In the Nation's Interest

Monthly Member Update Call: To Understand the Labor Market, We Need to Do More Than Just Count Jobs

Earlier this month the April employment report was released, and much to economists’ relief, the major indicators essentially matched expectations:

• Total nonfarm payroll employment increased by 223K—much improved from the downwardly-revised 85K gain in March.  (The biggest gains were in the business and health care services industries, and in construction’s “specialty trade contractors” jobs.)
• The unemployment rate was 5.4%, down slightly from 5.5% in March.
• The labor force participation rate edged up to 62.8% (from 62.7% in March), but has moved little over the past year, remaining in the 62.7-62.9% range.
• Average hourly earnings for all employees on private nonfarm payrolls rose 3 cents in April, and over the past year have increased by only 2.2%.

Overall, these numbers were as expected but nothing spectacular, and while they indicate the labor market continues to improve, economists continue to be somewhat disappointed in the lack of robustness in this recovery.  Economists also seem uncertain about how close the economy is to “fully recovered,” given how we seem to keep waiting for better news.

It’s an especially challenging time to understand what’s going on in the labor market given that our traditional economic measures don’t provide all the detail we would need to either: (i) separate demand-side from supply-side forces, or (ii) separate the effects of the “recovering” economy from various longer-term trends affecting the labor market.

(i) Separate labor demand-side from labor supply-side forces: In Econ 101 terms, all we see are the equilibrium outcomes which are where the labor markets clear—we don’t see the multitude of shapes that trace out all the demand and supply curves, by industries, by geographies, by skills and qualifications of the workforce, by demographics (age, gender, family types).  An increase in jobs can be consistent with either an increase in businesses’ demand for workers, or an increase in the supply (working-age population and participation rates).  Across industries and geographies, jobs can be increasing for different reasons.

(ii) Separate the trends reflecting the economic recovery from the Great Recession from those characterizing the “new normal” state of the labor market and economy.  Economists tend to focus on the movements of the major employment statistics (number of jobs, unemployment rate, labor force participation rate, hourly earnings) as indicating the robustness of the recovery.  But some of these statistics may be behaving differently in this recovery, because they may be headed to a “new normal” that’s quite different from the old, pre-recession normal. 

An example is the broader measures of unemployment, or “underemployment” that include people working part-time “for economic reasons”—which means because of “slack work or unfavorable business conditions, inability to find full-time work, or seasonal declines in demand.”  That labeling of part time status “for economic reasons” (suggesting involuntarily) is self-reported from the Current Population Survey of households, and economists view the steady and continued decline in that number over the recovery, yet still elevated level compared with prior to the recession, as evidence that the economy is still recovering and not yet fully recovered. At the same time, the number of people who work part-time “for noneconomic reasons”—which means due to “childcare problems, family or personal obligations, school or training, retirement or Social Security limits on earnings, and other reasons” (suggesting more out of choice) has increased quite steadily since the 1950s with few interruptions even during past recessions. This number actually dropped over the Great Recession as the involuntary or economic part-time job status increased, but now the voluntary/noneconomic part-time status has resumed the upward trend since 2010 and is almost (but not quite) back to its pre-recession peak.  In fact, over the past year the increase in those working part-time for noneconomic reasons has exceeded the drop in those working part-time for economic reasons.  Yet some of these “noneconomic” reasons may actually be more fundamental or longer-term economic reasons that are self-characterized as a voluntary choice, when in fact they reflect poor or weakened labor market prospects that cause people to choose to do child care, or go to school, or simply enjoy more leisure time (to consume experiences more than things!), as the opportunity costs of working part-time instead of full-time decrease—a labor supply-side phenomenon. But some of this increase in noneconomic part-time work is likely a demand-side phenomenon as well, because the service-providing sectors of the economy (where part-time work is more common) are growing as the goods-producing sectors are shrinking.

I will be elaborating a bit on the part-time work data on the CED blog this week (separating the various reasons why people report working part time), and in particular will look at whether female or older-worker participation explains or at least correlates with the trend in the noneconomic/voluntary part-time count.  Spoiler alert: it doesn’t explain it as much as it used to, probably because men are behaving more like women—from a labor-market perspective at least—and because the older people aren’t behaving so differently from the young (good news!).

So, how could we better separate the demand-side from supply-side factors, and the cyclical from longer-term factors in analyzing the state of the U.S. economy?  Well, a fundamental limitation of the monthly employment report is that these are aggregate and averaged numbers for the entire economy, which both dampens and obscures the movements happening at the more micro level; if two things move in opposite directions, on average or in aggregate, they don’t move at all.  It’s likely that labor markets are behaving differently in different metropolitan areas, in different types of industries, and among different types of people—of different ages, genders, skills, family composition, and just plain tastes for market work vs. home production or leisure.  Just like changing demographics seem to be affecting what’s going on with consumer spending, they seem likely to be affecting labor market trends as well.  Uncovering and sorting out these factors will require deeper dives into the labor market data at industry-, geographic-, and demographic-specific levels so that supply-side factors can be separated from demand-side factors and cyclical factors distinguished from the new longer-term trends.  Given that the macroeconomy is the sum of all its micro-components, which are constantly changing for different reasons at different places with different people at different times, I think macroeconomic forecasters would be wise to spend more time analyzing and explaining micro-level data rather than continuing to rely on time-series regression analysis that predicts the future of our economy using past macro data.

So counting jobs doesn’t tell us much about how the labor market and the broader economy are doing.  We need to take a closer look at what’s going on at the ground level—with high-resolution, zoomed-in video rather than low-resolution, wide-angle snapshots, in order to understand all the different ways and directions in which the economy is moving—and where it will likely settle when we are back to (a new) normal.  I don’t have a team of economic forecasters to direct, but I wish I did!