How a $64 Billion Pandemic Fraud Wave Exposed Weaknesses in Federal Relief Design

When the federal government launched the Paycheck Protection Program in the chaotic early weeks of the COVID-19 pandemic, speed was the overriding priority. Congress designed the program to push emergency loans to small businesses within days, before payroll obligations forced widespread layoffs and permanent closures. That urgency achieved its immediate goal, but according to new research, it also created one of the largest openings for financial fraud in the history of federal relief spending. A study published in September 2026 in the Journal of Accounting, Ethics & Public Policy examines exactly how that opening was exploited and what future emergency programs can learn from it.

The research was conducted by Dr. Katie Matt, Assistant Professor of Accounting at SUNY Polytechnic Institute, in collaboration with Daniel Hebert of Syracuse University. Their article, titled “Governance and Fraud Prevention in Federal Relief Programs: Insights from the Paycheck Protection Program,” analyzes the governance failures that allowed an estimated $64 billion in fraudulent claims, a figure reported by the U.S. Small Business Administration Inspector General. Rather than relying on statistical estimates alone, the researchers grounded their analysis in publicly prosecuted fraud cases, using court records to trace how specific schemes unfolded and which structural weaknesses made them possible.

The Paycheck Protection Program was administered through the Small Business Administration but delivered largely by private lenders, many of them financial technology companies that moved applications through automated pipelines at extraordinary speed. The researchers found that this delivery model, combined with limited verification requirements and minimal upfront screening, meant that applications could be approved with little more than self-reported figures. Borrowers attested to their payroll costs, employee counts, and business status, but in many cases no independent documentation was demanded before funds were disbursed. The design traded verification for velocity, and the study documents the consequences of that trade in detail.

Among the most common fraud patterns identified in the prosecuted cases were falsified financial records. Individuals inflated payroll expenses to qualify for larger loans, fabricated tax forms and bank statements to support their applications, and in some instances submitted the same fabricated documents to multiple lenders simultaneously. Because lenders were not systematically cross-checking applications against one another, duplicate and inflated requests could move through the system unnoticed. The researchers emphasize that these were not isolated acts of opportunism but recurring schemes that exploited predictable gaps in the program’s architecture.

Equally striking was the prevalence of fictitious businesses. In prosecuted cases, applicants created shell companies that existed only on paper, complete with fabricated employer identification numbers, invented employee rosters, and forged state registration documents. Because the program’s screening procedures were too weak to distinguish a legitimate small business from a convincing fabrication, these phantom enterprises received millions of dollars in federal funds. The study notes that once the money was disbursed, recovery became difficult, as proceeds were often spent quickly on personal purchases, real estate, luxury vehicles, and other expenditures far removed from payroll support.

The researchers argue that the root of the problem was not simply individual dishonesty but governance design. Emergency programs face an inherent tension: the faster aid must flow, the less time there is for verification, and the weaker the controls that can realistically be imposed. The PPP resolved that tension almost entirely in favor of speed, relying on after-the-fact audits and prosecutions to catch wrongdoing. The study shows that this retrospective approach caught many offenders, but only after billions of dollars had already left the treasury, and only against a fraction of the total fraud that investigators believe occurred.

From this analysis, Matt and Hebert derive a set of practical recommendations for policymakers who will inevitably design future disaster and economic relief programs. First, they call for improved applicant verification, including systematic cross-checking of applications against existing federal and state databases before disbursement. Second, they recommend centralized fraud prevention efforts, so that intelligence about suspicious applications is shared across lenders rather than trapped within individual institutions. The fragmented structure of the PPP meant that a fraudulent application rejected by one bank could simply be resubmitted elsewhere, a pattern the prosecuted cases illustrate repeatedly.

The recommendations also extend to the composition of the oversight workforce and the technology used to screen applications. The researchers advocate greater involvement of forensic accountants, whose training in detecting fabricated documents and tracing illicit fund flows is directly suited to the fraud patterns observed in the PPP cases. They further call for enhanced oversight of financial technology lenders, which processed a large share of program volume but operated under less traditional banking supervision. Finally, the study explores how artificial intelligence could strengthen future programs, using machine learning models to flag suspicious applications in real time by identifying anomalies in reported figures, duplicate documentation, and patterns inconsistent with a business’s stated industry or size.

The researchers are careful to frame these safeguards as complements to, rather than substitutes for, rapid aid delivery. Their central argument is that the choice between speed and accountability is a false one when controls are designed in advance. Automated identity verification, database cross-referencing, and AI-based anomaly detection can operate at the pace of an emergency rollout, catching the most blatant fraud without slowing legitimate applications. The lesson of the PPP, in their view, is that governance must be built into the architecture of a relief program from the start, not bolted on through prosecutions after the damage is done.

As federal agencies continue to recover funds and pursue cases stemming from pandemic-era programs, the study offers a timely evidence base for reform. The $64 billion in fraudulent claims represents taxpayer money intended for struggling businesses and their employees, and the researchers argue that preventing even a fraction of similar losses in future emergencies would justify the investment in stronger verification, centralized coordination, forensic expertise, and intelligent screening. Their work underscores a principle that extends well beyond one program: in national crises, the design of financial controls is not a bureaucratic afterthought but a determinant of whether aid reaches the people it was meant to protect.

Subject of Research: Fraud prevention and governance weaknesses in the federal Paycheck Protection Program

Article Title: Dr. Katie Matt examines fraud prevention in federal relief programs

Article References: Dr. Katie Matt examines fraud prevention in federal relief programs. (n.d.). Original publication

Image Credits: AI Generated

DOI: Not provided

Keywords: Paycheck Protection Program, fraud prevention, federal relief programs, COVID-19, small business loans, forensic accounting, artificial intelligence, government oversight, SUNY Polytechnic Institute, Small Business Administration, financial technology lenders, public accountability