The use of propensity score-matching methods in evaluation of active labour market programs in Serbia

Abstract

The aim of the paper is to evaluate impacts of active labour market measures in Serbia on the probability of the participants’ employment. In the estimation process of average treatment effects, propensity score-matching method was applied. Its use has considerably reduced the bias in evaluation of the average treatment effects, induced by systematic differences between samples of those who participated in the active labour market programmes and those who did not. The estimated impacts of active labour market programmes on the probability of employment were found to be both positive and statistically significant. .

Keywords

• active policy effects
• propensity score
• matching estimators
• nonexperimental evaluation
• labour market in Serbia