Where Does the Money Go? A Watchdog Just Found That $712 Billion in Federal Grants Can't Be Traced to a Place — and Why That Should Change How You Report

There is a question that sounds simple and turns out to be nearly unanswerable: when the federal government sends a grant to a state, a university, or a nonprofit, where does that money actually land? Not which agency signed the check, and not which organization cashed it — but which community, which county, which ZIP code ultimately received the benefit. In July 2026, the Project On Government Oversight (POGO) published a report that put a number on how badly the government fails to answer that question. The finding: more than 74% of federal grant spending — approximately $712.6 billion — could not be traced to a county, city, or ZIP code.

That number is now doing political work. It surfaced in the August appropriations fight, in transparency op-eds, and in the broader accountability debate that has followed federal spending since 2025. For grant seekers and grant recipients, it is easy to read a report like this as a Washington problem — a data-plumbing failure between agencies and USAspending.gov that has nothing to do with a program officer in the field. That reading is a mistake. The traceability gap is about to become your problem, because the fix for a transparency failure always flows downhill to the people who report. This is the deep dive on what POGO found, why it matters, and how to position for the reporting regime it is going to produce.

What the report actually measured

POGO examined the 60 largest domestically focused federal-assistance programs, which together accounted for 92.3% of all federal assistance spending in fiscal year 2024 — in other words, not a sample of odd corners but the overwhelming bulk of the money. It then asked a single, concrete question of each dollar: can this spending be located geographically, down to a county, city, or ZIP code?

The answer was mostly no. According to the report:

• Only 20.7% of grant spending was locally traceable through prime-award information.
• Subaward data added roughly another 5%, bringing the traceable total to about a quarter.
• The remaining ~74% — about $712.6 billion — went dark once it left the prime recipient.

The structural reason is the "pass-through" problem. A large share of federal grant money does not go directly to the entity that ultimately spends it. It flows to a state government, a state agency, a university, or a large intermediary — the prime recipient — which then re-grants it to subrecipients, contractors, and local programs. USAspending.gov captures the prime award cleanly. The subaward layer is reported inconsistently, and anything below that layer — the actual on-the-ground spending — is largely invisible. The money is not missing in the sense of being stolen; it is missing in the sense of being unlocatable. Watchdogs can see that $100 million went to a state agency; they often cannot see which towns the state agency spent it in.

Why an accounting gap becomes a political weapon

Transparency failures are quiet until someone decides to make them loud. In 2026, several forces have converged to make grant traceability loud:

The accountability politics of federal spending. Since 2025, "we can't even follow the money" has been a recurring line in the argument for tighter grant oversight — and it cuts across party lines. A finding that three-quarters of grant spending cannot be geolocated is precisely the kind of fact that gets cited by everyone from budget hawks arguing for cuts to good-government reformers arguing for better data. When a $712 billion number is available to both sides of an argument, it stops being a technical footnote and becomes a talking point.

The OMB rule's risk framework. The proposed federal grants regulation — currently frozen until December 11 but not gone — leans heavily on risk assessment, recipient compliance history, and the ability of agencies to justify that awards "effectuate program goals." An award you cannot trace to a community is an award you cannot easily defend as effective. The regulatory direction and the transparency finding point the same way: toward recipients being asked to prove, geographically and concretely, where their impact landed.

The technology finally exists. For years, the excuse for poor geospatial reporting was that it was too hard. That excuse is thinning. Geocoding, standardized location identifiers, and subaward reporting tools are mature. When capability catches up to demand, the burden of the gap shifts from "the system can't do this" to "why aren't recipients doing this."

What this means for grantees — the reporting turn is coming

Read the trend line and the conclusion is hard to avoid: the era of reporting grant outcomes in narrative summaries and aggregate dollar figures is ending, and an era of place-based, geocoded accountability is beginning. Recipients who get ahead of that shift will find it a competitive advantage; recipients who wait for it to be mandated will scramble.

Here is where to focus.

Know your own pass-through chain. If you are a prime recipient who re-grants to subrecipients, you are exactly the layer where the data goes dark. Start mapping, now, where your subawards actually land — by county and ZIP, not just by subrecipient name. If you are a subrecipient, understand that your prime is increasingly going to be asked to account for you geographically, and being the subrecipient who can hand over clean location data makes you easier to fund and easier to renew.

Geocode your impact by default. Build location into your outcome reporting as a habit, not an afterthought. For every dollar and every beneficiary you can, record the place: the ZIP code served, the county reached, the census tract of the program site. This is not busywork — it is the raw material of the story funders increasingly want, and it is exactly the data POGO found missing.

Turn traceability into a proposal asset. When you apply, do not just describe what you will do; describe where, with specificity, and commit to reporting it that way. In a funding environment shaped by "we can't follow the money," a proposal that says "here is precisely how we will show you which communities this reaches" is answering a question reviewers are under growing pressure to ask. It signals low risk and high accountability — two things every current policy signal rewards.

Instrument for subaward reporting before it's mandatory. The Federal Funding Accountability and Transparency Act already requires subaward reporting above certain thresholds; the POGO finding is, in part, a story about how unevenly that requirement is met. Assume enforcement and completeness expectations rise. Grantees who already have clean subaward and location data pipelines will absorb tightened rules with a shrug; those who don't will face a costly retrofit under deadline.

Watch for the data mandate in the rulemaking. The transparency gap is the kind of finding that gets converted into a reporting requirement. Whether it arrives through the OMB grant rule, through appropriations report language, or through agency-specific NOFO terms, expect geospatial and subaward reporting expectations to tighten over the next several cycles. Building the capability now is cheaper than building it against a compliance deadline.

The bigger picture

The POGO report is, on its surface, a critique of government systems — of USAspending.gov's blind spots and the pass-through architecture that hides the last mile of federal money. But the practical consequence lands on recipients. When a watchdog demonstrates that $712 billion in public spending cannot be located, the political and regulatory response is never "the government will simply tolerate that." It is "recipients will tell us where the money went." The organizations that thrive in the next few years of federal funding will be the ones that treat place-based transparency not as a compliance burden imposed from above but as a core part of how they prove their work matters. The money can be traced. Increasingly, you will be the one expected to trace it — and the grantees who start now will be the ones with the answer ready when the question is finally asked.