State Can't Trace a Third of Its $1.2B China Fund

Since fiscal year 2020, Congress has directed the State Department and USAID to spend a minimum of $1.625 billion⧉ countering the influence of the People's Republic of China worldwide, through the Countering PRC Influence Fund (CPRCIF) and its predecessor -- described by State officials as the only U.S. government money earmarked specifically for that purpose.▸ The two agencies funded about 470 projects worth roughly $1.2 billion in allocations⧉ between fiscal 2020 and 2023, according to the Government Accountability Office⧉ -- Congress's own auditor, whose recommendations agencies are expected to act on.

GAO's June 17, 2026 review found the interagency working group that runs the fund cannot produce a reliable record of what much of that money bought: no award number for roughly a fifth of the tracked projects, no start-or-end date for about a quarter, and no obligation record -- the point at which committed money becomes legally locked in, the step between a promise and a payment -- for about a third.▸ Nearly one in three of the estimated 430 approved funding proposals has no traceable link to the specific project it was supposed to become.

CPRCIF isn't a single pot of money -- it's an instruction layered onto five existing foreign-aid accounts (Economic Support Fund, Development Assistance, Foreign Military Financing, and two security accounts), each fiscal year, with Congress setting an overall floor and sometimes a floor for one account inside it.▸ For fiscal 2024, for instance, Congress required at least $400 million total and specified that at least $155 million of it come from the Economic Support Fund alone. An interagency working group -- State and USAID staff with no CPRCIF-dedicated budget of their own -- collects competing proposals from bureaus and posts worldwide each year, scores them against declared "lines of effort" like countering PRC economic coercion or military expansion, and decides what gets funded. Congress kept the mandate alive well past the years GAO reviewed: the enacted fiscal 2026 State/Foreign Operations bill⧉ locks in another $400 million for CPRCIF specifically. The tracking problem below sits underneath a fund that is still growing, not one being wound down.

The Economic Support Fund carried the largest share of the money GAO could account for -- $380 million across 220 projects -- ahead of Development Assistance, security-assistance accounts, and international-narcotics funding.▸ That much is traceable. What isn't: GAO asked the working group for the same basic fields it would need to manage any $1.2 billion portfolio -- project name, award number, implementing office, start and end dates, and how much of the money was actually obligated. Officials could not supply that data completely for a third to a fifth of the roughly 466 projects they could otherwise identify, depending on the field.

GAO's own assessment of the consequence is not about fraud or theft -- nothing in the report alleges either -- it's that working group officials lack the information to tell whether an actual project reflects what a bureau proposed, to track a project's status, or to avoid funding the same effort twice, which is exactly the standard Standards for Internal Control in the Federal Government sets for any agency managing public money: obtain and maintain reliable data in a timely way, so it can inform decisions.

GAO's own examples show the mechanism, not just the tally.▸ A post in the Western Hemisphere submitted a $500,000 proposal in November 2023⧉ to fund optical-telescope upgrades meant to prevent PRC espionage; the working group didn't approve it until May 2024 -- six months later, too late for the post to notify Congress and obligate the money on schedule. State eventually redirected the $500,000 to a different, already-approved cybersecurity project elsewhere in the region. Separately, a post in Europe identified cost savings on an approved project just before its 2023 obligation deadline, but the proposal calendar left no time to reallocate the freed-up money -- and as of March 17, 2026, those funds were still sitting unspent in State's budgetary system. Neither case is fraud; both are a governance structure moving slower than its own deadlines.

State did build a check meant to catch problems like these before they happen: 20 Regional China Officers, foreign service officers with China expertise posted across State's six global regions since 2019, specifically to weigh in on counter-PRC proposals.▸ Annual guidance has never required them to review one.

State's own annual evaluation plans -- the disclosures required under the 2019 Foundations for Evidence-Based Policymaking Act -- have not listed a single CPRCIF-specific evaluation since fiscal 2022, the earliest year for which those plans exist.▸ State and USAID started building a portfolio-wide scoring framework in 2023 to close that gap, holding workshops with bureaus through 2024 and beginning to gather feedback on draft performance indicators by February 2025. Then a foreign-aid funding pause tied to a January 2025 executive order hit: State put the framework contract on hold that same month. As of March 2026 -- more than a year later -- State officials told GAO they were still "assessing whether and how" the agency could resume the work. Until it does, there is no agency-wide way to say whether $1.2 billion in counter-China spending changed anything the fund was built to change.

GAO made five recommendations, all directed at the Secretary of State through the department's Office of China Coordination: release CPRCIF guidance earlier each year, require documented input from regional and issue-area experts on every proposal, require Regional China Officers to actually review proposals from their regions, consistently collect and update basic project data, and build the results-assessment process the agencies started and shelved.▸ State concurred with all five. Concurrence means State agreed with the finding -- it commits the department to nothing with a date attached. GAO's standard practice is to report back only once it has independently confirmed an agency actually took the action; as of this report, it had not yet done so for any of the five.