KEY TAKEAWAYS
  • Since 2000, labor force growth has come primarily from increases in workers ages 55 and older, while the number of young and prime-age workers has stayed flat
  • The proportion of long-tenured employees is rising because of worker aging
  • Service sectors are expected to continue to drive future employment growth, although the gap in employment growth rates between services and goods-producing sectors has diminished since 2010
  • Women compose about 55% of current and projected net additions to the labor force and have higher educational attainment than men at all ages below 60

The composition of the US labor force is always changing. There are twice as many people over 55 working today than in the year 2000, many of them long-tenured at their current jobs. New entrants into the labor force are more likely to be women and college-educated than ever before, and they are more likely to enter service sectors.

This changing labor force affects the workers compensation system in many ways, a point made by NCCI’s CEO, Bill Donnell, in his opening remarks at Annual Issues Symposium 2018. The amount and types of workplace injuries vary for workers of different ages, gender, and experience, and across different economic sectors. Payroll growth also depends on the size and skills of the labor force.

This issue of the Quarterly Economic Briefing (QEB) highlights demographic trends in the labor force that are important for workers compensation. The discussion here focuses on worker aging, labor force participation, tenure, economic sector, and gender. This issue does not directly address workers compensation outcomes, but forthcoming NCCI research will show the impacts of these changing demographics on workers compensation frequency. That study will be released later this year.

Why is the US Labor Force Adding So Many Older Workers?

The most prominent trend in the US labor force over the past two decades is the increasing share of older workers—from about 12% to 23% of the labor force. And population aging will continue to impact the labor force in future. There are two main reasons for this trend:

Baby Boomers

Baby Boomers, defined by the US Census Bureau as people born during the years 1946–1964, have had an outsized impact on the labor force for their entire working lives. The Baby Boomers are larger than any cohort before them. As a result, most of the growth in older workers since 2000—as well as the leveling off of middle-age workers—is due to Boomers’ aging into their fifties and sixties. We show this in Figure 1, which measures the size of the US labor force by year as estimated by the Current Population Survey.Figure 1 and other figures citing IPUMS-CPS are the authors’ calculations from the Integrated Public Use Microdata Series data set of the Current Population Survey produced by the University of Minnesota, www.ipums.org.

We divide the labor force into three age ranges: younger than 35, 35–54, and 55 and older. The oldest Baby Boomers turned 55 in 2001, which is marked by the shading on Figure 1. The figure shows that around this time, the size of the labor force age 35–54 leveled off, while the size of the labor force 55 and over was beginning its long increase. The youngest Boomers will turn 55 in 2019. However, the Bureau of Labor Statistics (BLS) projects the number of older workers to continue to increase, in part because of trends in labor force participation we discuss next.

Older Americans Have Been Working More Since the 1990s

The increase in older workers is not entirely due to population aging. It is also partially caused by increased labor force attachment among older people.

Figure 2 shows the labor force participation rates for the three age ranges. Since 1990, individuals age 55 and older have increased their labor force participation rate by one-third, from 30% to 41%. Most of this increase occurred between the mid-1990s to mid-2000s, meaning that this phenomenon is not primarily a matter of delayed retirement for people whose nest eggs were hurt by the Great Recession.

Over the same time period, the labor force participation rate of adults under 55 has declined. This decline is most pronounced for people under 35, from 76% to 69%, and is largely (but not entirely) due to increased time spent in school. Full-time students who are not working are not considered part of the labor force. The decline for individuals ages 35–54 is smaller. Their labor force participation has dropped from 83% in 1990 to around 81% now—a small decline in percentage terms, but still responsible for 1.5 million fewer people in the labor force than there would be at the 1990 participation rate.

What is the net effect of these changes within age groups, and of population aging generally? The overall US labor force participation rate—defined as the proportion of Americans age 16 and older either employed or looking for work—has decreased from a high of around 67% in the late 1990s to about 63% today.

About three-quarters of this decline took place from about 2008 to 2013. The rate of decline has decreased since 2013 despite the ongoing economic expansion. The biggest reason for this is the higher proportion today of older Americans. Declining labor force participation of younger and middle-age people also contributes. If not for the increased labor force participation rates for older Americans, the decline shown in the Figure 3 would be even sharper.

How Aging Affects Employee Tenure

How long do workers stay with employers? Employee tenure is an important aspect of the labor market because workers may develop firm-specific human capital, affecting their productivity and wages. More experienced workers also tend to have lower accident rates, making tenure particularly relevant to workers compensation.

Figure 4 shows employee headcounts by tenure from 1996–2016. These data, from the Job Tenure and Occupational Mobility supplement from the Current Population Survey, were collected in even-numbered years. The number of employees with over one year and over five years’ tenure increased during this period, while the number of workers with less than one year of tenure decreased slightly.

This result is not so surprising in view of the large increase in older workers over this time period, since older workers are likely to have longer job tenure.

Figure 5 shows the proportion of workers in each tenure category by age range. Job tenure patterns within age groups have not varied much over time, suggesting that the increase in average job tenure is mostly due to the workforce aging we discussed above.

Changing Sector Mix

Much has been made of the long-term trend in employment away from goods-producing sectors—especially manufacturing but also construction and resource extraction—to service sectors. This has important implications for workers compensation.

Construction, manufacturing, and natural resources and mining all have above-average injury rates. So does trade, transportation and utilities, which is classified as a service sector but whose transportation and utilities elements often complement resource or manufacturing production.

Figure 6 shows the increase in the employment share of service-sector jobs since 1990. The proportion of workers in service industries has continuously increased over this period, but since 2010 the share has nearly leveled off, rising only from 83.5% to 83.8% in 2017.

In terms of total employment, rather than employment share, long-run growth in the United States for decades has been almost entirely driven by service sectors. Fifty years ago, there were approximately 22 million nonagricultural jobs in goods-producing industries. In 2017, there were about 20 million such jobs. Over the same 50 years, service sector employment has nearly tripled.

Changing Gender Mix

There is also an important gender element to sector mix. Women now compose 47% of the total labor force and more than half of labor force growth. BLS statistics from August show that women currently make up 53% of private service-providing employment but only 22% of private, nonagricultural goods-producing employment.

In Table 1, we show total employment, BLS-projected employment growth, and the female share of employment, all broken down by sector. This illustrates the sharp contrast in goods and service sector employment by gender.

Construction and mining have the lowest percentages of female employees, and women also make up less than 30% of manufacturing workers. In contrast, women hold three-quarters of health care and social assistance jobs and over two-thirds of jobs in educational services, as noted in last year’s QEB issue about the impact of automation on employment.

Also, the five sectors that employ more women than men account for two-thirds of all projected employment growth through 2026.

Figure 7 shows overall patterns of labor market participation rates by gender and age.

The first thing to note is that the major demographic labor shift over the second half of the 20th century—increased labor market participation among women—has not continued since 2000. The overall labor force participation rate for women peaked in 1999 and has been in gentle decline for two decades. This decline is driven by younger and middle-age women, as reflected in the panels for women ages 16–34 and 35–54. Women 55 and older are the exception, with increased labor force participation continuing since 2000, a directional result that matches the Figure 2 result showing labor force participation rates for both men and women.

The highlighting on all three of Figure 7’s panels shows that trends in men’s and women’s labor force participation by age have roughly paralleled one another since 2000.

Another gender difference by age relates to educational attainment. Figure 8 shows the current proportion of US men and women who have four-year college degrees. We divide age groups into 5-year intervals, instead of the three large age bands used previously, to show the changing proportion of college graduates by gender within the larger age ranges. (We also exclude adults younger than 25, many of whom have not completed their schooling.) Another important point is that Figure 8 includes all men and women regardless of labor force attachment.

Women are more likely than men to be college graduates at every age below 60, and these education gaps are largest for younger cohorts. Men are much more likely than women to hold college degrees for cohorts age 65 and above, but women are much more likely than men to hold college degrees for cohorts younger than 45. Similar patterns between men and women hold at different levels of educational attainment. Women under 60 are also more likely than men of the same age to have graduated from high school, to have at least an associate’s degree, and to have a post-secondary degree.

As older workers retire and new cohorts of young people complete their schooling, we can expect the overall gender education gap of the population to increase in the coming years.

The Impact of Labor Force Demographics on Workers Compensation

What does this mean for workers compensation?

Changes in the workforce affect trends in workers compensation outcomes such as frequency and severity. NCCI’s 2011 research brief “Workers Compensation and the Aging Workforce” found that younger and older workers were becoming more similar to one another in injury frequency and cost per worker. Since the convergence was mostly due to rapid frequency decline for young workers, who previously had very high injury rates, this was good news. Looking ahead, however, this convergence invites a natural question whether continued workforce aging will have much impact on frequency in the future. This is one of the issues we will address in our forthcoming research brief on changing demographics and workers compensation frequency.

Our forthcoming brief will also expand our prior analyses beyond age, since workers compensation outcomes can also differ by other demographic characteristics discussed in this QEB. Historically, workers who have more tenure, who work in service sectors, and who are female are less likely to suffer workplace injuries. How much (or how little) of recent frequency declines can be attributed to these types of workers becoming more prevalent? And what do future projections mean for future injury outcomes?

Our brief will report and analyze recent data on these topics. We will also delve into the intersections of these demographic factors and examine other characteristics of work injuries to try to explain the key elements driving changes in frequency.

For example, how have the distributions of types of injuries suffered by younger and older workers contributed to the relative change in frequency by age over time? How much does the sorting of men and women into different sectors and occupations drive the observed gender difference in frequency? This discussion organizes what we know about recent and expected future changes in the demographic characteristics of US workers, and thus provides a foundation for us to investigate these and related questions about the relationship between changing demographics and workers compensation.