Department of Labor Logo United States Department of Labor
Dot gov

The .gov means it's official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you're on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Article
December 2016

Underemployment among Hispanics: the case of involuntary part-time work

The unemployment rate, a leading indicator of the nation’s economic health, has fallen steadily in the wake of the Great Recession of 2007–09. However, other indicators of labor force strength paint a more complex picture of how workers are faring economically. In this article, we use 1971–2014 data from the Current Population Survey to examine temporal changes in involuntary part-time work—an increasingly common type of underemployment. Our analysis identifies several shifts in involuntary part-time work, including high rates of such work among Hispanic workers since the late 1980s. While this form of underemployment grew substantially among all racial/ethnic groups during the Great Recession, it was especially prevalent among foreign-born Hispanics, in particular those without citizenship. Although our analyses of 2014 data suggest that educational attainment accounts for much of these racial/ethnic and nativity gaps, other factors—namely, job skill, industry of employment, and occupational composition—also help explain the observed differential rates of involuntary part-time work.

Underemployment is critical to understanding both the current labor force situation of many workers and the prospects for continued economic growth. Underemployed workers accrue less human capital than those who are fully employed.1 Underemployment thus strains the nation’s economy—especially during economic downturns—by diminishing workers’ economic capital (i.e., earnings), thereby decreasing consumer demand and lowering economic output. Andrew Sum and Ishwar Khatiwada have noted that underemployment during the Great Recession resulted in “slightly more than $68 billion in lost earnings.”2

Although the harmful effects of underemployment on economic output and workers’ financial well-being are well understood, much less is known about how underemployment intersects with the increased racial/ethnic diversity of the U.S. workforce over the past 40 years. Perhaps second only to women’s rapid entry into the labor force in recent decades, this increased diversity—which is largely due to the growing presence of Hispanics in the United States—has been one of the most dramatic demographic changes in the U.S. economy. Hispanics made up 14.5 percent of the nation’s labor force in 2010, compared with just 8.5 percent two decades earlier,3 and their share is expected to rise to 24 percent by 2050.4

Previous research has shown that underemployment, in its numerous forms,5 affects non-White workers substantially more than White workers. Since the 1970s, underemployment has been more persistent among Hispanic workers, and the gap between the underemployment rates of this group and those of non-Hispanic Whites has grown. This divide is especially evident in rates of involuntary part-time work, which is one of the most common types of underemployment.6 Hispanics were particularly hard hit during the Great Recession, with many of them seeing their rates of involuntary part-time work almost triple.7

Despite these disparities, the overrepresentation of Hispanics among involuntarily employed part-time workers has garnered little research attention. While some researchers have noted that Hispanics’ higher levels of underemployment are partly due to differences in human capital characteristics, such as education,8 no one has focused specifically on involuntary part-time work and its growth over time. In addition, no previous study has examined the effects of job skill and occupational segregation on this type of underemployment. Many investigations have also analyzed trends only up to the late 1990s or early 2000s, although major demographic and economic shifts have unfolded in the last decade.

In this article, we address these research gaps in four key ways. First, we trace the diverging trends in the prevalence of involuntary part-time work between Hispanic workers, on the one hand, and non-Hispanic White and Black workers, on the other, paying special attention to the effects of the Great Recession. Second, we examine how this form of underemployment has become stratified by nativity (i.e., foreign born vs. native born) and citizenship status among Hispanics. Third, using logistic regression models, we explore whether differences in education and industry of employment explain the observed disparities in the prevalence of involuntary part-time work. Lastly, we consider the mediating effects of two key measures of job characteristics that are often overlooked in the underemployment literature: job skill and occupational composition. Overall, our effort addresses the following three research questions:

  1. To what extent has involuntary part-time work increased since the 1970s?
  2. Do temporal changes in rates of involuntary part-time work differ by race/ethnicity, nativity, and citizenship?
  3. Do differences in education, industry of employment, job skill, and occupational composition explain demographic gaps in the prevalence of involuntary part-time work?

