide a clear
response to the question. On February 26, 2021, a J.P. Morgan analyst expressed concern
over the lack of disclosure, stating “we are concerned if this bucket of SG&A is starting to
become more variable versus fixed in nature.” When Carvana finally disclosed operations
expense per retail unit in November 2023, a J.P. Morgan analyst commented that “the
company finally provided disclosure on fixed vs. variable SG&A expenses.” The same
analyst revealed that, due to the opaque nature of Carvana’s Class Period SG&A disclosures,
his estimate of fixed versus variable costs was significantly off: “[T]he implied fixed cost
mix of ~55-60% . . . was lower than our prior assumption of ~70% that was driving our
[earnings model].” In other words, the analyst had viewed Carvana’s SG&A expenses as
predominantly fixed and was surprised to learn that, in reality, a significantly higher
proportion of Carvana’s SG&A expenses were variable operations expenses, which
increased with each incremental vehicle sold and negatively impacted Carvana’s retail
profitability (i.e., unit economics).
(f)
The operations expenses that were excluded from Retail GPU (and not
separately disclosed or quantified to allow investors to decipher Carvana’s overall
profitability of retail sales on their own) were directly tied to problems stemming from
Carvana’s undisclosed growth-at-any-cost scheme, as described herein. For example, by
concealing a complete picture of “multi-car logistics, last-mile delivery . . . [n]on-payroll
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 255 of 332
- 252 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
logistics expenses . . . third-party transport services . . . [t]transaction . . . expenses, [and]
title and registration expenses” that “were associated with completing retail . . . vehicle
sales,” Defendants were able to conceal numerous problems from investors, including that:
(i) Carvana was boosting retail sales with unprofitable sales in distant markets; (ii) Carvana’s
rapid increase in buying cars from customers left it with a glut of low-quality cars that were
causing massive logistics constraints and costs, and (iii) Carvana was experiencing
significant title and registration issues.
(g)
Defendants stated that increase in retail sales profitability was “driven
by an increase in the percentage of retail vehicles sourced from customers.” Meanwhile,
Defendants omitted the fact that buying cars from customers was also adversely impacting
operations expense and unit economics – another key measure of actual retail profitability.
As alleged herein, the rapid increase in buying low-quality cars from customers caused a
significant spike in wholesale vehicle sales. In Q4 2021, wholesale sales growth spiked over
100% year-over-year for the fifth straight quarter, far outpacing retail sales growth, as shown
in the first chart below. Meanwhile, wholesale sales made up 30% of the Company’s total
sales in Q3 2021, as shown in the second chart below.
Carvana later admitted that it experienced significant logistics constraints and costs as a
result of buying cars from customers and its corresponding surplus of wholesale vehicle
sales. For example, in Carvana’s May 2022 Operating Plan, Defendants admitted the need to
0%
100%
200%
300%
400%
500%
Q1 20Q2 20Q3 20Q4 20Q1 21Q2 21Q3 21Q4 21Q1 22
Wholesale Growth Exceeded Retail Growth
y/y growth (retail cars sold)
y/y growth (wholesale cars sold)
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 256 of 332
- 253 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
“[r]apidly reduce SG&A expense per retail unit sold” (i.e., operations expenses). Defendants
admitted that “[b]uying cars from customers growth” led to “additional wholesale volume”
which, in turn, led to “constraints in our nationwide logistics network” and a spike in SG&A
logistics costs. Defendants further disclosed that “wholesale units acquired from customers
have . . . increased complexity in our multi-car logistics network.” Nevertheless, the
majority of the negative cost impacts of these logistics constraints were not reflected in
Carvana’s Q4 2021 Retail GPU because Defendants categorized many of these costs as
operations expenses, which were buried in SG&A. Meanwhile, these costs did contribute to
the $2,743 in undisclosed “operations expense per retail unit sold” in Q4 2021 and, therefore,
negatively impacted Carvana’s retail “unit economics” (i.e., overall profitability per retail
vehicle sold). Thus, it was materially misleading to describe buying cars from customers as
a driver of an increase in Retail GPU – the only metric Defendants provided for retail
profitability in Q4 2021 – while simultaneously concealing that buying cars from customers
was also adversely impacting another key measure of profitability: unit economics per retail
vehicle sold.
(h)
The fact that Defendants created a misleading impression as to
Carvana’s actual retail sales profitability, or lack thereof, during the Class Period is further
supported by a series of revelations, and corresponding analyst reactions, that began during
2022. First, in April 2022, Carvana issued disappointing Q1 2022 financial results, which
partially revealed the truth about Carvana’s actual profitability, including the operations
expenses it had concealed from investors. For example, Defendants disclosed a quarterly
loss of $506 million and conceded that Carvana needed to purchase ADESA to “improve our
logistics network,” and bring down “shipping distances, times, and costs.” Analysts
commented on these revelations. For example, a J.P. Morgan analyst labeled the news
“‘confidence shattering’” and commented on Carvana’s opaque SG&A disclosures, which
encompassed operations expenses, stating: “‘we do not believe Carvana (CVNA) is likely to
get a free pass on SG&A any longer.’” Then, in May 2022, Carvana announced that it was
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 257 of 332
- 254 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
laying off approximately 2,500 employees “primarily in operational groups” and issued an
updated operating plan that emphasized the need to “[r]apidly reduce SG&A expense per
retail unit sold” (i.e., operations expenses) and make profitability a priority. Analysts
commented on this surprising news. For example, on May 21, 2022, Forbes cited to several
industry analysts who described Carvana’s “spendthrift business, whose growth-at-all-costs
mentality undermined business operations and sowed the seeds of its recent layoffs.” In
November 2022, Carvana issued disappointing Q3 2022 financial results and revealed that,
unbeknownst to investors, Carvana had been “frequently acquir[ing]” retail sales “that were
less profitable in the immediate period,” including “less profitable sales” in “markets with
lower profitability due to long distance from inventory.” Carvana further admitted it was
now taking actions “to improve profitability.” Analysts and the media reported on the news
regarding Carvana’s retail sales profitability. For example, a Needham & Company analyst
noted, “‘[t]he path forward for Carvana is to sell as many cars as possible, but to do so on a
profitable basis, versus prior it was more about selling as many cars as possible.’” Another
media outlet told investors to “[s]tep back and think about [Carvana’s disclosure] for a
moment” because “[i]t means that the company had historically grown in these markets
even though they weren’t all that attractive. It was growth for growth’s sake and not
necessarily growth to improve profitability.” Finally, on the last day of the Class Period,
Defendants revealed that Carvana was intentionally foregoing retail sales growth for the
foreseeable future to instead focus on driving “positive unit economics” (i.e., retail sales that
were actually profitable). Analysts commented on this disclosure. For example, a Morgan
Stanley analyst on February 24, 2023, explained that “4Q was a sizeable miss,” and
commented on Carvana’s “[s]hift in strategy away from growth and towards EBITDA
profitability.” The same day, a RBC Capital Markets analyst also noted that Carvana’s GPU
“miss[ed] Street expectations by 32% . . . due in large part to . . . lower unit economics from
cars.” Additionally, a Stephens analyst noted “we do not foresee the company achieving its
targeting unit economics this year.”
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 258 of 332
- 255 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
VIII. ADDITIONAL ALLEGATIONS OF SCIENTER
A.
Insider Stock Sales Support a Motive to Commit Fraud
261. Defendants were highly motivated to engage in their fraudulent scheme and
course of conduct to sell nearly 14.3 million shares of Carvana stock for proceeds of nearly
$3.76 billion. Notably, Garcia Senior and Jenkins sold 13.7 million shares – worth nearly
$3.6 billion – before the first partial disclosure of Defendants’ scheme to defraud or
misrepresentations and omissions.
1.
Garcia Senior’s Class Period Sales and Insider Trading
262. Garcia Senior was motivated to work hand-in-glove with his son, Garcia
Junior, to manipulate Carvana’s stock price. Indeed, he sold 13.95 million shares of his
personally held Carvana stock at prices as high as $376.55 per share – 43 times Carvana’s
stock price at the close of the Class Period – for nearly $3.7 billion in total proceeds.
263. The amount and time of Garcia Senior’s sales were highly suspicious. For
example, all of his 13.95 million sales – 15% of his substantial shareholdings – were sold in
a concentrated period between October 30, 2020 and August 23, 2021, when he sold every
single trading day.153 Suspiciously, Garcia Senior chose to sell for only nine trading days
following the first partial corrective disclosure; as noted by the media in the following
fourteen months as Carvana’s stock price collapsed, Garcia Senior never sold again during
the Class Period. Remarkably, Garcia Sr. sold 96% of his Carvana stock before the first
corrective disclosure.154
264. In stark contrast to his Class Period sales, in the nearly three years since
Carvana’s IPO, Garcia Senior did not sell a single share of his Carvana personal holdings.
Rather, he sold fewer than 10.9 million shares of Carvana stock that he owned indirectly
through DriveTime (of which Garcia Junior and his children own nearly a quarter through a
153 Garcia Senior’s first sale, which came after Defendants had pumped Carvana’s stock
more than 114% since the start of the Class Period, was larger than all his other sales.
154 John Hyatt, Carvana CEO’s Net Worth Skids But His Dad, Who Controls Company, Is
Worth Nearly $3 Billion, Forbes (Dec. 8, 2022) (the “December 8, 2022 Forbes article”).
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 259 of 332
- 256 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
trust) or his company Verde Investments for proceeds of $456.3 million – only 12% of the
personal proceeds he realized during the Class Period.155
265. Further, during the Class Period, Garcia Senior sold two million shares outside
of his trading plan, enabling him to pocket $478.4 million as Carvana’s stock price soared in
December 2020 as a result of the pump-and-dump scheme. In addition, Garcia Senior’s
trades pursuant to his 10b5-1 trading plan also support a strong inference as he modified his
June 15, 2020 10b5-1 plan twice within a year of its adoption date to accelerate the number
of shares he could sell at artificially inflated prices. Indeed, on November 4, 2020 – less than
four days after his first trade and less than five months into his new 10b5-1 trading plan –
Garcia Senior modified his trading plan while in possession of MNPI to “adjust certain
minimum trading price conditions,” which accelerated the rate per day at which he could sell
his Carvana shares to unsuspecting investors. See Carvana’s SEC Form SC 13D/A, filed on
November 6, 2020. Worse, less than six months after this suspicious modification, Garcia
Senior modified his trading plan for a second time on May 20, 2021 – at the height of the
scheme – to accelerate the number of shares he could sell per day. In fact, the day after this
modification became effective and the stock price reached $300 as a result of the scheme,
Garcia Senior began selling 60,000 Carvana shares per day, as compared to 30,000 a day –
his trading rate before any modifications. As shown below, Garcia Senior sold the vast
majority of his stock during the Class Period outside of a 10b5-1 plan altogether or pursuant
to these modified plans.
155 Because DriveTime and Verde are private companies, complete information regarding
their ownership is unavailable.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 260 of 332
- 257 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
266. Academics noted the unusual and suspicious nature of Garcia Senior’s
modifications. For example, as reported in a September 17, 2021 The Wall Street Journal
article, Professor Taylor of the Wharton School of Business explained that Garcia Senior’s
frequent modification of his plan raises serious red flags: “‘I’ve studied 20,000 10b5-1 plans,
[and] I can’t recall another of this size where there are modifications every six months.’”
Unsurprisingly, Professor Taylor stated: “‘What I’m saying is the Garcias knew it was
short-lived . . . . The Garcias knew the music would eventually end.’”156 Additionally, in
the December 8, 2022 Forbes Article, David F. Larcker, a professor of accounting at
Stanford Graduate School of Business, told Forbes: “The active modification of a 10b5-1
156 See John Hyatt, Carvana’s ‘Chaotic’ Zoom Firing Caps Company’s Struggles Amid
Market Downturn, Forbes (May 21, 2022).
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 261 of 332
- 258 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
plan raises the question of whether someone is trying to rebalance their portfolio, time the
market, . . . or are they doing it for other reasons that may not pass regulatory scrutiny.”157
267. In sum, Garcia Senior’s sales of a significant amount of his personal Carvana
shareholdings at artificially inflated prices while he was actively engaged in the fraudulent
scheme detailed herein and before the partial disclosures began further provides a compelling
inference of scienter.
2.
Jenkins’s Class Period Sales and Insider Trading
268. During the Class Period, Jenkins sold 336,922 shares of his Carvana stock,
48% of his holdings, for proceeds of over $79 million. These sales all occurred between
August 7, 2020 and November 8, 2021, just as Carvana’s stock peaked.158 In the nearly three
years since Carvana’s IPO, Jenkins sold 13,000 fewer shares of his Carvana stock for
proceeds of only $18.5 million – only 23% of the proceeds he realized during the Class
Period. Moreover, Jenkins’s annual salary for the fiscal year ended December 31, 2020 was
$384,395. Thus, the proceeds from his insider sales during the Class Period were over 206
times the amount he received in salary. His sales can be seen in the following chart:
157 In the April 2022 Offering, Garcia Senior purchased 5.1 million shares of Carvana stock
for $408 million and Garcia Junior purchased 2 million shares for $160 million. Their
purchases were necessary to ensure the execution of that offering. Garcia Senior also
purchased on June 10, 2022 (793,790 shares for $17.4 million) and on June 13, 2022 (less
than 1.2 million shares for $24.6 million) to artificially inflate Carvana’s stock price and
once again dump his over-valued shares.
158 Following November 8, 2021, Jenkins made one additional sale during the Class Period
on February 1, 2022. On this day, Jenkins sold a mere 7 shares for $1,153.46, which were
required sales to pay tax withholdings upon vesting of restricted stock units.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 262 of 332
- 259 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
269. Further, the timing of Jenkins’s trades are suspicious. After adopting a new
10b5-1 plan on June 15, 2020, Jenkins made his first sale on August 7, 2020 – an outsized
sale of 190,000 Carvana shares for proceeds of $38.5 million. Nine months later, Jenkins
adopted another 10b5-1 plan on March 15, 2021. There is no doubt that Jenkins was in
possession of MNPI at the time he entered his second plan in nine months. Indeed, Jenkins
had been meeting with the Board about title and registration issues since July 2020. See,
e.g., Ex. 1. Moreover, unbeknownst to investors, Ohio had already fined Carvana and
suspended its license, and MDOS was set to meet with Carvana’s executives within days of
his adoption after an investigation. By the time that this information was revealed Jenkins
had already sold $79.2 million worth of his Carvana stock. In sum, Jenkins’s sales of a
significant amount of his personal Carvana shareholdings at artificially inflated prices while
he was actively engaged in the fraudulent scheme and made materially misleading statements
and omissions detailed herein further provides a compelling inference of scienter.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 263 of 332
- 260 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
B.
Defendants’ 10b5-1 Plans Support a Strong Inference of Scienter
270. Although Garcia Senior and Jenkins each had a 10b5-1 trading plan in place
during the Class Period, they manipulated those plans while in possession of MNPI, which
further supports a strong inference of scienter. Indeed, during the Class Period, Jenkins
adopted two new plans within nine months; and Garcia Senior adopted a new plan, modified
his new plan, modified his plan again, and traded millions of shares outside of his modified
plan in less than one year. Thus, at every point that Defendants made the decision to trade,
they were aware of and participating in the Company’s pump-and-dump scheme and/or the
falsity of their statements.
271. Rule 10b5-1 was intended to provide a mechanism for insiders to sell while not
in possession of MNPI. By attempting to commit insiders to a predetermined plan, it moves
the relevant date at which the insider possessed material information from the date of the
trade to the date when the trade was planned. However, this purpose has frequently been
circumvented. A January 19, 2021 Stanford study, entitled “Gaming the System: Three “Red
Flags” of Potential 10b5-1 Abuse,” that examined over 20,000 10b5-1 plans between January
2016 and May 2020 found that a 10b5-1 plan would perform demonstrably better than its
counterparts when it had shorter periods between adoption or modification and the first trade,
which allows insiders to “systematically avoid losses.” The Stanford study concluded that
shorter periods between adoption or modification and the first trade were red flags of an
insider abusing his 10b5-1 plan. Additionally, a June 29, 2022 The Wall Street Journal
article entitled “CEO Stock Sales Raise Questions About Insider Trading,” analyzed 75,000
sales by corporate insiders from 2016 through 2021 and found that “insiders who sold within
60 days [of plan adoption] reaped $500 million more in profits than they would have if they
sold three months later.”
272. This manipulation, which defeats the purpose of a trading plan, has prompted
the SEC and the Department of Justice (“DOJ”) to take recent action. For example, in a
March 1, 2023 DOJ press release, the DOJ indicted a CEO for insider trading “exclusively
on [his] use of Rule 10b5-1 trading plans,” explaining it “‘will not allow corrupt executives
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 264 of 332
- 261 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
to misuse 10b5-1 plans as a shield for insider trading.’” Further, the SEC effectively
amended the Rule on February 26, 2023 to “target corporate insider trading” and “remove
many of the loopholes that allowed corporate insiders to hide behind these trading plans.”159
Notably, under these new rules, executives have to wait at least 90 days after starting or
modifying a 10b5-1 plan before they can trade under the plan. Id. Tellingly, certain sales by
both Garcia Senior and Jenkins during the Class Period would have been impermissible
under this new rule.
273. Because Garcia Senior and Jenkins manipulated their 10b5-1 plans to defeat
the very purpose of Rule 10b5-1, their plans provide even further evidence of an already
compelling inference of scienter.
C.
Multiple Confidential Witnesses Raised Quality, Indiscriminate
Growth, and Title Problems with Managers, Including Garcia
Junior and Jenkins
274. As the confidential witness accounts demonstrate, Defendants knew or
recklessly disregarded that Carvana had lowered its purchasing and verification standards
when buying cars from customers to induce trade-ins to boost retail sales and concealed this
information from investors. Defendants also knew or recklessly disregarded that the rapid
increase in buying cars from customers caused a spike in the number of cars that Carvana
was forced to sell wholesale at a loss. Indeed, Garcia Junior and Jenkins participated in calls
and attended meetings at which the low quality of cars being purchased from customers was
frequently raised and discussed.
