Technology gap, agglomeration and FDI spillovers: A survey of the literature

Recent decades have witnessed a rapid growth of the research field engaged with the empirical identification of spillover effects from Foreign Direct Investment (FDI), where productivity of domestic firms in a host economy is affected by the entrance and operations of foreign owned firms. However, the large body of evidence is very heterogeneous and has fostered limited consensus on the prevalence of positive FDI spillovers. In response to this, recent research is developing alternative research strategies to obtain improved evidence of FDI externalities. One of these strategies concerns the identification of endogenous factors that may influence the occurrence (and perhaps also the nature) of FDI spillovers. The purpose of this chapter is to provide a survey of FDI spillover studies that try to identify such endogenous factors that affect these externality effects. Previous surveys have focused mainly on the discussion on the overall prevalence of positive FDI spillovers. In comparison, only little attention has been paid to studies that try to identify determinants of FDI spillovers. In particular, the present survey focuses on studies that address the questions whether absorptive capacity of domestic firms is important for FDI spillovers to occur and whether and how agglomerations of economic activity within a host economy influence FDI spillovers. The main findings of the survey can be summarized as follows. Research on the effect of absorptive capacity of domestic firms uses the level of technological differences between FDI and domestic firms as a direct inverse indicator of absorptive capacity. A closer examination of the use of this technology gap shows however that this interpretation relates only partially to the underlying concept of catch up. This concept holds that, for meaningful spillovers to occur, a sufficient level of absorptive capacity needs to be accompanied by a sufficiently large technology gap. The survey shows that there is substantial empirical evidence that a large technology gap facilitates rather than hinders the occurrence of positive FDI spillovers, suggesting that the commonly used interpretation of the technology gap as direct inverse indicator of absorptive capacity is flawed. Instead, large technological differences are better interpreted as reflecting a substantial scope for positive effects to occur, offering incentives to domestic firms to improve their ability to absorb positive spillovers. Concerning the effect of agglomeration on FDI spillovers, the large degree of similarity between the mechanisms that create agglomeration economies and channels of FDI spillovers suggests strongly that agglomeration will enhance positive FDI externalities. The limited number of studies that look at this direct relation between agglomeration and FDI spillovers find corroborating evidence of such a positive effect. Clearly, more empirical research on this relationship is called for. Most studies that include assessments of spatial dimensions look at more general geographical dimensions of FDI spillovers. The inclusion of these spatial dimensions is welcome. The evidence shows that the scope for FDI spillovers can be much larger than envisaged originally, as spillovers may occur at the intra-and/or inter-regional level. Having said this, the survey also finds that the majority of the existing evidence on spatial FDI spillovers is characterized by several limitations. In particular, most studies fail to estimate for all types of intra-and inter-regional FDI effects. Furthermore, most studies give insufficient consideration to the variety of relations that may exist between FDI spillovers and geographical distance. FDI spillovers are transmitted via several channels that may be affected differently by distance decay effects. A fuller assessment of the variety of relationships that may exist between spatial spillovers and geographical distance is required to increase the robustness of the evidence on inter-regional FDI externalities.