Marginal Employment

views updated



In industrialized countries the principal way in which adults and their families make ends meet economically is through employment in the formal labor force. In the United States, earnings in the form of wages and salaries account for 71 percent of household income, on average. Even among individuals in poor families, half (49 percent) of all household income comes from earnings—their single most important source of income (Current Population Survey 1998). The centrality of work for economic well-being also is reflected in the close connection between finding gainful employment and exiting poverty on the one hand, and losing employment and sliding into poverty on the other. While formal employment is a necessary condition for economic well-being among almost all nonwealthy, able-bodied adults and their families, it is not a sufficient condition. Not all jobs pay well, many come without health and other benefits that workers must either pay for through their wages or do without, some are merely part-time, and some jobs are highly unstable and offer little long-term security. Marginal employment can be conceptually defined as the circumstance in which the formal employment of adults (or groups of adults within families) fails to generate the earnings needed to achieve a minimally acceptable standard of living, either because they work too few hours (insufficient labor supply), and/or because their wages are too low.

Two trends motivate increasing concern over marginal employment. The first is a shift in social welfare systems which, as in the United States, have come to place greater emphasis on formal employment as the route out of poverty and reliance on the state. In recent years industrialized nations have reoriented their social welfare programs to encourage employment and work preparedness (Cornea and Danziger 1997). In the United States, the Personal Responsibility and Work Opportunity Reconciliation Act of 1996 has placed strict time limits on Public Assistance receipt. The image of thrusting people off the welfare rolls to fend for themselves in the open labor market has generated deep concern over their employment prospects. Second, in recent decades the structure of employment opportunities has changed in ways detrimental to low-skilled workers. Whether due to changing technology (e.g., computerization), which is displacing labor; to the globalization of production and trade, which is leading to outsourcing of low-skilled work; to competition through increased immigration; or to changes in wage-setting institutions (e.g., declines in unionization or real minimum wages), lowskilled workers have slipped further behind their better-skilled counterparts in terms of wages and total compensation (Freeman and Gottschalk 1998). The point is, just when many governments are foisting more responsibility for the economic well-being of citizens on the labor market and workers themselves, questions are being raised about the implications of macroeconomic changes for the quality of jobs available, especially for low-skilled groups. As such, marginal employment—its severity, causes, and consequences—has become an important concern for sociologists and policy makers alike.


Marginal employment per se is not as yet a conventional concept in sociology and thus there are no standard empirical definitions of this term. As noted, we conceptualize it as consisting of both the degree of attachment to the labor force and the quality of jobs. A composite measure that incorporates both these dimensions is underemployment. Drawing on the Labor Utilization Framework of Phillip Hauser (1974), Clifford Clogg (1979) developed a measure of underemployment that consists of the following categories. The working poor are those people who are working full-time, full-year, but whose labor market earnings are less than 125 percent of the poverty threshold for single individuals. Involuntary part-time workers include those who are working fewer than full-time hours (thirtyfive-hours per week) only because they are unable to find full-time work. The unemployed are those who are not currently working and are either actively looking for work or on layoff. (The unemployment rate—a standard aggregate measure of the health of the economy—is simply calculated as the number unemployed divided by the number in the labor force, where the latter is the sum of those employed and those unemployed.) Finally, discouraged workers (sometimes referred to as the subunemployed) are neither working nor looking for work, but would nonetheless like to be working if they could find a job. All other adults, those who are not working and do not want to be, are considered not in the labor force and are excluded from analyses of underemployment. It is important to recognize that this operationalization of underemployment is an imperfect empirical measure of marginal employment as conceptually defined above. Discouraged or involuntary part-time workers, for example, could be in families that are otherwise well off economically. While this caveat needs to be borne in mind, the statistical patterns described below suggest underemployment is a reasonable measure of marginal employment.

Table 1 shows the distribution of underemployment and its types for the U.S. population as a whole, and for important demographic groups. The data are from the March 1998 Current Population Survey, a nationally representative survey of 64,659 U.S. households and a key source of official employment data (Current Population Survey 1998). In 1998, of all U.S. adults aged 18–64 who either were working or would like to have been working, 18.0 percent were underemployed. We stress that by this definition, over one in six U.S. workers or would-be workers are marginally employed. Of these, 4.7 percent were unemployed, 6.3 percent were working part-time but wanted full-time work, 6.1 percent were working full-time but for near- or below-poverty wages, and 0.8 percent had given up in the search for work, despite wanting a job.

Underemployment is not distributed equally across demographic groups, but parallels well-known correlates of economic well-being generally. At over one-third, the young (those aged 18–24) are far more likely to be underemployed than older adults. A more refined age breakdown for all adult age groups (not shown) suggests a curvilinear pattern, with young adults and elderly Americans at greater risk of underemployment than those in their prime working ages. Women have long been disadvantaged in the U.S. labor market with respect to wages and occupational prestige, so it is little surprise that there also is gender inequality in underemployment. In relative terms,

Table 1
Percentage distribution of underemployment by selected characteristics
 underemployedadequately employed
 totallow incomelow hoursunemployeddiscouraged 
note: percentage distribution of underemployment by selected characteristics.
source: original calculations from the march 1998 current population survey.
non-hispanic white15.
non-hispanic black25.
native american19.
marital status
never married17.
immigrant status
third (or higher) generation17.
second generation18.
first generation24.
less than high school36.3%12.2%11.5%10.5%2.2%63.7%
high school or ged20.
some college16.
college degree or more7.
central city20.
non-central city15.
not identified18.

women are 14 percent (2.3 percentage points) more likely than men to be underemployed, a difference that is due primarily to their greater risk of being working poor. Blacks, Hispanics and, to a lesser degree, Native Americans, are distinctly disadvantaged with respect to non-Hispanic whites. Indeed, about one-fifth of Native Americans, and over one-quarter of blacks and Hispanics, are underemployed, which compares to only 15 percent of non-Hispanic whites. Being married carries an advantage, while being divorced, separated, or widowed increases the risk of underemployment. With respect to immigrant status, first-generation adults (immigrants themselves), are far more likely than the native-born to be underemployed, and among natives, those with a foreign-born parent (the second generation), are slightly disadvantaged compared to those with native-born parents (the third and higher generations). (This pattern should not be taken as direct evidence of economic assimilation, however, given immigrant-cohort differences in factors that determine the risk of underemployment.) The rising wage inequality of the 1980s and 1990s is due largely to deteriorating conditions among low-skilled workers (Freeman and Gottschalk 1998). Indeed, as shown in Table 1, education is strongly associated with underemployment. Whereas 36.3 percent of those with less than a high school education are underemployed, only 7.5 percent of college graduates are similarly situated. Finally, the risk of underemployment is not distributed evenly across space. Despite legitimate and well-chronicled concern over declining employment opportunities in the inner city (Wilson 1996), at 21.6 percent the prevalence of underemployment is higher in nonmetropolitan (nonmetro) counties of the United States, than in metropolitan areas. Indeed, the nonmetro risk is even higher than that for those living in central cities of metro areas (20.5 percent). Thus, marginal employment is both an inner city and a rural problem.


Underemployment and its component parts are countercyclical. They rise in prevalence when the state of the wider economy declines, and they fall as the economy heats up. For example, above we noted the underemployment rate in 1998—a relatively good year economically—was 18.0 percent. In 1968, another good year, it stood at 15.2 percent. Amid recession in 1983, by comparison, 24.3 percent were underemployed. The rise and fall in the prevalence of marginal employment over time reflects the dynamic nature of labor markets. In large numbers individuals are constantly moving into and out of the labor force, between part-time and full-time work, and between good jobs and bad jobs.

While cross-sectional studies are useful for describing the broad contours of marginal employment and identifying sociodemographic groups that are more vulnerable, sociologists have begun to turn their attention to the nature and determinants of the movement between labor-force states. In an analysis of data from the Current Population Surveys of 1968 through 1993, Jensen and colleagues (1999) explored the determinants of transitions between adequate employment and underemployment. They found that adequately employed adults are more likely to slide into underemployment from one year to the next if they are black or Hispanic, female, less well educated, or young, or if they live in a nonmetropolitan area. The risk of becoming underemployed also was higher amid a souring economy, and lower when the economy was strengthening. Once underemployed, these same groups were found to be disadvantaged with respect to their likelihood of finding adequate employment. Similarly, using data from the Survey of Income and Program Participation, Hsueh and Tienda (1996) explored transitions between being employed, unemployed, and out of the labor force altogether. They document greater labor-force instability among women, blacks and Hispanics, young adults, those with only a high school degree or less, and those who are not married.

That minorities and women, particularly those who are less well educated, have greater labor-force instability is important for understanding persisting gender and racial or ethnic inequality in society. The checkered work history that comes from frequent transitions into and out of the labor force means less opportunity to gain valuable work experience, and may signal to prospective employers a questionable commitment to work (Tienda and Stier 1996). The negative implications of work instability can accumulate over time, darkening future labor market outcomes and increasing the likelihood of marginal employment (Clogg et al. 1990).


Even a stable work history is no guarantee against future employment hardship. An important determinant of labor-force instability is job displacement. Displaced workers are often technically defined as "persons 20 years and older who lost or left a job [which they had held for three or more years] because their plant or company closed or moved, there was insufficient work for them to do, or their positions or shifts were abolished" (Hipple 1997, p. 38). Sometimes referred to as "structurally unemployed" (because their unemployment is due to structural rather than cyclical shifts in the economy), displaced workers have little prospect of getting their old jobs back: Their job loss is permanent. To get a sense of the magnitude of the problem, consider that between 1993 and 1995, 15 percent of U.S. workers experienced job displacement (Kletzer 1998). The underlying causes of displacement are not well documented, but are likely rooted in technological change, rising foreign competition, and declining productivity within industries (Kletzer 1998). More is known about differences across sociodemographic groups in the risk of displacement. Recent evidence indicates that those who are black, less well educated, and employed in production occupations (as craftsmen, operatives, or laborers) and in manufacturing industries have higher rates of displacement (Kletzer 1998). Once displaced, there also is inequality in the prospects of finding new work after displacement. Here it helps to be white, male, and better educated.


Also contributing to volatility of labor-force attachment is the emergence of contingent labor and temporary workers. In the 1970s, international competition had begun to challenge U.S. industries, marking an end to two decades of economic expansion and growing corporate profits. In response U.S. firms began to restructure in order to maintain their competitive edge. By the mid-1980s, terms such as "mergers," "acquisitions," "deindustrialization," and "downsizing" had become commonplace (Harrison and Bluestone 1988). In the process, the traditional relationship between employer and employee began to change. Today, for a growing percentage of U.S. workers the postwar model of relatively stable and secure full-time employment with fringe benefits, steady wages, and the opportunity, if not expectation, of internal advancement within a firm, is no longer a reality (Henson 1996). Instead an increasing proportion of the American workforce is now engaged in contingent work.

To date there is no conventional definition of contingent work. The term has been used broadly in the literature to refer to a range of nonstandard employment practices including "part-time work, work performed by independent contractors and on-call workers, and work done by temporary workers, hired either directly for limited-duration projects or through temporary help firms" (Blank 1998, p. 258). In 1989, the Bureau of Labor Statistics (BLS) developed the following definition: "Contingent work is any job in which an individual does not have an explicit or implicit contract for long-term employment"(Polivka 1996, p. 3). In sum, despite disagreements over specifics, contingent work usually refers to nonstandard employment relationships that are associated with temporary or insecure employment duration, inadequate wages and benefits, and a conditional relationship that limits the attachment between employer and employee. Compared to permanent employees, contingent workers earn less and are less likely to be covered by employer health insurance or pension plans (Cohany et al. 1998; Hipple 1996). Depending upon the definition used, contingent workers composed between 2.2 and 16 percent of the labor force in 1995 (Barker and Christensen 1998).

The use of contingent workers has grown rapidly in recent decades. This growth has been driven primarily by a combination of employee desires (supply-side explanations) and employer strategies (demand-side explanations). On the supply side, contingent work often has features that some workers prefer, such as limited and flexible hours, independence, and variety. And some workers use contingent work as a way to shop around for good jobs and employers, in the hopes of obtaining more permanent work (Blank 1998). However, the idea that the rise in contingent work is being driven by worker preferences should not go too far since, by one estimate, two-thirds of contingent workers would prefer permanent work (Cohany et al. 1998). Demand-side factors include fluctuations in product demand which give rise to the need for flexibility in hiring and firing, reduced labor costs, and the ability to screen employees for permanent, full-time work (Blank 1998). Thus, in response to the growth of global competition and increasingly volatile consumer demand, employers have used contingent workers to shift the risk of market uncertainties from the firm to the worker. By reducing the obligation of the employer to the employee, firms are able to be more flexible in adjusting labor inputs and reducing labor costs. One of the primary reasons that firms have been able to cut costs at the expense of labor is the transition in the balance of power between labor and management in recent decades (Golden and Applebaum 1992). As labor unions have gradually lost bargaining power, management has increasingly been able to exercise cost cutting strategies at the expense of workers.

A subgroup of contingent workers that appear to be particularly marginalized are temporary help workers or "temps." Temps are typically employees of temporary help supply (THS) firms, which provide workers to other businesses on a contractual basis (Ofstead 1998). In most cases, THS firms remain the worker's legal employer while the worker provides services to another establishment. THS firms emerged as labor market actors in the 1940s, but have seen tremendous growth in recent decades. Research by the Bureau of Labor Statistics shows that the number of temporary jobs filled daily by THS firms has increased more than two and a half times since 1982 (Golden and Applebaum 1992). Further, research has found that over 90 percent of employers use temporary workers and that more than five million workers seek employment through THS firms in a given year (Golden and Applebaum 1992). THS firms typically specialize in worker placement in one of four occupational areas: office, industrial, medical, or engineering and technical, with office workers comprising over half of all temporary workers (Carey and Hazelbaker 1986).

Temporary workers are subject to an earnings penalty in comparison to permanent workers. Research has shown that temporary workers typically earn between 80 and 52 percent of the wage earned by their permanent counterparts (Barker and Christensen 1998). Further, research shows that the temporary workforce is disproportionately composed of individuals that face disadvantage in the labor force in general: women, minorities, and the young (Barker and Christensen 1998; Howe 1986). The predominance of women and the young among temporary workers is likely due in part to the flexibility allowed by the employment relationship. However, limited alternative employment opportunities and employer perceptions of what types of workers prove to be risky investments as permanent employees, also may be at play.

As it continues, the growth of contingent labor and proliferation of temporary workers will likely contribute significantly to the segmentation and bifurcation of the labor force. Permanent workers are able to realize higher wages, receive fringe benefits, receive on-the-job training, and have opportunities to climb the occupational ladder. On the other hand, contingent workers earn lower wages; receive few, if any, benefits; and enjoy little opportunity for advancement. In the case of those who engage in contingent employment because of personal preference, this may not present a problem. However, that many of those engaged in contingent work do so involuntarily is of concern. There is potential for the expansion of a fringe class of workers disenfranchised and marginalized from the mainstream labor force. At the very least, the growing use of contingent workers represents a fundamental shift in the relationship between employer and employee.

To summarize, marginal employment is a concept of increasing importance for sociologists. Structural economic shifts that are giving rise to job displacement and deterioration in the employment circumstances of low-skilled workers, combined with the increasing tendency for industrialized countries to link government help with work and/or work preparedness, have thrown a spotlight of concern on those most vulnerable to employment hardship. In addition to simply documenting the nature and trends in marginal employment, sociologists have begun to adopt more dynamic empirical approaches to model the movement of workers into and out of various states of marginal employment. All this work points to significant and persisting difficulties faced by women, racial and ethnic minorities, immigrants, young adults and elders, those with low achieved human capital, those who are unmarried, and rural or inner-city residents. Sociologists have also begun to explore the institutional changes that have contributed to the persistence of marginal employment and its unequal distribution across sociodemographic groups. Key among these is the rise in contingent work which is characterized by part-time and temporary employment arrangements, often at lower wages and lacking in the fringe benefits enjoyed by permanent workers.


Barker, Kathleen, and Kathleen Christensen 1998 Contingent Work: American Employment Relations in Transition. Ithaca, N.Y.: Cornell University Press.

Blank, Rebecca M. 1998 "Contingent Work in a Changing Labor Market." In Richard B. Freeman and Peter Gottschalk, eds., Generating Jobs: How to Increase Demand for Less-Skilled Workers. New York: Russell Sage Foundation.

Carey, Max L., and Kim L. Hazelbaker 1986 "Employment Growth in the Temporary Help Industry." Monthly Labor Review 109(4):37–43.

Clogg, C. C. 1979 Measuring Underemployment: Demographic Indicators for the United States. New York: Academic.

——, Scott R. Eliason, and Robert Wahl 1990 "Labor Market Experiences and Labor Force Outcomes." American Journal of Sociology 95:1536–1576.

Cohany, Sharon R., Steven F. Hipple, Thomas J. Nardone, Anne E. Polivka, and Jay C. Stewart 1998 "Counting the Workers: Results of a First Survey." In Kathleen Barker and Kathleen Christensen, eds., Contingent Work: American Employment Relations in Transition. Ithaca, N.Y.: Cornell University Press.

Cornea, Andrea Giovanni, and Sheldon Danziger 1997 Child Poverty and Deprivation in the Industrialized Countries, 1945–1995. Oxford: Clarendon.

Current Population Survey 1998 Conducted by the Bureau of the Census for the Bureau of Labor Statistics. Washington, D.C.: U.S. Bureau of the Census, March, 1998. (Machine-readable data file.)

Freeman, Richard B., and Peter Gottschalk 1998 Generating Jobs: How to Increase Demand for Less-Skilled Workers. New York: Russell Sage Foundation.

Golden, Lonnie, and Eileen Applebaum 1992 "What Was Driving the 1982–99 Boom in Temporary Employment?" American Journal of Economics and Sociology 51(4):473–493.

Harrison, Bennett, and Barry Bluestone 1988 The Great U-Turn: Corporate Restructuring and the Polarizing of America. New York: Basic.

Hauser, Philip M. 1974 "The Measurement of Labor Utilization." Malayan Economic Review 19:1–17.

Henson, Kevin D. 1996 Just a Temp. Philadelphia: Temple University Press.

Hipple, Steven F. 1996 "Earnings and Benefits of Contingent and Noncontingent Workers." Monthly Labor Review 119(10):22–30.

——1997 "Worker Displacement in an Expanding Economy." Monthly Labor Review (December):26–39.

Howe, Wayne J. 1986 "Temporary Help Workers: Who They Are, What Jobs They Hold." Monthly Labor Review 109(11):45–47.

Hsueh, Sheri, and Marta Tienda 1996 "Gender, Ethnicity, and Labor Force Instability." Social Science Research 25:73–94.

Jensen, Leif, Jill L. Findeis, Wan-Ling Hsu, and Jason P. Schachter 1999 "Slipping Into and Out of Underemployment: Another Disadvantage for Nonmetro Workers?" Rural Sociology 64(3):417–438.

Kletzer, Lori G. 1998 "Job Displacement." Journal of Economic Perspectives 12(1):115–136.

Ofstead, Cynthia 1998 Temporary Help Firms and the Job Matching Process. Unpublished doctoral dissertation. Department of Sociology, University of Wisconsin, Madison.

Polivka, Anne E. 1996 "Contingent and Alternative Work Arrangements Defined." Monthly Labor Review 119(10):3–9.

Tienda, Marta, and Haya Stier 1996 "Generating Labor Market Inequality: Employment Opportunities and the Accumulation of Disadvantage." Social Problems 43(2):147–165.

Wilson, William J. 1996 When Work Disappears: The World of the New Urban Poor. New York: Knopf.

Leif Jensen

Tim Slack