A bivariate probit approach to the determination of employment: a study of teen employment differentials in Los Angeles County

Following a bivariate probit approach and using the 1990 Public Use Microdata Samples (PUMS) for Los Angeles County, this study shows that the employment of teenage workers depends on both the worker's participation decision as well as the employer's hiring decision. Omission of the role of participation decision from the estimation of employment probability yields misleading evidence of hiring discrimination against Blacks. This evidence, however, disappears when the participation and hiring equations are estimated jointly in a bivariate framework. The study finds no evidence of discrimination against females and Latinos. In addition, the study shows that family, household and neighbourhood characteristics play significant roles in the determination of teen employment.