Thursday, June 11, 2015

What Lessons from Part-Timers about the US Labor Market?

The US unemployment rate has been in the range of 5.4-5.5% since February, which is clearly a vast improvement from its peak of 10% in October 2009. But of course, no single number will capture the health of the labor market. For example, the overall unemployment rate doesn't capture the rise in long-run unemployment (discussed here, here, and here). It doesn't capture concerns over whether are dropping out of the official labor force, and thus are no longer counted as unemployed (discussed here and here). It doesn't look at the lack of widespread wage growth. And of course, the unemployment rate doesn't look at part-time workers--because after all, it counts them as being employed.

However, some workers choose part-time status, while others have part-time status thrust upon them. The difference matters. Those who are working part-time, but would prefer to be working more hours, can be considered both as part-time workers, but also as part-time unemployed. Fortunately, the survey on which the unemployment rate is based asks questions about this distinction: that is, those who report working part-time are asked why. If they give a reason for working part-time that suggests they would prefer to work more hours but such jobs aren't available in their local economy, they are classified as part-time "for economic reasons." If they give a reason for working part-time that is based more on their personal circumstances--for example, family or personal obligations, being in school, being partially retired, and so on--then they are classified as part-time "for noneconomic reasons."

Here's the breakdown of US part-time workers in a figure created with the ever-useful FRED website run by the Federal Reserve Bank of St. Louis. The blue line on top shows the share of the civilian labor force working part-time for any reason, rising from about 14% back around 1970 to about 18% in more recent years. The green line in the middle shows the share of workers who are part-time for noneconomic reasons. That level doesn't show much trend: it was 12% of the civilian labor force for much of the 1970s and 1980s, up to about 13% in the 1990s, and then back down to 12% more recently. The red line on the bottom shows the share of workers who are part-time for economic reasons. In good years for the economy, this rate seems to be about 2% of the workforce. In bad recessions, like 1982 or 2009, it rises to about 6% of the workforce.



This share of part-time workers for economic reasons has declined since the end of the Great Recession, but it's still at about 4% of the workforce, a couple of percentage points above the level when the labor market is at its most robust. These workers do at least have some connection to the labor market, unlike the 5.5% of the labor force that is unemployed. But being part of the part-time labor force when you wold like to work more poses own set of challenges.

But there's one more additional distinction here. Is the higher level of part-time work for economic reasons because the broad labor market has not yet recovered? Or is it to some extent because employers have a greater preference for part-time workers, for reasons that have nothing to do with the sluggish recovery from the Great Recession? Rob Valetta and Catherine Van Der List tackle this issue in "Involuntary Part-Time Work: Here to Stay?" written as the "Economic Letter" for the Federal Reserve Bank of New York (June 8, 2015). They do a cross-state comparison over time, looking at factors that might affect the desire of employers to hire part-time labor: for example, wage levels and minimum wage levels, industry mix, and share of younger workers. Thus, they can capture to some extent whether part-time work for economic reasons is correlated with the unemployment rate, or with these structural economic factors. They write:
"From the base level of just under 3% in 2006, cyclical factors raised the rate of involuntary part-time work by slightly more than 2 percentage points at the peak in 2010, while structural factors raised it by a little over 1 percentage point. The cyclical component declined after 2010 and is likely to have continued falling beyond our sample period, while the structural component was relatively stable from 2009 through 2013."
For a final sense of perspective, here's one more figure generated from FRED website. It adds up three categories. One is the overall unemployment rate. A second category is those employed part-time for economic reasons, as discussed above. The third category is called the "marginally attached." It refers to those who to "persons who want a job, have searched for work during the prior 12 months, and were available to take a job during the reference week, but had not looked for work in the past 4 weeks." Because these people had not looked for work in the previous four weeks, they are counted as "out of the labor force," not as "unemployed," but their survey responses suggest that they would prefer to have work. The blue line shows the percentage of the labor force in these three categories combined. The red line shows the unemployment rate by itself. (The data on "marginally attached" is is only available back to 1994.)




The overall unemployment rate at 5.5% is fairly close to the sub-5% rates where it bottomed out in the previous recoveries. HAt the end of the Great Recession in 2009, the gap had soared to seven percentage points. The gap has now fallen back to about 5 percentage points, but it is still well above the gap of about three percentage points that was common from the mid-1990s through the 2001 recession and up to the start of the Great Recession. In this way, the decline in the official unemployment rate isn't  capturing some degrees of continuing weakness in the labor market.