275. For example, CW-5 attended a leadership summit with all Market Operations
Managers and senior leadership, including Garcia Junior, in April 2022 at Carvana’s
headquarters. During a Q&A session with Garcia Junior, CW-5 recalled that someone asked
159 Jonathan Weil, New SEC Rules Target Corporate Insider Trading: Loopholes will close
for executives selling company stock, Wall St. J. (Feb. 13, 2023). Under these new rules,
Garcia Senior and Jenkins would have had to certify that at the time of the adoption of their
new or modified plans they were: (1) not aware of MNPI about the issuer or its securities;
and (2) adopting the plan in good faith and not as part of a plan or scheme to evade the
prohibitions of Rule 10b-5.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 265 of 332
- 262 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
about what was going to be done regarding the quality issues affecting the vehicles that the
hubs were receiving from the IRCs.
276. CW-11, too, attended a meeting where the poor quality of vehicles purchased
from customers was discussed with Garcia Junior. Further, employees at CW-11’s level and
market team leaders attended Zoom calls with the Company’s C-level executives (including
Garcia Junior) probably every quarter. On one of these Zoom calls in early 2021 attended by
Garcia Junior, an attendee expressed concern about the poor quality of the cars Carvana was
purchasing. According to CW-11, one of the C-level executives responded: Carvana was
“just worried about growth and not procedural” operations. CW-11 recalled that this same
C-level executive spoke about the paramount importance of growth on that call. There was
time for questions and answers at the end of these calls, and during every call someone in a
market operations role complained about the poor quality of the vehicles they received from
the IRCs. CW-11 said that this was the number one complaint from the operations side of
the business.
277. Other confidential witnesses raised concerns about the eroding purchasing
standards and the mounting costs of excess inventory with director-level employees at
Carvana. Around the fall and winter of 2021, CW-8’s team began to raise concerns about
the high level of inventory to director-level employees because it was becoming difficult to
sell certain types of vehicles and CW-8’s team had to lower prices and margins were
decreasing. CW-8’s team had weekly meetings every Thursday with an Associate Director
where they would go over each segment and provide inventory updates. CW-8’s teammates
raised concerns about the excess inventory in a couple of these meetings and one of his/her
coworkers, who was a Financial Analyst, made a report about this topic. However, CW-8’s
team’s concerns were ignored, and the Buyers were told to continue buying vehicles.
Likewise, CW-1 said that, after operating under the directive “to purchase as much as was
humanly possible,” which resulted in Carvana purchasing “everything” regardless of the
quality, Carvana’s subpar and bloated inventory issues were “obvious to anyone with eyes.”
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 266 of 332
- 263 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Finally, when CW-2 reported his/her concerns that Carvana was purchasing vehicles that
were “not safe to drive,” “leadership would say take it” anyway.
278. In addition, the confidential witness accounts demonstrate that Defendants
knew or recklessly disregarded that Carvana was overrun with low-quality wholesale cars
and did not have a “capital-light” expansion model, such that Carvana had to issue billions of
dollars of high-interest junk bonds to acquire a nationwide auction house, ADESA. In fact,
CW-12 met with Garcia Junior, who informed CW-12 that Carvana’s management team
needed ADESA’s reconditioning centers located nationwide. CW-4 and CW-7 also stated
that Carvana’s acquisition of ADESA was part of Carvana’s expansion efforts.
279. The confidential witness accounts also show that Defendants knew or
recklessly disregarded that, in order to boost sales, Carvana purchased cars without title and
sold cars before it held title to those cars. For example, CW-9 described attending an all-
hands call in January 2022 in which employees told Garcia Junior that they were being
inundated with calls regarding registration problems. In addition, CW-3 discussed the title
problems with his/her manager and at weekly meetings with the Director of Wholesale, but
the Company brushed off CW-3’s concerns. CW-10 also raised the title problems with a
supervisor and then with a manager.
D.
The Fraud Involved the Company’s Core Operations, About
Which Defendants Held Themselves Out as Knowledgeable
280. Throughout the Class Period, Garcia Junior and Jenkins were CEO and CFO,
respectively. Moreover, as discussed in more detail below (§VIII.H.), Garcia Senior
controlled Carvana’s policies and operations. By virtue of their management and/or control
of the Company’s operations and their access to confidential proprietary information
concerning the Company’s operations and financial results, Defendants either knew about or
recklessly disregarded the alleged scheme and course of conduct, including Garcia Junior’s
and Jenkins’s material misstatements and omissions.
281. Defendants’ fraudulent course of conduct and material misstatements and
omissions concerned the very core of Carvana’s operations – retail sales growth – which
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 267 of 332
- 264 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Defendants emphasized as “the single most important metric in our business.” Indeed,
during the Class Period, Jenkins advised the SEC: “The Company’s primary business
objective is to sell used vehicles to retail customers and generate a profit doing so.”160
Further, Jenkins and Garcia Junior made and approved numerous specific statements
regarding the subject of the alleged fraudulent conduct and the information that Plaintiffs
allege was false or misleading. Indeed, Garcia Junior and Jenkins discussed retail unit sales
in every shareholder letter, SEC filing, and conference call. In addition, Garcia Junior and
Jenkins discussed Retail GPU in every shareholder letter and conference call. Moreover,
they repeatedly discussed title and registration risks and issues, buying cars from customers,
the Company’s capital-light, scalable expansion model, and the average days to sale metric
with their investors. These statements evince their knowledge of the subject matter and
support a strong inference of scienter.
282. Further, Defendants had actual access to the alleged fraudulent conduct and the
information that Plaintiffs allege was false or misleading. For example, a presentation from
a July 2020 Board Meeting attended by Jenkins and Garcia Junior reveals that Carvana had
undertaken an initiative “to reduce registration delays,” as part of the Company’s “Key
Focus Areas and Risks.” Ex. 1. Title and registration violations, including Carvana’s
suspensions and fines continued to be a prominent topic in every Board meeting until at least
December 2021. See, e.g., Ex. 2 (October 18, 2021 Board Minutes revealing that the Board
and Jenkins discussed how they can “explain the NC situation, amongst others?”); see also
Schertz, ECF 23. Notably, these Board meetings occurred before every quarterly earnings
release and were consistently attended by both Garcia Junior and Jenkins.
283. In addition, Defendants had access to data setting forth the total costs that
Carvana incurred to acquire, transport, and otherwise make cars available for sale, as well as
the final selling price of the vehicles, from which GPU was derived. CW-7 explained
Carvana used a Tableau data server which was updated daily and which revealed declining
160 August 23, 2022 letter from Jenkins to SEC, Division of Corporation Finance.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 268 of 332
- 265 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
GPU as the Company increasingly relied on purchasing vehicles from customers. The server
also contained data showing an increase in the average number of days on-site, which
measures the number of days between procuring and selling a vehicle. CW-8 and his/her
team also used Tableau, from which numerous reports could be derived, and Microsoft SQL,
which connected to a computer server where data was stored. Using this software enabled
CW-8’s team to see inventory data and various metrics, including how many vehicles
Carvana had in inventory, how long these vehicles had been sitting around without being
sold, Carvana’s website traffic, and supply/demand ratios.
284. Critically, on November 3, 2022, in response to an analyst, Garcia Junior
belatedly admitted that internally he viewed “the underlying costs of completing a sale” as a
critical component of Carvana’s actual retail vehicle profitability. Indeed, Garcia Junior
acknowledged:
When we think about trying to aim for more profitable sales, what does that
mean? There’s certainly variability in the kind of gross profit associated with
different types of sales. . . .
And then I think there’s a number of other dynamics kind of across car
type, et cetera. There’s also kind of variation in the underlying costs of
completing a sale.
Further, after the Class Period, in November 2023, Garcia Junior and Jenkins confirmed that,
internally, they viewed operations expenses (i.e., costs “associated with completing a retail
vehicle sale) as a “key driver[]” of “unit economics” (i.e., per-vehicle profitability).
285. Further, according to insiders, Carvana was run in a top-down fashion, with
“[t]he Garcias” maintaining an “iron grip” over the Company’s operations. For instance, the
December 8, 2022 Forbes Article explained:
The Garcias’ iron grip on Carvana’s is reflected in the younger
Garcia’s management style, according to one former Carvana executive who
spoke with Forbes on the condition of anonymity. “You will often hear
people talk about the good old boys club and it absolutely exists there,” says
the former employee. Higher ups who went against the CEO were quickly
shown the door, according to the former employee. “The culture is you either
are on board with us and everything we do and say and drink the Kool Aid,
or you are not and get out.”
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 269 of 332
- 266 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
286. To be sure, Defendants’ control and management of Carvana’s operations and
their access to confidential information concerning the Company’s operations and financial
results, further supports a strong inference of scienter.
E.
Analysts and Investors Frequently Asked Questions About, and
Garcia Junior and Jenkins Frequently Evinced Personal
Knowledge by Commenting on, the Subjects of Their Fraud
287. Throughout the Class Period, Wall Street analysts asked Garcia Junior and
Jenkins specific, direct questions about Carvana’s retail sales growth, buying cars from
customers, Carvana’s nationwide expansion, and title and registration issues. Garcia Junior
and Jenkins held themselves out to be knowledgeable about these subjects. For example:
(a)
On August 5, 2020, an analyst asked Jenkins, “on the retail GPU side,
you all had mentioned that you expected to see improvement . . . . Is that basically tied to the
increased production capacity . . . [o]r is there more that goes into that with pricing dynamics
and acquisition?” Jenkins responded:
So there are definitely some exciting trends in retail GPU. I think the most
exciting is that, as I mentioned earlier, in July, we bought more than 100% as
many cars from customers as we sold to customers. . . .
I think that’s obviously a positive, that drives wholesale GPU, that
drives incremental retail GPU, because cars that you acquire from
customers are typically more profitable than cars that you acquire at auction.
(b)
On October 29, 2020, an analyst asked Jenkins, “[i]s there a lot of extra
operating expense in that relative to buying at the auction?” Jenkins identified a number of
expenses and asserted, “[a]ll of those expenses are actually – those are reflected in our
wholesale gross profit number as well as in our retail gross profit number.”
(c)
On October 29, 2020, an analyst asked Garcia Junior to “give us a sense
of growth rates going on in [the Company’s older] markets.” Garcia Junior responded,
“internally, what we’re always focused on is what’s happening to underlying demand” and
that we’re “looking at what’s happening to sales and looking at the drivers of conversion
that impacts sales, which include where inventory is, how much inventory is close to
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 270 of 332
- 267 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
different markets, what delivery times are in different markets, what our inventory
distribution looks like.”
(d)
On March 1, 2021, an analyst asked, “how do you choose to expand
markets in any region specifically as you – on your way to getting to like that 95%
population coverage longer term?” Garcia Junior responded, “we’re largely in a place now
where if we wanted to just like open it all up, we probably could. It’s not operationally
difficult enough at this point where we couldn’t open up all of it very quickly.”
(e)
On May 6, 2021, an analyst asked Jenkins to “talk about the tight
inventory environment . . . and some of the measures Carvana has implemented to unlock
new channels for inventory acquisition and if you’re expecting those measures to have any
notable impacts on the near-term vehicle GPU.” Defendant Jenkins responded: “[O]ur #1
channel for sourcing inventory has now become sourcing cars directly from customers,
which has several advantages . . . from a sourcing perspective, our main area of focus is
going to be continue to grow the business of buying cars from customers.”
(f)
On August 5, 2021, an analyst asked Jenkins whether “there [is] any
way to unpack [the retail GPU number] a little bit in terms of how much of that is temporary
versus structural improvements . . . that might sustain not just like in the second half but into
next year and beyond?” Jenkins responded:
[L]et me start by taking the retail GPU question . . . [o]n retail GPU, we had
our first quarter over $2,000 retail GPU. We’ve come close a couple of
times before. I think we’ve done $1,850 before and $1,700, not including
adjustments also in 2020. But the $2,000 is a record for us. I think if we
think through the drivers, I think maybe it’s helpful to think through it
sequentially from Q1. The biggest driver there was certainly the
performance [of] buying cars from customers. . . .
[W]e feel really great about our retail GPU progress. Buying cars from
customers continues to be a strong driver of the improvements in the
business . . . .
(g)
On August 11, 2021, an analyst asked Jenkins about the North Carolina
suspension, and whether “you [are] comfortable that this is a very North Carolina specific
issue and not something you might need to investigate in like other states, just to make sure
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 271 of 332
- 268 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
it’s covered?” Jenkins misleadingly responded; “[W]e had quite a small fraction of
customers that were impacted by title and registration delays, . . . in the state of North
Carolina” and “[s]ure . . . our understanding is that this is quite unprecedented.”
(h)
On November 4, 2021, an analyst asked Garcia Junior to “talk about
how much of the increase in consumer purchases is just out of necessity . . . versus your view
that it’s just simply a more profitable and efficient channel for inventory?” Further, an
analyst asked, “do you envision pulling back on customer purchases perhaps to a point where
its back towards the levels that you were targeting at your Investor Day?” In a lengthy and
detailed response, Garcia Junior stated: “[Sourcing cars from customers] i[s] better. It allows
us to provide a high-quality experience to a customer that we’re buying a car from, and then
it gets us access to a higher-quality pool of inventory that is, on average, more profitable.”
(i)
On February 24, 2022, an analyst asked Garcia Junior about whether
Carvana would utilize ADESA’s title department “to make[] you more efficient on
processing titles out of your core business.” Garcia Junior responded:
Now because you brought it up, I do think the registration team and
title team inside Carvana . . . [has] done an unbelievable job over the last
probably 6 months . . . . We’re now kind of approaching similar levels of
success in title and registration to what we were experiencing pre-
pandemic. . . . So there’s a lot of great things going on there.
288. Garcia Junior and Jenkins demonstrated their visibility into and knowledge of
these critical areas at the heart of Defendants’ fraud, supporting a strong inference that they
issued the false and misleading statements alleged above with scienter. Further, if Garcia
Junior and Jenkins did not have detailed, up-to-date knowledge of these issues, then they
acted recklessly in holding themselves out as knowledgeable about these issues and in
discussing them in communications with investors and analysts.
F.
Defendants’ Decision to Cease Reporting Critical Metrics
Bolsters Their Scienter
289. Defendants’ scienter is further supported by their decision to cease reporting
key metrics that might have tipped investors off to their scheme.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 272 of 332
- 269 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
1.
Average Days to Sale
290. Average days to sale was one of Carvana’s six “Key Operating Metrics” since
becoming a public company. This makes sense as Carvana’s “business is dependent upon
our ability to expeditiously sell inventory” given that cars quickly depreciate. An “increase
[in] our average days to sale” served as a warning to investors that Carvana was failing to
rapidly sell its inventory, which “could have a material adverse effect” on Carvana’s margins
because of depreciation.
291. Although Defendants continued to view average days to sale as a key metric
during and after the Class Period, on May 6, 2021, as Carvana was facing a glut of inventory,
Defendants stopped reporting “average days to sale.” Worse, Defendants misleadingly
assured investors that it had replaced this “Key Operating Metric[]” with the number of IRCs
purportedly because “our number of IRCs is a more important metric than average days to
sale due to the impact of IRC capacity on retail units sold and the relative stability of
average days to sale over the past three years.” Defendants’ abrupt decision to pull an
admittedly key operating metric – and replace it with an entirely unrelated metric that
provided no visibility into whether Carvana was rapidly selling its inventory – supports an
inference that Defendants chose to conceal the metric because it was trending in the wrong
direction and would have alerted investors to their fraudulent conduct.
2.
Buying Cars from Customers
292. At Carvana’s Analyst Day, Jenkins and Garcia Junior announced Carvana’s
long-term goal to purchase 38%-52% of its vehicles from customers (as compared to
Carvana’s then-current level of 16% purchased from customers). In the quarters following,
the Company devoted an entire section of its shareholder letters to providing detailed metrics
on buying cars from customers to demonstrate the Company’s progress towards Defendants’
purported goal of growth and long-term profitability. These detailed metrics included
graphics regarding the number of customer vehicles acquired, the customer-sourced ratio,
and the percentage of customer vehicles acquired as a percent of retail units sold.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 273 of 332
- 270 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
293. Defendants’ scheme, however, depended on drastically increasing the volume
of cars purchased from customers far above the 38%-52% source ratio that Defendants had
publicly reported as their goal. Indeed, by Q4 2020, Carvana’s customer-source ratio had
already jumped to 65%, and had grown by 110% between 2019 and 2020. As the adverse
consequences of Defendants’ scheme mounted, Defendants abruptly stopped reporting these
metrics in order to hide the fraudulent scheme from investors.
294. On May 6, 2021, Defendants announced that they would no longer “provide
detailed statistics on buying cars from customers each quarter” as “[the Company’s] success
over the last 2 years has made the strength of our offering of buying cars from customers
clear.” Accordingly, as explained in a May 7, 2021 analyst report, while Carvana’s stock fell
on this news, investors continued to believe that Defendants’ strategy would contribute to the
Company’s growth and profitability. Upon maintaining its outperform rating, an analyst at
Cowen noted on May 7, 2021: “Cars Purchased from Customers Continues to Pace at ~100%
of Retail Units Sold, Mgmt to Stop Providing Detailed Metrics: Mgmt. noted they had
purchased approximately as many cars from customers as they had sold them . . . . Mgmt.
indicated that they believe their strategy of buying cars from customers had been fully
validated, and that they would stop providing detailed metrics on these ratios in the future.”
Upon revising its price target upward, an analyst at Truist Securities explained on May 7,
2021: “We believe the company has honed its efforts to efficiently acquire vehicles from the
consumer and believe this represents a GPU growth opportunity over time. Mgt noted that
the strength continued into April, as the company set a new record for cars bought from
customers . . . .”
G.
Garcia Junior’s and Jenkins’s SOX Certifications and Signing of
SEC Filings Support Scienter
295. Garcia Junior and Jenkins signed SOX certifications throughout the Class
Period, undertaking the affirmative obligation to ensure the Company’s disclosures to the
market were not materially false or misleading and so represented that they had evaluated the
effectiveness of Carvana’s disclosure controls.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 274 of 332
- 271 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
H.
Garcia Senior’s Control of and Related-Party Deals with
Carvana
296. Because of his felony bank fraud conviction and ban from the NYSE, Garcia
Senior is not identified in Carvana’s SEC filings as a Carvana employee, officer, or director.
