Follow the Money: Tackling Improper Payments, Hearing
Summary
A printed hearing, Follow the Money: Tackling Improper Payments, before the Subcommittee on Government Operations of the House Committee on Oversight and Reform, of March 31, 2022, Serial No. 117-75, with Gerald E. Connolly presiding as subcommittee chairman and Jody B. Hice as ranking member. Witnesses were Linda Miller of Grant Thornton, formerly of the Pandemic Response Accountability Committee, Scott Jensen, Rachel Greszler, Parker Gilkesson and Adrian Haro. Connolly's opening statement says that for Fiscal Year 2021, 86 high-risk federal programs estimated 7.2 percent of $3.9 trillion in outlays, $281 billion, were paid improperly; Hice's statement cites pandemic fraud estimates in the hundreds of billions. The hearing convened at 9:33 a.m. and adjourned at 11:42 a.m.
Summary drafted by a model from the document's text below and checked by script against that text before publication. It is a navigation aid, not a reading of what the document proves. Where AI is used
Full text
FOLLOW THE MONEY: TACKLING IMPROPER
PAYMENTS
HEARING
BEFORE THE
SUBCOMMITTEE ON GOVERNMENT OPERATIONS
OF THE
COMMITTEE ON OVERSIGHT AND
REFORM
HOUSE OF REPRESENTATIVES
ONE HUNDRED SEVENTEENTH CONGRESS
SECOND SESSION
MARCH 31, 2022
Serial No. 117–75
Printed for the use of the Committee on Oversight and Reform
(
Available at: govinfo.gov,
oversight.house.gov or
docs.house.gov
U.S. GOVERNMENT PUBLISHING OFFICE
47–264 PDF WASHINGTON : 2022
COMMITTEE ON OVERSIGHT AND REFORM
CAROLYN B. MALONEY, New York, Chairwoman
ELEANOR HOLMES NORTON, District of JAMES COMER, Kentucky, Ranking Minority
Columbia Member
STEPHEN F. LYNCH, Massachusetts JIM JORDAN, Ohio
JIM COOPER, Tennessee VIRGINIA FOXX, North Carolina
GERALD E. CONNOLLY, Virginia JODY B. HICE, Georgia
RAJA KRISHNAMOORTHI, Illinois GLENN GROTHMAN, Wisconsin
JAMIE RASKIN, Maryland MICHAEL CLOUD, Texas
RO KHANNA, California BOB GIBBS, Ohio
KWEISI MFUME, Maryland CLAY HIGGINS, Louisiana
ALEXANDRIA OCASIO-CORTEZ, New York RALPH NORMAN, South Carolina
RASHIDA TLAIB, Michigan PETE SESSIONS, Texas
KATIE PORTER, California FRED KELLER, Pennsylvania
CORI BUSH, Missouri ANDY BIGGS, Arizona
SHONTEL M. BROWN, Ohio ANDREW CLYDE, Georgia
DANNY K. DAVIS, Illinois NANCY MACE, South Carolina
DEBBIE WASSERMAN SCHULTZ, Florida SCOTT FRANKLIN, Florida
PETER WELCH, Vermont JAKE LATURNER, Kansas
HENRY C. ‘‘HANK’’ JOHNSON, JR., Georgia PAT FALLON, Texas
JOHN P. SARBANES, Maryland YVETTE HERRELL, New Mexico
JACKIE SPEIER, California BYRON DONALDS, Florida
ROBIN L. KELLY, Illinois VACANCY
BRENDA L. LAWRENCE, Michigan
MARK DESAULNIER, California
JIMMY GOMEZ, California
AYANNA PRESSLEY, Massachusetts
RUSSELL ANELLO, Staff Director
WENDY GINSBERG, Subcommittee Staff Director
AMY STRATTON, Deputy Chief Clerk
CONTACT NUMBER: 202-225-5051
MARK MARIN, Minority Staff Director
SUBCOMMITTEE ON GOVERNMENT OPERATIONS
GERALD E. CONNOLLY, Virginia, Chairman
ELEANOR HOLMES NORTON, District of JODY B. HICE, Georgia Ranking Minority
Columbia Member
DANNY K. DAVIS, Illinois FRED KELLER, Pennsylvania
JOHN P. SARBANES, Maryland ANDREW CLYDE, Georgia
BRENDA L. LAWRENCE, Michigan ANDY BIGGS, Arizona
STEPHEN F. LYNCH, Massachsetts NANCY MACE, South Carolina
JAMIE RASKIN, Maryland JAKE LATURNER, Kansas
RO KHANNA, California YVETTE HERRELL, New Mexico
KATIE PORTER, California
SHONTEL M. BROWN, Ohio
(II)
C O N T E N T S
Page
Hearing held on March 31, 2022 ............................................................................ 1
WITNESSES
Ms. Linda Miller, Principal, Advisory Services; Grant Thornton, Former Dep-
uty Executive Director, Pandemic Response Accountability Committee,
Former Assistant Director, Government Accountability Office
Oral Statement ........................................................................................................ 6
Mr. Scott Jensen, Chief Executive Officer and Vice President of External
Affairs, Research Improving People’s Lives, Former Director, Rhode Island
Department of Labor & Training
Oral Statement ........................................................................................................ 8
Ms. Rachel Greszler, Research Fellow, The Heritage Foundation
Oral Statement ........................................................................................................ 9
Ms. Parker Gilkesson, Senior Policy Analyst, Income and Work Supports,
The Center for Law and Social Policy
Oral Statement ........................................................................................................ 11
Adrian Haro, Chief Executive Officer, The Workers Lab
Oral Statement ........................................................................................................ 13
Written opening statements and statements for the witnesses are available
on the U.S. House of Representatives Document Repository at:
docs.house.gov.
(III)
INDEX OF DOCUMENTS
* Report, GAO-15-593SP, A Framework for Managing Fraud Risks in Fed-
eral Programs; submitted by Subcommittee Chairman Connolly.
* Business Insider, article, ‘‘A man was arrested after allegedly spending
$5 million in stolen COVID-19’’; submitted by Rep. Biggs.
* News report, ‘‘Biggest fraud in a generation: The looting of the Covid
relief plan known as PPP’’; submitted by Rep. Biggs.
* News reports, ‘‘Criminals have stolen nearly $100 billion in Covid relief
funds, Secret Service says’’; sbmitted by Rep. Biggs.
* News report, ‘‘Dr. Rand Paul Stops Democrats Attempt to Give Free
Money to Planned Parenthood’’; submitted by Rep. Biggs.
* Austin American-Statesman, article, ‘‘Fact-check Will prisoners also re-
ceive $1,400 stimulus checks’’; submitted by Rep. Biggs.
* CNBC News, news report, ‘‘Florida business owner got $3.9 million in
coronavirus relief money, allegedly spent’’; submitted by Rep. Biggs.
* News report, ‘‘Made out like bandits NY prisoners got $34M in federal
COVID stimulus funds’’; submitted by Rep. Biggs.
* NBC News, news report, ‘‘Man who used Covid funds on $57,000
Pokemon card gets 3 years in prison’’; submitted by Rep. Biggs.
* The New Jersey Monitor, article, ‘‘N.J. fisheries netted $2.4M in improper
COVID relief payments, watchdog says’’; submitted by Rep. Biggs.
* News report, ‘‘Stimulus turns political as SBA tries to claw back funding
from Planned Parenthood’’; submitted by Rep. Biggs.
* News report, ‘‘Trump Administration To Planned Parenthood: Return
Coronavirus Relief Funds’’; submitted by Rep. Biggs.
* Updated Data on Improper Payments; submitted by Rep. Biggs.
* Washington Examiner, article, ‘‘Planned Parenthood ‘illegally’ took $83
million in COVID stimulus: Senators’’; submitted by Rep. Biggs.
* News report, ‘‘Yes, Evers gave COVID relief money to Planned Parent-
hood - but it’s not used to fund abortions’’; submitted by Rep. Biggs.
* Questions for the Record: to Gilkesson; submitted by Chairman Connolly.
* Questions for the Record: to Jensen; submitted by Chairman Connolly.
* Questions for the Record: to Haro; submitted by Chairman Connolly.
* Questions for the Record: to Miller; submitted by Chairman Connolly.
The documents are available at: docs.house.gov.
(IV)
FOLLOW THE MONEY: TACKLING IMPROPER
PAYMENTS
Thursday, March 31, 2022
HOUSE OF REPRESENTATIVES
COMMITTEE ON OVERSIGHT AND REFORM
SUBCOMMITTEE ON GOVERNMENT OPERATIONS
Washington, D.C.
The subcommittee met, pursuant to notice, at 9:33 a.m., in room
2154, Rayburn House Office Building, and via Zoom; Hon. Gerald
E. Connolly (chairman of the subcommittee) presiding.
Present: Representatives Connolly, Norton, Davis, Sarbanes,
Lynch, Raskin, Porter, Brown, Hice, Keller, Clyde, Biggs, and
LaTurner.
Mr. CONNOLLY. The subcommittee will come to order.
Without objection, the chair is authorized to declare a recess of
the committee at any time.
I welcome everybody to the hearing which seeks to introduce—
I’m sorry, examine strategies the Federal Government could adopt
to better address improper payments. This is a subject of this sub-
committee going way back. In fact, one of my first hearings, I re-
member, was with a former chair of this subcommittee, Todd
Platts, talking about improper payments almost 14 years ago.
Hopefully things have improved, but we will see.
The chair now recognizes himself for my opening statement.
The Federal Government manages and allocates trillions of dol-
lars each year to programs that help individuals, families, and
businesses, whether that be Social Security, Medicare, crop insur-
ance for farmers, school lunch benefits, health insurance for chil-
dren, and a host of other essential programs, but that money does
not always go where Congress intends. For Fiscal Year 2021, 86
high-risk Federal programs estimated improper payments. Of their
$3.9 trillion in outlays, 7.2 percent were paid improperly, it is esti-
mated. That translates into $281 billion, more than any other Fis-
cal Year since 2003. Today’s hearing examines the causes of those
improper payments and explores solutions for reducing them.
The law defines an improper payment as ‘‘any payment that
should not have been made or that was made in an incorrect
amount, including an overpayment or an underpayment, under a
statutory, contractual, administrative, or other legally applicable
requirement.’’ Improper payments are an imperfect, but valuable,
way of measuring program integrity. In short, they help us to an-
swer the question, is the Federal Government distributing money
in the way Congress intended. If Federal improper payments were
low, the answer would clearly be ‘‘yes.’’ Unfortunately, they are not.
(1)
2
I should be clear. There are many causes of improper payments.
Some improper payments could be attributed to fraud, but many
are simply paperwork errors, for example, when an individual or
business accidentally checks a box on a form perhaps because of
confusing instructions, language barriers, or bad internet access,
and by checking that box, it affects the amount provided. Some im-
proper payments are even underpayments when the government
pays less than the amount for which they, in fact, qualified.
Unfortunately, more often than not, government does not know
the cause of any particular improper payment. Not knowing a pay-
ment’s error cause makes it difficult to distinguish fraud from any-
thing else. And without good data to help us identify root causes
of these improper payments, the government can’t make evidence-
driven decisions that enhance program integrity to prevent
fraudsters from gaining unlawful access to government services
and to ensure individuals and businesses access those services for
which they qualify. In short, bad data and deteriorating IT infra-
structure all too often makes it, in fact, near impossible to fix the
problem we see today without reinvigorating our IT investments.
History seems to agree.
Though improper payments have been a priority of Congress
since the beginning of the 21st century, they remain high and they
continue to grow as that chart, which I hope will be up, as that
chart shows.
[Chart.]
Mr. CONNOLLY. Thank you, Joshua. If you look at the screen, you
will see how improper payments have steadily grown since Con-
gress first required tracking and measurement under the Improper
Payments Information Act of 2002. As the figure shows, Congress
has enacted five bills into law that sought to identify and reduce
these improper payments in 2002, in 2010, in 2012, 2016, and
2019. Despite the efforts, improper payments are at an all-time
high, so we have gotten better at reporting, not so good at reduc-
ing.
According to the 2020 American Customer Satisfaction Index, the
Federal Government ranks among the bottom of all industries in
the United States in customer satisfaction. Public trust in govern-
ment remains near a historic low. As of April of last year, only 24
percent of individuals expressed trust in government. These low
scores are the result, all too often, of poor customer experience and
excessive improper payments. We can and must do better to en-
hance program integrity in the Federal Government and ensuring
Federal assistance goes to everyone Congress intended to receive it
and nobody we didn’t.
Clearly, things have to change. To help enact such change, I hope
today’s hearing will hit home three main points. First, we must
keep in mind our motivation for reducing improper payments. Put
simply, it is to ensure that money goes where Congress intends.
That means we must do everything we can to stop those elements
from stealing government funds, those criminal syndicates who are
organized to do that, and it also means we must ensure govern-
ment funds go to everyone Congress intends to reach, in an accu-
rate way. Second, there exists a false tradeoff in the minds of some
between preventing fraud and designing accessible, intuitive, and
3
equitable services. In fact, when done right, governments can de-
ploy modern technologies that prevent fraud and improve the cus-
tomer experience, designing more elegant and simple platforms by
which families, and businesses, and individuals can apply for and
access the services they need and for which they qualify. Govern-
ment programs can combat waste, fraud, and abuse while they
save and improve lives.
Finally, I want to highlight that good-quality data and the mod-
ern IT infrastructure that can support its collection sharing and
analysis are foundational to tackling improper payments. Data is
key to reducing paperwork errors and fraud without putting a bur-
den on the public. Using data as a solution is not a counter-visual
thing. Last July, the Office of Management and Budget and the
Pandemic Response Accountability Committee issued a payment in-
tegrity alert on the use of automation and data analytics to ‘‘sup-
port agency missions while mitigating payment integrity risks.’’
Using and matching datasets to identify potential improper pay-
ments, fraudulent or otherwise, is fundamental and nonpartisan.
