Job market continued to improve in 2017 as the unemployment rate declined to a 17-year low

In 2017, the U.S. economy completed its eighth year of expansion following the Great Recession, and most labor market measures continued to improve as well.¹ By the end of the year, the unemployment rate had hit a 17-year low.

This article highlights several important developments in the U.S. labor market in 2017 and explores how various demographic groups fared in the job market, with breakdowns by race and ethnicity, age, and gender. Additionally, we examine recent developments in employment and unemployment for veterans, the foreign born, and people with disabilities, as well as how hurricanes in the Gulf region affected the job market. The data are from the Current Population Survey (CPS).²

The employment situation improved in 2017

The number of unemployed people fell by 989,000 from a year earlier, to 6.6 million in the fourth quarter of 2017.³ The unemployment rate—the percentage of the labor force that is unemployed—was down by 0.6 percentage point over the year, reaching 4.1 percent at the close of 2017. (See table 1 and figure 1.) This is the lowest unemployment rate since the fourth quarter of 2000. The unemployment rate decline in 2017 was twice as large as the 2016 decline.

The number of employed people increased by 1.8 million over the year, to 153.9 million in the fourth quarter of 2017. This growth was smaller than the over-the-year increase of 2.5 million in 2016. As a result of the recent growth, the employment–population ratio continued to rise, reaching 60.1 percent by the fourth quarter of 2017. (See table 1 and figure 2.) The employment–population ratio has gradually trended up since 2014. In the third quarter of 2017, it reached its highest point (60.2 percent) since the recession ended. However, it remains well below levels that predated the Great Recession.

In 2017, the labor force participation rate, at 62.7 percent, was unchanged from a year earlier. The participation rate has remained around this level for the past 4 years. (See table 1 and figure 1.) In the third quarter of 2007—just before the recession—the participation rate was 65.9 percent.

Employment grew and unemployment fell for Whites, Blacks, and Hispanics; labor force participation changed little for all groups

Similar to employment growth in the overall labor market, employment growth among Whites raised the employment–population ratio for this group to 60.4 percent by the fourth quarter of 2017. A decline in the ranks of the unemployed also shrank the unemployment rate for Whites by 0.7 percentage point over the year, to 3.6 percent, a 17-year low. After declining from 2009 to 2014, the labor force participation rate for Whites changed little over the last 3 years; it was 62.7 percent at the end of 2017. (See table 1 and figure 3.)

Continuing the trend of strong employment gains since 2014, the employment–population ratio for Blacks rose by 0.9 percentage point over the year. The ratio in the fourth quarter 2017—57.8 percent—was the highest since the second quarter of 2008. The unemployment rate for Blacks declined by 1.0 percentage point over the year, to 7.1 percent in the fourth quarter of 2017, which marks a series low (comparable data became available in 1972). Despite this progress, the unemployment rate for Blacks remained about twice that of Whites.⁴ After trending up over the previous 3 years, the labor force participation rate for Blacks changed little in 2017, ending the year at 62.2 percent.

For Asians, most labor market indicators in 2017 were little different from those in 2016. The employment–population ratio, at 61.4 percent in the fourth quarter, changed little in 2017 after trending up in 2016. At 2.8 percent in the fourth quarter of 2017, the unemployment rate for Asians was little different from a year earlier. The unemployment rate for this group returned to its prerecession low in 2017.⁵ The labor force participation rate for Asians, at 63.2 percent, was unchanged from a year earlier.

People of Hispanic or Latino ethnicity also experienced employment growth and declining unemployment in 2017.⁶ Beginning in 2011, the employment–population ratio for Hispanics gradually trended up through the end of 2015. However, at 62.5 percent in the fourth quarter of 2017, this ratio changed little over the past 2 years. The unemployment rate for this group, at 4.8 percent in the fourth quarter of 2017, declined to a series low (comparable data became available in 1973). The labor force participation rate for Hispanics was relatively unchanged, at 65.6 percent in the fourth quarter of 2017, and has shown no clear trend since 2013.

