A critical reflection and some conclusions on OLI

Abstract

When I was in kindergarten, my teacher taught me that the way to learn about a new object was to ask five basic questions: Who or what, when, where, why and how? I have always believed that the OLI (or eclectic) paradigm was about a subset of those questions. The ‘who or what’ is, of course, the multinational enterprise (MNE). While John Dunning’s writings often start with a definition of the MNE, the focus of the eclectic paradigm from its beginnings up to the present day has been O (why?), L (where?) and I (how?). For me, the OLI paradigm is best understood by reading three out of the dozens of books that John Dunning has authored over the past 25 years: International Production and the Multinational Enterprise (1981b), Explaining International Production (1988a) and Multinational Enterprises and the Global Economy (1993a). Each book pulls together and synthesises Dunning’s writings from the previous period.1 In the concluding chapter to this book on the eclectic paradigm, what I would like to do is provide an overview of the development of the OLI paradigm as seen through my ‘lens’, and link the book chapters to its development. My thesis is that OLI should best be seen as a way of looking at the phenomenon of multinational enterprises and their activities. OLI addresses three of the five kindergarten questions – the why, where and how of MNE activities. Each of these questions can be addressed at a different level: macro (big picture, country), meso (mid picture, industry) or micro (firm, top management team). As examples of the why-where-how questions I list some of the research questions that have or currently do engage international business scholars in Table 13.1. I argue that the OLI paradigm has been most successful at the macro and meso levels, and less so at the micro level, but I do not see that as a Chapter Title 277

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 32 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.