Okun's Law is an empirical relationship first described by American economist Arthur M. Okun in a 1962 paper for the Cowles Foundation titled "Potential GNP: Its Measurement and Significance." Working at the U.S. Council of Economic Advisers under President Kennedy, Okun observed that for every roughly 1 percentage point increase in the U.S. unemployment rate, real output fell about 2 to 3 percentage points below its potential level. The original ratio he reported was close to 1:3, though later estimates for the U.S. cluster nearer 1:2.
The relationship is typically expressed in two forms:
- Gap version: the output gap (actual GDP minus potential GDP, as a share of potential) is proportional to the gap between actual and natural unemployment.
- Difference version: the year-on-year change in unemployment is a linear function of real GDP growth, with output needing to exceed some threshold growth rate (often estimated around 2–3% for the U.S.) before unemployment falls.
Okun's Law is not a law in the physical sense — it is a statistical regularity that varies across countries, labor-market institutions, and time periods. Economies with more rigid labor markets (much of continental Europe) historically show a flatter relationship, because firms hoard labor; the U.S. coefficient tends to be steeper. The relationship also weakened during and after the 2008 global financial crisis and again during the COVID-19 recession of 2020, when unemployment moved more sharply than output changes alone would predict.
For policy analysts and think-tank researchers, Okun's Law remains a quick back-of-the-envelope tool for translating growth forecasts into labor-market implications, and is frequently cited in IMF World Economic Outlook analyses, OECD employment reviews, and central bank communications when discussing slack, potential output, and the case for stimulus.
Example
In its April 2021 *World Economic Outlook*, the IMF invoked Okun's Law-style reasoning to argue that the post-pandemic U.S. recovery would close labor-market slack faster than in the eurozone, where the output–unemployment link is weaker.
Frequently asked questions
No. It is an empirical regularity, not a theoretical identity. The coefficient varies across countries and time periods and has weakened in several recent recessions.
Keep learning