Elsevier

Social Science Research

Volume 32, Issue 3, September 2003, Pages 347-375
Social Science Research

Racial/ethnic differences in voluntary and involuntary job mobility among young men

https://doi.org/10.1016/S0049-089X(02)00063-7Get rights and content

Abstract

Using the 1979–1994 waves of the NLSY-79 data, this study investigates racial/ethnic differences in the rates of voluntary and involuntary job mobility among young men. We find that there is no significant difference among racial/ethnic groups in the likelihood of voluntary job changing. However, blacks do suffer from job instability in that their likelihood of leaving jobs involuntarily is much higher than that of whites or Hispanics. Within the Hispanic population, Mexicans are more likely to experience involuntary job separation compared to whites, though they are not as likely to do so as are blacks. The results confirm the importance of separately analyzing the mechanisms and processes of voluntary and involuntary mobility in order to understand better the disadvantages of some groups in career development.

Introduction

A large body of literature has shown that high rates of joblessness and frequent job turnover are major features of the youth labor market. It is well known, furthermore, that the labor market in the US is distinctive in its relatively less stable nature: workers hold more jobs and change jobs more frequently than do their European counterparts, especially German workers (Blossfeld and Rohwer, 1995). In addition, economic changes since the mid-1970s, facilitated by global economic competition and technological improvements, led to a decline in traditional full-time employment. Instead, an increase in non-standard employment relations such as part-time, temporary and contract work (Kalleberg, 2000) has perhaps led to a further deterioration of employment stability for young people.

In this paper we ask whether the patterns of job mobility differ across racial and ethnic groups. We focus on young men since the patterns of job mobility for young women are quite different. Analyzing the patterns for men and women from different racial and ethnic groups in a careful way would go far beyond what we could accomplish in one paper. Have some racial/ethnic groups suffered from job instability more than have others? More specifically, we compare the rates of job separation among four groups of young men—blacks, Mexicans, other Hispanics, and whites—and examine the disadvantages of young Mexican, other Hispanic, and black men relative to white men in relation to the development of a stable career path. We focus our attention on various determinants that affect the rates of moving out of jobs among those four groups. This work complements other work that has examined other features of the labor market experiences of members of minority groups, including job acquisition (Elliott, 2000) and earnings (Hsueh and Tienda, 1996, Hsueh and Tienda, 1995)

As previous studies have shown, employment instability is an important mechanism though which racial and ethnic differences in labor market outcomes such as unemployment, earnings or promotion occur. Greater instability associated with more frequent job turnover among minorities discourages minority workers from actively participating in the labor force and prevents their accumulation of job-specific experiences and tenures (Hsueh and Tienda, 1996; Tienda and Stier, 1996). This demonstrates the importance of looking at the patterns of job mobility in order to better understand racial and ethnic inequality in the labor market.

Although a number of studies have compared the job shift behavior of black youth with that of whites (Donohue, 1988; Farber, 1992; Tiemeyer, 1993), far less research has investigated the extent to which Hispanic or Latino youth differ from blacks and whites. In most studies, Hispanics are either combined in the category of “non-white” with blacks, or they are totally excluded from the analysis. This absence of research on the job shift behavior of young Hispanic people as distinct from blacks and whites might be understood as a reflection of the lack of data on the work histories of Hispanics. The NLSY (National Longitudinal Survey of Youth) data we use in this paper, however, permit an examination of job changing for Hispanics as a distinct group.

As a number of researchers have pointed out, the umbrella Hispanic category disguises a great deal of heterogeneity. The Hispanics who were in the United States in 1979 when the NLSY was initiated represented many different countries of origin and could be recent immigrants or members of the second, third, or higher generation of their families in the United States. The size of the Latino sample does not permit us to look at all of the major Hispanic groups in the United States, but we are able to look at Mexicans and a residual category of other Hispanics. The Mexican situation is particularly important since they represent the largest single Hispanic group in the United States, and many Mexican individuals and families enter the United States specifically to take advantage of better employment opportunities here relative to Mexico. In some ways Mexican young men and black young men share similar disadvantages relative to white young men: they have relatively low levels of education and work in low-paying jobs. On the other hand, the experience of immigration is a factor that makes many Mexican young men very different from many black young men.