Employment rose and unemployment declined for nearly all age groups in 2017, and labor force participation rates generally showed little change

Among people of prime working age—defined as 25 to 54 years old—employment increased in 2017, and their employment–population ratio continued to climb, reaching 79.0 percent in the fourth quarter. Both prime-working-age women and prime-­­­­­­working-age men enjoyed employment gains during the year. The employment–population ratio for prime-working-age women increased by 0.9 percentage point over the year, to 72.4 percent in the fourth quarter of 2017, the same level as when the recession began in the fourth quarter of 2007. The ratio for prime-working-age men, at 85.8 percent in the fourth quarter of 2017, was up by 0.8 percentage point over the year but was still 1.3 percentage points lower than it was at the dawn of the recession. (See table 2.)

The unemployment rate for people of prime working age continued to decline in 2017, reaching 3.4 percent at the end of the year. Unemployment rates were down for both prime-working-age men and prime-working-age women over the year. At the close of 2017, the unemployment rate for people in this age group was about in line with its prerecession low (3.5 percent in the fourth quarter of 2006).

In 2017, labor force participation among people of prime working age edged up for women but not for men. After a prolonged upward trend, which saw the labor force participation rate for women peak at 77.1 percent in the third quarter of 1997, the rate for women began a modest decline, hitting a recent low of 73.4 percent in the third quarter of 2015. However, the labor force participation rate for women in this age group reached 75.0 percent in the fourth quarter of 2017. The labor force participation rate for men in this age group had generally trended down for the past five decades, but at 88.7 percent in the fourth quarter of 2017, has shown no clear trend in the past 2 years.

Among people age 55 and older, the employment–population ratio increased steadily for both men and women from the mid-1990s through the beginning of the Great Recession. Since the end of the recession, however, this metric took different trajectories for men and women. After declining in 2009, the employment–population ratio for older men resumed its upward trend, albeit at a slower pace than before the recession. This ratio for older men, at 44.5 percent at the close of 2017, was 1.9 percentage points higher than at the close of 2009. The employment–population ratio for older women showed relatively little movement following the recession, after gaining a percentage point or more each year during most of the early 2000s. This ratio for older women settled at 33.5 percent in the fourth quarter of 2017, which was only 1.0 percentage point higher than the ratio in the fourth quarter of 2009.

The unemployment rate for workers age 55 and older declined to 3.2 percent in 2017, following little change in the previous year. The unemployment rate for men edged down by 0.4 percentage point over the year, to 3.3 percent in the fourth quarter of 2017. The unemployment rate for women, at 3.0 percent, was down by 0.5 percentage point from the previous year. By the end of 2017, the unemployment rate for older men was still 0.4 percentage point above its prerecession low of 2.9 percent in the third quarter of 2006. The rate for older women was about in line with its prerecession low of 2.9 percent in the third quarter of 2007.

As shown in the labor force participation rate table below, the rate for people 55 years and older was unchanged over the year, at 39.9 percent in the fourth quarter of 2017:

In contrast to the rate for people of prime working age, the labor force participation rate for people 55 and older rose steadily from the mid-1990s until after the Great Recession. For the last 4 years, however, the rate has hovered around 40 percent. This general trend holds for both men and women in this age group, although older women are less likely to participate in the labor force than older men (34.6 percent compared with 46.1 percent in the fourth quarter of 2017).⁷

Within this age group, labor force participation trends have differed between those ages 55 to 64 and those age 65 and older. For those ages 55 to 64, labor force participation increased sharply from the mid-1980s through the end of the recession, hitting a high of 65.4 percent in the second quarter of 2009. (These data are not seasonally adjusted.) During the expansion, this rate has hovered in a narrow range of 63.6 percent to 65.4 percent and was 64.7 percent in the fourth quarter of 2017. For people age 65 and older, labor force participation began to increase sharply in the late 1990s and continued to rise through 2017. At 19.2 percent in the fourth quarter of 2017, the participation rate for people age 65 and older was 2.2 percentage points higher than it was at the end of the recession.

Among youth ages 16 to 24, unemployment was down in 2017. The unemployment rate for youth was 9.1 percent in the fourth quarter, after reaching a 17-year low in the third quarter of the year (9.0 percent). Within this age group, the unemployment rate for teens was considerably higher than that for young adults ages 20 to 24 (14.4 percent compared with 7.1 percent in the fourth quarter of 2017). The employment–population ratio of 16-to-24-year-olds was little changed over the year. However, at 49.9 percent in the fourth quarter of 2017, the ratio was 5.1 percentage points higher than the series low of 44.8 percent in the first quarter of 2010. Improvements in employment and unemployment among youth accrued to both young men and young women.

