Note: A Harvard faculty committee released an important proposal today, so we’re publishing this Special Edition Briefing now and we’ll move this Friday’s Weekly Briefing to Monday.
Harvard has been aware of grade inflation for years. Since 2010, the share of flat A’s at Harvard College has risen from about 34% to 66%. What Harvard has lacked is not data or (as of October) a diagnosis, but a proposal that squarely confronts the problem as a system failure rather than a failure of individual will.
That may have just changed. Earlier today, the faculty-led Subcommittee on Grading released a modestly titled document: “A proposal for updating grading policies.” On first read, it stands out as the most well-designed, internally coherent solution we’ve seen to grade inflation at any college.
The headline recommendation — capping A’s at 20% of enrollment, plus 4 additional per course — may sound dramatic. But its expected effect is a modest rollback, not a revolution: the proposal estimates the College’s share of flat A grades would drop to about 34%, roughly where they stood as recently as 2011.
The proposal's real significance lies in how it achieves this — by realigning the grading system with Harvard's mission through structural change, not individual appeals.
What You Need To Know
The two central reforms are: a cap limiting A grades to 20%+4 per course, and a shift from GPA to an Average Percentile Rank (APR) system for determining Harvard honors and prizes.
The proposal spotlights and reaffirms the existing Faculty of Arts & Sciences (FAS) definition of an A as signaling “extraordinary distinction.” Mastery alone may be conveyed with an A-.
The recommendations follow the Office of Undergraduate Education’s (OUE) October 2025 report on grading and workload and the College’s broader effort to “re-center academics” amid findings that flat-A grades now account for over 60% of all grades awarded.
The proposal was written by the faculty-led Subcommittee on Grading (“the committee”) of the Undergraduate Educational Policy Committee, which has representation across all four professorial divisions of the FAS: Humanities, Social Sciences, Science, and the School of Engineering & Applied Sciences (SEAS).
The recommendations haven’t been adopted yet. They must be approved by a vote of the FAS faculty. If approved, they would take effect in Fall 2026.
In This Special Edition:
1. Why Harvard’s Grading System Hasn’t Been Able To Fix Itself
Harvard’s current grading system fails on two fronts.
Externally, grades have lost meaning to employers, graduate schools, and fellowships.
Internally, the College can’t reliably distinguish students for honors and awards, and students lack an accurate understanding of their academic performance.
That breakdown has distorted incentives. Students feel pressure to preserve near-perfect transcripts, making them less willing to take academic risks or explore new fields. Faculty, in turn, feel pressure to inflate grades — whether to avoid negative evaluations, maintain enrollment, or both — even when the work doesn’t merit it. Extracurriculars, not academics, become the primary venue for distinction.
Grade inflation is often seen as a failure of standards or seriousness. But the committee’s proposal draws on the diagnosis on the October 2025 OUE report “Re-Centering Academics at Harvard College: Update on Grading and Workload,” which explains, as we did in our three-part Weekly Briefing series last fall, that grade inflation is a collective action problem at Harvard. The main issue isn’t that faculty don’t know what grades should mean. It’s that the current system makes it nearly impossible to uphold those meanings in practice.
Crucially, the proposal highlights an important point that has been left out of grade inflation discussions: FAS already has an official grading philosophy. The problem is that the system no longer makes that philosophy easy to apply consistently.
According to the existing FAS grading rubric, an A should signal “extraordinary distinction” — a standard that’s inherently relative. An A- should represent excellent, full mastery, which is an absolute standard every student can, in theory, achieve.
That framework is coherent in principle. In practice, sustained inflation has made it impossible for those meanings to be applied consistently, collapsing the distinction. As a result, the meaningful difference between an A and A- has collapsed.
The consequences show up most clearly in internal awards. The committee’s proposal charts the skyrocketing number of seniors each year who receive the Sophia Freund Prize for the highest GPA. In 2010, 1 student won the prize; in 2025, 55 students did.
The effect: Today’s honors and awards system appears precise but is increasingly arbitrary, relying on distinctions carried out to multiple decimal places because nothing else remains.
2. What the Faculty Committee Is Actually Proposing
Once grade inflation at Harvard is understood as a coordination problem rather than a failure of will, the contours of a workable solution become clearer. The committee’s proposal makes only two substantive moves. Everything else exists to support, balance, or constrain them.
