"Grade inflation" and "grade deflation" get thrown around as vague reputations -- Harvard is easy, Princeton is hard -- more often than they get checked against real numbers. This is that check. It's a different question from what "grade covering" is: covering hides a real grade from your transcript for one semester, usually your first; inflation and deflation are about what grade actually gets assigned in the first place, and whether a school is doing anything to change that trend.

The national trend, in real numbers

The most-cited long-running research on this comes from Stuart Rojstaczer (Duke) and Christopher Healy (Furman), who compiled GPA data spanning nearly 70 years across more than 400 US colleges and universities. Their headline finding: the national average college GPA rose from about 2.52 in the 1950s to roughly 3.11-3.15 in recent years -- nearly a third of a letter grade, system-wide, over seven decades. The trend isn't uniform: private schools average closer to 3.3, public schools closer to 3.0-3.1.

A real, single-school example of what that climb actually looks like: at Clemson University, about 24% of grades were A's in 1982; by 2001, that had risen to 38%. Research controlling for rising SAT scores and course selection finds those factors explain roughly half the national increase -- the rest is genuinely a relaxation of grading standards, not just better-prepared students. See Clemson's page for its current admissions profile.

The current flagship case: Harvard

Harvard is the sharpest, most current version of this story, and it's still unfolding. Two decades ago, about 25% of Harvard College grades were A's. By 2014-15 that had risen to 44%; the pandemic pushed it to 66% in 2020-21; and by 2024-25, A's still made up roughly 60-66% of all grades, with the median cumulative GPA at graduation climbing from 3.56 to 3.83 in just 15 years. The rise isn't even across departments: in 2024-25, 78% of Arts & Humanities grades were A's, against 57% in the Sciences and 56% in SEAS (engineering and applied sciences) -- a gap wide enough that a single university-wide fix affects departments very differently.

In October 2025, a faculty committee report concluded Harvard's grading was "too compressed and too inflated" and was "failing to perform the key functions of grading." A voluntary push that same fall cut the share of A's by nearly seven percentage points -- not enough, in the faculty's own view, to count as fixed. In May 2026, the Faculty of Arts and Sciences voted 458 to 201 (about 70%) to adopt an actual cap: no more than 20% of students in a course, plus four more, can receive a flat A (A-minus is uncapped). It takes effect fall 2027. This is a real, current, forward-looking deflation policy -- not a historical footnote -- and one of the very few examples of a major research university choosing to reverse its own inflation rather than just study it.

What "grade deflation" has actually looked like -- and why it rarely survives

Princeton is the name most people reach for when they say "grade deflation," but the policy that earned it that reputation is gone. In 2004, Princeton's faculty adopted a guideline capping A-range grades at 35% of grades per department. It worked, in the narrow sense: A's fell from 47% (2001-04) to 41.8% (2010-13). It also became genuinely unpopular -- students reported feeling like they were competing for a scarce resource of A's rather than being graded on their own work -- and in 2014 the faculty voted to abandon it. A's at Princeton have "skyrocketed" since; by recent reporting, roughly two-thirds of Princeton grades now fall in the A range, in the same ballpark as Harvard's current numbers.

Wellesley ran a separate policy, adopted in that same 2004 wave of national concern about grade inflation: a rule capping the mean grade at a B+ (3.33) in every 100- and 200-level course with 10 or more students, with instructors who exceeded it expected to explain why to the faculty curriculum committee. It ran from April 2004 to May 2019, when it was rescinded in favor of an annual department-level conversation about grading practices instead of a hard number. That's a different Wellesley policy from its still-active "shadow grading" for first-year students -- which hides a real grade from the transcript for one semester rather than capping what grade gets assigned -- covered in our full breakdown of grade covering.

The pattern across both real examples: an explicit numeric cap on A's is the one grading intervention schools have actually tried and then abandoned once students pushed back, not something that quietly stuck around. Harvard's 2027 policy is the first major test of whether that changes -- and whether a cap survives longer this time depends on whether it avoids the same "I'm competing against my classmates for a fixed number of A's" reaction that sank Princeton's version.

What this means for you

  • "Grade inflation" and "grade deflation" are both real, measurable, and specific to a school and era -- not a permanent reputation. Princeton's and Wellesley's deflation policies are both gone; check a school's current grading data, not its decades-old reputation.
  • Don't confuse a grading cap with grade covering. A cap changes what grade you're assigned; covering hides an assigned grade from your transcript. MIT, Swarthmore, Caltech, and Michigan's covering policies are a genuinely different mechanism from Harvard's, Princeton's, or Wellesley's caps.
  • If Harvard is on your list, the fall 2027 policy is worth knowing about specifically if you'd be enrolling under it -- a flat-A cap of "20% plus four" per course, with A-minus uncapped, is a real structural change to how a Harvard transcript will read going forward, not a minor adjustment.
  • A department-level split (like Harvard's Arts & Humanities vs. SEAS gap) means "the school's average GPA" can be a misleading number for your specific intended major -- the same institution can be meaningfully easier or harder to earn an A in depending on the department.
  • A capped or genuinely rigorous grading school isn't a red flag for your future GPA -- it's a real tradeoff worth knowing about going in, the same way a shorter or a harder-graded semester is worth planning for rather than being surprised by.

Sources