Data and methods

Data

The data for this article are drawn from the 1971–2014 Current Population Survey (CPS) Annual Social and Economic Supplement, which is provided by the Integrated Public Use Microdata Series (IPUMS) project at the University of Minnesota.9 The CPS, conducted by the U.S. Census Bureau for the U.S. Bureau of Labor Statistics (BLS), surveys approximately 50,000 households each month and is the primary source of data for key economic indicators, such as the official unemployment rate. Researchers and policymakers rely on the CPS and its supplements to gauge the state of the U.S. economy, including the incidence of involuntary part-time work.10

All of our data are weighted to account for sampling design. We use individual-level probability weights provided by IPUMS and account both for clustering of the sample within households and for stratification in sample design by geography.11 Our geographical stratification is based on Census divisions instead of states, because, until 1977, many states were grouped in pairings that changed over time; by contrast, the Census-division classification has remained the same for all years.

Sample

We restrict our sample to workers ages 25 and older, who are more likely to have completed their education. Because the CPS did not collect ethnicity detail before 1971—and ethnicity is one of our key measures of interest—our analyses use data collected from 1971 to 2014.

Dependent variable: involuntary part-time work

We limit our investigation of underemployment to involuntary part-time work (in the CPS, part-time work is defined as working less than 35 hours per week), since this form of economic disadvantage is both increasingly common and less studied than unemployment and other forms of underemployment. Workers are considered to be in an involuntary part-time status if they desire full-time work but do not work full time, either because their hours have been reduced as a result of lack of work or because they are searching for, but cannot find, a full-time position.

Explanatory variables

Race/ethnicity, foreign-born status, and citizenship. We include three racial/ethnic groups in our analysis: non-Hispanic Whites, non-Hispanic Blacks, and Hispanics (of any race). We adopt the U.S. Census Bureau’s definition of “foreign born,” which includes those who are not U.S. citizens at birth. The definition excludes people born in U.S. territories, such as Puerto Rico, Guam, and the U.S. Virgin Islands, or people whose citizenship status is listed as “born abroad to a U.S. citizen parent.”12 In estimating the effects of nativity and citizenship, we include binary variables for native-born Hispanics, foreign-born Hispanic citizens, foreign-born Hispanic noncitizens, native-born non-Hispanic Blacks, and native-born non-Hispanic Whites (the reference group).

Education. In examining involuntary part-time work, we include a series of binary variables representing the following educational categories: less than high school diploma; high school diploma or GED (general equivalency diploma); some college, no degree; associate’s degree; and bachelor’s degree or higher (including graduate or professional degree). The last category serves as the reference group.

Job skill. Formal education is but one dimension of human capital that might influence the incidence of involuntary part-time work. Another dimension is the skills required for a job. To capture this dimension, we use three binary variables to represent the following job-skill categories: low skill, middle skill, and high skill (the reference category). These categories are derived from the BLS education and training classification system.13 Low-skill jobs typically require less than a month of on-the-job training, no previous work experience, and no more than a high school diploma. Middle-skill jobs require at least one of the following: a month of on-the-job training, some experience in a related occupation, an apprenticeship, certification, or a 2-year degree (but no more than a 4-year degree). Lastly, high-skill jobs typically require at least a 4-year degree.

Because our job-skill measure captures occupational requirements beyond formal education, it is correlated with, but not directly based on, education. Job skill and education constitute separate concepts (for example, one could have a 4-year degree but still work in a low-skill job that requires no such credential). To test for the presence of collinearity, we calculated a variance inflation factor (VIF) for each category of skill. Our calculation yielded a maximum VIF value of 2.4 for our low-skill variable and a maximum VIF value of 2.2 for our middle-skill variable. Given these values, we find no reason to suspect that collinearity is an issue with respect to job skill.

Industry. Industry of employment can influence a worker’s economic security, including his or her odds of underemployment. We use the following binary variables to represent this dimension: agriculture, forestry, fishing, and mining; construction; manufacturing; transportation, utilities, and communication; trade; finance, insurance, and real estate; service; and public administration. As in previous research on underemployment,14 the service industry is the reference category.

Occupational composition. We include two ratio variables for occupational composition in our regression models: percent Hispanic and percent Black (non-Hispanic). These measures, estimated with data from the U.S. Census Bureau’s Equal Employment Opportunity tabulations (based on the 5-year American Community Survey, 2006–11),15 capture the percentage of, respectively, Hispanics and non-Hispanic Blacks in a respondent’s occupation. Our calculation of maximum VIF values (1.5 for percent Black and 2.1 for percent Hispanic) suggests that collinearity is not an issue with these measures.

Control variables

Gender. Gender is included in our logistic regression models as a binary variable with a value of 1 for women and 0 for men (the reference group).

