Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 165
Japan's Growing Technological Capability: Implications for the U.S. Economy Japan's Industrial Competitiveness and the Technological Capabilities of the Leading Japanese Firms JOHN CANTWELL NATIONAL SYSTEMS OF INNOVATION AND CHANGES IN TECHNOLOGICAL LEADERSHIP In an influential recent book Mokyr1 set out the case for what he termed "Cardwell's Law," based on an interpretation of Cardwell.2 This Law proposes that highly technologically creative societies only remain so for relatively short historical periods. At some stage the momentum that gathers behind technological advance becomes exhausted. In Mokyr's judgment, the Law has the status of an observed empirical regularity. Technological leadership changes from time to time, moving from one society to another. In recent history, technological leadership has passed from Britain to the United States, and in very recent times it has switched from the United States to Japan. A complementary perspective on these occasional changes in technological leadership has been provided by Schumpeterian economists such as Freeman and Perez.3 Schumpeter had held that long waves of economic development are initiated by pervasive new technologies that have an impact 1 J. Mokyr, The Lever of Riches: Technological Creativity and Economic Progress (Oxford: Oxford University Press, 1990). 2 D.S.L. Cardwell, Turning Points in Western Technology (New York: Neale Watson Science History Publication, 1972). 3 C. Freeman and C. Perez, "Structural Crises of Adjustment: Business Cycles and Investment Behaviour," in G. Dosi, C. Freeman, R.R. Nelson, G. Silverberg, and L.L.G. Soete, eds., Technical Change and Economic Theory (London: Frances Pinter, 1988).
OCR for page 166
Japan's Growing Technological Capability: Implications for the U.S. Economy on every major industry.4 The diffusion of steam power and electric power provide examples. According to Freeman and his associates, these periods of economic transformation depend upon the consolidation of a new technological system or techno-economic paradigm. The new system or paradigm encompasses a broad range of related technological development that goes beyond the best known major innovations that characterize the leading sectors. In this context, a technological paradigm is defined as a widespread cluster of innovations that represent a response to a related set of technological problems, based on a common set of scientific principles and on similar organizational methods.5 The organizational methods associated with different paradigms require the support of different kinds of social institutions. Therefore, it is not surprising that with the emergence of a new paradigm technological leadership tends to move away from a society whose institutions were particularly geared towards problem-solving activity within the confines of the previously prevailing paradigm. Leadership is likely to pass to a society whose institutions are more adaptable to and better represent the organizational structures needed to promote the most pervasive new technologies. In a new technology paradigm every country must adjust its national system of innovation. The national system of innovation is the network of institutions in the public and private sectors that support the initiation, modification and diffusion of new technologies.6 In the paradigm based on mass production that dates from the interwar period, U.S. institutions led the way. The typical national system of innovation relied on the establishment of specialized corporate R&D departments, increasing state involvement in civil science and technology, and the rapid expansion of secondary and higher education and industrial training. The new technology paradigm now taking shape is instead grounded on the economies of scope gained through an interaction between flexible but linked production facilities, in which individual plant flexibility and network linkages both depend upon the new information and communication technologies. The pioneers in this case are Japanese institutions. The appropriate national system of innovation today emphasizes the closer integration of R&D, production and marketing within firms, technological cooperation between firms, generalized education and training to provide a work force with multiple rather than specialized skills, and state support for generic technologies. 4 J.A. Schumpeter, Business Cycles: A Theoretical, Historical and Statistical Analysis of the Capitalist Process, 2 vols. (New York: McGraw Hill, 1939). 5 G. Dosi, Technical Change and Industrial Transformation (London: Macmillan, 1984). 6 C. Freeman, Technology Policy and Economic Performance: Lessons from Japan (London: Frances Pinter, 1987).
OCR for page 167
Japan's Growing Technological Capability: Implications for the U.S. Economy Each technological paradigm is characterized by a set of technological opportunities that can be very different from the opportunities that typified the paradigm it replaces. In the U.S.-led paradigm the greatest opportunities appeared in energy and oil-related technologies and in scale intensive systems. In the paradigm in which Japan is to the fore the best opportunities are in microelectronic technologies and in computerized systems. To take full advantage of these technological opportunities national systems of innovation must be adapted accordingly. However, this is a difficult process, especially in the countries that were the most heavily committed to the previously dominant production methods. Although all countries strive to adjust their national systems to the opportunities opened up by the new paradigm they have varying degrees of success, and significant differences in cross-country institutional structures remain. For this reason differences in industrial competitiveness between countries tend to persist over time within a paradigm, and substantial alterations in the competitive ranking of countries only tend to occur during the windows of opportunity provided by paradigm changes. Cross-country variations in the details of national systems of innovation are associated with fairly systematic differences in rates of technological change across countries once a new paradigm has become established. After a paradigm has settled down, these differences in national systems of innovation are sustained not so much by some natural or inherent cultural characteristics of different societies (despite the fact that this is a common interpretation in popular discussions), but rather result principally from the cumulative and path-dependent nature of technological change itself. While in some respects the international diffusion of new technology brings production systems closer together, in other respects technological development reinforces differences between countries and firms. In their seminal work in this field, Nelson and Winter laid the theoretical foundations for our understanding that technological change is typically cumulative, incremental and differentiated.7 Their theoretical conclusions were entirely consistent with the insights gained from historical studies of technological evolution by Rosenberg.8 Because technological change is differentiated between firms and between countries, the differences between them persist over time; and because technological change is cumulative and incremental, existing leaders tend to preserve their position within the confines of an unchanged paradigm. As will be explained at greater length in 7 R.R. Nelson and S.G. Winter, An Evolutionary Theory of Economic Change (Cambridge, Mass.: Harvard University Press, 1982). 8 N. Rosenberg, Perspectives on Technology, (Cambridge: Cambridge University Press, 1976) and Inside the Black Box: Technology and Economics (Cambridge: Cambridge University Press, 1982).
