Counterfactual impact evaluation of EU rural development programmes - Propensity Score Matching methodology applied to selected EU Member States. Volume 1: A micro-level approach

摘要

The main objective of this study is to show how various direct and indirect effects (e.g. deadweight loss, leverage effects, substitution and displacement effects) of EU RD programmes can be empirically estimated using recently developed advanced econometric evaluation methodologies. Answers to EU Common Evaluation Questions (CEQ) regarding effects of the RD programme at a farm level are provided by comparing changes in respective result indicators (e.g. profits, employment, gross-value added, labour productivity, etc.) in the group of programme beneficiaries with a control group (counterfactual analysis). Policy relevant direct programme effects are calculated on the basis of Average Treatment on Treated (ATT) indicators (for programme beneficiaries), Average Treatment Effects on Non-Treated (ATNT) indicators (for programme non-beneficiaries) and Average Treatment Effects (for both groups) using a combination of propensity score matching (PSM) and difference in differences (DID) methods. Furthermore, a modified conditional DID estimator (PSM-DID) is applied to confer about specific programme's general equilibrium effects (e.g. substitution and replacement effects). Robustness of obtained results (potential effect of a hidden bias) is analysed by applying a sensitivity analysis (Rosenbaum bounds). The empirical analysis - focused on the evaluation of an impact of two RD programmes implemented in one new and one old member states, i.e. the SAPARD programme in Slovakia (years 2002-2005) and the Agrarinvestitionsförderungsprogramm (AFP) in Schleswig Holstein, Germany (2000-2006) - is based on micro-economic data (balanced panels) of bookkeeping farms (including programme participants and programme non-participants). Furthermore, methodological and policy recommendations are provided on a general applicability of conventional and advanced evaluation methods, selection of variables and matching techniques, and the use of various data-bases for evaluations of RD programmes carried out at micro-economic levels.