USAID’s Award Cuts Look Different When Counted in Dollars: Two Regional Groups Reverse the Pattern

Two groups retain a majority of historical dollars, but a minority of awards

What does terminating most foreign-assistance awards tell the public about the financial size of the portfolio that remains? Less than the count alone suggests. What’s the Scoop With Broach compared the award-count and cumulative-obligation panels for all seven regional groups in a September 30 GAO report. In every group, terminated USAID awards outnumbered active awards. But in two groups—the Near East and the worldwide portfolio—the active awards carried more cumulative obligation value than the terminated ones.

That is the distinctive finding from our regional cross-comparison: a majority of awards ended in those two groups, yet a majority of the historical obligation value was attached to awards still active in March 2026. The Near East category includes the Middle East and Northern Africa. Worldwide is a separate portfolio grouping, not an eighth geographic region and not the overall total.

The other five groups had both more terminated awards and more cumulative obligation value associated with terminated awards. Counting program decisions and counting historical dollars answer different accountability questions. Neither measure in this chart establishes how many people lost services, how much money was saved, or how much funding remained unspent.

The remaining awards were much larger on average

The nationwide totals help explain the regional reversal. State Department data supplied to GAO show about 6,780 terminated USAID awards and about 730 active awards out of approximately 7,510 that were active on or after January 2025. Those two groups carried $79.1 billion and $52.1 billion, respectively, in cumulative obligations. The combined value was $131.2 billion.

Using those rounded figures, we calculate average cumulative obligations of approximately $71.4 million per active award, compared with $11.7 million per terminated award. The active-award average was roughly 6.1 times the terminated-award average. This is an arithmetic mean of historical obligations, not the typical award, a median or a measure of future spending commitments.

Active awards represented about 9.7% of the reported award count but 39.7% of cumulative obligation value. GAO itself summarizes those shares as about 10% and 40%. Our added calculation quantifies the difference in average financial scale and connects it to the two regional exceptions. A small number of large awards can carry a substantial share of a portfolio’s historical dollars. That fact does not establish whether retaining or terminating any particular award was justified.

The report separately notes that 639 transferred active awards expired between September 2025 and March 2026, according to State’s inspector general. We do not relabel those expirations as discretionary terminations or add them to the 6,780 termination total. An award reaching its end date is not automatically the same action as cancellation.

The $79.1 billion figure is not a savings estimate

The award start dates in this comparison span calendar years 2015 through 2026. Obligations accumulated from each award’s start until March 2026, or its termination date. The $79.1 billion therefore includes obligations made before the 2025 policy changes. It is not one year’s foreign-assistance budget.

GAO expressly cautions that cumulative obligations attached to terminated awards do not represent savings, money returned to the government or future obligations eliminated by termination. The corresponding $52.1 billion for active awards is likewise not a cash balance available for future distribution. Our headline’s reference to dollars describes which status group carries historical obligation value, not a current bank account.

Europe and Eurasia had the largest cumulative dollar amount associated with both terminated and active USAID awards. Sub-Saharan Africa had the largest terminated-award count in the chart. Those rankings are compatible: the region with the most ended awards need not be the region with the most historical dollars attached to ended awards. The figure does not identify the individual awards needed to assess concentration within each regional bar.

The agency inheriting the portfolio also had inconsistent status records

State announced that USAID ceased administering foreign assistance on July 1, 2025 and took responsibility for the remaining active awards. The same audit identifies a separate data problem within State’s own MyGrants system: more than 4,000 of around 10,000 grants and cooperative agreements supplied to GAO appeared as both active and terminated, despite being terminated.

Those MyGrants entries are not the 7,510-award USAID denominator in our calculations. Combining the two would mix distinct portfolios and produce an unsupported result. GAO obtained clarification from State that allowed it to identify terminated awards through other fields and remove duplicates, but State did not provide an updated, accurate dataset during the review.

State explained that some performance end dates had not been updated because administrative actions followed varying timelines. Officials also cited the volume of information requests. These explanations matter: the audit documents inconsistent status information, not proof that duplicate payments occurred or that officials deliberately concealed awards. GAO warned that poor status data can undermine oversight and decisions about which programs to continue.

State’s response and what this examination cannot establish

In its written response, State agreed to work with the MyGrants system managers and program bureaus to make status information accurate, consistent and current. It emphasized that program bureaus perform data validation. State also said it could generate detailed workforce information and would consider GAO’s guidance recommendation in future planning and budgeting. GAO acknowledged the response but said State had not demonstrated it could supply the requested workforce composition data during eleven months of requests. All three recommendations remained open on GAO’s public page when checked October 5.

This investigation is an analysis of published government records. We read the USAID methodology and count-and-dollar table, compared both bars in each of the seven regional rows, recalculated the nationwide shares and means from the printed totals, and examined the MyGrants findings and State’s written response. The regional classification uses only whether the active or terminated bar is larger; we do not manufacture exact regional numbers from unlabeled bar lengths.

The records establish an uneven financial composition of the surviving portfolio and a separate problem in the receiving department’s management data. They do not measure the humanitarian effects of the changes, certify cost savings or establish that March award statuses remain unchanged today. No interviews, visits or contacts by Billy Roach are claimed. Evaluating the merits of particular terminations would require award-level decisions, performance evidence and current implementation records beyond this report.