OCR for page 168
Japan's Growing Technological Capability: Implications for the U.S. Economy Section 2, leaders are defined here with reference to the current rate of technological change and not the absolute level of technology accumulated from the past. Economists often find it hard to understand this idea that persistent differences in the production methods of firms and countries may result from an interactive process of technological change. This is for a variety of reasons, but one is that economists tend to think of technology simply as a matter of applied science or engineering rather than as a matter of organization. Until recently, when thinking of economic organization economists have focused almost exclusively on the market as the means of organizing production and other economic activities, and have disregarded the role of firms and other institutions. Indeed, where they have considered firms they have usually depicted them as an alternative to markets,9 taking the market as a reference point instead of treating the technological and organizational evolution of firms in its own right. If technology can be reduced to scientific and engineering knowledge or information then on the condition that markets work well, technological differences will be short-lived, owing to the scope for trade in this information. Technology is more accurately described as consisting of two strictly complementary components.10 The first is the element that has been emphasized in the conventional economics literature, namely generic knowledge. Such knowledge has the characteristics of a latent public good, it is potentially tradable, and its public availability and diffusion bring the production systems of firms and countries closer together. The second element, which is emphasized in the new Schumpeterian and allied literature, is the tacit component of technology embodied in the collective skills and organizational routines of firms. This element is specific to the localized conditions under which technology is created and used, and while it can be imitated by others it cannot be directly copied in exactly the same form. It is therefore in itself nontradable, irrespective of how well markets work, although under agreements for technological cooperation, contracts can be devised for technical assistance that reduce the costs of imitation. So despite the diffusion of generic knowledge, technological differences between countries and firms remain. The tacit component of technology represents 9 R.H. Coase, ''The Nature of the Firm," Economica, vol. 4, no. 4., 1937 and O.E. Williamson, Markets and Hierarchies: Analysis and Antitrust Implications (New York: Free Press, 1975). 10 R.R. Nelson, "The Public and Private Elements of Technology," mimeo, (New York: Columbia University, 1990) and J.A. Cantwell, "The Theory of Technological Competence and Its Application to International Production," in D.G. McFetridge, ed., Foreign Investment, Technology and Growth (Toronto: University of Toronto Press, 1991).
OCR for page 169
Japan's Growing Technological Capability: Implications for the U.S. Economy the essential source of the continuing discrepancies in technological competitiveness or competence observed across countries and firms. 11 One important aspect of the differences in national systems of innovation is that countries and firms tend to evolve significantly different patterns of technological specialization. The distinctive composition of the distribution of technological activities in individual countries and firms again persists, especially over shorter periods of time. This is particularly true of countries or of national groups of firms. 12 These differences in national patterns of specialization provide a measure of how well a country and its leading firms have become accommodated to the prevailing technology paradigm. The paradigm is characterized by certain pervasive technologies in which opportunities (for development and application) are at their greatest. Technological activities that are the most closely related to these critical fields also offer attractive opportunities. A national system of innovation that is finely tuned to the new paradigm establishes a pattern of specialization that is relatively concentrated in these areas of strong potential growth. The leading role of Japan in the current technology paradigm can be examined in these terms. For some years now, Japan's specialization in technological activity has emphasised the fields that have become dominant in the new paradigm. The specialization that derives from Japan's national system of innovation is thus associated with a systematically higher overall rate of technological change. A specialization in what have become the pervasive new technologies directly raises the overall rate of technological progress since it entails a concentration of activity in branches in which growth is fastest. The overall rate of innovation also increases indirectly, as advances in these critical areas lead to greater spillover benefits in other fields. The favorable relationship between the Japanese pattern of technological specialization and the distribution of technological opportunities across fields of activity can be illustrated from an analysis of U.S. patent data. The composition of technological specialization can be measured through the construction of an index of what has been termed revealed technological advantage, or RTA.13 Revealed technological advantage is defined as the 11 D.J. Teece, G. Pisano, and A. Shuen, "Firm Capabilities, Resources and the Concept of Strategy," University of California at Berkeley Consortium on Competitiveness and Cooperation Working Papers, No. 90-8, December 1990 and Cantwell, op. cit., footnote 10. 12 J.A. Cantwell, "Historical Trends in International Patterns of Technological Innovation," in J. Foreman-Peck, ed., New Perspectives on the Late Victorian Economy: Essays in Quantitative Economic History 1860–1914 (Cambridge: Cambridge University Press, 1991) and P. Patel and K. Pavitt, "Large Firms in the Production of the World's Technology: An Important Case of Non-Globalisation," Journal of International Business Studies, vol. 22, no. 1, 1991. 13 L.L.G. Soete, "The Impact of Technological Innovation on International Trade Patterns: The Evidence Reconsidered," Research Policy, vol. 16, no. 1, 1987 and Cantwell, op. cit., footnote 12.
