Academic journal article
By Bernasek, Alexandra; Kinnear, Douglas
Journal of Economic Issues , Vol. 33, No. 2
The purpose of this paper is to explore the notion of workers' willingness to accept contingent employment. Contingent jobs are defined broadly as jobs that will not last indefinitely. Given the opposing views of these employment arrangements - on one hand, it is argued that they benefit workers by giving them greater flexibility, and on the other hand, it is argued that they hurt workers by subjecting them to greater insecurity and provide fewer job benefits - it seems likely that contingent arrangements are beneficial for some workers but not for others. The relevant questions, then, are which workers are helped, and which are hurt, by contingent employment? One way to investigate this is to compare workers who willingly accept contingent jobs with those who hold such jobs but would prefer permanent employment. If the two groups of workers are found to differ significantly in their personal and job characteristics, then there may be some basis for economic policies to assist workers who would like to move out of contingent jobs into permanent jobs. We are able to explore this issue of workers' willingness to accept contingent employment using data from the February 1995 Contingent Worker Supplement to the Current Population Survey.
Literature on Contingent Employment
While some researchers employ their own definitions, the Bureau of Labor Statistics (BLS) has provided a taxonomy for measuring contingency [Polivka 1996a, 4-6]. The BLS provides three different definitions of contingent employment; the present paper utilizes definition three, which is the broadest, as this encompasses most researchers' definitions of contingent employment and also allows us to use the largest number of observations from the data set. Definition three includes all workers who do not expect their jobs to last indefinitely, with the qualification that among the self-employed and independent contractors, only those who had been in their current arrangements for one year or less and expected them to last less than another year or less were included.
Much of the existing research on contingent labor provides descriptive statistics outlining the numbers and characteristics of workers in contingent arrangements. Research suggests that minorities, women, less educated, and young workers are disproportionately represented in contingent jobs. Using the same data set as the current paper, Anne Polivka [1996b, 11-12] found that women, blacks, and Hispanics were holding slightly higher proportions of contingent jobs than noncontingent jobs. It was also determined that workers between the ages of 16 and 24 were highly concentrated in contingent jobs: under definition three, workers in this age range occupied 30.5 percent of all contingent jobs but only 13.9 percent of noncontingent jobs [Polivka 1996b, 11-12]. Educational attainment was also correlated to contingent employment: workers lacking a high school diploma held 14.6 percent of contingent jobs under estimate three and only 10.5 percent of noncontingent jobs, although only 17.5 percent of contingent workers and 18.2 percent of noncontingent workers were college graduates. These findings on race, gender, age, and education have been generally confirmed by the Economic Policy Institute and the Women's Research and Education Institute [1997, 42-57], as they relate to "nonstandard work" arrangements,(1) and by Sharon Cohany [1996, 33-35], who found a correlation between these demographic characteristics and the proportions of workers in lower-quality contingent arrangements such as temporary work.
There has been relatively little empirical research attempting to explain why certain groups are disproportionately represented in contingent arrangements. We use a logit model to estimate the effects of contingent workers' personal and job characteristics on their probability of preferring noncontingent employment. This research is relevant for several reasons. It helps ascertain which workers prefer such arrangements and which may accept them due to a lack of alternatives. …