Content area
Full Text
Introduction
The impact of foreign direct investment (FDI) on host economies has been the domain of an extensive empirical literature. These studies have become notorious for their widely divergent findings, especially when longitudinal firm-level data are used to measure intra-industry effects of multinational presence on the productivity of domestic firms (Gorg and Greenaway, 2004; Smeets, 2008). This has led to a number of corrections in the modelling of productivity spillovers (see Driffield and Jindra, this issue ). However, extant literature has largely remained stuck to the externality framework, with no explicit criticism of the underlying conceptual and analytical issues. In the following commentary, I will support the more radical view that we should go beyond the mere concept of externality. In fact, externalities by definition entail the idea of 'not-paid-for' advantages/disadvantages accruing to one or more actors from somebody else's activity. I shall argue that although it may be roughly true that most of the negative consequences of multinational presence on host economies are not paid for by multinationals themselves, a substantial part of the advantages accruing to local firms and institutions are instead the result of costly efforts that the host country has to make. In other words, by restricting our attention to FDI externalities, as captured by models based on standard production function, one captures only a minor part of the story. That is, such an approach focuses on productivity increases that are obtained by local firms and institutions at no, or negligible, cost.
Indeed, to fully capture the links between FDI and development, one needs to look upon the more comprehensive analytical category of 'effects' of foreign presence, which cannot be tackled without taking into consideration costs that are paid for by both foreign and local actors, including firms and other institutions.
'Paid-For' and 'Not-Paid-For' Effects
To what extent are the consequences of multinational presence really 'not-paid-for' as assumed in a standard externality framework?
Indeed, the 'non-compensation' hypothesis might be acceptable in the case of disadvantages created by multinational presence: governments and other local counterparts - especially in LDCs - have limited bargaining power relative to foreign multinationals and hence few possibilities to force multinationals to pay for the negative consequences of their action, if any. To illustrate, Union Carbide...