Revisiting the Employability Effects of Training Programs for the Unemployed in Developing Countries

Data sets used for evaluations in developing countries do not lend themselves to measuring the impact of training programs on the re-employment dynamics of trainees. An exception is a data set collected for an evaluation conducted in 1994 on participants in a training program targeting the unemployed in Mexico. In addition to having a control group of eligible individuals who did not participate in the program, this data set is the only one with longitudinal data covering not only the length of unemployment episodes after the training of the respondent, but also the duration of his/her employment spells. We use this data and estimate the additional weeks individuals work as the result of training, relative to what would be the case without it. Based on hazard functions, we calculate a program`s impact on both the time spent searching for a job and the time spent in that job. We show that a failure to distinguish between finding a "sustained" job versus finding "a job" can lead to misleading conclusions about a program`s effectiveness. We also illustrate the need to correct for unobserved heterogeneity across individuals in hazard functions to avoid misleading implications in an evaluation