OCR for page 170
Japan's Growing Technological Capability: Implications for the U.S. Economy national share of patenting in a particular field (in this case, the share of patents granted attributable to inventors resident in Japan) relative to the national share of total patenting in all fields. The RTA index thus approximately varies around unity, with the highest values assigned to the fields of greatest local specialization. The fields of technological activity are derived from the U.S. patent class system. Figure 1 plots the relationship between the distribution of the Japanese RTA index in the 1960s and the growth of total U.S. patenting between the 1960s and the 1980s. The index was calculated across 33 sectors of technological activity, which for the purposes of the diagram have each been assigned to one of ten groups. The positive association between Japan's national pattern of technological specialization and the subsequent extent of technological opportunities across sectors is evident from the estimated regression line. In this regression the coefficient on RTA was significantly different from zero at the I percent level. The fields of greatest opportunity in which Japan was most heavily specialized were (in electrical equipment) computers, calculators and other office equipment, and image and sound equipment or radios and televisions, and (in instruments) photography and photocopying. It can also be shown that Japan's concentration of activity in the areas of the fastest technological change is much greater than for the United States or any other country, although this is partly attributable to the somewhat narrower focus of technological specialization in Japan than in the United States.14 Owing to the cumulative and incremental characteristics of technological change, the degree of mobility in Japan's national pattern of technological specialization over the last 20 or 30 years has not been very great. 15 Therefore, so long as the distribution of technological opportunities across sectors remains similar, Japan is likely to sustain her high rate of technological change. The distribution of opportunities is largely a function of the prevailing technological paradigm. Indeed Japan's position has improved in this respect, since the period between the 1960s and 1980s was one of transition from a paradigm that favored the U.S. national system of innovation to one that favors the Japanese. The high rate of technological change achieved in Japan and by Japanese based firms has had major economic consequences and will continue. It has supported a higher rate of economic growth such that Japan has steadily increased her share of world exports and (since local wages lag 14 D. Archibugi and M. Pianta, "Patterns of Technological Specialisation and Growth of Innovative Activities in Advanced Countries," in K. Hughes, ed., European Competitiveness (Cambridge: Cambridge University Press, 1992, forthcoming). 15 J.A. Cantwell, Technological Innovation and Multinational Corporations (Oxford: Basil Blackwell, 1989).
OCR for page 171
Japan's Growing Technological Capability: Implications for the U.S. Economy Figure 1 The relationship between the growth of total U.S. patenting from 1963– 1968 to 1981–1986 and Japanese revealed technological advantage (RTA) in 1963– 1968. Note: One observation hidden. Sector codes: 0, Food products; 1, Chemicals and allied; 2, Metals; 3, Mechanical engineering; 4, Electrical equipment; 5, Transport equipment; 6, Textiles and wood products; 7, Nonmetallic mineral products; 8, Professional and scientific instruments; 9, Other manufacturing and nonindustrial. behind productivity improvements) established a regular trade surplus. Meanwhile the leading Japanese companies have on average grown faster than their major international rivals, and have consistently increased their share of international markets through exports and international production. The position in international trade and production of national groups of firms depends upon their specific areas of technological strength.16 If the Schumpeterian perspective is a reasonable one the competitive success or failure of countries or firms (as measured by their growth rates and hence the change in their market shares) is essentially a function of the rate of technological change they are able to sustain. To formalize this 16 B. Kogut, "Country Patterns in International Competition: Appropriability and Oligopolistic Agreement," in N. Hood and J.E. Vahlne, eds., Strategies in Global Competition (London: Croom Helm, 1987) and G. Dosi, K. Pavitt, and L.L.G. Soete, The Economics of Technical Change and International Trade (Hemel Hempstead: Harvester Wheatsheaf, 1990).
OCR for page 172
Japan's Growing Technological Capability: Implications for the U.S. Economy view of the Japanese experience a simple model of the relationship between innovation and growth is set out in the next section. This is briefly related to the evidence on the association between technology and growth when comparing Japan with other countries. In the section following, this approach to technology and growth is extended to a consideration of evidence at the company level. It is also shown that the locally specific pattern of technological specialization of a national group of firms contributes to the rate of technological change that they achieve. When examining firms, comparisons of technological development and the evolution of market shares must be made at the industry level. To this end, data on the world's largest industrial firms are considered, given that these companies constitute one anothers' major competitors in the main international industries. In conclusion, some conclusions and suggestions pertaining to the likely future competitive performance of Japan and the United States (and of Japanese and American companies) are discussed. A MODEL OF THE ASSOCIATION BETWEEN TECHNOLOGICAL CHANGE, GROWTH AND COMPETITIVENESS The basic idea explored in the model formulated here is that the essential source of differences in the level of productivity across countries and firms (and hence differences in living standards in the national case) is the level of technological capability that they have accumulated from the past. Technological improvements are in part embodied in new capital equipment or other tangible assets, so the accumulation of technological ability is linked to the process of capital accumulation. However, technological progress is also partly disembodied in the form of collective skills and organizational expertise, and the relative significance of this tacit element of technological capability may increase over time. So in this model the level of accumulated technological capability determines the level of productivity and output. It follows that the rate of technological accumulation determines the rate of growth of productivity, the growth of output, and the rate of capital accumulation. This is a Schumpeterian model in the sense that the rate of technological innovation determines growth, and hence the competitive strength of countries and firms as measured by gains and losses in their market shares. However, it is a relatively simple model, and it is not intended as a formalization of Schumpeter's own views. Schumpeter himself emphasized the process by which competitors tend to catch up with innovative leaders, and in this respect at least he remained faithful to the conventional economist's view of the effect of technology diffusion. As already explained, the more recent Schumpeterian tradition allows that a continued divergence in rates of technological change may be expected to persist in the context of an unchanged paradigm.
