Pandemic Aid Fraud Is Theft From Taxpayers and Should Be Prosecuted

If you hand out money with the paperwork requirements of a library card, you should not be shocked when the people who show up include some who are not very interested in reading. That is not cynicism. It is arithmetic. Any program that pays on a claim, and checks the claim later, if at all, is a program that advertises itself to everyone with a printer and no scruples.

In the spring of 2020 the federal government, facing a sudden national shutdown, did exactly that on a scale with no precedent. It was understandable. It was also, in a predictable fraction of cases, an invitation. The prosecutions that followed are not a partisan sideshow. They are the correct response to theft, and they illustrate a principle that conservatives have argued for a long time: money that the government spends is money someone earned, and the first duty of anyone who spends it is to make sure it goes where it is supposed to.

Define the terms

Fraud is obtaining money or property by deliberate deception. It is a crime, and it is a species of theft. Federal prosecutors often charge it as wire fraud (18 U.S.C. 1343), which covers a scheme to defraud carried out using interstate wires, including email and electronic transfers, or as money laundering, conspiracy and related offenses.

Pandemic aid means the stream of federal money authorized beginning in March 2020, including the Coronavirus Aid, Relief, and Economic Security (CARES) Act, signed March 27, 2020. The money moved through many channels: loans to small businesses, expanded and extended unemployment insurance, direct payments, and food and nutrition programs whose rules were relaxed to keep them operating through the shutdown.

Improper payment is a payment that should not have been made, or was made in the wrong amount. Not all improper payments are fraud. Some are errors. Fraud is the portion that involves intent to deceive, and the two categories are often reported together, which makes the headline numbers hard to read. I will avoid quoting a single grand total for that reason and will name what I can verify.

The case that shows the problem: Feeding Our Future

The clearest single illustration is the federal case commonly known as Feeding Our Future, centered in Minnesota.

The facts, as set out in the federal charging documents, are these. The federal Child Nutrition Programs, which include the Child and Adult Care Food Program and the Summer Food Service Program, reimburse approved organizations for meals served to eligible children. The money flows from the United States Department of Agriculture to a state agency, in Minnesota the Department of Education, which approves sponsors, nonprofit organizations that in turn oversee meal sites and submit claims for reimbursement. A sponsor is paid per meal served. The system works on a form of trust: the sponsor says how many meals were served, and the state pays.

During the pandemic, the federal government waived a number of the program's normal requirements, among them rules that required meals to be served in a group setting, so that children could be fed while schools and community sites were closed. Sensible. And a sponsor named Feeding Our Future, a Minnesota nonprofit, saw its reimbursements climb dramatically.

In September 2022, federal prosecutors announced charges in the case, alleging a scheme to steal approximately $250 million from the federal child nutrition programs, and described it as the largest fraud scheme involving pandemic relief money that had been charged to that point. Dozens of defendants were charged. The indictment alleged that the defendants created sham meal sites, submitted claims for enormous numbers of meals that were never served, and used fake rosters and invoices to support the claims. The money, prosecutors said, was then spent on luxury items, real estate and cars rather than on feeding children. A number of defendants have pleaded guilty. As with any prosecution, a charge is an allegation and each defendant is entitled to the presumption of innocence until a court rules, and I will say only what the public record supports.

That caution is exactly the point of the Anglo-American tradition we defend. The crime is real, and the process matters.

The oversight failures

The prosecution is the end of the story. The beginning is how the money got out the door, and here the lessons are for the people who run the state agencies and the federal programs.

1. Payment came before verification. The structure of the programs paid on claims, with the checking done by site visits and paperwork that could be falsified. When the number of claims grew by orders of magnitude, the capacity to verify them did not grow with it.
2. Warning signs were reported. According to public reporting, state officials had identified concerns about the sponsor and some of its sites before the charges, and the sponsor's response included litigation accusing the state agency of unfair treatment. A government agency that is sued every time it asks a hard question will learn to ask fewer of them. That is a perverse incentive, and it is worth examining in any program that distributes money through intermediaries.
3. The waivers removed friction. The pandemic waivers eased some of the program's requirements for good reasons. But each requirement removed also removed a check, and no one weighed the trade honestly.
4. Responsibility was diffuse. The federal government funded the program, a state agency administered it, a sponsor supervised the sites and the sites served the children. When a layered system fails, every layer says that someone else was supposed to check.

