Understanding the Decline in US Labor Force Participation.

By Ben Griffy

The past 15 years have seen a striking decrease in the ratio of civilians employed to the size of the total working population. From 1948 (the earliest date at which statistics are available) to 2000, the labor force participation rate (LFPR) increased from 58.6 percent to 67.3 percent. Since, the LFPR has fallen to 62.4 percent, a rate last matched in the 1970s. us-lfpr.pngFurthermore, we have not seen a similar drop among Eurozone economies, creating some cause for concern that these changes are a sign of structural labor market weakness in the United States.LFPR.png

There are a number of factors that could be driving the current decline and the differences across countries. Demographic changes are certainly an important component, both in the increase in LFPR and the current decline. From 1948 to 2000, female labor force participation increased from 32 percent to 60.1 percent (see above), accounting for the increase in overall participation.  Male labor force participation declined during the same period: (see above). Another important factor is that the United States population has been aging during this time period, and different age groups exhibit different participation rates, as well as different numbers of hours worked at the intensive (i.e. number of hours worked at a single job) margin. This blog post by the St. Louis Fed suggests that if we control for the decline in “prime working age” adults, there has only been a small decline in the overall participation rates during the same time period. However, since 2010, there has still been a more than 2 percentage point decline in the LFPR. This may support the idea that labor markets have weakened in the United States.

Demographics alone do not explain why the labor market in the United States appears to be weakening. Europe has experienced an aging demography in larger magnitudes (see here) and have shown small, if any, declines in LFPR. A recent article in the Financial Times details the stark differences in labor force participation across countries, most notably the difference between the United States and Japan, an economy with a much older labor force. However, there are structural differences between the US and Europe that are worth addressing.

It is important to distinguish here between the intensive and extensive margins. Since 1980, the United States has seen only a small decline in average hours worked by an employed individual (see here), while there has been a large decline in the equivalent measures in both the UK and Japan. Here, it is important to note that the UK has a substantially larger share of individuals employed part-time, and the size of this share has accelerated relative to the United States. When we break employment down into part-time and full-time components, we see a different picture of the labor market. Namely, part-time employment has fallen in the US, while part-time employment has risen in comparable European economies.

This distinction is important when comparing the health of labor markets across countries. The United States has maintained strong full-time employment even following the Great Recession, while part-time continued a trend of longer-term decline. This translates into a very different story in terms of hours worked:hrs-wrked-fred.pngThe United States has had a small decline in hours worked over the past twenty years, while Europe has seen a drop from either similar levels to significantly lower levels than the United States, or previously higher levels than the United States to similar levels. The differences is that the United States has maintained higher levels of full-time employment, while Europe has averaged fewer hours at full-time work, and seen increases in part-time work.Full-Time.pngHours-Worked-Main.pngPart-Time.png

 

We see a similar story when we distinguish by age group. Economists are particularly concerned about the participation of “prime-aged male workers,” those from ages 25-54, when they could be expected to contributing most to the economy. However, there is little evidence of a decline in full-time employment among this age group:FT-by-age.png

We see that the United States has experienced little to no change in the percent of age 25-54 workers with full-time employment, outside of cyclical changes. Certainly, one cannot distinguish a long-term decline relative to European peers.

Europe and the United States are commonly juxtaposed to show that the US has suffered a weakened labor market. It is true that the United States has seen large declines in labor force participation when compared with the European Union. However, there are a number of factors at play that suggest the labor market has not weakened as substantially as the data on labor force participation might suggest. The United States population has aged during the same period, which accounts for some of the differences in LFPR. Additionally, the US appears to have undergone different structural changes than its European counterparts: most of the gains in participation rate in Europe have been through part-time jobs, while the United States has seen gains to both hours worked and full-time employment. Thus, while the participation rate suggests that the health of the labor market in the United States has declined relative to its European peers, other indicators make such a conclusion unclear.

 

Weak Employment on the Heels of weak GDP

by Thomas Cooley, Ben Griffy and Peter Rupert

Reading through today’s release of the employment situation from the BLS one is struck by how many times the phrase “little changed” shows up. Employment gains from the establishment survey were weak, up 151,000. Revisions over the past two months were almost off-setting, up 28,000 for November and down 30,000 for December, so, little change there.

empchgm-2016-02-05

Employment in mining and logging continues to decline, down 7,000 and down 29,000 over that past three months.
Manufacturing was up 29,000 and services up 118,000. However, temporary services declined 25,200. Health care and social assistance as well as leisure and hospitality rose by 44,000.

emp-mining-logging-2016-02-05

temps-2016-02-05

Average weekly hours were up to 34.6 but have bounced between 34.5 and 34.6 since March of 2014. Average hourly earnings rose from $25.27 to $25.39, up 2.5% over the year.

avghours-2016-02-05

The household survey also showed very little change in most headline numbers. The unemployment rate remained at 4.9% and the unemployment rate including marginally attached workers remained at 9.9%. Here are the numbers for the unemployment rate since September, with a lot of digits:

2015-09-01 0.05052050
2015-10-01 0.05028136
2015-11-01 0.05035331
2015-12-01 0.05007825
2016-01-01 0.04920580

uu6rate-2016-02-05

Given the amount of vacancies out there, however, the unemployment rate remains somewhat high, as can be seen in the Beveridge Curve.

beveridge-2016-02-05

There has been no significant change in the employment to population ratio or the labor force participation rate, those unemployed 27 weeks and longer, or those working part time for economic reasons.

epr-2016-02-05

lfp-2016-02-05

udur27-2016-02-05

parttimefrac-2016-02-05

Output per hour fell 3.0% at an annual rate in the fourth quarter of 2015 and has not shown any upward trend over the recovery. Indeed, it is only about 7% higher today compared to the peak before the great recession, by far the slowest rate of growth across all recessions and recoveries since the 1960’s, except for the 1973 recovery.

prodbar-2016-02-05

prod-cyc-2016-02-05

The BLS also undertook the annual revision to earlier data based on a more complete set of data. The annual revisions lowered December payrolls by 105k. The total nonfarm employment level for March 2015 was revised downward by
206,000 but this left a boost for job growth as the over-the-year change in total nonfarm employment for 2015 was revised from 2,650,000 to 2,735,000.

Overall, then, the slow GDP growth reported earlier this month along with the fairly weak labor report shows an economy that continues to grow at a fairly anemic pace. Moreover, it seems less likely the Fed will signal any moves in the near future.