OCR for page 173
Japan's Growing Technological Capability: Implications for the U.S. Economy In addition, the model here makes no reference to the role of individual entrepreneurs, which was important to Schumpeter. The model is sufficiently general that it can be applied either at the level of national economies or at the level of firms. It begins from a set of standard identities and definitions as follows: The value of output is denoted by Q; A is the value of labor productivity; N is employment; W is the total wage bill; π is the value of total profits; K is the value of accumulated capital stock; k* is the constant capital-output ratio; x is the share of wages in output; w is the wage rate; and r represents the rate of profit. The (approximate) constancy of the capital-output ratio is a stylized fact of normal economic development, which is used here to avoid unnecessarily complicating the analysis. On the further simplifying assumption that all profits are reinvested as set out in equation (8) (where I denotes the value of investment), the rate of economic growth can be derived: So Prime signs are used to depict derivatives (so K' = dK/dt and K" = d2K/ dt2) and dots over letters indicate proportional rates of growth (so Q = (1/ Q) (dQ/dt) = Q'/Q). Now two further equations, which represent the essence of the argument above, can be used to close the model and to define its dynamic properties: In equation (9), T is the accumulated stock of technology and u* is the constant coefficient of responsiveness of the current level of productivity to the technology stock. In equation (10), x* is the share of wages in output that would prevail if technological accumulation and thus productivity growth ceased. In this event the growth rate would be fixed at (I-x*)/k*. The
OCR for page 174
Japan's Growing Technological Capability: Implications for the U.S. Economy intuition that underpins equation (10) is that if innovation ceased the wage rate would be directly proportional to the level of productivity, but with technological change wages only follow productivity with a lag. Consequently, the higher the rate at which productivity growth is sustained, the lower is the ratio of the wage rate to productivity [where x = w/A, from equations (1), (4), and (6)]. Now with the rate of technological accumulation given by , this determines the rate of productivity growth, since from equation (9): By substituting into equation (10); The growth rate is now given by Hence, in this model output growth and capital accumulation as well as productivity growth come to depend upon the rate of technological accumulation. Countries or firms that achieve higher rates of technological advance also experience faster growth rates and so improve their market shares. The wage rate and its path over time can also be derived: If the rate of technological accumulation were constant (say, =c) then the other growth rates would also be fixed: Returning to an issue mentioned in passing earlier, the leading countries or firms in a technological paradigm are those that enjoy a consistently higher rate of technological accumulation, and thereby in this model sustain a faster rate of growth. However, especially in the early stages of a new paradigm, and if there has been a change in leadership from the previous paradigm, these new leaders are unlikely to be the countries or firms that have the largest accumulated technology stock. This is instead likely to be the province of the former leaders, who achieved a higher rate of accumulation in the past. High levels of technology stock, and productivity (and at a national level, living standards) reflect past rather than current performance.
OCR for page 175
Japan's Growing Technological Capability: Implications for the U.S. Economy The stock of the codifiable element of technology can be measured by the stock of patents or capitalized R&D expenditure, with an allowance for depreciation (in terms of their current effect upon productivity). The rate of technological accumulation may then be measured by the current increase in technology (the annual flow of patents or R&D expenditure) relative to the stock. The model just outlined suggests that this may be the best measure of technological competitiveness. Another commonly used measure of such competitiveness is the ratio of the increase in technology relative to the level of output (T'/Q in the notation above), such as R&D spending relative to the gross domestic product (GDP). Because of their role in empirical studies, it is worth exploring the implications of the system for the evolution of the (T'/Q) and (T/Q) ratios. With plausible values of the relevant parameters—namely x* close to unity and k* greater than unity—the (T/Q) ratio gradually rises over time. That is, there is a tendency for production to become steadily more ''technology-intensive." The path of the (T/Q) ratio is defined by So If x* ≥ 1 and k* > 1 then the expression on the right hand side is positive and the (T/Q) ratio rises. An assumption that x* is close to unity summarizes the Schumpeterian view. Since it follows that if with x* = 1 then . In other words, in this case if there is no innovation, growth ceases and the economy collapses into a stationary state. In this state profits fall to zero; profits depend upon regular technological improvements. Indeed, a country or firm that completely fails to innovate may even find that under competitive pressures its output declines (x* > 1). An assumption that k* > 1 is also reasonable. This simply states that the value of accumulated capital stock is greater than the value of current output. Under these conditions the (T/Q) ratio is a measure of past technological achievements. A stronger rate of technological accumulation in the past leads to a higher (T/Q) ratio. This helps to explain why the truly backward (with a low level of technological capability) may find it so difficult to catch up with the truly advanced (with a high level). If technology intensity rises with growth so does the contribution to output of the tacit component of technology, which cannot be directly traded or transferred between firms. The (T'/Q) ratio represents the combined product of current and past technological competitiveness as So the (T'/Q) ratio can be used as a measure of technological competi-
OCR for page 178
Japan's Growing Technological Capability: Implications for the U.S. Economy The analysis is based on the world's largest 792 industrial firms according to the value of their global sales in 1982.20 The value of these sales in 1972 and 1982 serves as a proxy for their output in the early 1970s and early 1980s respectively. The flow of new technology is measured by the number of patents granted in the United States to the same firms, consolidating patents granted to affiliated companies in the same fashion as in the calculation of global sales. Of the original 792 firms, 267 were excluded due to a lack of information on the value of their world sales in 1972, and a further 10 were excluded as they had no record of patenting in the United Staes. This left a final sample of 515 firms, of which 275 were U.S. owned, 78 were Japanese, 137 were European, and 25 originated from other countries (mainly from Canada). Each firm was also allocated to a primary industry of activity. It is not possible to form an accurate estimate of the rates of technological accumulation enjoyed by these firms without knowing their levels of patenting prior to 1969 (from which to calculate a technology stock). However, it can be asserted with a reasonable degree of confidence that by the late 1960s the rate of technological change in the leading Japanese firms already exceeded that in their major U.S. and European competitors. While the patenting of all Japanese residents in the United States was very low until the early 1960s, by 1970 the leading Japanese firms had attained substantial levels of U.S. patenting. The Japanese firms in the sample were granted an average of nearly 100 patents each in the United States in 1969–1972. Associated with their higher rates of technological change, this group of Japanese firms achieved a higher rate of output growth. Their combined sales grew more rapidly between 1972 and 1982 than the equivalent sales of their U.S. and European competitors in all industries except two. The two exceptions were pharmaceuticals (in which industry technological change may still have been higher in the United States and Europe), and textiles (in which comparisons may be difficult owing to the Japanese firms being more chemical oriented and the U.S. companies being more purely clothing and retail oriented). Moreover, it seems that during the 1969–1986 period the rate of technological accumulation of the same Japanese firms rose significantly, while for European firms the rate remained steady, and for U.S. firms it actually fell. This can be judged from the observation that the level of U.S. patenting of these Japanese companies increased dramatically between 1969–1972 and 1983–1986, for European firms it rose moderately, and for the leading U.S. firms the numbers of patents granted actually fell. 20 J.H. Dunning and R.D. Pearce, The World's Largest Industrial Enterprises, 1962–1983 (Aldershot: Gower, 1985).