Labor force participation among youth had dropped through the 1990s and 2000s, with their participation rate settling around 55 percent in 2010; in the fourth quarter of 2017, the rate was 55.0 percent. While this trend in labor force participation holds true for both teenagers and young adults, young adults are about twice as likely to participate in the labor force as teenagers. In the fourth quarter of 2017, 34.7 percent of teenagers and 71.0 percent of young adults participated in the labor force.

Unemployment rates were down for all levels of education

Among people age 25 and older, unemployment rates declined in 2017 for nearly every level of education. All rates were at or near their prerecession lows. The unemployment rate for those with less than a high school diploma was down by 1.8 percentage points over the year, to 5.9 percent in the fourth quarter of 2017, which marks a series low (comparable data became available in 1992). The jobless rate for high school graduates fell by 0.9 percentage point over the year, to 4.3 percent at the close of 2017, and the rate for people with a bachelor’s degree or higher, at 2.1 percent, declined by 0.4 percentage point from a year earlier. The unemployment rate for people with some college was little changed over the year, at 3.6 percent in the fourth quarter of 2017. (See table 3 and figure 4.)

Following a downward trend beginning in the 1990s, participation rates for high school graduates, people with some college, and those with college degrees all showed little or no net change from the fourth quarter of 2016 to the fourth quarter of 2017. The rate for those with less than a high school diploma, however, did not show the same trend. Their participation rate had risen through much of the 1990s through the mid-2000s. Participation among those with less than a high school diploma gradually declined in the aftermath of the recession, hitting a recent low in 2014. Since then, their participation rate has generally trended up, rising sporadically, to 45.4 percent in the fourth quarter of 2017. This growth likely reflects the increase in participation among older people and the older age profile of people with less than a high school diploma. Among people with less than a high school diploma, 3 in 10 were age 65 or older in 2017, compared with just over 2 in 10 of those with all other levels of education.

The labor market also improved for veterans, the foreign born, and people with disabilities

Unemployment rates declined for both veterans and nonveterans. Of the 19.8 million veterans in the civilian noninstitutional population in the fourth quarter of 2017, the largest share—about 2 in 5—served during World War II, the Korean War, and the Vietnam era (7.9 million). Another 4.1 million served during Gulf War era II, 3.1 million during Gulf War era I, and 4.6 million served on active duty outside these designated periods.⁸

The unemployment rate for veterans declined to 3.5 percent in the fourth quarter of 2017, and the rate for their nonveteran counterparts declined to 3.8 percent. (These data are not seasonally adjusted.) Among all veterans, the unemployment rate for men declined to 3.5 percent in the fourth quarter of 2017, while the rate for women declined to 3.1 percent. For Gulf War–era II veterans, the unemployment rate for men declined to 3.8 percent, but the rate for women, at 4.4 percent, changed little from the prior year. (See table 4.)

In the fourth quarter of 2017, 49.6 percent of veterans participated in the labor force, down by 1.1 percentage points from a year earlier. For nonveterans, 65.4 percent participated in the labor force in 2017. The percentage for nonveterans was unchanged from the previous year. The difference in participation rates for veterans and nonveterans reflects the age profile of veterans who served during World War II, the Korean War, and the Vietnam era. Veterans from these service periods are now all over age 60, and labor force participation tends to be lower for older people than for those of prime working age. In fact, the decline in the participation rate for all veterans was driven by a 1.8-percentage-point decline over the year in the participation rate for those who served in World War II, the Korean War, and the Vietnam era. The rate for these veterans fell to 21.8 percent in the fourth quarter of 2017. In contrast, the participation rate for Gulf War-era II veterans—who tend to be younger—was 81.1 percent at the end of 2017, little different from a year earlier.

Unemployment rates for both foreign- and native-born workers declined. At the end of 2017, the foreign born accounted for 17.0 percent of the U.S. civilian labor force age 16 and older.⁹ The unemployment rates for both foreign- and native-born individuals each declined by 0.6 percentage point over the year, to 3.7 percent and 4.0 percent, respectively, in the fourth quarter. (These data are not seasonally adjusted.) Foreign-born people continued to have a slightly higher labor force participation rate than native-born people in 2017. The labor force participation rate for the foreign born and native born, at 65.5 percent and 62.1 percent, respectively, in the fourth quarter of 2017, showed little or no change from the prior year. (See table 5.)

