While OMB's Grant Rewrite Sits Frozen, the Education Department Quietly Filed Its Own. EDGAR's Overhaul Closes Comments September 23 — and §75.228 Would Reward You for Underbilling Your Own Indirect Rate.

There is a version of this story where nothing happened. The Education Department published a notice of proposed rulemaking titled "Education Department General Administrative Regulations," a phrase engineered to repel readers, proposing what the preamble describes as technical alignment and streamlining. Most grants offices will file it unread.

That would be a mistake, and the timing is the tell.

OMB's rewrite of 2 CFR Part 200 — the government-wide Uniform Guidance overhaul that drew nearly half a million comments and was scheduled to bind on October 1 — is currently stalled by Section 157 of the Senate continuing resolution, which blocks implementation through December 11. The government-wide vehicle is parked.

So the Education Department drove around it. On August 24, 2026, ED published docket ED-2026-OPEPD-2542, RIN 1875-AA14, proposing amendments to 34 CFR parts 75, 76, 77, and 79 and to 2 CFR parts 3474 and 3485. Comments are due September 23, 2026 — a 30-day window for a rule that touches nearly every structural feature of how ED awards, continues, and terminates discretionary grants.

EDGAR is ED's own rulebook. It does not need OMB's rulemaking to move, and it is not covered by the CR provision holding OMB's back. An applicant to an ED discretionary competition lives under EDGAR far more intimately than under Part 200, because EDGAR is what governs the competition itself: the selection criteria, the priorities, the peer review, the continuation decision, the termination.

Here is what is actually in it.

§75.228: competitive preference for charging yourself less

This is the provision to read first, because it is new, it is short, and it restructures the economics of every ED competition.

The proposed §75.228 would authorize the Secretary to give a competitive preference to applicants that voluntarily propose to charge a lower indirect cost rate than their negotiated rate.

Understand what that does. An organization with a federally negotiated indirect cost rate agreement — a rate established through an audited, methodologically governed process with a cognizant agency — would be able to buy scoring points by declining to charge what that agreement entitles it to. The rate stops being a cost-recovery instrument and becomes a bidding variable.

The direct effects are predictable. In any competition where §75.228 is invoked, applicants will model the tradeoff: how many points is the preference worth, and how much indirect recovery must be surrendered to earn them. Sophisticated applicants with diversified revenue will discount aggressively, because they can absorb the shortfall from other sources. Applicants whose federal grants are their revenue — small nonprofits, community-based organizations, minority-serving institutions, rural districts — cannot discount without cutting the administrative capacity that makes them compliant grantees in the first place. The preference is regressive by construction, and it disadvantages precisely the applicants who have the least slack.

The policy context matters here too. The attempt to impose a flat 15 percent indirect rate cap on NIH research awards was enjoined in litigation, and parallel efforts at other agencies have met similar resistance, in substantial part because a categorical cap collides with the negotiated-rate framework in Part 200 and with appropriations riders. A competitive preference sidesteps that entirely. Nothing is capped. Nothing is imposed. Applicants simply volunteer — in a competition where volunteering is scored.

If your organization comments on one thing, comment on this. And comment with arithmetic: what your negotiated rate recovers in absolute dollars on a typical ED award, what functions that money pays for, and what happens to your single audit posture when those functions are unfunded.

§75.253: continuation awards become discretionary in a new way

Multi-year ED grants are made as a single award with annual continuations. The continuation decision has always had structure around it. The proposal loosens that structure and adds a procedure in its place.

The proposed changes remove the constraints in §75.253(c), clarify the Secretary's discretion over whether to continue, and confirm authority to reduce a continuation award without triggering the General Education Provisions Act procedures that attach to formal enforcement actions. Alongside that, the proposal adds verification of the quality of data submitted as a continuation input, and states that if the Department decides not to make a continuation award, the grantee may object under 2 CFR 200.341 through a request for reconsideration.

Read the two halves together. The substantive standard gets looser and the procedural remedy gets more explicit. That is a meaningful trade, and how good a trade depends entirely on what a "request for reconsideration" actually consists of — a question the proposal does not answer with a timeline, a decision-maker, or a standard of review. Reconsideration by the same office that made the decision, on the same record, with no deadline, is not a remedy. Reconsideration by an independent reviewer within a fixed period is.

That gap is the single most comment-worthy procedural defect in the rule, and it is fixable with two sentences. Ask for them.

Related and easy to miss: proposed §75.230 adds that approval of an application "does not obligate the Federal Government to provide additional funding for the award in the future," §75.251 adds authority to adjust awards based on unspent balances in multi-year projects, and a new §75.252 codifies frontloading procedures for multi-year funds. The unspent-balance provision deserves attention from anyone whose project has a slow first year by design — planning years, construction, hiring cycles. Underspending in year one has always been a conversation. It would now be a documented basis for adjustment.

§§75.100, 75.104, 75.105 and 79.3, 79.6, 79.8: the Federal Register exit

The proposal eliminates Federal Register publication requirements for application notices and annual priorities, consolidating them on Grants.gov. Parallel changes in Part 79 remove Federal Register publication of intergovernmental review program lists and comment deadlines. §75.222 removes the Federal Register reference for unsolicited applications and directs submissions to a Department email address. §75.224 removes the Federal Register notice requirement for multiple-tier review processes.

