|
|
|
|
Science-Based Economic Development edited by Susan Raymond
Lewis M. Branscomb
John F. Kennedy School of Government
Harvard University
This conference explores one of the several paths to development, a path in which scientific institutions play a significant role. In order to gain some insight into the importance of science in the development process, whether in highly industrialized societies or in those that are still struggling, it may be useful to explore some examples. Let us begin with two technical giants: the U.S. and the U.S.S.R.
The U.S. path to development in the first part of this century was based on bountiful natural resources, fertile land, a large integrated market, a relatively stable and open political system, and a people suspicious of old-world dependency but eager to borrow or buy the old world's technology and industrial knowledge. American scientists went back to Göttingen and Munich, to Oxford and Cambridge in the 1920s to learn the new European science and bring it home to be put to use.
This is the secret of American success: natural endowments and a culture that admired the practical and innovative. U.S. science bloomed early in the culture of Franklin and Jefferson, but not in the more academic European form. Rather it flourished in a home-grown style strongly imbedded in invention and technology. For a hundred years we can find a few outstanding native scientists of renown; J. Willard Gibbs comes to mind. But we can find a substantially larger number of immigrants such as Nikola Tesla, who designed much of George Westinghouse's electric power equipment.
Science started playing a really big role only in World War II, but for security, not development. Science was not institutionalized as an instrument of national development policy until about 1970, when incursions into U.S. markets by Japanese products of superior quality and lower cost shook up American entrepreneurs. But even then it was not science to which the firms turned; it was production and quality management, as well as improvements in productivity throughout the firm. A casual examination of congressional debate over U.S. development policy suggests a deep lack of consensus on whether, even today, government should make investments in research and development in support of economic growth.
How about the Soviet Union? Based on a form of state socialism that its supporters claim is more rational than capitalism, this society consistently allocated a high priority to science as a tool of national political and economic development. The USSR enshrined its most accomplished scientists in privileged positions. The Soviet Academy enjoyed enormous prestige; its members received a substantial income bonus simply for having been elected.
However, the USSR can hardly be cited as a successful case of science-based development. Or, if it is, then certain other conditions that are necessary for science to support development were clearly absent. One of those conditions is a free market economy, without administered prices and production, a floating currency, and reasonably open trade. In short, the USSR had the scientific capability, but it was not imbedded in a society that was able to make use of it in competition with free market countries.
Yuri Arbotov, in remarks at Harvard in April 1996, pointed out that a free market may indeed be a necessary condition for science-based development in Russia, but even it is not sufficient.
Functioning economic institutions that trust one another, an infrastructure for information and distribution that is flexible and inexpensive and reliable, an attitude toward the balance of personal reward and collective well-being that is conducive to responsible economic behavior-these are the requirements for science-based development. These attributes can be lumped under the general heading of "social capital." In short, Russia is in danger of losing its science before it has rebuilt the social capital that will allow them to "capitalize" on it.
How about Japan, Korea and Taiwan? Surely their rapid development was technology-based. All three relied initially almost entirely on imported technology but Japan enjoyed a much stronger and earlier capability for absorption and adaptation of imported technology. Japanese educational and industrial institutions were already at the world level- or close to it- before World War II.
Today, even as Korean expenditures on internal R&D are skyrocketing, Korean industry still spends 20 to 50 percent of the technology cost of new products on payments for imported technology and tools. 1 The Korean government plans that the national expenditures on R&D will surpass 4 percent of GDP by the turn of the century, heading toward 5 percent. Their goal is nothing less than to pass the U.K. in per capita economic performance and thus qualify to sit with the G-7 nations in discussions of the world economy. Thus Korea's economic miracle was not based on indigenous science- not at all. It was based on a high concentration of capital invested in imported technology and tools, pursuant to a strategy set by authoritarian governments, and resting fundamentally on a highly motivated Korean people who are determined to put their colonial past behind them and enter the family of leading free nations. (It is worth noting that the Koreans plan to hit 5 percent of GDP- mostly private investment- in about the same year conservatives in the U.S. House of Representatives want to see U.S. government nonmilitary R&D shrink from 1.8 percent of GDP to about 1.0 percent.)