First move: Restore the A as a signal of distinction, not baseline success.
As mentioned above, the official Harvard College grading rubric already defines an A as a mark of “extraordinary distinction,” separate from mastery. In principle, that distinction has always existed. In practice, it has disappeared.
The proposal restores that distinction by giving it a shared, quantitative interpretation: professors may award A grades to at most 20% of a course’s enrollment, plus four additional A’s.
Importantly, the cap applies only to the A grade. Other grades, including A-’s, remain uncapped, preserving the possibility that all students in a course may demonstrate mastery.
Why this matters: The proposal isn’t trying to redefine excellence; it’s trying to make the existing definition readable at scale. By anchoring “extraordinary distinction” to a common benchmark, it strengthens the informational value of grades while leaving pedagogy and assessment in individual professors’ hands.
Second move: Stop using GPA for distinctions it can no longer make.
GPA is currently used to determine honors, prizes, and eligibility for fellowships. That reliance makes sense only if GPA meaningfully reflects relative performance.
At the current levels of compression, it does not. As the committee notes, GPAs now cluster so tightly near 4.0 that remaining differences are “less reflective of genuine variation in academic performance than random noise in the grading process.”
The proposal replaces GPA with Average Percentile Rank (APR) for internal distinctions. APR measures how a student performs relative to peers in each course, then averages those rankings across courses. It does not separately account for a course’s level of difficulty beyond how difficulty is reflected in a course’s grade distribution.
Percentile rank is a good idea here precisely because it doesn’t depend on fixed cutoffs or absolute scales. It compares students directly to their peers and remains informative even when overall grade levels differ across courses or disciplines.
Why this matters: A metric built to measure relative performance is better suited to distinguishing students when absolute grades no longer separate them meaningfully.
The remaining proposal elements are structural reinforcements.
Once those two changes are in place, the rest of the proposal exists to stabilize them:
Faculty can opt out of the cap — but only by switching to SAT/UNSAT. Instructors who don’t want to follow the A-grade limit must convert their course to Satisfactory/Unsatisfactory (SAT/UNSAT), Harvard’s version of pass/fail grading. These courses are excluded from GPA, APR, and honors calculations. Opt-outs must be declared before registration so students can decide whether to enroll.
Faculty can submit numerical score-based rankings to supplement letter grades. While APR can be calculated from letter grades alone, instructors are encouraged to submit raw scores or rankings to improve accuracy (for internal use only).
Departments can reallocate A-grade slots between paired courses. While this option seems unlikely to be used, with departmental approval, instructors teaching easier and harder versions of the same subject can transfer A-slots between them, encouraging students to pursue the more rigorous option if they can.
Bottom line: The proposal’s power lies in its restraint. It doesn’t try to fix grading everywhere at once, just at the pressure points where individual choices break down: what an A means and signifies externally, and how the College measures distinction internally. By standardizing those pieces, it gives faculty and students a system they can act within, rather than work around. And importantly, it is a system built to withstand real pressures, not idealized behavior.
3. Why the Incentives Finally Line Up
Most colleges’ attempts to address grade inflation falter not because the goals are unclear, but because the systems designed to achieve them collapse under pressure. The strength of this faculty proposal lies in how deliberately it anticipates those pressures. Instead of assuming ideal behavior, it asks how real students and professors are likely to respond, and designs around that reality.
The committee is explicit about the problem it is trying to solve. As long as grade inflation is widespread, “no individual instructor has a strong incentive to change,” because unilateral restraint carries real costs. The proposal works because its components reinforce one another, making inflation harder to sustain and proper grading therefore the path of least resistance.
Faculty opt-outs come at a cost to students, encouraging their participation in the system.
Faculty may opt out of the A cap only by switching to SAT/UNSAT grading, but those courses are excluded from GPA, APR, and honors calculations. Because students know this before enrolling, many will likely avoid such courses.
The effect: Students’ incentives reinforce the policy. Faculty who want their courses to “count” will likely remain within the capped system.
Faculty are encouraged to grade with differentiation.
With no cap beyond the A, instructors could technically give everyone else an A-. But since APR ranks students relative to peers, undifferentiated grades risk lowering students’ standing.
The effect: Faculty face gentle, built-in incentives to distribute grades meaningfully. Students, in turn, may seek out courses where rankings reflect real distinctions.
Class size isn’t an easy way to game the system.