ED frames this as burden and cost reduction, and as a matter of printing bills it is. But Federal Register publication is not just a distribution channel. It is the legal mechanism of public notice — the citable, dated, permanently archived record that a priority was announced, on a particular day, in particular words. It is what a comment period attaches to, what a court looks at, and what a researcher reconstructing a program's history five years later can actually find.

Grants.gov is a posting board. Postings change. Attachments get replaced. Nothing versions.

For applicants the practical consequence is immediate: the monitoring habits that worked stop working. If your organization tracks ED competitions by watching the Federal Register — through a service, a saved search, or a person whose job includes it — that pipeline goes dark for application notices. Whatever replaces it has to be built against Grants.gov and simpler.grants.gov, and it has to be built before the final rule takes effect, not after you miss a competition.

§77.1: the What Works Clearinghouse comes out of the definitions

Proposed changes to §77.1 revise the evidence definitions that structure ED's evidence tiers — "experimental study," "moderate evidence," "strong evidence" — add a new "evidence framework" definition, and remove the requirement that studies meet What Works Clearinghouse standards, permitting alternative methods of verifying rigor.

For programs where evidence tiers determine eligibility or scoring — Education Innovation and Research most visibly, but also SEED, GEAR UP, and a widening set of competitions since ESSA — this is not a technical edit. WWC standards are demanding, slow, and frequently criticized as a bottleneck that privileges expensive randomized trials and the organizations that can afford them. There is a real case that a single clearinghouse should not be the sole gatekeeper of what counts as evidence.

There is an equally real case on the other side. WWC's value was never that it was fast. It was that it was one standard, applied by a third party, that an applicant could not grade itself against. "Alternative methods of verifying rigor" without a specified alternative is not a broader evidence base; it is an undefined one, and undefined standards get resolved by whoever is reviewing the application that day.

If you fund or conduct education research, this is your section. The comment that helps is not "keep WWC." It is: name the alternative standards, or name who decides whether an alternative is adequate, and when.

§75.500(f) and §76.500: merit practices as a grant condition

The proposal adds a new paragraph (f) to §75.500, with a parallel provision at §76.500, requiring that grantees ensure hiring, promotion, and compensation practices are merit-based and conducted "without regard to race, color, religion, sex, national origin, or proxies thereof," with limited exceptions for religious organizations and remedial action.

The operative phrase is "or proxies thereof," and it is doing an enormous amount of undefined work. A proxy is not a protected characteristic; it is a facially neutral criterion alleged to stand in for one. Whether a given recruitment practice, geographic preference, first-generation status criterion, or partnership with a particular institution constitutes a proxy is a contested legal question that this rule would import wholesale into the grant terms — where the enforcement mechanism is not a discrimination suit but an award condition, and where a false certification carries False Claims Act exposure.

Separately, §76.700 would add Executive Orders to the list of authorities grantees must comply with, alongside statutes and regulations. Executive orders are not published with implementing regulations, are not subject to notice and comment, and change without a transition period. Making compliance with them a condition of an award means the terms of your grant can change between the time you sign and the time you draw down.

The litigation-proofing tell

Two new sections — §75.4 and §76.3 — add severability provisions: if any part of the rule is held invalid, the rest survives.

Agencies add severability clauses when they expect to be sued and want a partial loss to stay partial. Nobody writes one for a technical alignment package. It is the most honest sentence in the document about what the Department thinks it has proposed. Read alongside §75.901, which clarifies termination-for-convenience authority for discretionary awards, and the architecture of the rule is not subtle: broaden discretion, narrow the paper trail, and insulate the whole thing from a single adverse ruling.

What to do before September 23

Model §75.228 against a real budget. Take your last ED application. Compute indirect recovery at your negotiated rate. Then compute it at 10 and 15 percent, and identify by name the positions and functions that stop being funded. That paragraph is your comment.

Rebuild your opportunity monitoring on Grants.gov now. Do not wait for the final rule. If application notices leave the Federal Register, any workflow anchored there fails silently — you will not get an error, you will get nothing, and you will find out when a competition closes.

Ask for continuation reconsideration mechanics. A decision-maker, a deadline, a standard. Three specifics, two sentences, entirely winnable.

Ask what replaces WWC. Name the alternative or name who decides.

Get the "proxies thereof" definition on the record. Every organization certifying compliance with §75.500(f) is certifying to a standard nobody has defined. Ask for the definition in the comment period rather than discovering it in an enforcement action.

File by September 23, 2026, at regulations.gov under ED-2026-OPEPD-2542. Thirty days is short for a rule of this reach, and the shortness is itself worth noting in a comment — a longer window is a routine, frequently granted request, and nobody gets it who does not ask.

The Department intends to finalize in late 2026. Whatever OMB's Part 200 rewrite eventually does or does not do in December, ED's grantees will be operating under a rewritten EDGAR well before that fight resolves.

For the government-wide picture, see our analyses of OMB's 2 CFR Part 200 overhaul and its October 1 effective date and the Senate CR provision that froze it until December 11.