Based on this literature, the current paper examines the impacts of job skill types (cognitive or “hard” skills and non-cognitive or “soft” skills) on the black/white pay differentials and occupational choices. Our theoretical analysis derives two main hypotheses. First, the more intensively “soft” skills are used in an occupation, the greater the racial income gap is there in that occupation. In other words, the racial income gap of an occupation depends on its relative requirement of “soft” skills versus “hard” skills. Second, in response to differential discrimination across occupations and the negative impacts of the stereotype of being disadvantaged on the formation of “soft” skills, ceteris paribus, blacks are more likely to self-select themselves into the jobs that use “hard” skills more intensively.
Generally speaking, occupations that use “soft” skills more intensively are also those with more necessary social interactions. Thus, the first hypothesis can be regarded as an extension and generalization of Holzer and Ihlanfeldt (1998), who investigates the effects of customer discrimination and the degree of direct customer contact on blacks’ earnings in the United States. Moreover, the greater racial wage gap in the occupations in which “soft” skills are more intensively used results from two possible sources. First, there may be differential “taste discrimination” across occupations. Second and perhaps more importantly, ethnic minorities, such as blacks, may face comparative disadvantage in accumulating “soft” skills. Carneiro, Heckman and Masterov (2003) have shown that family background is responsible for most of the non-cognitive skill gaps between blacks and whites in the US.
The recent study by Persico, Postlewaite and Silverman (2004) and some sociologists point out another possibility for the skill gaps, i.e. the stereotype of being disadvantaged in the society may reduce the accumulation of (pre-market) human capital, particularly non-cognitive skills, for the individuals from disadvantaged groups. Our theoretical hypotheses are tested using the National Longitudinal Survey of Youth (NLSY) data. Our sample is the pooled male white-collar workers from the 1982 to 2000 waves of NLSY. We first classify jobs into “hard” skill vs. “non-hard” skill ones and “soft” skill vs. “non-soft” skill based on the information provided by the Occupational Information Network (O*NET). Then, the regression analysis yields the following main findings. First, black/white pay differentials tend to be smaller for “hard” skill jobs than for “non-hard” skill jobs, and also smaller for “non-soft” skill jobs than for “soft” skill jobs. Second, based on the Heckman selectivity model, we find that the pattern of self selection implied by our theory indeed existed. Furthermore, the estimated results demonstrate that black white-collar workers tend to self-select themselves into “hard” jobs.
Thus, the empirical results provide clear support to our theoretical claims. In what follows, Section 2 provides the theoretical analysis and discusses the hypotheses; Section 3 describes the data used in the empirical analysis and discusses the empirical methodology employed. Section 4 reports and analyzes the empirical findings. Likely to have negative views of black men as workers.” presents further theoretical implications for our main findings.