We need the will to make it happen.
I plan to introduce legislation built upon these three principles.
The bill will establish an office dedicated to program integrity to
implement robust oversight of Federal agencies’ program integrity
efforts. Importantly, the office will shift from a compliance-based to
an action-oriented approach by requiring Federal programs with
significant risk of improper payments to implement proactive, data-
driven, and outcome-oriented anti-fraud controls. In addition, the
bill would ensure that agencies minimize the burden of their anti-
fraud controls on the public. By focusing efforts on data collection,
sharing, and analysis, the new office would enhance program integ-
rity in the highest-priority Federal programs without putting
undue burdens on those programs’ customers.
At its core, the bill would be about getting money to the places
Congress wants it to go and preventing that money from going to
places we don’t. Republicans or Democrats, I think, can all agree
that that is our ultimate goal. I look forward to working with my
colleagues on both sides of the aisle on this legislation and hope we
can pass it into law to get these numbers down.
I now recognize the ranking member for his opening statement.
Mr. HICE. Thank you so much, Chairman Connolly. I appreciate
you calling this long-overdue hearing. Republicans often are skep-
tical, I think it is fair to say, of government programs and how they
will work, and my Democratic colleagues often hold a different view
on these things. But we know with certainty that government as-
sistance will be totally ineffective if those funds never even reach
the intended recipients.
Improper payments have been a thorn in the side of our govern-
ment probably forever. The problem, however, is certainly getting
worse, and it was getting worse even before COVID. But the tril-
lions of dollars that were spent in pandemic assistance ushered in
a wave of fraud, quite frankly, the likes of which we have never
experienced before. Estimates literally range into the hundreds of
billions of dollars, and much of that is feared to have been lost to
organized crime in even countries like Nigeria, China, and Russia.
4
For all we know, American tax dollars could actually be helping to
fund Putin’s war against Ukraine.
Reports began to surface last year regarding the magnitude of
the problem, and yet here we are over a year since the American
Rescue Plan and another $2 trillion gone, signed into law, and this
is the first hearing we have had on this subject. If the state and
local funding of the American Rescue Plan is any guide, it appears
the Democratic plan to fight fraud is simply to let people spend it
any way they want to spend it: no rules, therefore, no fraud. Ear-
lier this month, the White House put out a fact sheet on what they
had done or what they were going to do around fraud and improper
payments. That is good. That is fine. That is great. But it talks
about how last May, they got a lot of people together from ARP,
the coordinator, PRAC, IGs, GAO, NSC, and others, in order to dis-
cuss governmentwide steps to prevent individuals from defrauding
public benefit programs.
But the question is what has been done. We don’t need more
meetings. We need action to stop the fraud and the abuse that has
been taking place. And, quite frankly with all, and I mean this sin-
cerely, with all due respect to every one of our witnesses that are
here today, frankly, Chairman, we should be hearing directly from
the Biden administration today. And then there is this that was
also in the fact sheet that I saw this morning: ‘‘Reestablished re-
spect for and transparency with the oversight community.’’ I guess
the result of that means more meetings, but in all honesty, give me
a break. We literally have no idea what is being done on anything
other than more talk that has been taking place.
This should be a top priority. This should be considered an emer-
gency for this administration. You know, appointing special pros-
ecutors is fine, and I am sure Democrats will talk about how they
have appropriated more money for inspectors generals, but, quite
frankly, that is like slamming the door on the barn after the horse
is already gone. The fact is the money is already gone. You might
get a little back, might get a little back, but for all intents and pur-
poses, the money is gone. There may have been funding for fraud
prevention in the American Rescue Plan and the omnibus, but the
question is how were those funds used. How will it be used? This
is basic oversight. This is our responsibility, and so far, as I can
tell, there has been no real oversight, nor is there any real over-
sight on the horizon.
Again, I would like to thank Chairman Connolly for at least pull-
ing together a conversation on this topic, and I am sure you are
under certain constraints as to who you can call as witnesses. But
I believe there is one thing that will be clear today is the tension
between fraud prevention and getting money to the recipients as
easily and quickly as possible. During the initial steps and stages
of the pandemic when we were all facing unprecedented action that
was needed to keep the economy afloat, it is understandable in that
type of circumstance that we would have erred on the side of get-
ting aid out. But as we will hear today, it does not have to be an
either/or choice.
There are actions that states could have taken, but they didn’t.
Why? We need to know. If we are talking about Federal dollars,
how much do states even care about program integrity, especially
5
if nobody is holding them accountable? There were certain things
that Federal agencies could have done, but they didn’t. Why not?
Was it a question of focus, was it a question of resources, or was
it a question of leadership? Congress first attempted to address im-
proper payments some 20 years ago at least, but those solutions ob-
viously are not working today. The pandemic laid bare the sorry
state of fraud prevention and program integrity at all levels of our
government. And to be very honest, this is true of Democratic ad-
ministrations, and it is also true of Republican administrations, but
something has to change.
Since I came to Congress, I have been amazed at how a simple
recitation of a problem is construed as insightful or is construed as,
in itself, problem solving. It is not enough just to talk about a prob-
lem. We have got to take action, and the reality is that agencies
don’t share data. Agencies don’t have access to data in many in-
stances, and you can’t do things across a whole of government.
There are no incentives. We could go on and on and on and on with
the problems that are out there, but, frankly, it is just admitting
defeat. We have literally swung wide the door of our Treasury and
invited our enemies to come in and help themselves, and if we are
not willing to fix the problem now, I fear that we will never be
ready to fix and address the problem. And, frankly, this is not an
issue to slip in halfway through the State of the Union. This is a
disgrace. It is an emergency. It is something that demands atten-
tion now. Chairman, we do hearings twice a year on FITARA. That
is fine, but this is an issue that needs to be highlighted twice a
year. We have got to do something about this problem.
I look forward to hearing from our witnesses today, and I appre-
ciate each of you for being here, but I really look forward to de-
manding answers from the administration and demanding action.
And with that, Mr. Chairman, I yield back.
Mr. CONNOLLY. I thank the gentleman. Let me now introduce our
witnesses.
Our first witness today is Linda Miller, principal of advisory
services at Grant Thornton. Formerly, Ms. Miller served as deputy
executive director of the Pandemic Response Accountability Com-
mittee, and as assistant director of the Government Accountability
Office where she worked over 10 years and specialized in fraud risk
and program integrity. Next, we will hear from Scott Jensen, chief
executive officer of Researching Improving People’s Lives, a non-
profit that works with governments to help them use data, science,
and technology to improve policy delivery. Formerly, Mr. Jensen
served as the director of Rhode Island’s Department of Labor and
Training.
Then we will hear from Adrian Haro, chief executive officer of
The Workers Lab——
Where is Adrian? I guess you are on virtually. Are you, Adrian?
Yes, there you are. OK.
—that invests in new ideas for the modern worker. And Adrian
prefers to be addressed by his first name, ‘‘Adrian.’’ Then we will
hear from Rachel Greszler, senior research fellow in economics,
budgets, and entitlements at the Heritage Foundation. And finally,
we will hear from Parker Gilkesson, senior policy analyst in the In-
come and Work Supports at the Center for Law and Social Policy.
6
So if you, Adrian, could unmute yourself and raise your right
hand. If you all would rise and raise your right hand to be sworn
in.
Do you swear or affirm that the testimony you are about to give
is the truth, the whole truth, and nothing but the truth, so help
you God?
[A chorus of ayes.]
Mr. CONNOLLY. Thank you. Let the record show the witnesses all
answered in the affirmative.
Without objection, your written statements will be entered into
the record in full.
With that, Ms. Miller, you are now recognized for your five min-
utes of summary testimony. Welcome.
STATEMENT OF LINDA MILLER, PRINCIPAL, ADVISORY SERV-
ICES, GRANT THORNTON, FORMER DEPUTY EXECUTIVE DI-
RECTOR, PANDEMIC RESPONSE ACCOUNTABILITY COM-
MITTEE, AND FORMER ASSISTANT DIRECTOR, GOVERNMENT
ACCOUNTABILITY OFFICE
Ms. MILLER. Thank you, Chairman Connolly, and Ranking Mem-
ber Hice, and esteemed members of the committee. It’s a pleasure
to be with you this morning.
I’m a principal at Grant Thornton where I lead our Fraud and
Financial Crimes Practice, but as Chairman Connolly mentioned,
today I’m going to discuss my experience both an executive at GAO
as well as my time spent being deputy executive director of the
PRAC.
Much has been written about the unprecedented fraud that oc-
curred during the pandemic. I spent almost 20 years in the anti-
fraud space, and there’s no question that this was the largest fraud
event in our Nation’s history. It comes as no surprise, though, to
those of us who have been working in the anti-fraud industry. We
live in the age of the data breach where criminals now have data
on literally every American citizen, and today’s fraud actors are a
sophisticated adversary. They take data purchased on the dark
web, often for less than a dollar a piece, and they apply advanced
tools and techniques: automated bots, artificial intelligence, and so-
phisticated counterfeiting tools, deep fake technology. They’ve
honed their craft.
And how do Federal and state agencies fight this adversary?
With very limited data, with rudimentary analytics, with IT sys-
tems that run on software developed in the 1960’s, with next to no
leadership commitment. If the pandemic fraud alarmed people,
that’s simply because they haven’t been paying attention. The fleec-
ing that we saw was the direct result of years of inattention from
senior leadership on the need for proactive fraud risk management.
Three key areas must urgently be addressed. The first is data.
By and large, agencies are struggling to effectively use data to fight
fraud and improper payments. Even the most basic use of data—
data matching—is not being done widely. California paid $810 mil-
lion in pandemic unemployment benefits to 45,000 prisoners be-
cause they hadn’t developed the capacity to match prisoner data
with their claims data. And 11 years after Treasury created the Do
7
Not Pay System, today, in 2022, that system still lacks the Social
Security Administration’s full death data or prisoner data.
Let that sink in. The fraudsters have detailed information about
you—your address, your date of birth, social security numbers, the
passwords to your most sensitive accounts—and Treasury cannot
even access data to find out if an applicant is alive. Our adver-
saries are using our own data against us. Those who say we must
choose between fraud prevention and privacy are creating a false
dichotomy. We have the technology to do both, but agencies need
the will to make fraud prevention on par with privacy protection.
The second area is leadership priority, which, I am afraid to say,
is woefully lacking in this area. Agency leaders have few, if any,
real incentives to prioritize integrity over the delivery of benefits.
It makes sense because they only get an angry call from a citizen
if a benefit has not been paid. Nobody calls to talk about fraud.
When I was at GAO, agency officials directed us to the OIG when
we wanted to talk about fraud risk management. They simply did
not see fraud prevention as their responsibility.
The third area is technology where government badly lags its pri-
vate sector peers. The banking sector has been using tools, like
Know Your Customer, that pull real-time, updated information in-
stantly on applicants and customers, and these tools have been in
place for years. What we saw during the pandemic was an almost
total removal of the guardrails. We asked people if they were eligi-
ble for a benefit, and when they said ‘‘yes,’’ we gave it to them.
Congress and the administration made the calculated decision that
the need justified the profound risk but make no mistake. There
was fraud and improper payments before the pandemic, and there
will be even more fraud and improper payments after the pandemic
if action is not taken.
The current approach is simply not working. There have been
five iterations of legislation focused on improper payments over the
last 20 years, and during that time, the improper payment rate has
steadily risen. As currently written, the Payment Integrity and In-
formation Act creates burdensome compliance requirements while
doing little to address the root cause of the problem. Most improper
payments are caused by a lack of data verification, yet when we
ask agencies to do check-the-box compliance, we fail to ask them
to address the root cause of the problem.
First and foremost, agency leadership needs to prioritize integ-
rity of the funds they disperse almost as much as they prioritize
getting those funds out the door. This imbalance must be ad-
dressed. And agencies need to proactively address the root causes
of the problem, not with more burdensome compliance activities,
but by using data and analytics that verifies the identity and eligi-
bility of an applicant: technology like device fingerprinting, and
threat intelligence from deep and dark web sources, and other
tools. These are complex problems, and there are no easy answers.
Therefore, a dedicated anti-fraud office must be created to work on
these issues solely. The United Kingdom established such an office
in 2018 and has seen great success.
Fraud is not going away with the end of the pandemic. Indeed
these fraud actors are only emboldened, and they will continue to
8
target these ill-prepared agencies. The time for action is now.
Thank you.
Mr. CONNOLLY. Thank you, Ms. Miller.
Mr. Jensen, you are now recognized for your five-minute opening
statement.
STATEMENT OF SCOTT JENSEN, CHIEF EXECUTIVE OFFICER
AND VICE PRESIDENT OF EXTERNAL AFFAIRS, RESEARCH
IMPROVING PEOPLE’S LIVES, AND FORMER DIRECTOR,
RHODE ISLAND DEPARTMENT OF LABOR & TRAINING
Mr. JENSEN. Thank you, Mr. Chairman and members of the com-
mittee. It is an honor to be here. My name’s Scott Jensen. I’m the
CEO of Research Improving People’s Lives, but I think I am here
today really to talk about my experience as the Rhode Island direc-
tor of Labor and Training during the pandemic, leading the unem-
ployment insurance effort there. In terms of leadership, that is all
we did for two years. As the director, I led it right from the front,
and I learned a lot, and I wanted to share just a couple of things
to add some color to great ideas that, you know, I’ve heard today.
To begin with this, it is not an either/or—either fraud or paying
folks. It has to be a both/and, and the one thing about my experi-
ence in Rhode Island’s pandemic was it was, I think, the hardest
C-minus I’ve ever earned in my life. That was a very, very difficult
time, and nobody, none of these programs were going to make it
through in one piece just because of the massive scale of what we
had to do during the pandemic. Five thousand people. It was the
biggest week Rhode Island had ever had in unemployment insur-
ance, and we were doing that every day for months: 5,000 new peo-
ple, 5,000 new people.
