The Unemployment Rate Doesn't Measure Labor Market Strength

Policy makers are desperately searching for the best methods of measuring the health and strength of the labor market because different data points, and even the same ones from month to month are flashing confusing signals at the moment. With the Federal Reserve preparing to raise interest rates as soon as they believe the labor market is strong enough, deciding on how to determine that state is important. One fact everyone should be able to agree on is that the official unemployment rate does not even attempt to measure the strength or health of the labor market. For example, it is open knowledge that Fed Chair Janet Yellen looks at a "dashboard" of at least nine labor market indicators.

The official unemployment rate (technically called U3) simply divides the number of people who are not working, want to work, and have been actively applying for jobs (defined as having applied to at least two different employers within the last month) by the sum of the people working and those defined as unemployed. Thus, lots of people who are unemployed by many reasonable definitions do not count as such in the official government statistic. Using the government's own definition, workers who are discouraged or marginally attached to the labor market do not count in the official unemployment rate. There are different, broader, unemployment measures available, but they do not get the headlines.

In particular, the unemployment rate is now back to a "normal" level of 5.5 percent. If that means the labor market is healthy and strong then wages should start rising. If the unemployment rate has little do with the state of the labor market, but rather we need to focus on the total potential supply of more workers including both the officially unemployed plus those currently not in the labor force (government acronym, NILF, meaning all people over age 16 who are not working or trying to gain work), then the labor market might still be quite weak with little prospect for higher wages.

A recent labor market trend is a rapidly rising number of people not in the labor force. The over 90 million Americans 16 years old or older that are not working fall into several categories: retirees, stay-at-home parents, students, and those who would prefer working but have given up on finding a job. Because government statistics do a very poor job of providing insight into how many of those not in the labor force are in each category, policy makers have been reduced to making educated guesses about the relative size of each subgroup of those not working and their potential to reenter the labor market as conditions improve.

If the unemployment rate is a poor measure of labor market strength, what is a better measure? Measures such as labor market churn, job openings, job leavings, and both gross employment and employment gain numbers are all much better measures of the health of the labor market.

Churn in the labor market focuses on the fact that while the net gain in jobs may be averaging around 200,000 per month, the fuller picture is that between 4.5 and 5 million people leave jobs and start new jobs every month. Data on the job openings, hires, and separations that make up labor market churn are all in a government report known as JOLTS. The actual changes in employment each month are a huge multiple of the net change in employment. Generally speaking, more churn signals a healthier labor market since it gives a much truer measure of the number of available jobs at any point in time. While churn in the labor market is still large compared to the net changes in jobs, churn today is lower than it used to be, having still not returned to pre-recession levels.

The actual employment numbers speak directly to the strength of the labor market. If demand for labor is increasing, more workers will be working. It really is pretty much that simple. Currently, six years from the end of the recession, total U.S. employment is 148.3 million compared to a pre-recession peak level of 146.6 million. Meanwhile the population of working age people has risen by 17.5 million, far more than the 1.7 million person gain in the number of people employed. A better economy should be employing more people, but when we look at changes in employment rather than changes in unemployment, we again see much less evidence of strength in the labor market.

The labor force participation rate measures the percentage of all eligible people who are either working or officially unemployed (meaning trying to work). Changes in the labor force participation rate, especially for younger workers, provide a consumer perception of labor market strength, albeit confounded with issues of social (and familial) safety net generosity. If the labor market is strong, good high-paying jobs will be available and people will choose to join the labor force, raising the labor force participation rate. The labor force participation rate has been on a twenty year slide, partially due to the rising share of retirees in our population; however, since the start of the last recession the labor force participation rate has dropped significantly and not recovered. The low participation rate is a clear sign of weakness in the labor market because a truly strong labor market would be attracting more people to join in. This link, shows the labor force participation rate for those ages 25-54, and shows a 2 percentage point drop since the recession.

Part-time workers, especially those working part-time for economic reasons tell you something important about the labor market. The government provides data on part-time workers in two categories: those doing so voluntarily and those who want full-time jobs but can only get part-time hours because businesses don't have enough demand for their products to make their workers full-time. When the number of part-time workers for economic reasons is rising, the labor market must be weak. If the number of part-time workers for economic reasons is falling, that is a sign of growing strength in the labor market. Part-time workers for economic reasons are falling, but are still 2 million above the level at the start of the recession.

Unemployment is really a measure of labor market disequilibrium; it measures the mismatch between employers' demand for labor of various types and workers' willingness and ability to supply that labor. Unemployment that is "too high" or "too low" in aggregate or in specific job categories is really about these mismatches, not the health of the labor market.

If 10 million people reentered the labor market and 9.5 million of them found jobs, the unemployment rate would be unchanged but the labor market should be characterized as much healthier (and the economy would be much stronger). Similarly, if the same 10 million people reentered the labor market (by applying for two jobs each) and none of them found jobs, the unemployment rate would rise to over 11 percent, yet total employment would be unchanged. The labor market would not be weaker; there would just be more people officially unemployed.

Policy makers should raise interest rates because the artificially low rates they have created are distorting capital allocation, punishing savers, and rewarding debtors (not incidentally including the federal government). Raising interest rates at a reasonable pace is unlikely to affect the labor market significantly. However, if policy makers wish to base their interest rate decisions on the health of the labor market, they must look well beyond the unemployment rate. The more carefully one looks at alternative measures, the weaker the labor market looks. This fact likely explains why the financial markets seem rather convinced that few, if any, interest rate hikes loom on the horizon.