APR balances incentives across class sizes. Smaller classes get more A slots in absolute terms, but their APR rankings have higher resolution. In a 10-person class, 6 students could get A’s, but moving up or down a single rank shifts a student’s APR in the course by 10 points. In larger courses, fewer A’s are given, but a strong performance can still translate into a high APR percentile.
The effect: Because the tradeoffs largely cancel each other out, students can focus on choosing classes for other, more substantive reasons like their interest in the course material.
Students across the board are pushed toward deeper engagement, and distinction remains meaningful at the top.
With A’s capped, students must outperform peers to earn one; meeting expectations isn’t enough. And because APR continues ranking students even within the A range, top performers also have a reason to not just meet the bar of an A, but to exceed it to stand out for honors and awards.
The effect: Students are motivated to engage more seriously with their academics, whether to earn an A or to distinguish themselves among A-level peers.
4. Why This Proposal Might Actually Stick
It learns from peers’ failures and avoids their mistakes. Other elite colleges have tried to curb grade inflation and stumbled. Princeton capped A-range averages across departments, but diffuse accountability made the policy hard to enforce. Wellesley College imposed GPA caps at the course level, forcing instructors to lower some students’ grades just to raise others. Harvard avoids both traps by designing a non-zero-sum system: the A cap applies within each course, without requiring coordination across departments. It’s simpler, more enforceable, and keeps incentives aligned.
It weighed alternatives and chose the clearest path. The committee considered flashier options like adding an A+ but concluded that doing so would ratchet up pressure, not rigor (“much as Spinal Tap tries to increase the volume by adding an ‘11’”). It also considered more complex internal metrics like z-scores and delta grading, ultimately favoring APR for its clarity and accessibility.
It reviewed potential unintended distributional effects. The proposal examined whether students from under-resourced high schools would be disproportionately affected. The evidence says no: FAS Institutional Research found no significant GPA gaps by background, and admissions and hiring leaders emphasized that what matters most is a student’s academic trajectory, not a static number like GPA.
It makes a clear, mission-driven case for the A cap. The proposal addresses concerns by noting that teaching already operates within institutional boundaries like meeting times, format requirements, and grading frameworks. The A cap simply operationalizes what the FAS faculty have already agreed to in the existing rubric: that an A means “extraordinary distinction.” Rather than undermining academic values, the committee argues, the cap restores them by making meaningful grading possible again.
Bottom line: The proposal’s strength lies in how deliberately it anticipates pressure points — from faculty incentives to potential distributional effects — and designs with them in mind. And it doesn’t just defend its choices; it shows its work. By walking through alternatives and hard questions, the committee signals confidence in the path it ultimately recommends.
5. What’s Next?
The elegance of the 19-page proposal reflects the nature of the committee that produced it. It was written by faculty, for adoption by faculty, recognizing the constraints of the grading culture that faculty currently face at Harvard. But a well-designed proposal doesn’t automatically go into effect.
The FAS professors still need to vote. The recommendations now go to the FAS Faculty Council, which can choose to put them on the docket for a meeting of the full FAS faculty. There, a vote of the faculty would be taken on whether to adopt this policy. If approved, the new system would take effect in Fall 2026. The Office of Undergraduate Education is organizing two Town Halls in February for instructors and students to discuss the proposal.
The College will need a real communications plan. As the report notes, success depends on helping students, employers, grad schools, and others understand the new standards. The proposal includes policy explanations on transcripts and outreach to post-grad admissions and hiring decisionmakers, but that can’t be the end of the message.
Harvard will need a sustained effort to explain not just what’s changing, but why. Early pushback is likely. (When the OUE report came out last October, the prospect of more rigorous grading pushed a student to tell The Crimson, “The whole entire day, I was crying…I skipped classes on Monday, and I was just sobbing in bed.”) But if implementation matches the clarity and conviction shown in this committee’s report, there’s a real shot at making lasting change.
In the end, the significance of this proposal lies less in any single policy choice than in the way it treats grading as shared academic infrastructure rather than a matter of individual virtue.
For years, Harvard has attempted to rely on informal norms to do work they can no longer do — asking individual students and faculty to absorb the costs of restraint in a system that quietly rewards inflation.
Whether or not it’s adopted, this proposal shows all colleges what a serious response to grade inflation looks like: not a demand for volunteers to be stricter, but a system that makes academic rigor easier to sustain.