Okun’s Law is a well-known proposition that relates the pace of economic growth to unemployment.
Traditionally, policymakers have used Okun’s Law as one indication of maximum sustainable growth. If
the pace of economic growth has a consistent impact on unemployment, then policymakers can target a
particular pace of growth that sets the unemployment rate to a level consistent with stable inflation.
On a seemingly regular basis, however, the economics profession comes to believe that Okun’s Law has
broken down. Most recently, this breakdown occurred in the wake of the 2007-09 recession. From the
Wall Street Journal:
Okun’s law has been a staple tool for economists ever since, but it’s been driving them crazy
lately because it doesn’t seem to be working all that well.
Similar concerns were voiced in the late 1970’s. From the Washington Post:
What has punched holes in Okun’s Law in recent months is the change in another somewhat
arcane-sounding measure – the size of the productivity gains that the nation can expect to
achieve over the long run.
In this project, you will explore the different versions of Okun’s Law over different time periods. The three
versions of Okun’s Law you will study are:
∆�# = �& + �(∆�#
�# = �& + �((�# − �#
∗
)
(�# − �#
∗) = �& + �((�# − �#
∗)
where:
�# is the unemployment rate.
∆�# is the change in the unemployment rate over the past year, ∆�# = �# − �#./ for quarterly
data.
�#
∗ is the natural rate of unemployment.
�# is real output (gross domestic product), logged.
∆�# is (logged) real output growth over the past year, measured as the log difference * 100, ∆�# =
(�# − �#./) ∗ 100 for quarterly data.
�#
∗ is real output (gross domestic product), logged.
All variables can be found on the FRED website. We are using quarterly data in this example
To begin, please read the following paper:
Knotek, Edward S. 2007. “How Useful is Okun’s Law,” Economic Review, Federal Reserve Bank
of Richmond, Fourth Quarter 2007.
Then do the following:
- Estimate each of the above forms of Okun’s law over the samples 1984:1-2007:4 and 2008:1-
2018:4, a total of six regressions. - Report the estimated coefficients, t-values, p-values, and �(’s for each equation. Identify which
coefficients are significantly different than zero. - Using the 2008:1-2018:4 sample, write the estimated equation for each form of Okun’s law.
Interpret each equation (how does the right-hand side impact the left-hand side?). Examples of
interpretation can be found in the reading (see pages 72-73, for example). - Consider the third version of Okun’s law above. This relates the unemployment gap to the output
gap. What is the theoretical value of �&? What are your estimates of �&? Impose the constraint the
�& = 0 (in other words, do not include a constant term in the regression). How does this constraint
change your estimate of �(? Complete for both sample periods. - Do your Okun’s Laws estimates change between the sample periods? How do they change (does
unemployment become more or less sensitive to output)? - According to Knoteck, what might account for changing estimates of Okun’s law?
- According to the Wall Street Journal article above, what might account for changing estimates of
Okun’s Law?