What is Four-fifths rule?
Also called 80% rule
The four-fifths rule is the rule of thumb in the United States Uniform Guidelines on Employee Selection Procedures: if a group's selection rate is less than four-fifths of the highest group's rate, federal enforcement agencies will generally regard the procedure as having adverse impact. It is a trigger for scrutiny, not a legal verdict.
What the Guidelines actually say
The rule appears at 29 CFR §1607.4(D), and its wording is deliberately hedged: a rate less than four-fifths 'will generally be regarded' as evidence of adverse impact, while a greater rate 'will generally not be regarded' as such. The same paragraph goes on to say that smaller differences may still be treated as adverse impact where they are statistically significant and practically significant, and that greater differences may not be where the numbers are small.
So the rule is a screening heuristic written into a guidance document, and it has never been the standard a court applies. Courts have relied on statistical significance tests for decades, and the four-fifths ratio is best understood as the number that decides whether anyone looks harder.
The two sentences that get cut, quoted in full
Read at source on 16 September 2026, the paragraph continues: “Smaller differences in selection rate may nevertheless constitute adverse impact, where they are significant in both statistical and practical terms or where a user’s actions have discouraged applicants disproportionately on grounds of race, sex, or ethnic group.” And then: “Greater differences in selection rate may not constitute adverse impact where the differences are based on small numbers and are not statistically significant, or where special recruiting or other programs cause the pool of minority or female candidates to be atypical of the normal pool of applicants from that group.”
Almost every summary in this market reproduces the first sentence and stops before those two. They are what makes the rule a heuristic in both directions, and they are why a ratio above 0.80 is not a finding of compliance. Clearing four-fifths is the absence of one enforcement trigger, not a verdict on the procedure.
The verb matters as much as the threshold. The regulation says a ratio below four-fifths “will generally be regarded by the Federal enforcement agencies as evidence of adverse impact”. Evidence of adverse impact opens the question of job-relatedness and business necessity. It does not answer it.
Using it without over-reading it
Compute it per stage, not over the whole funnel, and report the group sizes with it. Pair it with a two-proportion test so that a ratio computed on twelve people is not read as though it were computed on twelve hundred.
And treat the rule as jurisdictionally specific. It is a United States federal enforcement convention; it is widely borrowed elsewhere as a monitoring heuristic, but it carries no legal status outside that context and should not be presented as though it does.
The same impact ratio, four sample sizes — and the legal reading flips twice
- Two groups, each with the same number of applicants. Group A is selected at 50%, group B at 30%. The impact ratio is 0.30 / 0.50 = 0.60 in every row below, so the four-fifths rule returns the same verdict every time: fail.
- 10 applicants per group (5 of 10 against 3 of 10) — Fisher's exact test p = 0.65. Not significant.
- 40 per group (20 of 40 against 12 of 40) — p = 0.11. Still not significant.
- 100 per group (50 of 100 against 30 of 100) — p = 0.0059. Significant.
- 400 per group (200 of 400 against 120 of 400) — p < 0.0001. Significant.
- Now run it the other way, on a ratio that PASSES. Group A at 50%, group B at 41%: the ratio is 0.82, above four-fifths, so the rule returns pass in every row.
- 200 per group — p = 0.088. Not significant, and the rule and the test agree.
- 500 per group — p = 0.0052. Significant, while the four-fifths rule still reads pass.
- 2,000 per group — p < 0.0001. Significant, and the rule still reads pass.
Both failures are in the regulation's own text rather than in a critique of it. §1607.4(D) says smaller differences may nevertheless constitute adverse impact where they are significant in both statistical and practical terms, and that greater differences may not where they rest on small numbers and are not statistically significant. The two sentences are quoted in full in the [evidence index](/guides/evidence/uniform-guidelines-four-fifths-rule). So the ratio is a screening heuristic whose answer is independent of the one quantity that decides whether the gap is real — which is why a vendor reporting only an impact ratio, with no applicant counts beside it, has handed you a number you cannot act on. Ask for the denominators. Figures computed with Fisher's exact test, two-sided.
Not the same as statistical significance
The four-fifths rule asks whether the gap is large. A significance test asks whether the gap is more than sampling noise. Small samples routinely fail the first and pass the second; large samples routinely do the opposite.
Sources
Read next
Related terms
More terms beginning with F
Check a selection process against the four-fifths rule
Free, no signup, computed in your browser — with the remedy, not just the verdict.
Last reviewed