Job Queues, Discrimination, and Affirmative Action
Bisping, Timothy O., Fain, James R., Economic Inquiry
JAMES R. FAIN [*]
If employers have different Becker-type discrimination coefficients for different demographic groups, then the implementation of affirmative action may have a differential impact on those groups. We estimate two vector autoregressive models of the US. economy, including the unemployment rates of four demographic groups. We find that a job queue existed before the implementation of affirmative action and that affirmative action changed the ordering of the job queue in manner that had a negative impact on nonwhite males. We find evidence that affirmative action may have increased the unemployment rate of nonwhite males by increasing their duration of unemployment. (JEL J71, J78, J64, J68)
The relative status of demographic groups in the labor force has been undergoing significant change since the early 1960s. Throughout this period white females and racial minorities have experienced changes not only relative to white males but also relative to one another. One such change is exhibited in relative unemployment rates.
Figure 1 shows differentials in the smoothed unemployment rates (using a 12 X 2 centered moving average) for white females, nonwhite males, and nonwhite females for the years 1959--83. One can easily see that the relative unemployment burdens have varied over time. Focusing first on the two lines that are positive over the entire range, one may see that prior to the late 1960s the status of nonwhite individuals, especially nonwhite males, relative to white females was improving. However, beginning in the late 1960s, this trend reversed. The unemployment rate differential between nonwhite males and white females increased from a low of 1.03 in July 1969 to a high of 11.12 in March 1983. Similarly, after declining during the 1960s, the unemployment rate differential between nonwhite females and white females increased from a low of 3.49 in October 1969 to a high of 9.16 in August 1983. These reversals in trend indicate the status of white females relative to both nonwhite males and nonwhite females began improving in the late 1960s. Figure 1 also shows the status of nonwhite females improved relative to that of nonwhite males during the same time period: the unemployment rate differential between nonwhite males and nonwhite females increased from a low of - 3.13 in February 1968 to a high of 2.38 in February 1983.
These three unemployment rate differentials all declined in the 1960s, a time of tremendous social change and strong economic growth, but all of these trends reversed direction in the 20-month period from February 1968 to October 1969. After the trend reversal, the status of white females improved relative to both nonwhite males and nonwhite females, and the status of nonwhite females improved relative to that of nonwhite males. To put it another way, nonwhite males lost ground relative to both nonwhite females and white females. Our goal is to examine what role, if any, the implementation of affirmative action in 1968 had in reversing these trends.
Other possible explanations of the above-noted reversal in trends certainly exist, but several factors point to affirmative action as a potential cause of these changes. First, the timing is correct: the trends reversed in close proximity to the implementation of affirmative action. Second, whatever caused these reversals affected all minority groups, did so within a short (20-month) time span, and its impact was sustained over a number of years. A new Federal policy designed to change the way firms hire minorities should have exactly this type of effect. Finally, as we will argue below, imposing affirmative action on discriminating employers will not benefit all demographic groups equally: discriminating employers will continue to discriminate, though the pattern of discrimination may change.
Many of the traditional explanations of changes in relative unemployment rates are unlikely candidates to explain these significant and sustained reversals of longstanding trends. …