شرکت های چند ملیتی و سرریزهای بهره وری: یک فراتحلیل Multinational Companies And Productivity Spillovers: A Meta-Analysis

The increasing importance of multinational companies (MNCs) and associated foreign direct investment (FDI) for international production has prompted considerable interest in the effects of MNCs on host countries.1 One of the most frequently referred to positive effect is the presence of technological externalities, which can lead to productivity spillovers from MNCs to domestic Firms in the host country. Since MNCs use a higher level of technology, and technology, or knowledge, has certain characteristics of public goods (Caves, 1996, Markusen, 1995), there is scope for technological externalities and indigenous Firms may benefit through spillovers from MNCs. If there are productivity spillovers, the presence of MNCs leads to productivity increases in domestic Firms , allowing them to become more efficient.2 Such spillovers can occur through three main channels (BlomstroÈm and Kokko, 1998). First, if there are movements of highly skilled staff from MNCs to domestic Firms , these employees may take with them knowledge which may be usefully applied in the domestic Firm. Second, there may be so-called `demonstration effects’ if there are arm’s-length relationships between MNCs and domestic Firms  and domestic Firms  learn superior production technologies from MNCs. Third, competition from MNCs may force domestic rivals to update production technologies and techniques to become more productive. This is frequently referred to as a `competition effect’. As Aitken and Harrison (1999) point out, however, this competition effect may also reduce productivity in domestic Firms , if MNCs attract away demand from their domestic competitors.  Productivity spillovers are difficult to measure, since, as Krugman (1991, p. 53) points out, `knowledge ¯ows … leave no paper trail by which they may be measured and tracked’. The approach adopted in the empirical literature therefore largely avoids the (arguably difficult to answer) question as to how productivity spillovers actually take place, but focuses on the simpler issue of whether the presence of MNCs affects productivity in domestic Firms . This is usually done in the framework of an econometric analysis in which labour productivity or total factor productivity in domestic Firms  is regressed on a number of covariates assumed to have an effect on productivity, one of which is the presence of foreign Firms . If the estimate of the coefficient on the foreign presence variable turns out to have a positive and statistically signifi- cant sign, this is taken as evidence that spillovers have taken place from MNCs to domestic Firms . The empirical results on the presence of spillovers are mixed. In the First empirical study of this kind, Caves (1974), using cross-sectional data for Australia, Finds evidence of positive spillovers. His initial approach has been refined and extended subsequently by, for example, Globerman (1979) for Canada, and BlomstroÈm and Persson (1983), BlomstroÈm (1986), BlomstroÈm and Wolff (1994) and Kokko (1994, 1996) using data for Mexico. These studies, all of which use cross-sectional data, also Find statistically significant positive effects of the presence of MNCs on productivity in domestic Firms . Haddad and Harrison (1993) appears to be the first paper which benefits from the availability of Firm level data for several years, and newly developed panel data econometric techniques to analyse productivity spillovers from MNCs in Morocco. Using such highly disaggregated data and the appropriate estimation techniques, they Find evidence for negative spillover effects of multinationals, ie ceteris paribus, the presence of MNCs in Morocco reduces productivity in domestic Firms . Table 1 lists the results of our literature search, described in more detail in section 1, on papers on productivity spillovers from MNCs. It is apparent that, while there have been a number of spillover studies since Caves (1974), recent years have seen a surge of such studies. This re¯ects, on the one hand, the growing interest, not least from policy makers, in the effects of MNCs on host countries in order to justify policy incentives aimed at attracting MNCs. On the other hand, it also bears testimony to the fact that disaggregated data over longer time periods and estimation techniques for the analysis of such data allowing the investigation of this issue have become widely available to researchers. One should note from the table that, since Kokko et al. (1996), all but two studies listed had available Firm level data. Also, the table shows that all but one study using panel data Find statistically significant negative or statistically insignificant effects of MNC presence on domestic productivity, while all but one cross-sectional study Find statistically significant positive effects. This suggests that the availability of longitudinal Firms  level data, and the appropriate statistical techniques for analysing them, have had profound effects on the results obtained in spillover studies. Various explanations are put forward in the literature to explain statistically insignificant or negative results. For example, the presence of foreign Firms  can reduce productivity of domestic Firms , as pointed out by Aitken and Harrison (1999). Since foreign Firms  can frequently be assumed to posses some sort of Firms -specific assets (Caves, 1996) which allow them to use a superior production technology, they have lower marginal cost than a domestic competitor and can attract demand away from domestic Firms . Thus, productivity in domestic Firms  falls, at least in the short run, because of competition with MNCs. It is also argued in the literature that positive spillovers only affect a certain group of Firms  and aggregate studies may, therefore, underestimate the true significance of such effects. Kokko et al. (1996) Find evidence for productivity spillovers only to domestic Firms  with moderate technology gaps vis-aÁ-vis foreign Firms, ie domestic Firms  with at least some capability of being able to make use of the spillover effects. They do not Find evidence for spillovers from MNCs to domestic Firms  which use considerably lower levels of technology. Aitken and Harrison (1999) Find that productivity in small Venezuelan Firms  (with less than 50 employees) has increased following the presence of MNCs, while there does not appear to be a similar effect on large domestic Firms.