Harvard Capped the A. History Says the Hard Part Comes Next

On May 20, 2026, Harvard's Faculty of Arts and Sciences announced the result of a weeklong ballot: by 458 votes to 201, professors agreed to limit how many A grades an undergraduate course can award^[1][2]. From fall 2027, an instructor may give a flat A to no more than 20 percent of the undergraduates in a course, plus four more students. There is no limit on A-minuses or any lower grade^[2].

Faculty also voted 498 to 157 to stop using grade point average (GPA) for internal honors and prizes and to use each student's average percentile rank instead, a figure that will not appear on transcripts. They rejected, 292 to 364, a new satisfactory-plus grade for courses that opt out of the cap^[1][2].

From one in four to six in ten

The vote followed an October 2025 report by Amanda Claybaugh, Harvard's dean of undergraduate education. Its central numbers: A's made up 24 percent of all grades awarded in Harvard College in 2005, 40.3 percent in 2015 and 60.2 percent in 2025^[3]. Among letter grades alone, almost 85 percent were an A or A-minus^[4]. By spring 2025 the median GPA at graduation had risen from 3.56 to 3.83 in 15 years^[2], and the cutoff for summa cum laude, the highest honor, reached 3.989 in 2025^[3].

Harvard has company. A Yale committee reported in April 2026 that 10 percent of Yale College grades were an A or A-minus in 1963, against 79 percent in 2022-23^[9]. At Princeton, a December 2025 report to faculty found that two-thirds of 2024-25 course grades were in the A range^[12]. Older national data point the same way: Stuart Rojstaczer, who has compiled grade records from more than 400 institutions, found that C was the most common college grade until the Vietnam War and that by 2013 about 45 percent of all letter grades were A's^[13].

Better students do not explain it, at least nationally. Economist Jeffrey Denning and four co-authors compared two national cohorts and found that first-year college GPA rose from 2.44 to 2.65 even though entering students' math percentiles fell^[8]. Their paper, published in 2022, concludes that grade inflation explains most of the rise in US college completion rates since the 1990s^[7]. The authors add a caution: the welfare effects are ambiguous, because looser standards may also have helped meet demand for graduates^[8].

What a compressed transcript costs

When most grades are A's, a transcript stops sorting people. Harvard's subcommittee reported that eligibility for summa now turns on GPAs carried to five decimal places, and that prize committees phone around to learn who a department's real standouts are^[4].

Outsiders have noticed. Admissions deans at the law and medical schools where Harvard students most often apply were unanimous, the subcommittee reported, that inflation is a serious problem, and they lean on LSAT and MCAT scores more than they would like^[4]. Employers have drifted from grades: citing national data, the subcommittee said 42 percent now screen on GPA, down from 73 percent before the pandemic, and that among employers recruiting at Harvard only 10.5 percent of internships and 3.0 percent of jobs set a minimum^[4]. Its fairness argument: when grades carry no signal, selection shifts to informal networks, which favor well-connected students^[4].

Claybaugh's report adds a cost inside the classroom: students said they fear the A-minus and choose courses for safe grades rather than interest^[3].

“We owe our students a functioning grading system.”

What happened the last two times

Start with Princeton. In 2004 its faculty adopted a policy recommending that each department award no more than 35 percent A-range grades in coursework^[5]. A decade later a faculty committee recommended removing the numbers, and Princeton dropped them^[5][11]. Three of its findings bear on Harvard:

• Grades started falling a year before the policy took effect. The committee's reading was that the dean's conversations with departments had already raised awareness^[5].
• It found no measurable effect on medical school admission, and concluded that the best reasons for change were psychological, not tangible harm to graduates' prospects^[5].
• Students read the target as a quota and a zero-sum contest. The dean of admission called it the most discussed topic among admitted students, and students at rival universities used it to recruit against Princeton^[5].

From 2004 to 2019, Wellesley College required courses with more than 10 students to average no higher than a B-plus^[10]. A 2014 study in the Journal of Economic Perspectives by Kristin Butcher, Patrick McEwan and Akila Weerapana found that professors complied and compression at the top eased. But in the departments where the cap was binding, enrollments and majors fell, student ratings of professors dropped and racial gaps in grades widened^[6]. Weerapana, a Wellesley professor, told The Harvard Crimson of the new plan: "I honestly don't think that the 20 percent policy will work"^[10]. Henry Farber, an emeritus Princeton economist who served on the 2014 committee, told the paper the lesson he drew was that no other school followed^[10].