None of this excuses the defendants. A person who steals meal money intended for hungry children made his own choice. But a system that makes theft easy is a system that gets more of it, and that is a structural failure that the officials who ran it are answerable for. In a republic, those who spend the public's money answer to the public.

The same pattern in other programs

Feeding Our Future is vivid, but it is not an isolated case, and other prosecutions show the same pattern.

The unemployment insurance programs. Congress expanded unemployment benefits dramatically in 2020, with added weekly payments and eligibility for people, including the self-employed, who were not normally covered. States, which administer unemployment insurance, were overwhelmed by claims. Fraudsters, including organized rings, filed claims using stolen identities. The Labor Department's Office of Inspector General estimated that the amount of improper payments in the pandemic unemployment programs was very large, in the range of well over a hundred billion dollars, with a substantial share attributable to fraud. I give that as an order of magnitude, because estimates differ and include errors as well as fraud, and a careful reader should check the inspector general's current reports. Whatever the precise figure, it is an enormous sum of money paid to people who were not entitled to it.

The small business loan programs. The Paycheck Protection Program and the Economic Injury Disaster Loan program were designed to move money to small businesses within days. Prosecutors have charged numerous defendants with obtaining loans using fabricated businesses or inflated payrolls. Congress recognized the scale of the problem: in August 2022, it extended the statute of limitations for fraud in these two programs to ten years, giving investigators and prosecutors far more time to bring cases than the normal five-year federal limit.

The common thread is speed over verification. When the design pays first and asks later, the fraud runs faster than the investigators.

The steelman: speed saved families

Now the other side, stated as its defenders would want it. It is a strong argument, and conservatives should not wave it away.

In the spring of 2020, tens of millions of Americans lost work in a matter of weeks. Small businesses faced closure and mass layoffs. Food banks and school meal programs scrambled to feed children who could no longer come to the cafeteria. A government that demanded the verification we would expect in ordinary times would have sent money after the businesses had failed and the families had gone hungry. In an emergency, the cost of delay is measurable in evictions and empty refrigerators. The correct comparison is not fraud versus no fraud. It is fraud and rescue versus no rescue. A system tuned to catch every cheater would have let many honest people go under. Some leakage was the price of speed, and it was a price worth paying.

That argument has real force, and I accept the first half of it: in an emergency, speed has value and some leakage is inevitable. A government cannot run a program with zero loss.

Here is the answer, in numbered points.

1. Speed and basic controls are not opposites. Many of the cheapest controls take almost no time. Checking an applicant's Social Security number against records, comparing a claimed payroll to tax filings, flagging a thousand claims from one address or one bank account, and verifying that a meal site exists are things a database can do in seconds. Speed argues against site-by-site paperwork, not against automated checks.
2. Leakage at this scale is not "some." A leak that reaches hundreds of millions in a single program, and a far greater sum across all of them, is not a rounding error. The honest recipients were not helped by the thieves. They were hurt, because the money stolen was money that did not reach them, and money that will be repaid by taxpayers, current and future, as part of the national debt. You can follow that debt on the national debt clock.
3. The people who pay for fraud are the people the program was meant to help. The child who did not get the meal, the honest small business that was denied a loan because the fund ran out, the worker whose claim was delayed because the system was drowning in fake filings, all of them are victims of the fraudster as surely as the Treasury is.
4. Prosecution is how the deterrent is built. A crime that is not prosecuted is a crime that is advertised. If the next emergency program is also designed to pay first, the people planning to exploit it will look at how the last one ended. A credible threat of prosecution, with a long limitations period, is the only thing that stands between an emergency program and an open till.

Why this is a conservative cause, twice over

Two conservative principles meet here.

