تاثیر انتقال فن آوری و تحقیق و توسعه در رشد بهره وری در صنعت تایوان: تجزیه و تحلیل اقتصاد خرد با استفاده از اطلاعات در سطح بنگاه و کارخانه

This paper conducts a comparative empirical analysis of the impact of R & D spending and purchases of foreign technology on output and productivity in Taiwanese industry. We employ data from two different sources, providing an econometric perspective on this question at two different levels of aggregation. We first conduct empirical analysis using data from the Taiwanese government's industrial census of technological activities at the plant level. This study is, to the best of our knowledge, the first empirical analysis using these data. We complement these results with analyses of data at the firm level. The results of our regression analyses generally support the conclusion that both R & D spending and purchases of foreign technology have contributed positively to Taiwanese productivity growth. J. Japanese Int. Economies20 (2) (2006) 177–192.

Recent analyses have emphasized the skill of Taiwanese firms in obtaining and successfully utilizing technology developed abroad. This “absorptive capacity” is believed to be a key element in Taiwan's export success in increasingly technology-intensive manufacturing industries. Our paper undertakes an empirical study of formal technology purchases from foreign sources in Taiwanese industry, in an effort to better understand the role this foreign technology has played in Taiwanese technological development and economic growth. We also examine the role played by the R & D spending of Taiwanese firms. Using data from the Taiwanese government's industrial census of technological activities, we conduct empirical analyses of these issues at the plant level, with a focus on Taiwan's dynamic and competitive electronics industry. This study is, to the best of our knowledge, the first empirical analysis using these data. We also examine these issues using data at the firm level, drawing upon a sample of publicly traded firms that is not limited to, but heavily populated by electronics firms. Our empirical results to date suggest that the impact of foreign technology imports on productivity growth at the plant level have been positive and significant. Furthermore, the industries and plants that spend the most on R & D also tend to spend the most on technology imports. We acknowledge at the outset that formal technology purchases from foreign firms are only one of the means by which Taiwanese firms have acquired foreign technology. Other important channels include the technical information provided by foreign purchasers through “OEM” contract manufacturing agreements, the firms' own efforts at reverse engineering and those undertaken by government-affiliated research institutes, and the “embodied human capital” which thousands of Taiwanese students brought with them when they returned to Taiwan after graduate study abroad. Unfortunately, none of these other channels leave a “paper trail” by which economic analysis might hope to measure knowledge flows and estimate their effects. Formal technology purchases do leave such a paper trail, and we exploit that fact in this paper. 1 We believe this paper will contribute to the empirical literature on purchases of foreign technology by industrial firms in developing countries. Earlier work by Braga and Willmore (1991) examined the cross-sectional determinants of technology imports in a sample of Brazilian establishments, using data drawn from a survey undertaken in a single year. The study also examined the impact of technology imports on “technological effort” and quality control. Because data were only available for a single year, this study was unable to include firm fixed effects in its analysis. The study by Basant and Fikkert (1996) examined the impact of R & D, formal purchases of foreign technology, and a number of other variables on the productivity of Indian firms from the mid 1970s through the early 1980s. While the Basant and Fikkert study possessed many advantages over the Braga and Willmore study in terms of data availability, it is also set in the context of a country whose government intervened heavily in the market for technology. Both Brazil and India placed significant limits on the activities of foreign firms. Furthermore, the intellectual property regime in both countries was intentionally designed in ways that limited the ability of incumbent patent holders to negotiate from a position of strength with licensees. The exchange rate regimes in both countries also had features which arguably distorted industrial firms' technology purchase decisions. It is also fair to say that neither Brazil nor India experienced a great deal of success—at least over the sample periods examined in these papers—in terms of closing the technological gap with the West. As we noted at the outset, Taiwan is one of the world's resounding successes in terms of successful technological adoption.2 Taiwan's industrial and technological dynamism was achieved under substantially different conditions, including an FDI regime that was more open to foreign multinationals and an intellectual property regime that, at least from the late 1980s, was arguably stronger than that prevailing in the contexts studied by these earlier papers. For these reasons, we think a study of Taiwan may prove informative

The results from our micro data suggest that, despite the ongoing increase in R & D spending, which itself reflects an increasing technological maturity on the part of Taiwanese firms, technology imports still “matter” in terms of explaining the productivity growth of Taiwanese firms. One might have expected the Taiwanese electronics sector to be divided into a technologically “backward” group of subsectors that did little R & D, relying extensively on foreign technology, and a technologically “progressive” group of subsectors that relied principally on its own R & D. Instead, the most R & D-intensive plants and industry segments are also the ones responsible for the vast majority of foreign technology purchases. Perhaps the success of Taiwan's leading high-tech firms comes not from “technological independence” from the US and Japan, which they have manifestly failed to achieve, but by pursuing R & D that complements the technological developments in these two major Pacific technological superpowers. 26 One could think of this as a policy of “technological integration” rather than technological independence. If this view is correct, then it suggests that efforts by developing country governments to “wean” their firms of dependence on foreign technology are misguided. Several next steps immediately suggest themselves. One is to look at the relationship between technology purchases and capital accumulation. The work of Young (1995), Kim and Lau (1994), and others has suggested that it is largely physical and human capital investment, rather than productivity growth per se, which has been the most important driver of economic growth in East Asia. Krugman (1994) has used this result to argue that there is no Asian miracle. However, this result merely begs the question of why the East Asian nations have been able to invest so much more than other developing nations. Is it possible that technology purchases increased the efficiency with which capital could be used, increasing its marginal product? If so, then we may be able to observe, at the micro level, a link between capital intensity and technology purchases. One interesting next step would be to explore this linkage using a variety of empirical techniques. Another interesting next step would be to link the plants in our data set together into business groups. We have treated the plants in our data set as autonomous economic agents selling into a competitive product market, but many of the plants are actually units of larger enterprises. It seems reasonable to expect that at least a fraction of the useful technology purchased by one plant within a multi-plant firm (or a multi-firm business group, such as the Tatung Group) or generated through R & D spending, would be deliberately transferred to the other plants within the firm. This would tend to weaken the econometric relationship between R & D or technology purchases estimated at the plant level. But by grouping plants together into firms, we could begin to examine these technology transfers within Taiwanese enterprises. All of these next steps will require considerable additional investment in data collection and processing, but the payoff to this investment is likely to be quite high. Understanding how Taiwan's electronics industry managed to develop so quickly is important, not only for that country's political leaders, but also for other small open economies around the world who seek to duplicate Taiwan's success. We hope that the research agenda outlined in this section will bring us closer to that level of understanding.