The Atlanta Fed's macroblog provides commentary and analysis on economic topics including monetary policy, macroeconomic developments, inflation, labor economics, and financial issues.
Comments are moderated and will not appear until the moderator has approved them.
Please submit appropriate comments. Inappropriate comments include content that is abusive, harassing, or threatening; obscene, vulgar, or profane; an attack of a personal nature; or overtly political.
In addition, no off-topic remarks or spam is permitted.
November 22, 2016
Outside Looking In: Why Has Labor Force Participation Increased?
The labor force participation rate (LFPR) is an estimate of the share of the population actively engaged in the labor market. The LFPR has increased about 30 basis points over the past year (from the third quarter of 2015 to the third quarter of 2016)—a modest reversal in the precipitous decline in the LFPR that began in 2008. What accounts for this stabilization and—given the demographic and cyclical forces in play—how much longer can it last?
The following is perspective through the lens of the reasons people give for not participating in the labor force. Perhaps the component most responsive to changes in labor market conditions is what I will refer to as the "shadow labor force," which is made up of people who are not in the official labor force and are not actively seeking employment, but who say they want a job. (This group includes people discouraged over job prospects.) During tough times, the share of the population in the shadows rises, and during good times it falls. In the third quarter of 2016, about 2.3 percent of the population fell into this category—down from a high of 2.8 percent but still a bit above prerecession levels (see the chart).
But focusing solely on the decline in the shadow labor force to explain the recent reversal in the LFPR would be a mistake. In fact, high unemployment in the aftermath of the Great Recession was accompanied not only by a rise in the share of the shadow labor force, but also by an increase in the share of the population who said they didn't currently want a job—because of either a health issue or engagement in some other activity. Although some of this likely reflects trends already at work before the recession, some of it was also probably a cyclical response to weak job opportunities.
The chart below shows how these various factors cumulatively contribute to the decline in the LFPR between the third quarter of 2007 and the third quarter of 2016. It shows that, in addition to a larger share in the shadow labor force, the reasons for the decline between 2007 and 2016 also stemmed from a greater age-adjusted share who were too sick or disabled to work (purple) or in school instead of working (light blue). Interestingly, the share out of the labor force but wanting a job (dark blue) actually exerted the smallest downward force on LFPR of all of these three reasons. The green section represents the impact of the baby boomers: an increasing share of the population of retirement age. Partly offsetting this shift in the age distribution was a decrease in the propensity of these workers to actually retire (orange).
The next chart shows that almost all the nonparticipation factors that had put downward pressure on the LFPR since 2007 have reversed course and contributed positively to an increase in the LFPR during the past year. In particular, there was a decline in the share of the population who cited nonparticipation because of poor health or enrollment in school or were otherwise wanting but not looking for work. This decline in the schooling and illness nonparticipation rates is particularly noteworthy because it stands in contrast to the increasing trends that were in place prior to the recession (to read more, visit our LFP Dynamics page).
The only significant factor continuing to depress the LFPR during the past year is the impact of an increasing share of the population in age groups with relatively low labor force attachment. This factor brings me to the second question I posed earlier in this post: What will this picture look like going forward? Unfortunately, I think the answer is that it's very hard to say.
Other things equal, it seems reasonable to think that the nonparticipation rates attributable to age-adjusted schooling and poor health will eventually revert to the upward trends occurring before the recession, a reversion that will push down the LFPR. Probably the biggest wild card for the future is what will happen with decisions concerning retirement (and hence older individuals' LFPR). The trend toward retiring later in life has risen and fallen a couple of times during the past two decades. The positive role that later retirement has played in mitigating the overall decline in the LFPR in recent years—coupled with the steadily increasing share of the population approaching traditional retirement age—suggests that deferred retirement will be an especially important factor to keep an eye on. This point is nicely illustrated in this piece by our colleagues at the Kansas City Fed, who look at the role of later retirement in reducing the rate of outflow of people from the labor force.
Note: Data on the reasons for not participating are available in the data download section of the Labor Force Dynamics page, which we will update every quarter (data for the third quarter are coming soon).
October 5, 2016
The Slump in Undocumented Immigration to the United States
Immigration is a challenging and often controversial topic. We have written some on the economic benefits and costs associated with the inflows of low-skilled (possibly undocumented) immigrant workers into the United States here and here. In this macroblog post, we discuss some interesting trends in undocumented immigration.
There are no official records of undocumented immigration flows into the United States. However, one crude proxy for this flow is the number of apprehensions at the U.S. border. As pointed out in Hanson (2006), the number of individuals arrested when attempting to cross the U.S.-Mexico border, provided by the Department of Homeland Security (DHS), is likely to be positively correlated with the flows of attempted illegal border crossings (see chart 1).