Garcia Senior undoubtedly worked behind the scenes and controlled Carvana and its
operations, however, and knew of or recklessly disregarded Defendants’ fraudulent scheme
and course of conduct.
297. As Carvana’s largest shareholder that held approximately 84% of the
Company’s voting power, Carvana’s SEC filings make clear: “The Garcia Parties control
us” and can, among other things, “elect all of the members of our Board and thereby
effectively control our policies and operations, including the appointment of management
[and] future issuances of our Class A common stock or other securities.” 2020 10-K; 2021
10-K; February 23, 2023 annual report on Form 10-K for the fiscal year ended December 31,
2022. Moreover, as reported in a September 17, 2021 The Wall Street Journal article,
according to Professor Taylor, because Carvana utilizes a capital structure that enables
Garcia Junior and Garcia Senior to maintain supermajority voting control largely untethered
from their economic interest in the Company, Carvana’s “‘structure has allowed [the
Garcias] to run this $60 billion public company as if it’s a family firm and for the family’s
benefit.’” Accordingly, Garcia Senior could and did install his son as Carvana’s CEO and
his loyal cronies and former employees to the Company’s Board. Critically, he caused
Carvana to do numerous related party deals with his privately held companies that were
worth hundreds of millions of dollars.
298. As reported by The Wall Street Journal:
The companies themselves are as intertwined as the Garcia men. Carvana
leases properties and sources cars from its onetime parent. Other companies in
the web of family businesses service Carvana’s loans and sell insurance
products to Carvana customers. The deals between the companies generate
hundreds of millions of dollars of revenue for the family business.161
161 Ben Foldy, For Clues About Carvana’s Future, Look to Its Past, Wall St. J. (July 17,
2023).
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 275 of 332
- 272 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
299. Indeed, as detailed below, leading up to and throughout the Class Period,
Garcia Senior engaged in countless related-party deals with Carvana – admittedly “not
negotiated at arm’s length” – demonstrating that he was integral to the fraudulent scheme
and possessed MNPI concerning Carvana. 2020 10-K; 2022 Proxy Statement; March 2023
Form DEF 14-A Proxy Statement.162
IRC Leases: Carvana leases IRCs in
New Jersey, Georgia, Tennessee,
Texas, and Ohio from DriveTime.
Shared Services Agreement: Carvana
pays DriveTime for certain accounting
and tax, legal and compliance,
information technology, and other
services.
IRC Leases: Carvana leased and later
purchased an Arizona IRC from Verde
Investments.
Servicing Agreement: Carvana pays
DriveTime to perform certain
servicing and functions on automotive
finance receivables.
Lease Guarantee: DriveTime
guarantees Carvana’s vending machine
lease in Houston, Texas, its hub in
Atlanta, Georgia, and part of its
headquarters in Tempe, Arizona.
Transfer Agreements: DriveTime
services receivables in Carvana trusts.
Corporate HQ Sublease: Carvana
subleases office space from DriveTime
in Tempe, Arizona.
Credit Facilities: Carvana engages
DriveTime as servicer of receivables.
Office Lease: Carvana leases office
space from Verde Investments in
Tempe, Arizona.
GAP Waiver Insurance Policy:
Carvana purchases insurance policies
from an affiliate of DriveTime.
Hub Lease Agreement: Carvana leases
office space and parking spaces at
various DriveTime IRCs and retail
facilities.
Master Dealer Agreement: DriveTime
administers the VSCs and a portion of
any and GAP waiver coverage sold to
Carvana customers.
Master Purchase and Sale Agreement
(“MPSA”): DriveTime is engaged as
servicer of receivables under
Carvana’s MPSA.
Limited Warranty Agreement:
DriveTime administers the limited
warranties provided to all Carvana
customers.
162 In 2018, when Garcia Senior’s company, SilverRock, “hit snags renewing an Arizona
license in 2018 because it was in negative equity,” it recovered by advising the “department
that its financial position was improving from contracts with Carvana . . . and that it would
soon be profitable.” December 17, 2021 The Wall Street Journal article.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 276 of 332
- 273 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Retail Vehicle Acquisition
Agreements: Carvana purchases
reconditioned vehicles from
DriveTime.
Profit Sharing Arrangement: Carvana
sells customers Road Hazard and Pre-
Paid Maintenance contracts
administered by DriveTime.
Wholesale Vehicle Purchase
Agreements: DriveTime purchases
wholesale vehicles from Carvana.
Insurance Services and Purchase
Agreement: Carvana purchases certain
technology assets and ancillary
services from DriveTime.
Aircraft Time Sharing Agreement:
Carvana leases DriveTime’s private
aircrafts.
300. Through these transactions, Garcia Senior’s companies received approximately
$534 million from Carvana through their deals during the Class Period.163
2021
2022
Lease Agreements
$5.7 million
$4.9 million
Corporate Office Leases
$2 million
$2 million
Retail Vehicle Acquisitions
$168 million
$22 million
Master Dealer Agreement
$15.3 million
$18.1 million
Servicing and Administrative Fees
$81.7 million
$125.8 million
Shared Services Agreement
$1 million
$1 million
Private Plane Sharing Agreement
$0.5 million
$1 million
Total
$274.2 million
$174.8 million
301. Unsurprisingly, one of Carvana’s creditors, which include Apollo Global
Management Inc., Ares Management Corp., and Pacific Investment Management Inc.,
163 In 2020, Garcia Senior’s companies generated $85 million in revenue from providing
extended warranties to Carvana’s customers, collecting on customer loans, and selling or
leasing real estate to Carvana. December 17, 2021 The Wall Street Journal article. Further,
in 2021 and 2022, his companies brought in at least $449 million from its deals with
Carvana. Carvana’s Related Party disclosures in its 2021 10-K, 2021 Proxy Statement; 2022
10-K; 2022 Proxy Statement. And Garcia Senior’s Bridgecrest saw its loan servicing
portfolio more than double to over $10 billion due to its deals with Carvana. December 17,
2021 The Wall Street Journal article. Many of the above transactions are material to
Carvana’s operations and financials. For example, Carvana sold $2.1 billion and $3.8 billion
in principal balances of finance receivables under the Master Purchase and Sale Agreement,
which DriveTime services, during 2021 and 2022, respectively.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 277 of 332
- 274 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
asserted “[a]s long as Ernie Sr. is involved in DriveTime, he’s directly involved in
Carvana.”164
302. Garcia Senior is also Carvana’s largest beneficiary of the Company’s atypical
tax receivable agreement. Under the agreement, Carvana is required to pay 85% of the
benefit of its tax assets – usually generated to reduce future tax bills – to Carvana Group
LLC Unitholders, the largest of which is Garcia Senior. During the Class Period, the portion
of the Company’s tax receivable agreement owed to the Carvana Group LLC Unitholders
was $1.1 billion.
303. In sum, according to Carvana’s SEC filings, the Company’s own creditors, and
academics, Garcia Senior designed Carvana, controls Carvana, “is directly involved in
Carvana,” and runs Carvana to benefit himself and his family, which further supports a
strong inference that he was aware of MNPI throughout the Class Period, including the
scheme to defraud in which he participated.165
I.
Garcia Senior’s History of Similar and Fraudulent Misconduct
Supports a Strong Inference of Scienter
304. Garcia Senior’s prior similar criminal conduct further supports an already
compelling inference of scienter. As explained in the 1990 L.A. Times article, Garcia Senior
pled guilty to felony bank fraud in federal court as he “fraudulently obtained a $30-million
line of credit in a series of transactions that also helped Lincoln [Savings & Loan] hide its
ownership . . . from regulators.” Following his felony fraud conviction, Garcia Senior
purchased Ugly Duckling and took it public on the NASDAQ (Garcia Senior was banned
from the NYSE due to his fraud conviction). As with Carvana, Garcia Senior maintained a
164 As noted in a July 17, 2023 The Wall Street Journal article, Carvana’s creditors “took the
unusual step of signing a pact to cooperate against the family, wary of company efforts to
remove assets from Carvana.”
165 Notably, an internal email between Garcia Senior and Garcia Junior confirm that he ran
Carvana behind the scenes with his son. For example, he coordinated and emailed with
Garcia Junior about the list of potential investors for a 500 million dollar offering, and the
need to “‘figure out [a] plan on [his] money’” prior to the offering. In re Carvana Co.
S’holders Litig., No. 2020-0415-KSM, Transmittal Affidavit of R. Garrett Rice in
Connection with Nominal Defendant Carvana Co.’s Unopposed Motion for Continued
Confidential Treatment, Ex. 2 at 58 (Del. Ch. Mar. 5, 2024) (Dkt. 155).
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 278 of 332
- 275 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
majority ownership of Ugly Duckling when it went public and touted the company’s growth
and innovation, claiming that “while the goal of CarMax and AutoNation is to restructure the
retail auto business, Ugly Duckling is breaking new ground on the financing side.”166 Not
surprisingly, its stock price skyrocketed – like Carvana’s did. But just a few years later,
Ugly Duckling was drowning. The price of Ugly Duckling sank from $25 per share to just
$2.50 per share, and Garcia Senior bought all the publicly owned shares for a pittance.
305. As with Carvana, during Ugly Duckling’s dive, Garcia Senior lined his
pockets; at the time, “Garcia ha[d] been especially adept at feathering his own nest with Ugly
Duckling’s assets” with many investors calling “‘fowl.’”167 Thereafter, Garcia Senior was
compelled to swiftly settle a number of shareholder actions. Garcia Senior’s conviction for
felony bank fraud and involvement in other similar misconduct further supports a strong
inference of scienter.
J.
Corporate Scienter
306. The allegations above also establish a strong inference that Carvana as an
entity acted with corporate scienter throughout the Class Period, as Defendants, its officers,
management, and agents, had actual knowledge of the misrepresentations and omissions of
material facts set forth herein (for which they had a duty to disclose), or acted with reckless
disregard for the truth because they failed to ascertain and to disclose such facts, even though
such facts were available to them. Such material misrepresentations and/or omissions were
made knowingly or recklessly, and without a reasonable basis, for the purpose and effect of
concealing the fraudulent scheme from the investing public. By concealing these material
facts from investors, Carvana maintained and/or increased its artificially inflated common
stock prices throughout the Class Period.
166 Jerry Knight, To Wall Street, It’s a Swan, Wash. Post (Sept. 26, 1997).
167 Nathan Vardi, Feathered Nest, Forbes (Nov. 26, 2001).
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 279 of 332
- 276 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
IX.
LOSS CAUSATION
307. Defendants’ wrongful conduct, as alleged herein, directly and proximately
caused Plaintiffs’ and Class members’ economic loss. Plaintiffs’ claims for securities fraud
are asserted under the fraud-on-the-market and Affiliated Ute Citizens of Utah v. United
States, 406 U.S. 128 (1972) theories of reliance. The markets for Carvana’s stock were
open, well-developed, and efficient at all relevant times. During the Class Period, as detailed
herein, Defendants engaged in a scheme and made materially misleading statements and
omissions. Defendants’ conduct artificially inflated the price of Carvana stock and operated
as a fraud or deceit on the Class.
308. The Class Period inflation in Carvana’s stock price was removed when
information concealed by Defendants’ scheme and misleading statements and omissions was
revealed to the market. The information was disseminated through partial disclosures that
revealed the nature and effect of Defendants’ alleged misconduct. These disclosures, as
more particularly described below, removed artificial inflation from Carvana’s stock price,
causing economic injury to Plaintiffs and other members of the Class.
309. The corrective impact of the partial disclosures during the Class Period alleged
herein, however, was tempered by Defendants’ continued scheme and misleading statements
and omissions that continued to conceal the true nature of Defendants’ fraud. Each partial
disclosure did not on its own fully remove the inflation from Carvana’s stock price, because
it only partially revealed the nature and ramification of Defendants’ previously
misrepresented and concealed conduct. Defendants’ continued scheme and
misrepresentations and omissions maintained the price of Carvana common stock at a level
that was inflated by fraud and induced members of the Class to continue purchasing shares in
Carvana even after Defendants’ partial disclosures.
310. The disclosures that corrected the market price to reduce the inflation
maintained by Defendants’ fraud are detailed below. These stock price declines were due to
firm-specific, fraud-related disclosures and were not the result of market, industry, or firm-
specific, non-fraud factors. The following partial disclosures, resulting stock price declines,
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 280 of 332
- 277 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
and descriptions thereof are not necessarily comprehensive because fact and expert discovery
are not complete.
311. On August 10, 2021, after the market closed, local media in North Carolina
reported that Carvana’s dealer license in Raleigh, North Carolina was suspended for six
months for violating the state’s title and registration laws. ABC 11 detailed the
“suspen[sion],” noting “[d]ocuments show the hearing determined that Carvana failed to
deliver titles to the DMV, sold motor vehicles without a state inspection, and issued out-of-
state temporary tags and plates for vehicles sold to customers in North Carolina.” As a result
of this partial disclosure, the price of Carvana stock declined by $9.40 per share, or 2.5%, to
close at $360.70 per share on August 11, 2021. In contrast to the decline in Carvana stock,
the Standard & Poor’s Composite Stock Index (“S&P 500”) increased by 0.2% during this
period.168
312. This partial disclosure did not on its own fully remove the inflation from
Carvana’s stock price because it only partially revealed the nature and extent of Defendants’
previously misrepresented and concealed conduct. In addition, on August 11, 2021, at the
J.P. Morgan 2021 Auto Conference, analysts questioned Defendants about the media reports
of title and registration violations and the prospect of issues in other states. In response,
Defendants misleadingly reassured investors that the problems were a North Carolina
specific issue and that state action “was a relatively unusual action, but is also pretty small in
scope.” That same day, securities analysts at William Blair noted: “After speaking with
management . . . we are optimistic that this will be an isolated incident.” Thus, the price of
Carvana’s stock remained artificially inflated.
313. On October 22, 2021, the WSJ exposé was published. The article exposed in
part certain of the Company’s recent problems with various state title and registration laws,
revealing that “[a]t least four states have disciplined Carvana or are investigating the
company for violating vehicle-sales rules. . . . These actions haven’t been previously
168 For purposes of comparing its stock price performance vis-à-vis its peers and relevant
market, Carvana referred investors to the S&P 500 Index.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 281 of 332
- 278 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
reported.” Following this news, the price of Carvana stock declined 1.4% on Friday,
October 22, 2021, and declined an additional 1.8% on Monday, October 25, 2021. In
contrast to the total two-trading-day decline of 3.2% in Carvana common stock, the S&P 500
increased by 0.4% during this period.
314. On April 20, 2022, after the market closed, Carvana issued disappointing Q1
2022 financial results and a stock offering to support its purchase of ADESA, which partially
revealed the ramifications and negative economic conditions caused by Defendants’
fraudulent scheme and course of conduct. Among other things alleged herein, Defendants
disclosed that: (i) retail sales growth was hampered that quarter by significant “logistics
network constraints,” “reconditioning” expenses, and “higher wholesale volume from buying
more cars from customers”; (ii) Carvana no longer expected to sell 550,000 retail cars in
2022 as previously stated; and (iii) Carvana needed to purchase ADESA to “improve our
logistics network,” by placing the Company “within 200 miles of 94%” of the population,
which “will have the benefit of reducing shipping distances, times, and costs.” In addition,
while certain aspects of Defendants’ fraud remained concealed, the ramifications and
negative economic conditions caused by their fraud contributed to Carvana’s disappointing
Q1 2022 financial results. For example, as later revealed, Carvana’s Q1 2022 financial
results were negatively impacted by several factors, including: (i) significant logistics
constraints and cost overruns associated with its rapid nationwide expansion and surplus of
wholesale vehicles from buying cars from customers; (ii) “frequently acquir[ing] sales that
were less profitable in the immediate period,” including sales “in markets with lower
profitability due to long distance from inventory (e.g., the Pacific Northwest)”; (iii) a spike
in aged inventory, which negatively impacted GPU and caused Carvana’s average days to
sale metric to spike nearly 37% year-over-year in Q1 2022; (iv) the use of costly third-party
providers due to Carvana’s significant logistics constraints, which caused reconditioning and
transport costs per retail vehicle to spike over $570 year-over-year in Q1 2022; and (v) a
62% year-over-year spike in operations expenses per retail vehicle – a key driver of “unit
economics” – in Q1 2022, and a near 80% spike since the start of the Class Period. In the
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 282 of 332
- 279 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
days following the Q1 2022 results, Carvana disclosed that it was raising additional capital,
including $1.25 billion of stock and $3.275 billion of high-interest rate debt, to fund the
ADESA purchase and for general corporate purposes. As a result of this partial disclosure,
the price of Carvana’s stock declined 10.1% on volume of more than 17 million shares to
close at $83.14 per share on April 21, 2022, and continued to steadily decline, falling another
30.3% over the following six trading days to close at $57.96 per share on April 29, 2022. In
contrast, the S&P 500 declined by 7.4% during this period.
315. Analysts commented on these revelations and tied Carvana’s sudden reset to its
“growth-at-all-costs mentality” that had “undermined business operations.” For example, a
J.P. Morgan analyst labeled the earnings call and follow-ups “‘confidence shattering’” and
added “‘we do not believe Carvana (CVNA) is likely to get a free pass on SG&A any
longer.’” A Deutsche Bank Securities Inc. analyst noted that Carvana “surprised investors
with new details around capital raise” and linked Defendants’ disclosures to the negative
economic conditions caused by Defendants’ unsustainable nationwide expansion. The same
analyst noted: “Q1 results mainly demonstrate that the ongoing constraints throughout the
company’s logistics network is starting to take a material toll on the company’s performance
as it continues to grow its footprint rapidly.” In addition, on April 25, 2022, a Deutsche
Bank securities analyst noted that Carvana’s announcements regarding its urgent need to
raise capital in order to purchase ADESA “surprised investors,” and were related to the fact
that Carvana’s business model was “capital-intensive,” not “capital-light.” Wedbush
analysts stated on May 3, 2022: “Undoubtedly, this is a capital-intensive business (even
without dealerships). . . . ADESA Transaction Adds Strategic Value . . . The company
estimates that sales completed today to customers within 200 miles of the IRC where a
customer’s car is stored will cost $750/unit less in COGS and SG&A than its average sale.
