Economic Research Publications
National Economic Update
Recent Unemployment Increases Break the Law
June 29, 2009
The recent recession has hit the U.S. economy hard. Since December 2007, 6 million jobs have been lost, the unemployment rate has nearly doubled and real GDP growth is well into negative territory. Recent reports, however, show most measures of real economic activity are either stabilizing or declining less rapidly, and forecasters have become more optimistic about near-term growth prospects. Usually, when the growth outlook improves, forecasted unemployment rates decline. This isn't happening. In fact, unemployment forecasts have been revised upward. How can the outlook for real activity improve, while the outlook for the labor market worsens?
Real GDP and Unemployment
In fourth quarter 2008, real GDP fell at a staggering 6.3 percent annual rate, the steepest decline in 27 years. First quarter 2009 growth was slightly less terrible at –5.5 percent, and forecasts for the remainder of the year are relatively encouraging, calling for 1.9 percent growth in the fourth quarter according to the Blue Chip Consensus Forecasts (Chart 1). Meanwhile, the unemployment rate is expected to continue rising, reaching 9.8 percent by the fourth quarter (Chart 2).
There's nothing especially unusual about rising output combined with rising unemployment: That's a coupling we see whenever growth isn't rapid enough to absorb new entrants into the labor force. What is unusual is to see output and unemployment forecasts revised upward at the same time, as occurred between April and June of this year (Table 1). Stronger output growth ordinarily means stronger jobs growth and, hence, a lower unemployment rate. This empirical relationship is so strong and so reliable that it has come to be called a law—"Okun's law," after Kennedy administration economist Arthur Okun.
|Blue-Chip Consensus Forecasts|
|NOTES: GDP growth is seasonally adjusted annualized percentage rates. The unemployment rate entries are percentages, measured at the end of the period.|
|SOURCE: Blue Chip Economic Indicators.|
Okun argued that there is a tight negative link between output growth, relative to potential, and changes in the unemployment rate relative to the "natural," or long-term equilibrium, unemployment rate. Empirically, the unemployment rate rises at a rate of 1 percentage point per year, relative to the natural rate, for each 2.5 percentage points that realized output growth falls short of potential.
Ordinarily, changes in the natural unemployment rate occur gradually enough that one doesn't need to worry about them, leaving a relationship between changes in the output gap and changes in the unemployment rate.
Chart 3 shows Okun's law in operation. On the vertical axis is the percentage-point change in the unemployment rate. On the horizontal axis is the percentage-point change in the output gap, as estimated by the Congressional Budget Office (CBO). Each point in the scatter diagram matches a four-quarter unemployment change with a four-quarter output-gap change one quarter earlier. (The one-quarter lag reflects the sluggishness of unemployment's response to changes in output growth.)
Notice how all points cluster closely around the fitted line—all except the point that corresponds to 2009. The unemployment rate rose much more rapidly between first quarter 2008 and first quarter 2009 than was predicted by Okun's law. Unemployment increases in April and May 2009 were also unusually large, hence the upward revisions to the Blue Chip's unemployment rate forecasts since April.
So, what's gone wrong? There are several, not mutually exclusive possibilities:
- Okun's law has broken down, if only temporarily. Firms, responding to unusual cash-flow pressures arising from dysfunctional financial markets, have been unusually quick to lay off workers. This may keep unemployment rates well above normal until financial markets get back on track, even if the economy as a whole is recovering in the meantime.
- There has been a significant upward jump in the natural rate of unemployment caused by rising frictional unemployment. This explanation is certainly plausible. Frictional unemployment describes the temporary unemployment created as workers change jobs. In the current recession, there have been big, long-lasting reductions in the number of jobs available in construction, real estate, finance and auto manufacturing. It takes time for those who have lost their jobs to find new ones in other fields or to decide to leave the labor force. Even labor mobility has been impaired because so many mortgages are underwater. Older workers may be delaying planned retirements, and some retirees may even be re-entering the labor force. Young people may be having trouble financing college, due in large part to the financial discord, and turning to the labor market instead. All of these situations point to an increase in labor supply coupled with a decrease in labor demand, which significantly increases unemployment. The rise in the natural rate, however, would have to be enormous to explain the large discrepancy noted in Chart 3, and although this is certainly possible, it's not likely to be the whole explanation.
- GDP growth has been overestimated. Future revisions of real GDP may show slower growth during 2008 than is indicated by current official data. Real GDP, however, would need to be revised downward nearly 4 percentage points to help explain the discrepancy alone; this is unlikely to be the case.
- The CBO has substantially underestimated the economy's growth potential, so the true output gap has widened by more than the current CBO estimates suggest. Estimating potential GDP is as much an art as it is a science, especially since productivity growth, labor force growth and net capital formation estimates—significant explanatory factors of potential GDP—are subject to large after-the-fact revisions. The CBO estimates, however, would need to be understated by 4 percentage points to account for the large recent divergence from Okun's law.
Data from the Job Openings and Labor Turnover Survey (JOLTS) lend some credence to the first two explanations: Cash-flow pressures have accelerated layoffs, and there has been an increase in frictional unemployment. According to the JOLTS, hiring and quit rates have drifted downward in parallel during the current recession, much like they did in 2001 (Chart 4). However, the "layoffs and discharges" rate, which was basically constant during the 2001 recession, soared in fourth quarter 2008. The timing of the rise in the layoffs and discharges rate suggests that dysfunctional credit markets may have driven firms to shed workers at an accelerated pace. Also, because it takes longer to find a new job after being laid off than after quitting, it is likely that frictional unemployment has increased along with the layoffs and discharges rate.
What Does the Future Hold?
Blue Chip forecasters are assuming that Okun's law will be reestablished in 2010 (Chart 5): GDP declines are expected to be every bit as severe during 2009 as 2008, but the resultant rise in the unemployment rate is expected to be much more nearly "normal" between first quarter 2009 and first quarter 2010. Implicitly, forecasters are assuming that the unusually large unemployment increases we've seen over the past year were not the result of a "front-loading" of layoffs—for if that were the case, the green triangle in Chart 5 ought to lie below the Okun's law line, rather than (more or less) on it.
—Nicole Ball and Evan F. Koenig
About the Author
Ball is a senior research analyst, and Koenig is a vice president and senior policy advisor in the Research Department at the Federal Reserve Bank of Dallas.