The apprehensions series displays spikes that coincide with well-known episodes of increased illegal immigration into the United States, such as after the financial crisis in Mexico in 1995 or during the U.S. housing boom in the early 2000s. Importantly, the series also shows a sharp decline in the flows of illegal immigration at the U.S.-Mexico border during the last recession, and those flows have remained at historically low levels since then.
A better proxy for illegal immigration from Mexico would adjust the number of apprehensions for the intensity of U.S. border enforcement (for example, the number of border patrol officers). The intuition is straightforward: for the same level of attempted illegal crossings, greater enforcement is likely to result in more apprehensions. Chart 2 shows the border patrol staffing levels as an indicator of enforcement intensity.
As the chart shows, the sharp decrease in apprehensions after the Great Recession occurred despite a remarkable increase in border enforcement, indicating that the decline in migration flows in recent years may have been even more abrupt than implied by the (unadjusted) border apprehensions shown in chart 1.
The measure of inflows shown in chart 1 is largely consistent with estimates of the stock of undocumented immigrants in the United States, such as those provided by a new study by the Pew Research Center based on data from the U.S. Census Bureau. After having peaked at 12.2 million in 2007, the stock of unauthorized immigrants fell during the Great Recession and remained stable afterwards, most recently at 11.1 million in 2014. Also, the composition of this stock has shifted since the Great Recession. Although the population of undocumented Mexican immigrants fell by more than one million from its 6.9 million peak in 2007, the number of undocumented immigrants from Asia, Central America, and sub-Saharan Africa remained relatively steady as of 2014 and even increased in some cases. For example, the population of unauthorized immigrants from India rose by about 130,000 between 2009 and 2014. However, a lot of this type of unauthorized immigration is a result of overstayed visas rather than from people crossing the border without a visa.
What do these numbers suggest about the future? It is likely that the flows of undocumented immigrant labor between Mexico and the United States reflect differences in demographic patterns and economic opportunities between the two economies. In the United States, the baby boom came to an abrupt halt in the 1960s, causing a notable slowdown in the native-born labor supply two decades later. In contrast, Mexico's fertility rate remained high for much longer, hovering at 6.7 births per woman in 1970 versus 2.5 in the United States (see chart 3).
Mexico's labor force expanded rapidly during the 1980s, which, juxtaposed with the Mexican economic slump of the early 1980s, unleashed a wave of Mexican migration to the United States (Hanson and McIntosh, 2010). Also encouraging this flow was the steady U.S. economic growth during the "Great Moderation" period from the mid-1980s up through 2007 (Bernanke, 2004). More recently, however, Mexico's fertility rate has fallen (as in some Central American economies), and economic growth there has mostly outpaced that of the United States. Therefore, it is perhaps not too surprising that demographic trends—along with greater enforcement—have caused the inflows of undocumented migration at the U.S.-Mexico border to slow in recent years. Shifts in demographic and economic factors across countries are likely to continue to influence undocumented immigration in the United States.
Note: The views expressed here are those of the authors and do not necessarily reflect the views of the Federal Reserve Banks of Atlanta or Boston.
July 15, 2016
How Will Employers Respond to New Overtime Regulations?
As of December 1, 2016, employers will face expanded coverage of overtime regulations. Most hourly workers are already, and will continue to be, eligible to receive overtime pay for work over 40 hours a week. However, under the new rules, most salaried workers making less than $47,476 ($22.83 per hour for a full-time, full-year worker) will be eligible for overtime pay. Currently, the maximum salary for qualifying for overtime pay is $23,660, or $11.38 per hour.
The Labor Department estimates that the new rule would currently apply to about 4.2 million salaried workers who earn above the old threshold but below the new one. But how many workers are actually affected by the new rule and what happens to the overall demand for labor will depend a lot on how employers respond.
At this stage, it's not clear just how employers will respond. But based on our conversations with local businesses, employers seem to be considering several options for workers whom the new rule would cover. These include:
- Keeping their salary the same but monitoring and paying for the overtime hours worked.
- Increasing their salary to just above the threshold to avoid paying overtime.
- Splitting the hours worked for the job across more people, possibly by hiring additional staff to work the overtime hours.
- Converting salaried employees to hourly and reducing their base hourly rate so that their total pay will remain the same as under their current salary.
- Curtailing certain business activities, such as networking and training activities, that might occur outside of the standard eight-hour day.
- Reducing staff levels elsewhere in the business and/or cutting other employee expenses to offset the increased cost of overtime.
