The Albanian Ndihma Ekonomike is one of the first poverty reduction programmes launched in transitional economies. Its record has been judged positively during the recession period of the 1990s and negatively during the more recent growth phase. This article reconsiders the programme using a regression-adjusted local linear matching estimator first suggested by Heckman et al. (1997, 1998). We find the programme to have a weak targeting capacity and a non-significant impact on different household outcomes.