Essays on the youth and low-skilled labor market

Abstract

This dissertation consists of three chapters on the youth and low-skilled labor markets. In Chapter 1, I show that teen employment is significantly more responsive than adult employment to immigration, and that growth in low-skilled immigration appears to be a partial explanation for recent declines in teen employment rates. Using variation in immigrant shares across metropolitan areas between 1980 and 2000, I demonstrate that the impact of immigration on youth employment is at least twice as large as the impact on adults, and that immigration affects school enrollment decisions and the type of jobs held by native youth. These effects are strongest for black youth and youth from poorer and less educated families. The estimates suggest that a 10 percentage point increase in the immigrant share of a city's low-skilled population reduces the teen employment rate by 5 percentage points, implying that between one-third and one-half of the fall in teen employment between 1990 and 2005 can be explained by increased immigration. In Chapter 2, co-authored with David H. Autor and Alan Manning, we offer a fresh analysis of the effect of state and federal minimum wages on earnings inequality over 1979 to 2007, exploiting substantially longer state-level wage panels than were available to earlier analyses as well as a proliferation of recent state minimum wage laws. We obtain identification using cross-state and over-time variation in the 'bite' of federal and applicable state minimum wages, as per influential studies by Lee (1999) and Teulings (2000, 2003).

(cont.) Distinct from this work, we use statutory minimum wages as instrumental variables for the bite of the minimum wage, thereby purging simultaneity bias stemming from errors-in-variables, which we hypothesize causes upward bias in prior OLS estimates. While we uphold the finding that the minimum wage reduces inequality in the lower tail of the wage distribution, we estimate that earlier OLS models overestimate this impact greatly-by 150 to 450 percent. Models purged of simultaneity bias indicate that the minimum wage explains at most one-third of the rapid rise in inequality during the 1980s, and a comparable share of the more modest subsequent rise. These impacts are still larger than would be implied by a simple mechanical application of the minimum to the distribution, suggesting spillovers. We identify these spillovers by structurally estimating the latent wage distribution, calculating the mechanical effect of the minimum wage through truncation, and inferring spillovers by comparison of the mechanical and observed distributions. Spillovers account for one-third to one-half of the minimum's modest impact on percentiles in the lower tail of the wage distribution. Their magnitude has declined in parallel with the direct effects of the minimum, though their share of the total effect has risen.

(cont.) In Chapter 3, I explore the extent to which polarization in the adult labor market-i.e. a gradual increase in the share of adults working in the highest and lowest paying occupations, caused by technology-induced (computers) changes in labor demand-has impacted youth employment. I show that, since 1980, teen employment rates fell more in states and commuting zones for which the share of adults in low-paying occupations increased the most. I also find that this measure of polarization is strongly associated with lower teen and low-skilled adult wages, and more weakly associated with lower employment rates for low-skilled adults. These results can be rationalized in a model of local labor markets for which a reduction in the price of computing capital reduces labor demand for middle -income, routine-task intensive (manufacturing) jobs, pushing these workers into lower-paying service jobs. This chapter therefore provides evidence that a portion of the recent decline in youth employment is attributable to a reduction in labor demand for youth, due to an increase in the supply of substitutable labor (i.e. the gradual movement of less-educated adults from middle-paying to lower-paying occupations).