ERSA – 2017 Barcelona Workshop on Regional and Urban Economics: Causal Inference in Regional and Urban Economics. Methods and Applications

November 23rd-24th 2015, Faculty of Economics and Business, University of Barcelona

On November 23rd and 24th 2017, AQR-IREA hosted the ERSA – 2017 Barcelona Workshop on Causal Inference in Regional and Urban Economics at the Faculty of Economics and Business of the University of Barcelona.

As in past editions, the 2017 Barcelona Workshop on Regional and Urban Economics organised by the AQR research group in collaboration with the ERSA, brought together researchers interested in spatial economics, as well as in other interrelated disciplines such as development and conflict, housing economics, labour markets, or innovation. This year the Workshop focused on the discussion of empirical methods and strategies that allow estimating in a consistent and robust way causal effects of interest in the field of Regional and Urban Economics. As widely acknowledged, causal evidence is crucial to derive sound conclusions and to guide policies with a spatial dimension.

Seven papers that include cutting-edge methodologies to identify causal effects in a spatial context were selected by the Scientific Committee to be presented in this year’s Workshop. It must be said that, due to the high average quality of the submitted papers (about 40), the Scientific Committee faced a difficult task in selecting the papers to be accepted. Together with the authors of the papers from several countries, Peter Egger (KOF Swiss Economic Institute, ETH Zürich) contributed to the Workshop with a Keynote Speech on the consideration of spatial effects in selection and treatment models in a dynamic framework. Overall, around 45 scholars and researchers attended the three thematic sessions and the keynote speech. They contributed to the lively discussions during the presentation of the papers and even during the coffee breaks and the Workshop lunch.

On November the 23rd, the agenda of the event started with an opening ceremony, in which the Dean of the Faculty of Economics and Business of the University of Barcelona, Dr. Ramón Alemany, welcomed all presenters and the audience. After this welcoming words, the chair of the Organising Committee, Dr. Enrique López-Bazo, introduced the workshop and acknowledged the work done by the members of the Organising and Scientific Committees and, particularly, by Dr. Antonio Di Paolo. Shortly after the protocolary ceremony, Dr. Peter Egger gave his keynote speech on the importance of accounting for spatial effects, through the standard mechanisms of spillovers and diffusion, in dynamic models of endogenous selection and treatment. He discussed the specifications that include these spatial effects and derived the corresponding estimators to measure their key parameters in a consistent manner.

The three contributed sessions were organised around the broad methodological approaches used to identify the causal effects of interest. The first one focused on “Instrumental Variables”, chaired by Dr. Jordi Jofre (IEB-Universitat de Barcelona), included two presentations by Federico Curci (Universidad Carlos III de Madrid) and Kai Gehring (University of Zürich), respectively. The first paper used the temporal variation in lead poisoning and geographical differences in soil quality in different USA cities, and its link with mental disorders, to identify the causal effect of violent crime on suburbanization. The second examines the causal relationship between opium and the geography of conflict in Afghanistan, combining temporal variation in international drug prices with spatial variation in opium suitability.

On the 24th, we had two papers in the second session on the application of “Regression Discontinuity Design” in a spatial context. The first one, presented by Mara Guia (Roma Tre University), exploits the spatial discontinuity between highly funded and less funded wards in some UK regions to analyse the role played by the EU Structural Funds on the UK’s decision to leave the European Union (i.e. the so-called Brexit). In turn, Felipe Carozzi (London School of Economics and Political Science) estimates in his paper the price elasticity of land re-development for residential purposes. It does so by using a spatial regression discontinuity design based on school quality information and school admission boundaries that correlates with demand but are orthogonal to redevelopment.

Finally, the application of “Difference-in-Difference” methods in a spatial context was discussed in the third session of the Workshop. Three papers were presented in this section. The first one by Alexander Stepanov (European Bank for Reconstruction and Development) compared the current socio-economic outcomes of “Science Cities” created in former Soviet Russia to those of matched localities that were similar at the time of establishment. The second, presented by Sander Rambier (Katholieke Universiteit Leuven), evaluates the effectiveness of place-based policies in Flanders using details of the process of selection of eligible areas for regional aid and differences with respect to the aid map in the previous period as key information for the diff-in-diff exercise. This session ended with the presentation by Daniele Bondonio (Università del Piemonte Orientale). Daniele’s paper uses a novel dynamic propensity score matching approach for multiple cohorts of US counties to examine the resilience of local economies to natural disasters under the existing relief assistance provisions. The effect of interest is estimated using a difference-in-difference estimator that compares trends in county-level post-disaster outcomes with counterfactual trends of non-disaster comparison counties.

More detailed information about this last edition of the Workshop is available at the website:

We hope to meet you again in Barcelona for the 2016 edition, which will be focused on another hot topic related with Urban and Regional Economics.


Download the summary of the workshop here.