. . . [b]ut [c]omes at a [s]teep [c]ost.” (Emphasis in original.) The next day, Morgan
Stanley analysts noted, upon downgrading Carvana, that “[i]n recent weeks, CVNA found
itself in the [unfortunate] position of raising capital at a time when they really needed it.”
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 283 of 332
- 280 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
316. Defendants’ partial disclosure did not on its own fully remove the inflation
from Carvana’s stock price because it only partially revealed the nature and extent of
Defendants’ previously misrepresented and concealed conduct. In addition, Defendants
falsely reassured investors that Carvana’s issues were “transitory.” On April 21, 2022, BofA
Securities analysts stated: “GPU misses on macro headwind . . . Omicron remained a
headwind in 1Q, pressuring production and logistics capacity.” Thus, the price of Carvana’s
stock remained artificially inflated.
317. On May 10, 2022, Carvana issued an 8-K announcing that it was: (i) laying off
approximately 2,500 employees “primarily in operational groups in connection with its
previously announced plans to better align staffing and expense levels with sales volumes”;
(ii) transitioning operations away from some logistical hubs and an IRC “[i]n connection
with these right-sizing initiatives”; and (iii) that it would release “materials on Carvana’s
[updated] operating plan” later that week. In addition, while certain aspects of Defendants’
fraud remained concealed from investors, the ramifications and negative economic
conditions caused by their fraud contributed to the right-sizing initiatives, the layoffs, and the
need to issue a new operating plan that Carvana disclosed. For example, as later revealed,
Carvana’s financial condition in May 2022 was negatively impacted by several factors,
including: (i) significant logistics constraints and cost overruns associated with its rapid
nationwide expansion and surplus of wholesale vehicles from buying cars from customers
(which was revealed mere days later); (ii) “frequently acquir[ing] sales that were less
profitable in the immediate period,” including sales “in markets with lower profitability due
to long distance from inventory (e.g., the Pacific Northwest)”; (iii) a spike in aged inventory,
which negatively impacted GPU and caused Carvana’s average days to sale metric to spike
76% year-over-year in Q2 2022; (iv) utilizing costly third-party providers due to its
significant logistics constraints, which caused reconditioning and transport costs per retail
vehicle to spike over $585 year-over-year in Q2 2022; and (v) a 62% year-over-year spike in
operations expenses per retail vehicle as of May 2022 and a nearly 80% spike since the start
of the Class Period. As a result of this partial disclosure, the price of Carvana’s stock
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 284 of 332
- 281 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
declined 5.4% on volume of more than 15.7 million shares on May 10, 2022, and declined an
additional 18.2% on volume of nearly 23 million shares on May 11, 2022. In contrast to the
total two-day decline of 22.6% in Carvana common stock, the S&P 500 declined a modest
1.4% during this period.
318. Analysts commented on this corrective disclosure, linking it to the negative
economic conditions caused by Defendants’ scheme. For example, an article published by
Forbes on May 21, 2022 reported: “Carvana’s mass firing was a sign of much bigger
problems at the company, according to . . . several industry analysts. They describe a
spendthrift business, whose growth-at-all-costs mentality undermined business operations
and sowed the seeds of its recent layoffs.” Further, a Wedbush analyst explained on May
11, 2022, that it was “surprised by the workforce reduction,” given Defendants’ contrary
statements in “late April.” Similarly, on May 10, 2022, Morgan Stanley reported on the
Form 8-K, noting, that it was likely the “first of many potential steps to address the cost
structure,” and that “[b]eyond the actions disclosed in today’s 8-K, we believe transparency
around future actions and management’s disclosure and messaging around the path forward
will be critical to re-establishing investor confidence which has been lost over the past few
months.”
319. Defendants’ partial disclosure did not on its own fully remove the inflation
from Carvana’s stock price because it only partially revealed the nature and extent of
Defendants’ previously misrepresented and concealed conduct. In addition, Defendants
falsely reassured investors in a Form 8-K issued May 10, 2022 that the changes were made
in connection with the Company’s “previously announced plans to better align staffing and
expense levels with sales volumes” and the “macroeconomic environment.” Thus, the price
of Carvana’s stock remained inflated.
320. On June 24, 2022, after the market closed, Barron’s published an exposé,
which included interviews with Carvana customers and former employees. The exposé
further revealed the Company’s problems regarding title and registration and its violations of
various state laws and regulations. Indeed, it revealed for the first time that Arizona had
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 285 of 332
- 282 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
canceled Carvana’s temporary plate program, that “[i]nterviews with state officials . . . show
the issue is wide[]-reaching,” and, that Carvana had revamped its title department three
times and created an “‘undriveable-car task force.’” On the following two trading days, the
price of Carvana stock declined 21% to close at $24.74 per share on June 28, 2022, in
contrast to the S&P 500, which declined 2.3% during the same period.
321. This partial disclosure did not on its own fully remove the inflation from
Carvana’s stock price because it only partially revealed the nature and extent of Defendants’
previously misrepresented and concealed conduct. In addition, Defendants falsely reassured
investors by downplaying the Company’s practices and problems with title and registration.
Indeed, in the Barron’s exposé, Defendants claimed only that “‘[i]n a very small percentage
of a very small percentage of instances, customers did not receive permanent license plates
or transferred title within the time frame set forth by the respective states.’” Further, they
misleadingly assured investors that “‘[w]e’ve had productive conversations with regulators
in all of those states and feel very confident about our operations going forward.’” Thus, the
price of Carvana’s stock remained inflated.
322. On October 7, 2022, MDOS announced that it had suspended the license of the
Carvana dealership in Novi, Michigan for “imminent harm to the public.” The suspension
resulted from Carvana’s recent title and licensing violations in Michigan. Following this
news, the price of Carvana stock declined 5.5% on volume of nearly 9.6 million shares to
close at $18.21 per share on October 10, 2022. In contrast to the decline of 5.5% in Carvana
common stock, the S&P 500 declined a modest 0.8% during this period. This partial
disclosure did not on its own fully remove the inflation from Carvana’s stock price because it
only partially revealed the nature and extent of Defendants’ previously misrepresented and
concealed conduct.
323. On November 3, 2022, after the market closed, Carvana issued disappointing
Q3 2022 financial results, which further revealed the unraveling of Defendants’ fraudulent
course of conduct and its negative economic conditions. Among other things alleged herein,
Defendants disclosed that Carvana: (i) suffered an 8% decline in retail unit sales from the
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 286 of 332
- 283 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
prior year; (ii) had been “frequently acquir[ing] sales that were less profitable in the
immediate period but . . . [c]urrently, as we are primarily focused on profitability, we are
intentionally acquiring fewer of these sales”; (iii) had reduced and was continuing to
“reduce[] . . . advertising in distant markets with less profitable sales,” including “markets
with lower profitability due to long distance from inventory (e.g., the Pacific Northwest)”;
(iv) had “take[n] other actions to improve profitability, such as increasing long-distance
shipping fees, in these [new, far-flung] markets”; (v) “reduced . . . inventory by -10%” in Q3
2022, which was one of the “largest drivers” of Carvana’s retail unit sales decline that
quarter; and (vi) would be “continuing to normalize our inventory size and expect to further
reduce inventory in Q4 [2022].” In addition, while certain aspects of Defendants’ fraud
remained concealed from investors, the ramifications and negative economic conditions
caused by their fraud contributed to the disappointing Q3 2022 financial results. For
example, as later revealed, Carvana’s Q3 2022 results were negatively impacted by several
factors, including: (i) a spike in aged inventory, which negatively impacted GPU and caused
Carvana’s average days to sale metric to spike over 60% year-over-year in Q3 2022; (ii)
significant logistics constraints that compelled Carvana to utilize costly third-party providers,
and caused its reconditioning and transport costs per retail vehicle to spike over $400 year-
over-year in Q3 2022; and (iii) an over 40% spike in operations expenses per retail vehicle
since the start of the Class Period.
324. As a result of this corrective disclosure, the price of Carvana stock declined
39% on volume of more than 71 million shares on November 4, 2022, and declined an
additional 15.6% on volume of more than 52 million shares on November 7, 2022 to close at
$7.39 per share. In contrast to the total two day decline of 54.6% in Carvana common stock,
the S&P 500 increased by 2.4% during this period.
325. Analysts and the media reported on Defendants’ November 3, 2022 corrective
disclosure. For example, a Needham & Company analyst noted, “‘[t]he path forward for
Carvana is to sell as many cars as possible, but to do so on a profitable basis, versus prior it
was more about selling as many cars as possible.’” On November 4, 2023, D.A. Davidson
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 287 of 332
- 284 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Institutional Equity Research stated: “CVNA is . . . shift[ing] from just growing sales to
becoming profitable.” The Hit the Wall Article told investors to “[s]tep back and think about
[Carvana’s disclosure] for a moment” because “[i]t means that the company had historically
grown in these markets even though they weren’t all that attractive. It was growth for
growth’s sake and not necessarily growth to improve profitability.”
326. Defendants’ partial disclosure did not on its own fully remove the inflation
from Carvana’s stock price because it only partially revealed the nature and extent of
Defendants’ previously misrepresented and concealed conduct. Thus, the price of Carvana’s
stock remained inflated.
327. On February 23, 2023, after the market closed, Carvana issued disappointing
Q4 2022 financial results that disclosed the unraveling of Defendants’ fraudulent scheme and
course of conduct, including its negative economic conditions. Among other things alleged
herein, Defendants disclosed that: (i) retail unit sales declined 23% year-over-year – a “first
. . . in [Carvana’s] history”; (ii) “it was the first year we stepped back on the key metrics of
retail units sold”; (iii) gross profit declined by 63% year-over-year; (iv) Carvana had
“significantly accelerated . . . reductions in inventory”; (v) Carvana had written off over $50
million of aged and impaired vehicle inventory that was worth less than Carvana had paid to
acquire it and prepare it for sale; (vi) Carvana was “significantly reducing retail vehicle
acquisitions”; and (vii) Carvana was intentionally foregoing retail sales growth for the
foreseeable future to instead focus on driving “positive unit economics” (i.e., retail sales that
were actually profitable). The ramifications and negative economic conditions caused by
Defendants’ fraudulent scheme and course of conduct contributed to the disappointing Q4
2022 financial results. For example, as later revealed, Carvana’s Q4 2022 results were
negatively impacted by several factors, including: (i) a near 40% year-over-year spike in
Carvana’s average days to sale metric; (ii) significant logistics constraints that forced
Carvana to utilize costly third-party providers and caused reconditioning and transport costs
per retail vehicle to remain elevated in Q4 2022, up $377 since Q2 2021; and (iii) a 50%
spike in operations expenses per retail vehicle since the start of the Class Period. As a result
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 288 of 332
- 285 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
of this corrective disclosure, the price of Carvana stock declined 20.5% on volume of more
than 35 million shares on February 24, 2022.169 In contrast, the S&P 500 declined by 1.0%
during this period.
328. Analysts reported on Defendants’ February 23, 2023 corrective disclosure. For
example, a BNP Paribas analyst noted that “[v]irtually every item of Carvana’s metrics
worsened materially in Q4 as units softened, retail [GPU]’s were the lowest since 2016 due
primarily from inventory write-downs and aged inventory.” A J.P. Morgan analyst also
noted that Retail GPU of $425 was now 60% below estimates and “included a ~$598
inventory write-down,” which “reflects aging inventory.” In addition, a William Blair
analyst noted that “Retail GPU was also penalized by aged inventory, with retail cars sold
within 90 days of acquisition netting about $600 in higher GPU versus retail units in
aggregate.” A Morgan Stanley analyst explained that “4Q was a sizeable miss” and noted
Carvana’s “[s]hift in strategy away from growth and towards EBITDA profitability.” A
RBC Capital Markets analyst noted that GPU “miss[ed] Street expectations by 32% . . . due
in large part to retail inventory allowance and wholesale adjustments . . . and lower unit
economics from cars held more than 90+ days.” A Stephens analyst stated: “[W]e do not
foresee the company achieving its targeting unit economics this year.” Further, a Deutsche
Bank analyst observed: “As the company continues to focus on right-sizing its business and
the costs “associated, we believe there is naturally a transition period that may last for a
couple of years before it can refocus on top-line growth.” A D.A. Davidson Institutional
Equity Research analyst noted: “CVNA moved to fix their inventory position this quarter”
after “carrying too much inventory . . . which forced a decrease in Retail GPUs,” and stated
“CVNA is still a long way from EBITDA profitability.” A BofA Securities analyst also
stated: “We now believe that without a cash infusion, Carvana is likely to run out of cash by
the end of 2023.”
169 The closing price of $8.01 per share on February 24, 2023 represented a total decline of
98% from the Class Period high of $376.83 per share in August 2021.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 289 of 332
- 286 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
X.
APPLICABILITY OF THE PRESUMPTION OF RELIANCE AND
THE FRAUD-ON-THE MARKET DOCTRINE
329. Plaintiffs will rely, in part, upon the presumption of reliance established by the
fraud-on-the-market doctrine in that: (i) Defendants made public misrepresentations or failed
to disclose material facts during the Class Period; (ii) the omissions and misrepresentations
were material; (iii) Carvana stock is traded in an efficient market; (iv) the Company’s stock
is liquid and was heavily traded during the Class Period; (v) the Company’s stock was traded
on the NYSE and was covered by multiple analysts; and (vi) Plaintiffs and members of the
Class purchased, acquired, and/or sold Carvana stock between the time Defendants failed to
disclose or misrepresented material facts and the time the true facts were disclosed, without
knowledge of the omitted or misrepresented facts.
330. Based upon the foregoing, Plaintiffs and the members of the Class are entitled
to a presumption of reliance upon the integrity of the market.
331. Plaintiffs and the members of the Class are entitled to the presumption of
reliance established by the Supreme Court in Affiliated Ute Citizens of Utah v. United States,
406 U.S. 128 (1972), as Defendants omitted material information in their Class Period
statements in violation of a duty to disclose such information, as detailed above.
XI.
NO SAFE HARBOR
332. Many (if not all) of Defendants’ false and misleading statements and omissions
during the Class Period were not forward-looking statements (“FLS”) and/or identified as
such by Defendants, and thus did not fall within any “Safe Harbor.”
333. Defendants’ verbal “Safe Harbor” warnings accompanying its oral FLS issued
during the Class Period were ineffective to shield those statements from liability.
334. Defendants are also liable for any false or misleading FLS pleaded because, at
the time each FLS was made, the speaker knew the FLS was false or misleading and the FLS
was authorized and/or approved by an executive officer of Carvana who knew that the FLS
was false or misleading. Further, none of the historic or present tense statements made by
Defendants were assumptions underlying or relating to any plan, projection, or statement of
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 290 of 332
- 287 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
future economic performance, as they were not stated to be such assumptions underlying or
relating to any projection or statement of future economic performance when made.
XII. CLASS ACTION ALLEGATIONS
335. Plaintiffs bring this action as a class action pursuant to Rule 23 of the Federal
Rules of Civil Procedure on behalf of a Class, consisting of all those who purchased or
otherwise acquired Carvana stock during the Class Period (the “Class”). Excluded from the
Class are Defendants and their families, the officers and directors of the Company, at all
relevant times, members of their immediate families and their legal representatives, heirs,
successors or assigns, and any entity in which Defendants have or had a controlling interest.
336. The members of the Class are so numerous that joinder of all members is
impracticable. Throughout the Class Period, Carvana stock was actively traded on the
NYSE. While the exact number of Class members is unknown to Plaintiffs at this time and
can only be ascertained through appropriate discovery, Plaintiffs believe that there are
hundreds or thousands of members in the proposed Class. Record owners and other
members of the Class may be identified from records maintained by Carvana or its transfer
agents and may be notified of the pendency of this action by mail, using the form of notice
similar to that customarily used in securities class actions.
337. Common questions of law and fact predominate and include: (i) whether
Defendants’ acts violated the federal securities laws; (ii) whether Defendants engaged in a
scheme to defraud; (iii) whether Jenkins and Garcia Junior omitted and/or misrepresented
material facts; (iv) whether Defendants acted knowingly or recklessly disregarded the truth;
(v) whether the prices of Carvana stock during the Class Period were artificially inflated
because of Defendants’ conduct complained of herein; and (vi) whether the members of the
Class have sustained damages and, if so, what is the proper measure of damages.
338. Plaintiffs’ claims are typical of the claims of the members of the Class as all
members of the Class are similarly affected by Defendants’ wrongful conduct in violation of
federal law that is complained of herein.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 291 of 332
- 288 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
339. Plaintiffs will fairly and adequately protect the interests of the members of the
Class and have retained counsel competent and experienced in class and securities litigation.
340. A class action is superior to all other available methods for the fair and
efficient adjudication of this controversy since joinder of all members is impracticable.
Furthermore, as the damages suffered by individual Class members may be relatively small
per Class member, the expense and burden of individual litigation make it impossible for
members of the Class to individually redress the wrongs done to them. There will be no
difficulty in the management of this action as a class action.
XIII. CLAIMS FOR RELIEF UNDER THE EXCHANGE ACT
COUNT I
Violations of §10(b) of the Exchange Act and Rule
10b-5 Promulgated Thereunder
(Against the Exchange Act Defendants)
341. Plaintiffs repeat and allege each and every allegation set forth above as if fully
set forth herein.
342. This Count is based upon §10(b) of the Exchange Act, 15 U.S.C. §78j(b), and
SEC Rule 10b-5 promulgated thereunder.
343. During the Class Period, Defendants engaged in a scheme and wrongful course
of conduct pursuant to which they knowingly or recklessly engaged in acts, transactions,
practices, and a course of business which operated as a fraud and deceit upon Plaintiffs and
the other members of the Class, made various untrue statements of material fact and omitted
to state material facts necessary in order to make the statements made, in light of the
circumstances under which they were made, not misleading, and employed a scheme to
defraud in connection with the purchase and sale of Carvana stock. Defendants’ scheme was
intended to, and, throughout the Class Period, did: (i) deceive the investing public, including
Plaintiffs and other Class members, as alleged herein; (ii) artificially inflate and maintain the
market price of Carvana stock; and (iii) cause Plaintiffs and other members of the Class to
purchase or otherwise acquire Carvana stock at artificially inflated prices. In furtherance of
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 292 of 332
- 289 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
this unlawful scheme and course of conduct, Defendants, and each of them, took the actions
set forth herein.
