Yesterday Federal Reserve Vice Chairman Stanley Fischer gave a speech entitled The Great Recession: Moving Ahead. A key topic is the question of long-term structural changes to the economy — whether we're experiencing economic weakness with deeper roots than the cyclical effect of the last recession.
Despite Mr. Fischer's ambivalence toward, as he put it, "the relative importance of cyclical (short-term) versus structural (long-term) factors", I believe there is profound evidence that the U.S. workforce has undergone structural changes more fundamental than the cyclical impact of a recession — even that of the Great Recession.
The Unemployment Rate: Between 2.8 and 3.7 Million Jobs Shy
The headline unemployment rate is one of the most watched economic numbers. It is a calculation of the percentage of the Civilian Labor Force, age 16 and older, currently unemployed. The first chart below illustrates this monthly data point since 1990. Today's Civilian Employed would require 2.8 million additional job holders to match its interim low in 2007, and we would need 3.7 million to match the lowest rate in 2000.
For a somewhat more optimistic spin, let's look at the same statistic for the core workforce, ages 25-54. This cohort leaves out the employment volatility of the college years, the lower employment of the retirement years and also the decade when many in the workforce begin transitioning to retirement. Today's age 25-54 labor force would require 1.8 million additional employed to match its interim low in 2006 and 2.3 million to match the lowest rate in 2000.
Labor Force Participation Rate: A More Sobering Measure
A wildcard in the two snapshots above is the volatility of the Civilian Labor Force — most notably the subset of people who move in and out of the workforce for various reasons, not least of which is discouragement during business cycle downturns. The chart below continues to focus on our 25-54 core cohort with a broader measure: The Labor Force Participation Rate (LFPR). The LFPR is calculated as the Civilian Labor Force divided by the Civilian Noninstitutional Population (i.e., not in the military or institutionalized). Because of the extreme volatility of the metric, I've included a 12-month moving average.
Based on the moving average, today's age 25-54 cohort would require 2.8 million additional people in the labor force to match its interim peak participation rate in 2008 and 4.1 million to match the peak rate around the turn of the century. Why are so many more labor force participants needed for a complete LFPR recovery? When the economy is going gangbusters, as in the late 1990s, jobs are abundant, which encourages the population on the workforce sidelines to join the ranks of the employed. Today's economy doesn't offer that sort of encouragement.
Employment-to-Population Ratio: Even More Sobering
The final chart below, again focused on our ages 25-54 cohort, is calculated as the Civilian Employed divided by the Civilian Noninstitutional Population. Again I've included a 12-month moving average. A significant feature of the Employment-to-Population Ratio is that isn't affected by the volatility of labor force participants who, for various reasons, are unemployed.
First the good news: This metric began to rebound from its post-recession trough in late 2012. However, the more disturbing news is that the current age 25-54 cohort would require an increase of 4.7 million employed participants to match its ratio peak in 2007. To match its mid-2000 peak would require a 6.6 million increase.
A Structural Change in the U.S. Economy
The charts above offer strong evidence that our economy is in the midst of a structural change. The three mainstream employment statistics — unemployment, labor force participation and employment-to-population — all document an ongoing economic weakness far deeper than the result of a business cycle downturn.
In his speech on the aftermath of the Great Recession, Federal Reserve Vice Chairman Fischer had some specific observations on employment [bolding added by me]:
The considerable slowdown in the growth rate of labor supply observed over the past decade is a source of concern for the prospects of U.S. output growth. There has been a steady decrease in the labor force participation rate since 2000. Although this reduction in labor supply largely reflects demographic factors — such as the aging of the population — participation has fallen more than many observers expected and the interpretation of these movements remains subject to considerable uncertainty. For instance, there are good reasons to believe that some of the surprising weakness in labor force participation reflects still poor cyclical conditions. Many of those who dropped out of the labor force may be discouraged workers. Further strengthening of the economy will likely pull some of these workers back into the labor market, although skills and networks may have depreciated some over the past years.
In order to discount Fischer's point about the aging population, I have focused on the 25-54 age group. Also, by excluding the age 55-64 decade associated with early or pre-retirement, I've eliminated a cohort that might include a major source of discouraged or less-determined workers.
The Growth of the Elderly Workforce and Its Causes
I'll close this analysis with a chart that essentially demolishes Fischer's view of our aging population as a demographic drag on labor supply. Here is the ratio of the 65-and-over cohort as a percent of the employed civilian population all the way back to 1948, the earliest year of BLS employment data. Mind you — these people are not only in the workforce, but also actually employed.
The percentage of elderly employment is at its historic high — now double its low in the mid-1980s. This is a trend with multiple root causes, most notably longer lifespans, the decline in private sector pensions and frequent cases of insufficient financial planning. Another major cause, I would argue, is the often surprising discovery by many of the elderly that the "golden years of retirement" might be less personally satisfying than productive employment. Note that the growth acceleration began in the late 1990s, prior to the last two business cycle downturns (aka "recessions").
In Conclusion...
I would dismiss Vice Chairman Fischer's reference to "considerable uncertainty" in the interpretation of labor supply weakness as routine Fedspeak. We are indeed experiencing a structural change in employment, one that is a major drag on the overall economy. The fact this this change was (not surprisingly) exacerbated by a business cycle downturn should not blind us to its structural foundation. While this change won't be permanent, it will be a burden on economic growth for many years to come.
Note from dshort: For some related analysis, see the following periodic updates:
- Structural Trends in Employment by Age Group
- Demographic Trends in the 50-and-Older Work Force
- Multiple Jobholders as a Percent of the Employed: Two Decades of Trends
