Counterfactual impact evaluation of EU rural development programmes - Propensity Score Matching methodology applied to selected EU Member States. Volume 2: A regional approach

Abstract

objective of this study is to analyze the impact of EU RD programmes on rural regions. Aggregated effects of a given RD programme at regional levels are estimated using the Rural Development Index (RDI) – a proxy describing the overall quality of life in individual rural areas. The impacts of individual RD measures are analysed by means of a counterfactual analysis by applying combination of the Propensity Score Matching (PSM) (e.g. Kernel matching) and difference-in-differences (DID) methods (i.e. by comparing supported regions and matched control group, prior to the programme and after it). Evaluation of programme effects (by programme measures) at regional level is carried out on the basis of the estimated policy parameters: Average Treatment Effects (ATE), Average Treatment on Treated (ATT) and Average Treatment on Untreated (ATU) effects by using the RDI Index and unemployment ratios as impact indicators. Given information on regional intensity to programme exposure (financial input flows by regions) the overall impact of obtained support via a given RD programme is estimated by means of a dose-response function and derivative dose-response function within the framework of a generalized propensity score matching (GPS). Furthermore, sensitivity analysis (Rosenbaum bounds) is carried out in order to assess a possible influence of unobservables on obtained results (under a binary PSM methodology). Above methodologies are empirically applied to evaluation of the impact of the SAPARD programme in Poland and Slovakia in years 2002-2005 at NUTS-4 level. Results show a full applicability of proposed approach to the measurement of the impact of rural development and structural programmes.