On 20 August 2026 the Department of Education published a Notice of Proposed Rulemaking, “Accreditation, Innovation, and Modernization,” rewriting the federal regulation — 34 CFR Part 602 — that determines which organizations get to decide whether a college is real enough to receive federal student aid (Established; 91 Fed. Reg. 53940, document 2026-17001, 34 CFR Parts 600, 602, and 668). The rule does three distinct things: it lowers the barrier for new accreditors to enter the market, it adds mandatory outcomes metrics like completion and post-graduation earnings, and it requires every recognized accreditor to evaluate colleges on “intellectual diversity” and academic freedom, including surveyed perceptions of viewpoint diversity among students and faculty. Those three changes are being sold as one package. They are not equally contested, and they are not equally exposed to the same risk.
How the gate currently works
Accreditation is the federal government’s proxy for quality control in higher education. A school cannot enroll a single student on a federal Pell grant or loan without being accredited by an agency the Education Department itself recognizes — which makes accreditors gatekeepers of roughly $100 billion or more in annual federal student aid, a figure both the Department’s own announcement and AEI’s independent analysis of Title IV flows put in that range (Established). Since 2020, the regulatory distinction between “regional” and “national” accreditors has been eliminated; both compete, on paper, under one federal standard, though a small number of legacy agencies still accredit most degree-granting institutions in practice (Established, per the Department’s own overview of the system).
What the NPRM actually proposes
Reading the rule itself and the law-firm summaries closest to its text, three structural moves stand out. First, entry: agencies could seek federal recognition before accrediting a single institution, colleges could hold accreditation from more than one agency at once, and the Department’s prior approval would no longer be required for an institution to switch accreditors — changes aimed at breaking accreditors’ territorial habits (Established, Nixon Peabody and Faegre Drinker regulatory alerts; Tier 2). Second, outcomes: accreditors would have to evaluate program-level completion, retention, licensure pass rates, post-completion employment, and “educational and economic returns” against cost of attendance, using standardized rather than self-reported data (Established, same sourcing). Third, viewpoint: accreditors would have to require institutions to adopt policies protecting “intellectual diversity and the free exchange of ideas” and to measure student and faculty perceptions of the range of perspectives on campus — a standard the Department frames as protecting speech evenhandedly across ideology but leaves substantially undefined in the binding text, offering only a nonbinding preamble definition (Established that the provision exists as described; Assessed with moderate confidence that its practical bite will depend on how individual accreditors operationalize “perception” measurement, which the rule does not specify). A fourth, less-noticed provision strips pre-1991 accreditors of a waiver letting them argue recognition was a protected property interest — a real due-process change (Established per Faegre Drinker; Assessed with moderate confidence, per Century Foundation fellow Robert Shireman, that this invites prolonged litigation over any future deregistration). Comments are due 21 September 2026; the Department targets a final rule for 1 November 2026, effective 1 July 2027 (Established).
The reform case, steel-manned
The claim that accreditation functions like a cartel is not confined to one political camp. Legal scholars Roger Meiners and Andrew Morriss have applied antitrust frameworks directly to accreditation’s territorial, entry-restricting behavior; libertarian-leaning groups from the Mises Institute to the Heritage Foundation have made versions of the same argument for years. The accountability half of the critique is broader still: the centrist Third Way has documented that fewer than 3 percent of accreditor enforcement actions penalize an institution for poor outcomes, that under 1 percent of institutions ever lose accreditation, and that more than a third of accredited colleges fail to graduate half their students while drawing over $20 billion a year in federal aid. AEI’s independent count found roughly 1 percent of schools lost accreditation against an 8 percent sanction rate over a five-year sample — a wide gap between flagging a problem and doing anything about it. That is a documented dysfunction, and it is fair to say plainly: on the accountability metric alone, the current system underperforms its own stated mission (Assessed with high confidence, based on convergent findings from ideologically opposed sources using different methods).
The concern case, steel-manned
Robert Shireman — a Century Foundation fellow who has advised NACIQI, the Department’s own accreditation oversight committee — called an earlier draft of the rule “a cluster bomb being dropped on American higher education” in an April 9, 2026 Substack post. Antoinette Flores of New America and Emily Rounds of Third Way — an organization that itself wants accreditation held to tighter outcome standards — made similar objections to Inside Higher Ed on April 13, 2026, when an earlier 151-page draft was under discussion: Flores called it an “extreme overcorrection,” and Rounds warned that lowering entry barriers while leaving outcome thresholds to individual accreditors’ discretion could trigger “a race to the bottom” in which struggling institutions shop for the most permissive accreditor. The final NPRM published in August tracks the substance of that April/May draft closely on the provisions these three critiqued, per the Nixon Peabody and Faegre Drinker regulatory alerts, though none of the three is confirmed to have commented specifically on the final 354-page text. AAUP has objected that any federal definition of academic freedom must track its own 1940 statement, not a Department-drafted alternative. And Inside Higher Ed reported the Department seated a political appointee, Jeffrey Andrade, in the negotiating seat historically reserved for a career civil servant — a procedural detail that does not by itself prove bad faith but is the kind of structural signal that shapes how a technical rulemaking is read (Established that these statements were made, as reported; Assessed with moderate confidence on what the appointee-seat change portends, since one seating decision is a thin basis for a durable prediction).
Assessment
The entry-liberalization and outcomes provisions answer a real, bipartisan-documented accountability failure; on that ground the NPRM reads as a genuine, if incomplete, attempt to fix dysfunctional gatekeeping architecture (Assessed with moderate-to-high confidence). The viewpoint-diversity mandate is a different animal: it was not part of the accountability critique made by Third Way or AEI, it relies on subjective perception surveys the rule does not standardize, and it arrives bundled with weaker due-process protections for accreditors the Department wants to remove — a combination that hands a future Secretary more discretionary leverage over which accreditors survive, on grounds that are not primarily about completion or debt outcomes. Both readings hold simultaneously: this is a rule that addresses a documented problem and creates a new lever for a different kind of institutional influence over who gatekeeps higher education (Assessed with moderate confidence; the eventual balance depends on choices — how “viewpoint diversity” is operationalized, how due-process changes play out — that cannot be resolved from the proposed text alone. We do not know how the Department will handle comments on the viewpoint-diversity metric specifically, since the preamble language is explicitly nonbinding and could tighten or loosen in the final rule).
What to watch
The comment window closes 21 September 2026, through regulations.gov under this docket; expect substantive filings from CHEA and the major legacy accreditors defending due-process protections, from AAUP and higher-ed faculty groups on the academic-freedom definition, and from Third Way, AEI, and New America pressing — from different directions — on whether the outcomes standards are specific enough to bite. The Department’s stated target for a final rule is 1 November 2026, with a 1 July 2027 effective date; whether that timeline holds, and whether the viewpoint-diversity and due-process provisions survive comment and likely litigation in something close to their current form, is what will determine which of the two readings above turns out to be the story.