What Is Okun's Law? Unemployment and GDP Explained

Search Fintrev

Search our guides, reviews and glossary.

Okun’s law is an empirically observed relationship between unemployment and economic output: when the unemployment rate rises above its natural rate, real gross domestic product (GDP) falls below its potential level, and the reverse holds when unemployment falls. In Arthur Okun’s original estimate, each 1 percentage point of unemployment above the natural rate corresponded to output about 3 percentage points below potential. Later estimates for the US have tended to be lower, which is why the rule is commonly quoted as a 2 to 3 point output loss per point of excess unemployment.

Despite the name, Okun’s law is a statistical regularity, not a law of economics in the strict sense. Economists often call it Okun’s rule of thumb because it comes from observed data rather than a formal theory, and its size changes across countries and time periods.

What Okun’s Law Says

The intuition is simple: output depends on how much labor goes into production, so fewer people working means less output. Okun put a rough number on that link using US data.

Okun’s law is used to estimate how much GDP may be lost when unemployment is above its natural rate. Two reference points matter here:

• Potential GDP is the output an economy can produce when its resources (the factors of production: land, labor, capital and entrepreneurship) are fully employed at a sustainable pace. It is an estimate, not a measured figure.
• The natural rate of unemployment is the unemployment rate consistent with that sustainable level of output. It is never zero, because people are always changing jobs, entering the workforce or moving between industries.

The gap between actual and potential output is the output gap. Okun’s law says the output gap and the unemployment gap (actual unemployment minus the natural rate) move in opposite directions, with output moving by a larger amount.

The Okun’s Law Equation

There is no single Okun’s law equation. Economists use several versions, and the coefficient takes a different value in each, so a coefficient estimated for one version cannot be plugged into another.

Gap version

The gap version relates levels:

(Y − Y*) / Y* = −c × (u − u*)

where Y is actual real GDP, Y* is potential GDP, u is the actual unemployment rate, u* is the natural rate, and c is Okun’s coefficient. Okun’s original estimate put c near 3. Many later US estimates put it closer to 2.

Rearranged, this version answers a common practical question: given the natural unemployment rate and the output gap, what unemployment rate would you expect?

u ≈ u* − (output gap ÷ c)

Difference version

The difference version relates changes from one period to the next:

Δu = −k × (g − g*)

where Δu is the change in the unemployment rate, g is actual real GDP growth, g* is the normal or potential growth rate, and k is the coefficient. This version says unemployment stays roughly flat when the economy grows at its normal pace, rises when growth falls short, and falls when growth runs faster. Because k measures how much unemployment moves per point of growth, it is a fraction (well below 1) rather than a number like 2 or 3.

A simple regression form used in Federal Reserve research is:

U = a + b × G

Here U is the change in the unemployment rate from one quarter to the next, G is real GDP growth in that quarter, and b is Okun’s coefficient, the slope of the relationship. The constant a, combined with b, implies the growth rate at which unemployment stays unchanged.

Worked Examples

The figures below are hypothetical and chosen only to show the mechanics.

Example 1: estimating the output loss. Assume the natural rate of unemployment is 4% and the actual rate is 6%, so unemployment is 2 points above its natural level. Using the gap version with an assumed coefficient of 2:

Output gap = −2 × (6 − 4) = −4%

Output is about 4% below potential. If potential GDP were $20 trillion, that gap would amount to roughly $800 billion of goods and services not produced in a year. This forgone output is the economy-wide opportunity cost of the idle labor.

Example 2: estimating unemployment from the output gap. Assume the natural rate is 4.5%, the output gap is −3%, and the coefficient is 2. Then:

u ≈ 4.5 − (−3 ÷ 2) = 4.5 + 1.5 = 6%

Example 3: the difference version. Assume normal growth is 2% a year, actual growth is 0%, and k is 0.5. Then:

Δu = −0.5 × (0 − 2) = +1 point

Unemployment would be expected to rise by about 1 percentage point. If growth instead ran at 4%, the same assumptions imply unemployment falling by about 1 point.

Why Output Moves More Than Unemployment

If output depended only on the number of people employed, a 1% drop in output would roughly match a 1% drop in employment. Okun’s ratio is larger because firms and workers adjust along several margins at once when demand changes.

