Hours worked by women age 55 to 61 confound labor market analysis
November 26, 2019
The U.S. economy accomplished an exceptional feat in September 2019—a record, 123-month expansion, according to the National Bureau of Economic Research.
The achievement was all the more remarkable because it occurred with the unemployment rate at 3.5 percent, a half-century low. Historically, low unemployment rates have been associated with tight labor markets and higher wages that push inflation well above target. Oddly, that wasn’t the case in September. The year-over-year increase in the personal consumption expenditures price index, an indicator of U.S. inflation, came in at 1.3 percent, significantly below the Federal Reserve’s 2 percent inflation target.
The puzzle has prompted researchers to look at U.S. labor market conditions using less-common indicators, such as the labor input utilization rate (LIUR). The LIUR is the share of available time devoted to work by the working-age population (age 16 and older).
This indicator’s readings of labor market tightness are more ambiguous than what the unemployment rate suggests. The reason: The seemingly anomalous low LIUR of one particular demographic group—women age 55 to 61 years old—poses a challenge for assessing the extent to which the U.S. economy is currently close to full employment.
Identifying the labor input utilization rate
More specifically, the LIUR is the fraction of the discretionary available time that working-age individuals are actually at work on average, as opposed to engaged in non-market activities such as meeting physiological and immediate needs, including sleeping, eating, health care and commuting to and from work.
The LIUR has its foundations in the observation that, historically, the share of discretionary time that the working-age population is at work has typically moved around a constant value, which could be regarded as the underlying trend in countries with stable workforce demographics. Therefore, the gap between the actual value of the LIUR and its trend value at a given period can be interpreted as an indicator of labor market conditions.
However, the demographic structure of the U.S. has changed dramatically in recent decades, owing to the unusually large cohorts of baby boomers at retirement age or approaching it. Thus, the gap between actual LIUR and a demographically adjusted trend value is more appropriate for assessing whether U.S. labor markets are operating above or below normal.
The Dallas Fed in December 2018 estimated this trend by calculating the LIUR specific to working-age individuals of each gender in relevant age brackets and by inferring the underlying trend of the LIUR for each of those groups from data prior to the 2008–09 Great Recession. The demographically adjusted overall trend of the LIUR in Chart 1 (blue line) was obtained by simply adding up the LIUR trend specific to each demographic group, weighted by the share of the corresponding group in the overall working-age population.
As is apparent from the chart, the labor market by second quarter 2019 was not as tight as suggested by the prevailing near-record low unemployment rate because actual LIUR was about 1 percent below its demographically adjusted trend.
The exception: women age 55 to 61
Confidence in such an assessment hinges critically on the extent to which the trend of the LIUR specific to each demographic group prevailing before the Great Recession was correctly identified by this methodology. In fact, more recent data suggest that the LIUR value for each demographic group in the second quarter is rather close to what would have been predicted by its corresponding pre-Great Recession trend (Chart 2).
There was one quantitatively relevant exception to that apparent uniform behavior. The LIUR for women age 55 to 61 in the second quarter of this year was far below what would be expected from the trend before the Great Recession (Chart 3). The gap between the actual LIUR and the trend value for this group has apparently been widening, in sharp contrast with the progressive shrinking of the corresponding gap for most of the other groups shown in Chart 2.
This raises the possibility that the trend for women 55 to 61 years old experienced a structural break after the Great Recession and is no longer the dashed, upward-sloping blue line in Chart 3 but rather the much flatter dashed red line.
Under this alternative assumption, the group’s LIUR was much closer to its trend value in second quarter 2019. Thus, it would be fair to conjecture that the gap detected earlier for the entire working-age population would nearly vanish. In fact, that is the case when the overall demographically adjusted LIUR trend in the recent quarter is recalculated—assuming that for women 55 to 61 years old, the LIUR trend value corresponds to the lower one to which this population subset appears to have shifted after the Great Recession. To be more specific, when reestimated this way, the second-quarter value of the demographically adjusted LIUR trend falls almost exactly on top of the corresponding observed LIUR value of 0.228 reported in Chart 1.
A further, closer look at the data failed to identify a specific reason—such as educational or racial attributes—for such a trend shift for women 55 to 61 years old. If it did occur, the apparent structural break means that women in this age group are devoting less available time to work, which has implications for understanding the level of full employment in the U.S. However, affirming or denying the presence of such a structural break would require additional research and analysis.
Shrinking labor market slack
An updated reading of labor market conditions using the LIUR indicator suggests that some slack remained in the U.S. labor market by second quarter 2019, although not as much as in the prior year. But this assessment is subject to an important caveat: Data released in the intervening 12 months strongly hint at the possibility that the LIUR trend of a specific demographic group—women age 55 to 61—experienced a structural break following the Great Recession.
If more conclusively established by subsequent studies and evidence, such a structural break would imply there was no gap between the overall actual LIUR and its demographically adjusted trend by mid–2019, supporting the view of labor market tightness.
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.