Individual disadvantage and policy performance: The makings of model-based composite indicators

In evaluating a policy, it is fundamental to represent its multiple dimensions and the targets it affects. Indeed, the impact of a policy generally involves a combination of socio-economic aspects that are difficult to represent. In this study, regional training policies are addressed, which are aimed at recovering the huge gaps in employability and social inclusion of weak Italian trainees. Previous counterfactual estimates of the net impact of regional training policies show the mess to observe and take into account the manifold aspects of trainees' weakness. In fact, the target population consists of very disadvantaged individuals, who experience hard situations in the labour market. To overcome this shortfall, the present paper proposes a Structural Equation Model, that considers the impact of trainees' socio-economic conditions on the policy outcome itself. In particular, the ex ante human capital is estimated from educational, social and individual backgrounds. Then, labour and training policies augment the individual human capital, affecting labour market outcomes jointly with individual job search behaviour. All these phenomena are expressed by a wide set of manifest variables and synthesised by composite indicators calculated with Partial Least Squares SEM. The makings of SEM are appraised, applied to the case of trainees in compulsory education.