- Okun’s law reports a negative relationship between the unemployment rate and GDP growth.
- In spite of several episodes of "jobless recovery" it remains valid.
- The link has even strengthened after the Great Recession, which could comfort the Fed in its long status quo.
Following my post on the labor mismatch, I focus today on a relationship that was first reported by Arthur Okun in 1962. The Okun's law reports a negative relationship between the unemployment rate (or positive with job creations) and GDP growth.
A succession of jobless recoveries has put into question the validity and the stability of the relationship. The reason why the curve matters is linked to the implementation of economic policy - monetary in particular. If the relationship does not hold, then endeavors to foster domestic demand will falter and prove useless. If, on the contrary, the link remains robust, then the Fed has the good strategy to focus on aggregate demand.
I estimated three different kinds of Okun's law over 10-year rolling periods, starting with the 1960/70 time frame:
i. The 1-quarter and 4-quarter log change of Nonfarm payrolls (Establishment survey) and total employment (Households survey);
ii. The 1-quarter and 4-quarter absolute change of the unemployment rate (U-3 - see BLS for alternative measures).
All against U.S. GDP growth. The charts below present the results.
The link between GDP growth and Nonfarm payrolls has remained constant through the period, even though it weakened somewhat in the 1990s and early 2000s. It can be explained by the change of labor force management by businesses through the economic cycle (enhanced focus on the stability of profitability).
There is a striking difference in the late period when the Households survey is taken into account. As can be seen below, the link between growth and employment surged during the Great Recession, highlighting the violence of layoffs in the aftermath of the financial crisis. The reason why this feature is visible only for the Households survey is unclear, as there is no change in the fact that the Households survey is structurally more volatile than the NFP.
Given that the unemployment rate is calculated with this series, it is no surprise to see a strengthening of the Okun's law over the recent past. This means that after one decade and a half when the link weakened (curves drifting upward), the recent strengthening of the relationship highlight the renewed strength between growth and unemployment (I find the same pattern if I roll the estimate over a 5-year period).
The same pattern is visible for the broadest measure of unemployment (U-6: long term / structural unemployment).
Bottom line: The link between GDP growth and changes in unemployment has strengthened over the recent years. This comforts the Fed in its view that a stronger aggregate demand reduces unemployment. Some investors might be concerned by the possibility that the relationship may have been skewed by the behavior of the participation rate (sharp decline since 2008). I ran an estimate using the 1-year change of unemployment with constant participation rate (starting in late 2008). The link has strengthened after the Great Recession too.
No hesitation here: Okun's law is still valid and calls for growth to reduce unemployment. It also suggests that the Fed will take time before it tightens.