Why Harvard thinks this time is different

Harvard's designers argue that their rule differs in kind. It applies course by course rather than as a departmental average, caps only the flat A rather than the whole A range or a course mean, and is paired with percentile ranking to blunt the incentive to hunt for easy courses^[4][10].

It is also looser than the headline. A 10-student seminar can award six A's and a 150-student lecture 34^[2]. The subcommittee estimated that A's would settle at about 34 percent of grades, roughly the 2011 distribution, and said about 60 percent of courses already met the cap^[4].

Skeptics remain. Nearly 85 percent of respondents to a February student government survey opposed the proposal^[1]. Faculty at Cornell and Swarthmore warned the Crimson that a course-level cap could turn collaborative classes into contests and pile grades up just below the line^[10]. Duke public policy professor Nick Carnes praised the aim but asked what happens when a whole seminar does exemplary work^[14]. Enforcement is unsettled: in August, Claybaugh said automatically converting a noncompliant course's A's to A-minuses was one option being considered^[11].

One early data point echoes Princeton's finding that attention moves grades. Harvard's share of A's fell from 60.2 percent to roughly 53 percent in fall 2025, before any cap existed^[1][3][11].

The alternatives on the table

Yale's committee recommended a college-wide standard such as a 3.0 mean and, as an immediate step, printing course percentiles on transcripts^[9]. Dartmouth and Cornell tried printing course medians on transcripts; Claybaugh's report says only the most dedicated students responded by choosing harder courses^[3][4]. Princeton's 2014 committee favored standards written by each department and regular review of grade data^[5]. A decade on, two-thirds of Princeton's grades are in the A range^[12].

The WAC view

A grade is a shared currency, and Harvard is right that no single instructor can defend its value alone. But Princeton and Wellesley show that the arithmetic of a cap is the easy part. What gave way was student consent, and no peer moved with them. Harvard's cap will last only if assessments truly separate excellent work from very good work, and if other selective universities act too.

What to do with this

Sources

1. 70% of Faculty Vote to Overhaul Harvard Grading With A Cap. The Harvard Crimson, May 20, 2026.
2. Faculty decisively approve grading changes. FAS Current (Harvard Faculty of Arts and Sciences), May 20, 2026.
3. Re-Centering Academics at Harvard College: Update on Grading and Workload (Amanda Claybaugh). Harvard College Office of Undergraduate Education, October 2025.
4. A Proposal for Updating Grading Policies. Harvard College Office of Undergraduate Education, Subcommittee on Grading, February 27, 2026.
5. Report from the Ad Hoc Committee to Review Policies Regarding Assessment and Grading. Princeton University, August 5, 2014.
6. The Effects of an Anti-Grade-Inflation Policy at Wellesley College (Butcher, McEwan, Weerapana). Journal of Economic Perspectives (American Economic Association), 2014.
7. Why Have College Completion Rates Increased? (Denning, Eide, Mumford, Patterson, Warnick). American Economic Journal: Applied Economics (American Economic Association), July 2022.
8. Why Have College Completion Rates Increased? An Analysis of Rising Grades (Working Paper 28710). National Bureau of Economic Research, April 2021.
9. Report of the Yale Committee on Trust in Higher Education. Yale University, Office of the President, April 10, 2026.
10. After Mixed Results Elsewhere, Faculty at Peer Schools Caution Harvard on Rigid A-Grade Cap. The Harvard Crimson, February 26, 2026.
11. Harvard Weighs Automatically Converting A's to A-'s to Enforce Grade Cap. The Harvard Crimson, August 28, 2026.
12. Two-thirds of undergrad grades at Princeton last year were A-range, faculty report says. The Daily Princetonian, December 2, 2025.
13. National Trends in Grade Inflation, American Colleges and Universities. GradeInflation.com (Stuart Rojstaczer), March 29, 2016.
14. Harvard capped A's. Duke faculty and students are skeptical. The Duke Chronicle, May 21, 2026.

Written by the WAC editorial team. Facts are drawn from the numbered sources above and were checked against them before publication; opinions appear only under “The WAC view”. The organisations and people named are not affiliated with the World Assessment Council unless stated. Send a correction