The Okun's law calculator above applies the empirical relationship between unemployment and output that Arthur Okun described in 1962. It handles both standard forms — the gap version, which links the output gap to the unemployment gap, and the difference version, which links the change in unemployment to growth relative to trend — and it takes the coefficient as your input rather than supplying one.
Arb Digital publishes it with a caveat that belongs in the opening rather than in a footnote. Okun's law is a statistical regularity, not a law and not a structural model. Its coefficient is estimated from data, differs between countries and between decades, and is disputed in the literature. Nothing this page produces is a forecast, and a figure computed here is a statement about a historical correlation applied to numbers you supplied.
What This Okun's Law Calculator Does
In gap form it takes the difference between the unemployment rate and an estimate of the natural rate, multiplies it by your coefficient, and reports the implied output gap as a percentage of potential output. Given a potential output level it also expresses that gap as a level, which is the form in which it usually gets quoted.
In difference form it takes real GDP growth relative to trend growth, divides by your coefficient, and reports the implied change in the unemployment rate over the period, along with the rate that would result.
Every input is yours. The tool publishes no unemployment rate, no growth figure, no natural rate and no coefficient default that pretends to be authoritative — the value in the box is a placeholder so the page loads with a worked example. Take the real figures from your national statistics office, and say which series and which vintage you used, because both get revised.
How to Use It
- Choose the form your source uses. A coefficient estimated in gap form is not valid in difference form.
- Enter the coefficient from the study you are working with, including its estimation period and country.
- Enter the rates for the form you selected, all as percentages.
- Read the output as an implication of the relationship, not as a prediction of what will happen.
- Vary the coefficient across the range your source gives and note how far the answer moves. That spread is part of the result.
The Formula and How It Is Calculated
The gap form states that the percentage shortfall of output below potential is proportional to the excess of unemployment above its natural rate: (Y − Y*) ÷ Y* = −c × (u − u*), where c is Okun's coefficient. A coefficient of 2 means every percentage point of excess unemployment is associated with output about two percent below potential.
The difference form avoids the unobservable natural rate by working in changes: the change in the unemployment rate equals minus one over c times the amount by which real growth exceeds trend growth. Equivalently, growth above trend by c percentage points is associated with a one-point fall in unemployment.
The Federal Reserve Bank of San Francisco's Economic Letter "Interpreting Deviations from Okun's Law" by Mary Daly, John Fernald, Òscar Jordà and Fernanda Nechio states the rule of thumb in this form: a two percent drop in inflation-adjusted GDP growth relative to trend is associated with about a one percentage point increase in the unemployment rate. They add that Okun's law is a simple statistical correlation which has held up surprisingly well over time — a description that is precise about what kind of claim it is.
Why The Coefficient Is Disputed
Okun's original estimate for the United States was near 3. Later work has produced estimates clustering closer to 2, with substantial variation, and there is no single agreed figure even for one country over one period.
Three reasons the estimates move. Estimation method: the gap form requires estimates of both potential output and the natural rate, neither of which is observed, so the coefficient inherits whatever filtering method produced those series — and different filters give different answers on the same data. Period: several studies find the coefficient's magnitude changed in the 1990s and did not return, and results generally suggest the relationship behaves differently in expansions than in contractions. Country: labour market institutions differ so widely that estimates for different economies are not comparable at all — where dismissal is costly and hours adjust instead of headcount, the same output shortfall produces far less measured unemployment.
The practical implication is that a coefficient must always be quoted with its country, its period, its form and its estimation method attached. A number carried across any of those boundaries is not being used for what it measured. That is why this tool takes it as an input and offers no default it stands behind.
Why The Relationship Exists At All
The correlation is not a coincidence, and understanding the mechanism explains where it breaks.
Output falling by more than employment is the normal pattern because firms adjust several margins before they cut headcount. Hours per worker fall first. Overtime disappears. Labour productivity dips as retained staff work below capacity, which economists call labour hoarding: firing and rehiring is expensive, so firms hold trained workers through a downturn they expect to be short. Only after those margins are exhausted does the headcount move.
Participation is the other margin, and it is why the measured unemployment rate can understate a downturn badly. People who stop looking for work leave the labour force and stop being counted as unemployed, so the rate can fall while employment is also falling. Our unemployment rate calculator covers the definitional choices that determine who ends up in that denominator, and the choice matters here: an Okun coefficient estimated on one unemployment definition should not be applied to another.