Age and age-squared. We include two age variables: age and age-squared. The latter variable allows us to test whether the age effect on underemployment is concave (i.e., underemployment declines as workers approach middle age and rises as they near retirement age). To reduce collinearity, we center both variables at their means.

Region. We control for region by including binary variables for four regions: Midwest, South, West, and Northeast (the reference category).

Marital status. To control for marital status, we include three binary variables: separated, divorced, or widowed; single or never married; and married (the reference category).

Residency status. The CPS includes a variable describing whether respondents, given their county of residence, live in metropolitan areas or outside of them, in rural areas. To control for residency status, we use a binary variable coded 1 for those living outside metropolitan areas and 0 for urban residents (the reference group). Urban residents are defined as those who live in metropolitan areas, whether in central cities, outside of central cities, or in metropolitan locations whose central-city status is not known.

Analytic approach

We begin with a descriptive analysis of underemployment patterns by race/ethnicity, focusing on the period 1971–2014. We then describe underemployment differences by citizenship and nativity status, restricting the period of analysis to the years 1994–2014 (measures of citizenship and nativity were not included in the CPS until 1994). Finally, we estimate logistic regression models to examine whether the gaps revealed in our descriptive analyses are explained by human capital and job characteristics.

Descriptive analyses

Involuntary part-time work by race/ethnicity

Figure 1 shows the proportion of workers in the labor force who are engaged in involuntary part-time work, by race and ethnicity. Blacks had the highest rates of involuntary part-time work before the mid-1980s (6.8 percent in 1971), but after that time the rates of Hispanics overtook those of Blacks—a pattern that remains today. Whites have always enjoyed the lowest rates of underemployment (2.4 percent in 1971 and 4.0 percent in 2013). In each year displayed in figure 1, the underemployment differences between Whites, on the one hand, and Blacks or Hispanics, on the other, were statistically significant (two-tailed t-tests, p < .05). With respect to differences between Black and Hispanic workers, only in 1971, 1977, 1979, and 1981 did Black workers have rates of involuntary part-time work significantly higher than those of their Hispanic counterparts.

A general pattern evident in figure 1 is that increases in underemployment are more dramatic during recession years, with that impact being stronger among Blacks and Hispanics than among Whites. Although the Great Recession deepened the racial/ethnic disparities in underemployment, these disparities date back decades. Tim Slack and Leif Jensen, for example, examined a broad measure of underemployment, often referred to as “inadequate employment,” and found that, between 1968 and 1998, underemployment remained much higher among Blacks and Hispanics than among Whites.16 For all groups, underemployment followed cycles of economic decline. The authors also observed that, while the Black–White gap remained stagnant, the Hispanic–White gap widened from the mid-1980s onward, a trend consistent with the economic restructuring that detrimentally affected less educated workers.17

Since the recession of 1990–91, involuntary part-time work has been most widespread among Hispanics. Before that time, there was only one year (1988) in which Hispanics had significantly higher rates of this form of underemployment than did Black workers. In 1989 and 1990, the differences between these two groups were not significant. From 1991 to 2014, however, the Hispanic rates of involuntary part-time work were significantly higher than those of Black workers, except for 1999, when the difference was not significant. During the Great Recession, the Hispanic rates tripled, from 4.2 percent in 2007 to 12.5 percent in 2009; neither White nor Black workers experienced as sharp an increase during the recession. By March 2014, the incidences of involuntary part-time work for all three groups had yet to return to their prerecession levels. During the entire postrecession period, Black and Hispanic workers continued to have higher rates of underemployment than did White workers, with the rates of Hispanics approaching, but still remaining significantly higher than, those of Blacks (p < .001).

Involuntary part-time work by nativity and citizenship status

Figures 2, 3, and 4 show the incidence of involuntary part-time work, by nativity and citizenship status, for non-Hispanic Whites, non-Hispanic Blacks, and Hispanics (of any race), respectively.18 The data are limited to the years 1994–2014, because this is the period for which information on nativity and citizenship is available. Before, during, and after the Great Recession, non-Hispanic Whites continued to exhibit the lowest rates of underemployment, irrespective of nativity and citizenship status. Differences in involuntary part-time work between U.S.-born Black and Hispanic workers were significant only in 2005. Compared with U.S.-born Hispanics, however, foreign-born Hispanic noncitizens had higher rates of involuntary part-time work in each year between 1994 and 2014; the rates of foreign-born Hispanic citizens were significantly higher than those of U.S.-born Hispanics in every year since 2009. During the last two recessions (2001 and 2007–09), much of the increase in involuntary part-time work was concentrated among the foreign born, especially those without citizenship, and this noncitizen disadvantage was particularly stark among Hispanics.