OCR for page 179
Japan's Growing Technological Capability: Implications for the U.S. Economy While this particular measure understates American achievements due to the rising propensity to patent abroad (U.S. company patenting also increased in Europe and Japan), all such measures suggest that Japanese firms tended to enjoy a superior rate of technological change. As a result of these changes in the flow of new technology (T' above) there was a shift in the ranking of these groups of firms in terms of their patent to sales ratios (T'/Q above). This shift is illustrated in Table 1. For each group of firms the values of sales in 1972 and 1982 were aggregated and compared with their combined patenting in the United States in 1969–1972 and 1983–1986. There was a tendency for patent to sales ratios to fall simply because sales were measured in nominal and not real terms; due to the effect of the oil price rise this effect was especially noticeable in the coal and petroleum products industry. Despite this, overall the average patent to sales ratio of Japanese firms increased slightly from 67.0 to 70.3. The U.S. company ratio fell substantially from 182.8 to 41.5. Therefore, by the early 1980s the average Japanese TABLE 1 Ratio of U.S. Patents Granted to Global Sales (in billion dollars) of the leading U.S., Japanese, and European Industrial Firms in 1969–1972 and 1983–1986 United States Japan Europe 1969–1972 1983–1986 1969–1972 1983–1986 1969–1972 1983–1986 Food products 43.2 7.6 19.6 9.4 16.0 7.4 Chemicals 465.0 111.2 173.4 133.7 245.9 82.9 Pharmaceuticals 246.6 74.9 155.2 76.5 248.8 84.8 Metals 106.0 21.3 24.4 24.0 58.0 24.0 Mechanical engineering 273.9 68.9 16.5 32.2 113.5 57.3 Electrical equipment 296.1 90.4 142.2 148.2 151.5 55.3 Office equipment 325.6 82.2 206.5 276.5 179.6 40.3 Motor vehicles 84.7 26.1 43.1 73.0 73.5 28.8 Textiles 25.7 7.8 67.1 60.5 31.4 10.2 Paper products 62.1 22.4 2.5 8.5 16.3 6.2 Printing and publishing 16.3 3.3 10.1 24.5 1.7 0.7 Rubber products 166.5 50.4 79.8 82.2 34.2 6.1 Nonmetallic mineral products 215.5 63.5 118.8 39.0 43.8 15.5 Coal and petroleum products 126.7 18.8 7.9 4.2 58.4 7.5 Total manufacturing 182.8 41.5 67.0 70.3 100.5 30.8 SOURCE: Data base on the world's largest industrial firms held at the University of Reading.
OCR for page 180
Japan's Growing Technological Capability: Implications for the U.S. Economy company's patent to sales ratio had risen above the U.S. equivalent, just as Japan's ratio of nondefense R&D to GDP rose above the U.S. level in the early 1970s. In fact, the catching up of the Japanese (T'/Q) ratio may well have happened at around the same time at both the firm level and the country level, given that U.S. patenting is a technology measure that is weighted in favor of U.S. firms. U.S. firms are likely to have a higher propensity to patent in the United States than are firms based elsewhere. By 1983–1986, the average patent to sales ratio of the leading Japanese firms had risen above the U.S. company ratio in all industries except mechanical engineering, paper products, nonmetallic mineral products, and coal and petroleum products. U.S. firms were not very far behind, though, in food products, pharmaceuticals and metals. These are all branches in which U.S. firms had retained a reasonable rate of technological accumulation by international standards. Similarly, it is not surprising that the highest patent to sales ratios of Japanese firms (reaching around 150 patents per $1 billion of sales or greater) were obtained in the industries of their highest rate of innovation, that is electrical equipment and office equipment or computers. So the expected relationship between the rate of technological accumulation, growth, and the ratio of new technology to output seems to hold up not only at the level of national comparisons, but also when these are extended to the industry level. The innovative record of national groups of firms varies across industries according to their comparative advantage in technological activity. The rate of technological change of Japanese firms has been greatest in the electrical equipment, office equipment and motor vehicles sectors, and in these industries they have experienced the fastest growth and increase in market shares. It is also feasible to examine the comparative advantage of firms within industries, as measured by their pattern of specialization in technological activity compared to other companies in the same industry. For this purpose consider firms in the industries with the highest levels of patenting: chemicals and pharmaceuticals, taken together, and electrical equipment and office equipment, taken together. In the introduction the recent link between Japan's overall pattern of technological specialization (RTA) and the structure of technological opportunities (or the growth of patenting) was demonstrated. It is possible to carry out a similar exercise at the industry level, from an analysis of the patenting of firms in the chosen industry considered separately. To this end, total industry patent growth between 1969–1972 and 1983–1986 was regressed on the RTA of Japanese firms in the same industry at the start of the period. This depends upon the distinction between the classification of firms by the industry for which they produce and (in any industry) the classification of their patenting by types of technological activity. Denoting the growth of total patenting in a branch of technological activity i by
OCR for page 181
Japan's Growing Technological Capability: Implications for the U.S. Economy GTi, and Japanese RTA by RTAi, the form of the cross-section regression in each industry was as follows: As in Figure 1, patents were allocated to one of 33 types of technological activity (sectors i), but branches were excluded if firms in the industry in question had no activity in that sector. This left 25 sectors in the chemical and pharmaceutical industry, and 27 technological sectors in the case of firms manufacturing electrical and computing equipment. As shown in Table 2, the competitive strength of Japanese firms in terms of the beneficial nature of their technological specialization is clearly more evident in the electrical and office equipment industry than it is in chemicals and pharmaceuticals. In the electrical industry the RTA of Japanese firms in 1969–1972 was positively and significantly related to the distribution of subsequent patent growth in the industry as a whole from 1969–1972 to 1983–1986. In other words, Japanese firms specialized in fields in which, at an industry level, technological opportunities were at their greatest. In the electrical equipment industry the fields of greatest patent growth and high Japanese RTA were road vehicles and engines, and image and sound equipment. This is indicative of a crucial area of Japanese success under the new technology paradigm. Under a new paradigm pervasive technologies gradually help to transform industries outside those in which they were originally developed. In this case under Japanese leadership the infusion of new electronic technology reinvigorated the motor vehicle industry, which had previously been regarded as a ''mature" and noninnovative industry. The vehicles industry now symbolizes the transition between paradigms: from the production of vehicles intensive in their use of energy by scale-intensive methods, towards customer-designed vehicles produced by computerized systems. The major Japanese electrical equipment producers were geared up for this new technological opportunity at an early stage. TABLE 2 The Cross-Sectoral Regression of Total Industry Patent Growth from 1969–1972 to 19832–1986 on the RTA of Japanese Firms in the Same Industry tα tβ Chemicals and pharmaceuticals 12.189 8.164 0.53 0.50 Electrical and office equipment -22.777 52.310 -1.11 3.28a a Denotes coefficient significantly different from zero at the 1% level.
