Full text
Testimony by Professor John M. Griffin, Hearing on “Fintech and Transparency in Small Business
Lending” by The Small Business Subcommittee on Oversight, Investigations, and Regulations
Chairman Phillips, Ranking Member Van Duyne, and members of the Committee, thank you for inviting
me to appear before The Small Business Subcommittee on Oversight, Investigations, and Regulations
today to speak with you about FinTech lending. My name is John M. Griffin, I am a finance professor in
the McCombs School of Business at The University of Texas at Austin. I am also the CEO of Integra FEC, a
small financial consulting firm in Austin, Texas that focuses on investigating complex financial fraud. As a
finance professor that focuses on forensic finance, or the investigation of fraud and potential misreporting
in financial market, I think it is important to thoroughly investigate the past in order to guide justice and
future policy.
This testimony is based on my paper with my coauthors, Professor Samuel Kruger and Prateek Mahajan,
entitled, “Did FinTech Lenders Facilitate PPP Fraud?” which is found on the Social Science Research
Network (SSRN) at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3906395. I will summarize
some of the main findings from our paper and then briefly discuss potential policy implications.
Our paper analyzes the SBA’s Paycheck Protection Program (PPP) based on four main metrics/flags and
seven additional flags related to potential misreporting. Four main findings of the paper are: 1) The flags
for potential misreporting consistently concentrates in FinTech, with the 12 lenders with the most
misreporting all being FinTech lenders. Overall, FinTech loans are 6.52 times as likely to be fraudulent. 2)
Misreporting is not a simple function of getting money out the door quickly in 2020. In fact, the fraud at
the end of the program in May 2021 is four times the level at the start of the program. 3) The amount of
likely fraud is quite high. Our four main measures, which are cross-verified by seven additional measures,
place the likely level of misreporting at 1.41 million questionable loans representing $64.2 billion in
capital. However, when we take a broader look at potential misreporting, we find a total suspicious
lending estimate of $117.3 billion. Since these analyses use only public data and take a conservative
approach, the total amount is likely even larger. 4) Suspicious loans are being overwhelmingly forgiven at
similar rates to other loans, and very few are being prosecuted, indicating that substantial reforms in SBA
lending are needed.
Let me give some brief background of FinTech lending in the PPP. The PPP rapidly distributed over $793
billion of funds in three short rounds between April 2020 and May 2021. Although FinTech lenders began
slowly with less than 5% of loans in round 1, they ramped up to over 80% of loans by May 2021,
highlighting their growing importance. FinTech lending was recognized by the media and academics for
broadening access to PPP loans, particularly to smaller firms without pre-existing lending relationships
with traditional banks, and for facilitating quick and efficient lending when many small businesses were
in need of funds due to the COVID-19 pandemic. However, the rapid expansion of FinTech lending may
have come at the expense of underwriting standards. Whereas traditional banks have established
borrower relationships and extensive Bank Secrecy Act (BSA) compliance programs, many FinTech lenders
had few established relationships and may have been less diligent when establishing formal procedures
with little reputation to protect.
Our paper examined whether FinTech lenders prevent or facilitate fraud in the PPP, and how potential
fraud and misreporting vary across individual lenders. To investigate these questions, we perform a big
data analysis of loan features for the 11.5 million PPP loans with eight disparate datasets. We introduce
four primary and four secondary indicators of whether a loan is potentially fraudulent, which we then
validate with three independent external measures. Each indicator creates an inference that a loan is
1
suspicious but is not proof of fraud on its own. The four primary measures are non-registered businesses,
multiple loans at a residential address, abnormally high implied compensation relative to industry by core-
based statistical area (CBSA) averages, and large inconsistencies (as large as tenfold) between the jobs
reported by borrowers on their PPP application and jobs reported to another contemporaneous
government program application with a different incentive structure. Suspicious lending rates vary
substantially across lenders, with potential misreporting rates in excess of 25% for ten large FinTech
lenders as can be seen in Figure 2 from our paper (reproduced below). The red and light yellow are non-
bank and bank FinTech lenders, which are mostly on the left with high levels of misreporting indicators.
Interestingly, not all FinTech lenders have high misreporting rates. In particular, Square and Intuit have
among the lowest rates of potential misreporting. These lenders also benefited from established
reputations and broad existing customer relationships before the PPP. This indicates that online lending
in and of itself does not appear to be the problem. Additionally, the paper shows that the loans are not
simply a function of loan composition. FinTech lenders exhibited consistently higher misreporting rates
across all borrower types and even after controlling for a rich set of loan and borrower characteristics,
indicating that the higher misreporting in FinTech loans was not driven by composition.