So the central challenge that I think we faced that allowed us
to get a C-minus, and make it through the pandemic, and give de-
cent service but also do a fairly decent job with fraud, was that,
you know, we used UI data, the data the UI program itself gen-
erates, in order to improve the administration of the program. And
let me try to explain that a little bit with some specificity.
So the first thing, when we talk about technology and the old
technology that UI programs are faced with, is absolutely true, of
course. New technology is a necessary, but not sufficient, condition
to solve the problems that we’re looking at. You can’t just buy an
off-the-shelf new computer, put it in, and this problem is going to
go away. So let me give you some context about how UI works and
how we might do this.
First off, there are three steps to unemployment insurance. Very
simple. One, get information from somebody so you can figure out
whether they are eligible or they’re not eligible; two, use that infor-
mation to make a judgment about the benefit; and three, either pay
it or don’t pay it. And while those three things seem rather
straightforward, they’re hard, and in some instances, it’s very dif-
ficult. And there’s tough cases and you need a real expert to take
a look at that, and in other cases it’s pretty straightforward: a
straightforward ‘‘yes’’ or ‘‘no.’’
What we faced in Rhode Island and everywhere else was that ef-
fort, when you’re relying on people to make judgments like that, it
doesn’t scale. So you have a lot of new claimants coming in, and
9
you can’t hire quick enough to train folks quick enough, to get
them to be able to keep up with the volume, so then you begin to
get swamped. You know, in all of your states, I’m sure you have
a lot of experience with swamped UI programs.
So what we tried to do, and it was a little naive, we tried to at-
tack both fraud and deal with pain quickly. With fraud, when
somebody comes to apply for unemployment insurance, they leave
a footprint in the metadata of your system, and if you can distin-
guish between the fraudster and the right person by using sophisti-
cated artificial intelligence and other techniques, it’s like probable
cause. You can really scrutinize somebody who’s presenting as if
they’re a fraudster, and you can be a little bit more sure that some-
body else who’s presenting normally, you can get them paid
quicker. So that was broadly the strategy on the fraud side. And
on the payment side, we were trying very hard to distinguish be-
tween a clean claim, which is an obvious ‘‘yes’’ or ‘‘no,’’ and a com-
plex claim that is going to require a person to solve it.
Data is crucial to do either of those things, and they’re both an
enormous undertaking. If you don’t have data, you can’t make the
difference between either one. And the biggest problem we had dur-
ing the pandemic, we’re one of those states with those old AS400
computer systems from the 1980’s. In order to get data out of that
thing, you had to code it from scratch, so a really, really inefficient
way, impossible really, to do continuous improvement and change
the way you’re re administering these benefits.
So it was clear to——
Mr. CONNOLLY. If you could sum up, Mr. Jensen. Your five min-
utes is up.
Mr. JENSEN. The summary is simply we have to use data better,
and we need the technology to do it. But then the further step is
continuous improvements so we can achieve the kind of customer
service we’re all getting used to in the private sector.
Mr. CONNOLLY. Thank you.
Mr. JENSEN. Yep.
Mr. CONNOLLY. And I got to admit, the image of a footprint in
the metadata system will stick with me.
Mr. JENSEN. Good.
Mr. CONNOLLY. OK. Adrian, you are recognized, Mr. Adrian
Haro, who prefers to be known by his first name, ‘‘Adrian.’’ Adrian,
you are recognized for your five-minute opening statement. Adrian,
you need to unmute yourself. Adrian, we cannot hear you. All right,
Adrian. Let’s try to work that technical problem out, and we will
go to our next witness.
Ms. Greszler, you are recognized for your five-minute opening
statement. Welcome.
STATEMENT OF RACHEL GRESZLER, RESEARCH FELLOW IN
ECONOMICS, BUDGET AND ENTITLEMENTS, THE HERITAGE
FOUNDATION
Ms. GRESZLER. Good morning and thank you for the opportunity
to be here today.
We’ve heard about the recent rise in improper payments, but I
wanted to highlight this graph here to help convey the magnitude
of that increase.
10
[Chart.]
Ms. GRESZLER. We have here a chart between 2005 and 2021, so
on the bottom is GDP growth. That grew 76 percent, the green line
there. Total program outlays grew by 221 percent. That’s the green
line, so those are the government payments. The red line sky-
rocketed 633 percent. That’s improper payments, so improper pay-
ments grew at eight times the rate of GDP growth. If improper
payments had their own budget category, it would’ve been the Fed-
eral Government’s fifth largest expenditure in 2021.
Multiple factors have contributed to this rise, including the in-
creased use of online government services, the failure of agencies
to utilize sufficient security measures, and the pure enormity of
government transfer payments. Three things are crucial to reduc-
ing improper payments. First is to identify that people are who
they say they, second is to make sure that people are eligible for
the benefits they claim, and third is to minimize the number of
government programs that individuals have to rely on for their or-
dinary needs.
The Pandemic Unemployment Insurance programs, which will
likely go down as the biggest fraud in U.S. history, provides some
important lessons. In March 2020, Congress passed a massive ex-
pansion in state unemployment benefit programs, and it was sup-
posed to last for only four-and-a-half months but instead lasted for
18. These programs were highly problematic. The $600 bonus pay-
ments on top of existing state payments created a high-value target
for fraud. Benefits were extended to people who weren’t part of the
unemployment system and for whom it was nearly impossible to
verify their eligibility. And then broad qualifications meant that
nearly anybody could come up with a seemingly legitimate claim.
These expansions opened the floodgates to woefully outdated and
ill-prepared state UI systems, inviting massive fraud and abuse,
and allowing international crime rings, even state-sponsored Chi-
nese hackers, to steal Americans’ identities and taxpayers’ money.
I estimate that the pandemic unemployment benefits likely ex-
ceeded a 40-percent improper payment rate. That’s based on the
data that shows that we paid out 1.365 billion unemployment
checks, but had we simply paid out 100 percent of people who are
actually unemployed, that would’ve cost—required 807 million
checks. So what that means is those excess checks amounted to
$357 billion of taxpayers’ money that went above and beyond actu-
ally covering every single person who was unemployed. That is a
$2,700 cost per household in the U.S. just for the pandemic unem-
ployment fraud. Unfortunately, there are many other government
programs that regularly impose massive improper payment bills on
Americans. The earned income tax credit, Medicaid, and CHIP all
have improper payment rates between 22 percent and 32 percent,
and there’s a total of $123 billion in improper payments in those
three programs alone.
So what should be done? For starters, government programs
should be required to follow security measures similar to the NIST
standards. For example, Arizona had a time when it was receiving
more unemployment insurance claims than it had people in its en-
tire work force, but after it partnered with a NIST-certified pro-
vider, its claims fell by 99 percent. There’s a 2019 OMB memo that
11
requires Federal agencies to utilize these standards, but most are
not doing it. Next, the Federal Government should optimize secure
information sharing across government agencies through its Do
Not Pay data base. Lawmakers need to improve eligibility
verification and maintain a competitive market for cybersecurity so
that taxpayers can benefit from the most responsive and innovative
systems out there.
Perhaps most important, however, is the need to limit the size
and the scope of government programs. In 2022, the U.S. will
spend $3.9 trillion in transfer payments. That is $4 out of every $5
tax dollars that we’ll collect. Constantly expanding entitlement pro-
grams without any consideration of what they’ll cost is
unsustainable, and taking people’s money and requiring them to
jump through bureaucratic hoops to get some of it back from pro-
posed government programs, like government-directed childcare,
government-determined paid family leave, or purchasing govern-
ment-approved electric vehicles, is inefficient, and it will leave peo-
ple with lower incomes and less control over their lives.
A few steps to reduce the Federal Government’s tax and transfer
footprint and empower people instead of bureaucrats include elimi-
nating tax credits in favor of lowering overall tax rates, enacting
universal savings accounts to make savings simpler and more af-
fordable, rejecting new entitlement programs and reforming exist-
ing ones.
Thank you.
Mr. CONNOLLY. Thank you very much, and you had time to
spare. Great job.
Have we solved with the problem Adrian? Have we solved the
problem?
Voice. Not yet.
Mr. CONNOLLY. Thank you. Ms. Gilkesson, you are recognized for
your five-minute opening statement.
STATEMENT OF PARKER GILKESSON, SENIOR POLICY ANA-
LYST, INCOME AND WORK SUPPORTS, THE CENTER FOR
LAW AND SOCIAL POLICY
Ms. GILKESSON. Good morning, Chairman Connolly, Ranking
Member Hice, and distinguished members of this committee.
Thank you for inviting me today. My name is Parker Gilkesson,
and I’m senior policy analyst at the Center for Law and Social Pol-
icy. My experience as a North Carolina case worker led to my com-
mitment to become a researcher/advocate and expert on SNAP,
TANF, and Medicaid. CLASP is a national, nonpartisan, anti-pov-
erty nonprofit advancing policy solutions for people with low in-
comes. My testimony summarizes a detailed paper I published ana-
lyzing the racialized history of fraud in SNAP.
I’d like to make four key points. One, the intense focus on fraud
in SNAP is tied closely to a history of racist stereotypes; two, ac-
tual SNAP fraud is rare; three, the focus on fraud makes programs
less effective at reducing hunger; and four, Federal and state agen-
cies have many practical fixes to reduce the barriers to SNAP
caused by unnecessary fraud provisions.
Healthcare, food, secure housing, and a livable wage are basic
human needs, and seeking the help you need to succeed is a state-
12
ment of human dignity and justice. However, coded language, dog
whistling, and racist stereotypes have reinforced the lie that folks
receiving public benefits are exaggerating how poor they are and
are likely committing fraud. The perception of fraudulence based
on poverty and race has a long history here. Anti-black images
borne out of slavery labeled black people as biologically inferior, in-
herently lazy, promiscuous, and untrustworthy. These harmful
tropes were used to justify the forced labor of black people and are
still reflected in many U.S. policies today.
In the last 50 years, policymakers have used tropes like ‘‘welfare
queen’’ to insinuate that people experiencing poverty, especially
black people, are fraudulent and must be forced to work. Legisla-
tors have used this dog whistle to justify benefit cuts while increas-
ing spending to protect program integrity. Even those promoting
expansions for public benefits have reinforced stereotypes by em-
phasizing how these programs would only benefit the so-called de-
serving poor.
Fraud is quite rare, contrary to media portrayals, yet despite
that rarity, Federal and state legislators prioritize fraud prevention
with millions in grants and targeted funding for program integrity.
States even have tip lines for the public to report suspected fraud,
although most reports are based on personal vendettas or a belief
that a person using SNAP is too nicely dressed to qualify for bene-
fits. Also, due to complex SNAP rules, many honest mistakes get
lumped in with fraud because recipients often misunderstand re-
porting rules. Eligibility workers are often better trained to look for
fraud than to provide trauma-informed care or refer families to
other necessary assistance based on the many challenges of living
in poverty. This isn’t surprising as Federal rules penalize states for
any benefits granted in error but not for failing to help people in
need. Also, difficult and time-consuming applications meant to
catch people attempting to commit fraud often present barriers to
applicants and caseworkers, reducing access and increasing stigma
and shame.
Of course no program can survive if it doesn’t take abuses seri-
ously. However, when actions taken to improve program integrity
have a strong negative and racially skewed impact that decreases
access to basic life necessities, policymakers have a moral obliga-
tion to find less harmful ways to fight fraud. They also must repair
the harm of these actions on innocent recipients’ dignity and trust
in government. Recipients accused of fraud are often considered
guilty until proven innocent. While they’re supposed to be able to
defend themselves in a hearing, they aren’t entitled to representa-
tion, and some states have coerced people into signing away their
rights by threatening criminal prosecution.
Historically, anti-hunger advocates have been afraid to criticize
the negative consequences of the focus on program integrity and its
disproportionate impact on people of color for fear of being accused
of defending fraud or legitimizing racist tropes. However, we must
discuss and address fraud, program integrity, and the over policing
of people experiencing poverty to achieve policies that reflect eq-
uity, trust, and truth, instead of mistrust, mistreatment, and sys-
temic oppression.
13
Here are just a few of the equitable solutions state and Federal
policymakers should try: simplifying SNAP rules so clients under-
stand what changes to their income need to be reported, especially
for the growing share of workers in the gig economy. Many im-
proper payments are caused by confusion and honest mistakes; re-
thinking quality controls so we measure timeliness and accuracy
while also reaching all eligible households and ensuring customer
service satisfaction; providing oversight and accountability to states
for incorrect determinations of fraud sanctions and overpayments,
as well as for violations of recipients’ right to due process.
For more detailed research, I would like my report to be officially
submitted into the congressional record. Thank you, and I look for-
ward to your questions.
Mr. CONNOLLY. Without objection, it is so ordered.
Mr. CONNOLLY. Perfect timing. You had one second left. Thank
you, Ms. Gilkesson.
[Laughter.]
Mr. CONNOLLY. Adrian, are you ready now?
Adrian HARO. I believe I am, Chairman. Can you hear me?
Mr. CONNOLLY. Yes, you got it. You are recognized for your five-
minute opening statement.
Adrian HARO. OK. Terrific.
STATEMENT OF ADRIAN HARO, CHIEF EXECUTIVE OFFICER,
THE WORKERS LAB
Adrian HARO. Good morning, Chairman Connolly, Ranking Mem-
ber Hice, and all the members of this Subcommittee on Govern-
ment Operations. My name is Adrian Haro, and I’m the chief exec-
utive officer of an organization called The Workers Lab. Thank you
for the opportunity to share how our work is contributing to the
important enterprise of tackling improper payments.
So many of the ways that we serve and take care of workers in
this country are outdated and exclusive, so we need new ideas
about how to better serve more and every kind of worker in this
country. That’s why at The Workers Lab we are on a mission to
give new ideas for and with workers a chance to succeed. Over the
last year, one of those ideas has been about how automating the
process of verifying income in the administration of unemployment
benefits could both increase the efficiency of the system and make
it more equitable and accessible for non-traditional gig and contract
workers that have historically been excluded. And what we’re
learning is that automating income verification in the administra-
tion of unemployment benefits can be a win-win, a win for gig
workers in that it can make the process of applying for unemploy-
ment benefits faster and more user-friendly, and a win for govern-
ment in that it can increase the efficiency of the process and allow
for better service delivery to its constituents.