OCR for page 182
Japan's Growing Technological Capability: Implications for the U.S. Economy In the chemical industry the largest Japanese firms also specialized in electrical technology. The positive association between Japanese RTA and patent growth can be related to the fast growth in the technological activity of the major chemical companies in computers and image and sound equipment (presumably as means of reorganizing chemical production plants, with computerized control technologies that met their own specific objectives). However, the fit between Japanese RTA and patent growth was not significant as Japanese chemical firms had no specialization in the other leading fields of technological opportunity in the industry; namely, in agricultural chemicals, pharmaceuticals and biotechnology. So the specialization of Japanese firms gave a general impetus to their overall rate of technological accumulation, but the effectiveness of this varied across industries. It has been at its strongest in industries in which technologies can be most clearly related to electrical and computing systems. There is another way of viewing this connection, however. As the new paradigm has taken shape, so the composition of technological opportunities has been gradually changed in each industry. These opportunities have steadily shifted in favor of the key fields of the new paradigm, in which high growth has already been experienced in the leading industries, as spillover benefits from these industries and the potential for new forms of technology fusion begin to influence other industries. If Japanese firms have led these changes, then they can be expected to have been the first to have taken advantage of the new opportunities created in a broad range of industries. To test this, the growth of patenting across different fields of activity was compared in two different periods. Considering again firms in the electrical and chemical industries, total industry patent growth between 1978–1982 and 1983–1986 was regressed on the growth of patenting by Japanese companies in the same industry between 1969–1972 and 1973–1977. Denoting the growth of Japanese firm patenting in the technological field i by GJi, and using subscript t and t-1 for the later and earlier periods respectively, the simple cross-section regression was Excluding sectors with very low levels of patenting left 24 branches of activity in the chemical case and 23 in the electrical and computing equipment industry. The results are reported in Table 3, and they illustrate a further element of the technological leadership already being exercised by Japanese firms. Even in the chemical industry the fields in which Japanese companies concentrated their efforts in the early 1970s were significantly related to those that were to become the major areas of technological opportunity in the industry in the early 1980s. Of course, this might only mean that technological opportunities were
OCR for page 183
Japan's Growing Technological Capability: Implications for the U.S. Economy Table 3 The Cross-Sectoral Regression of Total Industry Patent Growth from 1978–1982 to 1983–1986 on the Growth of Patenting by Japanese Firms in the Same Industry from 1969–1972 to 1973–1977 tα tβ Chemicals and pharmaceuticals -26.656 0.180 -3.14a 4.47a Electrical and office equipment -13.531 0.136 -1.83b 2.57c a Denotes coefficient significantly different from zero at the 1% level. b Denotes coefficient significantly different from zero at the 10% level. c Denotes coefficient significantly different from zero at the 5% level. greatest in the same branches of activity in both periods. There is an element of truth in this in the electrical and computing equipment industry. However, Japanese firms still demonstrated some leadership in taking up activities within the electrical industry at this time. The pattern of patent growth in Japanese companies early in the period provided a better explanation of the distribution of total patent growth at the end of the period than did total patent growth at the start of the period. Total electrical industry patent growth from 1978–1982 to 1983–1986 was positively correlated with the equivalent Japanese patent growth from 1969–1972 to 1973–1977 at the 5 percent level (see Table 3), but it was correlated with total industry patent growth in the earlier period only (just) at the 10 percent level. In the chemical industry, though, there is no correlation at all between total patent growth in the two periods; the distribution of technological opportunities underwent a substantial change. This is consistent with the view that a new technology paradigm begins by affecting leading industries before moving out to influence others. While the new fields of technological opportunity had already become fairly settled in the electrical industry between the 1970s and 1980s, opportunities in the chemical industry began to shift at that time. Japanese firms seem to have helped lead this switch. In the electrical equipment industry the fields of high patent growth in both 1969–1977 and 1978–1986 have already been mentioned. These are road vehicles and engines and image and sound equipment. Japanese firms were not only specialized in these areas in 1969–1972 (as commented on above), but they also witnessed very fast growth in the same areas through to 1973–1977, so that the extent of their specialization (RTA) in these fields actually increased. In the chemical and pharmaceutical industry, the technological activities in which Japanese companies expanded their interests fastest, to be followed subsequently by their competitors, were in computers, image and sound equipment, and electronic communications. This illustrates how the
OCR for page 184
