Sunday, August 23, 2015

Does Lower Unemployment Imply a Stronger Economy?

With all the buzz about a $15 minimum wage, or a $10.10 minimum wage, there's been a lot of discussion about the effects of minimum wage on the labor market. While some of the empirical work on the subject says the minimum wage doesn't affect employment, most of it says otherwise.

Another popular topic these days is Federal Reserve policy; specifically how and when the Fed will raise interest rates. A major concern is that raising rates "too soon" will cause unemployment to stop falling or start rising again. The Fed has cited improvements in labor markets as a sign that it could start raising interest rates soon. Looking only at the unemployment rate, the idea that labor markets are improving makes sense.

In this post I'll discuss unemployment and labor force participation rate data since the recession, then give a numerical example that shows that even if the unemployment rate is falling, the labor market and the economy overall may not be improving as much as we'd like to think.

Just looking at this one chart, we might get the impression that the labor market is improving. I think, however, it's misleading. We need to look at the labor force participation rate as well.

Here's a chart of the labor force participation rate in the U.S. from 2005 to 2015. The labor force participation rate basically tells us how many people are either working or looking for work as a percentage of everyone who, theoretically, can work. I'll make this clearer later when we get to my numerical example.

What you should notice here is that the percentage of people who are either working or looking for work has fallen from about 66.25% before the recession to 62.6% in July of this year. The drop has been more extreme in some groups than others (16-20 year olds, 20-24 year olds, and 24-54 year olds). Since these younger age groups have experienced serious declines in labor force participation, we know that the overall decline isn't just the Baby Boomers retiring.

So is the labor market really stronger? We have declining unemployment, but labor force participation has also declined. What do these statistics tell us about what's actually happening?

Below are some tables with hypothetical employment statistics. Table 1 is the baseline scenario and Table 2 tells us what's happening behind the scenes when both unemployment and labor force participation are falling.

First, some definitions. In the baseline scenario there are 500 people employed (Line A). That just means they have a job. There are 25 people who are unemployed (Line B). Those people are not working but are actively seeking work. There are 225 people who theoretically can work (that is, they aren't "institutionalized") but aren't working or actively seeking work (Line C). The labor force participation rate shows the percentage of people working or unemployed and seeking work as a share of the total labor force (A+B+C). Here's a more thorough description with a couple of detailed examples. Finally, the unemployment rate is the percentage of people unemployed as a share of those who are employed and unemployed and looking for work (B divided by A+B).

Now, look what happens when 5 unemployed people in group B stop looking for work and move to group C:

Look at that! The unemployment rate fell. Good news, right? Maybe. Those people might have stopped looking for work because things got rosier for them. However, given that this recovery has dragged on much longer than other recent recoveries, I'd wager most of those 5 people were just giving up.

This story also explains the drop in the labor force participation rate. Since some people who were in the labor force (unemployed and looking for work) left, labor force participation fell also. I think this scenario explains what we're seeing today quite well. The decline in the unemployment rate might make some people feel better about conditions in the economy, but I think they're mistaken. The adjustments needed to restore a productive economy haven't been allowed to take place.

Of course, there are other explanations. An increase in population could increase C and not affect A or B, meaning that the labor force participation rate would fall independently of the unemployment rate. Since that basically means people are growing up and are unable to find jobs, I don't  think it's much better than the scenario I discussed above.

So next time you're told that the declining unemployment rate means a strengthening economy, you can give a more sophisticated explanation that takes other factors into account.


  1. The other thing concerning is that Yellen seems to be adhering to a naive Phillips Curve view and is nervous about inflation because she thinks we're near the NAIRU. Even if U3 were an accurate measure of labor surplus, a stronger economy-faster real growth-puts downward pressure on prices, not the reverse. Besides which, even if tighter labor markets did drive up prices, shouldn't we be seeing accelerating inflation if that was true? But we really aren't, just the opposite in fact.

    1. Anonymous,

      Good points.

      I think the inflation picture is really a function of the money multiplier. If you look at M2/Monetary Base, it crashed during the recession when the Fed really cranked up its balance sheet. The last time the multiplier crashed that hard was in the 1930s. We didn't see price inflation until the 1960s and 1970s when the multiplier had recovered (i.e. M2 had grown and caught up with the base). I don't think we'll see significant price inflation until M2 is able to "catch up" with the base this time. Of course, we've definitely seen prices skyrocket in asset markets thanks to the low rates over the last 7 years.