Our learning journey started at the beginning of the pandemic
when, for the first time ever, gig workers were made eligible for
unemployment benefits, a tremendous and novel achievement but
one that presented many administrative challenges for states all
across the country. An estimated 57.3 million Americans generate
income from gig work, which represents 35 percent of the U.S.
work force and is expected to grow to 50 percent by 2025. One of
14
the biggest challenges for states was how to make sense of and
verify income from gig work, which can often be generated from
multiple sources at multiple times, sometimes even within the
same day. Our unemployment system was simply not designed to
verify income that looks like this. So we brought gig workers to-
gether with Steady, a platform that helps give workers track, un-
derstand, and optimize their income to develop what we call an In-
come Passport to help fix this problem.
At The Workers Lab, we believe deeply in user-centered design
and that any innovation for workers should be done in partnership
with workers. The Income Passport automates an otherwise man-
ual process for verifying income in unemployment benefits applica-
tions. Rather than relying on the often complex and sprawling
paper trail that documents income generated from gig work, it in-
stead pulls that income data directly from trusted financial
sources. We tested the Income Passport with hundreds of workers
in Alabama and Louisiana to demonstrate that automating income
verification can make the process of administering unemployment
benefits easier and more efficient for government staff and gig
workers.
Here is what we are learning about how automating the process
of verifying income can help mitigate improper payments. First, the
Income Passport reduces the potential for fraud. Because the solu-
tion safely and securely pulls income data directly from trusted
sources and then transmits that income data directly to the state,
it leaves very little room for fraudulent activity. Initial findings
show that of applicants that chose not to use the passport, 90 per-
cent of them were likely bad actors. Second, the Income Passport
increases accuracy in the determination of benefit amounts for the
state in that it has access to a trusted and tailored record of in-
come generated from gig work, and for gig workers in that they
have unprecedented access to see past income from multiple
sources and, thus, more control over how their income is reported.
Third, the Income Passport optimizes efficiency for states in that
it automates an otherwise manual process, reducing the potential
for human error, reducing processing time from as much as half an
hour to 1 to 2 minutes and allowing for benefits payouts in less
than 24 hours, and for workers in that it simplifies the daunting
task of organizing income from gig work, thus enhancing the qual-
ity of the data submitted to the state. Finally, and perhaps most
importantly, we’re learning that the Income Passport holds promise
beyond temporary pandemic assistance programs to automate the
verification of income from gig work across the entire menu of pub-
lic benefits programs in states all around the country.
The Federal Government can do three things to help states make
this happen. First, it should use its resources to evaluate potential
technology solutions and support states that want to test them,
providing funding and advice so that states can take advantage of
innovation. Second, it can also establish policy guidelines and re-
quirements across Federal agencies that incentivize states to adopt
more innovative worker-centered technologies. And third, it can
share best practices and successful technologies across states and
help it to be adopted at scale through funding and other incentives.
And, of course, The Workers Lab stands ready and willing to help.
15
Thank you again for this opportunity, and I look forward to your
questions.
Mr. CONNOLLY. Wow. You had five seconds left. Great job, Adri-
an. Thank you.
Adrian HARO. Thank you, Chairman.
Mr. CONNOLLY. We now enter into member questioning. The
chair recognizes the distinguished Congresswoman from the Dis-
trict of Columbia, Eleanor Holmes Norton, for her five minutes of
questioning. Welcome, Eleanor.
Ms. NORTON. Hi. I appreciate the chair, my good friend, my next-
door-neighbor recognizing me.
Maybe we need—we have been thinking about improper pay-
ments all wrong. If solving the problem of improper payments
means ensuring that government benefits, like Social Security and
Medicare, military pay, reach those who Congress intended, per-
haps we should focus more on those who cannot even access these
benefits. Let me use a couple of examples. Nee Hall from Mount
Hall, New Jersey was a victim of fraud and could not access unem-
ployment insurance for seven months because the system locked
him out. Refavo from Paris, Tennessee, a 45-year-old retail worker
who worked in both New Jersey and Tennessee, got laid off in July
2020. He is still waiting for the state of Tennessee to send the state
of New Jersey information on his earnings so he can access benefits
he qualifies for.
Mr. Jensen, you have seen firsthand cases where incorrectly
checked boxes, language barriers, agency processes and operations,
and confusing program applications prevented an individual from
accessing a benefit. In your experience, who are the individuals and
families most hurt by byzantine applications or other barriers to
program access?
Mr. JENSEN. Thanks for the question. I think that in the pan-
demic, one thing that we saw and so many people heard across the
board: folks who couldn’t use technology well, folks who weren’t fa-
miliar with the byzantine structures that you are describing be-
cause they are byzantine. But we have to, as a number of folks
have said here on this panel, focus on a user-centered design when
we are designing these programs because our mission has to be
right down the middle.
We have to make it easy and straightforward to get access to
benefits when you are deserving of them, when you qualify for
them, and keep others who aren’t qualifying from getting paid. And
that that really requires a lot more thinking about how we are
going to make sure that folks who, just because they were confused
by the technology, just because they were defrauded and had to
wait for months and months for benefits. That is not acceptable.
Ms. NORTON. Ms. Miller, does the Federal Government even
know how many people are eligible for benefits and cannot access
them?
Ms. MILLER. No, the government has not undertaken any effort
to try to estimate how many people would need to access benefits.
That would be a very difficult calculation, I think, for the govern-
ment to even try to undertake.
Ms. NORTON. I guess it would be. So, Adrian, in your testimony,
you stated that your projects create a win-win for workers and the
16
government. Can you talk a bit more about what you mean by
‘‘win-win?’’
Adrian HARO. Absolutely. Thank you for the question. I guess I
mean that what we have demonstrated in this project is that you
can actually do both, right: increase access to public benefits, and,
at the same time, also increase efficiency inside government simul-
taneously. Like it has been said before today, it is a false tradeoff
to say that you have to sacrifice efficiency to increase access. I be-
lieve that doing so can stifle innovation in the public sector, and
we should be doing everything we can to encourage it.
Ms. NORTON. Thank you very much. I yield back.
Mr. CONNOLLY. Thank you. Thank you, Ms. Norton.
Mr. Hice, you are recognized for your five minutes of questioning.
Mr. HICE. Thank you, Mr. Chairman. Ms. Greszler, let me start
with you, and it may be kind of a loaded question because there
is a lot to answer within this. But it is no secret the rampant fraud
came about through the pandemic programs, but also unemploy-
ment insurance, Medicaid, and a host of other things. Are there
any actions that the Federal Government could be taking that they
just simply are not?
Ms. GRESZLER. Yes, I think there are definitely standards out
there that exist that say what are the right steps the program
should be taking. They are simply not using them. I mean, they
have been directed to do this, and yet they are not. And I think
this comes down to an accountability problem because, at the end
of the day, if there is no consequence for not reducing your own im-
proper payments or staying within a limit, then the programs are
not going to take them. I mean, in the UI payments, there were
multiple problems there, and certainly the Federal Government put
mandates on the states that they weren’t capable of enforcing those
identity checks and those eligibility checks.
Looking to going forward with you UI, I think it would be very
problematic to open this up to a group of workers who you really
can’t make those proper eligibility checks there. For those who
don’t have an employer, there is no verification in the process.
Mr. HICE. OK. So they have been directed to take certain steps,
but they have not, and there is no accountability. Why?
Ms. GRESZLER. Well, and this gets to the nature of government
programs. A lot of that $3.9 trillion is entitlement programs. The
money is essentially there no matter what. It doesn’t matter how
many claims are made and improper payments come out of the
door, there is unlimited access to taxpayer money even if it is def-
icit finance. You know, if you tell me I need to lock my door so that
my money and my things don’t get stolen, but I don’t lock it and
somebody comes in and robs me, but by the time I get back, all my
money and my things have been replaced, I am not going to lock
my door in the future. And so we need to have accountability in
place. The standards exist, but if nothing happens when the pro-
grams don’t meet those standards, we are not going to get to a bet-
ter solution here.
Mr. HICE. This is so infuriating to hear this kind of stuff. What
needs to happen for our agencies to care? I mean, what is behind
the lack of attention being directed toward fraud?
17
Ms. GRESZLER. It is the lack of accountability. You know, you
here in Congress are not in charge of micromanaging those agen-
cies. You know, administrations are the ones that are in charge of
running the programs, and they are not taking the measures they
should, but what could you do is, Congress, you have the power of
the purse. And so I would recommend that there need to be some
limits placed on there, whether it is setting a pathway toward re-
ducing improper payments and actually having penalties that hap-
pen if they don’t achieve those standards.
Mr. CONNOLLY. Would the gentleman——
Ms. GRESZLER. Unless there is a consequence——
Mr. CONNOLLY. I am sorry.
Ms. GRESZLER. Go ahead.
Mr. HICE. I will yield.
Mr. CONNOLLY. If the gentleman would yield without penalty, if
we could freeze his time just for a second. I am also infuriated
when I think about it, as Mr. Hice is. Ms. Miller suggested that
one of the problems we face is that managers don’t see it as their
responsibility. That is someone else’s problem, so I don’t have to
worry about improper payments, whether it be fraud or just mis-
takes. That goes to the IG or somebody else.
Ms. GRESZLER. Mm-hmm.
Mr. CONNOLLY. Is that, from your point of view, a big part of the
problem or at least part of the problem, and is that corrected some-
how by our building in evaluation criteria so that, no, it is your
problem. It isn’t just someone else’s problem. If the gentleman
would allow me——
Mr. HICE. Yes.
Mr. CONNOLLY [continuing]. Because I think this is consistent
with where you are headed to. Thank you. Thank you for yielding.
Ms. GRESZLER. Yes. I mean, at the end of the day, if whoever is
in charge of that agency knows that there are going to be con-
sequences—perhaps it is their job, perhaps it is a reduction in
funding, and then they can’t actually perform the service that they
are supposed to do—then they are going to do more to reduce those
improper payments. And it is unique across every program. It is
not to say that there is one set of rules that should apply holis-
tically because Medicaid is different than SNAP benefits, than the
EITC, or the UI. Most agencies and programs need to have the
flexibility to determine what are the right steps that they need to
take. They have different eligibility criteria. They have different
people they are serving. They have different ways of making the
payments. So that flexibility still needs to be there, but it is the
accountability that is missing.
Mr. HICE. Thank you, Mr. Chairman. Listen, accountability
starts right here in this committee, Mr. Chairman. We have got to
do our business of oversight and hold accountable these agencies
that are not doing their job to protect the American tax dollars.
This is inexcusable to me. Ms. Miller, let me just ask your opinion.
Is there anything you would add to this in terms of what are some
of the obstacles that are keeping our agencies from protecting
funds and preventing fraud?
Ms. MILLER. Yes. Look, I think people respond to whatever they
are being measured on, right, and so most Federal agency leaders
18
see themselves as their job is delivering the benefits. When you ask
someone what is my mission in this agency, it is delivering bene-
fits, getting benefits out the door. If you ask them is my mission
to ensure that only eligible beneficiaries get it, they don’t see that
as much as their mission. They see it as, sure, of course, it is nec-
essary. But if you are incentivizing people to adjudicate a certain
number of applications in a given period of time, they aren’t
incentivized to slow down and look and see if this application looks
like fraud and go check some things, because at the end of the day,
it is the number of applications that they approved. So we have to
change the incentive structure from the top all the way down.
Mr. HICE. Absolutely. OK. So while you were at the PRAC, was
there anything, any steps from Federal agencies that moved in the
direction of preventing fraud and abuse?
Ms. MILLER. Yes. The Small Business Administration, when they
went into the second round of PPP funding, they started employing
some Know Your Customer tools that the banks employ, and that
did help with identity theft. I think that Ms. Greszler is right.
Identity theft was the biggest fraud issue in the pandemic, and so,
you know, Know Your Customer, the concept that banks use, is an
effective tool to identify whether someone is who they say they are.
Identity theft is a growing problem in fraud, but I want to make
sure that we are clear that the pandemic fraud is in the past, and
we can talk about all of the problems that happened, but really, we
are looking forward. How are agencies going to solve these prob-
lems going forward? They are going to have to take a much more
proactive approach using data.
Mr. HICE. Well, it is not in the past to ignore it. We can’t go
down that path, but thank you, Mr. Chairman. I yield back.
Mr. CONNOLLY. I thank the chair—I thank the gentleman and
promise, let’s find some common ground to look at metrics that ac-
tually incorporate responsibility and accountability for improper
payments across the board so that managers have an incentive to
pay attention to this, because right now they don’t have much, as
the data would show. I thank the gentleman.
The gentleman from Illinois, Mr. Davis, is recognized for his five
minutes of questioning.
Mr. DAVIS. Thank you, Mr. Chairman, and I also want to thank
all of the witnesses for their informative testimony.
You know, during this year’s State of the Union Address, Presi-
dent Biden announced his plan to appoint a chief prosecutor to
focus on the most egregious forms of pandemic fraud by criminal
syndicates. The White House published an estimate that between
Fiscal Year 2020 and Fiscal Year 2021, the governmentwide im-
proper payment rate rose from 5.6 percent to 7.2 percent. Clearly,
addressing fraud then is important. Ms. Miller, let me ask you,
what steps can Federal managers take to achieve a more strategic,
risk-based approach to managing fraud and developing effective
anti-fraud control?