Japan's Growing Technological Capability: Implications for the U.S. Economy effects of the new technology paradigm had begun to influence the pattern of technological change in the chemical industry by the early 1980s. The leading Japanese chemical companies, perhaps drawing on collaborative arrangements with the Japanese electrical industry, were in the forefront of initiating this new development. The change in technology paradigm may have assisted the rate of technological accumulation of Japanese firms in some industries more than others (in electrical equipment more than in chemicals), but it has given Japanese companies some impetus in most industries. SOME CONCLUSIONS AND SUGGESTIONS ON THE IMPLICATIONS FOR JAPAN'S FUTURE TECHNOLOGICAL COMPETITIVENESS On this subject it is perhaps rather dangerous to speculate too much about the future. However, some suggestions do emerge if bets about the future must be made. The most significant is that the rate of growth of Japanese industry and of the leading Japanese firms is likely to continue to exceed the equivalent U.S. growth so long as the Japanese sustain a higher rate of technological change; and the Japanese rate of technological accumulation is likely to remain higher so long as the now prevailing technology paradigm continues in place. The Japanese national system of innovation and the accompanying pattern of technological specialization of her firms have come to represent the very expression of this paradigm. By contrast, U.S. institutions remain to some extent locked into structures associated with a previous era of technological opportunities, just as British institutions had been at the turn of the century.21 As technological change is cumulative and incremental, organizational routines adapt only slowly, and institutional structures can only change gradually, countries and firms become locked into some technological course. The distribution of technological opportunities then favors some countries and firms rather than others. In the near future the current technology paradigm with its particular spread of opportunities is likely to be consolidated further rather than reversed, and this will work to the advantage of the Japanese. The changes that will occur in the composition of technological opportunities are likely to be mainly of an incremental kind, moving from the fast growth fields of the 1980s into related areas. Japanese firms are likely to lead such incremental shifts. So long as they do, Japan will open up a new lead not only in its rate of technological accumulation but also in its overall level of technological capability. The model described above suggests that there is no relationship 21 Cantwell, op. cit., footnote 12.
OCR for page 185
Japan's Growing Technological Capability: Implications for the U.S. Economy between the proportional rate of technological accumulation (and hence growth) and the current level of technological capability. Just as there was no necessary reason why the United States had to have the highest rate of innovation when it had the largest technological capability, so now there is no reason why Japan should not continue to sustain its high rate of technological change with the greatest overall level of capability. It is only likely to fall back if there is another change in the technology paradigm that provides a new window of opportunity to others on a different technological course. This offers a perspective at odds with those who have claimed that Japan's high rate of innovation was due to having a lower level of technological capability than the United States, offering the scope to catch up rapidly through the import and adaptation of foreign technology. A rather more sophisticated variant of this argument is that the United States still retains the leadership in science even if she has lost it in technology, so Japan remains dependent on her ability to commercialize U.S. scientific achievements. These contentions are misleading insofar as they rely on what has recently been called (by Rosenberg, among others) the linear model of technological development. In the linear model, there is a unidirectional causal chain that runs from scientific advance to technology pioneered by an innovative leader, to diffusion to a wider circle of firms and to other industries. In fact, there is more likely to be a regular interchange between the problem-solving activities of different companies, combining to generate a series of complementary technological improvements. In the process they all contribute to an underlying stock of generic knowledge that is to some extent held in common. Even the direction of scientific advance comes to depend upon the issues raised through technological problem-solving activity, or more directly upon technology itself (as in the case of the creation of new and more precise scientific instruments). So if Japanese technology is dependent upon U.S. science, so too is U.S. science dependent upon Japanese technology. To put matters another way, the distinction between innovation and imitation is blurred to the point where it may be analytically unhelpful. Japanese firms were more successful in their imitation than others that lay the same distance behind the ''technology frontier" because this imitation went alongside their own high rate of technological innovation. This is illustrated by the study of patent citations by Narin and his colleagues, who have shown that Japanese patents have on average been of higher quality than others; they do not simply represent minor adaptations. Since, as outlined in the introduction, technology is localized, imitation is often just as costly as innovation, and it is sometimes more so (where the imitator begins from a technological base that is poorly related to the field in question, or
OCR for page 186
Japan's Growing Technological Capability: Implications for the U.S. Economy where it is operating in a less helpful environment).22 The imitation of related technologies supports the cumulative development of the firms' own technology; imitation and innovation are complements rather than alternatives. The interplay between imitation, innovation, and science and the increased importance of basic R&D in the new paradigm help to explain why Japanese firms have been investing heavily in basic research to enhance the further development of their strengths in production engineering, computerized systems and new organizational methods. Firms carry out basic research partly to increase their ability to monitor their external environment, and to help them identify opportunities that might otherwise be missed.23 Although generic knowledge is not itself responsible for lasting differences in technological competence between countries and firms, a capacity to generate and understand such knowledge helps to support local competence. Moreover, even though in the core fields of the new technology paradigm Japan's absolute level of technological capability may exceed that of the United States, the scope for Japanese imitation is not over. Since technological development is localized and differentiated, the course followed by Japanese firms is to some extent different from the path of U.S. firms in the same industry. Therefore, they still have much to learn from one another, whether through agreements for technological cooperation or otherwise. This helps to explain the trend toward international production by the leading multinationals in foreign centers of excellence. The U.S.