Figures 2–4 also show that, in the last two decades, the foreign-born, noncitizen disadvantage persisted only among Hispanics. This disadvantage is much less evident among Black and White workers. For example, throughout the period, Whites born in the United States had lower rates of underemployment than White noncitizens, although the rates of these two groups often converged (as they did in 2000 and between 2004 and 2008).19 Similarly, the rates of foreign-born Black noncitizens were lower than those of native-born Blacks in a number of years (1995, 1999–2000, and 2006–07). By comparison, in every year between 1994 and 2014, foreign-born Hispanics without citizenship reported significantly higher rates of involuntary part-time work than did U.S.-born Hispanics, and their rates were significantly higher than those of foreign-born Hispanic citizens in each year except 2014. In addition, throughout the entire period, foreign-born Hispanic noncitizens experienced involuntary part-time work more often than foreign-born White noncitizens (p < .01), and their underemployment rates were higher than those of foreign-born Black noncitizens in 11 years of the period.20

Regression analysis

Our descriptive analyses reveal important differences in underemployment among racial/ethnic, nativity, and citizenship groups. But to what extent are these differences explained by measures of education, industry of employment, job skill, and occupational composition? To answer this question, we estimate logistic regression models that control for group differences in those measures. Because our goal is to compare Hispanic workers of different nativity with native-born White and Black workers (this is the dimension along which underemployment gaps are more pronounced in recent years), we exclude foreign-born (non-Hispanic) Whites and foreign-born (non-Hispanic) Blacks from the analysis. Table 1 presents our regression results for 2014. We first discuss the effects of human capital and job characteristics on the incidence of involuntary part-time work, and then examine how these characteristics attenuate the underemployment gaps along racial/ethnic, nativity, and citizenship lines.

Table 1. Nested models predicting involuntary part-time work, odds ratios and robust standard errors (RSEs), 2014 (n = 434,581)
VariableModel 1Model 2Model 3Model 4Model 5Model 6Model 7
Race/ethnicity, nativity, and citizenshipEducationEducation and industryEducation, industry, and job skillEducation, industry, and occupational compositionEducation, industry, job skill, and occupational compositionAll controls
Odds ratioRSEOdds ratioRSEOdds ratioRSEOdds ratioRSEOdds ratioRSEOdds ratioRSEOdds ratioRSE

Race/ethnicity, nativity, and citizenship (White, native born)(1)

Black, native born

1.51***0.141.34**0.121.38***0.121.31**0.121.130.101.140.101.040.10

Hispanic, native born

1.32**.131.11.111.10.111.05.10.99.10.98.10.85.09

Hispanic, foreign born, citizen

2.30***.301.63***.221.63***.221.55**.211.34*.181.36*.191.29.18

Hispanic, foreign born, noncitizen

2.93***.261.73***.181.64***.171.47***.161.16.131.19.131.14.13

Education (bachelor's degree or higher)

Less than high school

3.50***.383.25***.372.35***.281.82***.221.74***.221.78***.22

High school or GED

2.33***.192.22***.191.66***.161.46***.131.37**.131.39**.14

Some college

2.21***.202.15***.201.67***.171.61***.151.51***.151.44***.15

Associate's degree

1.54***.181.51***.171.23.151.20.141.15.141.12.14

Industry of employment (service)

Agriculture, forestry, fishing, and mining

.76.12.68*.11.71.13.72.13.76.14

Construction

1.26*.131.34**.141.34*.151.48**.171.64***.20

Manufacturing

.37***.05.41***.06.40***.06.42***.06.45***.06

Transportation, utilities, and communication

.48***.09.49***.09.51***.09.51***.09.56**.10

Trade

1.50***.111.30***.091.60***.121.48***.111.46***.11

Finance, insurance, and real estate

.39***.07.40***.07.48***.08.47***.08.48***.08

Public administration

.32***.08.31***.07.35***.08.34***.08.36***.08

Job skill (high skill)

