174 H Dunning luternational Business Review 9 2000)163-190 of interest is similar to that of the resource and evolutionary theories,the emphasis of this kind of approach is on the capabilities of management to orchestrate and integrate the resources it can intemally upgrade or innovate,or externally acquire. rather than on the resources themselves.But,as with the resource based and evol- utionary theories.the objective of the decision taker is assumed to be as much directed to explaining the from gi qu ex the theories relating to the origin nd cont ent advanta Table cont po ent wit wh h。 ecle paradi m wa they at expla if ent pher y,rathe than a explanation phenome a.It is true the unit analysis was frequently different,and that the underlying philosophy and some of the assumptions of indus trial organization theory were different than those of resource based theories (Pauwells Matthyssens,1997).But,in general,within their specified analytical frameworks,the predictions of the various theories were consistent with those of a general 'envelope'paradigm,and also the more specific predictions of the O sub- paradigm about the kind of competitive advantages likely to be possessed by MNEs, and the industrial sectors and countries in which their affiliates were likely to record superior levels of performance relative to those of their indigenous competitors (Dunning 1993:Caves.1996) 3.The locational sub-paradigm of countries(and regions) For the most part,until recently,neither the economics nor the business literature gave much attention as to how the emergence and growth of the cross-border activi- ties of firms might be explained by the kind of location-related theories which were initially designed to explain the siting of production within a nation state;nor,indeed of how the spatial dimension of fdi might affect the competitiveness of the investing entities.In the last decade or so,however,there has been a renaissance of interest by economists,(e.g.Audretsch,1998;Krugman 1991,1993;Venables,1998).and industrial geographers (e.g.Scott,1996.Storper.1995:Storper&Scott,1995)in the spatial concentration and clustering of son e kinds of economic activity:by econ omists in the role of exchange rates in affecting the extent g ography and tim of fdi (Cushman.1985:Froot Stein.1991:Rangan.1998).and by ess schola (Porter 1994 1996 Enright 1991 1998)in the idea that an optimum locational portfolio of assets is a etitive advantag in its right par wa s recognize importance fthe loca s as a key det n or M Moreover,since the 1930s there have been numerou here the taken to be indepndent174 J.H. Dunning / International Business Review 9 (2000) 163–190 of interest is similar to that of the resource and evolutionary theories, the emphasis of this kind of approach is on the capabilities of management to orchestrate and integrate the resources it can internally upgrade or innovate, or externally acquire, rather than on the resources themselves. But, as with the resource based and evol￾utionary theories, the objective of the decision taker is assumed to be as much directed to explaining the growth of firm specific assets, as to optimizing the income stream from a given set of assets. The question now arises. To what extent are the theories relating to the origin and content of O specific advantages, as set out in Table 1 — and particularly their contemporary versions — consistent with, or antagonistic to, each other? Our reading is that, when the eclectic paradigm was first propounded, they were largely aimed at explaining different phenomena, or offered complementary, rather than alternative, explanations for the same phenomena. It is true the unit of analysis was frequently different; and that the underlying philosophy and some of the assumptions of indus￾trial organization theory were different than those of resource based theories (Pauwells & Matthyssens, 1997). But, in general, within their specified analytical frameworks, the predictions of the various theories were consistent with those of a general ‘envelope’ paradigm, and also the more specific predictions of the O sub￾paradigm about the kind of competitive advantages likely to be possessed by MNEs, and the industrial sectors and countries in which their affiliates were likely to record superior levels of performance relative to those of their indigenous competitors (Dunning, 1993; Caves, 1996). 3. The locational sub-paradigm of countries (and regions) For the most part, until recently, neither the economics nor the business literature gave much attention as to how the emergence and growth of the cross-border activi￾ties of firms might be explained by the kind of location-related theories which were initially designed to explain the siting of production within a nation state; nor, indeed, of how the spatial dimension of fdi might affect the competitiveness of the investing entities. In the last decade or so, however, there has been a renaissance of interest by economists, (e.g. Audretsch, 1998; Krugman 1991, 1993; Venables, 1998), and industrial geographers (e.g. Scott, 1996; Storper, 1995; Storper & Scott, 1995) in the spatial concentration and clustering of some kinds of economic activity; by econ￾omists in the role of exchange rates in affecting the extent, geography and timing of fdi (Cushman, 1985; Froot & Stein, 1991; Rangan, 1998); and by business scholars (Porter 1994, 1996; Enright 1991, 1998), in the idea that an optimum locational portfolio of assets is a competitive advantage in its own right. The eclectic paradigm has always recognized the importance of the locational advantages of countries as a key determinant of the foreign production of MNEs (Dunning, 1998).18 Moreover, since the 1930s, at least, there have been numerous 18 Unlike with internalization theory, where the locational decision is normally taken to be independent of the modality of resource transference