July 21, 2015
Last quarter the civilian unemployment rate fell to 5.41 percent, the lowest rate since the second quarter of 2008. This represented a drop of 0.16 percentage points from the first quarter, when unemployment was 5.57 percent.
According to the Congressional Budget Office (CBO), this also meant that the economy nearly achieved full employment. The CBO estimates that the natural rate of unemployment, an estimate of the unemployment rate that excludes cyclical effects, was 5.38 percent in the second quarter. (The CBO publishes a short-term as well as a long-term rate; both were 5.38 percent in the second quarter. The two rates are expected to be the same going forward.)
The CBO’s estimates of the natural rate of unemployment are inconsistent with its estimates of the output gap. The output gap measures the difference between GDP and what GDP could be if there were no cyclical weakness in the economy. There should therefore be no output gap when the economy has hit its natural rate of unemployment.
In the first quarter of 2015, CBO pegged the natural rate at 5.39 percent. Okun’s Law—an empirical regularity which associates a 2 percent increase in output with a 1 percentage point drop in the unemployment rate—tells us that the 0.18 percentage point gap between the unemployment rate and the natural rate of unemployment implies lost output of 0.36 percent of GDP. Given that GDP in the first quarter was $17.7 trillion,* this would imply an output gap of $63.7 billion. However, CBO’s own estimate of potential GDP assumes an output gap of over $600 billion, nearly ten times the gap consistent with its estimated natural rate of unemployment.
The GDP data for the second quarter haven’t been released yet (they come out on July 30), but it’s hard to imagine that the newest figures will resolve this discrepancy. With the unemployment rate falling to 5.41 percent and the natural rate falling to 5.38 percent, the difference between the two rates was just 0.03 percentage points in the second quarter. Okun’s Law tells us that this represents an output gap of 0.06 percent of GDP. In order for the output gap to fall to just 0.06 percent of GDP given the CBO’s second-quarter potential GDP estimate of $18.4 trillion, nominal output will have to grow at an annualized rate of 17.4 percent between the first and second quarters.
The CBO’s estimates of the output gap and the natural rate of unemployment appear inconsistent. Notably, the discrepancy disappears when we examine other measures of employment. For example, the prime-age employment rate, which I’ve previously argued is the single best measure of labor market slack, appears highly consistent with the CBO’s estimated output gap. The first-quarter output gap of $605 billion is compatible with a prime-age employment rate of 79.8 percent.** CBO estimates that the last time the U.S. didn’t have an output gap was the second quarter of 2006, when the prime-age employment rate was 79.7 percent (and unemployment was 4.7 percent). This implies that the prime-age employment rate, unlike the unemployment rate, is a consistent predictor of the output gap. Instead of being quite close to full employment—as CBO’s 5.38 percent natural rate of unemployment implies—both the output gap and the prime-age employment rate suggest that we are quite far from full employment. We aren’t likely to achieve full employment any time this year. At the rate that prime-age employment has been increasing, we are likely years away from a healthy labor market.
*Note that this is an annualized figure. All figures for both quarterly GDP and the output gap represent annualized levels.
**This was estimated through the following set of equations:
- First quarter GDP ($17.7 trillion) / First quarter workforce (148.3 million workers) = $119,327 per worker
- First quarter output gap ($604.7 billion) / $119,327 per worker = 5.1 million non-employed workers
- First quarter prime-age workforce (96.5 million workers) / Total workforce (148.3 million workers) = Prime-age workers are 65.1% of the workforce
- 65.1% of the 5.1 million non-employed employed workers is 3.3 million, implying that if 5.1 million Americans were to gain employment, 3.3 million would be prime-age workers
- 3.3 million non-employed prime-age workers / 125.0 million prime-age Americans = 2.6% of prime-age population
- 77.2% prime-age employment rate + 2.6% of population not employed due to output gap = 79.8% prime-age employment rate with no output gap
