At next week’s FOMC meeting, the state of the labor market will play a key role in policy deliberations. But there’s a lot more going on underneath top line unemployment numbers that make them a bad tool for monetary policy decision-making.
The May employment report is a conundrum. Employment growth and the unemployment rate sent opposing signals about labor market conditions — much like they have been doing throughout the recovery. The economy added 138,000 jobs last month, with the three-month average only at 121,000 jobs, suggesting labor market weakness.
By contrast, the unemployment rate fell to 4.3 percent — the lowest reading in 16 years. Additionally, job openings are near an all-time high. And voluntary quit rates are up. These data all suggest tight labor market conditions.
The weak employment growth is consistent with the sub-par economic growth we have experienced since the recession. But deep recessions, like the one we just experienced, are normally followed by a stronger than average recovery, not a weak one. There has been no calendar year during the recovery in which real GDP grew at three percent — a desultory performance.
The week recovery has managed a fairly steady, if very gradual, fall in unemployment, bringing it to a 16-year low. As a result, many observers declare we are at full employment. But unemployment is so low because of the length of the recovery, not overall economic strength. We have had a long-lived, weak expansion — with unemployment statistics themselves masking weaknesses in the labor market.
“Unemployment” is a term of art and does not mean simply the number of people not working. It comprises the number of people not working and who are looking for a job. The unemployment rate is the number of people unemployed (in the technical sense) divided by the labor force. If people cease looking for a job, they are removed from both the unemployment measure and the labor force. They are subtracted from both the numerator and denominator; the numerator is always a smaller number. That causes the unemployment rate to fall, even if no one is getting a new job. The unemployment rate is therefore not a clear indicator of the state of the labor markets.
There is a cyclical pattern of people leaving the labor force in recessions because they are “discouraged workers” and give up looking for work. When economic recovery begins, many discouraged workers return to the labor force and secure employment. In this recovery, however, the pattern has not held up as usual.
Because of the decline in labor force participation a significant portion of the drop in the unemployment rate reflects not strength but something entirely different: a decline in the willingness to work.
Numerous causes have been offered up for the decline, with demographics frequently mentioned. The labor force includes those aged 16-64. So, with more people staying in school longer and baby boomers retiring, the labor force participation rate declines for these demographic reasons.
Demographics have much less explanatory power for people aged 25-54, the prime-aged workforce. Movement out of the labor force has been concentrated among men, with nearly 7 million from the prime-aged cohort. Nicholas Eberstadt describes them as the “vast army of jobless men who are no longer even looking for work.” Fifty years ago, the percentage of men in the prime-age category not looking for work was 6 percent. In 2015, it was two and half times larger, at 15 percent.
Millions more would be employed today if the labor-force participation rate were as high as it was in 2000. This undercuts the claim that we are at the cusp of full employment.
What is driving this?
Economic policy certainly influences work decisions. With a substantial growth in those claiming disability since the recession, Social Security Disability Insurance (SSDI) may be viewed as an alternative to rejoining the work force. Early retirement under Social Security at age 62 is also a factor. In this last recession, jobless benefits were extended to as long as 99 weeks. Coming off nearly two years of not working, lost skills and the need to retool to seek new employment make it much harder to reenter the job market.
Charles Murray emphasizes cultural and value changes in Coming Apart. It is a story of cultural and economic decline in which blue collar, white men have lost the work ethic. A more popular rendition of the story appears in J. D. Vance’s Hillbilly Elegy.
My purpose here is not to assign weights to the various explanations for men dropping out of the labor force. My goal, as stated from the outset, is to gauge the state of the labor market. What I hope to have shown is that the unemployment rate, whatever its merits at one time, is now a misleading measure. It can no longer reliably indicate what is going on in the labor market. It certainly should not be used by the Federal Reserve as either a target or indicator of monetary policy. Monetary policy is impotent in the face of declining willingness to work.
There are profound consequences to declining male participation in the work force. Economically, it is a drag on employment growth and GDP growth. The social costs are even higher. Men without work are men without dignity. That leads to substance abuse and other ills
There are certainly policy steps that can be taken to address these issues (for example, addressing SSDI benefits). Monetary policy is not one of them.
Ultimately, weak employment growth is the consequence, not of weak demand as some suppose, but tight supply — caused largely by a declining willingness to work among men. We are a long way from full employment in any meaningful sense. But we may be at the fullest employment possible given underlying social trends and existing economic entitlements.
Whatever the FOMC decides at its meeting next week, it should not be influenced by a number — the unemployment rate — whose movements no longer provide clear signals of labor market conditions. That conclusion follows regardless of one’s views on the dual mandate for the Federal Reserve.
Commentary by Gerald P. O’Driscoll. Originally published at Cato @ Liberty. This work by Cato Institute is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License.