Ms. MILLER. Sure. Well, there is a framework. GAO put the
framework out in 2015. It is called the Framework for Managing
Fraud Risks in Federal Programs, and it outlines steps for taking
a proactive, strategic approach to managing fraud risks. If agencies
had done these kinds of activities prior to the pandemic, they
19
would have recognized that things, like not collecting device iden-
tity information, IP addresses, was going to create significant fraud
issues during a large event like the pandemic. Fraud risk manage-
ment needs to be proactive and needs to be at the management
level, and it needs to be in a strategic, not after the fact, recovery
focus, which is where we are now in the pay-and-chase mode.
I also want to be a little provocative here and suggest that we
might consider combining improper payments and fraud for the
purposes of agencies’ implementation because they see these as two
unfunded mandates, and they don’t take either of them seriously.
They take a very compliance orientation to both. And if we were
to reduce the separate and burdensome compliance activities
around improper payments and fraud, combine them and require
them to be more strategic, proactive, and outcome oriented, I think
we could see a lot more progress.
Mr. DAVIS. Well, thank you very much. You know, program in-
tegrity means more than just preventing fraud. It also means get-
ting money to everyone Congress intended. Ms. Gilkesson, what
happens to lifesaving Federal programs and the individuals and
families who rely on them when policymakers focus solely on reduc-
ing fraud by adding to an application an eligibility review process?
Ms. GILKESSON. Thank you so much for that question. When
there is a focus on fraud, it causes unnecessary burdens on both
clients and case workers, and, particularly, in terms of public bene-
fits like SNAP, TANF, and Medicaid. Instead of spending time ca-
tering to the needs of their customers, case workers have to be
hyper-vigilant about finding any fraud or risk losing their jobs. For
example, I used to be a case worker for the state of North Carolina,
and one thing that I do know is that quality control measures do
exist, and states are held really tightly to that and are often held
accountable to pay if they aren’t held to that. But some of those
quality control measures don’t allow for people to be able to get
their jobs done effectively.
And one thing that I would say is that public benefit recipients
have to endure over-policing, privacy invasions, difficult and time-
consuming applications, unnecessary verifications, all to prove that
they aren’t committing fraud, and this burden causes errors and
burdens to access. Thank you.
Mr. DAVIS. Well, thank you very much, and let me just say that
I think we surely need to spend as much time, energy, and effort
on the big guys as we do the small guys, and much more needs to
be done. Thank you both, and I yield back.
Mr. CONNOLLY. If the gentleman would yield to me for a minute.
Let me just say I agree with you completely. We need to be looking
at every level of improper payments. It is not just fraud. But Ms.
Gilkesson, I would just say, and I think you would agree, there is
an opportunity cost, however, when mistakes are made or fraud oc-
curs. So if I have $100 for a good program and $20 of it is siphoned
off improperly, whatever category that is $20 I am not providing
to the people who need it. And so it seems to me that, you know,
every effort we can make to try to make sure that the integrity of
the program is preserved so that the benefits are flowing to those
who need it and are eligible for it is a good public policy goal. And
I am sure you agree with that.
20
Ms. GILKESSON. Yes, absolutely.
Mr. DAVIS. Thank you, Mr. Chairman, and I agree whole-
heartedly.
Mr. CONNOLLY. Thank you, Mr. Davis. Yes, ma’am?
Ms. GILKESSON. Yes, I definitely agree with that, and I want to
make sure that we are understanding that having streamlined ac-
cess for people to actually be able to have access to the programs
is super important. And I love all of the technology fixes that a lot
of my co-testimonials here are mentioning because it really is im-
portant to streamline access and make sure that people actually
have the ability to get access, because there so many barriers with
applications, with verifications, that folks just, some of them walk
away and say they are not even going to try to go through with
it
Mr. CONNOLLY. Yes.
Ms. GILKESSON. And we want to make sure that all of the tech-
nology put in place is something that is encouraging people who
are eligible to apply.
Mr. CONNOLLY. Yep. Thank you. I think that is an important
point.
The gentleman from Pennsylvania, Mr. Keller, is recognized for
his five minutes of questioning.
Mr. KELLER. Thank you, Mr. Chairman, and thank you to Rank-
ing Member Hice and all our witnesses for being here at today’s
hearing.
Improper payments continue to be an area of concern in the Fed-
eral Government. According to the GAO, improper payments have
been estimated to cost $1.7 trillion governmentwide from 2003
through Fiscal Year 2019. These improper payments were con-
centrated in three program areas: Medicaid, Medicare, and the
Earned Income Tax Credit. As spending for Medicare continues to
increase, it is crucial that these funds are properly managed and
distributed because, as it was mentioned, improper payments take
benefits away from people who are truly in need.
According to the Centers for Medicaid and Medicare Services, in
Fiscal Year 2019, Medicaid made $57 billion in improper payments.
This amount rose to $87 billion in Fiscal Year 2020, meaning more
than 1 out of 5 payments under the Medicaid program was im-
proper. I think we all agree it is unacceptable that any funds are
going to ineligible recipients or made in incorrect amounts. We
need more accountability and oversight of the outlay funds so we
can ensure we are using taxpayer dollars appropriately. Reducing
these improper payments is crucial to protecting Federal programs
that serve Americans.
So if we take a look at all the things that have happened, and
I guess I will just ask Ms. Greszler. Have we seen payments in-
crease over the years, or have we seen the scorecard getting any
better with improper payments?
Ms. GRESZLER. No. We have seen kind of a double whammy be-
cause both the number of payments that are going out are increas-
ing, and also the rate of improper payments are also increasing
there. And so that is why we have seen 633-percent growth in the
number of improper payments, you know, just since 2005.
21
Mr. KELLER. And this actually was happening prior to the pan-
demic.
Ms. GRESZLER. Yes. The UI was specific to the pandemic because
just the size and scope of it, but all these other improper payments
were already increasing prior to.
Mr. KELLER. So had we put in place plans prior to the pandemic,
we wouldn’t have seen so much trouble through the pandemic in
getting payments made properly?
Ms. GRESZLER. Correct.
Mr. KELLER. OK. And that, I guess, is the point. We have these
government agencies that are responsible to make payments. So
the same people that run these programs and responsible for mak-
ing the accurate payments, are they the same people that have
been in charge of them for many years?
Ms. GRESZLER. I think that the problem is that the people who
are in charge of the program’s mission, as Ms. Miller pointed out,
are getting the payments out the door. And I see that problem in
two ways.
Mr. KELLER. No, but I would disagree with that. Getting the pay-
ments out the door accurately is the mission.
Ms. GRESZLER. It should be. It should be.
Mr. KELLER. Let me ask you a question. If you had an account-
ant that you were running a business and they were making im-
proper payments, their mission is to get the payment out the door,
right? But if your accountant isn’t doing their job properly, would
you continue to hire them and employ them?
Ms. GRESZLER. And that is exactly the point here is that——
Mr. KELLER. OK.
Ms. GRESZLER [continuing]. Is that we don’t have the account-
ability. There is no consequence when improper payments are——
Mr. KELLER. Well, that goes back to my point. Is it the same peo-
ple in these government agencies that have been overseeing these
payments for years at the head of the agencies? I mean, I guess
I am just——
Ms. GRESZLER. Yes. I mean, they change with administrations,
but the same ones are in there.
Mr. KELLER. Yes, and what needs to happen is some of them
need to lose their jobs because they are not doing them properly.
I mean, government is the only entity where we do less with more,
right?
Ms. GRESZLER. Mm-hmm.
Mr. KELLER. And we reward bad behavior. I mean, I guess I
would just look at that, even with unemployment insurance. I am
sure states have had issues. I know Pennsylvania has had issues.
I toured Pennsylvania’s Unemployment Division and Department
of Labor when I was a state rep——
Ms. GRESZLER. Mm-hmm.
Mr. KELLER [continuing]. And I looked at the inefficiencies there
at that point in time. Would it be safe to say that some of the pay-
ments are made improperly because we hand the money over to the
states, and then the states don’t make sure that——
Ms. GRESZLER. That was a big problem with the Pandemic Un-
employment Insurance benefits is that it was Federal money, and
there was no consequence to just simply approving those claims be-
22
cause they were paying the $600 and they were paying the ordi-
nary benefit. And so even if you thought, well, hey, maybe some
of this money is going to get spent in my economy in a good way
as opposed to going to criminal enterprises overseas, the incentive
was there to utilize those and to send the payments out the door.
Mr. KELLER. So would it make sense that when we give money
to states, if the state isn’t able to distribute that money effectively,
they shouldn’t get the dollars?
Ms. GRESZLER. Exactly. There have to be requirements there
that if you are getting this Federal money, you are going to be pro-
tecting Federal taxpayers through that.
Mr. KELLER. I mean, because we can talk about putting all this
money out, but, ultimately, our kids and grandkids are going to be
the ones paying it back because we are borrowing it. So I think it
should be imperative on all of us, a priority of all of us to make
sure it gets done effectively.
So I am out of time. Thank you, Mr. Chairman.
Mr. CONNOLLY. Thank you. The gentleman from Maryland, Mr.
Raskin, is recognized for his five minutes.
Mr. RASKIN. Mr. Chairman, thanks so much. You know, many of
our Federal programs and benefit systems have not been updated
to serve the needs of the contemporary economy and work force.
Consider how different our service economy and work force are
today compared to just 10 or 15 years ago. Many workers are in
the gig economy now, like Uber drivers, graphic designers, free-
lance writers, web developers. The Fed estimated that nearly a
third of adults engage in gig economy work. Ms. Miller, let me ask
you. Are current Federal laws and benefit structures conducive to
serving this population of today’s workers?
Ms. MILLER. I don’t believe they are. I mean, this is a very new
area that we are moving into where we have got lots more people
working remotely and gig workers. And, I mean, the systems that
are in place now aren’t even really set up all that well to work.
They are traditional systems. So, no, I think we have a ways to go
to adapt to this new work force.
Mr. RASKIN. Yes.
Ms. MILLER. And I don’t think we are even close to where we
need to be.
Mr. RASKIN. Well so, Secretary Jensen, it is good to see you. I
assume that gig workers in Rhode Island benefited from some of
the big improvements to state programs that took place under your
leadership. How did you work with an existing state law or change
state laws in order to design better systems to make it easier for
people to access while also preserving the integrity of the pro-
grams?
Mr. JENSEN. Thanks for the question. What we did with the pan-
demic unemployment assistance verification was we did a data
sharing agreement with the state income tax department and used
cloud computing to check to make sure that gig workers, when they
applied for PUA benefits, that their gross income was accurate.
And if they were accurate, we were able to pay that person quickly,
and if they weren’t, that person had to go through a much more
arduous process, which, you know, we were taking very seriously
23
because fraud is something that, I think, my colleagues definitely
cared about, and so did I, during the pandemic.
Mr. RASKIN. So, Ms. Gilkesson, can you tell us briefly what are
the first steps Congress should take now to renovate our laws to
make them more applicable to the real economic situation of work-
ers, families, and businesses?
Ms. GILKESSON. Thank you, Congressman. It is important for
Congress to think about people who are benefiting from these pro-
grams. So many of them experience the worst moment of their lives
seeking help because of sickness, a job loss, fleeing domestic vio-
lence, and so much more. They deserve to be treated with respect
and dignity, and they deserve to be defined by who they are, not
by stereotypes or false media portrayals. And last, they deserve to
have a seat at the table, not just to share their stories but to share
with you all policy changes that they know would help to serve
them and their families. So it is also important to keep in mind
how essential these programs are to people’s livelihoods. They help
to bring so many people out of poverty and really are programs
that are helpful to people when they are going through some of the
worst moments in their lives.
So I think that Congress should consider really working along
with their constituents to hear their thoughts on how they are
being treated at the public benefits agencies, and really create poli-
cies that are centered in trust and truth and what they believe and
deserve is best for them.
Mr. RASKIN. Thank you. And, Adrian, in your project on helping
gig workers receive unemployment benefits, did you see an actual
increase in people’s access to the benefits, and did you see improper
payments go down?
Adrian HARO. Yes and yes.
Mr. RASKIN. And how do you make sure that the projects you are
investing in are sustainable and effective in the long term?
Adrian HARO. That is a great question, Congressman. Thank
you. So really quickly, we follow the data, and in the case of gig
work and this project, the data show that there will be growth in
this part of the economy, right? We also know that these workers
have very little access to public benefits programs because of chal-
lenges like verification of their income. So we stood up this project
to help, both in the crisis moment but also over the long haul. And
as I mentioned in my testimony, the Income Passport holds prom-
ise beyond the temporary pandemic assistance programs, right, to
income across the entire menu of public benefits. That is every-
thing from SNAP, to Medicaid, to TANF, CHIP, right?
We could take the Medicaid example really quickly and just say,
unlike unemployment benefits in which eligibility is tied to work
status and dependent on employers paying in on behalf of workers,
and where gig workers do not currently qualify for the most part,
Medicaid, on the other hand, is largely depended on income level,
not work status. So all forms of non-traditional income—1099, self-
employment, et cetera—count toward the thresholds for eligibility.
So we need a better and more efficient way to verify this kind
of income in these types of programs so that qualified people are
getting the benefits they are legally entitled to, no more, no less.
Mr. CONNOLLY. Thank you.
24
Mr. RASKIN. Well, thank you. Thank you, Mr. Chairman, and I
yield back.
Mr. CONNOLLY. I thank the gentleman.
The gentleman from Arizona, Mr. Biggs, is recognized for five
minutes.
Mr. BIGGS. Thanks, Mr. Chairman. I am grateful that you are
having this hearing today. I think is an important topic.
Ms. Greszler, is there a focus in our agencies, our executive agen-
cies, on fraud prevention?
Ms. GRESZLER. I think there is—you know, at a surface level
they will say that they are looking into it, but I don’t think at the
core that they actually are. And I think that the problem there is
that they are looking more toward serving customers, toward mak-
ing sure that they are getting payments out as opposed to pro-
tecting the integrity of those payments.
Mr. BIGGS. So it is not—there is a superficial level. There is a
verbal commitment but no actual commitment. Ms. Miller, would
you agree with that?
Ms. MILLER. Absolutely.