Low skill

2.60***.261.70***.191.62***.18

Middle skill

1.33**.131.09.111.08.11

Occupational composition

Percent Black in occupation

1.04***.011.04***.011.03***.01

Percent Hispanic in occupation

1.04***.001.03***.001.03***.00

Region (Northeast)

Midwest

.97.09

South

.89.08

West

1.25*.12

Age (centered)

Age

1.03.02

Age-squared

1.00.00

Gender (male)

Female

1.18**.08

Marital status (married)

Separated, divorced, or widowed

1.76***.14

Single or never married

1.97***.15

Residency status (metropolitan)

Rural

1.00.08

Constant

.04***.00.02***.00.02***.00.02***.00.01***.00.01***.00.01***.00

F-statistic

41.37***39.12***34.36***39.30***43.90***41.00***34.71***

Notes:

(1) Reference categories shown in parentheses.

*Statistically significant at p < .05.

**Statistically significant at p < .01.

***Statistically significant at p < .001.

Note: According to adjusted Wald tests, boldfaced odds ratios are significantly different from their corresponding odds ratios in previous model (p < .05).

Source: U.S. Bureau of Labor Statistics and authors' calculations.

Underemployment effects of education, industry, job skill, and occupational composition

The full model (model 7) in table 1 presents results for the relationship between human capital and job characteristics, on the one hand, and the incidence of involuntary part-time work, on the other. With respect to education, our full model indicates that underemployment occurs more frequently among workers with less than a bachelor's degree. This gap is most evident for those with less than a high school diploma, whose likelihood of working part time involuntarily is nearly 1.8 times that of college-educated workers. Industry of employment matters as well. Workers in agriculture and extractive industries face odds of underemployment similar to those of workers in the service industry (the reference group). However, workers in construction and trade are more likely to work fewer hours than the reference group, and workers in manufacturing; transportation, utilities, and communication; finance, insurance, and real estate; and public administration have greater odds of working more hours.

Our regression results also illustrate the important role of job skill in underemployment. Workers employed in low-skill jobs, which require less than a month of on-the-job training and no formal education or previous experience, have odds of underemployment 70 percent higher than those of workers in high-skill jobs, which require at least a 4-year degree. Interestingly, workers in middle-skill jobs face odds of involuntary part-time work that are not significantly higher than those of workers in high-skill jobs. Although existing research has yet to establish the potential role of middle-skill jobs in keeping workers out of some forms of underemployment (in this case, involuntary part-time work), our findings suggest that this role may be substantial.21

Both of our measures of occupational composition—percent Black and percent Hispanic—are positive and significant, even after controlling for education, job skill, industry of employment, and a number of other demographic characteristics. Being overrepresented in the most volatile industries and occupations, minority workers are at a greater risk of involuntary part-time work, especially during times of economic uncertainty or decline, when employers seek to shed jobs (and hours). The high concentration of Hispanics in low-skill occupations—a phenomenon often referred to as “occupational crowding”—may be another channel driving this group’s higher incidence of involuntary part-time work.

Conditional effects of race/ethnicity, nativity, and citizenship

Our base model (model 1 in table 1), which does not control for education, job attributes, and other demographic characteristics, reveals notable racial/ethnic, nativity, and citizenship disparities in underemployment. Compared with native-born non-Hispanic Whites (the reference group), native-born non-Hispanic Blacks face 51 percent higher odds of involuntary part-time work and native-born Hispanics face odds 32 percent higher. The gap is much larger for Hispanics born outside the United States, however. Foreign-born Hispanic citizens exhibit odds of involuntary part-time work 2.3 times larger than those of native-born Whites, and foreign-born Hispanic noncitizens exhibit odds nearly 3 times larger.

Successive model iterations reveal the mediating effects of our control variables. Adding education (model 2) significantly attenuates the underemployment gap between non-Hispanic Whites and all other groups, including the two groups of foreign-born Hispanics. The odds ratios for foreign-born Hispanic citizens and noncitizens are reduced to 1.63 and 1.73, respectively. Moreover, education alone appears to explain the (native-born) White–Hispanic underemployment gap, with our binary variable for native-born Hispanics dropping from significance in model 2. Taken together, these results indicate that education disparities between foreign-born Hispanics and native-born Whites contribute substantially to the growing Hispanic underemployment disadvantage.22

Introducing industry of employment (model 3) alters the relationship between race, ethnicity, and citizenship, on the one hand, and underemployment, on the other, only for Hispanic foreign-born noncitizens. The mediating effect is one of reducing this group’s odds of involuntary part-time work. The odds are significantly lowered for all groups when we introduce measures of either job skill (model 4) or occupational composition (model 5)—an effect illustrating the importance of both measures. Further, in model 5, the difference between (native-born) White and Black workers dissipates, which demonstrates the potentially unique role that occupational composition plays in perpetuating underemployment gaps between these two groups. Similarly, the model’s indicator for foreign-born Hispanic noncitizens is no longer significant, which suggests that occupational composition plays a critical part in shaping this group’s employment outcomes. The results for model 6, which includes controls for both job skill and occupational composition, are similar to those reported for model 5.