344. Pursuant to the above wrongful course of conduct, each of the Defendants
participated directly or indirectly in: (i) Defendants’ scheme; and/or (ii) the preparation
and/or issuance of the quarterly and annual reports, SEC filings, press releases, and other
statements and documents described above, including statements made to securities analysts
and the media that were designed to influence the market for Carvana stock. Such reports,
filings, releases, and statements were materially false and misleading in that they failed to
disclose material adverse information and misrepresented the truth about Carvana’s finances
and business prospects.
345. By virtue of their ownership and/or positions at Carvana, Defendants had
actual knowledge of the materially false and misleading statements and material omissions
alleged herein and intended thereby to deceive Plaintiffs and the other members of the Class,
or, in the alternative, Defendants acted with reckless disregard for the truth in that they failed
or refused to ascertain and disclose such facts as would reveal the false and misleading
nature of the statements made, although such facts were readily available to Defendants.
Said acts and omissions were committed willfully or with reckless disregard for the truth.
346. Information showing that Defendants acted knowingly or with reckless
disregard for the truth is peculiarly within Defendants’ knowledge and control. As the
controlling shareholder, senior managers, and/or directors of Carvana, Defendants had
knowledge of the details of Carvana’s internal affairs.
347. The Defendants are directly and indirectly liable for the wrongs complained of
herein. Because of their positions of control and authority, Defendants were able to and did,
directly or indirectly, control the content of the statements of Carvana. As controlling
shareholders, officers and/or directors of a publicly held company, Defendants had a duty to
disseminate timely, accurate, and truthful information with respect to Carvana’s businesses,
operations, future financial condition, and future prospects. As a result of Defendants’
misconduct, the market price of Carvana’s stock was artificially inflated throughout the
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 293 of 332
- 290 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Class Period. In ignorance of the adverse facts concerning Carvana’s business and financial
condition which were concealed by the Individual Defendants, Plaintiffs and the other
members of the Class purchased or otherwise acquired Carvana stock at artificially inflated
prices and in doing so relied upon the price of the securities, the integrity of the market for
the securities, and/or upon statements disseminated by the Individual Defendants, and were
damaged thereby.
348. During the Class Period, Carvana stock was traded on an active and efficient
market. Plaintiffs and the other members of the Class, relying on the materially false and
misleading statements described herein, which Defendants made, issued, or caused to be
disseminated, or relying upon the integrity of the market, purchased or otherwise acquired
Carvana stock at prices artificially inflated by Defendants’ wrongful conduct. Had Plaintiffs
and the other members of the Class known the truth, they would not have purchased or
otherwise acquired said stock, or would not have purchased or otherwise acquired it at the
inflated prices paid. At the time of the purchases and/or acquisitions by Plaintiffs and the
Class, the true value of Carvana stock was substantially lower than the prices paid by
members of the Class. The market price of Carvana stock declined upon public disclosure of
the facts alleged herein to the injury of Plaintiffs and Class members.
349. By reason of the conduct alleged herein, Defendants knowingly or recklessly,
directly or indirectly, have violated §10(b) of the Exchange Act and Rule 10b-5 promulgated
thereunder.
350. As a direct and proximate result of Defendants’ wrongful conduct, Plaintiffs
and the other members of the Class suffered damages in connection with their respective
purchases, acquisitions, and/or sales of Carvana stock during the Class Period, as the truth
about Carvana’s operations and prospects began to be disclosed to the investing public.
COUNT II
Violations of §20(a) of the Exchange Act
(Against the Exchange Act Defendants)
351. Plaintiffs repeat and reallege each and every allegation set forth above as if
fully set forth herein.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 294 of 332
- 291 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
352. During the Class Period, the Individual Defendants participated in and oversaw
the operation and management of Carvana, and conducted and participated, directly and
indirectly, in the conduct of Carvana’s business affairs.
353. During all relevant times during the Class Period, Garcia Senior owned a
majority of the voting power of all outstanding shares of the Company’s common stock,
effectively giving him full control over the Company’s most important decisions, such that
Carvana was a “controlled company” within the meaning of the corporate governance
standards of the NYSE.
354. As officers and/or directors of a publicly owned company, the Individual
Defendants had a duty to disseminate accurate and truthful information with respect to
Carvana’s financial condition and results of operations, and to correct promptly any public
statements issued by Carvana which had become materially false or misleading.
355. Each of the Individual Defendants, therefore, acted as a controlling person of
Carvana. By reason of their senior management positions, and/or being directors, and/or the
controlling shareholder of Carvana, each of the Individual Defendants had the power to
direct the actions of, and exercised the same to cause, Carvana to engage in the unlawful acts
and conduct complained of herein. Each of the Individual Defendants exercised control over
Carvana’s operations and possessed the power to control the specific activities which
comprise the primary violations about which Plaintiffs and the other members of the Class
complain. In addition, Garcia Junior and Garcia Senior owned a majority of Carvana’s Class
B shares, which shares guarantee Garcia Junior and Garcia Senior ten votes per share for so
long as the Garcia Parties maintain direct or indirect beneficial ownership of at least 25% of
the outstanding shares of stock. Throughout the Class Period, the Individual Defendants
exercised their power and authority to cause Carvana to engage in the wrongful acts
complained of herein. The Individual Defendants, therefore, were “controlling persons” of
Carvana within the meaning of §20(a) of the Exchange Act.
356. Carvana had the power to control and influence the Individual Defendants, and
other Company executives through its power to hire, fire, supervise, and otherwise control
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 295 of 332
- 292 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
the actions of its employees and their salaries, bonuses, incentive compensation, and other
employment considerations. By virtue of the foregoing, Carvana had the power to influence
and control, and did influence and control, directly or indirectly, the decision-making of the
Individual Defendants, including the content of their public statements.
357. By reason of the above conduct, the Individual Defendants are liable pursuant
to §20(a) of the Exchange Act for the violations committed by Carvana.
COUNT III
Violations of §20(A) of the Exchange Act
(Against Garcia Senior and Jenkins)
358. Plaintiffs repeat and reallege each and every allegation set forth above as if
fully set forth herein.
359. Count III is brought pursuant to §20(A) of the Exchange Act against Garcia
Senior and Jenkins on behalf of Plaintiffs and members of the Class who were damaged by
each of these Defendant’s insider trading.
360. During the Class Period, Garcia Senior sold nearly $3.7 billion of his Carvana
stock to unknowing investors. Further, as detailed above, Garcia Senor possessed MNPI
concerning Carvana when he sold his stock, and he took advantage of his possession of
MNPI regarding Carvana to obtain billions of dollars of insider trading profits during the
Class Period.
361. During the Class Period, Jenkins sold over $79 million of his stock to
unknowing investors. As detailed above, Jenkins was in possession of MNPI concerning
Carvana when he sold his stock, and he took advantage of his possession of MNPI regarding
Carvana to obtain tens of millions of dollars in insider trading profits during the Class
Period.
362. Defendants’ sales of Carvana stock during the Class Period (set forth in
Appendix A attached hereto) were made contemporaneously with Plaintiffs’ purchases of
Carvana stock (set forth in Appendix B and in Plaintiffs’ certifications (attached hereto)
during the Class Period.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 296 of 332
- 293 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
363. As Carvana’s controlling shareholder, Garcia Senior owed a duty to Carvana
and its shareholders to maintain the MNPI in confidence and not trade on the basis of it.
Jenkins, as Carvana’s CFO, also owed a duty to Carvana and its shareholders to maintain the
MNPI in confidence and not trade on the basis of it.
364. Plaintiffs and members of the Class who purchased shares of Carvana stock
contemporaneously with sales by Garcia Senior and Jenkins suffered damages because: (i) in
reliance on the integrity of the market, they paid artificially inflated prices as a result of the
violations of §§10(b) and 20(a) of the Exchange Act as alleged herein; and (ii) they would
not have purchased Carvana stock at the prices they paid, or at all, if they had been aware
that the market prices had been artificially inflated by the false and misleading statements
and concealment alleged herein.
XIV. SECURITIES ACT ALLEGATIONS
365. In this section of the complaint, Plaintiffs assert a series of strict liability and
negligence claims on behalf of purchasers in the 2022 Public Offering for violations of the
Securities Act (the “Securities Act Class”). Plaintiffs expressly disclaim any allegations of
knowing or reckless misconduct or fraud.
A.
Plaintiffs’ Purchases in the 2022 Public Offering
366. On April 22, 2022, Plaintiffs purchased shares of Class A common stock
directly in the 2022 Public Offering from underwriter defendant Citigroup for $80.00 per
share.
B.
Securities Act Defendants
367. Defendant Carvana was the issuer of the 2022 Public Offering.
368. Defendants Ernest Garcia III (“Garcia Junior”) and Mark Jenkins (“Jenkins”)
signed the Registration Statement (defined below).
369. Defendant Stephen Palmer (“Palmer”) served as Carvana’s Vice President of
Accounting and Finance and signed the Registration Statement.
370. Defendants Michael Maroone (“Maroone”), Neha Parikh (“Parikh”), Ira Platt
(“Platt”), and Greg Sullivan (“Sullivan”) each served as members of Carvana’s Board and
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 297 of 332
- 294 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
signed the Registration Statement. Defendants Maroone, Parikh, Platt, and Sullivan are
collectively referred to herein as the “Director Defendants” and, together with defendants
Garcia Junior, Jenkins, and Palmer, as the “Individual Securities Act Defendants.”
371. Defendants Citigroup and J.P. Morgan Securities LLC (“J.P. Morgan”) acted
as underwriters of, and as sellers in, Carvana’s 2022 Public Offering (collectively, the
“Underwriter Defendants”). Each of the Underwriter Defendants served as underwriters
and/or underwriter representatives for the 2022 Public Offering.
372. Defendants Carvana, the Individual Securities Act Defendants, and the
Underwriter Defendants are collectively referred to herein as the “Securities Act
Defendants.”
C.
Background of the 2022 Public Offering
373. On April 20, 2022, Carvana filed SEC Form S-3 Registration Statement (the
“Registration Statement”), an automatically effective shelf registration statement pursuant to
which Carvana could offer to sell shares of Class A common stock on terms to be determined
at the time of each offering, pursuant to a prospectus supplement to be filed with the SEC in
connection with each offering. Five days later, on April 25, 2022, Carvana filed SEC Form
424B2 (the “Prospectus Supplement”) offering to sell 15,625,000 shares of Class A common
stock at a price of $80.00 per share. The Registration Statement, Prospectus Supplement,
and all documents incorporated therein are collectively referred to as the “Registration
Statement” or the “Offering Documents.”
374. The Offering Documents incorporated a number of previously filed documents
by reference, including the 2021 10-K.
375. On April 26, 2022, Carvana announced the completion of the 2022 Public
Offering, which raised proceeds of $1,225,460,000.
376. Prior to the 2022 Public Offering, and unbeknownst to investors, Carvana
faced numerous challenges that rendered the Company’s growth unsustainable, including
worsening logistical constraints in acquiring cars, an increasing supply of low-quality
vehicles, and widespread title and registration issues. Negligently failing to disclose these
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 298 of 332
- 295 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
problems, the Registration Statement made a number of false and misleading statements
about: (i) Carvana’s retail units sold; (ii) the quality of Carvana’s vehicle inventory; and (iii)
the purported risks surrounding regulatory compliance.
377. Carvana’s rapid market expansion created logistical constraints and additional
costs to support new markets at greater distances, which created mounting expenses. As
Carvana later disclosed in the Company’s May 2022 Operating Plan, following the 2022
Public Offering, Carvana’s rapid “[l]ocation growth” created “[m]ore vehicle moves, more
miles traveled . . . [h]igher number of constrained routes,” and a “[h]igher degree of backlog
on constrained routes.” Carvana’s logistical constraints were increasingly costly as the
Company was forced to rely on expensive third-party providers for vital logistics and
reconditioning functions as the Company’s own critical infrastructure lagged behind its
nationwide expansion.
378. Carvana also had amassed a substantial inventory of vehicles purchased from
consumers, without regard to the vehicle’s quality. The impact of this was two-fold. First,
the increasing inventory of vehicles from consumers worsened Carvana’s logistical
constraints as cars had to be obtained from these individuals and then transported and stored
at Carvana IRCs – creating more vehicle moves, more miles traveled, constrained routes, a
higher backlog on constrained routes, poor performance on operational metrics, and the
inefficient use of Carvana’s nationwide logistics network. Second, there was an
overwhelming increase in low-quality vehicles that could only be sold at wholesale, which
was often unprofitable when taking into account the additional expenses of obtaining and
processing the vehicles. This was especially true in light of the fact that Carvana frequently
overpaid for low-quality vehicles, as consumers selling their vehicles as trade-ins would
misrepresent the vehicle quality, with no reduction in trade-in value or other consequence
from Carvana.
379. Finally, the massive influx of vehicles purchased from consumers exacerbated
widespread vehicle titling issues at Carvana, and ultimately resulted in countless vehicles
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 299 of 332
- 296 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
being sold to customers before title was procured and transferred, often in violation of state
law.
D.
The Registration Statement Contained Materially False and
Misleading Statements and Omitted Material Facts Required to
be Stated Therein
380. The Registration Statement was negligently prepared and, as a result: (i)
contained untrue statements of material fact; (ii) omitted to disclose material information
necessary to make statements made therein not misleading; and (iii) was not prepared in
accordance with SEC rules and regulations governing its preparation.
381. Statement A: With regard to vehicle sourcing and acquisition, the 2021 10-K
stated:
Vehicle acquisition. . . . For vehicles sold to us through our website,
we use proprietary algorithms to determine an appropriate offer. We assess
vehicles on the basis of quality, inventory fit, consumer desirability, relative
value, expected reconditioning costs, and vehicle location to identify what we
believe represent the most in-demand and profitable vehicles to acquire for
inventory. We utilize a broad range of data sources, including proprietary site
data, and a variety of external data sources to support our assessments.
382. As detailed below, Statement A was materially false and misleading when
made because it negligently failed to disclose the following material facts necessary to be
disclosed in order to make the Registration Statement not misleading:
(a)
Carvana abruptly and drastically lowered the Company’s purchasing
and verification standards when buying cars from customers in order to induce trade-ins and
increase retail sales. CW-4 confirmed that, to boost retail sales through trade-ins, Carvana
bought cars from customers that plainly did not meet its purchasing and verification
standards. CW-11 said that Carvana was buying vehicles “sight unseen” and his/her team
was not expected to do much to verify the condition of the vehicles. CW-11 observed that
Carvana did not actually care about the condition of the vehicles and just wanted to “get as
many cars as it could.” CW-10 corroborated this account, explaining that Carvana was
buying cars at breakneck speed and not adequately inspecting them. In addition, CW-2 was
overruled when voicing concerns about the quality of vehicles as “leadership would say take
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 300 of 332
- 297 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
it.” Similarly, CW-5 stated that there was only a 50/50 chance that a vehicle would be
reappraised if it did not match the seller’s description. In short, contrary to the statement that
Carvana was “assess[ing] vehicles on the basis of quality, inventory fit, consumer
desirability, relative value, expected reconditioning costs, and vehicle location to identify
what we believe represent the most in-demand and profitable vehicles to acquire for
inventory,” Carvana largely disregarded the condition of the vehicles they purchased as
CWs-1, 2, 3, 4, 5, 10, and 11 reported.
(b)
Carvana later admitted that buying cars from customers adversely
impacted the Company’s bottom line by flooding Carvana with less profitable wholesale cars
that had to be sold at a loss and crippling its logistics network, as detailed below.
(i)
Because Carvana was “intentionally buying lower quality
cars,”170 it was flooded with cars that did not meet its retail standards and, thus, had to be
sold wholesale. Indeed, by lowering its purchasing and verification standards, Carvana’s
wholesale growth began to far outpace its retail sales growth. As shown in the charts below,
immediately following Carvana’s spike in the number of cars purchased from customers
during late 2020 and early 2021, its wholesale sales spiked by over 100% in each of the next
five quarters, including 115% in Q4 2021. Meanwhile, wholesale sales made up roughly
30% of the Company’s total sales in Q4 2021, nearly double the ratio at the beginning of the
Class Period.
170 Defendants judicially admitted this omitted fact by including it in a brief in this action.
ECF 58 at 12.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 301 of 332
- 298 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
(ii)
This massive increase in wholesale vehicles adversely impacted
Carvana’s bottom line because, unbeknownst to investors, Carvana lost money on wholesale
vehicle sales, as detailed below:
(1)
Carvana sold cars at wholesale at significantly lower
prices than those sold at retail. Carvana, however, still incurred many of the same costs
involved in selling retail vehicles, including inbound transportation, reconditioning, and
other logistics. Thus, the lower average sales price combined with customary selling costs
resulted in significantly lower reported gross profit per unit on wholesale cars. In Q4 2021,
wholesale GPU was $1,396 while Retail GPU was $1,495.171
(2)
Although Carvana disclosed that it generated a modest
gross profit on wholesale vehicle sales, its actual profitability – or lack thereof – of
wholesale sales was not adequately disclosed because it: (i) excluded certain per-vehicle
operations expenses from its calculation of wholesale GPU; and (ii) did not disclose or
quantify these excluded costs separately so investors could decipher overall profitability of
wholesale sales on their own. These per-vehicle operations expenses, which were only
separately broken out and quantified for the first time after the Class Period, included
171 Carvana also did not earn any ancillary revenue or profit on wholesale vehicle sales, such
as warranty, insurance, or financing-related revenue. These “other” revenue streams
represented the vast majority of Carvana’s reported total GPU on its retail sales, but
generated $0 on wholesale sales.