In addition to exploring the ways in which the job shift behavior of Mexican and other Hispanic men differs from that of blacks and whites, we distinguish between voluntary and involuntary moves. The decision to leave a job can be made either by the jobholder, or the decision may be forced upon the worker by the employer’s decision to terminate employment. It is widely recognized that the processes and determinants of involuntary moves are different from those of voluntary moves (Breen, 1992; Hachen, 1990; Tuma, 1976). In recognition of the significant distinction between leaving a job voluntarily and being fired, Tuma (1976) decomposed the rate of job mobility into three parts, including the rates of quitting and being fired as well as a residual rate. Furthermore, she anticipated that determinants of job mobility should affect the two events in the opposite way, e.g., factors that increased the rate of voluntary job mobility would decrease the rate of involuntary job mobility.

Most analyses of job changing have not distinguished between voluntary and involuntary events, or they have focused only on voluntary shifts in spite of general agreement on the necessity of separate analyses (Rosenfeld, 1992).1 For example, Light and Ureta (1992) and Farber (1992) treated all job separations as being alike. Alternatively Trevor (2001) examined only voluntary job changes.

The literature separately considering voluntary and involuntary job separations is very limited. Among the exceptions are the studies of Jackson and Montgomery (1986) and of Hachen (1990).2 The former showed that blacks are more likely to be discharged than whites, though there is no significant difference in the likelihood of quitting. Utilizing the hazard models of job separation, Hachen’s research confirms the conclusion that the explanatory variables pertaining to job exits have different influences on voluntary and involuntary moves. Hachen noted, however, that the small number of involuntary exits in the data limited his analysis and conclusions. Thus, he suggested that the results should be interpreted with caution because the relative frequency of involuntary moves was small. Therefore, we still need further research that models separately voluntary and involuntary mobility with enough cases of both events in order to reach a reasonable conclusion on the issue.

In this paper we examine voluntary and involuntary mobility as defined by the individual. We are, in effect, examining the extent to which individuals feel they have control over whether or not they left their job. Employers, of course, might have a different perspective, e.g., an employer might define a voluntary move as one that occurred in anticipation of a lay-off or termination. Nonetheless, looking at the degree of choice that the individual had over ending a job is an important piece of information that reflects the extent to which an individual is in charge of his employment situation or at the mercy of an employer. There are other dimensions of job changing that are important. Some voluntary job exits are followed by upward mobility in wages or prestige while other job changes are followed by no employment or downward mobility. Voluntary job changes that appear to result in downward mobility in wages may, however, actually be made because other features of the new employment situation may be better than those of the old job and thus on balance the move results in an improved job. Or, a voluntary move to a job with lower or similar wages may actually reflect a move to an employer who provides more job security or opportunities for advancement in the future. Our research investigates the determinants of both voluntary and involuntary movement and thus illuminates one of the mechanisms that generate inequality in career paths among young men from different racial/ethnic groups. The NLSY79 contains relatively large number of cases of both voluntary and involuntary job mobility that enable us to look at both types of mobility.

More specifically, we look at racial/ethnic differences in the patterns of moving out of one’s first job held after the completion of education. We limit ourselves to considering only movements out of relatively stable jobs that continue at least 6 months.3 We want to compare the rates of both involuntary and voluntary mobility among young men who have completed their schooling and have become attached to and engaged in the labor market. Previous studies have shown that jobs held while in school are often qualitatively different from jobs undertaken after entering the full-time labor force. Therefore, we may find very different mobility patterns if we combine the two types of jobs than if we focus only on jobs after the completion of schooling (Donohue, 1988; Tuma and Hannan, 1984).

It is possible that there is much less racial/ethnic difference in the rates of job moves once workers achieve stable employment, even though a certain group might have more severe job instability before they have settled into stable jobs. If this is the case, policy efforts should pay most attention to creating stable employment early in the work career for members of disadvantaged groups. However, if significant inequality with respect to job stability remains even after people have been in jobs for several months, a more complicated and broader approach is required to address labor market stability.

