Pamina Koenig

Structural gravity equations with intensive and extensive margins

Recent trade models with heterogeneous firms have changed the interpretation of gravity equations. Chaney (2008) shows that the effect of distance on the number of exporters and average exports depends on key parameters characterizing the elements of market structure. We use firm-level export data to estimate the structural parameters of Chaneys model. Controlling for the fixed costs of exporting, our estimated parameters match, for 28 out of 34 industries, the models theoretical predictions. Our industry parameters allow us to evaluate separately the effect of transport costs and tariffs on trade, without having to resort to detailed data on trade frictions.

Cluster policies and firm selection : evidence from France

In this paper, we shed light on the selection of the benefi ciaries from the French competitiveness cluster policy which was launched in 2005 and extended to 2012. We disentangle the selection and self-selection eff ects, as emphasized in the theoretical literature on regional and industrial policy. Our main conclusion is that winners were (self-)selected at both steps of the procedure, and that this holds for the three cluster types: worldwide clusters , potentially worldwide clusters and national clusters . We thus provide a methodology which allows us to contrast the e ffective outcomes of the selection process and the official objectives of cluster policies in terms of targeting, and which thus helps in their econometric evaluation.

Clustering the Winners: The French Policy of Competitiveness Clusters

In 2005 the French government launched a policy of competitiveness clusters, giving subsidies for innovative projects managed locally and collectively by firms, research centers and universities. This paper proposes an ex-ante analysis of the outcome of the selection process that took place before the implementation of the subsidies program, in order to assess whether the policy ended up in choosing winners or losers. We first ask how the clusters have been selected, and then focus on the selection of firms within the clusters, using export and productivity as a measure of performance. Our main conclusion is that public authorities have chosen the winners during the two-step selection procedure. Export premium, beyond what individual characteristics would predict, is however most visible within the category of clusters having no international ambition, where heterogeneity among firms is the largest.

Regulatory policy and the location of bio-pharmaceutical foreign direct investment in Europe

This paper examines the relationship between cross-country differences in drug price regulation and the location of biopharmaceutical Foreign Direct Investment (FDI) in Europe. Simple theory predicts that price regulation in one country might affect total investment, but not the location of that investment, if sales are global. Nevertheless, some manufacturers threaten that the introduction of price regulation in a country will motivate them to move their investments to other countries. Are such threats cheap talk, or is there evidence that firms avoid price-controlling countries when making FDI location choices? We use data on 527 investments initiated in 27 European countries between 2002 and 2009 and find that investors are less likely to choose countries with price controls, after controlling for other determinants of investment. We also observe a relative decline in investment in countries that increased the stringency of regulatory regimes during our sample period. The effect is restricted to non-manufacturing investments and is most robust for those related to administrative functions.

Cluster Policies and Firm Selection: Evidence from France

In this paper, we shed light on the selection of the benefi ciaries from the French competitiveness cluster policy which was launched in 2005 and extended to 2012. We disentangle the selection and self-selection eff ects, as emphasized in the theoretical literature on regional and industrial policy. Our main conclusion is that winners were (self-)selected at both steps of the procedure, and that this holds for the three cluster types: worldwide clusters , potentially worldwide clusters and national clusters . We thus provide a methodology which allows us to contrast the e ffective outcomes of the selection process and the official objectives of cluster policies in terms of targeting, and which thus helps in their econometric evaluation.