The Federal Research Funding Pipeline Is Breaking While the Cash Still Flows
The headline numbers on federal science funding look almost calm. The pipeline underneath them does not.
I presented these slides at the NCRAA conference last month at UNC-Chapel Hill: a data drill-down on federal and foundation funding for university research, with federal data through June 2026.
What the data shows:
1. Total federal obligations to universities are down only 7 to 8 percent. That is the number that can make leadership feel safe, and it is the wrong number to watch. Obligations are dominated by continuations on awards signed years ago. They are backward-looking.
2. The leading indicator is new-award commitments, and those are down roughly 20 percent per year, 35 percent over two years. I re-ran the analysis this week using the August USAspending data release, and the decline holds. Nothing in the latest month suggests a reversal. The pipeline is breaking while the cash still flows.
3. And the break is not evenly distributed. NIH institutes moved more than 100 percentage points apart (NHLBI new-award dollars down 69 percent). Every NSF directorate fell, some by more than 90 percent. Every NIH grant type declined in count, even as front-loaded awards kept research-project dollars roughly flat. The FY25 R01 success rate was 13 percent, NIH's own figure.
4. Can philanthropy fill the gap? Partly, and only with precision. Universities received about $22.6 billion in foundation grants in 2023, but strip out captive university foundations, corporate operating entities, DAFs, and pass-throughs, and the prospectable flow is about $10.5 billion. Relief, not replacement.
The deck ends with the playbook we discussed: route federal effort toward the agencies, institutes, and mechanisms still functioning; target philanthropy at the specific gaps federal opened; treat philanthropic grants as seed money, not as a substitute for recurring federal support.
If you lead research administration, sponsored programs, or development at a university, the slides are meant to be borrowed from. Take what is useful.
Disclosure: Independent analysis, presented in a personal capacity, not as part of my Duke University role. SciRise is an independent venture and is not affiliated with, sponsored by, or endorsed by Duke University.