Unemployment rates declined for both people with and people without disabilities. Although the job market remains challenging for people with disabilities, labor market indicators for this group improved for the second consecutive year. The labor force participation rate for people with disabilities increased by 0.9 percentage point over the year, to 20.9 percent in the fourth quarter of 2017, in contrast to participation trends for most other groups in 2017. (These data are not seasonally adjusted.) The employment–population ratio for this group also increased, by 1.1 percentage points over the year, to 19.2 percent. Despite this improvement in their employment–population ratio, people with disabilities were still less than one-third as likely to be employed as people with no disabilities. Although the unemployment rate for people with disabilities decreased by 1.4 percentage points over the year, to 8.2 percent in the fourth quarter of 2017, the rate was still more than twice as high as the rate for people without disabilities. (See table 6.)

While the unemployment rate for those with no disabilities improved, both their labor force participation rate and employment–population ratio remained relatively unchanged over the year. The unemployment rate for people without disabilities decreased by 0.6 percentage point over the year, to 3.7 percent. Their labor force participation rate was 68.2 percent, and their employment–population ratio was 65.6 percent.

Unemployment declined to a 17-year low in 2017

During the period leading up to the Great Recession, the unemployment rate was at a low of 4.5 percent for the three consecutive quarters that ended with the second quarter of 2007. Unemployment then rose sharply with the onset of the recession, reaching a high of 9.9 percent in the fourth quarter of 2009, before beginning a steady descent. At the end of 2017, 8 years into the expansion, the unemployment rate fell to 4.1 percent—a level last experienced shortly before the 2001 recession.

Changes in the unemployment rate are often analyzed with the use of data showing labor force status flows, which show the current employment status of people who were unemployed in the previous month. Each month, millions of people move between labor force statuses (for example, from unemployed to employed). The labor force status flows data measure these dynamic movements.¹⁰ In 2017, the percentage of unemployed people who remained unemployed from month to month continued its downward trend and was below 50 percent for much of the year. (These data are seasonally adjusted 3-month moving averages.) The percentage of unemployed people who found jobs in 2017 continued to inch up, reaching 27 percent for the first time since October 2007. Recently, unemployed people have been more likely to find a job than they were to leave the labor force, in contrast to earlier years of the expansion. (See figure 5.)

From job losers to first-time job hunters, the ranks of the unemployed fell for all groups

In the CPS, unemployed people are grouped according to what they were doing before they began looking for work. People are unemployed because they either (1) lost their job or completed a temporary job, (2) left their job, (3) reentered the labor force, or (4) entered the labor force for the first time. At the end of 2017, unemployment totaled 6.6 million, and all four categories declined over the year. (See table 7 and figure 6.)

The number of unemployed people who lost their job declined by 427,000 over the year, to 3.2 million. This group accounted for nearly one-half of the unemployed. The next largest group among unemployed people were reentrants—people who had been in the labor force previously, had spent time out of the labor force, and were actively seeking work once again. People who reentered the labor force accounted for about 31 percent of the unemployed in 2017. The number of reentrants, at 2.0 million, declined by 251,000 over the year. There were 729,000 job leavers (people who voluntarily left their job) by the end of the year—189,000 fewer than a year earlier. Lastly, the number of new entrants (unemployed people looking for a job for the first time) declined by 138,000 from a year earlier, to 635,000.

Long-term unemployment remained elevated

The number of people who were long-term unemployed (those who had been looking for work for 27 weeks or longer) declined in the last quarter of 2017, yet their proportion among the unemployed remained high by historical standards.¹¹ Long-term unemployment, at about 1.6 million people in the fourth quarter of 2017, was 320,000 less than a year earlier. However, the proportion of unemployed people who were jobless for 27 weeks or longer was little changed, at 23.9 percent. This proportion, although much lower than the historical high of 45.1 percent in the second quarter of 2010, was still notably higher than the prerecession level of 17.8 percent in the third quarter of 2007. This indicates that some people are still having great difficulty finding a job despite the prolonged expansion. In the fourth quarter of 2017, about 16 percent of unemployed people were searching for work for a year or more and about 7 percent were searching for more than 99 weeks, or almost 2 years. Both measures remain elevated 8 years after the recession ended. (These data are not seasonally adjusted.) (See table 7 and figure 7.)

