The Treasury Department intercepted about $99 million in federal payments associated with people listed as deceased before the money left government accounts. More than 4,900 disbursements were flagged for agency review. Treasury disclosed the screening results on July 21, 2026, after applying the expanded check to payments processed since March 2025.
The review covered 885 million transactions with a combined value near $2.7 trillion. That total describes the payment stream examined, not the amount suspected of being improper. The stopped sum represented only about 0.0036% of the money screened. A very small mismatch rate can still produce a substantial dollar total when the underlying flow reaches trillions, which is why automated comparison is useful even when almost every transaction clears.
Officials did not simply cancel each transaction after a computer match. The Bureau of the Fiscal Service returned flagged items to the federal agencies that originated them. Those agencies must determine whether the intended recipient had died, whether records were current and whether another person was attempting to collect money improperly. The $99 million is therefore best understood as money held before release, not as a final finding that every recipient committed fraud. That distinction matters when agencies report prevention as savings. Treasury did not publish a program-by-program breakdown.
Death-File Access Moves the Check Ahead of Disbursement
The screening uses the federal Do Not Pay program, which checks identity, eligibility and bank information before an award or payment is released. Treasury combined that system with additional verification tools. The operational change is the timing: an agency receives a warning while it can still stop the transfer rather than trying to recover money after it reaches an account.
Death records are a central part of the comparison. Congress opened the death database to Treasury on a temporary basis in 2021. The records come from Social Security’s Full Death Master File. A law signed in February made the connection permanent. The legal change removes an expiration point from a data link used to identify recipients who may no longer be eligible. It also allows the control to become a standing part of payment processing rather than a temporary project awaiting renewal.
Permanent access does not eliminate the need for review. Names, dates of birth and Social Security records can contain errors, while benefits may lawfully pass to an estate or another eligible person after a death. Returning a payment to its agency creates a verification step between an automated alert and a final decision.
The flagged group averaged roughly $20,000 per transaction when the approximate $99 million total is divided by 4,900 payments. That figure is only a scale indicator because Treasury reported more than 4,900 cases rather than an exact count. It also does not reveal whether the distribution was dominated by a small number of large payments or spread evenly across programs.
Small Error Rate Still Produces a Large Recovery Target
Treasury said the blocked amount was more than triple the value identified before the current administration took office. The comparison indicates that wider screening found more questionable payments, but it does not by itself measure the underlying fraud rate. The department did not provide the earlier transaction count or screening volume needed for a like-for-like rate comparison. A larger total can result from better detection, a broader search or growth in the payment volume being examined.
The department projects $330 million in net benefits from reducing payments tied to deceased recipients. That projection is larger than the first reported block because it covers expected performance over time rather than a single review period. Net benefit should account for the cost of operating the controls as well as money prevented from leaving the government. Confirming the estimate will require published results showing how many flags agencies uphold, how many they clear and how much remains unrecoverable after mistaken payments.
The program’s value lies in turning a fragmented set of agency payments into a common verification problem. A shared death-file check can catch inconsistencies that individual programs miss, while agency review protects legitimate recipients from automatic denial. The durable measure is whether Treasury can document accurate flags, low false-positive rates and sustained savings without delaying valid federal payments, not the size of one announcement.