Introducing other demographic correlates of underemployment (model 7) significantly lowers the odds of involuntary part-time work for native-born Black and Hispanic workers (once we account for education and job characteristics). These findings suggest that present-day disparities in involuntary part-time work between native-born White and foreign-born Hispanic noncitizen workers are driven not only by education and job characteristics, but also by demographic differences between these groups (i.e., differences by age, gender, marital status, and rural or urban residence).

The emergent pattern across our models is that minority workers—especially Hispanic foreign-born workers—are more disadvantaged in terms of involuntary part-time work. Foreign-born Hispanic noncitizens are particularly at risk. Because our analysis lacks an indicator for workers’ legal status, we are unable to determine whether or how much such status affects the rates of involuntary part-time work for this group.23

Conclusion

Our findings reveal a persistent Hispanic disadvantage with respect to one form of underemployment—involuntary part-time work. Although this disadvantage is largely explained by education—a variable whose role in underemployment has been established in previous research—it also reflects the strong causal influence of other factors, namely, job skill, occupational composition, and industry of employment. In addition, our results suggest that higher rates of involuntary part-time work occur more frequently among foreign-born Hispanic noncitizens, especially during economic downturns. We provide some evidence that the underemployment impacts of the Great Recession were much harsher among Hispanic noncitizens than among other workers.

Suggested citation:

Justin R. Young and Marybeth J. Mattingly, "Underemployment among Hispanics: the case of involuntary part-time work," Monthly Labor Review, U.S. Bureau of Labor Statistics, December 2016, https://doi.org/10.21916/mlr.2016.55

Notes


1 Barry T. Hirsch, “Why do part-time workers earn less? The role of worker and job skills,” Industrial and Labor Relations Review, vol. 58, no.4, July 2005, pp. 525–551.

2 Andrew Sum and Ishwar Khatiwada, “The nation’s underemployed in the ‘Great Recession’ of 2007–09,” Monthly Labor Review, vol. 133, no. 11, November 2010, pp. 3–15, https://www.bls.gov/opub/mlr/2010/11/art1full.pdf.

3 Mitra Toossi, “Labor force projections to 2020: a more slowly growing workforce,” Monthly Labor Review, January 2012, https://www.bls.gov/opub/mlr/2012/01/art3full.pdf.

4 Mitra Toossi, “A century of change: the U.S. labor force, 1950–2050,” Monthly Labor Review, vol. 125, no. 5, May 2002, pp. 15–28, https://www.bls.gov/opub/mlr/2002/05/art2full.pdf.

5 Underemployed workers include highly skilled full-time workers employed in low-paying or low-skill jobs and workers who work part time involuntarily. See Tim Slack and Leif Jensen, “Underemployment across immigrant generations,” Social Science Research, vol. 36, no. 4, December 2007, pp. 1415–1430.

6 Ibid.

7 See Sum and Khatiwada, “The nation’s underemployed”; and Justin R. Young, “Underemployment in urban and rural America: 2005–2012,” Issue Brief no. 55 (Durham, HN: Carsey Institute, University of New Hampshire, 2012).

8 Daniel T. Lichter, “Racial differences in underemployment in American cities,” American Journal of Sociology, vol. 93, no. 1, January 1988, pp. 771–792; Gordon F. De Jong and Anna B. Madamba, “A double disadvantage? Minority group, immigrant status, and underemployment in the United States,” Social Science Quarterly, vol. 82, no. 1, February 2001, pp. 117–130; Tim Slack and Leif Jensen, “Race, ethnicity, and underemployment in nonmetropolitan America: a 30-year profile,” Rural Sociology, vol. 67, no. 2, June 2002, pp. 208–233; Slack and Jensen, “Underemployment across immigrant generations”; and Tim Slack and Leif Jensen, “Underemployment among minorities and immigrants,” in D. C. Maynard and D.C. Feldman, eds., Underemployment: psychological, economic, and social challenges (New York: Springer, 2011).

