Good Intentions and Matthew Effects: Access Biases in Participation in Active Labour Market Policies

Over the last few years we have witnessed the emergence of a new orientation in social policy based on the idea that help for disadvantaged individuals should take the form of enabling interventions. This approach, commonly known as social investment, has been criticised for having benefitted the middle class relative to other more disadvantaged groups. In the social policy literature such a bias is referred to as Matthew effect.

Against this background, the objective of this paper is to investigate whether ALMPs manage to reach the most disadvantaged individuals or they are subjected to Matthew effects. We investigate this question for two typically disadvantaged groups of unemployed people: the low-skilled and immigrants. Our analysis is based on a systematic review of evaluations of ALMPs. In general, these studies provide information with regard to who the participants are relative to non-participants. We use this information to ascertain whether or not an access bias is present in these programmes. Our results provide evidence that a Matthew effect is present only in some programmes (typically training and wage subsidies). In other programmes, disadvantaged individuals are overrepresented. Our overall conclusion is that policies are generally explicitly targeted on the most disadvantaged (good intentions) but other factors limit their participation, something which explains the mixed pattern that we observe. We also show that the programmes that are more easily accessible to disadvantaged individuals tend to be less effective.