Economic Indicators | Jobs and Unemployment

A Tale of Two Surveys: Today’s Jobs Report Paints a Murky Picture

The October jobs report released this morning by the Bureau of Labor Statistics (BLS) was one of those cases where the two surveys covered in the report—the establishment survey and the household survey—told very different stories. The establishment survey showed a relatively strong gain of 204,000 jobs in October. That, combined with an upward revision of 60,000 to prior months’ data, brought the average monthly growth rate over the last year to 194,000. There appears to be no discernible impact on the establishment numbers of the partial government shutdown in October; in the establishment survey, federal employees on furlough during the partial government shutdown were still considered employed.

The household survey, on the other hand, showed a 735,000 employment decline. The partial government shutdown likely contributed about 200,000 to that drop (see question 5 here), since federal employees on furlough during the partial government shutdown should have been counted as unemployed on temporary layoff in the household survey. But that means that removing the effect of the furlough, the household survey still showed an employment decline of more than 500,000.

The rule of thumb when the two surveys tell different stories is to put much more weight on what the establishment survey says, because it is a larger survey with much less month-to-month volatility. However, the weak household survey dampens the optimism about the relatively large employment gain in the establishment survey.

At a time like this, when the month-to-month signal is potentially murky, it’s useful to step back and take stock of the larger picture. The larger picture, however, is grim. For example, we need 8.0 million jobs to get back to the prerecession unemployment rate, and at the average rate of growth of the last 12 months, that won’t happen for another five years. Other trends include:

  1. In October, the labor force participation rate hit its lowest point in more than 35 years
  2. Long-term unemployment remains extremely high
  3. Unemployment is elevated across the board—across all categories of education, age, gender, race/ethnicity, and occupation.
  4. Wage growth is weak
  5. Because the labor market is weak, there is disproportionate job growth in low-wage industries

We now examine each of these in turn.

1. In October, the labor force participation rate hit its lowest point in more than 35 years

The labor force participation rate dropped 0.4 percentage points to 62.8 percent in October, its lowest point since March 1978. It should be noted that according to BLS, the drop in labor force participation in October was not due to the partial government shutdown. The labor force series are very volatile month-to-month, and drops of this magnitude are rare but not unheard of.

The decline in labor force participation has big implications for the unemployment rate. The unemployment rate has improved substantially from its peak exactly four years ago at 10 percent in October 2009 to 7.3 percent today. However, most of that improvement was not for good reasons; rather, it was due to the growth in the number of “missing workers”—people who have dropped out of, or never entered, the labor market because jobs opportunities are so weak. I estimate that there are currently roughly 6.1 million missing workers, and if these workers were in the labor force looking for work and therefore counted as unemployed, the unemployment rate would be 10.8 percent instead of 7.3 percent.

2. Long-term unemployment remains extremely high

The share of unemployed workers who have been unemployed for more than six months decreased in October from 36.9 percent to 36.1 percent. However, the long-term unemployed share is still far above normal (in 2007 the share averaged 17.5 percent; today we are at more than double that rate). Federal unemployment insurance benefit extensions are set to expire at the end of this year. See this report for more information on the economic effects of allowing these extensions to expire.

3. Unemployment is elevated across the board

The table below shows the current unemployment rate and the unemployment rate in 2007, along with the ratio of those two values, for various demographic and occupational categories. There is currently substantial variation in unemployment rates across groups, as is always the case (note, for example, that it was true in 2007, before the recession began). A key message from this table is that the unemployment rate is between 1.3 and 1.9 times as high now as it was six years ago for all groups. Today’s sustained high unemployment relative to 2007 across all age, education, occupation, gender, and racial and ethnic groups underscores that the jobs crisis stems from a broad-based lack of demand. In particular, unemployment is not high because workers lack adequate education or skills; rather, a lack of demand for goods and services makes it unnecessary for employers to significantly ramp up hiring.

Jobs Day

Unemployment rates of various demographic groups, 2007 and today

2007 October 2013 Ratio
All 4.6 7.3 1.6
Male 4.7 7.6 1.6
Female 4.5 6.9 1.5
White 4.1 6.3 1.5
Black 8.3 13.1 1.6
Hispanic 5.6 9.1 1.6
Age 16–24 10.5 15.1 1.4
Age 25–54 3.7 6.3 1.7
Age 55+ 3.1 5.4 1.7
Workers age 25 and older
Less than high school 7.1 10.9 1.5
High school 4.4 7.3 1.7
Some college 3.6 6.3 1.8
Bachelor’s and advanced degree 2.0 3.8 1.9
Workers under age 25, not enrolled in further schooling
High school degree 12.0 19.2* 1.6
Bachelor’s and advanced degree 5.4 7.1* 1.3
Management, professional, and related occupations 2.1 3.7* 1.8
Service occupations 5.9 8.8* 1.5
Sales and office occupations 4.3 7.3* 1.7
Construction and extraction occupations 7.6 12.9* 1.7
Installation, maintenance, and repair occupations 3.4 5.5* 1.6
Production, transportation, and material moving occupations 5.8 9.2* 1.6

* This is a 12-month average (November 2012–October 2013), since this series is not seasonally adjusted.

Source: Author's analysis of the Current Population Survey public data series

Copy the code below to embed this chart on your website.

4. Wage growth is weak

Average hourly wages for all private-sector workers increased by two cents in October. Over the last three months, average hourly wages grew at a 1.8 percent annualized rate, a substantial decline from the prerecession rate of wage growth. Average weekly wages, a measure that combines both hours and hourly wages, increased by 69 cents in October and also grew at an annualized rate of 1.8 percent over the last three months. Consumer spending is unlikely to rise unless wage growth picks up, but wage growth is unlikely to pick up without robust job growth. The economic link between high unemployment and low wage growth is straightforward; employers do not need to pay sizable wage increases to get and keep the workers they need when job opportunities are so weak that workers do not have other options.

5. Because the labor market is weak, there is disproportionate job growth in low-wage industries

Job growth in low-wage sectors was again dominant in October; this is to be expected when the labor market is as weak as it has been throughout this sluggish recovery. One reason for this is because, with nearly three unemployed workers for every job opening, there are many desperate job seekers who have no other choice but to accept low-wage jobs that they would not have to accept if job opportunities were strong. The two lowest-paid major industries are retail trade and leisure/hospitality (of which restaurants and bars make up the largest share). Retail trade added 44,000 jobs in October, somewhat higher that the average growth of the prior three months, 33,900. The increase could be due in part to seasonal issues, with holiday hiring starting earlier than expected (that is what happened in October of last year). Restaurants and bars increased by 29,300 jobs in October, somewhat higher than the average growth rate of the prior three months, 25,600. Health care saw an increase of 15,000 in October, in line with the average growth rate of 14,300 over the prior three months.

Some good news was in manufacturing, a relatively high-paid sector, which added 19,000 jobs in October, more than the total number of manufacturing jobs added in the first nine months of the year (16,000). Construction—another relatively high-paid sector—was up 11,000, in line with its average of the first nine months of the year, 12,400.

With research assistance from William Kimball and Hilary Wething

See related work on Jobs and Unemployment

See more work by Heidi Shierholz