Key takeaway: Federal agencies move toward verifying payments before they go out rather than clawing back funds afterward. Payment processors and risk teams working with government-linked programs should expect tighter identity- and account-verification requirements. They can also look to Treasury’s expanded data-sharing model as a reference point for reducing fraud and improper-payment losses on their own payment rails.
The U.S. Department of the Treasury’s new payment verification process identified and returned 13,500 payments worth $175 million that would have gone to deceased individuals during fiscal year 2026. This is a part of a broader fraud- and improper-payment-screening push that verified $3.7 trillion in federal payments, Treasury Secretary Scott Bessent said in an Oct. 6 announcement.
The results come from two parallel efforts the Treasury advanced during the fiscal year that ended Sept. 30. These are the existing Do Not Pay program and a new, separate payment verification process.
Through Do Not Pay, Treasury expanded federal programs’ access to its fraud-screening data more than twentyfold, from about 4% of programs at the end of fiscal 2025 to roughly 99% now, with most remaining programs expected to gain access in early fiscal 2027. The department also added nine new data sources to Do Not Pay. These include company registration records from OpenCorporates and select verification against the Social Security Administration’s Numident data. They also screened more than 2.3 billion records against Do Not Pay data sources, nearly four times the 641 million screened the year before.
Separately, the new payment verification process screened more than 1.1 billion payments, totaling the $3.7 trillion verified. And that screening is what caught the 13,500 payments worth $175 million headed to deceased recipients. Bessent said the department is moving beyond “pay and chase” and making prevention the government’s first line of defense.
Why it matters: Treasury’s shift from recovering improper payments after the fact to verifying them before disbursement mirrors the proactive screening approach increasingly expected across the payments industry. With the Government Accountability Office estimating federal fraud losses of $233 billion to $521 billion a year, and the Consumer Federation of America estimating consumers lose roughly $119 billion annually to scams, Treasury’s expanded data-sharing model offers a case study in using identity verification to stop improper and fraudulent payments before funds move.