Mr. BIGGS. So when I look at this, we currently have more than
a $30 trillion national debt, and in 2021, there was more than $280
billion in improper payments. The recent omnibus gave DOJ $35
billion in discretionary funding. The legislative branch receives $6
billion. The President requested $773 billion for an agency that
wastes nearly one-third of it, the Department of Defense. And ac-
cording to the White House’s own analysis, between Fiscal Year
2020 and 2021, the governmentwide improper payment rate rose
from 5.6 to 7.2 percent, and it really skyrocketed—UI fraud and
improper payments—during the pandemic, which I never use that
term, but from the 12-month non-pandemic compared to the 12-
month normal period.
So we look at this, and we say DOJ has brought over 1,000 crimi-
nal cases related to pandemic relief programs totaling over a bil-
lion, another 1,800 civil cases for an additional $6 billion, and, ac-
cording to the Secret Service, more than $100 billion was illegally
obtained from COVID relief programs. So, one of the things that
you testified to, Ms. Miller, is that we don’t have a balance, and
that is point I really want to get to today before I submit things
for the record, but this balance of putting money or benefits out the
door versus preventing fraud and improper payments.
And you have said that there are, since 2015, a GAO fraud risk,
basically a template. Tell me a bit about that, if you will.
Ms. MILLER. Sure. Yes. And I will say that, you know, in the pri-
vate sector, I have a number of private sector clients, and they
would go out of business, as has been mentioned earlier. It is not
like you can’t do this. It is not as though you can’t safeguard tax-
payer funding while also effectively getting benefits out the door.
It is just a matter of the will to do so.
The Fraud Risk Management Framework that GAO put out has
four areas to focus on. The first one, frankly, the one that I would
like to focus on the most is commit. That is the very first one. Des-
ignate an entity who will be committed to managing fraud risks.
Every private sector company that I work with as a client has a
dedicated fraud program. They have a fraud office. They have a
25
special investigations unit, someone whose job it is to make sure
that their money isn’t getting stolen. No agency in any Federal
agency, aside from the Center for Medicare and Medicaid Services,
has such an entity. It is going to be really important to make this
on par with getting benefits out the door if we want to see some-
thing change.
Mr. BIGGS. So when you mention that, implicit in that statement
and something that Ms. Greszler talked about earlier as well is in-
centives to find and stop fraud. And there seems to be very little
incentive within Federal agencies to stop fraud. They are rewarded,
and Ms. Gilkesson actually testified to this as well. They are re-
warded by moving benefits out, but there is no carrot and stick ap-
proach, quite frankly, to stopping improper payments and fraud.
I think that this hearing, Mr. Chairman, and I am glad we are
having it, but I think we need to expand this so we can actually
get some more specific, concrete evidence. I mean, me trying to get
five minutes, and then trying to get to everybody who is sitting
there, and trying to get some response, this is not the most condu-
cive way to tackle something, I think, is a really, really serious
problem. And I am hoping that we will revisit this again, Mr.
Chairman. I know you are going to say something. Do you want to
respond to that?
Mr. CONNOLLY. I agree with you.
Mr. BIGGS. OK.
Mr. CONNOLLY. And let me just tell you, this has been one of my
kind of pet topics, and I see Mr. Lynch is on. He will recall when
Todd Platts, Republican chairman of this subcommittee, was heav-
ily involved in this along with Mr. Lynch and got me involved. And
here we are, four bills later, you know, 14 years later, and things
are not better, and so we do need a new approach. And I do think
this provides an opportunity for bipartisan cooperation to figure out
what is that approach that hopefully will be efficacious and bring
that number and percentage down and incentivize Federal man-
agers to take this seriously.
Mr. BIGGS. Yes, I think that this is a nonpartisan issue. This is
an American issue. And so I need to read into the record just a cou-
ple things. I yielded to you.
Mr. CONNOLLY. Absolutely.
Mr. BIGGS. Thanks. I want a piece from The New Jersey Monitor
that says New Jersey fisheries net $2.4 million in improper COVID
payments. This is a Business Insider piece that says a man was ar-
rested after allegedly spending $5 million in COVID–19 relief
money on a Ferrari, Bentley, and Lamborghini sports cars. At least
he had good taste. Then we move to the next one, Austin American-
Statesman says, ‘‘Will prisoners also receive 1,400 stimulus checks?
Mostly true.’’
Mr. CONNOLLY. Is the gentleman seeking unanimous consent?
Mr. BIGGS. Yes, I am on these. I am sorry.
Mr. CONNOLLY. Without objection, it is so ordered, under-
standing, however, that when we are talking about improper pay-
ments, it is very rare. It is about a Lamborghini.
[Laughter.]
Mr. BIGGS. And continuing on, Mr. Chairman.
Mr. CONNOLLY. Yes.
26
Mr. BIGGS. Another one. ‘‘Rand Paul Stops Democrats’ Attempt
to Get Free Money to Planned Parenthood,’’ and I have a series of
pieces on that. Moving to, ‘‘Stimulus Turns Political as SBA Tries
to Claw Back Funding from Planned Parenthood.’’ Some more on
that. And then how about this one: ‘‘Made Out Like Bandits: New
York Prisoners Got $34 million in COVID Stimulus Funds’’. ‘‘Big-
gest Fraud in a Generation, the looting of the COVID Relief Plan
Known as PPP.’’ ‘‘Criminals Have Stolen Nearly a Hundred Billion
Dollars in COVID Relief Funds, Secret Service Says.’’ That is from
NBC.
Another, NBC News, ‘‘Man Who Used COVID Funds on a
$57,000 Pokemon Card Received 3 Years in Prison.’’ What the
heck, a Pokemon card. And then the next one from CNBC: ‘‘Florida
Businessowner Got $3.9 Million in Coronavirus Relief Money.
Spent It on a Lamborghini and Luxury Hotels.’’ And it goes on, and
I got a whole stack of these things. They are just a small droplet
in the big bucket of fraud and waste that occurred in the skyrocket
of the last two years. But, believe me, as the testimony goes, Mr.
Chairman, and you know this, this was on the rise before we ever
hit the end of COVID relief.
So thank you, Mr. Chairman.
Mr. CONNOLLY. Without objection, it is so ordered into the
record.
Mr. CONNOLLY. I would also ask unanimous consent to enter the
GAO’s report, ‘‘A Framework for Managing Fraud Risk in Federal
Programs’’ into the record as well.
Without objection, it is so ordered.
Mr. CONNOLLY. The gentleman from Maryland, Mr. Sarbanes, is
recognized for his five minutes. Welcome.
Mr. SARBANES. Yes, Mr. Chairman. Thanks very much, and I
want to thank our witnesses.
I want to make sure I understand something. Maybe, Ms. Miller,
you could help with this. You know, there are obviously situations
in which payments go out to people that are intending to commit
fraud against the system, and recovering that becomes very impor-
tant, and having accountability measures in place up front is vital.
But a lot of this overpayment is going to people that are deserving
of the dollars. It is just the efforts to get those out the door in a
time-sensitive way. Some of the problems with the technologies and
systems that are delivering those dollars are such that an overpay-
ment may occur to someone who is deserving, and then we have
the challenge of trying to recover it. Is that lion’s share of what the
overpayments look like, Ms. Miller?
Ms. MILLER. Well, I think the short answer is ‘‘no’’ and ‘‘no one
really knows’’ because I think estimation of improper payments and
certainly root causes is extremely difficult, and agencies have been
unsuccessful to date in doing that effectively. But I would argue
that, in my experience at GAO, the majority of improper payments,
while we don’t know, and I don’t think it is really important to try
to understand someone’s intent for the purposes of trying to make
sure that money that someone is not getting a benefit that they are
entitled to, all you really need to know is did they circumvent a
control, advertently or inadvertently, that enabled them to get
something they weren’t entitled to. I think getting intent out of this
27
discussion is actually helpful because we are really just trying to
make sure we are safeguarding the integrity of the funds, not nec-
essarily determining whether somebody had malintent.
Mr. SARBANES. Which is why building is these protections on the
front end is so important because, as you indicate, you are not de-
pendent on the intent in terms of the analysis you are doing
around the overpayment. Mr. Jensen, why did the pay-and-chase
approach become the norm for governments? Give me your perspec-
tive on that.
Mr. JENSEN. Thanks, Congressman. I don’t think it is the norm.
I think that, daily, people running unemployment insurance pro-
grams, at least is what I am familiar with, are trying to do every-
thing they can to stop improper payments. The entire UI system
is meant to do that, and so is, you know, trying to catch identity
fraudsters and others. You know, I think that we wind up in that
circumstance too often similar to, you know, police departments
trying to stop burglaries. But, you know, I think it is an overstate-
ment to say that no one is motivated to stop fraud, and no one is
doing it.
Mr. SARBANES. OK. Let me ask you and Ms. Miller this question.
We have talked about if there was more accountability in terms of
these program managers and others who are in a position to push
dollars out the door, that that would be helpful, and then we also
know that technology is a huge part of this. What is the relative
impact of those two things? In other words, you could have pro-
gram managers who know that they are, you know, they are ac-
countable, and they could have a terrible system available to them
in terms of the technology, and there is not a lot that they can do,
and you could also have the alternative scenario. How much is the
technology preventing us from getting to a place where, with real
accountability in place, we would actually fix this problem? And,
Ms. Miller, maybe you can start, and, Mr. Jensen, it would inter-
esting to hear your perspective from the field.
Ms. MILLER. Sure. I think that technology is a significant limita-
tion in doing more to prevent and detect fraud and improper pay-
ments. I think it is technology, and I think it is data. Honestly,
data sharing is so broken in government right now that even if you
had technology, frankly, you wouldn’t really be able to use it very
well because there is so little data that agencies can access and use
to meaningfully understand the extent that a payment might be
improper. But technology, from a perspective of, you know, the
pandemic and the new wave that we are in now, the way that that
fraud is happening through cyber channels. So fraudsters are using
all kinds of very advanced techniques to steal identities, and the
agencies are still using, in many cases, knowledge-based authen-
tication. They are not even using multi-factor authentication to
verify the identity of an applicant. So it is a technology and a data
problem, absolutely.
Mr. CONNOLLY. Thank you.
Mr. SARBANES. Mr. Jensen?
Mr. CONNOLLY. Mr. Jensen, real briefly if you want to respond
because the gentleman’s time has expired.
28
Mr. JENSEN. Sure. I think it is an application of check technology
challenge. So you need the technology, and you need to apply it
prudently to what you are trying to do.
Mr. CONNOLLY. Thank you. We got to apply things.
Mr. SARBANES. Thank you.
Mr. CONNOLLY. That is another watchword, ‘‘apply things pru-
dently.’’ We should all live by that.
All right. The gentleman from Georgia, Mr. Clyde, is recognized.
Mr. CLYDE. Thank you, Chairman Connolly and Ranking Mem-
ber Hice, for holding this very important hearing today.
Part of this committee’s mission is to look for and prevent gov-
ernment waste, fraud, and abuse, and today we are finally holding
a hearing that is highlighting the Biden administration’s failure to
conduct oversight of government programs awarding improper pay-
ments, costing the Federal Government and American taxpayers
billions of dollars every year.
Improper payments have cost American taxpayers $281 billion
alone in Fiscal Year 2021. This is an increase from $108 billion in
Fiscal Year 2012. Just imagine if the Federal Government spent
more time ensuring proper payments were being made. Our na-
tional debt would probably not be as high as it is today. It is imper-
ative that the Federal Government be good stewards of taxpayer
funds as each improper payment further erodes the trust of the
American people and the security in the Federal Government.
While Federal agencies were quick to deliver financial relief to
American taxpayers when COVID hit, it is extremely important to
note that many of these agencies did not have the safeguards in
place to forestall improper payments. In fact, in Fiscal Year 2020
and Fiscal Year 2021, the governmentwide improper payment rate
rose from 5.6 percent to 7.2 percent, largely due to problems with
COVID-related relief funds. For example, experts believe that mil-
lions of dollars of unemployment benefits went to bad actors, and
there is proof of it. Just yesterday, a man in Rhode Island was in-
dicted for using stolen identities of others to file for COVID-related
unemployment benefits.
Mr. Miller, a question for you, ma’am. During his State of the
Union address, President Biden acknowledged that fraud within
the Unemployment Insurance Program was egregious. What ac-
tions must the Biden administration immediately take to better in-
vestigate and to retrieve improperly received taxpayer funds dis-
tributed for Pandemic Unemployment Insurance?
Ms. MILLER. Well, I believe that there are a number of initiatives
under way right now to do recovery. He obviously identified a Fed-
eral prosecutor that would be in charge of COVID funding. There
are also several law enforcement agencies. I worked at one of those
oversight entities, the PRAC, which coordinates the work of several
law enforcement agencies. There is a fraud task force that PRAC
has created. The DOJ has multiple task forces, Secret Service, FBI.
There are a variety of law enforcement agencies currently going
after the fraud in the pandemic.
The problem is there is not a limited number of oversight. At this
point, there is actually a limited number of people that can actually
go back and do all of the investigative work that is going to be
needed, time to build cases to get prosecutions, but, honestly, very
29
little of that money is going to get to get recovered. Most of the
pandemic spending was identity theft-based fraud. Identity theft,
you are not going to recover that. Those guys are gone. They were
not using identities that can be identified through a law enforce-
ment investigation, unless they can find a criminal ring, which is
extremely challenging for them to find, so especially when they are
offshore.
So, I mean, to me, I don’t think it was a problem with the pros-
ecutor, and I think it is great that we are going to try to claw back
as much as we can. But going forward, I think the focus has to be
on prevention and not so much on recovery.
Mr. CLYDE. Thank you, ma’am. So you think that pretty much
what is out there that is gone is gone already, and there is not
much chance of getting it back?
Ms. MILLER. I mean, I think some programs are more likely to
recover more funding. PPP loans in particular are ones that are
going to be more likely to get money back, but I think a lot of the
unemployment assistance payments are going to be very difficult,
if not impossible, to recover.