Unemployment rates were down regardless of occupation

Jobless rates declined in all five major occupational categories in 2017. (These data are annual averages; the unemployed are classified by occupation according to their last job, which may or may not be similar to the job they are currently looking for.) The rate continued to be lowest in management, professional, and related occupations, declining by 0.3 percentage point from 2016, to 2.2 percent in 2017. The rate for sales and office occupations was down by 0.5 percentage point, to 4.1 percent. Production, transportation, and material moving occupations and service occupations both had an unemployment rate of 5.4 percent; rates in these occupations declined by 0.7 percentage point and 0.4 percentage point, respectively, from the previous year. The rate for natural resources, construction, and maintenance occupations declined by 0.4 percentage point, to 6.0 percent, in 2017. (See table 8.)

Employment growth in 2017

Management, professional, and related occupations accounted for most of the employment growth

In 2017, there were 1.3 percent more employed people than in the previous year, following a 1.7-percent increase in 2016. (These data are annual averages.) Employment expanded in two of the five major occupation groups. Employment in management, professional, and related occupations rose by 1.5 million, to 60.9 million, from 2016 to 2017; employment in this occupation grew by 2.5 percent in 2017, the same as in 2016. Natural resources, construction, and maintenance occupations also experienced employment gains in 2017, increasing by 289,000 to 14.2 million; employment in this occupation grew by 2.1 percent over the year after changing little in 2016. Employment in service occupations; sales and office occupations; and production, transportation, and material moving occupations changed little in 2017. (See table 9.)

Earnings growth continued

Median weekly earnings of full-time wage and salary workers were up 3.4 percent from 2016 to $860 in 2017.¹² During the same period, the Consumer Price Index for All Urban Consumers rose by 2.1 percent. (These data are annual averages.) The earnings comparisons here are on a broad level and do not control for many factors that can be important in explaining earnings differences, such as job skills and responsibilities, work experience, and specialization. (See table 10.)

Median weekly earnings for men reached $941, and women’s earnings rose to $770. Although earnings grew at the same rate (2.8 percent) for men and women in 2017, women earned 81.8 percent as much as men. This ratio has been in the range of 81 to 83 percent since 2010. (See figure 8.)

Among full-time wage and salary workers in the major race and ethnicity groups, median weekly earnings continued to be higher for Asians ($1,043) and Whites ($890) than for Blacks ($682) and Hispanics ($655). Over the year, Hispanics had a 5.0-percent increase in median weekly earnings. Earnings for Whites grew by 3.2 percent, while Asians’ earnings increased by 2.2 percent. Blacks experienced the smallest over-the-year increase in earnings at 0.6 percent.

Among workers age 25 and older, those with at least a bachelor’s degree continued to have the highest median weekly earnings at $1,279 (up by 1.6 percent) in 2017. Workers with some college or an associate’s degree had weekly earnings of $798 (an increase of 2.4 percent), while those with only a high school diploma had earnings of $712 (a 2.9-percent increase). Earnings remained the lowest for workers with less than a high school diploma at $520 (up by 3.2 percent).

Other measures of slack in the labor market improved in 2017

The number of involuntary part-time workers decreased

People who worked part time for economic reasons, often referred to as involuntary part-time workers, have been of particular interest in the aftermath of the recession. The count of involuntary part-time workers—that is, people who worked 1 to 34 hours a week for an economic reason—includes both people who usually work full time and people who usually work part time. The number of involuntary part-time workers fell over the year to 4.9 million in the fourth quarter of 2017. This measure hit a high of 9.1 million in the fourth quarter of 2009. Involuntary part-time workers accounted for 3.2 percent of total employment in 2017. In the third quarter of 2007, just before the onset of the Great Recession, there were 4.5 million involuntary part-time workers, who accounted for 3.1 percent of total employment. (See figure 9.)

Involuntary part-time workers were at work less than 35 hours mainly because of slack work or business conditions or because they could only find part-time work.¹³ The number of involuntary part-time workers in both of these categories decreased in 2017. As a share of total employment, both of these categories of involuntary part-time work have decreased in recent years.