Japan rebuilt her industries using the talents of a gifted and highly motivated group of technical leaders, mostly elite younger engineers who were protected from field service in the military in World War II and developed Japan's advanced weaponry. This is a story they are quite proud of; they see it as Japan's success with a "dual use" technology strategy. While Japan's development was based on incremental, adaptive innovation from the beginning, it can be said to be science-based now. Having caught up with North America and Europe technologically, Japan, like its competition, must look to higher value added in the economy, more reliance on advanced skills, and less reliance on raw materials, cheap capital, and cheap, compliant labor.
Taiwan is an equally interesting case, an extraordinarily successful economy based primarily on a highly productive set of small and medium-sized enterprises, many of which benefit from Chinese family networks. These networks consist of sons and daughters with Ph.D.s from U.S. universities and businesses in the U.S., perhaps an uncle in Hong Kong with access to large amounts of capital, perhaps through a brother in a Singapore bank, and other members of the extended family with low case manufacturing facilities in Taiwan or Malaysia, or indeed in mainland China
Finally we should observe that Japan, Korea, Taiwan, China, India and Iran (before the Ayatollah) shared one other attribute: they sent their best science and engineering students to the U.S. for advanced degrees and made a substantial effort to get them back home afterwards.
From this brief tour through a few of the success stories, let me tell you my conclusion, then go into just a few points in more detail: What we must mean by "science-based development" is a strategy for development that rests on human and social capital in a global economy. Institutional capacity, efficient relationships based on trust, a functioning market for intellectual assets- these are as important as Nobel Prizes, ingenious inventors, and great engineers.
The reason I qualify the definition with the words "in a global economy" is that cheap transportation and efficient information infrastructure tends to equalize the cost of raw materials and of capital all around the world. Furthermore it permits a degree of access to knowledge, to technology, and to methods of production and management that allows the rapid diffusion of industrial capability. Those who would differentiate their products and services from the competition can start from a moderately level playing field, but must exhibit superior performance in innovation rate, in service, in cost, and in responsiveness to market opportunities.
As Pavitt and Patel point out in an interesting paper, 2 the economic theory prevailing in the 1960s predicted that buoyant demand and an open trading system would allow the international (and domestic) diffusion of technology, and this would lead to equalization of technological performance at the national level. This prediction was based on a flawed model of science-based development, indeed of technological change. It presupposed that:
If this model were equally applicable to all countries in a similar state of development, it would follow that through markets for machinery, free access to codified technical knowledge, and a rapid process of learning by doing, the gaps between the U.S. economy and those of Japan, the U.K., Germany, and France should have closed rather rapidly.
It has not happened. Japan and Germany have moved ahead, and the U.K. and France have fallen behind. Taiwan, Korea, and Singapore have leapt ahead from a very backward state thirty years ago. Brazil, Mexico, and India have failed to do so, although they show signs of progress recently.
These three assumptions in the 1960s economic model have persisted, in more recent years, in the form of a cold war paradigm of the innovation process. It was widely believed in the U.S. that one could gain commercial advantage by force-feed ing industrial S&T development in pursuit of government missions. This is the "spinoff" theory of defense industrial research. Defense research investments would bring into existence new machinery, abundant and accessible new technical knowledge, and the private sector could quickly assimilate the imbedded knowledge by "learning by doing."
All three of the assumptions are, of course, reasonable pictures of the technology diffusion process. The problem is that the efficiency of each of these processes varies strongly from one institutional setting to another. Patel and Pavitt conclude that technology diffusion, productivity learning, and transfer of embodied technology have large transaction costs when they encounter cultural, managerial, and institutional barriers. It is the absence of these barriers that represents "social capital."
Where will one find examples of high levels of social capital? In the relatively wealthy, well industrialized smaller nations of Europe- Switzerland, Sweden, and the Netherlands, for example. As Henry Ergas points out, these small countries compete with Germany, Japan, and the U.S. quite well. They do it with an industrial structure that has a few very large, global firms (think of Phillips, Hoffman La Roche, ASFA Brown Bovari) supported by a large, diverse, and technically very advanced group of small and medium firms collaborating and competing through a system of industry associations that manage the markets that are critical to the economy. The level of trust is high; the government supports the industry strategies without dispute; the business transaction costs are low. As relatively small, mature societies, these countries have the social capital that allows them to concentrate their resources, gain efficiencies that larger economies may not have, and attack world markets successfully.
