Mutual Productivity Spillovers and Clusters in Eastern Europe : Some Empirical Evidence

In recent years FDI inflows towards transition countries have progressively increased, further stimulated by the entrance of some countries in the European Union. Traditional theoretical studies predict that foreign-owned companies serve as a very important source of technology transfer and productivity spillovers for the host countries. At the same time, the current applied literature finds mixed results with regards to the actual spillover effects from foreign-owned to local companies. This strand of literature on FDI is based on the fact that the MNEs' motivations for investing abroad are characterized mostly by the possibility of exploiting their pre-existing ownership advantages. However, a new approach towards MNEs as asset-seeking entities has been recently growing. This paper tests for the presence of traditional direct horizontal spillovers from foreign to domestic firm as well as of reverse horizontal spillovers from domestic to foreign firm in the context of two Eastern European countries (Poland and Romania). We have further introduced the concept of regional clusters as industrial environments theoretically more prone to induce mutual spillovers between foreign and domestic firms. In this respect, this paper examines two hypotheses: firstly, whether the overall effect of direct spillover is greater for firms in clusters compared to non-clustered firms, and secondly whether the reverse spillover effect actually takes place and if clusters have any role. The econometric analysis is based on a sample of more than 7000 manufacturing firms in the two Eastern European countries. By comparing two estimation methodologies, OLS first difference (after having first estimated the coefficients of the production function with the Levinsohn and Petrin estimator), and a dynamic system GMM, we test the presence of direct and/or reverse spillovers between foreign and domestic firms considering the role of regional clusters. The evidence found with reference to direct spillovers in clusters is in some cases positive confirming the hypothesis of a presence of a cluster effect. Reverse spillovers effects have been found in some cases, both in cluster and outside clusters, and even in low-tech sectors. This suggests that the presence of clusters could be a determinant in FDI localization decisions as there is a possibility of reverse spillovers, even if the host country does not possess higher technological capacity. Furthermore, clusters do seem to be industrial environments where the possibility of direct spillovers is considerable