Despite extensive research, economists have been unable to explain three recent youth employment phenomena. First, why did overall youth employment recover between 1965 and 1978, after a considerable decline in the previous decade? Second, why did nonwhite employment rates fall precipitously relative to white employment rates from 1965-78? Third, why are current nonwhite employment rates dramatically lower than white employment rates?

This dissertation expands on past theoretical models of youth employment to address these puzzles. The specification stresses the heterogeneity and educational decisions of teen-agers and the overriding influence of family on youth employment and enrollment.

Previous empirical studies were based on highly collinear national aggregate time-series data and neglected important family-effect variables. This thesis uses a more suitable data base to distinguish the various influences on youth employment: decennial census data for one hundred SMSAs in 1960 and 1970. Further, a more extensive set of explanatory variables, including measures of family structure, is employed.

The empirical results elucidate all three puzzles, challenge the conventional wisdom on key explanatory variables, and confirm the paramount importance of family influences. The baby boom accounts for almost the entire reduction in employment and increase in school enrollment experienced from 1956-65. The 1965-78 recovery in white teen-age employment stemmed from a gradual adjustment of reservation wages by later baby boom youths and, more significantly, the increasing weight placed on working by family members, as reflected by higher adult female participation. Finally, both the failure of nonwhite teen-age employment to recover between 1965 and 1978 and the white/nonwhite employment gap can be traced to family-related causes. The 20 percentage point drop in two-parent families among nonwhites accounts for virtually the entire decline in nonwhite employment rates since 1965 and the 33 percentage point white/nonwhite differential in two-parent families accounts for the current racial employment disparities.