Many governments in Central and Eastern European Countries (CEECs) have offered significant incentives in order to attract foreign investments, motivated by expectations on possible spillover benefits. FDI is usually perceived as a vehicle for transferring technology not only across national boundaries but also between firms, i.e. between foreign and domestic firms. When a foreign firm enters a new market, it generates different reactions from domestic firms. Additional competition pushes for efficiency improvements, which become necessary if firms want to keep their market shares. Domestic firms may learn from foreign companies about new products, production techniques and organization skills, thus increasing their performance. This transfer of benefits may occur either voluntarily, through input output linkages between domestic and foreign firms, or involuntarily through competition, imitation and training. The final result, however, is the same: domestic firms become more productive and efficient, thus fostering local industrial development, as suggested by several economists, from Hirschman (1954) to Markusen and Venables (1999). Despite this long theoretical tradition, there is little conclusive evidence supporting this claim. This paper focuses on the role played by multinational enterprises in fostering the economic development of the hosting regions in Central and Eastern European Countries (CEECs) by testing whether and to what extent technology transfer between domestic and foreign firms does occur. We distinguish not only between horizontal and vertical effects, but also, within the latter, between contacts between domestic suppliers of intermediate inputs and their multinational customers, and contacts between foreign suppliers of intermediates inputs to their domestic clients. The analysis is based on an unbalance panel of about 30,000 domestic firms and about 7,000 foreign firms operating in the manufacturing sector in Bulgaria, Czech R., Hungary, Poland and Romania. The years covered are 1993 through 2002. From a methodological point of view, we adopt the semiparametric estimation method suggested by Olley and Pakes (1996) to account for endogeneity of input demand, and test for the presence of several specific effects, such as country, region, time and sector effects.
