Allocated and Unaccounted: The Federal Research Dollars That Simply Cease to Exist
In the federal government's accounting architecture, there exists a peculiar category of expenditure: the project that is funded, named, and then effectively abandoned to administrative silence. No published findings. No documented outcomes. No returned appropriations. Just a line item in a budget request, a disbursement confirmation, and then — nothing. For researchers, watchdog organizations, and ordinary citizens attempting to trace where their tax contributions actually go, this pattern is not an anomaly. It is a system.
The scale of unaccounted federal research spending is difficult to quantify precisely — and that difficulty is itself part of the problem. The Government Accountability Office has issued repeated warnings over the past two decades about the inability of major agencies to reconcile their own financial records. In 2023 alone, the Department of Defense failed its sixth consecutive audit, unable to account for trillions in assets and transactions. Within that broader accounting failure, research and development expenditures — which totaled more than $160 billion across the federal government in fiscal year 2022 — represent a particularly opaque subset.
The Phantom Project Problem
What distinguishes phantom research projects from ordinary bureaucratic inefficiency is the completeness of their erasure. Inefficient programs typically leave behind documentation: progress reports, contractor communications, internal evaluations. Phantom projects, by contrast, generate funding authorizations and little else.
Freedom of Information Act requests filed by independent researchers and journalists have repeatedly surfaced this pattern. An agency receives an appropriation for a defined research initiative. A project code is assigned. In some cases, a contractor or university is identified as the recipient of funds. Then the trail fractures. Deliverables are either never specified in the original contract language, or they are specified in terms so broad — "exploratory research," "preliminary assessment," "feasibility review" — that virtually any output, including no output, satisfies the agreement.
The National Institutes of Health, the Department of Energy, and the Defense Advanced Research Projects Agency have all faced scrutiny over grants that produced no publishable science, no patent applications, no technical reports filed with the awarding agency. In several documented instances, FOIA requesters have received responses confirming that responsive records simply do not exist — meaning the federal government itself cannot produce documentation of what a funded project actually accomplished.
How the Structure Enables Disappearance
The conditions that allow research funding to vanish without consequence are not accidental. They are embedded in the procurement and grant management frameworks that govern federal science spending.
First, the classification mechanism. A portion of federal research — particularly that administered through the intelligence community and certain defense programs — is shielded from public disclosure by classification designations. This is appropriate in some contexts. It is also a mechanism that can be applied preemptively to research that has no obvious national security dimension, effectively sealing the financial and scientific record from public view before any questions can be raised.
Second, the contractor intermediary. A significant share of federal research funding does not flow directly to scientists or institutions. It passes through prime contractors — large defense and consulting firms — who then subcontract work to smaller entities. Each layer of subcontracting introduces additional opacity. The federal agency's FOIA obligations apply to its own records, not to the internal documentation of private contractors. A project can be almost entirely insulated from public accountability simply by virtue of how its funding chain is structured.
Third, the reporting gap. Federal agencies are required to submit research outcomes to centralized repositories such as the NIH's PubMed database or the Department of Energy's Office of Scientific and Technical Information. These requirements, however, are inconsistently enforced. A 2021 review found that a substantial share of NIH-funded studies never resulted in a publicly accessible publication, and that the agency lacked the internal mechanisms to systematically identify or address non-compliance.
What FOIA Requests Actually Reveal
The Freedom of Information Act, in theory, provides citizens with the legal authority to demand accountability for how their money is spent. In practice, its utility in tracing phantom research funding is constrained by several factors that agencies exploit with apparent comfort.
Requests for financial records related to specific research projects are frequently met with one of three responses: a claim that the records are exempt from disclosure under Exemption 3 (statutory protections), Exemption 4 (commercial or financial information from contractors), or Exemption 1 (classified information); a response confirming that the agency conducted a search and located no responsive records; or a partial release that confirms a project existed and was funded, but withholds all documentation of its scope, methodology, or outcomes.
Each of these responses, taken individually, might represent a legitimate application of FOIA's exemption framework. Taken collectively, across hundreds of requests filed by different requesters targeting different agencies and different fiscal years, they describe something more systemic: a federal research enterprise in which accountability is structurally optional.
Independent analysis of USASpending.gov data — the federal government's primary public-facing expenditure database — reveals thousands of research awards for which outcome data is either absent or limited to a single sentence of project description. The database, while useful, was not designed to track scientific outputs. It tracks transactions. The gap between what was paid for and what was produced is not something USASpending.gov is equipped to close.
The Consequence of No Consequence
Perhaps the most significant finding in any examination of phantom research funding is the absence of institutional consequence for the agencies and contractors involved. Federal procurement regulations do include provisions for clawing back funds from contractors who fail to deliver on their obligations. In practice, these provisions are rarely invoked in the research context, particularly when the original contract language was sufficiently vague to make non-performance difficult to establish.
The result is a feedback loop. Agencies that fail to account for research expenditures face no meaningful penalty. Contractors who receive payment for work that produces no documented output continue to receive future awards. The researchers and institutions that attempt to surface these discrepancies through FOIA or public records requests encounter delays, exemptions, and stone walls. And Congress, which holds the formal power of the purse, receives budget justification documents that aggregate expenditures at a level of generality that makes project-level scrutiny essentially impossible without dedicated investigative resources.
The Public's Right to a Full Accounting
Federal research investment is not a discretionary corporate expenditure. It is a public commitment, made with public funds, in pursuit of public benefit. The scientific knowledge that emerges from that investment — or fails to emerge — belongs to the American people in a meaningful sense that extends beyond legal ownership.
When that investment produces nothing visible, and when the records that might explain why produce nothing retrievable, the public is left with a straightforward question that the government has so far declined to answer with any clarity: where did the money go?
The Full Disclosure Project will continue filing FOIA requests, cross-referencing expenditure databases, and examining the contractual structures that permit federal research dollars to move from appropriation to oblivion without leaving a trace. The accounting failures documented here are not inevitable features of large-scale science administration. They are choices — choices made by agencies, contractors, and oversight bodies that have collectively decided that transparency carries a lower priority than convenience.
That calculation deserves to be challenged, publicly and persistently.