The Production System

SET policy must address the production system specifically, not just by focusing on the commercialization of research results. In this area it must address the following issues:

• a deliberate effort to harmonize SET policy with trade, foreign-exchange, fiscal, monetary, and other economic policies to ensure synergy and mutual reinforcement (this is what the East Asian countries, Japan, South Korea, and Taiwan, seem to have done remarkably well);
• a targeting of policies to ensure the commercialization of research results;
• technological forecasting and a deliberate choice of a niche for the nation in the global economy;
• a deliberate choice of development strategy and the design of appropriate policies to implement that strategy—import substitution or export orientation;
• methods of infant-industry protection to foster targeted industrial technologies—i.e., creating winners, not merely choosing winners, since some new industries do not have a chance to win from their infancy without support;
• policies that will foster the growth of a dynamic capacity to manage technological change;
• industrial property protection;
• strategic alliances among private sector firms and between the private sector and the government;
• issues of the policy process on matters of SET, especially the extent of participation of all stake holders in the policy process;
• deliberate policies that foster innovative behavior, interfirm competition, and penalizes non-innovative behavior; and
• government procurement policies for the promotion of targeted technologies.

The Consequences of Technical Change

SET policy must address the pressing developmental issues of society and seek to minimize the adverse consequences of technical change while exploiting the opportunities that new and emerging technologies offer. SET policies in this area must:

• address the basic needs of society—food, security, shelter, basic health, infrastructure, etc.;
• seek to alleviate poverty;
• reduce the negative effects of technical change on the environment in terms of pollution, desertification, environmental degradation, waste management. etc.;
• be gender-sensitive in order to minimize any negative consequences;
• anticipate and exploit the opportunities of new and emerging technologies such as:
  • New materials
    titanium steel alloys; graphite and epoxy composites; doped silicons; aluminum; ceramic foam; spider silk (which is five times stronger, gram for gram, than steel).
  • Information revolution
    virtual reality; universal personal telephones; information super highways.
  • High-temperature superconductors
  • Biotechnology
  • Genetic engineering
  • Voice-activated computers
  • Optical electronics
  • Nanotechnology

Regional Development

While the world is fast turning into a global village, a number of regional alliances and trading blocs are emerging. SET policies must address the issues that they raise. National SET policy in this area must deal with:

• joint development of strategic technologies that promote the particular niche chosen within each region
• SET policy harmonization within the various regions;
• the harmonization of SET policies with macroeconomic policies within the regions or trading blocs; and
• strategic alliances of firms, SET institutions, and governments within regions to solve particular SET problems.

The International Arena

International competition is no longer based on factor endowments (whether dynamic or static). International competitiveness is no longer seen In the traditional terms of prices, costs, and exchange rates, nor simply in terms of differences in productivity growth. Some studies have shown that a rise in relative unit wage costs and export prices accompanied an increase in Germany and Japan's export shares, and that rising productivity levels went hand in hand with declining sectoral competitiveness. Growth in market share is being increasingly associated with technological competitiveness and the ability to compete on delivery.6 This ability is shown to depend on a conscious effort to deepen the industrial process and widen industry's linkage to the SET infrastructure. In the international arena, SET must address:

• the neoliberal policies of the World Bank and International Monetary Fund in the context of the need to develop infant industries;
• the stronger global enforcement of intellectual property rights, which introduces new rules of the game regarding reverse engineering, technology pirating, etc.—practices that helped some of the newly industrialized countries earlier in their development history;
• the decline in foreign direct investment (FDI) of multinationals in SSA as a result of the new global free trade trends and the drive against protectionism, which seems to have brought them to SSA in the past (in the globl village, production facilities do not have to be located behind protective walls, but according to dynamic technological advantage);
• the acquisition, adoption, adaptation, diffusion, and innovation from borrowed technology; and
• the interventionist role of the state in the context of global realities.

When examined from this broad scope, we can say that all countries have a science policy. Ruivo reaches this conclusion from an analysis of several developed countries. ₇ She argues that national science policies keep changing over time; and that there are marked similarities across countries, or a visible process of internationalization of science policy as a result of intergovernmental interactions in conferences, seminars, and other fora. She argues that science policy has evolved through three recognizable paradigms:

• science as a motor of progress; then
• science as a problem solver; and finally
• science as a source of strategic opportunity.

In this sense of the term, it is difficult to find countries without science policy. It seems more plausible to discuss innovation, in spite of, or not driven by science policy.

Over time, two trends are recognizable in the innovation area. Many countries have become highly interventionist in SET policy, and some have shown highly positive results from this interventionist stance. Notable among these are the newly industrializing countries (NICs) of East and Southeast Asia. Even the U.S. has become highly interventionist in spite of the rhetoric and ideology of nonintervention. The second trend is that innovation initiatives are becoming increasingly globalized or regionalized through cooperative effort among institutions, corporations, national governments, and trading blocs.

The Link Between Science and Technology and Economic Growth

Is science and technology linked in any causal way to economic growth? Many scholars do not see this link, especially in the short run. This researcher is convinced that there is a causal link both in the short and in the long run.