We spend substantial time in the paper assessing our indicators. I don’t have time to overview much of
our analysis in the 56 page paper and 55 page Internet Appendix, but it is safe to say that indicators are
quite robust. For example, we found that many borrowers reported having 10 employees to the EIDL
Advance program while only themselves when applying in the PPP.
As shown in Figure 5 from the paper (reproduced below), the percentage of flagged loans increased as
one approached the perceived $100k threshold. This is true for traditional lenders in grey dots, but much
more prevalent for FinTech lenders in red dots.
2
A basic interpretation of this figure is that fraudsters maxed out the amount of cash they could receive,
and this was particularly true for loans through FinTech lenders. Additionally, in many U.S. counties the
number of businesses in an industry-county pair frequently exceeds the number of establishments listed
for that industry and county in U.S. Census data. For FinTech lenders, 39.3\% of loans exceed industry-
county establishment counts, and 32.6\% of loans exceed industry-county establishment counts by a
factor of more than two. Additionally, we found repeated use of recurring loan features, and in ways that
were likely improbable. For example, we found 938 loans all in downtown Chicago for miscellaneous crop
farming, all for $20,000. These patterns were not isolated and are present in many zip codes and counties
across the U.S., indicating the potential spread through criminal networks. Finally, FinTech borrowers are
more than 3.4 times as likely to have a felony record, and borrowers flagged for potential misreporting
based on the primary and other secondary measures are also more likely to have felony records.
A common reaction to our paper is that the higher level of fraud may be ok because the design of the
program was to get money out the door quickly. However, an interesting finding is that the fraud through
FinTech lenders got worse throughout the program. There were particularly high rates in the last month
of round 3 (25.0%) as can be shown in Figure 10 of the paper (reproduced below).
3
Lenders with high rates of misreporting in rounds 1 and 2 increased both their misreporting rates and
their loan volume in round 3, which indicates that their due diligence process was not improving. For
example, the four largest FinTech lenders exhibited high rates of misreporting and large lending volume
growth while generating approximately a billion dollars in processing fees each. Finally, FinTech lenders
often doubled, tripled, or even quadrupled their lending in zip codes with high levels of potential
misreporting in rounds 1 and 2 while also substantially increasing their misreporting percentages.
Overall, we find 1.41 million questionable loans representing $64.2 billion in capital with our primary
measures. While these measures inevitably contain some false positives, the measures also miss many
forms of suspicious lending. Slightly lowering the threshold on the high implied compensation and
considering excess loans in industry-county pairs beyond the number of establishments reported by the
U.S. Census results in a total suspicious lending estimate of $117.3 billion. Moreover, this sensitivity
analysis is along only two limited dimensions. It is important to note that our estimates take a very
conservative view of potential misreporting. For example, none of our measures consider whether the
loan request was “necessary to support the ongoing operations” or whether the funds were used for
“unauthorized purposes.”
Our findings have important policy implications. First, the PPP did not include robust verification
requirements, and traditional banks may have been more apt to follow standard lending practices than
new FinTech lenders. The lack of rigorous verification for PPP loans seems to have led to substantial costs
to taxpayers, especially in 2021 when there was likely also less urgency to the loans. Second, FinTech
lending, though found to be successful at adapting to new environments and quickly disbursing funds in
other papers, needs to improve due diligence practices. Two established FinTech lenders persistently have
low rates of misreporting, indicating that FinTech lending need not be substandard. Third, our evidence,
along with evidence that the PPP saved relatively few jobs at a high cost, (Autor et al. 2022; Chetty et al.
2020; Granja et al. 2020), providing growing evidence that the PPP may not have been an efficient source
of capital allocation and perhaps indicate that SBA lending may not be effective source of government
spending. Fourth, incentives in the PPP appear misaligned in that FinTech lenders with widespread
indicators of misreporting made billions of dollars dispersing loans with apparently lax oversight
procedures. The FinTech organizations who facilitated such massive frauds should be thoroughly
investigated and not be allowed to engage in future SBA lending. Fifth, the increasing scale of misreporting
through time indicates that current penalty and enforcement systems are not effective. If the system is
not changed for future SBA lending programs, the most likely outcome is even more of the same.
Government agencies can assist this transparency goal by making detailed data available to the public.
Finally, much more needs to be done to align future incentives by prosecuting past financial crime,
particularly in the PPP. Tax payers have been defrauded billions of dollars and our analysis shows that less
than one in 10,000 loans that we identify has been prosecuted. To the contrary, over 90% of suspicious
loans are being forgiven by the SBA. Much more can be done. Without prosecuting the organizations and
networks of individuals who stole billions of dollars from U.S. taxpayers, justice is not served and wrong
incentives are created for the future. If history is any guide, these same individuals and organizations will
defraud tax payers again. This again is just a high-level summary, and many more detail can be found in
our paper. Thank you again for your attention to these important issues. I look forward to further
questions.
4