What are the social capital requirements for science-based development? What is the relevant venue for assessing the adequacy of social capital? Requirements vary both in space and time, and it is not obvious that the nation-state is the right unit of analysis. Nor do all parts of an economy advance at the same rate. There is a popular notion of a unit of economic analysis called the "National System of Innovation," best known perhaps from the book edited and authored by Dick Nelson. 3 To the extent that governments play a critical role in guiding and financing economic development, they will understandably focus on trying to benefit their own citizens. However we see many interesting case studies that reflect a strong local or regional focus to economic development in the U.S. Studies show that indeed, university-industry collaboration is preferentially concentrated on firms no more than a hundred miles from the university. 4 While an important fraction of technical talent is highly mobile, most students look for work in their own region, and increasingly so as both members of a family are professionally employed.
In a small developing economy there are inverse advantages to scale. Singapore and Hong Kong leap to mind. What strategies are available to large countries like Brazil, India, or China, which can ill afford to develop all parts of the economy at the same rate?
The Chinese have a system of government that allows the leadership to declare, simply, that certain regions on the East Coast are to receive preferred treatment in capital allocation, in access to FDI, in mobility of the work force, and in access to overseas markets. The center of the nation is to be the breadbasket and natural resource economy with a lower wage structure. The West hosts huge military regions that hold the local standard of living at historic levels.
Democracies cannot get away with that strategy of deliberately favoring one part of the economy at the expense of others in order to concentrate the investment in both social and economic infrastructure, at least not as politically overt policy. Some, however, pursue it de facto. Compare the state of Sao Paulo, Brazil, with the Amazon Valley; there are huge differences in per capita income, in infrastructural development, in human resources and in economic and technological institutions. If human and social capital are in short supply, one obvious strategy is to try to create these conditions in a restricted area and hope that the institutional and human learning will diffuse to other parts of the country. While they are not a product of national policy, one can say that Silicon Valley, Route 128, and the Research Triangle represent analogous strategies of focused development within California, Massachusetts, and North Carolina, just as China has its Hong Kong and Southeast Asia its Singapore.
Development strategies also vary in time. Sri Lanka, Thailand, Korea, and Spain are all in stages of strong development, but at very different stages. What role does science- or indeed universities- play in each level of development? I will not attempt that analysis now; it is too lengthy a process. But it is my view that there is one common denominator to the role of science in all these levels. It is a view that is almost never reflected in development economics thinking.
What is the role of science itself in the development process? It is my contention that in all levels of economic activity the most valuable contribution science can make is to inform decisions. A secondary role is the creation of indigenous technologies or technological improvements.
Often those engaged in scientific research and who, through that work, are very well-informed about the future evolutionary tracks of technology, can inform technology decisions. Having spent fourteen years as IBM's chief technical officer, I can assure you that the most valuable contribution of the IBM's TJ Watson Research Laboratories was to help the company understand the likelihood of alternative technological trajectories and try to get the company on the right one. In this case a few scientists' informed judgement can make the difference between a billion dollar triumph and a billion dollar disaster. Of course someone must listen to them, and the scientists must be able to give advice in context, i.e., understanding costs, markets, and innovative demands and capabilities.
I would assert that a quite similar situation exists when a Korea or Thailand is deciding what their high-tech strategy should be, even though the strategy does not involve attempting to exploit indigenous innovations based on in-country research. Thus research-informed technological choice is probably more critical in countries or regions at an early stage of development than it is in the firms of a highly industrialized nation. But where will we find this talent and how can we insure that it can give advice in context, and that the advice will be listened to?
