The Invisible Crisis: 95,000 Federal Scientists Gone, and Your Grant Is Stuck in a System Running on Fumes

Congress preserved the budgets. It did not preserve the people who run them.

That is the dissonance at the center of the federal research enterprise in March 2026. The FY2026 appropriations package gave NIH $48.7 billion — a $415 million increase. NSF held near $8.8 billion. NASA's science directorate secured $7.25 billion, the largest allocation in nearly three decades. On paper, the money is there. In practice, the agencies tasked with reviewing proposals, managing awards, and disbursing funds have lost between 20 and 35 percent of their workforces in the past 14 months, and the consequences are now hitting grant seekers in ways that no budget line item can fix. (Granted News)

The Numbers Behind the Collapse

Between September 2024 and December 2025, approximately 95,000 employees departed federal science agencies, according to a Partnership for Public Service analysis — an 11.9 percent reduction across the affected workforce. More than one-third of all federal layoffs during this period fell on science-related positions.

The damage was not distributed evenly. The National Science Foundation lost over 30 percent of its staff, dropping from roughly 2,000 employees to about 1,300. NIH, NOAA, and USGS each shed approximately 20 percent. NASA lost 4,890 people, including more than 2,000 senior-level staff. The Department of Energy and NIST saw reductions of 17 and 15 percent, respectively.

The mechanisms varied — deferred resignation offers, probationary employee firings (later ruled illegal by courts, though reinstatement was not mandated for many), formal reductions-in-force, and a hiring freeze that capped new hires at no more than 2 percent of the total workforce while separation rates ran between 7 and 30 percent. The result was a one-way door: experienced personnel left and were not replaced.

Among the departures were more than 10,000 individuals holding STEM PhDs, according to Science magazine — program managers, grant officers, peer review coordinators, and the institutional memory that keeps a $192 billion federal R&D enterprise functioning.

What Happens When the Reviewers Disappear

The downstream effects for grant seekers are now measurable. Since the beginning of this administration, NIH has posted only 84 Notices of Funding Opportunities, compared with 787 in the prior year. That is not a rounding error. It is a 90 percent reduction in the pace at which new funding competitions reach the research community.

The backlog is compounding. Large academic research programs may not receive renewal decisions until 2027 — assuming those notices are approved at all. At the time of writing, $1.7 billion in NIH funding has been formally withheld, and over 2,200 grants totaling $3.8 billion have been canceled outright. Universities have responded by freezing hiring, delaying clinical trials, and, in some cases, shutting down laboratories entirely.

At NSF, the picture is structurally different but equally constrained. The agency acknowledged that its 1,300-person headcount is "too low" and announced plans to both boost staffing and reduce grant solicitations to half or fewer of historical levels. Chief Management Officer Micah Cheatham framed the reduction as efficiency — "the fewer solicitations you have, the less time grant applicants have to figure out which of our pigeonholes they fit into" — but the practical effect is that researchers will have fewer entry points to compete for funding, with each competition absorbing a larger and more heterogeneous applicant pool.

The loss of program officers is particularly acute. Two directors of separate NSF AI Research Institutes reported losing at least one program officer — the experienced staff who help grantees conceptualize research designs, offer feedback during the award period, and serve as institutional bridges between researchers and the agency's strategic goals. These roles take years to develop. The expertise walks out the door in an afternoon.

The Budget Paradox

The irony is sharp. Scientific R&D contracts across the federal government declined 23 percent in FY2025 compared to the prior year, dropping to $16.5 billion. Science agency project grants fell 24 percent, to $112.6 billion. CDC research contracts cratered 79 percent. The Fish and Wildlife Service eliminated its research contracting entirely — a 100 percent reduction.

Yet the appropriations bills that fund these agencies did not demand these cuts. Congress explicitly rejected the administration's proposed 40 percent reduction to NIH. It blocked the 15 percent cap on university indirect cost reimbursement rates. It preserved Pell Grants, TRIO programs, and research agency baselines.

The disconnect exists because the executive branch controls hiring, organizational structure, and the pace of obligation — the mechanics of translating appropriated dollars into issued awards. Funds can be appropriated and still not spent. Positions can be authorized and still not filled. The result is an agency system that has the money on its balance sheet but lacks the personnel to process it through the pipeline.

What This Means for Your Next Proposal

For researchers navigating this environment, several practical realities have changed.

Expect longer timelines. Award notification cycles that historically ran 6 to 9 months may now stretch to 12 to 18 months, particularly for new and renewal R01-equivalent awards at NIH. Budget accordingly — gap funding from institutional sources, bridge grants, or diversified portfolios that do not depend on a single agency's timeline are no longer nice-to-have.

Fewer competitions mean fiercer ones. When NSF consolidates solicitations, the applicant pool for each remaining competition grows. Broader solicitations demand proposals that clearly articulate both intellectual merit and programmatic fit, because program officers triaging a larger inbox have less time per application. Sharpen the first page.

Program officer relationships matter more, not less. The surviving program staff are carrying heavier portfolios. A well-timed inquiry — asking about programmatic priorities before submitting, confirming that your research fits the solicitation's intent — signals professionalism and reduces the officer's burden. Do not waste their time with generic questions that the solicitation answers. Do invest in a pre-submission conversation that helps both of you.

Diversify across agencies and mechanisms. If your research touches energy, health, and national security — many AI, materials, and climate projects do — submit to DOE, NIH, and DOD in parallel rather than depending on a single pipeline. Each agency's staffing situation is different. DOE's Office of Science is still running active solicitations. DOD's SBIR restart is underway. NIH's timeline is the most constrained.

Track state-level alternatives. Massachusetts, Texas, New York, Pennsylvania, and California are collectively mobilizing billions in state research funding specifically because the federal pipeline is constricted. These programs prioritize in-state institutions and research with regional economic relevance — framing your work in those terms opens doors that did not exist two years ago.

The Longer-Term Risk

Federal science agencies are not just understaffed. They are losing the knowledge infrastructure that makes grant programs function — the institutional memory of how review panels are assembled, how programmatic priorities translate into funding announcements, how interagency coordination works on cross-cutting initiatives.

Rebuilding this capacity, even under the most favorable conditions, takes years. New program officers need two to three years to develop the expertise and professional networks that make them effective. Study section coordinators, grants management specialists, and compliance officers require similar ramp-up periods. The 95,000 departures represent not just headcount but accumulated judgment, relationships, and operational know-how.

For the research community, the message is not to wait for the system to heal. It is to build resilience into every funding strategy — diversifying sources, maintaining strong relationships with the program staff who remain, and planning for a grant cycle that is slower, more competitive, and less predictable than anything in recent memory.

The budgets were preserved. The capacity to execute them was not. Adapting to that reality is now the central challenge facing every institution, laboratory, and principal investigator competing for federal research dollars, and tools like Granted can help researchers identify alternative funding pathways while the traditional pipelines rebuild.