Yts, Employment and Sample Selection Bias

In this paper, data from the first Youth Cohort Study (YCS) are used to analyze the effectiveness of the Youth Training Scheme (YTS) in England and Wales in terms of its impact, at an individual level, on the employment prospects of young people. The analysis goes beyond previous work in that it explicitly deals with sample selection bias and heteroscedasticity. A switching bivariate probit model is estimated and is found to be preferred to both the univariate probit model and the bivariate probit model with a YTS participation dummy. The analysis suggests that previous studies of YTS may have underestimated both the size and variability of the effect that the scheme had on participants' chances of finding employment.