256 BULLETIN on the special costs and uncertainties of operating a subsidiary in a foreign environment.3 One such special strength is an innovational lead. The product cycle hypothesis begins with the assumption that the stimulus to innovation is typically provided by some threat or promise in the market.4 But according to the hypothesis.firms are acutely myopic:their managers tend to be stimulated by the needs and opportunities of the market closest at hand,the home market. The home market in fact plays a dual role in the hypothesis.Not only is it the source of stimulus for the innovating firm:it is also the preferred location for the actual development of the innovation.The first factor that has pushed innovating firms to do their development work in the home market has been simply the need for engineers and scientists with the requisite skills.That requirement.when gauged through the eyes of the typical innovating form,has tended to rule out sites in most developing countries and has narrowed the choice to some site in the advanced industrialized world.As between such advanced country sites,the home market has generally prevailed.s Locating in the home market.engineers and scientists can interact easily with the prospective customers whose needs they hope to satisfy.and can check constantly with (or be checked by)the specialists at headquarters who are concerned with financial and production planning. The propensity to cluster in the home market is fortified by the fact that there are some well-recognized economies to be captured by an innovating team that is brought together at a common location.s These include the usual advantages that go with subdividing any task among a number of specialists.and the added advantages of maintaining efficiency of communication among the research specialists.7 The upshot is that the innovations of firms headquartered in some given market tend to reflect the characteristics of that market.Historically,therefore.US firms have developed and produced products that were labour-saving or responded to high-income wants:continental European firms.products and processes that were material-saving and capital-saving:and Japanese firms,products that conserved not only material and capital but also space.s 3 That is a central proposition of the S.H.Hymer work,cited earlier.See also my 'The Location of Economic Activity'.in J.H.Dunning.Economic Analysis and the Multinational Enterprise (London: George Allen and Unwin,1970).pp.83-114. Various empirical studies demonstrate that innovations which do not arise out of a market stimulus-innovations,for instance,that are dreamed up by the laboratory as a clever application of some new scientific capability-have a relatively low chance of industrial success.See for instance Sumner Myers and Donald Marquis,Successful Industrial Innouations,National Science Foundation Report No.69-17,G.P.O.,Washington,1969,p.31. s For econometric evidence of the tie between the choice of a production location,skills and innovation,see Sanjaya Lall.'Monopolistic Advantages and Foreign Involvement by US.Manufacturing Industry'.Oxford Economic Papers,forthcoming,March 1980. For evidence of such clustering.see D.B.Creamer,Overseas Research and Development by United States Multinationals,1966-1975(New York:The Conference Board,1976):Robert Ronstadt,Research and Development Abroad by U.S.Multinalionals (New York:Praeger Publishers,1977):and Vernon. Storm Over the Multinationals.pp.43-45. See especially T.J.Allen,Managing the Flow of Technology (Cambridge:MIT Press,1978).An important exception is pharmaceuticals,a case in which US regulation has driven the innovation process abroad.See e.g.H.G.Grabowski and J.M.Vernon,'Innovation and Invention:Consumer Protection Regulation in Ethical Drugs'.American Economic Review,vol.67.no.1.1977,pp.359-364. 8 For evidence,see W.H.Davidson.'Patterns of Factor-Saving Innovation in the Industrialized World',European Economic Review,No.8.1976.pp.207-217.256 BULLETIN on the special costs and uncertainties of operating a subsidiary in a foreign environment3 One such special strength is an innovational lead. The product cycle hypothesis begins with the assumption that the stimulus to innovation is typically provided by some threat or promise in the market4 But according to the hypothesis, firms are acutely myopic; their managers tend to be stimulated by the needs and opportunities of the market closest at hand, the home market. The home market in fact plays a dual role in the hypothesis. Not only is it the source of stimulus for the innovating firm; it is also the preferred location for the actual development of the innovation. The first factor that has pushed innovating firms to do their development work in the home market has been simply the need for engineers and scientists with the requisite skills. That requirement, when gauged through the eyes of the typical innovating form, has tended to rule out sites in most developing countries and has narrowed the choice to some site in the advanced industrialized world. As between such advanced country sites, the home market has generally prevailed.5 Locating in the home market, engineers and scientists can interact easily with the prospective customers whose needs they hope to satisfy, and can check constantly with (or be checked by) the specialists at headquarters who are concerned with financial and production planning. The propensity to cluster in the home market is fortified by the fact that there are some well-recognized economies to be captured by an innovating team that is brought together at a common location.6 These include the usual advantages that go with subdividing any task among a number of specialists, and the added advantages of maintaining efficiency of communication among the research specialists.7 The upshot is that the innovations of firms headquartered in some given market tend to reflect the characteristics of that market. Historically, therefore, US firms have developed and produced products that were labour-saving or responded to high-income wants; continental European firms, products and processes that were material-saving and capital-saving; and Japanese firms, products that conserved not only material and capital but also space.8 That is a central proposition of the S. H. Hymer work, cited earlier. See also my 'The Location of Economic Activity', in J. H. Dunning, Economic Analysis and the Multinational Enterprise (London: George Allen and Unwin, 1970), pp.83-114. Various empirical studies demonstrate that innovations which do not arise out of a market stimulusinnovations, for instance, that are dreamed up by the laboratory as a clever application of some new scientific capabilityhave a relatively low chance of industrial success. See for instance Sumner Myers and Donald Marquis, Successful Industrial Innovations, National Science Foundation Report No. 69-17, G.P.O., Washington, 1969, P. 31. For econometric evidence of the tie between the choice of a production location, skills and innovation, see Sanjaya Lau, 'Monopolistic Advantages and Foreign Involvement by U.S. Manufacturing Industry', Oxford Economic Papers, forthcoming, March 1980. 6 For evidence of such clustering, see D. B. Creamer, Overseas Research and Development by Unite4 States Multinationals, 1966-1975 (New York: The Conference Board, 1976); Robert Ronstadt, Research and Development Abroad by U.S. Multinationals (New York: Praeger Publishers, 1977): and Vernon, Storm Over the Multinationals, pp. 43-45. 7See especially T. J. Allen, Managing the Flow of Technology (Cambridge: MIT Press, 1978). An important exception is pharmaceuticals, a case in which US regulation has driven the innovation process abroad. See e.g. H. G. Grabowski and J. M. Vernon, 'Innovation and Invention: Consumer Protection Regulation in Ethical Drugs', American Economic Review, vol. 67, no. 1, 1977, pp. 359-364. 8 For evidence, see W. H. Davidson, 'Patterns of Factor-Saving Innovation in the Industrialized World', European Economic Review, No. 8, 1976, pp. 207-217