Mr. CLYDE. So, therefore, we need to increase the safeguards on
those programs right now, make them much more robust. All right.
Well, thank you. A question for Ms. Greszler. Do you believe the
lack of safeguards implemented within the Unemployment Insur-
ance Program made these programs more vulnerable to bad actors,
or, if not, what do you think did make them more vulnerable?
Ms. GRESZLER. Yes, I think it was absolutely that lack of safe-
guards and lack of any consequence when the payments went out
the door. What I think is particularly troubling going forward is,
yes, we have to address the existing improper payments within
government programs already, but we are seeing from this Admin-
istration proposals to make Americans dependent on the govern-
ment, not just for things like unemployment insurance and Social
Security, but for every part of their life: for childcare, for paid fam-
ily leave, for pre-K. And so the bigger that the government grows,
the more improper payments that are out there, but it is also not
helpful for people.
We have heard from Ms. Gilkesson, these programs are not easy
to navigate. Who wants to go and fill out a bunch of paperwork and
submit medical information to be able to receive paid family leave
with what we have seen in some of the states that have that?
Washington state, for example, people were waiting weeks, if not
months, when they weren’t even able to apply until they had a
need for leave, and so they didn’t find out until months after they
needed the money whether or not they were eligible for it. And so
expanding government’s reach into people’s lives is not what em-
powers them. We need to look toward changing the nature of the
programs that are out there and also toward not expanding them
any further.
Mr. CONNOLLY. Thank you.
Mr. CLYDE. I would agree with that. Thank you very much, and
with that, Mr. Chairman, I yield back.
Mr. CONNOLLY. Thank you, Mr. Clyde. Just for the record, I want
to be clear that this hearing is not about any particular adminis-
tration. It is a hearing in a long series of hearings this committee
30
has had over the years looking at a chronic problem that has af-
flicted every administration, namely improper payments, and our
hope is that we can arrive at some nonpartisan solutions that get
at this. I thank the gentleman.
The gentlelady from Ohio, and forgive me if I overlooked you pre-
viously, Ms. Brown. I didn’t mean to do that. You are recognized
for your five minutes of questioning. Welcome.
Ms. BROWN. No worries. Thank you, Chairman Connolly, for
holding this hearing, and thank you to all the witnesses for joining
us today.
The Do Not Pay Program within the Bureau of the Fiscal Service
at the Treasury Department enables agencies, and even states, to
access data bases from across the government to verify payment
criteria. Ms. Miller, can you please explain how Federal and state
agencies can utilize Do Not Pay to identify and prevent improper
payments?
Ms. MILLER. Yes. Well, agencies are encouraged, and actually al-
most required by the Payment Integrity Information Act, to use Do
Not Pay to the extent that they can prior to making a payment.
GAO has reported numerously over the years on agencies’ limited
ability to use Do Not Pay and the limitations within Do Not Pay.
It is my opinion that Do Not Pay is broken and it needs significant
enhancing, and if a centralized anti-fraud office were created, I
think it would be really important to assign responsibility for im-
proving that dramatically.
The Pandemic Response Accountability Committee, the PRAC,
where I was a deputy executive director, created a center of excel-
lence called the Pandemic Accountability Center of Excellence. And
in just a short period of time, less than a year, the PACE, as it is
known, has been able to pull in numerous data sources from the
private sector and across government, including a lot of hotline
data that has been really valuable to them in their oversight ef-
forts. I believe that Do Not Pay really needs significant overhaul.
The data bases in it need to be expanded, and agencies need to
have more ability to access those datasets in order to be able to
identify fraud because, right now, that system is not working any-
where nearly as well as it was intended to work.
Ms. BROWN. Do Not Pay has identified or prevented $780 million
in improper payments in its lifetime, and that is a lot of improper
payments avoided. Yet only 40 percent of states, and not many
more than 20 Federal agencies out of hundreds that exists, cur-
rently use the program. My question to this panel is why. Why are
many states and so many Federal agencies choosing to forgo Do
Not Pay?
Ms. MILLER. Well, I will just say that I think it is because a lot
of them don’t even know how to access it, and because, right now,
there are only limited numbers of datasets that they could use to
access them. But I think also, as Mr. Jensen has mentioned, a lot
of times at the state level, the data systems that states use, and
Mr. Jensen can probably speak better to this, can have challenges
accessing some of these data sources that the Federal Government
offers. There needs to be a lot better Federal and state data shar-
ing and accessing each other’s systems, and I think that would im-
prove the use of Do Not Pay.
31
Mr. JENSEN. And data infrastructure as well. During the pan-
demic, there are other systems like that to check on wages that
were earned in other states and unemployment insurance. That im-
mediately crashed, and if you had a claim that was in two states,
you were not going to get paid for months and months until that
system came back up. So, to me, that is a function of not being on
the cloud, not using contemporary technology.
Ms. GILKESSON. One thing that I would also mention is that the
states have a focus on fraud in terms of individual people, but like
Ms. Miller has talked about today, she has mentioned identity
stealing, cyberattacks. The technology that states have right now
are not meant to catch people who are doing cyberattacks or who
are stealing identities. It is meant to catch individual people. And
oftentimes, the technology that is in place is catching people who
are actually innocent.
So there are a lot of people who are being accused of fraud who
have benefits taken away from them who actually were eligible and
didn’t do anything wrong. But the systems that are set up are
catching those individual people versus focusing on international
rings that have been mentioned today. And I think that is impor-
tant to have on the record because what we have that exists are
solely based on, like I talked about in my oral testimony, stereo-
types and things of that nature that are trying to catch people com-
mitting fraud that are not committing fraud, right?
And so the increase in improper payments is based on not nec-
essarily individual people who are all deserving, but based on, like
Ms. Miller talked about, identity stealing, cyberattacks, the dark
web. And right now, I can tell you that state systems are not
equipped whatsoever to take on those threats and attacks.
Ms. BROWN. Thank you so much. My final question is for Ms.
Miller. What can Congress do to encourage or require agency and
state participation in Do Not Pay?
Ms. MILLER. Well, I am not sure that this is necessarily a Con-
gress solution. I think that Treasury needs to improve the access
to data that exists right now in Do Not Pay. I think OMB really
needs to prioritize putting datasets into Do Not Pay, and Congress
has already told agencies to use Do Not Pay prior to making pay-
ments. GAO has found that Do Not Pay actually matches payments
after they are already made, so it is not even actually a prepay-
ment system. So I am not sure, besides Congress dedicating fund-
ing and leadership to overhauling and fixing Do Not Pay, I am not
sure how much more Congress can do.
Mr. CONNOLLY. Thank you. I thank my friend from Ohio.
The gentleman from Kansas, Mr. LaTurner, is recognized for his
five minutes.
[No response.]
Mr. CONNOLLY. Mr. LaTurner is not here?
Voice.
[Inaudible.]
Mr. CONNOLLY. OK. All right. The gentleman from Massachu-
setts, Mr. Lynch, is recognized for his five minutes, a longtime pro-
ponent of addressing this subject, improper payments. Mr. Lynch.
32
Mr. LYNCH. You are too kind, Mr. Chairman. Thank you very
much, and I want to compliment you as well for pulling together
such great witnesses who have been very helpful.
I do want to remind my colleagues on the other side of the aisle
that they seem to have forgotten the CARES Act, $2 trillion was
actually signed into law and deployed by the Trump administra-
tion, so I appreciate the bipartisan tone that the chairman has set.
There should be room for facts as well.
After the CARES Act was passed, we got a report, I believe it
was from GAO that 75 percent of the fraud committed in connec-
tion with that funding occurred on financial technology platforms
and APIs. So I would like to know from Ms. Miller, are we focusing
on that fact and targeting where most of this fraud occurred? Is
there any hope that perhaps two-factor authentication might be re-
quired by recipients in order to more deeply secure the funding
flow of people who are deserving of those benefits?
Ms. MILLER. Yes, I believe that multi-factor authentication is one
of many tools that agencies need to employ to ensure that appli-
cants are who they say they are. There has been some NIST guid-
ance. I think there has been some White House executive orders
recently that have required agencies to use multi-factor authentica-
tion as well as things like Zero Trust architecture. So there are a
variety of cybersecurity initiatives underway.
What I believe is there needs to be coordination between an anti-
fraud entity that hopefully will exist at some point if some legisla-
tion is introduced that creates that office. That office needs to co-
ordinate with the Cybersecurity Group, with CISA, and with NIST
to make sure that there are anti-fraud controls that are required
to be put in place, like the private sector currently has, things like
device fingerprinting, things like IP address checking, to make sure
that somebody who is using a phone, or a computer is who they say
they are. You can even use more advanced tools like behavioral bio-
metrics. And so it is really important, I think, for this coordination
to happen between the cybersecurity offices and the anti-fraud ef-
fort in order to make sure that these technologies are put in place
and being able to be used to their greatest effect.
Mr. LYNCH. Well, I appreciate that. Ms. Miller, I would just note
that we have been at this, at least Mr. Connolly and I, for about
14 years. So, you know, my own observation is that the government
has not been known for its speed in this area. You know, multi-
factor authentication is something that we can adopt very simply.
It is at the front end of this process. So while certainly the need
for an anti-fraud agency won’t go away, I think we can stop at the
front end a lot of this happening. If 75 percent of the fraud oc-
curred on these, you know, Fintech applications and Fintech plat-
forms that should tell us something, right? The balance of it oc-
curred, you know, through manual processes.
You know, the SBA was trying to push out the money as quickly
as possible. It was emergency funding and it needed to get out, so,
you know, we own part of that because of the haste at which we
acted and the President acted. But, Ms. Greszler, do you have any
different perspective on, you know, multi-factor authentication that
might prevent some of this stuff at the front end?
33
Ms. GRESZLER. I think just allowing the agencies and the pro-
grams to determine what steps are necessary in their unique pro-
gram. Sometimes it is as simple as verifying an ID and an income
that might be already on a W–2 form. Other cases it is verifying
is this child, you know, the legal guardian of this person. Are they
living in the household? It gets more and more complicated along
the way across these various programs. And so ensuring that the
agencies can have the flexibility to choose which metrics are right
for them, but that at the end of the day they are held accountable
to meet certain criteria to protect the integrity of those programs.
Mr. LYNCH. Thank you. And, Ms. Gilkesson, do you have any
thoughts on this as well?
Ms. GILKESSON. Thank you so much. Yes, I do. I think it is really
important for us to begin to simplify access for people by stream-
lining eligibility. I think it is a really great point that you men-
tioned that 75 percent, just like I have mentioned before, of the
fraud that was taking place was indeed technology systems. So I
think it is important, as we simplify access for actual people and
open the door for them to be able to have easier access, be able to
submit their verifications more easily, that we can get access to
folks better, and then be able to prevent those cybersecurity sys-
tems from causing fraud.
Mr. LYNCH. Thank you so much, Mr. Chairman. I yield back.
Thank you.
Mr. CONNOLLY. I thank the gentleman and look forward to work-
ing with him on the legislation I mentioned a little bit earlier. You
know, this is ground he has plowed before, and I know we all share
the same goal of ultimately, in theory, trying to get improper pay-
ments down to zero. So we will work together in that. I thank him.
The chair recognizes himself for his five minutes.
Mr. Jensen, if you would put your state hat on for a minute.
Does the Federal Government run Rhode Island’s unemployment
insurance system?
Mr. JENSEN. No.
Mr. CONNOLLY. Who runs it?
Mr. JENSEN. The state runs it.
Mr. CONNOLLY. Would that be true in, say, Georgia, too?
Mr. JENSEN. That would be true in every state.
Mr. CONNOLLY. Every state. So we have 50 different unemploy-
ment insurance systems. Is that correct?
Mr. JENSEN. That is right.
Mr. CONNOLLY. And how would you characterize the data base
management, the IT systems, of those 50 states? Would you say
they are in great shape and pretty efficient in catching errors?
Mr. JENSEN. They are old, many of them, and they are modern-
izing right now. And the virtue of some of the old ones is the people
running those programs know how to use them, but there needs to
be improvement in the tech.
Mr. CONNOLLY. I think you would concede, let’s say you got a
system based on COBOL. The population that knows COBOL is
dwindling.
Mr. JENSEN. That is correct.
Mr. CONNOLLY. Yes, and I can tell you that, to Ms. Gilkesson’s
point, as someone doing constituent work, when we had the begin-
34
ning of the pandemic when there were tens of millions of people fil-
ing for unemployment, even though it is a state responsibility, not
a Federal responsibility, I was inundated. And all of the cases were
people who qualify but weren’t getting their benefits, to your point,
Ms. Gilkesson. So, I mean, we can’t lose sight while we are also
trying to make sure every penny is accounted for, but there were
lots of people desperate for that assistance. And because of the IT
systems that could not handle the programming changes we made
at the Federal level, and then the volume of demand, it just over-
whelmed the system, and it was really quite problematic. I don’t
know if that reflected your experience.
Mr. JENSEN. Absolutely. That is why I referred to our experience
in Rhode Island as the hardest C-minus I have ever earned.
Mr. CONNOLLY. Yes, thank you. I can tell you that was true in
Virginia, too. Ms. Miller, talking about fraud, a few years ago, the
U.S. attorney in Boston decided to make Medicare fraud a priority
for her office. And in that one year, they identified and mostly re-
covered about $3 billion in Medicare fraud. Now, that is 1 of 99
U.S. attorneys across the country. Do you believe that U.S. attor-
neys are taking this kind of fraud, whether it be Medicare, or Med-
icaid, or anything else for that matter, seriously? I mean, is it a
high enough priority, because when I saw that, I thought, well,
that is pretty low-hanging fruit apparently, and could that be rep-
licated in other offices. And are we missing an opportunity here to,
in fact, both make a statement in terms of criminal prosecution to
disincentivize people from cheating or frauding, but also recovering
taxpayer dollars?
