Moderate Wage Growth Spurs Search for ‘Hidden Slack’ in Labor Market
Some analysts have been puzzled by what they see as moderate wage growth at a time of historically low unemployment. In this article, we search for “hidden slack” in the labor market as a reason for this puzzle and find that underemployment doesn’t explain why wages haven’t risen more.
In recent years, much has been made about the idea of hidden slack—unused labor capacity not captured by the unemployment rate. The financial press has extensively discussed this notion. We have previously explored the conundrum of why wages haven’t risen more from both the perspective of wage-growth measurement and the relationship between wage growth and unemployment.
In this article, we consider the unemployment gap, the traditional measure of the slack or tightness in the labor market, defined as the difference between the reported unemployment rate and the “natural rate” of unemployment. (The natural rate is an estimate of the unemployment rate that would exist in a “neutral” labor market with the economy operating at its potential level of output.) We then compare the unemployment gap to an analogous gap measure constructed for underemployment.We argue that the unemployment gap using the traditional unemployment rate performs well at measuring the degree of slack or tightness in the labor market and that a broader measure based on underemployment adds little useful information.
Difficulty of Measuring Full Employment
The unemployment gap is criticized for missing possible sources of labor market underutilization not reflected in the unemployment rate. To be counted as unemployed, an individual must be actively looking for a job. The rate can mismeasure labor market slack for individuals in several ways:
- Part-Time Workers for Economic Reasons—Individuals who are employed but not working as many hours as they want.
- Marginally Attached Workers—Individuals who want full-time work and have looked for work in the last year but are not currently looking.
Chart 1 shows these two categories of workers, plus unemployed workers, which together we refer to as “underemployed” workers.
While unemployment in August stood at 3.7 percent, underemployment was significantly higher, 7.3 percent. Most of the difference reflects individuals with jobs but who are not working as much as they would like. “Unemployment” and “part time for economic reasons” make up almost all of the underemployment (6.4 percent out of 7.3 percent in August).
Natural Rate of Underemployment
To measure labor market slack with the underemployment rate, we calculate a “natural rate of underemployment” using the same approach we previously applied to the unemployment rate. We estimate life-cycle underemployment profiles for 360 cohorts (each with a distinct race, gender, age and education) that allows us to remove the effects of the business cycle. An example of one of our cohorts is white males born in the 1950s with a high school education. These life-cycle profiles indicate the underemployment rate for individuals in that cohort at each age in a neutral labor market.Chart 2 shows an example of these life-cycle underemployment profiles for the cohorts of white males born in the 1950s, broken down by two different education levels.
The degree of underemployment relative to unemployment is larger for less-educated workers. Surprisingly, the degree of underemployment declines only slightly over workers’ careers up until they reach their 50s. At that time, both unemployment and underemployment increase over roughly the next 10 years.We use these underemployment profiles to generate our natural rate of underemployment. Using this measure, we can derive the slack or tightness in the labor force by comparing the actual underemployment rate with the estimated natural rate of underemployment.
Comparing Underemployment, Unemployment
Let’s return to the original issue of whether underemployment gives us valuable additional information about labor market conditions. To begin, we can examine the actual and natural rate for both underemployment and unemployment (Chart 3).
Demographic changes have pushed down the natural rate of underemployment and unemployment over the past 40 years. These shifts have affected the two natural rates in a very similar way, with the difference between the two only narrowing slightly over the 40 years. The difference between the observed rate and the natural rate for both underemployment and unemployment is the key to understanding current conditions in the labor market (Chart 4).
These gap measures indicate the extent of slack or tightness in the labor market, and they tell the same story. Positive gaps indicate periods of slack labor markets, and negative gaps indicate periods of tight labor markets. Over the past 40 years, the timing of when slack has been removed from the labor market (when the gap moves from being positive to zero) is similar using both gap measures, typically differing by six months or less.
In the current expansion, the difference in timing of when the labor market reached neutral is the longest over the period studied, at slightly over a year. The unemployment gap indicates that the labor market reached neutral in August 2015, while the underemployment gap indicates that the labor market reached neutral in November 2016. Overall, adding in underemployment does not materially change when we would classify the labor market as being slack or tight. In particular, the underemployment gap does not support the view that currently there is still hidden slack in the labor market.
Assessing the Utility of the Underemployment Rate
Does using the broader measure of underemployment help explain wage behavior? For this to be the case, there would need to be useful information provided by the underemployment gap that is not conveyed by the unemployment gap.
However, a simple regression shows that the unemployment gap explains 96 percent of the movement in the underemployment gap. That is, once we control for the unemployment gap, the underemployment gap does not add additional useful information about the state of the labor market.
Is there hidden slack in the labor market? After constructing an underemployment gap and comparing it to the unemployment gap, we see no evidence of hidden slack. The traditional unemployment gap can continue to serve as a principal measure of labor market conditions.
About the Authors
The views expressed are those of the authors and should not be attributed to the Federal Reserve Bank of Dallas or the Federal Reserve System.