0%
100%
200%
300%
400%
500%
Q1 20Q2 20Q3 20Q4 20Q1 21Q2 21Q3 21Q4 21Q1 22
Wholesale Growth Exceeded Retail Growth
y/y growth (retail cars sold)
y/y growth (wholesale cars sold)
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 302 of 332
- 299 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
material costs associated with completing a wholesale sale, such as the full cost to ship a car
to a wholesale auction site and title and registration costs.172 Tellingly, Carvana admitted
after the Class Period in its Q3 2023 Shareholder Letter and November 2, 2023 Cost
Structure Details presentation that there are “two key drivers” of “unit economics” (i.e., per-
vehicle profitability): (i) cost of sales, which were the expenses Carvana had always included
in its reported GPU calculations; and (ii) “operations expenses” per vehicle sold, which
consisted of per-vehicle selling expenses that were not included in Carvana’s reported GPU
and had never been separately broken out to investors.173 Instead, during the Class Period,
these expenses that drove Carvana’s actual profit per car sold, were lumped into and
comingled with other SG&A expenses.174 Carvana later admitted, however, that these
“operations expenses” were not typical fixed or corporate overhead SG&A costs, conceding:
“Operations expenses include the fulfillment, customer service, and transaction expenses
associated with completing retail and wholesale vehicle sales.”175 Indeed, these vehicle
172 In its November 2, 2023 Cost Structure Details presentation, Carvana disclosed and
quantified retail and wholesale vehicle operations expense.
173 After disclosing and quantifying operations expenses for the first time in November
2023, Jenkins acknowledged that the additional disclosure “allows a much deeper dive into
the economics of Carvana.” Jenkins added: “we’ve really been focused on . . . improving the
unit economics of the business. In other words, driving down variable costs per car,” which
he confirmed included “operations expenses, which are the more variable component of our
selling, general and administrative expenses.” Jenkins continued: “And the reason that we’re
focused on that is obviously, the more gains we make on reducing variable cost per car
today, the more profitable our growth can be in the future . . . .” Jenkins also described
“driving down operational expenses per unit” as one of Carvana’s “profitability initiatives,
and in particular, . . . our unit economics initiatives.” Thus, according to Jenkins, operations
expenses were a critical component of assessing Carvana’s profitability.
174 While Carvana included per-vehicle operations expenses in SG&A expenses (rather than
GPU), it intentionally did not separately disclose or quantify these expenses in a manner that
would have allowed investors to decipher the overall per-vehicle profitability on their own.
In fact, as detailed in ¶260(e), the disclosures Carvana did make regarding SG&A expenses
during Q4 2021 did more to mislead investors than to provide transparency around
Carvana’s true profitability per vehicle sold.
175 At the time of the 2022 Public Offering, Defendants described Carvana’s SG&A costs as
“expenses associated with advertising and providing customer service to customers,
operating our vending machines and hubs, operating our logistics and fulfillment network
and other corporate overhead expenses, including expenses associated with information
technology, product development, engineering, legal, accounting, finance, and business
development.” Moreover, while Defendants made vague Class Period disclosures indicating
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 303 of 332
- 300 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
selling expenses that Carvana had arbitrarily excluded from its reported GPU included: (i)
“customer care, multi-car logistics, last-mile delivery, and other operations payroll”
“associated with completing . . . wholesale vehicle sales”; (ii) “non-payroll logistics
expenses, including fuel, repairs and maintenance, and third-party transport services”
“associated with completing . . . wholesale vehicle sales”; and (iii) “[t]ransaction and other
expenses, including limited warranty, title and registration, and finance platform expenses”
“associated with completing . . . wholesale vehicle sales.” When describing Carvana’s
wholesale profitability, the Registration Statement should have disclosed “operations
expenses” because, internally at Carvana, “the underlying costs of completing a sale” were
viewed as a critical component of Carvana’s actual wholesale vehicle profitability. Indeed,
as Garcia Junior later acknowledged:
When we think about trying to aim for more profitable sales, what does that
mean? There’s certainly variability in the kind of gross profit associated with
different types of sales. . . .
And then I think there’s a number of other dynamics kind of across car
type, et cetera. There’s also kind of variation in the underlying costs of
completing a sale.
Because the Registration Statement did not specifically break out operations expenses (i.e.,
the underlying costs of completing a sale) and instead buried and comingled them among
SG&A expenses, investors could not calculate the actual overall profitability of Carvana’s
wholesale vehicle sales, like Garcia Junior did internally. As demonstrated in the chart
below, upon incorporating all of the estimated vehicle operations expenses (i.e., the actual
costs Carvana incurred in completing wholesale vehicle sales), Carvana lost money on
wholesale vehicle sales every quarter during the Class Period.176 For example, in Q4 2021,
that certain of expenses were recorded in SG&A, until the end of the Class Period Carvana
never: (i) broke out “operations expenses” to investors; (ii) described that these vehicle
“operations expenses” were direct per-vehicle costs “associated with completing . . .
wholesale vehicle sales”; nor (iii) described that, in addition to the selling costs included in
Carvana’s GPU metric, these additional per-vehicle “operations expenses,” were one of “two
key drivers” of per-unit profitability.
176 Estimated operations expense per wholesale vehicle is calculated as follows: total retail
and wholesale vehicle operations expenses in Q4 2021 divided by total retail and wholesale
vehicles sold in Q4 2021.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 304 of 332
- 301 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Carvana’s average selling price per wholesale vehicle was $12,211 while its total costs (cost
of sales in addition to estimated undisclosed operations expense) totaled $12,830, for a total
loss of $619 per vehicle.
(iii)
The Registration Statement negligently omitted that the surplus
of wholesale cars resulting from Carvana “intentionally buying lower quality cars” also led
to significant logistics constraints. For example, in its May 2022 Operating Plan, Carvana
admitted that “[b]uying cars from customers” led to “additional wholesale volume” which, in
turn, led to “constraints in our nationwide logistics network.” Carvana also admitted that
“wholesale units acquired from customers have typically been transported to the nearest
Carvana IRC, generating additional vehicles [sic] moves and increased complexity in our
multi-car logistics network.”
(iv)
The Registration Statement negligently omitted that Carvana
lacked critical infrastructure necessary to process the huge surplus of wholesale vehicles,
which is what led to Carvana’s acquisition of ADESA, a nationwide wholesale auction
house. In describing his rationale for the purchase, Garcia Junior pointed to ADESA’s 56
brick-and-mortar locations, consisting of 6.5 million square feet of buildings on more than
4,000 acres of land, and as reported in a September 13, 2022 Vehicle Remarketing article,
explained “‘[y]ou need a place to stage the logistics, and there’s no better infrastructure than
auctions to serve that purpose.’” Indeed, Carvana later revealed that, prior to acquiring
ADESA, Carvana “frequently ha[d] our last mile delivery advocates pick these cars up from
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 305 of 332
- 302 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
customer’s homes and transport them to a market hub and then transport these cars long
distances from their origin to our nearest IRC where they are either reconditioned or held
until they are wholesaled.” Further, Carvana admitted that “[w]ith this acquisition [of
ADESA], we can now position advocates at the auction locations and transport cars from our
customer’s driveways directly to an ADESA U.S. location without any additional moves.”
383. Statement B: In the 2021 10-K, Carvana reported:
[A]n increase in the number of used vehicles sold to 425,237 from 244,111
during the years ended December 31, 2021 and 2020, respectively. The
increase in units sold was driven by growth in existing markets due to
expanded inventory selection, enhanced marketing efforts, increased brand
awareness, customer referrals, and growth in market population coverage to
81.0% as of December 31, 2021 from 73.7% as of December 31, 2020.
384. As detailed below, Statement B was materially false and misleading when
made because it negligently failed to disclose the following material facts necessary to be
disclosed in order to make the Registration Statement not misleading:
(a)
Although the Registration Statement described the drivers of retail unit
growth, it did not disclose that this growth was primarily driven by a number of
unsustainable practices. Specifically, and discussed in more detail below, the stated retail
unit sales growth was primarily fueled by: (i) sales to customers awaiting proper title and
registration processing; (ii) “less profitable sales” in “markets with lower profitability due to
long distance from inventory”; (iii) trade-in sales resulting from Carvana’s lowered
purchasing and verification standards; (iv) sales, pursuant to a pass-through sales agreement
with Garcia Senior’s DriveTime, that lacked economic substance for Carvana; and (iv) “sales
that were less profitable in the immediate period.” For example:
(i)
Statement B omitted that, in violation of state laws, Carvana
routinely sold cars to customers before it held title to those cars and faster than it could get
them registered to their new owners. For example, CWs-3, 4, 8, 9, 10, and 11 universally
reported that Carvana would sell these vehicles before they received title. Specifically, CW-
10 described instances where titles on vehicles Carvana sold could not be obtained for over a
year, which meant the buyers were unable to drive their cars. CW-8 said that Carvana’s
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 306 of 332
- 303 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
practice was simply to assume that titles for vehicles could be acquired later if they were not
available at the time of purchase. CW-4 said Carvana had a Google spreadsheet that tracked
vehicles without titles for his/her entire IRC and “there were a ridiculous amount of cars on
the list.” CW-9 said, beginning in Q1 2022, registration delays amounted to roughly half of
the calls the team had handled. The CWs’ accounts are corroborated by state investigations,
findings, and suspensions of Carvana’s dealership licenses in North Carolina, Michigan,
Illinois, Arizona, Pennsylvania, Florida, Texas, Maryland, Georgia, and Ohio. In fact: (i) the
Ohio BMV had investigated Carvana, suspended its temporary tag issuance privileges for all
Ohio locations, and subjected the Company to increased oversight in December 2020; (ii)
following additional meetings between MDOS and Carvana, Carvana “admi[tted] . . . several
more violations of [Michigan’s] Code,” faced thousands of dollars in additional fines, and
had its 18-month probation extended on February 7, 2022; and (iii) Illinois had begun
investigating Carvana in February 2022 for issuing out-of-state temporary registration
permits and for failing to transfer titles in a timely manner in violation of its laws. These
omitted facts were material as Defendants admitted that Carvana’s retail sales growth was
significantly hindered by Defendants’ belated enactment of “buffer” periods near the end of
the Class Period to allow for sufficient time to secure title.
(ii)
Statement B omitted that Carvana’s retail sales growth was
largely driven by less profitable sales in far-flung markets in connection with Carvana’s
nationwide expansion. In fact, in FY 2021, 100% of the new markets that Carvana entered
were over 200 miles from an existing IRC, which cost Carvana an additional $750 more per
retail car sold. These costs were material. Indeed, near the end of the Class Period, Carvana
was forced to reduce sales in these markets. As one misled commentator realized near the
end of the Class Period, “[Carvana’s] was growth for growth’s sake and not necessarily
growth to improve profitability.”
(iii)
Statement B omitted that Carvana’s retail sales growth was
supported by the lowering of the Company’s purchasing and verification standards when
buying cars from customers. Indeed, CW-4 confirmed that, to boost retail sales through
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 307 of 332
- 304 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
trade-ins, Carvana bought cars from customers that did not meet its purchasing and
verification standards. CW-11 said that Carvana was buying vehicles “sight unseen” and
his/her team was not expected to do much to verify the condition of the vehicles. CW-11
observed that Carvana did not actually care about the condition of the vehicles and just
wanted to “get as many cars as it could.” CW-10 corroborated this account, explaining that
Carvana was buying cars at breakneck speed and not adequately inspecting them. In
addition, CW-2 was overruled when voicing concerns about the quality of vehicles as
“leadership would say take it.” Similarly, CW-5 stated that there was only a 50/50 chance
that a vehicle would be reappraised if it did not match the seller’s description. This omitted
fact was material as Carvana was forced to meter (i.e., not sell) some of its purchases of cars
from customers because of the strain it placed on its logistics network, which would hinder
Carvana’s retail growth.
(iv)
Statement B omitted that Carvana “frequently acquired sales that
were less profitable in the immediate period.” Carvana’s internal sales’ calculation
“incorporated the value of future sales” as Carvana singularly focused on growth without
regard to profitability. These sales were material; near the end of the Class Period, Carvana
would be forced to forego these sales to survive, which adversely impacted Carvana’s retail
sales growth.
(v)
Statement B omitted that Carvana had entered a sham pass-
through arrangement with DriveTime. The 2021 10-K omitted that, for many of these sales,
Carvana kept none of the proceeds, but still recorded the revenue and retail unit sale on its
books. These vacuous sales were material as they made up 169% of Carvana’s reported
sequential growth in Q4 2021 and 19% in Q3 2021.
(b)
As a misled Needham & Company analyst noted near the end of the
Class Period, “‘[t]he [only] path forward for Carvana is to sell as many cars as possible, but
to do so on a profitable basis, versus prior it was more about selling as many cars as
possible.’”
385. Statement C: The 2021 10-K also contained the following risk disclosure:
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 308 of 332
- 305 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
We operate in several highly regulated industries and are subject to a
wide range of federal, state, and local laws and regulations. . . . [o]ur failure
to comply [with these laws and regulations], could have a material adverse
effect on our business, results of operations, and financial condition.
We are subject to a wide range of evolving federal, state, and local laws
and regulations, many of which may have limited to no interpretation
precedent as it relates to our business model. Our sale and purchase of used
vehicles and related activities, including the sale of complementary products
and services, are subject to state and local licensing requirements, state laws,
regulations, and systems and process requirements related to title and
registration . . . .
*
*
*
The violation of any of these laws or regulations could result in
administrative, civil, or criminal penalties or in a cease-and-desist order
against some or all of our business activities, any of which could damage
our reputation and have a material adverse effect on our business, sales, and
results of operations. Additionally, even an allegation that we violated these
laws, by regulators, competitors, individuals, or consumers, could result in
costly litigation with uncertain results.
386. As detailed below, Statement C was materially false and misleading when
made because it negligently failed to disclose the following material facts necessary to be
disclosed in order to make the Registration Statement not misleading:
(a)
The disclosed risk – that the violation of any state regulations and laws
“could result in administrative, civil, or criminal penalties or in a cease-and-desist order against
some or all of our business activities, any of which could damage our reputation and have a
material adverse effect on our business, sales, and results of operations” – was misleading
because it omitted that the risks it was warning investors about had already come to fruition.
In fact: (i) the Ohio BMV had investigated Carvana, suspended its temporary tag issuance
privileges for all Ohio locations, and subjected the Company to increased oversight in
December 2020; (ii) following additional meetings between MDOS and Carvana, Carvana
“admi[tted] . . . several more violations of [Michigan’s] Code,” faced thousands of dollars in
additional fines, and had its 18-month probation extended on February 7, 2022; and (iii)
Illinois had begun investigating Carvana in February 2022 for issuing out-of-state temporary
registration permits and for failing to transfer titles in a timely manner in violation of its
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 309 of 332
- 306 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
laws. Defendants also concealed the inevitably of the foregoing risks and identified harms
materializing because of the above omitted facts.
(b)
Statement C presented Carvana’s “failure to comply” with title and
registration laws and others’ “allegation[s] that we violated these laws,” as hypothetical risks
that could materially harm Carvana if they materialized when, in truth, these risks had
already come to fruition. Carvana routinely and repeatedly sold cars before holding title to
the vehicles, failed to register cars within the legally required timeframe, and issued out-of-
state temporary tags and/or license plates in violation of numerous state laws. Indeed, CWs-
3, 4, 8, 9, 10, and 11 universally reported that Carvana would sell these vehicles before they
received title. Specifically, CW-10 described instances where titles on vehicles Carvana sold
could not be obtained for over a year, which meant the buyers were unable to drive their cars.
CW-8 said that Carvana’s practice was simply to assume that titles for vehicles could be
acquired later if they were not available at the time of purchase. CW-4 said Carvana had a
Google spreadsheet that tracked vehicles without titles for his/her entire IRC and “there were
a ridiculous amount of cars on the list.” CW-9 said, beginning in Q1 2022, registration
delays amounted to roughly half of the calls the team had handled. Further, from June 2021
to July 2022, Carvana failed to timely register approximately 10% of its sales in Maryland,
which would later result in additional fines. In addition, the Barron’s expose would later
reveal that Carvana’s violations were systematic and wide-reaching. Carvana also concealed
the inevitably of the foregoing risks and identified harms materializing because of the above
omitted facts.
(c)
A reasonable investor would have viewed the above omitted facts as
important for the reasons stated herein. Indeed, the Exchange Act Defendants admitted in an
October 2022 court filing that “‘the damage to Carvana’s reputation and goodwill posed
by’” even one “‘suspension order is incalculable and irreparable.’”
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 310 of 332
- 307 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
E.
The Registration Statement Contained Deficient and Inaccurate
Risk Disclosures
387. Item 3 of Form S-1 required the Registration Statement to furnish the
information called for under Item 105 of Regulation S-K, namely, “a discussion of the
material factors that make an investment in the registrant or offering speculative or risky.”
17 C.F.R. §229.105. None of the risk disclosures in the Registration Statement meaningfully
disclosed the risks associated with Carvana’s mounting titling issues and its practice of
issuing temporary out-of-state licenses to customers for whom it was unable to procure
vehicle title.
388. The 2021 10-K warned that the Company is “subject to state and local
licensing requirements, state laws, regulations, and systems and process requirements related
to title and registration,” and that “violation of any of these laws or regulations could result
in administrative, civil, or criminal penalties or in a cease-and-desist order against some or
all of our business activities, any of which could damage our reputation and have material
adverse effect on our business, sales, and results of operations.”
389. The Registration Statement did not warn, however, that, at the time of the 2022
Public Offering, Carvana was already experiencing widespread issues complying with state
and local title and registration requirements, which the Company often addressed by issuing
illegal, out-of-state temporary license plates to customers while they attempted to procure
proper title.
390. As a result of these practices, Carvana faced a mounting risk of regulatory
action across the country, not merely as a result of being “subject to” state and local
licensing requirements, but as a result of the Company’s active violations of those
requirements. Moreover, the Company was already experiencing damage to its reputation as
a result of the title and licensing issues as customers lodged numerous complaints, reported
Carvana to local authorities, and increasingly sold their purchased vehicles back to Carvana
after they were unable to obtain proper title.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 311 of 332
- 308 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
391. Accordingly, the Registration Statement negligently failed to disclose the
material factors that made an investment in the offering risky or speculative, as required by
Item 3 of Form S-1.
F.
Events Subsequent to the 2022 Public Offering
392. Less than a month after the 2022 Public Offering, on May 10, 2022, Carvana
issued an 8-K announcing “a workforce reduction . . . in operational groups in connection
with . . . plans to better align staffing and expense levels with sale volumes.” Carvana
further announced, in connection with “these right-sizing initiatives,” that Carvana “will be
transitioning operations away from” an IRC in Ohio as well as “a few logistics hubs.”
