What's behind the surprising large drop in the unemployment rate?
- The dramatic fall in the unemployment rate of a full percentage point from December 2010 through March 2011 caught many analysts and policymakers by surprise.
- Employment growth coming out of a recession is normal, but the continued decline in the labor force is a puzzle. Typically as the economy rebounds, people are enticed into the labor market to take advantage of new job opportunities in the growing economy.
- Two explanations that have been offered for the declining labor force have been the availability of extensive federally funded Emergency Unemployment Compensation and an increase in the number of workers filing for disability insurance; neither explanation can account for the observed decline in the labor force.
- The bottom line: among the potential behavioral or policy-related explanations for the surprising drop in the unemployment rate of a full percentage point over a four-month period of time, the most appealing is that it was the result of a misalignment of usual forces playing on a labor market in recovery.
After a relatively flat period of little action, the unemployment rate dropped by a whopping (that's a technical term) amount of four-tenths of a percentage point from November 2010 to December 2010 and by an additional four-tenths of a percentage point from December 2010 to January 2011. After dropping yet another tenth of a percentage point each month from January to March, the unemployment rate stands at 8.8 percent.
This drop in the unemployment rate caught many by surprise. This article considers the role that persistent declines in the labor force played in this three-month nine-tenths of a percentage point fall in the unemployment rate.
A tale of three statistics: Employment, the labor force, and the unemployment rate
Two factors drive movement in the unemployment rate: the number of jobs created—which directly affects the number of people unemployed—and the number of people in the labor force —that is, the sum of those employed and unemployed, which are those people without a job but looking for one. Chart 1 plots the monthly change in employment and the monthly change in the labor force, along with the monthly change in the unemployment rate, from January 2010 through February 2011. Typically, whenever the change in employment exceeds (either increases by more or decreases by less) the change in the labor force, the unemployment rate falls.
Chart 1 has two primary features. First, the change in employment and change in labor force in December and January were not wildly out of sync with changes that occurred during all of 2010. (Note that these employment numbers are from the U.S. Bureau of Labor Statistics' Household Survey data.) Second is that, up until December and January, changes in employment and labor force were tracking each other but they diverged during those months, with employment increasing and the labor force decreasing. The net result was the dramatic decline in the unemployment rate by four-tenths of a percentage point in both months.
Persistent decline in the labor force
While we normally expect to see employment growth coming out of a recession, the continued decline in the labor force is a puzzle. Typically as the economy rebounds, people are enticed into the labor market to take advantage of new job opportunities in the growing economy.
Chart 2 highlights the uniqueness of the present persistent decline in the labor force. It plots the seasonally adjusted labor force, indexed to the trough of the recession for the past four recessions, starting six months before the trough and continuing through 20 months after the trough. For the most recent recession, this represents the period December 2008 through February 2010, with the trough at June 2009. The 2007–9 recovery has been unique in the continued decline in the labor force, even if we look at earlier recessions not shown here. Of course, this picture begs the question of what is behind this ongoing decline.
Long-term trends in the labor force participation rate (LFPR)
Most experts generally accept that we are facing a persistent decline in the labor force participation rate because of long-term demographic changes in the population: the baby boom generation is aging, so we would expect a lower aggregate rate of labor force participation.
Applying the consensus estimates of LFPRs1 to projected growth in the population (from U.S. Census Bureau estimates), chart 3 plots the total civilian labor force, actual and projected. The first thing noticeable in this chart is that, despite the expectation that the LFPR will not return to prerecession levels, this does not mean the labor force will continue to stagnate. The second conclusion from this chart is that the labor force stagnation during the recovery is not the result of declining LFPRs, as predicted by changing demographics.
Is extended unemployment insurance to blame for ongoing labor force declines?
During the most recent recession and recovery, a popular scapegoat of many labor market outcomes—such as very long spells of unemployment—has been the availability of extensive federally funded Emergency Unemployment Compensation (EUC). Theoretically, extended EUC could also be contributing to this ongoing decline in labor force participation. The theory goes like this: some individuals receiving extended EUC have not really wanted a job but go through the motions of looking for one in order to continue receiving the benefit. This situation has kept the number of unemployed elevated. As the "false" unemployed roll off EUC, they stop looking for work, as there's no reason to keep up pretenses any more. Since they are no longer looking for work, they are no longer counted in the labor force. Consequently, the labor force declines.
