Working Paper: CEPR ID: DP8474
Authors: Sascha O. Becker; Peter Egger; Maximilian von Ehrlich
Abstract: Transfers to individuals, firms, and regions are often regulated by threshold rules, giving rise to a regression discontinuity design. An example are transfers provided by the European Commission to regions of EU member states below a certain income level. Researchers have focused on estimation of the average treatment effect of this program, assuming that it does not vary in a systematic way across units. We suggest a regression discontinuity design which allows for parametric or nonparametric identification of heterogeneous average treatment effects that systematically vary with observable characteristics in order to shed light on the role of absorptive capacity in determining the impact of regional transfers on economic growth across regions in the European Union. The results suggest that only about 47% of the regions, namely those with a sufficiently high endowment with human capital and a high quality of government, are able to turn transfers under the Union's Objective 1 Structural Funds programme into faster growth. Those regions are the ones which are responsible for a positive average effect of the programme.
Keywords: absorptive capacity; heterogeneous local average treatment effects; regional transfers; regression discontinuity design
JEL Codes: C21; H54; O40; R11
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
Cause | Effect |
---|---|
Absorptive capacity (F35) | Effectiveness of regional transfers (F16) |
Human capital (J24) | Economic growth (O00) |
Quality of government (H11) | Economic growth (O00) |
Regional transfers (J62) | Economic growth (O49) |
Absorptive capacity (F35) | Positive growth effects from transfers (F16) |
Human capital and quality of government (J24) | Local average treatment effect (LATE) of transfers (F16) |
Regions with above-average absorptive capacity (R53) | Positive LATE of transfers (F16) |
Regions with below-average absorptive capacity (F35) | Insignificant LATE of transfers (F16) |