An Empirical Study of the Internationalisation of Research and Development by Leading Firms

As we saw in the previous chapter a number of studies have tackled the problem of distinguishing the determinants of overseas R&D by MNEs. In the course of these studies a number of clear hypotheses have been defined. Notable data problems have, however, been encountered in subjecting the approaches to empirical testing. Several studies (e.g. Hewitt 1980; Hirschey and Caves 1981; Lall 1979) used industry level, rather than firm level, data, so that, in Hewitt’s words (1980, p. 318), ‘industries must be treated as if they were firms.’ Where firm level data has been used (Mansfield, Teece and Romeo 1979; Håkanson 1981) the sample has been quite small and restricted to firms from one country. The data source studied in this chapter has the advantage of being relatively large (122 firms) and incorporating non-US as well as US enterprises. However, it too has its deficiencies, notably in failing to cover several variables (e.g. age of overseas operations, cost influences, direct measures of R&D scale economies) whose relevance has been demonstrated by earlier researchers. Nevertheless the information available to us does permit the testing of a number of hypotheses derived in the pioneering literature from a different data perspective.