Content provided in partnership with
Business Services Industry
Reemployment differences among dislocated and other workers
American Journal of Economics and Sociology, The, Jan, 1997 by Mary Ellen Benedict, Peter VanderHart
How do They Adapt to Job Losses?
I
Introduction
There are many reasons why workers lose their jobs, but the one that has raised national concern during the past two decades has been worker dislocation. The federal definition classifies a worker as dislocated if that person has permanently lost a job held for at least three years due to slack work, abolition of the job, or closing of the plant (Flaim and Sehgal, 1985). In the survey years of 1984, 1986, 1988 and 1990, the federal government classified about 10 million workers as dislocated (Podgursky, 1992). These large numbers, combined with concern over future changes in product demand, global competition and technological advancement, have made worker dislocation of great interest to policy-makers.
- Most Popular Articles in News
- The Ten Best Laptop bags
- Tata plans cheapest-ever car for Indian market
- GLOBALIZATION AND THE DEVELOPMENT OF UNDERDEVELOPMENT OF THE THIRD WORLD
- Corn is good for you; Corn is not only a tasty treat, but also a cereal that ...
- THE 50 BEST STYLISH HANDBAGS TO CARRY
- More »
Worker dislocation would not be so troublesome if those displaced could find new jobs quickly and at an equivalent rate of compensation, particularly if the new jobs were in stable or growing industries. However, it is not clear whether these growth jobs are matched to dislocated workers or to other types, such as workers new to the workforce or those who voluntarily quit a job. If workers displaced from declining industries are unable or unwilling to move into stable or growth industries, or if these jobs do not compensate as well as the previous jobs did, then worker dislocation can result in high costs for the displaced individual (high search costs and loss of industry-specific skills) and possibly for society (the net loss of productivity due to lost human capital and the cost of publicly funded programs which aid the unemployed).
This research will address these issues by investigating the following question: What are the reemployment propensities of dislocated workers and how do they compare with those of others in the labor force? This study uses a multi-sector empirical model which includes five potential outcomes: remaining unemployed, exiting the labor force, or taking a job in one of three types of industries (declining, stable, or growth). A multinomial logit regression with year-to-year labor market transition data is employed to examine movement from unemployment to the other labor market states. Data from the Panel Study of Income Dynamics (PSID) are used because it contains work history information on individuals over many years and allows for comparisons between those who nave been dislocated and other groups of the unemployed.
II
Evidence of Worker Adjustment After Dislocation
The worker dislocation literature focuses primarily on the relationship between dislocation and economic loss. Many researchers have examined the size of earnings losses due to dislocation and the relative magnitude of these losses across various subgroups.(1) Only recently have researchers examined the impact of dislocation on reemployment. Ruhm (1991) uses PSID data for 1971-75 to find that while dislocated workers experience only one more week of unemployment than their nondislocated counterparts, the dislocated workers experience a lasting wage reduction due to attachment to specific jobs. Using various statistical techniques on data from the Dislocated Worker Survey (DWS) supplements to the Current Population Survey (CPS), Kletzer (1992, 1996), Gibbons and Katz (1992) and Neal (1995) find that when dislocated workers incur economic losses it is in large part due to the workers' lack of industrial mobility. This literature suggests that sector-specific skills play a crucial role - workers reemployed in the same or similar industries suffer lower losses than the dislocated who switch industries.
Fallick (1993) examines the mobility of dislocated workers more explicitly by estimating transition rates of such workers from unemployment in their last industry to employment in a new industry. His work suggests dislocated workers tend to seek reemployment in their pre-displacement industries. Reemployment probabilities estimated by Seitchik and Zornitsky (1989) further support this result. Their estimates suggest that moving from declining to growth jobs lowers the probability of reemployment for dislocated workers.
The results using the DWS pertain only to dislocated workers, and thus tell us nothing about how they compare to other types of the unemployed. As part of a larger study, Farber (1993) uses data from five full CPSs to test for differences in reemployment outcomes between dislocated and nondislocated workers. Using a binomial logit model to estimate reemployment probabilities, he finds that dislocation in the year before the survey date lowers the reemployment probability substantially (25% for men, 15% for women), and that education and race are significant regardless of dislocation status. However his data have two limitations. First, his nondislocated group includes those who do not work at all and workers who do not experience any unemployment, as well as the nondislocated unemployed. This group therefore is an overly broad group to compare to dislocated workers, who experience unemployment almost by definition.(2) A second problem occurs because the CPS lacks tenure information so that the distinct effects of dislocation and sector-specific skills are not clearly separated.
