I will survive! The impact of place-based policies when public transfers fade out

ABSTRACT

Are place-based policies capable of taking lagging areas to a higher growth trajectory permanently? We answer this crucial question by investigating what happens when strongly subsidized regions experience a substantial reduction in funding. By analysing an extensive database via the mean balancing approach, we estimate the causal impact of exiting the convergence region status of the European Union regional policy. We find that only regions that experienced a considerable reduction in funding during a recession suffered a negative impact on economic growth. However, the impact varies with the features of the regions and the local economic context.

KEYWORDS:

• place-based policy
• European Union
• mean balancing
• regional growth

JEL:

• C23
• O47
• R11

ACKNOWLEDGEMENTS

An earlier version of this study was published as a working paper in the DISSE Working Papers Series No. 20/2020. We are grateful to Antonella Ferrara, Silvia Granato, the participants at conferences in Rome, L’Aquila and Ispra, and two anonymous reviewers for their comments and suggestions.

DISCLOSURE STATEMENT

No potential conflict of interest was reported by the authors.

Notes

1. There is vast heterogeneity in the amount of funds per capita among treated regions with the least intensively treated regions that could receive more than 10 times less per capita public transfers than the most intensively treated regions. Such differences reflect the decision to allocate more resources to those regions that are particularly needy, to sustain areas experiencing economic and social distress and to maintain qualitative judgment by the EU and individual member states (Cerqua & Pellegrini, 2018).

2. The 75% rule is not always observed. Political negotiations among the member states have often influenced the allocation of the EU budget. Consequently, in the different PPs, some regions have been entitled to receive assistance within the Objective 1 framework, even if they did not comply with the criterion set in the regulations (Pellegrini et al., 2013). Since the 2014–20 PP, there has been a change in terminology. The whole policy has been renamed European Structural and Investment Funds (ESIF) and it is made up of four funds: ERDF, ESF the European Agricultural Fund for Rural Development, and the European Maritime and Fisheries Fund. Furthermore, convergence regions have been renamed less developed regions.

3. Transitional programmes can be of two types: the ‘phasing-out’ programme targets those regions that would have been eligible for funding in terms of the EU-15, but with a GDP per capita higher than the threshold of 75% in terms of the EU-27, while the ‘phasing-in’ programme targets those regions that exited the convergence status as they now have a GDP per capita higher than the threshold of 75% in terms of the EU-15.

4. This is a relevant point as Figures A1–A3 in Appendix A in the supplemental data online show that when controlling for the pre-treatment value of the EU expenditure per capita, the extent and the sign of the estimates change in several instances.

5. However, considering the limited number of treated units, it is not guaranteed that the fuzzy RDD perfectly balance observed and unobserved covariates at the threshold, especially country-specific effects which might be predominant at a time of economic crisis.

6. This variable provides citizen-based perception and experience concerning corruption, quality and impartiality in terms of education, public health care and law enforcement. Although this variable is publicly available only from 2010 onwards, we have been able to use the 2000 and the 2008 values of the EQI thanks to the adaptation and interpolation of 16 survey questions of the EQI dataset with four of the six institutional pillars defining the country-level Worldwide Governance Indicators dataset developed by the World Bank. See Rodríguez-Pose and Ganau (2022) for a more detailed description of this procedure. We thank Andres Rodríguez-Pose for providing us with this data.

7. Real payments typically take place earlier and end earlier than the EU payments. In the analysis, we use the modelled real expenditure as the lag in EU payments may distort the economic analysis of the effect on the investments.

8. The definition of the latter three dummies follows the country clustering suggested by Artis and Zhang (2002).

9. None of these five regions lost the convergence status in the PPs under analysis. In the analysis concerning regions that lost the convergence status in the programming period 2000–06, we exclude from the analysis the German region (DE40) which had the convergence status only for a part of its territory.

10. However, although mean balancing has strongly reduced the imbalance in pre-treatment EUF per capita, they are still about 25% larger in the treatment group. In the robustness analysis we will show that a better matching on pre-treatment EUF leads to almost identical estimates. Table A1 in Appendix A in the supplemental data online displays the weights of each control region in the synthetic exiting region.

11. In Figures A2 and A3, the counterfactual estimate in the case we do not control for pre-treatment EUF is also reported. These estimates show that when controlling for the pre-treatment value of the EU expenditure per capita, the extent and the sign of the estimates change in several instances. In particular, there is a significant change for the Portuguese region of Lisboa (PT17), Southern and Eastern Ireland (IE02), the Italian region of Basilicata (ITF5), the Spanish region of Ciudad Autónoma de Melilla (ES64) and the Swedish region of Övre Norrland (SE33). This analysis confirms the importance of controlling for pre-treatment EUF in this strand of literature.

12. For these regions it was not possible to build a valid counterfactual scenario, as shown in Figures A2 and A3 in Appendix A in the supplemental data online.

13. EUF per capita change over PPs is defined as the average change over time in EUF per capita for each treatment region, compared with the average change over time for the relative synthetic region. The pre-treatment differences in EUF per capita between treated regions and their counterfactuals are more prominent in the most subsidized convergence regions (e.g., the Portuguese region of Madeira, and the Spanish regions of Ciudad Autónoma de Ceuta and de Melilla). Indeed, for such regions, there were no untreated regions with similar pre-treatment EUF per capita.