9 Miriam King, Steven Ruggles, J. Trent Alexander, Sarah Flood, Katie Genadek, Matthew B. Schroeder, Brandon Trampe, and Rebecca Vick, Integrated Public Use Microdata Series, Current Population Survey: Version 3.0 [machine-readable database] (Minneapolis, MN: Minnesota Population Center, 2010).

10 Researchers have used the CPS to measure involuntary part-time work and other types of underemployment. See Slack and Jensen, “Race, ethnicity, and underemployment”; Leif Jensen, Jill L. Findeis, Wan-Ling Hsu, and Jason P. Schachter, “Slipping into and out of underemployment: another disadvantage for nonmetropolitan workers?” Rural Sociology, vol. 64, no. 3, 1999, pp. 417–438; Sum and Khatiwada, “The nation’s underemployed”; and Lichter, “Racial differences in underemployment.”

11 For a more in-depth discussion of the weighting procedures used in the CPS, see King et al., Integrated Public Use Microdata Series.

12 Profile of the foreign-born population in the United States: 2000 (U.S. Census Bureau, December 2001), www.census.gov/prod/2002pubs/p23-206.pdf.

13 “Education and training data definitions” (U.S. Bureau of Labor Statistics), https://www.bls.gov/emp/documentation/definitions.htm. The BLS classification system specifies education and training requirements typical for entry into an occupation, but workers in that occupation often possess more education or training than the minimum requirement.

14 See Slack and Jensen, “Underemployment across immigrant generations.”

15 “Tables for EEO tabulation 2006–2010 (5-year ACS data)” (U.S. Census Bureau), https://www.census.gov/data/tables/time-series/demo/eeo/acs-2006-2010.html.

16 Slack and Jensen, “Race, ethnicity, and underemployment.” Slack and Jensen’s measure of underemployment includes involuntary part-time work, inadequate wages, unemployment, and discouragement.

17 For a discussion of the economic restructuring that took place after the mid-1980s, see Arne L. Kalleberg, Good jobs, bad jobs: the rise of polarized and precarious employment systems in the United States, 1970s to 2000s (New York: Russell Sage, 2011).

18 Because of small sample size, estimates for Black foreign-born citizens and noncitizens should be interpreted with caution, particularly for the period 1994–2000, for which our sample sizes for these groups were sometimes lower than 200.

19 In two-tailed t-tests, the rates of involuntary part-time work for native-born White workers and White noncitizens were not significantly different in 2000, 2004–08, and 2014. Compared with foreign-born White citizens, White noncitizens had higher rates in all years (p < .05) except 1995, 1996, 2000, 2005–08, and 2011–14.

20 The rates of involuntary part-time work for Black and Hispanic noncitizens were not significantly different in 1994, 1996–99, 2001, 2004, 2008, 2013, and 2014.

21 Many middle-skill jobs have been on the decline, especially since the Great Recession. See Justin R. Young, “Middle-skill jobs remain more common among rural workers,” Issue Brief no. 63 (Durham, NH: Carsey Institute, University of New Hampshire, 2013).

22 Not only do immigrant workers typically possess lower levels of education, they also enter a labor market in which the lack of educational credentials puts workers at a strong disadvantage. See Julian R. Betts and Magnus Lofstrom, “The educational attainment of immigrants: trends and implications,” in George J. Borjas, ed., Issues in the economics of immigration (Chicago, IL: University of Chicago Press, 2000).

23 Previous research has indicated that unauthorized workers tend to have worse labor market outcomes and working conditions than authorized workers. See Katharine M. Donato and Blake Sisk, “Shifts in the employment outcomes among Mexican migrants to the United States, 1976–2009,” Research in Social Stratification and Mobility, vol. 30, no. 1, March 2012, pp. 63–77.

article image
About the Author

Justin R. Young
justin.young@unh.edu

Justin R. Young is a postdoctoral researcher in the UNH ADVANCE program, University of New Hampshire, Durham, NH.

Marybeth J. Mattingly
beth.mattingly@unh.edu

Marybeth J. Mattingly is director of research on vulnerable families at the Carsey School of Public Policy, University of New Hampshire, Durham, NH.

close or Esc Key