Ms. MILLER. I mean, I definitely think that there is no question
that if you persevere and that is your focus, you are going to have
results, and that is the main reason why we see such disparate re-
sults across agencies as well as, you know, U.S. attorneys or law
enforcement agencies, U.S. attorneys’ offices, from my experience at
the PRAC, working with other law enforcement agencies to refer
cases, you know, they are also, like many other agencies, very, very
overwhelmed with cases. And so what they want is a really win-
nable case to get onto their desk.
If we can develop really, really winnable cases to turn over to the
U.S. attorney’s office and the U.S. attorney makes that their pri-
ority, yes, I think we could see significant increases in prosecution.
But, again, I do believe prevention is a better area for us to focus
on.
Mr. CONNOLLY. I agree with you, but I am harkening back to
something you also said. If we don’t make this a priority for people,
it won’t be a priority. And it seems to me, you know, protecting tax-
payer investments so that the people who need it are getting it as
opposed to it being diverted by fraudulent characters or even just
mistakes, is in the public interest.
Ms. MILLER. For sure, and I don’t think it needs to be even,
right? Like, I think there is this balance, and we are seeing agen-
cies, the scale is toward get money out the door, and payment in-
tegrity, program integrity is very, very little. I don’t think we are
going to get it to 50/50, but we really need to move the needle. And
again as Ms. Greszler was saying earlier, we don’t see anyone sit-
ting up here today, you know, having to be held to account for why
35
their improper payments rates are continuing to rise. They are not
addressing the root causes of these problems. They are just report-
ing higher and higher rates, and there needs to be more account-
ability, I agree. And there needs to be a focus on incentives and
performance metrics that prioritize integrity to the degree that
they also prioritize mission delivery.
Mr. CONNOLLY. And real briefly, if I could indulge just one more
question for you. I mentioned that we have passed, you know, four
bills in this time period that Mr. Lynch and others——
I just see Ms. Porter coming. OK. I am going to recognize you
in a second, Katie.
OMB, pursuant to those laws, requires agencies to have a correc-
tive action plan. Given your experience when you were at GAO,
how well are those working, because we don’t seem to see the num-
bers coming down, and have we just succeeded in sort of getting
more accurate data about how bad it is as opposed to actually get-
ting that number down, because that would seem to be what the
charts, both Ms. Greszler’s and my own, would seem to suggest.
Ms. MILLER. Yes, I would say that we have gotten better at ad-
miring the problem, but we have made almost no progress in actu-
ally fixing the problem.
Mr. CONNOLLY. Thank you. My time has expired.
Ms. Porter, you are recognized for your five minutes of ques-
tioning. Welcome.
Ms. PORTER. Thank you very much, and thank you, Mr. Chair
and to our staff on the committee for pulling together this hearing.
I really appreciate it.
Policymakers and our minority witness often talked about the
tradeoffs between fighting fraud and expanding access to programs.
They say we can do one or the other but not both, but when done
right, governments can design programs to prevent improper pay-
ments and fulfill customer needs. Adrian, our witness from The
Workers Lab, stated that government and the people it serves can
both win.
In December, President Biden issued an executive order with the
title, ‘‘Transforming Federal Customer Experience and Service De-
livery to Rebuild Trust in Government.’’ The executive order di-
rected Cabinet secretaries and 17 agencies to implement reforms to
make government service easier to use and more transparent.
These reforms have the potential to improve the Federal customer
experience and reduce improper payments. This executive order is
a great step forward that we should be building on. In the past,
the government has been way too slow to adapt new technologies,
to streamline government operations, make it easier to administer
programs, and prevent fraud.
Mr. Jensen, your organization is focused on helping the govern-
ment use data and technology to improve programs. Mr. Jensen,
can you provide an example of how technology can help the govern-
ment reduce fraud while also providing benefits to those who qual-
ify?
Mr. JENSEN. Sure. Thanks for the question. You know, I would
like to congratulate Adrian on the solution they came up with be-
cause that is exactly the kind of thing that needs to be done. For
example, you know, I mentioned this earlier, but to verify to verify
36
income for the PUA program, we were able to configure a way on
the cloud where both the labor department and the tax department
could say with a straight face that PII was not being shared be-
tween the two, so that both departments could be safe just in terms
of not having any data breaches in that context. It worked great.
So what we need to do is prioritize the kind of performance metrics
that do require both a fraud detection and service delivery, and it
can be done.
Ms. PORTER. Ms. Miller, what private sector practices can gov-
ernment adopt or borrow to improve service delivery and program
integrity?
Ms. MILLER. The private sector focuses on something called fric-
tion when they apply technology to consider their customer user ex-
perience. The goal is to reduce friction to the lowest level possible
because you will lose a customer, right? If you make it too difficult
to get onto your app on your phone, or to transfer a payment, or
to access any kind of service or benefit in a private sector company,
they run the risk of losing that customer. So that friction piece is
incredibly important.
The government can learn from the private sector in the tools
that they have employed that reduce friction, and today, we can
use real-time data checks. There are data providers out there that
have an enormous amount of data on every American citizen, data
that can be checked within milliseconds of when you apply for
something. So as soon as you log into an application, onto a
website, it can tell you really quickly whether or not that is me or
not. And these kinds of tools, they are used every single day in pri-
vate sector companies across the country and the globe.
And these are the kinds of things we need to start to push gov-
ernment agencies to do so that they are not making it harder for
citizens to access benefits, but at the same time, they are reducing
the likelihood that they are actually a fraud actor and not a legiti-
mate beneficiary.
Ms. PORTER. I completely agree, Ms. Miller. And, you know, the
Federal Government has to store and process massive amounts of
information, massive amounts of data created by applicants for
Federal programs, but automation, data sharing, the techniques
that you mentioned, including reducing friction, can help govern-
ment agencies handle this challenge. There are just so many exam-
ples in this hearing of how government can use modern technology
to both reduce fraud and improve customer experience for tax-
payers.
Ms. Gilkesson, can you give an example or talk about from, your
perspective, is this possible? Can we reduce improper payments
without making government services more difficult to access?
Ms. GILKESSON. Absolutely. I definitely believe we can reduce—
could you repeat the question one more time? I apologize.
Mr. CONNOLLY. I think the question was, can’t we reduce the
probability of fraud or improper diversion of payments without
making it harder for people who need the benefits to apply and get
them?
Ms. GILKESSON. Absolutely. I think it goes back to the point that
I made earlier of us shifting our focus on individual people who are
innocent, and shifting our focus toward, like Ms. Miller talked
37
about, the computer systems. And I love that you made the point
about reducing friction—I am going to take that word, that phrase
that you said—the ability to reduce friction and make it easier for
folks to access benefits.
In an OMB report, it reports that administrative burden and dif-
ficult applications often block those who need benefits the most
from receiving them and exacerbates racial inequity. So making the
application process, making it harder to access benefits, that is the
issue that we are experiencing right now as to why improper pay-
ments are higher because it increases the error rate. So reducing
friction, like Ms. Miller mentioned, I think will be great in helping
to make sure that we reduce fraud and also increase access.
Mr. CONNOLLY. Thank you, and I thank the gentlelady.
Mr. LaTurner, you are back. You are recognized——
Mr. LATURNER. I am back, Mr. Chairman.
Mr. CONNOLLY. You are back, so you are recognized for your five
minutes of questioning. Welcome.
Mr. LATURNER. Thank you. Good morning, and thank you for
having this hearing, Mr. Chairman. I really do appreciate it.
My first questions are for Ms. Greszler. Ms. Greszler, the pan-
demic relief funds were disproportionately susceptible to fraud and
abuse, which many of my Republican colleagues predicted prior to
their passage. However, fraud and overpayments have been a con-
sistent problem in other Federal programs for years. In your testi-
mony, you use the Earned Income Tax Credit as one example,
pointing out that, on average, about a quarter of EITC dollars were
claimed incorrectly over the past decade. You talked about possible
actions the administration could take, but what legislative actions
have been taken in the past or could be taken now to mitigate
these issues?
Ms. GRESZLER. Yes. So specifically dealing with the Earned In-
come Tax Credit, it is verifying the income that was made, and it
also depends on whether or not you are married and the number
of children who are in the home, and so that that is where it gets
difficult. And this is not an easy thing to design a government pro-
gram and to get the benefits to the people you need when those
needs get more specific in terms of what you want because it gets
to the eligibility side. And so ID checks are pretty simple. That is
something that across the board we should absolutely be doing. But
then you have the eligibility standards unless we get to a point
where people are going to be comfortable with the government hav-
ing direct access to their bank account, having access to their med-
ical records, perhaps even having a video into their home to see is
this child who you say is living with you actually living there. You
know, those are the types of things there has to be a check that
that child exists, they are living there, and then also the income
that is coming in, and not all the time is that income going to be
on a W–2 Statement, you know.
As Adrian mentioned, we have more than a third of the work
force now that are working independently in these flexible type
jobs. Some of the time, that will show up easily on an app if you
are driving for Uber where all the money is in one place, but if you
are doing things on the side through your own business and maybe
it was a cash payment, it is a lot more difficult to track. And hav-
38
ing to gather all of that information and to submit it is difficult,
and that is where we hear it. People don’t get the benefits that
they may be entitled to because it is a cumbersome process, and
oftentimes it is the people who need them most who find it the
most difficult to do that, and it is the criminals who understand
the system who can get access to the benefits easiest.
Mr. LATURNER. Yes. No, I understand what you are saying, and
it certainly is not without a lot of difficulty.
At the beginning of this month, as I know you are aware, the
White House released a statement which claimed the Biden admin-
istration had begun to take actions to combat fraud in the pan-
demic relief programs. Beyond implementing new retroactive re-
porting requirements and payment audits, how can we leverage ex-
isting Federal information security infrastructure to make more ac-
curate eligibility determinations?
Ms. GRESZLER. I think it is important that the administration
tasks these agencies and the programs that they are running with
clear standards and what is their expectation. If your improper
payment rate is 22 percent this year, it should be expected to get
down to 10 percent over X time period, and what will be the con-
sequences if it doesn’t in terms of your funding levels. There need
to be actual teeth put on these so that people will be held account-
able. Otherwise, you are just not going to get to the outcomes.
Mr. LATURNER. I appreciate that and thank you for your time.
Ms. Miller, I would like to speak to you on your experience as a
former assistant director in the Forensic Audits and Investigative
Services Division of the Government Accountability Office. When
making eligibility determinations for Federal programs, agencies
often need access to sensitive data—we were just talking about
this—such as income, Social Security numbers, tax information. In
your opinion, is our current cybersecurity infrastructure prepared
to accommodate increased data sharing across Federal agencies?
Ms. MILLER. I don’t think that data sharing is a cybersecurity
issue. I think getting access to data and being able to create the
secure portals to use that data, the ability to mask personally iden-
tifiable information, none of this is beyond the capabilities of Fed-
eral agencies. They are all able to do this. The key is to require
them to do it, and the challenge that they have is these data shar-
ing agreements, they take a very long time to access.
I mean I was speaking with my former chief data officer at the
PRAC, and he said anytime I ask for a dataset, the first answer
is always ‘‘no.’’ They start with ‘‘no,’’ and then we try to get to
‘‘yes.’’ I mean, that is what so challenging. We are not setting up
a system for data sharing in the government that encourages it.
There is a deep discouragement of sharing data, and you have to
have enormous amounts of will to overcome the political, the legis-
lative, and the technological and bureaucratic barriers in order to
be able to do it, and we have to remove those barriers. So what can
Congress do? Help remove those barriers.
Mr. LATURNER. I understand what you are saying, and my time
is running out, but really quickly. I don’t want to mischaracterize
what you were saying, but we have the ability to do it and we
should do more of it, but you don’t have concerns about the security
of that data once we have it or once we have more of it.
39
Ms. MILLER. I don’t want to get technical here, but we can hash
data. Many agencies hash data now. IRS hashes data. So there are
tools that Federal agencies can use to mask personally identifiable
information to encrypt it so it can be only used—you can use public
encryption keys. Agencies have the technology right now. It is not
that they don’t have that technology. It is that they aren’t using
it.
Mr. LATURNER. Thank you. Thank you, Mr. Chairman. I yield
back.
Mr. CONNOLLY. I thank you Mr. Turner, and to your point, Ms.
Miller, I remember when the OPM—the Office of Personnel Man-
agement—breach occurred two administrations ago. They had not
yet installed the Einstein 2 encryption that could have helped pre-
vent the breach or mitigated the breach. And as a result, of course,
the data of 24 million fellow citizens was compromised.
Ms. MILLER. I do want to note that data breaches are going to
happen. They are just a fact of life. We aren’t going to be able to
say—the solution is not let’s not share personally identifiable infor-
mation because we don’t want to be compromised.
Mr. CONNOLLY. Right.
Ms. MILLER. It is going to get compromised, and every single
U.S. citizen’s data has already been compromised and for sale right
now on the dark web. The important thing here today, I think, is
to understand that they are already using this data. They have the
data. They are using the data against us, and we are limiting our
ability to fight them by protecting privacy to the degree that we are
protecting it right now at the expense of being able to use tech-
nology to fight fraud.
Mr. CONNOLLY. Thank you. Anything for the record, Mr. Hice?
Mr. HICE. No, I am good.
Mr. CONNOLLY. Seeing and hearing none, without objection, all
members will have five days in which to submit through the chair
additional questions for our witnesses. And I would ask should
such questions be presented to you, if you could try to answer them
in an expeditious fashion for the record that would be great.
Mr. CONNOLLY. If there is nothing else to come before us, we are
adjourned. I thank our witnesses.
[Whereupon, at 11:42 a.m., the subcommittee was adjourned.]
Æ
File and source
- File
- chrg_chrg-117hhrg47264_chrg-117hhrg47264.pdf
- Size
- 222,818 bytes
- SHA-256
- 426123ce60cfcefad6f47b9412b4390eaf836162a0347d8732b3218803a65f10
- Original
- www.govinfo.gov