The first two responses outlined above will result in additional employee costs. But trying to avoid these higher costs could itself prove to be expensive. For example, hiring additional workers to cover the overtime comes with fixed staffing costs—including state and federal unemployment insurance tax on new workers, and perhaps benefits—not to mention any hiring and training costs involved in recruiting new workers. In addition, there is a risk that splitting a job between multiple workers or hiring less experienced workers to cover the extra hours will reduce productivity. Also, staff morale could suffer as a result of any actions perceived as infringing on employee rights or status.
These new rules have many dimensions and potential implications (see, for example, here, here, and here for some discussion), and getting a handle on the effects is complicated by the fact that employer responses will likely differ across industries and possibly even across jobs within a firm. Hopefully, a somewhat clearer picture of the ramifications of the new overtime rule will begin to emerge as the time of implementation gets nearer.
July 15, 2016
How Good Is The Employment Trend? Decide for Yourself
The post-announcement commentary on last Friday's June employment report strikes us as about right: Not as spectacular as the 287,000 number the Bureau of Labor Statistics (BLS) reported for the month, but much better than the worst of our fears.
From the Wall Street Journal's wrap-up of economist reaction, here's Joseph Brusuelas:
The 147,000 three-month average is a fair representation what an economy at full employment looks like late in the U.S. business cycle. We anticipate that as the business cycle enters the final innings of the cyclical expansion that monthly job growth will slow towards 100,000, which represents the number necessary to stabilize the unemployment rate, which climbed to 4.9% in June due to an increase of 417,000 individuals that entered the workforce.
The consensus opinion is that observers should focus less on the monthly number and more on the three-month average, a vantage point we certainly endorse. We also think the reference point of the "number necessary to stabilize the unemployment rate" is the right way to decide whether a number like 147,000 net job gains is strong or not so strong.
The 100,000 unemployment-stabilizing job-gains statistic seems reasonable to us, but the average and median estimate from an April Wall Street Journal survey pegged the same statistic at 145,000. The three-month average job gain is comfortably above the former estimate but not the latter.
Where you stand on the number of job gains required to stabilize the unemployment rate is determined by your assumptions about the pace of civilian population growth (ages 16 and above), the labor force participation rate (LFPR), and the relationship between the payroll employment numbers and the comparable household survey statistic (from whence the unemployment rate is derived). Of course, you can always go to the Atlanta Fed's very own Jobs Calculator and input your assumptions yourself. But if you are like us, you may be more inclined to think in terms of a range of plausible numbers.
Here's our take on what some reasonable bounds on these assumptions might look like.
With respect to population growth, we assume a baseline growth rate equal to the same 1.0 percent annual rate that it has grown over the past year—after accounting for the artificially large population increase of 461,000 in January resulting from the BLS incorporating updated population estimates from the U.S. Census Bureau—with high and low growth alternatives of plus and minus one-tenth of a percentage point.
Second, our baseline for the LFPR is a decline of 0.226 percentage points per year, essentially the impact that we would attribute to age- and sex-related demographic changes over the past two years. Our low-side alternative assumption would be a larger 0.386 annual percentage decline in the LFPR, which adds in the average decline in the participation rate since February 2008 not due to demographic changes. Our high-side assumption is that the LFPR remains at its current level.
Finally, we note that the ratio of employment measured by the BLS payroll survey to employment measured by the household survey has been drifting up for several years. We have chosen a baseline assumption equal to the trend in this ratio since August 2005, and a high-side assumption chooses the steeper trajectory realized since February 2008. Since both August 2005 and February 2008, the unemployment rate has been unchanged, on balance.
The three scenarios for each assumption, in all combinations, yield 27 different implications for the number of payroll jobs required to maintain the unemployment rate at its current level (see the table).
These calculations generate a range of about 40,000 jobs per month to about 140,000 jobs per month. Our baseline assumptions suggest the unemployment rate would stabilize at payroll gains of about 80,000 per month, making the roughly 150,000 monthly average seen during the past quarter of a year look pretty good.
But we're not here to convince you of that today. You've got the numbers above. As we said at the outset, you can decide for yourself.
- Business Cycles
- Business Inflation Expectations
- Capital and Investment
- Capital Markets
- Data Releases
- Economic conditions
- Economic Growth and Development
- Exchange Rates and the Dollar
- Fed Funds Futures
- Federal Debt and Deficits
- Federal Reserve and Monetary Policy
- Financial System
- Fiscal Policy
- Health Care
- Inflation Expectations
- Interest Rates
- Labor Markets
- Latin AmericaSouth America
- Monetary Policy
- Money Markets
- Real Estate
- Saving Capital and Investment
- Small Business
- Social Security
- This That and the Other
- Trade Deficit
- Wage Growth