In the section that follows, we briefly discuss various determinants of job mobility based upon previous studies estimating the effects of those factors on either voluntary or involuntary movements. Then we review previous empirical studies addressing the issue of racial/ethnic differences in the probability of leaving jobs and draw some hypotheses for our current research. After we introduce our statistical models for estimating the rates of job mobility and the data we show the empirical results. Finally, we conclude our discussion with a summary of major findings and some suggestions for further research.

Section snippets

A theoretical model of voluntary and involuntary mobility

Past work on job mobility has relied primarily on the reward–resource model for explaining why individuals leave jobs voluntarily or involuntarily. This model assumes that individuals make voluntary moves in order to improve their wages immediately or for an increased opportunity of future advancement with another employer, while involuntary moves are not the choice of the individual. The model suggests that there are a number of reasons for why black, Mexican, other Hispanic, and white young

Using the reward/resource model to explore racial and ethnic differences in job mobility

Our discussion of the reward–resource model of job mobility may be used to draw some hypotheses about racial/ethnic differences in the likelihood of leaving jobs voluntarily or involuntarily. This model suggests that whites who enjoy relative advantages in individual resources like education will be more likely than members of minority groups to leave their present jobs voluntarily. In addition, their greater human capital also protects whites from being fired or laid off and thus lowers

Methods

In order to examine the rates of exiting the first job, we employ hazard models, which estimate the log of job mobility rates as a linear function of a vector of explanatory variables and duration terms. Rather than choosing a specific parametric form for the transition rate, we utilize the piecewise exponential model, which imposes the fewest shape assumptions on the baseline distribution. As our discussion in the previous section shows, there is disagreement on the specific parametric form of

Data and sample

The data used in this analysis come from the National Longitudinal Survey of Youth collected between 1979 and 1994. The NLSY is a nationally representative sample of 12,686 young males and females who were 14–21 years of age at the end of 1978. The respondents were first interviewed in 1979, and they have been re-interviewed each year up to 1994 and every two years after that. Specially designed to provide information on labor force behavior and education, the data contain the detailed work

A comparison of survivor functions

The product-limit (or Kaplan–Meier) estimation method of the survivor function provides a general description of the process under study and is useful for graphically comparing survivor functions among two or more groups (Blossfeld and Rohwer, 1995). The survivor functions show the proportions of workers who still remain in the same job over time in the job. Before entering into a discussion of the effects of individual, job, and local labor market characteristics on job mobility, we describe

Determinants of job mobility

We now turn to multivariate analyses of both voluntary and involuntary job moves, based on the hazard models. We apply both the piecewise exponential model and the log-logistic model to our data. Table 2 presents the maximum likelihood estimates of the covariates from the two models for the rates of involuntary and voluntary moves, respectively. It is evident that in both types of mobility, the estimates of covariates are very similar between the piecewise exponential model and log-logistic

Racial and ethnic differences after controls

Now let’s turn to our concern with racial and ethnic differences in job mobility. In the previous analysis of survivor functions, we identified considerable racial/ethnic differences in the rates of involuntary moves, while there was no marked difference in the rates of voluntary moves. Now we examine the extent to which such observed differences in the likelihood of involuntary moves among the four groups are altered after controlling for racial/ethnic differences in various individual, job,

How do these factors help account for the group differences in job mobility?

The results in Table 2 show the remaining racial/ethnic differences in involuntary moves after observed differences in characteristics are controlled. We are also interested in exploring the extent to which different covariates help account for the observed racial and ethnic differences. Table 3 reveals the contribution of different sets of variables to explaining the relationship between race/ethnicity and mobility. In Model 1 we include only race, age, and duration as covariates. The results

Summary and discussion

In this paper, we hoped to extend previous studies of racial/ethnic differences in job mobility patterns in two important ways: first, we examined the job shift behaviors of Mexicans and other Hispanics as well as blacks and whites. Second, we investigated racial/ethnic differences in voluntary and involuntary moves. Our empirical findings highlight the importance of looking at racial/ethnic differences in different dimensions of career development. Minority disadvantages in the labor market

Acknowledgements

Research on this paper was supported by a grant (U01-HD37566) from the National Institute of Child Health and Human Development to Gary D. Sandefur. A version of this paper was presented at the 2001 American Sociological Association Annual Meeting. We thank Steven T. Cook for his helpful advice on computer programming.

References (40)

  • D.S. Hachen

    Gender differences in job mobility rates in the United States

    Social Science Research

    (1988)
  • S. Hsueh et al.

    Gender, ethnicity and labor force instability

    Social Science Research

    (1996)
  • P.D. Allison

    Event History Analysis: Regression for Longitudinal Event Data

    (1984)
  • F.D. Blau et al.

    Race and sex differences in quits by young workers

    Industrial and Labor Relations Review

    (1981)
  • H.-P. Blossfeld et al.

    Techniques of Event History Modeling

    (1995)
  • A.L. Booth et al.

    Job tenure and job mobility in Britain

    Industrial and Labor Relations Review

    (1999)
  • R. Breen

    Job changing and job loss in the Irish youth labour-market: a test of a general model

    European Sociological Review

    (1992)
  • Bureau of Labor Statistics, 1993. Work and family: turning thirty-job mobility and labor market attachment. Report...
  • Center for Human Resource Research (CHRR) 1999. NLSY79 User’s Guide. The Ohio State...
  • J.J. Donohue

    Determinants of job turnover of young men and women in the United States

    Research in Population Economics

    (1988)
  • J.R. Elliott

    Class, race, and job matching in contemporary urban labor markets

    Social Science Quarterly

    (2000)
  • Farber, H.S. 1992. Evaluating competing theories of worker mobility. National Bureau of Economic Research Discussion...
  • R. Ferguson et al.

    Do better jobs make better workers? Absenteeism from work among inner-city black youths

  • D.S. Hachen

    Three models of job mobility in labor markets

    Work and Occupations

    (1990)
  • M.D. Hayward et al.

    The significance of socioeconomic status in explaining the racial gap in chronic health conditions

    American Sociological Review

    (2000)
  • Holzer, H., LaLonde, R.J., 1999. Job change and job stability among less-skilled young workers. Paper Presented for the...
  • S. Hsueh et al.

    Earnings consequences of employment instability among minority men

    Research in Social Stratification and Mobility

    (1995)
  • P. Jackson et al.

    Layoffs, discharges and youth unemployment

  • A.L. Kalleberg

    Nonstandard employment relations: part-time, temporary and contract work

    Annual Review of Sociology

    (2000)
  • A.L. Kalleberg et al.

    Organizational size, layoffs, and quits in Norway

    Social Forces

    (1998)
  • Cited by (17)

    • Men's mobility into management from blue collar and white collar jobs: Race differences across the early work-career

      2014, Social Science Research
      Citation Excerpt :

      The “particularistic mobility thesis” maintains that across the early career years there are race-specific prospects for mobility into management when tracked from similar white collar and blue collar jobs. This perspective distills studies with a structural and social psychological bent over the last several decades regarding the race-based recruitment and promotion practices of managers and executives primarily within White owned and managed workplaces (e.g. Pettigrew 1985; Dobbin 2009; Smith 2012; Smith, 2005; Smith, 2002; Sheriden 1997; Hite 2007; Baldi and McBrier, 1998; Powell and Butterfield 1994; Roscigno 2007; Federal Glass Ceiling Commission 1995; Tsui and O’Reilly, 1989; Smith and Elliott 2002; Wilson et al., 1999; Maume 1999; Castilla, 2008; Park and Sandefur 2003; Mundra et al., 1995; Tomkiewitz and Brenner, 1996). Overall, encompassing case studies, treatises, and survey-based analyses, approximately one-fourth of the studies comprising this perspective reference, but do not explicitly examine early career dynamics or conduct job-level analyses.

    • Changes in precarious employment in the united states: A longitudinal analysis

      2021, Scandinavian Journal of Work, Environment and Health
    View all citing articles on Scopus
    View full text