NSF Pulled the OMB-Only Language Out of Its New Grant Rulebook. The Retreat Is Conditional, and the Last Comment Window Closes in Late November
November 4, 2026 · 8 min read
Granted Research Team · Editorial policy
The most important change to NSF grant policy this fall was announced in a two-page Paperwork Reduction Act notice. NSF published it in the Federal Register on October 28, 2026 (91 FR 67875), and it is the agency's first real step back from the government-wide grant overhaul that has worried research institutions since spring.
The notice says NSF has sent its revised Guidance on Financial Assistance (GFA), the plain-language replacement for the Proposal and Award Policies and Procedures Guide (PAPPG), to the Office of Management and Budget for clearance. It also says the agency received 3,277 responses to the June draft and publicly posted 3,075 of them. The rest were duplicates or out of scope. "All in-scope comments have been considered in the development of the proposed version," the notice says.
According to the American Institute of Physics' FYI policy service, which reviewed the agency documents, the revised GFA removes language NSF had taken solely from OMB's proposed rewrite of 2 CFR Part 200. Four provisions are gone:
- The elimination of fixed-price awards
- The disallowance of publication costs
- Termination for convenience
- Making indirect-cost recovery subject to the availability of funds
NSF also says it "will continue to follow its established merit-review process," and that it will deal with future Uniform Guidance changes only after the government-wide requirements are final.
This is a meaningful reversal. It is also a conditional one, and the conditions matter more than the headline.
How we got here
In September we described the gap that this revision partly closes (Congress Froze the OMB Grant Rule Until December 11. NSF and the Education Department Are Writing It Into Their Own Rulebooks Anyway). The sequence:
| Date | Event |
|---|---|
| May 29, 2026 | OMB proposes a rewrite of the Uniform Guidance (2 CFR 200) that implements Executive Order 14332, including political-appointee review of funding decisions |
| June 24, 2026 | NSF publishes the draft GFA for 60 days of comment (91 FR 38030, docket NSF-2026-OTR-0001) |
| July 13, 2026 | Comments close on the OMB proposal |
| August 24, 2026 | Comments close on the NSF draft |
| September 2, 2026 | H.R. 6500, the stopgap appropriations law, is signed. It bars OMB from finalizing or implementing its rule through December 11 |
| October 28, 2026 | NSF publishes its second PRA notice and sends the revised GFA to OMB |
The June draft did more than rename and reorganize the PAPPG. Its own summary of changes said Guide 12 "disallows publication costs consistent with proposed revisions to 2 CFR 200." Its termination section aligned NSF with OMB's proposed 2 CFR 200.220, which would let the agency end awards that no longer serve its priorities. Another provision let NSF suspend or terminate "to protect the interests of the government" without first issuing a notice of non-compliance or offering a corrective action plan. Guide 1 removed references to fixed-price awards. Comment letters from AIMBE, AIBS, APS, FASEB, the American Council on Education and others also flagged a preference, all else equal, for institutions with lower indirect-cost rates.
So the stopgap bill froze OMB's rule, but it did not freeze NSF's version of the same provisions, which was moving on its own schedule under separate authority. The October revision removes the four provisions named above from that version.
What actually changed, and what we can't confirm
Three things are clear from the notice and AIP's reporting.
1. The four OMB-derived provisions are out of the proposed text. For grantees, the most concrete effects are on publication costs and fixed-price awards. NSF's public access policy requires immediate open access. A rule that mandated open access while banning article processing charges as a direct cost would have pushed those fees onto departmental budgets or principal investigators. Fixed-price awards are the simplest instrument for fellowships, travel grants, conferences and small organizations without sophisticated cost accounting. Removing them would have hit the smallest awardees hardest. AIMBE made exactly this point.
2. Merit review stays the official decision path. The revised GFA reaffirms the established process. That matters because the OMB proposal's most debated feature was pre-issuance review by political appointees.
3. The changes apply prospectively. They would cover new awards and new funding increments on existing awards. NSF says it "will allow sufficient time" between publishing the final GFA and implementing it. Earlier NSF materials targeted issuance for "early to mid FY2027," which means roughly January through April 2027.
Here is what we could not confirm from public documents as of November 4:
- Whether the lower-indirect-rate preference survived. AIP's summary lists indirect-cost recovery "subject to the availability of funds" among the removed items. It does not mention the separate preference for lower-rate institutions. Read the revised Guide text before assuming it is gone.
- Whether the "interests of the government" termination language changed. Removing termination for convenience is not the same as removing every expanded termination ground. Check Guide 25 against the June draft.
- Whether the clause making 2 CFR 200 take precedence over the GFA is still there. The June draft said the Uniform Guidance controls where the two conflict. If that clause remains, any provision OMB eventually finalizes will reach NSF grantees through 2 CFR 200 regardless of what the GFA says.
