Okun’s law is a rough empirical regularity linking changes in unemployment to changes in real output (usually real GDP). In plain language: when growth is stronger than usual, unemployment often falls; when growth is weaker than usual, unemployment often rises. It is called a “law” in the textbook sense of a recurring pattern, not a physical law. The slope and intercept shift across countries and decades. This explainer shows the intuition, a simple arithmetic sketch, common pitfalls, and how UK and US readers can use the idea when reading releases. Educational only; not investment advice.

The core intuition

Firms produce more when demand is firm. Producing more usually means more hours and eventually more jobs. Producing less means fewer hours and, with a lag, fewer jobs or slower hiring. Because the labour force and productivity also move, the mapping from GDP growth to unemployment changes is not one for one. Okun’s law summarises that mapping as a rule of thumb for teaching and rough scenario thinking.

A common classroom form says that for each percentage point by which real GDP growth exceeds its trend (or potential) rate, the unemployment rate falls by a fraction of a percentage point. Older US illustrations sometimes used a factor near one half. Modern estimates vary. Always treat any single coefficient as sample dependent.

A worked arithmetic sketch

Example (illustrative)

Suppose a country’s trend real GDP growth is about 2 percent a year, and a simple Okun rule of thumb says each 1 percentage point of GDP growth above trend lowers unemployment by about 0.4 percentage points (a made up teaching coefficient). If real GDP grows 4 percent in a year, that is 2 points above trend, so the rule of thumb would point to unemployment falling by roughly 0.8 percentage points, other things equal. If GDP grows 0 percent, that is 2 points below trend, and the same rule would point to unemployment rising by about 0.8 points. Real economies add productivity shifts, participation changes, and hours adjustment, so live BLS, BEA, ONS, and OBR figures will not match this cartoon.

Okun’s law is a bridge between the GDP pillar and the unemployment pillar. It is not a forecasting machine.

Why the relationship is loose

  • Hours before heads. Firms often cut or raise hours before they fire or hire. Unemployment can lag output.
  • Productivity. If output rises mainly because each worker produces more, unemployment may not fall much.
  • Participation. People entering or leaving the labour force change unemployment even when employment is stable.
  • Composition. Sector mixes differ. A goods heavy rebound can show different Okun behaviour than a services heavy one.
  • Measurement. GDP revisions and survey unemployment revisions can rewrite last year’s “Okun residual.”

UK and US framing

US readers usually pair BEA real GDP (quarterly) with BLS unemployment. FRED makes both easy to chart side by side. UK readers pair ONS real GDP with Labour Force Survey unemployment, remembering that UK GDP is often discussed in monthly as well as quarterly form and that survey methods evolve. Cross country Okun slopes are not interchangeable: labour market institutions, short time working schemes, and data definitions differ.

Related explainers: recession definitions, natural rate of unemployment, Phillips curve. Glossary: GDP, natural rate.

Common myths

  • Myth: Okun’s law gives a precise unemployment forecast from one GDP print. It gives a directional sketch with wide noise.
  • Myth: the coefficient is universal. US historical samples differ from UK samples, and both differ from emerging economies.
  • Myth: if GDP rises, unemployment must fall immediately. Lags, hours, and participation can break that month by month.
  • Myth: Okun’s law replaces labour market detail. Vacancies, pay growth, and underemployment still matter for inflation and welfare.

How policymakers gesture at the idea

Forecast staffs embed some output to unemployment mapping inside larger models. Stress tests and scenario boxes often ask: if growth undershoots, how much might unemployment rise? That is Okun style reasoning even when the label is not used. Monetary policy cares because labour market slack feeds the inflation outlook with a lag. Fiscal policy cares because unemployment benefits and tax receipts move with the cycle. See monetary and fiscal policy and fiscal multipliers.

Reading data without chart traps

Compare growth rates with clear units (quarter on quarter annualised versus year over year). Align unemployment as a level or a change and say which. Prefer primary series from BEA, BLS, ONS, or FRED mirrors over anonymous screenshots. When someone claims “Okun says unemployment will hit X,” ask which coefficient, which trend growth assumption, and which sample. For growth measurement differences across agencies, see BEA versus ONS GDP.

One careful habit: use Okun’s law to sanity check stories that claim huge job gains with almost no growth, or deep downturns with no labour market cost. Large residuals invite a second look at productivity, hours, or data revisions.

Sources