What The Numbers Cannot Support
A figure from this page is a correlation applied to inputs, and three claims it cannot support are worth naming.
It is not causal. The relationship says nothing about which direction the causation runs, and both plausibly run at once — weak demand raises unemployment, and high unemployment weakens demand. A policy conclusion of the form "raise output by x and unemployment falls by y" is not supported by a correlation, however stable that correlation has been.
It is not a forecast. Applying the relationship to a projected growth number produces an implication conditional on the relationship continuing to hold with the coefficient you entered. Both conditions have failed in the past, most visibly during the jobless recoveries when unemployment stayed high through periods of growth.
And it does not survive structural change. A relationship estimated over a period of stable labour force participation and stable industry composition is describing that period. Where composition shifts — different sectors, different contract types, different measured hours — the historical coefficient describes an economy that no longer exists. The same caution applies to every macro figure you feed it: real GDP series are revised, sometimes substantially, and the United States Bureau of Economic Analysis publishes its gross domestic product data in successive estimates for exactly that reason. An output gap computed from a first estimate can change with the second.
How It Fits With The Other Macro Relationships
Okun's law is one of three simple empirical relationships that together form the skeleton of an introductory macro model. The Phillips curve links unemployment to inflation. The Taylor rule links inflation and the output gap to a policy interest rate, and our Taylor rule calculator handles that one, which shares the same dependence on an unobservable output gap. Okun's law is the link that converts between the output gap and the labour market, which is what lets the three connect.
All three share the same weakness, and it is worth stating once: each depends on a quantity nobody can measure. The natural rate of unemployment, potential output and the neutral real interest rate are all estimated, all revised, and all capable of moving a conclusion substantially. Chaining the three together compounds those uncertainties rather than cancelling them. Our GDP calculator works with the expenditure identity that defines the output side, and our inflation calculator handles the price adjustment that converts nominal figures to real ones before any of this arithmetic can begin.
Arb Digital produces explanatory content that separates what a figure measures from what it is being used to argue, with every source named.
Browse All Free Tools Talk To Our TeamCommon Mistakes to Avoid
- Treating the coefficient as a constant — it is an estimate that differs by country, period, form and method, and quoting it without those is quoting nothing.
- Mixing the two forms — a coefficient fitted in gap form does not belong in the difference form, and the two are different regressions.
- Reading the result as causal — the relationship is a correlation, and causation plausibly runs in both directions at once.
- Ignoring the unobservable inputs — potential output and the natural rate are estimates that get revised, and the answer moves with them.
- Presenting the output as a forecast — it is conditional on a historical relationship holding, which during jobless recoveries it did not.
Related Free Tools From Arb Digital
Work the labour force arithmetic with the unemployment rate calculator, build the output side with the GDP calculator, apply the policy rule with the Taylor rule calculator, convert nominal figures to real using the inflation calculator, or handle the underlying change arithmetic with the percentage change calculator. The free online tools hub lists every economics tool we publish.
Frequently Asked Questions
An empirical relationship, described by Arthur Okun in 1962, between the shortfall of output below potential and the excess of unemployment above its natural rate. It is a statistical regularity rather than a structural model.
The one estimated for the country, period and form you are working with. Okun's original US estimate was near 3 and later work clusters closer to 2, with substantial variation and no single agreed figure.
The gap form relates levels, using estimates of potential output and the natural rate. The difference form relates changes, avoiding those unobservables. They are different regressions and their coefficients are not interchangeable.
No. Applying it produces an implication conditional on the relationship continuing to hold with the coefficient you entered. It has broken down before, most visibly during jobless recoveries.
Because firms cut hours, overtime and effective productivity before headcount. Retaining trained staff through a short downturn is cheaper than firing and rehiring, which economists call labour hoarding.
Substantially. Labour market institutions determine whether firms adjust hours or headcount, so the same output shortfall shows up as very different measured unemployment in different economies.
Not on its own. The relationship is a correlation and says nothing about the direction of causation, so it cannot establish that a change in one quantity would produce a change in the other.
This page explains an empirical macroeconomic relationship for educational purposes. It is not financial advice, not a forecast and not a policy recommendation; every figure it produces depends on a coefficient and on unobservable estimates that you supply, and a qualified economist should be consulted before relying on any of it.