The number of people who were not in the labor force but wanted a job decreased

People who are not in the labor force are neither working nor looking for work.¹⁴ In the fourth quarter of 2017, 95.7 million people were not in the labor force, little changed from a year earlier. (See table 11.) (These data are not seasonally adjusted.) Of those who were not in the labor force, about 2 in 5 were 65 years and older, similar to the proportion before the recession.

People who are not in the labor force are asked if they currently want a job. By the end of 2017, 5.0 million of those not in the labor force indicated that they wanted a job even though they were not currently looking for one. Over the year, this metric declined by 567,000. People who wanted a job accounted for 5.2 percent of those not in the labor force in the fourth quarter of 2017. This percentage was a series low—the smallest since comparable data became available in 1994.  

Among the 5.0 million people who wanted a job in the fourth quarter of 2017, 1.5 million were available for work and had looked for a job sometime in the prior 12 months but not in the past 4 weeks. This group is referred to as marginally attached to the labor force. Because these individuals did not actively search for work in the past 4 weeks, they were not classified as unemployed. The number of individuals marginally attached to the labor force declined by 226,000 over the year, after little change in 2016. People marginally attached to the labor force accounted for a similar share of those not in the labor force in the fourth quarter of 2017 as they did leading up to the recession.

Discouraged workers, a subset of the marginally attached, are people not currently looking for work specifically because they are discouraged over their job prospects. In the last quarter of 2017, the number of discouraged workers (489,000) was essentially unchanged from a year earlier.

The remaining 1.1 million people marginally attached to the labor force in the fourth quarter of 2017 had not searched for work for reasons other than discouragement, such as school attendance, family responsibilities, and health-related issues. The number of these individuals declined by 214,000 over the year.

Most alternative measures of labor underutilization have returned to their prerecession lows

The alternative measures of labor underutilization portray a broad range of labor market challenges faced by today’s workers, from various aspects of unemployment to insufficient hours of work for the employed.¹⁵ U-1 and U-2 are defined more narrowly than the official unemployment rate (which is included in this range as U-3) and include only subsets of unemployed people who are counted in the overall unemployment rate. U-4 through U-6 are defined more broadly than the official unemployment rate and include some people who are either employed or not in the labor force in addition to all of those who are unemployed. (See figure 10.)

In the fourth quarter of 2017, all alternative measures U-1 through U-6 declined. By the end of the year, all measures were back to, or very near, their prerecession lows.

Hurricanes Harvey and Irma swelled the number of people absent from work because of bad weather in September

Natural disasters and weather-related events adversely affect local economies in many ways. One obvious example is how these events disrupt people’s ability to work. In the third quarter of 2017, on August 25, 2017, Hurricane Harvey made landfall in Texas as a category 4 storm; Harvey affected Southeast Texas and the surrounding areas of the Gulf Coast. Shortly afterwards, on September 10, 2017, Hurricane Irma made landfall also as a category 4 storm, mainly affecting Florida and nearby states. Irma struck during the household survey reference week (the week that generally includes the 12th of the month). Because people are counted as employed in the CPS even if they were absent from their job for the entire reference week (regardless of whether they were paid), weather events typically do not have a discernible effect on total employment.

Of interest, however, is that the CPS collects data on weather-related work absences.¹⁶ In September 2017, 1.5 million full- and part-time workers had a job but were not at work for any part of the reference week because of bad weather. This figure is exceptionally high for September by historical standards. (See figure 11.) An additional 2.9 million full-time workers—those who usually work more than 35 hours per week—worked less than 35 hours during the survey reference week because of bad weather. (These are monthly data that measure what happened specifically during the survey reference and collection weeks and do not capture the full effect of weather-related events.)

Summary

The job market continued to improve for the eighth consecutive year in 2017. Employment–population ratios rose and unemployment rates declined for most demographic groups. Employment growth was largely concentrated in management, professional, and related occupations in 2017, and usual weekly earnings of full-time workers increased. The overall unemployment rate hit a 17-year low of 4.1 percent in the fourth quarter of 2017. In addition to the unemployment rate, other measures of labor market slack—such as the share of employment made up of involuntary part-time workers or the percentage of people not working or looking for work who want a job—hovered around or just below prerecession lows in 2017. Although unemployment and other measures of labor underutilization improved overall, long-term unemployment remained high. The labor force participation rate showed little change over the year.