Why the retreat is conditional
NSF's wording is precise. It is not saying it rejects fixed-price elimination, publication-cost disallowance, termination for convenience or funds-available indirect costs on the merits. It is saying it won't adopt them before OMB finalizes them, and that it will address Uniform Guidance changes "after final government-wide requirements are issued."
In practice, that means NSF's rules on these four points now depend on what happens to OMB's rule, and OMB's rule depends on Congress. The H.R. 6500 restriction expires December 11. AIP reports that Congress returns from recess on November 9 with most federal funding expiring the same week in December. Lawmakers have to pass full-year appropriations, another continuing resolution or some combination, and each vehicle could extend the OMB restriction, make it permanent or let it lapse.
Yesterday's midterm elections change the politics of that decision but not the calendar. The current Congress writes the December bill. A new House majority would not take its seats until January, after the deadline.
There are three plausible outcomes:
| Scenario | What happens to OMB's rule | What happens to NSF's four provisions |
|---|---|---|
| The December bill extends the block (another CR or omnibus rider) | Stays frozen | Stay out of the GFA until the next deadline |
| The block lapses December 11 | OMB can finalize, possibly within weeks given that comments closed in July | NSF has said it will conform after finalization. They could come back through the precedence clause or a later GFA revision |
| Congress blocks the rule permanently, or a future administration withdraws it | Dead | NSF's revised text becomes the durable baseline |
Under the second scenario, the October revision only delays the four provisions. Institutions should budget and write proposals with that possibility in mind.
The last comment window
The October 28 notice opens a 30-day comment period on the information collection, submitted through reginfo.gov. Search for OMB control number 3145-0058 under "Currently under 30-day Review — Open for Public Comments." Thirty days from publication is November 27, 2026. Some trackers list November 28. Treat the 27th as the deadline, which falls the day after Thanksgiving.
NSF's earlier materials described this as the one additional 30-day comment period before finalization. After OMB clears the collection, NSF can finalize the GFA without another notice.
The notice invites comments on burden, utility and clarity, plus "feedback on the new design and display of information." NSF asks commenters to tag each comment with its guide and section number in brackets, for example [G2.A.Term 2]. The format matters because NSF is processing thousands of submissions.
NSF's own burden estimate is useful evidence if you are writing a comment. The agency says each proposal takes 120 to 360 hours to prepare and that it receives about 43,000 proposals a year for about 8,300 awards. That totals 5.16 million to 15.48 million burden hours annually across roughly 3,000 institutions. A comment that ties a GFA provision to measurable added hours for your institution, such as an unclear termination clause that requires legal review or a cost rule that requires rebudgeting, is the kind PRA review is designed to weigh.
What research offices should do now
1. Get the revised text and run a redline against the June draft. The October notice says a summary of significant changes is built into the document. Pay closest attention to Guide 1 (award types), Guide 12 (costs, especially publication and data-deposit costs), Guide 25 (suspension and termination), and any indirect-cost language. Don't rely on secondhand summaries, including this one, for the details.
2. File a short, specific comment by November 27. If your institution commented in August, check whether your issues were resolved and comment again on anything that wasn't. The most useful comments this round will:
- Ask NSF to state its termination standard in the GFA itself instead of cross-referencing 2 CFR 200
- Ask for publication and data-deposit costs to be stated as explicitly allowable in the body of Guide 12, not just omitted from the disallowance list
- Ask NSF to clarify how a later OMB final rule would be incorporated, including whether that would take a new GFA notice and comment period or happen automatically through the precedence clause
3. Treat December 11 as the second deadline. Watch the December funding bill text for the OMB rider. If it lapses, assume OMB can finalize quickly and that NSF will follow.
4. Keep open-access costs in proposal budgets. The PAPPG still governs current solicitations, and nothing in force today bans article processing charges. Taking them out of budgets in anticipation of a ban that has now been withdrawn would mean less money for publishing.
5. Look at NASA and other agencies too. NSF's revision applies only to NSF. AIP reports that a coalition of scientific societies and universities has asked NASA to rescind grant-policy changes that track the OMB proposal and to pay out funds on a predictable schedule. The Education Department's EDGAR rewrite was on a similar path. If your portfolio crosses agencies, map which rulebook is borrowing from OMB, and on what timeline, for each one.
What it means
The October revision shows that the comment process can still change agency rules. More than 3,000 comments, plus pressure from lawmakers of both parties, moved a science agency off four specific provisions in four months. Most agency PRA dockets get a few dozen responses.
The revision also shows what comments can't do. NSF did not commit to anything that would bind it after OMB acts. That decision is up to Congress in the next five weeks. Institutions that comment by November 27 and keep watching the December rider will be ready for whichever outcome they get.
While the rulebook is still being decided, you can keep proposals moving by searching current NSF solicitations, which still run under the PAPPG, on Granted.