Grade inflation is headline news again. Harvard’s decision to cap flat A’s has revived familiar arguments about weakened standards, distorted course choices and diminished incentives to excel. Those concerns are real, but the attention they receive can obscure a deeper social problem. When grades lose much of their power to distinguish students, inequality at the admissions gate travels farther into graduate education and employment.
College performance creates a new record through which students can distinguish themselves after admission. The diploma identifies the institution; the transcript should distinguish the student. When grades cluster near the top, that balance shifts toward institutional reputation. A solid but unexceptional graduate of a prestigious university retains the benefit of its name, while an exceptional graduate elsewhere loses one of the clearest ways to demonstrate superior achievement.
Grade inflation does not invent the college brand; it gives the brand less competition. The student graduates are credentialed and branded.
Beginning in fall 2027, Harvard College courses will limit flat A’s to 20 percent of enrollment plus four additional students; A-minuses will remain unrestricted. Faculty approved the cap by 458–201. More than 60 percent of undergraduate grades, according to Harvard data, were flat A’s in 2025, compared with about one-third in 2010.
Inflation raises average grades over time; compression concentrates them near the ceiling. A GPA may retain numerical precision while losing comparative value. A 3.8 can place a student near the top of one program and near the middle of another, depending on the field, course and instructor. A bare GPA does not supply the distribution needed to tell the difference.
A 2026 George Mason University review examined 1,801,024 final grades awarded from 2017 through 2023. A was the most common undergraduate grade, and A-minus or higher accounted for 48 percent of undergraduate grades. Average undergraduate grades ranged from 2.92 in the College of Science to 3.60 in Health and Human Services. The report noted that A has been the most frequently awarded grade nationally since at least 2009.
Harvard and Mason occupy different positions in American higher education, but their data expose the same measurement problem. Grades are assigned locally and read in a common market. Employers and graduate schools compare GPAs drawn from universities, fields, courses and instructors with different grade distributions. The same number can represent very different relative performance.
This measurement problem is magnified by unequal access to institutional prestige. A 2026 Quarterly Journal of Economics study by Raj Chetty, David Deming and John Friedman found that children from families in the top 1 percent are more than twice as likely to attend an Ivy-Plus college as middle-class students with comparable SAT or ACT scores. Two-thirds of the gap comes from higher admission rates among high-income applicants with comparable scores.
The researchers also estimate that attending an Ivy-Plus college rather than the average flagship public university nearly doubles the chance of attending an elite graduate school and almost triples the chance of working at a prestigious firm. Admission decisions largely reflect records assembled by age 17 or 18 under unequal economic and social conditions. College performance supplies later evidence that can revise the expectations attached to the institution’s name. When grade compression makes that evidence less informative, more of the sorting completed at admission carries forward into graduate selection and employment.
I teach at George Mason, where some students produce work that would distinguish them at any university. Their transcripts should reveal how unusual that performance was within the courses and programs they completed; a bare GPA cannot do that by itself.
Labor-market evidence shows why the loss of comparative information matters most as graduates enter employment. A 2024 Journal of Labor Economics study examined a nationwide Danish reform that mechanically converted previously earned grades to a new scale. The reform raised some students’ reported GPAs independently of their work or measured ability. Those increases raised earnings during the first two years after graduation, but the effect disappeared by the third year and was strongest among graduates with fewer alternative signals. The authors interpret the temporary premium as employer learning. Employers initially rely on the reported grade, then revise their judgments as evidence from the workplace accumulates.
A 2020 Economics of Education Review study of Chilean graduates found a similar early-career effect for institutional prestige. College graduates barely admitted to one of the country’s two most prestigious universities earned 13 percent more during their first year in the labor market than those barely rejected. After six or more years, the wage premium had fallen to 4 percent. The result suggests that the university name carries its greatest value when employers possess the least direct information; its influence declines as graduates show what they can do.
Neither study measures grade compression directly. Together they show why information about academic performance has its greatest value before a graduate has established a work record. When transcripts provide less comparative evidence, employers have greater reason to rely on institutional reputation. That shift can affect the first interview, the first offer, the first salary and the first opportunity to build experience.
Harvard’s faculty also approved an internal average percentile rank for Latin honors and prizes. The ranking will not appear on transcripts. Harvard will use richer comparative evidence for its own distinctions, while employers and graduate schools will receive the conventional record.
Universities should extend that context to the official transcript. Each grade could be accompanied by the course median, grade distribution and enrollment. Programs could place students within broad performance bands and identify demanding capstones, substantial research, honors awarded on comparative criteria and work assessed against common standards. For small classes, distributions could be aggregated across offerings or reported in broad bands to protect privacy. Such reporting would preserve legitimate disciplinary variation while making it visible.
Employers and graduate schools would then have a stronger basis for comparing achievement across institutions and fields. Performance after admission would carry greater influence over graduate study and early career opportunities. Universities should preserve the distinctions that make accomplishment visible. The diploma identifies the institution; the transcript should reveal what the student achieved.
Read Full Article »