Carvana explained that it believed these initiatives “will result in Carvana restoring a better
balance between its sales volumes and staffing levels and facilitate Carvana returning to
efficient growth.” In addition, Carvana announced that it would release “materials on
Carvana’s [updated] operating plan” later that week. On May 10, 2020, analyst Morgan
Stanley also reported on the 8-K, noting, that it was likely the “first of many potential steps
to address the cost structure,” and that “[b]eyond the actions disclosed in today’s 8-K, we
believe transparency around future actions and management’s disclosure and messaging
around the path forward will be critical to re-establishing investor confidence which has been
lost over the past few months.”
393. Forbes reported on the announcement, explaining, “Carvana’s mass firing was
a sign of much bigger problems at the company, according to 10 former employees . . . and
several industry analysts. They describe a spendthrift business, whose growth-at-all-costs
mentality undermined business operations and sowed the seeds of its recent layoffs.”
394. On May 12, 2022, Carvana’s license was suspended in Illinois for continued
violation of title and registration requirements.
395. On June 24, 2022, the Barron’s exposé was published. The article discussed
Carvana’s problems regarding title and registration and its violations of state laws and
regulations, citing interviews with Carvana customers and former employees. The article
revealed for the first time that Arizona had canceled Carvana’s temporary plate program, that
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 312 of 332
- 309 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
“[i]nterviews with state officials . . . show the issue is wide[]-reaching,” and, that Carvana had
revamped its title department three times and created an “undriveable-car task force.”
396. On October 7, 2022, MDOS announced that it had suspended Carvana’s
license for its dealership in Novi, Michigan due to recent title and licensing violations.
MDOS cited “imminent harm to the public” in suspending Carvana’s license.
397. On November 3, 2022, Carvana announced its Q3 2022 financial results,
disclosing that: (i) retail car sales had declined 8% from the prior year; (ii) Carvana had been
“frequently acquiring sales that were less profitable in the immediate period but . . . .
[c]urrently, as we are primarily focused on profitability, we are intentionally acquiring fewer
of these sales”; (iii) Carvana had been “reducing advertising in [its new,] distant markets
with less profitable sales”; and (iv) Carvana had “take[n] other actions to improve
profitability, such as increasing long-distance shipping fees, in these [new, far-flung]
markets.”
398. On February 23, 2023, Carvana announced its Q4 2022 financial results,
disclosing that: (i) retail unit sales had declined 23% year-over-year – a “first . . . in
[Carvana’s] history”; (ii) “it was the first year [Carvana] stepped back on the key metrics of
retail units sold”; (iii) gross profit had declined by 63% year-over-year; (iv) Carvana had
“significantly accelerated . . . reductions in inventory”; (v) Carvana had written off over $50
million of aged and impaired vehicle inventory that was worth less than Carvana had paid to
acquire it and prepare it for sale; (vi) Carvana was “significantly reducing retail vehicle
acquisitions”; and (vii) Carvana was intentionally foregoing retail sales growth for the
foreseeable future to instead focus on driving “positive unit economics.”
399. Purchasers in the 2022 Public Offering bought their shares for $80.00 per
share. On the date this action was filed, August 3, 2022, Carvana’s common stock traded in
a range of $35.59 to $32.50 per share. On February 14, 2023, the day the Consolidated
Complaint was filed, Carvana’s common stock traded in a range of $11.75 to $10.06 per
share.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 313 of 332
- 310 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
G.
The Role of the Underwriter Defendants in Connection with the
2022 Public Offering
400. The 2022 Public Offering was a firm commitment offering conducted by and
through the Underwriter Defendants.
401. The Registration Statement filed by Carvana in connection with the 2022
Public Offering explained: “We may sell these securities on a continuous or delayed basis,
directly, through agents, dealers or underwriters as designated from time to time, or through
a combination of these methods. If any agents, dealers, or underwriters are involved in the
sale of any securities, the applicable prospectus supplement will set forth their names and
any applicable commissions and discounts.”
402. Carvana’s Prospectus Supplement filed in connection with the 2022 Public
Offering explained that Citigroup and J.P. Morgan would act as underwriters in the 2022
Public Offering, and further represented that they would collectively earn more than $24
million in underwriting discounts and commissions as a result of their efforts in the 2022
Public Offering.
403. The Underwriter Defendants thus participated in the violations complained of
herein as detailed below.
404. The Underwriter Defendants are investment banking houses that, among other
things, specialize in underwriting public offerings of securities.
405. The Underwriter Defendants acted as Carvana’s financial advisors at the time
of the 2022 Public Offering.
406. In connection with the 2022 Public Offering, the Underwriter Defendants
marketed Carvana common stock to potential investors using materially false or misleading
information about the Company, and/or omitted material information required to be
disclosed in the Offering Documents. The Underwriter Defendants also caused the Offering
Documents to be filed with the SEC and to be declared effective in connection with the 2022
Public Offering. The Underwriter Defendants are liable to Plaintiffs and those similarly
situated under the Securities Act.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 314 of 332
- 311 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
407. The Underwriter Defendants’ participation in and solicitation of purchases of
Carvana Class A common stock in the 2022 Public Offering was motivated by their financial
interest. They served as the underwriters of the 2022 Public Offering and shared nearly
$25,000,000 in fees collectively for doing so.
408. The Underwriter Defendants also obtained an agreement from Carvana to
indemnify and hold the Underwriter Defendants harmless from any liability under the
Securities Act.
409. In return, the Underwriter Defendants were willing to promote and sell
Carvana’s Class A common stock in the 2022 Public Offering.
410. Concurrently with the 2022 Public Offering, Carvana received committed
financing from J.P. Morgan Chase Bank N.A., an affiliate of J.P. Morgan, and an affiliate of
Citigroup and conducted a senior unsecured notes offering, both of which were to help fund
the acquisition and integration of ADESA. The Underwriter Defendants acted as financial
advisors in Carvana in structuring these finance arrangements, and similarly advised Carvana
as to the 2022 Public Offering.
411. According to an April 17, 2022 The Wall Street Journal article, entitled
“Carvana, Once a Market Darling, Forced to Turn to Apollo for Cash,” around the time of
the 2022 Public Offering, “Carvana hired JP Morgan Chase & Co. to raise billions of dollars
in debt and equity to pay for the purchase and subsequent integration” of ADESA.
412. And according to an April 30, 2020 Bloomberg article, entitled “How Apollo’s
Last-Minute Twist Salvaged Carvana’s Debt Sale,” Carvana’s disappointing financial results
and worsening operational struggles made it difficult to find investors in Carvana’s debt
offering, which put “JPMorgan Chase & Co. and Citigroup Inc. . . . in a bind.”
413. Thus, on information and belief, the Underwriter Defendants were incentivized
– by their connections with Carvana, including their financial entanglements with Carvana’s
acquisition of ADESA and their positions as Carvana’s financial advisors – to maximize the
value received in the 2022 Public Offering, which was a key source of financing for the
Company at the time.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 315 of 332
- 312 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
414. Upon information and belief, representatives of the Underwriter Defendants
also assisted Carvana in planning the 2022 Public Offering, and had access to confidential
corporate information concerning Carvana’s operations and financial prospects.
415. In addition to availing themselves of virtually unlimited access to internal
corporate documents, on information and belief, agents of the Underwriter Defendants met
with Carvana’s lawyers, management, and top executives prior to the 2022 Public Offering.
416. Upon information and belief, during these meetings, agreements were reached
as to: (i) the strategy to best accomplish the 2022 Public Offering; (ii) the terms of the 2022
Public Offering, including the price at which Carvana common stock would be sold; (iii) the
language to be used in the Offering Documents; (iv) what disclosures would be made in the
Offering Documents; and (v) what responses would be made to the SEC in connection with
its review of the Offering Documents.
417. As a result of those frequent contacts and communications between the
Underwriter Defendants, Carvana, and the Individual Securities Act Defendants (as well as
the Underwriter Defendants’ direct involvement in material issues requiring disclosure,
including Carvana’s business performance and financials), the Underwriter Defendants knew
of, or in the exercise of reasonable care should have known of, the existing yet undisclosed
conditions and material risks detailed herein, which were either misrepresented in or omitted
from the Offering Documents.
418. At a minimum, the Underwriter Defendants were negligent in not knowing,
and failing to disclose in connection with the 2022 Public Offering, the adverse information
alleged herein that was contrary to the disclosures in the Offering Documents, the omission
of which rendered the Offering Documents false and misleading at the time it was made
effective.
419. Moreover, the Underwriter Defendants were required to investigate with due
diligence the representations in the Offering Documents to confirm that they did not contain
materially misleading statements or omit material facts. None of the Underwriter
Defendants made a reasonable investigation or possessed reasonable grounds for the belief
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 316 of 332
- 313 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
that the statements in the Offering Documents (described herein) were true, were without
omission of any material facts, and/or were not misleading. The Underwriter Defendants
thus negligently failed to conduct proper due diligence, and were negligent in not knowing of
the materially untrue statements and omissions contained in the Offering Documents, as
detailed herein.
420. The Underwriter Defendants helped cause the Offering Documents to be filed
with the SEC and declared effective in connection with the offer and sale of the shares of
Carvana common stock registered thereby, including those shares purchased by Plaintiffs
and other members of the Securities Act Class.
H.
Class Action Allegations for Securities Act Claims
421. Plaintiffs bring this action as a class action pursuant to Federal Rules of Civil
Procedure 23(a) and 23(b)(3) on behalf of all persons who purchased Carvana common stock
directly in the 2022 Public Offering pursuant to the Registration Statement. This class
asserts claims only for violations of §§11, 12(a)(2), and 15 of the Securities Act, 15 U.S.C.
§§77k, 77l, and 77o. This class does not assert any claims sounding in fraud.
422. Any person who did not purchase or acquire Carvana shares directly in the
2022 Public Offering and pursuant to the Registration Statement is not included in the
Securities Act Class. Also excluded from the Securities Act Class are Defendants, the
officers and directors of the Company, members of their immediate families and their legal
representatives, heirs, successors or assigns, and any entity in which Defendants have or had
a controlling interest.
423. The members of the Securities Act Class are so numerous that joinder is
impracticable. The 2022 Public Offering involved the issuance and sale of millions of shares
of Carvana stock, which were publicly traded on the NYSE following the 2022 Public
Offering. While the exact number of Securities Act Class members is unknown to Plaintiffs
at this time, Plaintiffs believe that there are thousands of members in the proposed Securities
Act Class. Record owners and other members of the Securities Act Class may be identified
from records maintained by Carvana or its transfer agent and may be notified of the
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 317 of 332
- 314 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
pendency of this action by mail, using the form of notice similar to that customarily used in
securities class actions.
424. Plaintiffs’ claims are typical of the claims of the Securities Act Class, as all
Securities Act Class members were and are similarly affected by Defendants’ conduct.
425. Plaintiffs will fairly and adequately protect the interests of Securities Act Class
members and have retained counsel competent and experienced in securities class action
litigation.
426. Common questions of law and fact exist as to all Securities Act Class members
and predominate over any questions solely affecting individual Securities Act Class
members. Among the common questions of law and fact are:
(a)
whether the Securities Act Defendants violated the Securities Act;
(b)
whether the Registration Statement misrepresented and/or omitted
material facts in violation of the Securities Act; and
(c)
whether and to what extent Securities Act Class members have
sustained damages and the proper measure of damages.
427. A class action is superior to all other available methods for the fair and
efficient adjudication of this controversy. Because the damages suffered by individual
Securities Act Class members may be relatively small, the expense and burden of individual
litigation make it exceedingly difficult, if not impossible and impracticable, for Securities
Act Class members to individually redress the alleged wrongs done to them. There will be
no difficulty in managing this action as a class action.
XV. CLAIMS FOR RELIEF UNDER THE SECURITIES ACT
COUNT IV
Violations of §11 of the Securities Act
(Against the Securities Act Defendants)
428. Plaintiffs incorporate ¶¶19-23, 32-33, 39-42, 44, and 365-427 by reference.
429. This Count is brought pursuant to §11 of the Securities Act, 15 U.S.C. §77k, on
behalf of the Class, against the Securities Act Defendants.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 318 of 332
- 315 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
430. The Securities Act Defendants are liable under theories of strict liability for
their violations of §11 of the Securities Act. Plaintiffs do not allege that the Securities Act
Defendants had scienter or fraudulent intent in connection with this Count.
431. The Registration Statement used to complete the 2022 Public Offering
contained untrue statements of material fact, omitted to state material facts required to be
stated therein and/or omitted to state other facts necessary to make the statements made
therein not misleading.
432. By reason of the conduct herein alleged, each Securities Act Defendant named
herein violated §11 of the Securities Act.
433. Plaintiffs purchased Carvana common stock in the 2022 Public Offering.
Specifically: (a) UANPF and SHEPP purchased Carvana Class A common stock directly in
the April 2022 Public Offering; (b) UANPF and SHEPP purchased Carvana Class A
common stock on April 22, 2022 at $80.00 per share; and (c) UANPF and SHEPP purchased
Carvana Class A common stock directly from underwriter defendant Citigroup.
434. Plaintiffs and the Class sustained damages as the value of the stock issued in
the 2022 Public Offering declined due to the Securities Act Defendants’ violations.
435. At the time of their purchases of Carvana Class A common stock issued in the
2022 Public Offering, Plaintiffs and other members of the Class were without knowledge of
the facts concerning the wrongful conduct alleged herein and could not have reasonably
discovered those facts prior to the disclosures herein. Less than one year has elapsed from
the time that Plaintiffs discovered or reasonably could have discovered the facts upon which
this Complaint is based to the time that Plaintiffs commenced this action. Less than three
years have elapsed between the time that the stock upon which this Count is brought were
offered to the public and the time Plaintiffs commenced this action.
COUNT V
Violations of §12(a)(2) of the Securities Act
(Against the Securities Act Defendants)
436. Plaintiffs repeat and re-allege the above allegations in ¶¶19-23, 32-33, 39-42,
44, and 365-435 as if fully set forth herein.
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 319 of 332
- 316 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
437. This Count is brought pursuant to §12(a)(2) of the Securities Act, 15 U.S.C.
§771(a)(2), on behalf of the Securities Act Class, against the Securities Act Defendants.
This Count does not allege, and does not intend to allege, fraud or fraudulent intent, which is
not a required element of §12(a)(2), and any implication of fraud or fraudulent intent is
hereby expressly disclaimed.
438. Section 12(a)(2) gives rise to liability as to “[a]ny person who . . . offers or
sells a security . . . by means of a prospectus or oral communication, which includes an
untrue statement of a material fact or omits to state a material fact necessary in order to make
the statements, in the light of the circumstances under which they were made, not
misleading.” 15 U.S.C. §771(a)(2). Liability for a violation of §12(a)(2) extends to those, at
a minimum, who passed title to the security to the purchaser, as well as those who solicited
the purchase.
439. By means of the defective Prospectus Supplement, which was incorporated in
and formed part of the Registration Statement for the 2022 Public Offering, these Defendants
promoted and sold, for the benefit of themselves and their associates, Carvana common stock
to the Plaintiffs and other members of the Securities Act Class. In the absence of their
efforts to publicize the 2022 Public Offering and solicit Carvana common stock purchasers,
the 2022 Public Offering could not have occurred.
440. The Underwriter Defendants participated in the preparation and dissemination
of the defective and inaccurate Offering Documents for their own financial benefit. But for
their participation in the 2022 Public Offering, including their solicitation, the 2022 Public
Offering could not, and would not, have been accomplished.
441. The Offering Documents contained untrue statements of material fact, and/or
concealed or failed to disclose material facts, as detailed above. The Securities Act
Defendants owed Plaintiffs and the other members of the Class who purchased Carvana
common stock pursuant to the Offering Documents the duty to make a reasonable and
diligent investigation of the statements contained in the Offering Documents to ensure that
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 320 of 332
- 317 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
such statements were true and that there was no omission to state a material fact required to
be stated in order to make the statements contained therein not misleading.
442. By reason of the conduct alleged herein, the Securities Act Defendants violated
§12(a)(2) of the Securities Act. As a direct and proximate result of such violations, Plaintiffs
and the other members of the Class who purchased Carvana common stock pursuant to the
Prospectus Supplement sustained damages in connection with their purchases. Accordingly,
Plaintiffs and the other members of the Class who hold the common stock issued pursuant to
the Prospectus Supplement have the right to rescind and recover the consideration paid for
their shares, and hereby tender their common stock to the defendants sued herein. Class
members who have sold their common stock seek damages to the extent permitted by law.
COUNT VI
Violations of §15 of the Securities Act
(Against the Individual Securities Act Defendants)
443. Plaintiffs repeat and re-allege the above allegations in ¶¶19-23, 32-33, 39-42,
44, and 365-442 as if fully set forth herein.
444. This Count is brought pursuant to §15 of the Securities Act, 15 U.S.C. §77o,
against the Individual Securities Act Defendants. This Count does not allege, and does not
intend to allege, fraud or fraudulent intent, which is not a required element of §15, and any
implication of fraudulent intent is hereby expressly disclaimed.
445. Where a violation of §11 or §12(a)(2) occurs, §15 gives rise to liability as to
“[e]very person who, by or through stock ownership, agency, or otherwise, or who, pursuant
to or in connection with an agreement or understanding with one or more other persons by or
through stock ownership, agency, or otherwise, controls any person liable under Sections
77k or 77l [§11 or §12(a)(2)].” 15 U.S.C. §77o(a). Control persons under §15 are “liable
jointly and severally with and to the same extent as such controlled person to any person to
whom such controlled person is liable.” Id.
446. As detailed herein, each of the Individual Securities Act Defendants committed
primary violations of the Securities Act and is directly responsible and primarily liable for
any such violations. In addition, the Individual Securities Act Defendants acted as
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 321 of 332
- 318 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
controlling persons of Carvana within the meaning of §15 of the Securities Act by virtue of
their position as a director and/or senior officer of Carvana. By reason of their senior
management positions and/or directorships at the Company, as alleged above, these
Individual Securities Act Defendants, individually and acting pursuant to a common plan,
had the power to influence and exercised the same to cause Carvana to engage in the conduct
complained of herein. By reason of such conduct, the Individual Securities Act Defendants
are liable pursuant to §15 of the Securities Act.
