The latin model of welfare: Do ‘insertion contracts’ reduce long-term dependence?

This paper aims to present an assessment of the welfare policies implemented in most South European countries. Welfare programs in these countries try to combine a basic level of economic protection and measures favoring life and labor skills (‘insertion benefits’) of low-income households. We focus on a specific program set up with the twofold strategy of cash and ‘insertion benefits’ (Madrid's IMI) and, more precisely, on the so-called ‘insertion projects’, consisting in a gradual mix of job search assistance, training and subsidized jobs. We evaluate the effects of these ‘insertion projects’ on welfare recidivism and the duration of off-welfare spells using propensity score-matching methods. Our results suggest that propensity score estimators appear to reduce selectivity due to non-random participation. Both recidivism rates as well as the duration of off-welfare spells suggest potentially successful interventions.     

A fully Bayesian tracking algorithm for mitigating disparate prediction misclassification

We develop a fully Bayesian tracking algorithm with the purpose of providing classification prediction results that are unbiased when applied uniformly to individuals with differing sensitive variable values, e.g., of different races, sexes, etc. Here, we consider bias in the form of group-level differences in false prediction rates between the different sensitive variable groups. Given that the method is fully Bayesian, it is well suited for situations where group parameters or regression coefficients are dynamic quantities. We illustrate our method, in comparison to others, on simulated datasets and two real-world datasets.