This brings me to the universities, and a kind of good-news, bad-news conclusion. The research university is a remarkably flexible institution and can be critical in the national development process. Within one Institution one may have the full range of knowledge, from science to engineering to management and economics, even to public policy, law and social analysis. In most countries there are traditional protections for radical, or at least challenging thinking in universities- in Europe with long-standing historical roots, in Asia through the high level of respect Confucian societies accord to higher learning.
Unfortunately, possessing all these skills and the independence to break out of conventional thinking does not create the integrated capability, the political sophistication, or the knowledge of how private industrial innovation works. Indeed the independence of the university can become both an asset and an impediment to the level of trust that both industry and government must have if they are to benefit strategically from university resources.
If there is a solution to this dilemma, it is likely to require a great deal of local adaptation. In Korea, the most prestigious university, without question, is Seoul National. Its graduates enjoy the kind of entrée into key jobs and the support of a "school tie" network that is rivaled only by les Grandes Ecoles in France. But the Ministry of Science created its own university- KAIST- modeled on MIT, which is enormously successful. KAIST produces most of the science and engineering doctorates in Korea and contributes importantly to Korea's high-tech industry. A second model, even more impressive in that it is a private university created by a single large industrial corporation, is Pohang University of Science and Technology.
For universities to serve the goal of development, both their levels of expertise and the linkage mechanisms to the private sector and the government must be appropriate. In some cases their level of sophistication in technology (which they must enjoy to access world knowledge) is a serious mismatch to their knowledge of business institutions and processes, and it impairs their ability to help. In others their knowledge is too high-level to be assimilated by the firms they attempt to help. This is especially likely for small and medium-sized establishments (SMEs) in traditional manufacturing industries.
In a recent World Bank study of how small to medium firms get S&T support in eight countries, it was found that local industrial extension services, sometimes offered by business or trade associations were most helpful. Government-funded S&T institutions tended to be remote from the firms' problems (although the industrial extension service institutions may benefit from support by sophisticated technical institutions). Indeed, when one looks at the SME's that support the more technically advanced industries, the unit of government most relevant is usually provincial or local. The U.S., Japan, and many other countries have found this to be true.
However, those larger, more successful firms that can afford a corporate research lab may indeed use leading universities or government laboratories as a window on the world of S&T. An excellent research university in the neighborhood can amplify that window to a house full of windows from which to assess the state of the art in many fields. Those relationships, however, must be developed- through students employed in the firm, through joint studies or research, and through personal consulting by faculty through which they may gain significant incremental income.
In the end, the university must become an important node in the social capital of the country. Its faculty and students represent one of the few concentrations of individuals motivated to identify, and even to stimulate radical change, including new institutional relationships. The good-news bad-news dilemma, of course, is that the university owes this change-inducing capability to its isolation from the vagaries of politics and markets. But protecting that independence from the pressures of society makes universities understandably reluctant to engage fully the problems of society for fear of losing the very attribute that makes it valuable.
To summarize, science-based development does not imply development based on products that derive primarily from indigenous science. Indeed, I do not know a single example of a country that fits this description, certainly not the U.S. But it does imply smart choices of technology made by open-minded, innovative, and trusting people working in institutional settings that are appropriate to the development conditions at hand. In seeking to create this kind of infrastructure or social capital, nations must remember that just as innovators create rapid, sometimes radical change in institutions and their roles, and create entirely new demands for human skills, the educational institutions that can drive this change in the human resource base take decades to develop and change. Thus a high level of intellectual effort- call it science if you wish- needs to be applied to anticipating the revolutions in order to prepare to support their further evolution when the time comes. Very few universities can claim to have even attempted this, much less feel that they have succeeded.
Note 1: Branscomb, Lewis M. and Young-Huan choi, Korea at the Throing Point: Innovation- Based Strategies for Development (Greenwich, CT: Praeger Press, 1996). Back.
Note 2: Patel, Pari and Kieth Pavitr, "The Nature and Economic Importance of National Innovation Systems." STI Review, No.14, pp.9-32. Back.
Note 3: Nelson, Richard R. National Systems of Innovation (New York: Oxford University Press, 1993). Back.
Note 4: Jaffe, Adam B., "Real Effect of Academic Research." The American Econmic Review, Back.
Email this citation