Law and order. Fraud is theft, and theft deserves prosecution. A conservative who thinks that shoplifting should be enforced should think that a ten-million-dollar scheme to steal from a children's meal program should be enforced harder. The principle is the same, and the scale is larger. We take up the retail version in Organized Retail Theft Is a Business Model, Not a Cry for Help.

Limited government. The conservative critique of government programs has always included the point that money spent by officials who did not earn it, on recipients they do not know, with no personal stake in the result, is spent less carefully than money spent by its owner. Milton Friedman's famous taxonomy of spending, which sorts spending by whose money it is and for whom it is spent, makes the same point in a sentence. Programs that distribute money through layers of intermediaries, and that reimburse claims rather than results, have exactly the weakness that theory predicts. Fraud is not an accident of the pandemic programs. It is an expected feature of that kind of design.

The Founders cared about this, too. The Constitution's Appropriations Clause, in Article I, Section 9, provides that "No Money shall be drawn from the Treasury, but in Consequence of Appropriations made by Law; and a regular Statement and Account of the Receipts and Expenditures of all public Money shall be published from time to time." A regular statement and account. The Framers understood that a republic keeps its government honest by keeping its books open. The principle applies to emergency spending as much as to ordinary budgets.

Controls that should come before the money

If prosecution is the answer after the fact, design is the answer before it. Here is a list of the controls that conservatives, and honest reformers of any party, should want built into every future emergency program, so that speed and integrity can coexist.

• Identity verification at the front door. Match applicants against Social Security and tax records, and use the same checks the private sector uses for credit.
• Automated cross-checks. Flag claims that share bank accounts, addresses, phone numbers or devices. Duplicate-detection software is cheap and effective.
• Pay on results where possible. Reimburse verified meals, not claimed meals; verify before the second payment even if the first goes out fast.
• Proportionate scrutiny. Large and fast-growing claims should draw more review, not less. A sponsor whose claims multiply tenfold in a year is not proof of success. It is a reason to look.
• Real consequences for the agency that fails to look. Inspectors general, legislative oversight and public reporting should follow the money and name who approved it. The federal government's oversight resources are described at the oversight page, and every citizen can follow the spending there.
• A long limitations period and dedicated prosecutors. Congress has already extended the period for the loan programs. The same logic applies elsewhere.

None of these requires a new bureaucracy. Most of them require a decision, before the emergency, that the integrity of the money is part of the design and not an afterthought.

The verdict

Prosecute the thieves. That is not harsh. It is justice for the taxpayers who funded the programs and for the honest recipients who were crowded out. And then fix the machine that let them in, because a country that responds to every fraud with a prosecution and never with a redesign will keep prosecuting the same crime forever.

A free government, Madison told us in Federalist No. 51, must first be able to control the governed, and then must be obliged to control itself. Pandemic fraud was a failure of the second half of that sentence. We can be generous in an emergency and still insist that the generosity come with a receipt.

For related reading on how the law treats the wider family of crimes that prey on public and private money, see Prosecutorial Discretion Has Limits, and Blanket Non-Prosecution Crosses Them and Equal Justice Under Law, From the Fourteenth Amendment to the Courthouse.

The case in brief

1. Fraud is theft, and theft from a program that feeds children or keeps small businesses open deserves prosecution, whatever the politics of the moment.
2. The Feeding Our Future case, charged in 2022 and alleging about $250 million in false claims on a federal child nutrition program in Minnesota, shows how a system that pays on claims and verifies later can be exploited at scale, and the same pattern appears in unemployment insurance and small business loan programs.
3. The steelman is that speed saved families, and that is true, but basic automated controls cost almost no time, and the victims of fraud include the honest recipients crowded out.
4. This is a conservative cause twice over: law and order demands prosecution, and limited government demands that money be spent carefully, with the Constitution's requirement of a "regular Statement and Account" as the standard.
5. The remedy is both prosecution and design: verify identity, cross-check claims, scrutinize fast growth, extend limitations periods and hold the agencies accountable.

What you can do: read your state's inspector general or auditor reports on how emergency money was spent, and ask your legislators what pre-payment controls they will require in the next program.