447. Each of the Individual Securities Act Defendants was a culpable participant in
the violations of §11 alleged in Count IV above. Because of their senior executive
management and/or director positions with Carvana, they each had access to the undisclosed
adverse information about the Company’s business, operations, market trends, and present
and future business prospects via internal corporate documents, conversations, and
connections with other corporate officers and employees, attendance at management and/or
Board meetings and committees thereof. Based on this knowledge and: (i) having signed
and/or authorized the signing of the Registration Statement; (ii) being named in the
Registration Statement and identified as a director and/or executive officer; and/or (iii)
playing a material role or otherwise participating in the process which allowed the 2022
Public Offering to be successfully completed, each of the Individual Securities Act
Defendants was a culpable participant in the violations of §11 alleged in Count IV above.
448. By reason of the conduct alleged herein, the Individual Securities Act
Defendants violated §15 of the Securities Act, and Plaintiffs and the Class have suffered
harm as a result.
PRAYER FOR RELIEF
WHEREFORE, Plaintiffs demand judgment against Defendants as follows:
A.
Declaring this action to be a class action properly maintained pursuant to Rule
23(a) and (b)(3) of the Federal Rules of Civil Procedure and certifying Plaintiffs as Class
Representatives and Robbins Geller Rudman & Dowd LLP as Class Counsel;
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 322 of 332
- 319 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
B.
Awarding compensatory damages in favor of Plaintiffs and the other members
of the Class against all Defendants, jointly and severally, for all damages sustained as a
result of Defendants’ wrongdoing, in an amount to be proven at trial, including interest
thereon;
C.
Awarding Plaintiffs and the Class rescission or a rescissory measure of
damages;
D.
Awarding Plaintiffs and the Class their reasonable costs and expenses incurred
in this action, including reasonable attorneys’ fees, accountants’ fees, and experts’ fees, and
other costs and disbursements; and
E.
Awarding Plaintiffs and other members of the Class such other injunctive or
equitable relief, including disgorgement and/or the imposition of a constructive trust, that
may be deemed just and proper by the Court.
DEMAND FOR TRIAL BY JURY
Plaintiffs demand a trial by jury.
DATED: March 29, 2024
ROBBINS GELLER RUDMAN
& DOWD LLP
DANIEL S. DROSMAN
(Admitted pro hac vice)
RACHEL A. COCALIS
(Admitted pro hac vice)
SARAH A. FALLON
(Admitted pro hac vice)
s/ DANIEL S. DROSMAN
DANIEL S. DROSMAN
655 West Broadway, Suite 1900
San Diego, CA 92101
Telephone: 619/231-1058
619/231-7423 (fax)
ddrosman@rgrdlaw.com
rcocalis@rgrdlaw.com
sfallon@rgrdlaw.com
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 323 of 332
- 320 -
4890-0865-5022.v1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
ROBBINS GELLER RUDMAN
& DOWD LLP
ROBERT M. ROTHMAN
(Admitted pro hac vice)
DAVID A. ROSENFELD
(Admitted pro hac vice)
BRENT E. MITCHELL
(Admitted pro hac vice)
58 South Service Road, Suite 200
Melville, NY 11747
Telephone: 631/367-7100
631/367-1173 (fax)
rrothman@rgrdlaw.com
drosenfeld@rgrdlaw.com
bmitchell@rgrdlaw.com
Lead Counsel for Lead Plaintiffs
O’DONOGHUE & O’DONOGHUE LLP
DINAH S. LEVENTHAL
5301 Wisconsin Avenue, N.W., Suite 800
Washington, DC 20015
Telephone: 202/362-0041
202/362-2640 (fax)
dleventhal@odonoghuelaw.com
Additional Counsel for Lead Plaintiffs
BONNETT FAIRBOURN FRIEDMAN
& BALINT PC
ANDREW FRIEDMAN
7301 N. 16th Street, Suite 102
Phoenix, AZ 85020
Telephone: 602/274-1100
602/274-1199 (fax)
afriedman@bffb.com
Local Counsel
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 324 of 332
CARVANA
CERTIFICATION PURSUANT TO FEDERAL SECURITIES LAWS
United Association National Pension Fund (“Plaintiff”) declares:
1.
Plaintiff has reviewed a complaint and authorized its filing. Plaintiff
has authorized the filing of a motion for appointment as lead plaintiff.
2.
Plaintiff did not acquire the security that is the subject of this action at
the direction of plaintiff’s counsel or in order to participate in this private action or
any other litigation under the federal securities laws.
3.
Plaintiff is willing to serve as a representative party on behalf of the
class, including providing testimony at deposition and trial, if necessary.
4.
Plaintiff has made the following transaction(s) during the Class Period
in the securities that are the subject of this action: See attached Schedule A.
5.
Plaintiff has not sought to serve or served as a representative party in a
class action that was filed under the federal securities laws within the three-year
period prior to the date of this Certification except as detailed below: None.
6.
Plaintiff will not accept any payment for serving as a representative
party on behalf of the class beyond the Plaintiff’s pro rata share of any recovery,
except such reasonable costs and expenses (including lost wages) directly relating
to the representation of the class as ordered or approved by the court.
I declare under penalty of perjury that the foregoing is true and correct.
Executed this ____ day of March, 2024.
United Association National Pension Fund
By:
Toni C. Inscoe, Fund Administrator
27th
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 325 of 332
Common Stock
Date
Amount of
Acquired
Shares Acquired
Price
09/24/2020
2
$214.97
09/24/2020
74
$208.45
09/24/2020
198
$215.97
09/25/2020
126
$203.41
09/28/2020
226
$217.56
09/29/2020
74
$223.24
10/01/2020
600
$237.67
10/15/2020
700
$224.04
10/16/2020
300
$221.71
10/30/2020
800
$184.01
11/04/2020
466
$200.58
11/05/2020
234
$206.89
11/06/2020
31
$209.86
11/06/2020
188
$209.35
11/09/2020
112
$201.41
11/09/2020
169
$200.39
11/12/2020
400
$214.52
11/18/2020
217
$224.12
11/19/2020
583
$227.72
11/23/2020
731
$242.08
12/08/2020
300
$262.09
01/06/2021
1,175
$253.43
01/06/2021
6,285
$252.00
01/07/2021
17
$265.48
01/08/2021
400
$278.64
01/12/2021
292
$288.68
01/12/2021
447
$287.20
01/12/2021
544
$285.94
01/12/2021
1,968
$287.76
01/13/2021
445
$292.57
01/13/2021
607
$287.68
01/13/2021
633
$289.26
01/13/2021
1,772
$289.52
01/14/2021
104
$297.93
01/14/2021
200
$297.39
01/14/2021
3,591
$297.36
01/21/2021
1,618
$273.56
01/29/2021
400
$262.34
02/10/2021
300
$297.30
02/26/2021
167
$288.01
02/26/2021
332
$278.31
02/26/2021
3,220
$281.61
03/12/2021
200
$287.43
04/06/2021
360
$280.73
04/15/2021
300
$283.15
05/24/2021
400
$253.93
SCHEDULE A
SECURITIES TRANSACTIONS
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 326 of 332
Date
Amount of
Acquired
Shares Acquired
Price
06/11/2021
200
$274.28
07/01/2021
106
$305.42
07/02/2021
106
$311.95
07/02/2021
395
$309.73
07/06/2021
314
$316.17
07/06/2021
319
$317.89
07/07/2021
302
$316.17
07/08/2021
131
$320.62
07/08/2021
228
$312.87
07/08/2021
368
$322.51
07/09/2021
74
$325.17
07/09/2021
249
$322.53
07/12/2021
105
$323.00
07/13/2021
97
$326.12
07/13/2021
99
$324.44
07/13/2021
438
$323.61
07/14/2021
222
$316.61
07/14/2021
661
$318.54
07/15/2021
33
$310.83
07/15/2021
529
$311.42
07/16/2021
166
$313.61
07/16/2021
666
$314.16
07/19/2021
23
$305.67
07/19/2021
209
$306.80
07/20/2021
66
$325.34
07/20/2021
312
$320.69
07/21/2021
19
$330.32
07/21/2021
30
$332.05
07/21/2021
54
$330.40
07/21/2021
63
$330.38
07/21/2021
76
$325.14
07/22/2021
22
$334.62
07/22/2021
202
$335.28
07/23/2021
49
$338.92
07/23/2021
218
$334.39
07/26/2021
323
$337.23
07/27/2021
69
$330.14
07/27/2021
72
$330.27
07/27/2021
788
$331.51
07/28/2021
163
$337.62
07/28/2021
263
$340.62
07/29/2021
416
$338.40
07/29/2021
875
$338.70
07/30/2021
36
$336.17
07/30/2021
58
$337.56
07/30/2021
115
$336.77
07/30/2021
206
$340.07
08/02/2021
14
$335.97
08/09/2021
120
$352.05
08/11/2021
3
$356.97
08/12/2021
3
$356.99
08/16/2021
13
$356.91
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 327 of 332
Date
Amount of
Acquired
Shares Acquired
Price
08/16/2021
64
$356.22
08/16/2021
90
$356.93
08/17/2021
74
$352.22
08/17/2021
233
$352.22
08/30/2021
100
$334.85
08/31/2021
200
$330.93
10/08/2021
200
$284.63
12/17/2021
500
$214.07
01/10/2022
162
$173.71
01/10/2022
438
$173.39
01/12/2022
700
$179.37
02/25/2022
374
$153.32
02/25/2022
426
$146.79
03/01/2022
300
$139.04
03/04/2022
300
$113.88
04/22/2022
1,455
$80.00
Date
Amount of
Disposed
Shares Disposed
Price
04/08/2021
10,517
$270.76
05/10/2021
279
$245.11
10/12/2021
80
$282.67
10/12/2021
157
$282.29
10/13/2021
17
$281.90
10/13/2021
480
$282.41
10/14/2021
9
$283.93
10/14/2021
221
$286.00
10/15/2021
35
$290.50
10/15/2021
99
$287.61
10/15/2021
256
$288.11
10/18/2021
69
$289.45
10/18/2021
117
$288.82
10/18/2021
152
$288.90
10/19/2021
20
$292.69
10/19/2021
118
$293.46
10/20/2021
21
$291.29
10/20/2021
68
$292.95
10/20/2021
253
$289.55
10/21/2021
19
$302.00
11/02/2021
191
$300.61
11/16/2021
399
$298.31
11/30/2021
95
$283.46
11/30/2021
105
$279.68
03/22/2022
14,007
$136.63
04/28/2022
8,262
$61.88
11/07/2022
993
$7.51
11/07/2022
1,348
$7.48
11/07/2022
4,051
$7.40
11/08/2022
6,174
$7.12
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 328 of 332
Bond
Date
Type of
Face
Acquired
Debt
Amount
Price
08/12/2021
770,000
$100.00
09/20/2021
400,000
$99.63
12/20/2021
20,000
$93.13
01/05/2022
75,000
$94.75
01/06/2022
130,000
$94.25
01/26/2022
120,000
$91.25
03/25/2021
580,000
$100.00
05/03/2021
105,000
$102.38
05/04/2021
45,000
$102.38
05/04/2021
65,000
$102.38
05/05/2021
25,000
$102.63
05/06/2021
50,000
$103.25
05/10/2021
105,000
$103.50
05/11/2021
65,000
$103.25
01/26/2022
75,000
$94.25
04/30/2021
505,000
$103.25
12/20/2021
105,000
$99.00
01/26/2022
275,000
$95.38
Date
Type of
Face
Disposed
Debt
Amount
Price
05/11/2022
35,000
$65.50
05/11/2022
40,000
$63.75
05/26/2022
80,000
$66.00
07/05/2022
270,000
$56.25
08/11/2022
170,000
$65.75
09/22/2022
65,000
$52.00
09/23/2022
855,000
$50.00
04/20/2022
125,000
$86.00
04/20/2022
310,000
$81.75
04/26/2022
120,000
$82.50
05/26/2022
75,000
$72.00
07/05/2022
345,000
$64.50
09/23/2022
140,000
$56.75
04/27/2022
70,000
$88.33
04/27/2022
110,000
$88.25
04/27/2022
160,000
$88.00
05/11/2022
165,000
$75.50
03/16/2022
105,000
$87.00
03/17/2022
35,000
$87.50
03/17/2022
80,000
$87.50
03/17/2022
160,000
$87.00
Prices listed are rounded to two decimal places.
5.625% due 10/01/2025
5.875% due 10/01/2028
5.875% due 10/01/2028
5.875% due 10/01/2028
5.875% due 10/01/2028
5.5% due 04/15/2027
5.5% due 04/15/2027
5.5% due 04/15/2027
5.625% due 10/01/2025
5.625% due 10/01/2025
5.625% due 10/01/2025
4.875% due 09/01/2029
4.875% due 09/01/2029
4.875% due 09/01/2029
5.5% due 04/15/2027
5.5% due 04/15/2027
5.5% due 04/15/2027
5.875% due 10/01/2028
5.875% due 10/01/2028
4.875% due 09/01/2029
4.875% due 09/01/2029
4.875% due 09/01/2029
4.875% due 09/01/2029
5.5% due 04/15/2027
5.5% due 04/15/2027
5.5% due 04/15/2027
5.5% due 04/15/2027
5.5% due 04/15/2027
5.625% due 10/01/2025
4.875% due 09/01/2029
4.875% due 09/01/2029
5.5% due 04/15/2027
5.5% due 04/15/2027
5.5% due 04/15/2027
5.5% due 04/15/2027
4.875% due 09/01/2029
4.875% due 09/01/2029
4.875% due 09/01/2029
4.875% due 09/01/2029
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 329 of 332
CARVANA
CERTIFICATION PURSUANT TO FEDERAL SECURITIES LAWS
Saskatchewan Healthcare Employees' Pension Plan (“Plaintiff”) declares:
1.
Plaintiff has reviewed a complaint and authorized its filing. Plaintiff
has authorized the filing of a motion for appointment as lead plaintiff.
2.
Plaintiff did not acquire the security that is the subject of this action at
the direction of plaintiff’s counsel or in order to participate in this private action or
any other litigation under the federal securities laws.
3.
Plaintiff is willing to serve as a representative party on behalf of the
class, including providing testimony at deposition and trial, if necessary.
4.
Plaintiff has made the following transaction(s) during the Class Period
in the securities that are the subject of this action: See attached Schedule A.
5.
Plaintiff has not sought to serve or served as a representative party in a
class action that was filed under the federal securities laws within the three-year
period prior to the date of this Certification except as detailed below:
Chin v. KE Holdings Inc., No. 1:21-cv-11196 (S.D.N.Y.)
Retail Wholesale Dept. Store Union Local 338 Ret. Fund v. Beyond Meat, Inc., No. 2:23-cv-03602 (C.D. Cal.)
6.
Plaintiff will not accept any payment for serving as a representative
party on behalf of the class beyond the Plaintiff’s pro rata share of any recovery,
except such reasonable costs and expenses (including lost wages) directly relating
to the representation of the class as ordered or approved by the court.
I declare under penalty of perjury under the laws of the United States of
America that the foregoing is true and correct. Executed this ____ day of March,
2024.
Saskatchewan Healthcare Employees' Pension
Plan
By:
Dale Markewich, Chief Financial
Officer
28
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 330 of 332
Stock
Date
Acquired
Price
08/05/2020
1,181
$173.54
08/05/2020
2,229
$174.85
08/06/2020
1,230
$206.05
08/06/2020
1,923
$202.24
08/07/2020
648
$203.91
08/07/2020
3,677
$203.27
08/10/2020
972
$189.18
08/11/2020
810
$188.84
08/12/2020
147
$191.97
08/13/2020
250
$193.84
08/14/2020
180
$195.03
08/17/2020
1,664
$195.41
08/19/2020
264
$199.39
08/20/2020
55
$200.92
08/21/2020
607
$203.34
08/24/2020
1,114
$199.60
08/25/2020
490
$200.21
08/26/2020
581
$205.36
08/27/2020
1,400
$207.83
08/28/2020
2,163
$216.98
11/25/2020
516
$241.12
10/20/2021
217
$289.28
10/22/2021
130
$296.58
10/25/2021
104
$292.22
10/26/2021
138
$292.34
10/27/2021
281
$295.21
11/18/2021
166
$289.28
11/18/2021
202
$286.09
11/19/2021
569
$291.63
11/22/2021
2,426
$284.57
11/23/2021
1,495
$281.02
11/24/2021
1,121
$287.73
11/26/2021
363
$289.60
11/29/2021
734
$292.23
11/30/2021
221
$279.68
11/30/2021
2,167
$283.65
12/01/2021
538
$278.76
01/14/2022
2,084
$156.39
04/04/2022
36
$132.97
04/04/2022
183
$130.31
04/04/2022
505
$132.81
04/04/2022
1,885
$130.19
04/05/2022
2,413
$121.78
04/22/2022
3,838
$80.00
06/27/2022
1,803
$29.87
SCHEDULE A
SECURITIES TRANSACTIONS
Amount of
Shares Acquired
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 331 of 332
Date
Acquired
Price
06/28/2022
1,803
$26.79
07/20/2022
2,569
$26.73
07/20/2022
6,897
$26.28
07/21/2022
1,695
$27.80
07/21/2022
3,921
$27.66
07/22/2022
945
$25.59
07/22/2022
1,219
$27.70
07/22/2022
8,650
$25.91
07/25/2022
2,421
$25.18
07/26/2022
4,486
$23.85
Date
Disposed
Price
10/21/2020
3,224
$195.69
12/11/2020
1,553
$259.91
02/12/2021
1,611
$296.99
Prices listed are rounded to two decimal places.
Amount of
Shares Disposed
Amount of
Shares Acquired
Case 2:22-cv-02126-MTL Document 71 Filed 03/29/24 Page 332 of 332File and source
- File
- gov.uscourts.azd.1319766.71.0.pdf
- Size
- 5,480,704 bytes
- SHA-256
- c66f737c729da9e0dcf8b84e98a3895a7fffd8d770e53f1ff660a253f49e5cdb
- Original
- archive.org