-located affiliates of Japanese firms provide their parent companies with a stream of complementary technologies, derived from the local characteristics of U.S. production. 24 European firms in the United States and U.S. firms in Europe have followed similar strategies for international technological development for some years. Mutual strategies for imitation do not in themselves provide a threat to an innovative leader. Firms that sustain a higher rate of technological change will also tend to have a greater capacity to imitate others where appropriate and to build upon opportunities in fields related to their own. It may well be in the Japanese interest to help to promote a wider diffusion (through localized adaptation) of the technologies they have pioneered. This may improve the positive interaction between U.S. and Japanese firms, 22 E. Mansfield, M. Schwartz, and S. Wagner, "Imitation Costs and Patents: An Emipirical Study," Economic Journal, vol. 91, no. 4., 1981. 23 W.M. Cohen and D.A. Levinthal, "Innovation and Learning: The Two Faces of R&D," Economic Journal, vol. 99, No. 3., 1989 and N. Rosenberg, "Why Do Firms Do Basic Research (With Their Own Money)?" Research Policy, vol. 19, no. 2., 1990. 24 Cantwell, op. cit., footnote 15 and B. Kogut and S.J. Chang, "Technological Capabilities and Japanese Foreign Direct Investment in the United States," Review of Economics and Statistics, vol. 73, no. 3, 1991.
OCR for page 187
Japan's Growing Technological Capability: Implications for the U.S. Economy allowing them both to raise their rate of technological accumulation. While mutual imitation between U.S. and Japanese firms, whether through cooperative agreements or otherwise, increases their rates of technological change, it will not necessarily affect the differential between them. However, in his contribution to this volume Professor Nelson suggests that such greater international interchange will also reduce the differential in rates of innovation between countries or national groups of firms, but the basis for this argument is essentially that in the close combination of imitation and innovation there has been a shift towards imitation and intercompany linkages. The growth in cross-border exchanges between firms, formalized in some cases by strategic alliances, may be partly explained by the increasing interrelatedness between formerly separate types of technology, such that imitation becomes an even greater mutual benefit or necessity. Although Japanese firms are likely to continue to sustain an overall rate of technological accumulation higher than their major rivals in the immediate future, this will vary across industries. They are likely to continue to lead in electrical equipment and motor vehicles, but they will not necessarily do so in chemicals, even if their rate of technological change has risen relative to their competitors (the new paradigm has had an effect). Freeman has discussed the likelihood of Japanese firms further raising their rate of technological progress in the chemical industry.25 Whether they manage this or not, it seems unlikely that the overall pattern of technological specialization of Japanese companies will change too dramatically. Given the cumulative nature of technological change, they are likely to achieve more success in building upon their core strengths in electronic-related technologies than they are liable to accomplish in a new and largely unrelated field such as biotechnology. Very little has been said so far about technology policy. This is partly because a great deal has been written on Japanese technology policy elsewhere, and on its role in helping to ensure Japan's leadership in the new technology paradigm.26 Japan has for some years emphasized long-term industrial policies rather than policies of short-term macroeconomic management. She has emphasized investment in education, training, and the scientific and allied infrastructure. The Japanese government has regularly collaborated with industry to try and ensure that the longer term market forces they have identified work to their advantage. This compares favorably with the prevalent attitude in the United States and Britain especially in the 1980s, which has held that governments should steer completely clear 25 C. Freeman, "Technical Innovation in the World Chemical Industry and Changes of Techno-Economic Paradigm," in C. Freeman and L.L.G. Soete, eds., New Explorations in the Economics of Technical Change (London: Frances Pinter, 1990). 26 Freeman, op. cit., footnote 6.
OCR for page 188
Japan's Growing Technological Capability: Implications for the U.S. Economy of private corporate decisions, and allow markets and firms to operate entirely independently. The result is that where strategic government support for industry is required to foster technological development, the U.S. government may come to pick up the signals to this effect only late in the day: "American government officials and businessmen negotiating economic matters feel at a great disadvantage because Japanese officials are much better informed, not only about Japanese companies, but often about American companies."27 Economists and policy advisors tend to overemphasize the efficacy of their policy recommendations, since it is in this way that they persuade their audience and ultimately the policymakers. It seems unlikely that through policy changes alone the United States could raise the rate of technological advance in her industry to Japanese levels. The structure of institutions and local technological competence is different in the United States, and it is simply much more costly for her to undertake the deep shift required to bring her closer to a structure that better fits the new technology paradigm. However, this is not a recipe for doing nothing, as under these conditions the U.S. position may deteriorate further. With the appropriate technology policy; support for science, education, and training; encouragement of new organizational structures and industrial relations systems; and new forms of association between finance and industry, the costs of the transition to the new technology paradigm will be reduced, the rate of technological change will rise, and longer-term benefits will follow. This is one of the lessons of the Japanese experience. 27 E.F. Vogel, Japan as Number One (Tokyo: Turtle, 1980) cited in Freeman, op. cit., footnote 6.
Representative terms from entire chapter: