About


The Atlanta Fed's macroblog provides commentary and analysis on economic topics including monetary policy, macroeconomic developments, inflation, labor economics, and financial issues.

Authors for macroblog are Dave Altig, John Robertson, and other Atlanta Fed economists and researchers.

Comment Standards:
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.

March 26, 2019

Young Hispanic Women Investing More in Education: Good News for Labor Force Participation

In a recent recent macroblog post, my colleague John Robertson found that the recent rise in female prime-age (ages 25 to 54 years) labor force participation (LFP) over the last few years has been driven in large part by increased participation among Hispanic women. (Hispanic refers to people of Cuban, Mexican, Puerto Rican, South or Central American, or other Spanish culture or origin regardless of race.) Much of the LFP improvement among Hispanic women has come as they've shifted away from household duties.

To understand this development and determine whether it's a trend likely to continue, we look at trends in the activities of younger Hispanic women. In particular, we look at the so-called NEETs rate among women ages 16 to 24. The NEETs rate is the share of the youth population that is "Not Employed or pursing Education or Training." This group is sometimes referred to as "disconnected youth" or "opportunity youth" because they are generally less likely to be attached to the labor force as they move into their prime working years and are at higher risk of experiencing long-term unemployment, persistent poverty, poor health, and criminal behavior.

A look at the next chart shows substantial improvement in the NEETs rate among young Hispanic women over the last two decades. The gap has narrowed considerably and in recent years has tracked much more closely with black non-Hispanic women.

The declining NEETs rate for young Hispanic women primarily reflects shifting preferences toward more education and away from household responsibilities. As you can see in the next chart, the share of young Hispanic women who are in education or training has risen over the last two decades, up nearly 19 percentage points since 2000. Their share now more closely matches that of young black and white non-Hispanic women.

Mirroring the rise in educational activities has been a shrinking share of young Hispanic women who are not in the labor force because they are taking care of home or family, as the following chart shows.

Young Hispanic women have invested increased time in their education over the last two decades and as a result have higher average levels of educational attainment than earlier cohorts moving into their prime working years. To see this, the next chart shows the distribution of educational attainment over time for Hispanic women aged 25.

The higher levels of LFP in recent years among prime-age Hispanic women partly reflects the greater investment in education by younger Hispanic women. If this trend continues—and there is no obvious reason why it wouldn't—then it will help drive even higher labor force attachment for prime-age Hispanic women in the years to come.

 

August 8, 2018

Immigration and Hispanics' Educational Attainment

In a previous macroblog post, Whitney Mancuso and I wrote about the improved labor market outcomes for workers with the least amount of formal education. We attributed this improvement mostly to a combination of a secular decline in the supply of these workers over time and a shift in the composition of the low-skilled workforce toward Hispanic immigrants—a group that has an especially high rate of workforce attachment.

In a related article by colleagues at the St. Louis Fed, Alexander Monge-Naranjo and Juan Ignacio Vizcaino explore how the employment characteristics of the Hispanic population have grown increasingly concentrated in low-skilled occupations over time, and they relate this to the relatively smaller gains in the average educational attainment of the Hispanic population.

The authors ask why the education level of Hispanics has lagged behind other groups and suggest that it could be a consequence of intergenerational persistence; it takes a while for the children of poorly educated immigrants to catch up with the rest of the population. This explanation is likely to play a role, especially when considering why a relatively smaller share of U.S.-born Hispanics go to college. The study also notes differences across gender, showing that Hispanic men are less likely than Hispanic women to continue their education after high school, and although the college rate has been rising for all Hispanics, it is growing faster for women.

I also want to note that a large share of the Hispanic population in the United States are foreign born, and these immigrants have a much lower average level of educational attainment than do U.S.-born Hispanics. This observation is evident in table 1, which is based on data on individuals aged 25-54 (prime age) from the Current Population Survey. For instance, in 2017, 57 percent of the U.S. prime-age Hispanic population was foreign born, and 21 percent of these prime-age foreign born Hispanics had a college degree (associate degree or higher). In contrast, 36 percent of U.S.-born prime-age Hispanics had a degree.

Table 1: Selected Characteristics of the U.S. Prime-age Population (percent)

 

Foreign born

Completed a college/associate degree

 

Hispanic

Non-Hispanic

Hispanic

Non-Hispanic

 

 

 

Foreign born

U.S. born

Foreign born

U.S. born

1997

62

9

13

22

48

37

2007

64

12

15

29

56

43

2017

57

14

21

36

64

51

Source: Current Population Survey, author's calculations

As the St. Louis Fed study concludes, a primary factor distinguishing the Hispanic workforce in the United States is their lower average level of educational attainment. Further distinguishing between foreign and U.S.-born Hispanics shows the role that immigration has played in holding down the average education level since a large fraction of Hispanic immigrants have less education.

The Hispanic/non-Hispanic college completion gap remains large and has not closed over time. However, there has been relative improvement in high school completion, as table 2 shows.

Table 2: Selected Characteristics of the U.S. Prime-age Population (percent)

 

Foreign born

Completed 12th grade

 

Hispanic

Non-Hispanic

Hispanic

Non-Hispanic

 

 

 

Foreign born

U.S. born

Foreign born

U.S. born

1997

62

9

50

81

92

92

2007

64

12

55

87

93

94

2017

57

14

65

92

95

96

Source: Current Population Survey, author's calculations

Since 1997, the share of the prime-age foreign born Hispanic population who have finished 12th grade has increased by 15 percentage points. At the same time, the share of prime-age U.S.-born Hispanics completing high school has increased by 11 percentage points and is now not much lower than for non-Hispanics. While relatively low college attendance remains a major obstacle, greater high school completion is encouraging for Hispanics' future role in the workforce.

August 30, 2017

Is Poor Health Hindering Economic Growth?

It is well known that poor health is bad for an individual's income, partially because it can lower the propensity to participate in the labor market. In fact, 5.4 percent of prime-age individuals (those 25–54 years old) reported being too sick or disabled to work in the second quarter of 2017. This is the most commonly cited reason prime-age men do not want a job, and for prime-age women, it is the second most often cited reason behind family responsibilities (see the chart). (Throughout this article, I use the measure "not wanting a job because of poor health or disability" as a proxy for serious health problems.)

In addition to being prevalent, the share of the prime-age population citing poor health or disability as the main reason for not wanting a job has increased significantly during the past two decades and tends to be higher among those with less education (see the chart).

Yet by some standards, the health of Americans is improving. For example, compared to two decades ago the average American is living two years longer, and the likelihood of dying from cancer or cardiovascular disease has fallen. These specific outcomes, however, may have more to do with improvements in the treatment of chronic disease (and the resulting reduction in mortality rates) than improvements in the incidence of health problems.

Another puzzle—which is perhaps also a clue—is the considerable variation across states in the rates of being too sick or disabled to work. For example, people living in Mississippi, Alabama, Kentucky, or West Virginia in 2016 were more than three times likelier to indicate being too sick or disabled to work than residents of Utah, North Dakota, Iowa, or Minnesota (see the maps below).

This cross-state variation is useful because it allows state-by-state comparisons of the prevalence of specific health problems. Among a list of more than 30 health indicators, the two factors that most correlate with the share of a state's population too sick or disabled to work were high blood pressure (a correlation of 0.86) and diabetes (a correlation of 0.83). Both of these conditions are associated with risk factors such as family history, race, inactivity, poor diet, and obesity. Both of these health issues have increased significantly on a national basis in recent years.

So how might poor health hinder economic growth? Health factors account for a significant part of the decline in labor force participation since at least the late 1990s. After controlling for demographic changes, the share of people too sick or disabled to work is about 1.6 percentage points higher today than it was two decades ago (see the interactive charts on our website). Other things equal, if this trend reversed itself during the next year, it could increase the workforce by up to 4 million people, and add around 2.6 percentage points to gross domestic product (calculated using our Labor Market Sliders).

Of course, such a sudden and large reversal in health is highly unlikely. Nonetheless, significant improvements to the health of the working-age population would help lessen the drag on growth of the labor supply coming from an aging population. Public policy efforts centered on both prevention and treatment of work-impeding health conditions could play an important role in bolstering the nation's workforce.


April 11, 2017

Going to School on Labor Force Participation

In the aftermath of the Great Recession, labor force attachment declined. However, that pattern has been reversing itself lately. In particular, the labor force participation rate (LFPR) of the prime-age (25 to 54 years old) population, the core segment of the workforce, has been moving higher since late 2015. While this is good news, the prime-age LFPR remains well below prerecession levels, meaning that there are more than two million fewer prime-age people participating in the labor force. What factors have contributed to that decline? Where did those people go?

The Atlanta Fed LFP dynamics web page has an interactive tool that allows users to drill down into the drivers of the change in LFPR. The tool breaks the change in LFPR into two parts. The first part is the effect of shifts in the share of the population in different age groups (we use five-year age groups). The second part is the change attributable to shifts in the rate of nonparticipation. Using a methodology described here, we can drill deeper into the second part to learn more about the reasons for not participating in the labor force.

The U.S. Census Bureau will make the first quarter 2017 microdata on the reasons for nonparticipation available in a few weeks, so the following chart shows a decomposition of the 1.8 percentage point decline in the prime-age LFPR (not seasonally adjusted) between the fourth quarters of 2007 and 2016.

In this chart, "residual" pertains to the part of the total change in the LFPR that is attributable to the simultaneous shifts in both age-group population shares and age-group participation rates. In the present case, the residual is zero.

Because we are examining changes in prime-age participation, and all age groups within prime-age have reasonably similar participation rates, a change in the composition of ages tends to have little impact on the overall prime-age LFPR. Instead, the decline is due to shifts in the nonparticipation rate within age groups (the orange bar). In particular, the decline could indicate an increased likelihood of being in school, having family responsibilities that prevent participation, being in the shadow labor force (wanting a job but not actively looking), and a disability or poor health.

Although all these factors put downward pressure on participation, an important countervailing influence is that the education level of the population has been rising over time, and participation tends to increase with more education. In 2007, 41.0 percent of the prime-age population had a college degree, and they had an 88.3 percent participation rate versus 79.5 percent participation for those without a degree. By the end of 2016, the fraction with a degree had increased to 47.3 percent, and that cohort's participation rate had declined 1 percentage point, to 87.3 percent, versus a drop of 3.5 percentage points, to 76.0 percent, for those without a degree.

To see the importance of rising education on participation, the following chart shows the decomposition of the 1.8 percentage point decline in prime-age LFPR based on education-group population shares (degree and nondegree) instead of age-group shares.

In this chart, "residual" indicates the part of the total change in LFPR due to the simultaneous shifts in both education-group population shares and education-group participation rates.

As the chart shows, the shift in the education distribution of the prime-age population from 2007 to 2016 by itself would have increased the prime-age participation rate by about 0.7 percentage points (the green bar). Conversely, if education levels had not increased then the participation rate would have decreased by even more than it actually did. The nonparticipation effect would be larger for most nonparticipation reasons and especially for reasons of disability or poor health (−0.8 percentage points versus −0.5 percentage points). See the charts and analysis in the "health problems" section of the Labor Force Dynamics web page for more information on health-related nonparticipation by education.

Despite some partial reversal over the last year and a half, the prime-age LFPR is still lower than it had been prior to the recession. However, the decline in participation could have been even larger if the education level of the population had not also increased. Rising education is associated with a lower incidence of nonparticipation than otherwise would be the case, and it's principally associated with less nonparticipation attributable to disability or poor health. While researchers agree on the positive association between education and health, pinning down the specific reasons for this remains somewhat elusive. Factors such as income, informational, and occupational differences—as well as public policy choices—all play a role. Recent research by Nobel laureate Angus Deaton and Anne Case suggests that both education and racial differences are important considerations—emphasizing the sharply rising incidence of health problems among middle-age, white families with lower levels of education—and this Washington Post article highlights rising disability rates in rural America.