Ask an Economist

Michael Owyang is an economist at the St. Louis Fed. His research interests are time series econometrics, forecasting and regional analysis. He likes pepperoni on his pizza and drinks too much coffee. For more on Owyang's work, see http://research.stlouisfed.org/econ/owyang/.

Q. What is potential output, and how is it measured?

A.When discussing the performance of the U.S. economy, people sometimes cite the output gap, which is the difference between actual and potential output. But what is potential output? A common misperception is that it is the maximum output the economy could produce if everyone were employed and all capital were used. Economists define potential output as what can be produced if the economy were operating at maximum sustainable employment, where unemployment is at its natural rate.^[1] Therefore, actual output can be either above or below potential output.

Unlike actual GDP, we cannot observe potential GDP and must estimate it. As a result, different economists can have different views of potential output. One way to construct potential GDP is by fitting a trend line through actual GDP. Looking at a short sample period, however, may lead to an inaccurate estimate of potential. For instance, starting in 2000 would lead to a trend line that is defined by the expansion period and is relatively steep. If, on the other hand, output rose above potential during the expansion period, then the trend line would be slightly flatter. The latter case implies that output would have been above potential during the boom period and perhaps not quite so far below potential during the recession.

Many people believe that the previous decade had a housing bubble, with construction much higher than in normal times. If that is correct, the notion that the economy was producing output above potential prior to the recession does not seem that far-fetched. In that case, actual output today may not be as far below potential as a lot of people think.

Endnotes

1. See Okun, Arthur M. "Potential GNP: Its Measurement and Significance," Cowles Foundation Paper 190, reprinted from the 1962 Proceedings of the Business and Economic Statistics Section of the American Statistical Association. See http://cowles.econ.yale.edu/P/cp/p01b/p0190.pdf. [back to text]