Every time a major fraud case involving taxpayer dollars makes news, public reaction follows a familiar pattern: outrage that funds were stolen, disbelief that the government “didn’t see it coming,” and demands for agencies to tighten oversight.
Those reactions are often justified. But they glide past an important reality: Preventing fraud and error requires tradeoffs that affect everyone, not just bad actors.
That dynamic was on display at a recent House Judiciary hearing focused on fraud cases in Minnesota—including a pandemic-era program that prosecutors allege is one of the country’s largest fraud schemes ever. Now the Senate is ramping up scrutiny with the announcement of a new fraud task force, and a trial is underway in Mississippi involving a professional wrestler allegedly misappropriating public funds intended for needy families.
These government probes serve an important role in our system, providing accountability over how taxpayer dollars are spent. What these exercises in public scrutiny too often lack, however, is a clear accounting for the tradeoffs necessary to stop fraud.
Preventing erroneous payments almost always means asking more of everyone, including legitimate recipients—more reporting, more documentation, and more delays to allow for intensified government review. When we choose to limit those burdens and get payments out the door quickly, we choose to accept a higher risk of fraud and error.
Emergency programs starkly expose the tension between burden and prevention. In crises, speed is paramount. Payments go out before controls are in place. Bad actors notice this and exploit the government’s fast response. That’s what appears to have happened in Minnesota.
In my various roles with the US Office of Management and Budget, or OMB, and later as IRS commissioner, these tradeoffs surfaced again and again in efforts to prevent fraud and error. Technology and data science have helped at the margins, but they haven’t eliminated the underlying tension between speed and risk.
The earned income tax credit, or EITC, was a recurring example. This program has one of the highest payment error rates of any federal program—not due to bureaucratic ineptitude, but because eligibility turns on facts the government can’t independently verify.
To claim the credit, a taxpayer must have lived with a qualifying child for more than six months of the year. Yet there’s no national database that tracks where children live.
I used to say—in jest—that the IRS could solve the EITC error problem by issuing every parent a card and requiring them to swipe when they entered or exited their home with their child. The IRS could then verify child residency in real time.
Of course, this would rightly strike most Americans as an unacceptable invasion of privacy. But the hypothetical makes a worthwhile point: If residency is required and we reject systems that could verify it, we knowingly accept some degree of error. Therefore, not all forms of payment error are a failure—some are the result of a choice.
Minnesota is also grappling with additional accusations of potential fraud and error in publicly subsidized childcare, where payments allegedly were made based on attendance that never occurred. Government authorities have opened investigations, but the scope of the alleged abuses, including claims circulating online, haven’t yet been verified.
Regardless, some have asked why governments don’t require more rigorous, real-time verification of child attendance. Some states already do—something uncomfortably close to the very idea I just described as politically untenable.
Back in my “improper payments czar” days at OMB, officials from Oklahoma came to Washington, DC, to share a best practice. The state had implemented card-swipe technology for subsidized childcare, requiring parents to physically swipe their child in and out of childcare centers to validate attendance.
During the briefing, I asked them the obvious questions: Did parents complain about privacy? Was there outrage about government tracking their children?
The answer surprised me. The biggest complaint, the Oklahomans reported, wasn’t about intrusive data collection; it was that parents no longer could do “kiss and ride.” They had to park, walk in, and swipe.
Despite this, as well as frustrations over periodic app glitches, and the need to avoid penalties for routine child absences (given subsidies are often tied to enrollment), state officials concluded the tradeoff was worth it. After the success of this program, New Jersey adopted a similar approach.
Notably, even though the bad actors in these particular cases are a small number of dishonest service providers, it is the families that shoulder some of the extra burden when addressing the abuse.
The situation in Minnesota has grown tragically more complex. The fatal shootings of Renee Nicole Good and Alex Pretti by federal immigration enforcement agents have rightly become the central focus of public attention and debate over how government power is exercised, particularly in the realm of policing and enforcement. The situation is a sobering reminder that government action carries consequences far beyond balance sheets.
Even as these tragic deaths rightfully command the national conversation, the ongoing congressional probe into fraud and error in Minnesota, Mississippi, and other states will have significant implications. Decisions, such as moratoriums on federal childcare subsidies, will affect honest and well-meaning families and providers across the country.
We shouldn’t anchor decisions on how to address fraud and error in emotion or politics. These discussions demand a reflection on what history has taught us about the tradeoffs involved in prevention.
I hope the debate over fraud and error becomes more candid about those tradeoffs. Speed, privacy, and simplicity aren’t free—and reducing fraud requires deliberate choices about burden and delay.
We’ve made this choice before with notable results. The American Recovery and Reinvestment Act—a nearly $800 billion emergency stimulus program enacted in 2009—had comparatively low levels of fraud and error. This wasn’t by accident; the law required, as a central feature, extensive and frequent reporting by recipients that gave federal officials and auditors the information they needed to spot problems early.
For states now wrestling with how to respond to fraud concerns, it may be worth revisiting what Oklahoma learned years ago: Sometimes the most workable spot on the continuum isn’t the one that eliminates all risk, but the one the public is willing to live with.
Danny Werfel has twice served as IRS commissioner, most recently from 2023 to 2025. He is now executive in residence at the Johns Hopkins School of Government and Policy and a distinguished fellow at the Polis Center for Politics at Duke University, writing about the intersection of tax and policy.
Read More By All Accounts
To contact the editors responsible for this story: