Association for Public Policy Analysis & Management

Whereas schooling levels have been going up, on average, in the United States over time, the overall trends mask differing underlying patterns across groups based on race and ethnicity. There are level differences in the dropout rates between the different groups with Hispanic trends clearly standing out. Not only are the Hispanics more likely to drop out of High School at any given point in time, they also experience bigger changes in their dropout rates over time as compared to other groups (Whites, Blacks, and Asians).

Existing studies looking at the effect of labor demand shocks on schooling decisions treat all groups as a whole. Given that labor demand shocks such as the Great Recession affected schooling decision through multiple channels and these effects are likely to be differentially important across sub-groups, this paper motivates analysis separately by race and ethnicity. This paper improves upon existing measures of labor demand by constructing group-specific Bartik measures at the MSA level in order to mitigate any concerns regarding employment changes being driven by labor supply responses.

The main result of the paper is that young native Hispanics exhibit higher degree of responsiveness in their schooling decision to labor demand shocks even after we account for differences in intensity of the labor demand shock experienced by the different groups. Whereas Hispanic and Blacks are affected more by labor demand shocks relative to Whites, Hispanics are most responsive to these shocks and respond to adverse shocks by increasing their educational attainment in hardest hit areas. This effects holds true both at High School dropout margin as well as for college attendance. It is worth noting that results from the full sample largely mimic those of Whites implying that any analysis which treats all groups as one, is likely to pick up effects (or lack thereof) driven mainly by the White sub-samples. The results of this paper are therefore consistent with and bring to fore the differential importance of different factors for each of the sub-group and therefore motivate a move away from looking at labor demand shocks from a singular lens.