The Eclectic (OLI) Paradigm of International Production: Past, Present and Future JOHN H. DUNNING
ABSTRACT This article describes the origins, and traces the subsequent evolution of the eclectic paradigm from the mid-1950s to the present day. It does so in the light of the changing characteristics of MNE activity and of the global economic scenario. The article concludes by asserting that the eclectic paradigm still remains a powerful and robust framework for examining contextual specific theories of foreign direct investment and international production. Key words: Eclectic paradigm; FDI; MNEs; Strategy; International production; Alliances. JEL classifications: F21, F23, M21.
1. Its Origins Although the eclectic paradigm (or the eclectic theory as it was initially called) of international production was first put forward by the present author at a Nobel Symposium in Stockholm in 1976, its origins can be traced back to the mid-1950s. At that time, I was writing my PhD thesis, later to be published as a book (Dunning, 1958), on US direct investment in British manufacturing industry. Earlier research by Rostas (1948), Frankel (1955) and some Anglo ± American study teams1 had shown that the labour productivity in US manufacturing industry was, on average, 2 to 5 times higher than that in UK industry. The question this fact posed in my mind was: was this difference in productivity a reflection of the superior indigenous (and immobile) resources of the US (cf. the UK) economy; or was it due to the more proficient way in which the managers of US firms (cf. UK firms) harnessed and organised these resources? ± a capability which, I argued, at least to some extent, might be transferable across national boundaries. This article draws on various past contributions of the author, but most particularly on those of Dunning (2000a and b). Professor J.H. Dunning, Holly Dell, Satwell Close, Rotherfield Greys, Henley-on-Thames, Oxon RG9 4QT; fax: + 44 (0)1491 628 902 International Journal of the Economics of Business ISSN 1357-1516 print/ISSN 1466-1829 online 2001 Taylor & Francis Ltd http://www.tandf.co.uk/journals DOI: 10.1080/13571510110051441
J. H. Dunning
The hypothesis of my thesis then was, if the superior productivity was entirely managerially related, US manufacturing affiliates in the UK should perform at least as well as their parent companies, and fare considerably better than their indigenous competitors. This I identified as the ownership-specific effect, as the productivity differences were presumed to rest on the spatially transferable intangible assets of the parent companies. If, however, the US affiliates in the UK recorded no better performances than their UK competitors, and hence, much poorer than that of their parent companies, I hypothesised that this would be due to the non-transferable (i.e. immobile) characteristics of the US economy. This I called the location specific component of any productivity differential. As might be expected, I discovered that US affiliates were not as productive as their parent companies, but were more productive than their local competitors. This then suggested that, in the 1950s, at least, Anglo ± American productivity differences were partly explainable by location (L) and partly by ownership (O) specific characteristics. However, my study omitted to ask a follow-up question, namely, to what extent was the origin of the O advantages of US firms themselves home country specific? Neither did it attempt to distinguish between those O advantages that arose as a consequence of US direct investment in the UK, and those that the US firms possessed prior to engaging in foreign production. I took up the theme of ownership and location advantages again in two papers written in the early 1970s. The first (Dunning, 1972), concerned the likely impact of Britain’s membership of the European Common Market...