Okun broke down what happens when GDP rises 3 percentage points relative to its long-run level. In his estimate, that increase was associated with:

1. A 0.5 point rise in the labor force participation rate, as more people look for work.
2. A 0.5 point rise in hours worked per employee.
3. A 1 point rise in labor productivity (output per worker per hour), partly because firms use existing staff and equipment more intensively.
4. A 1 point fall in the unemployment rate.

So only part of the output swing shows up in the unemployment rate. The rest is absorbed by participation, hours and productivity. This also explains why the coefficient varies across countries. Where firing is costly or work-sharing is common, firms adjust hours rather than headcount, and unemployment moves less for the same change in output.

Arthur Okun and the Origin of the Law

Okun proposed the relationship in the early 1960s while working on how to measure potential GNP, the output measure then in common use. His aim was practical: policymakers wanted to know how much output the US was losing by running with high unemployment, and his rule gave a quick way to translate an unemployment gap into lost production.

How Okun’s Law Is Used

Estimating the cost of a downturn. The original use remains the most common: converting excess unemployment into an estimate of lost output, as in Example 1.

Checking forecasts for consistency. Forecasters who project GDP growth and unemployment separately can use Okun’s law to see whether the two projections fit together. A forecast of weak growth alongside sharply falling unemployment would stand out, and the forecaster would need a reason, such as falling labor force participation, to justify it.

Estimating the output gap. Unemployment data are released quickly and revised less than GDP. Analysts sometimes run the gap version in reverse, using the unemployment gap to infer how far output is from potential. That estimate feeds into policy rules such as the Taylor rule, which links interest rate settings to the output gap and to inflation.

Monetary and fiscal policy. Central banks track the output gap because persistent slack tends to restrain inflation, while output above potential tends to add to price pressure. Okun’s law is one tool for gauging that slack. It works alongside inflation measures such as the Consumer Price Index, not in place of them.

How Well Okun’s Law Holds Up

A 2007 Federal Reserve Bank of Kansas City review of quarterly unemployment changes and real output growth found the law largely accurate but “not a tight relationship.” Unemployment often failed to move as predicted and the coefficient shifted over time. Even so, growth slowdowns typically coincided with rising unemployment, so the rule can still help policymakers who allow for that instability.

A 2014 Federal Reserve Bank of San Francisco analysis of the Great Recession found that early GDP figures suggested unemployment had risen far more than Okun’s law implied. After GDP revisions the gap largely closed, and the relationship looked similar to earlier deep recessions.

A Federal Reserve Bank of Cleveland commentary was more skeptical. It found “rolling instability” across several versions of the equation, with observed changes sometimes many times larger than predicted, and concluded that a rule of thumb with many exceptions is not much of a rule.

Limitations of Okun’s Law

• No agreed coefficient. Economists accept that output and unemployment move together but disagree on the size of the link, and estimates change with the sample period, data frequency and version of the equation.
• Unobservable inputs. Potential GDP and the natural rate of unemployment cannot be measured directly, and errors in those estimates carry straight into any Okun’s law calculation.
• Data revisions. GDP figures are often revised substantially. As the Great Recession case showed, an apparent breakdown can disappear once the data are corrected, and the reverse can also happen.
• Other drivers of output. Productivity, technology, capacity use, hours and participation trends affect output independently of unemployment, which weakens the rule as a standalone forecast.
• Country differences. Labor market rules, industry mix and work-sharing produce different coefficients, so a US estimate does not transfer to France, Germany or elsewhere.
• Correlation, not causation. The law describes how the two series have moved together. Neither necessarily drives the other; in practice both respond to the same shifts in demand.

Okun’s Law vs. Related Concepts

Phillips curve. The Phillips curve links unemployment to inflation. Okun’s law links unemployment to output. Economists often combine them: Okun’s law translates an output gap into an unemployment gap, and the Phillips curve translates that into expected pressure on inflation.

Output gap. The output gap is a measurement (actual minus potential GDP, as a share of potential). Okun’s law is a relationship that connects the output gap to the unemployment gap.

Inequality measures. Okun’s law does not predict that a wealth or income gap narrows when GDP rises. Any link between growth and inequality is a separate question with its own research.

The Bottom Line

Okun's law matters most to forecasters, analysts and policymakers who need a quick way to connect unemployment with lost output or the output gap. It works as a rough guide. The coefficient shifts across countries, time periods and equation versions, and it relies on estimates of potential GDP and the natural rate, so results should be checked against other data and revised GDP figures.