Although the eligibility period for receiving EUC was extended through January 3, 2012, with June 9, 2012, being the last date to collect the benefits, a significant number of individuals are reaching the maximum allowed benefit time period.2 It's possible that December and January saw a significant rise in the number of people coming off EUC who chose just to leave the labor force. If this were the case, we would naturally expect to see a significant flow of people moving from unemployment to out of the labor force in December and January.
Chart 4, which is from a report by the Council of Economic Advisors and based on data from the U.S. Department of Labor, illustrates the number of people that would be exhausting EUC under the extension. (This report was prepared in support of passing that extension.) The report estimates that with the extension, one million people would exhaust benefits in January 2011. This one million, out of a total estimated 14 million recipients as of October 2010, is as it turns out not a terribly significant number, suggesting this is not the likely culprit for the persistent declines in the labor force.
Chart 5 highlights another indicator that EUC exhaustion is not likely to be blamed for the labor force dynamics in December and January. This chart plots the monthly flow of individuals from unemployment to out of the labor force, indexed to the trough of each of the last three recessions. There is no clear deviation of these flows during the most recent recovery compared to the last two, and no dramatic increase in this particular flow during months 18 (December) and 19 (January) that would be consistent with the declines in the labor force seen during those two months.
Those not in the labor force are staying put
The one flow dynamic that does appear to have deviated during this recovery, particularly in recent months, is the number of individuals who are out of the labor force and staying out of the labor force (see chart 6). The increases in this flow since the recession trough mirror those seen after the 2001 recession but are higher and appear to have slightly accelerated at month 16 (October 2010). The higher growth in the number of people out of the labor force staying out of the labor force during this recovery is consistent with reports of increases in disability claims that Social Security officials blame on the combination of the deep recession and aging baby boomers.3
Chart 7 shows that since June 2008, when the earliest data are available, there have been two significant upticks in the share of people not in the labor force that are disabled. The first started in January 2009, the other in August 2010. However, the share has been on a fairly steady overall decline during the second half of the recession and the recovery to date, suggesting that increases in workers filing for disability benefits is not likely to be the driving force behind the stagnation of people out of the labor force.
It's also of interest to note that there doesn't appear to be any one age group that was driving the growth in nonparticipants in December 2010 or January 2011. Chart 8 plots, since January 2008, the monthly change in the number of people not in the labor force along with the share of that change contributed by each age group. Since the start of the recession, those aged 16–24 have consistently dominated the contribution to labor force nonparticipant growth, but not in January or February 2011.
Bottom line and looking ahead
Among the potential behavioral or policy-related explanations for the surprising drop in the unemployment rate of a full percentage point over a four-month period of time, the most appealing is that it was the result of a misalignment of usual forces playing on a labor market in recovery. As this recession and recovery have offered their fair share of unique events, this shouldn't be too alarming. Chart 1 appears to illustrate the entire story. Despite the relatively robust job growth, the decline in the labor force persisted through December and January.
Reasons offered for the continued growth in nonparticipation don't appear to hold water. The Emergency Unemployment Compensation program was extended through January 2012, greatly reducing the potential contribution of individuals rolling off EUC to the ranks of not-in-the-labor-force. Further, despite the sporadic upticks over the last couple of years in the share of those out of the labor force who are disabled, the share is on a consistent downward trend. There also doesn't appear to be any particular age group that contributed to the most recent increases in nonparticipants. Finally, the current stagnation in the growth in the labor force cannot be explained by longer-term trends in labor force participation rates.
As the recovery strengthens, more jobs are expected to be created, putting additional downward pressure on the unemployment rate. At the same time, a strengthening labor market provides even greater incentive for entrance into the labor market, which will serve as a counterforce against further declines in the unemployment rate. Looking again at chart 1, the uptick in the labor force in February and March counterbalanced the significant number of jobs created and arrested the dramatic decline in the unemployment rate seen the previous two months. The question remains whether the economy will continue to create enough jobs both to entice people back into the labor market and continue to reduce unemployment at the same time.
By Julie Hotchkiss, Atlanta Fed research economist and policy adviser
1 See Julie Hotchkiss, "Changes in the Aggregate Labor Force Participation Rate," Economic Review 4 (2009): 1–6, .
2 See U.S. Department of Labor, "Emergency Unemployment Compensation 2008 (EUC) Program," .
3 "Baby Boomers send disability benefit claims on the rise," USA Today, July 31, 2009, .