Science and technology generate innovations. Continuous innovation (technological, organizational, managerial) generated by learning entities have been shown to be responsible not only for productivity increases but also for dynamic competitive advantage of firms, industries and nations in the unfolding global economy. ₈ Economic growth is the result of this dynamic process. The innovation process is nonlinear, multidirectional and is central to a complex of factors which are each necessary but alone insufficient to generate growth. When innovation is missing from the development process, development becomes unsustainable in the long run. Because innovation has many sources, R&Amp;D expenditure is only a small tip of the innovation iceberg. Failure to recognize this fact may invalidate the inconclusive results obtained from cross-country regression models which find a positive but statistically insignificant relationship between R&Amp;D expenditure and economic change or growth. ₉ Easterly and Levine, using similar cross-country regression techniques in explaining the development "tragedy" of SSA, conclude that the low growth of SSA is statistically explained by: low schooling, political instability, poor financial depth, foreign exchange overvaluation, high government deficits, low infrastructure, ethnic fractionalization, and the negative contagion effect of neighbors. ₁₀ A part from low schooling, none of the eight factors said to explain SSA's poor performance has any bearing on science and technology. Some researchers have even gone a step further to argue that it is growth which creates the absorption of and innovation in science and technology, not the other way around.

These regression and similar neoclassical studies are based on equilibrium models which assume perfect competition and constant returns to scale, and which treat technology as exogenous to the development equation. Attempts to incorporate technical change in the equation and to engage the empirical fact that technical change creates situations of increasing returns have largely been made within the neoclassical tradition, which is static (or at best compares static conditions at different points in time) and does not conform to observed evolutionary process of economic change, which innovation entails. The present reality of constant technological progress is characterized by continuous disequilibria.

Innovative firms thrive on this disequilibria because, as Cooper points out, innovation is firm-specific (localized), cumulative (with firms building on their historical core competence), and appropriable. ₁₁ These characteristics make it possible for firms to thrive on temporary rents from continuous innovative activities. In other words, innovation from this standpoint does not conform to the allocative efficiency or pareto optimality of neoclassical economics.

If we view economic development as an integrated process of structural change, characterized less by the rate of growth of economic quantities than by the emergence of different types of industrial and social institutions, the corresponding advent of new technologies, changes in patterns of land use, new structures of governance, and other alterations in the pattern of the social fabric, then the most appropriate metaphor is evolutionary. It is customary to label this approach structuralist. However, we believe that the alternative neoclassical treatment of technical change or innovation is "a prisoner of its own toolbox." ₁₂ For example, commenting on the mainstream economic position, Just man and Teubal argue that the problem is partly epistemological and partly ideological. On the former they argue that "neoclassical analysis views structural change, not as a central phenomenon of economic growth but as a consequence of the process of physical capital accumulation and the consequent rise in income." Here "change is always incremental and always feasible the marginal product of capital diminishes everywhere and all future production possibilities are known from the outset." ₁₃

In this model researchers confine themselves mainly to the efficiency of resource allocation within marginal analysis—i.e., allocative efficiency in the context of static or intertemporal comparative advantage. And since growth is attributable only to the identified factors of production—land, labor, capital, and entrepreneurship— the issue has usually been one of how to allocate given investment funds between capital and labor. The problem of "technology" is thus reduced to one of choice of technique. It is a question of choice between labor or capital intensity, however defined, and the central mediating factor is the market, which when unhindered achieves allocative efficiency automatically.

On the ideological front, Justman and Teubal note that the idealization of market forces inhibits analysis of the interventionist role of the state in economic development, emphasizing instead the role of the market in promoting growth and limiting the government's role to providing general infrastructure for growth—preserving law and order, providing general education, maintaining a favorable macroeconomic environment, and correcting for "rare" market failures through neutral policy measures. ₁₄ In this model, all economic activities are equal in importance, so there cannot be preferred or strategic activities to be selectively promoted through policy. ₁₅ Finally, because it is an optimizing model, it has difficulty producing testable propositions arising from the disequilibria which technological changes introduce.

Lall has shown extensively that assumptions about perfect markets, especially with respect to less developed countries, do not conform to reality. He has shown that market failure is pervasive in product, input, capital, technology, and skills markets. ₁₆ This position draws validity from emerging evidence.

The position taken in this paper is that innovation is not only the main engine of economic growth, but also that it does not automatically accompany capital accumulation; that change is neither smooth nor automatic; that inefficiency of resource allocation is not rare but pervasive in the context of technological change; that although it is the appropriable returns accompanying innovation that encourage innovative behavior, important returns may be inappropriable (the externalities in economic parlance), therefore necessitating intervention to correct for market failures; and that there is nothing in the nature of free markets that guarantees innovative behavior by economic agents.

On both theoretical and empirical grounds, the tools of neoclassical economic analysis are inappropriate for determining the role of innovation in economic development. Theoretically, economic growth is a long-term process and innovative activities have varying periods of gestation depending on particular activities; change is not instantaneous: the innovation process is characterized by uncertainty. risk. and unpredictability of outcomes; the change process is not always linear, nor unidirectional; and the behavior of change agents (firms, farmers, individuals) is embedded in their history, organizational practices, strategies, and particular objectives (which do not have to be profit maximization), so that the same market signals can be interpreted differently by different change agents. These empirically observable characteristics fly in the face of mainstream neoclassical economic theory, and the tools of that discipline have not yet engaged these empirical realities. What we have now are simplifying assumptions which allow the application of mathematical models to dynamic interactive and complex systems. The realities remain, and our understanding of them seems to worsen rather than improve. I cannot agree more with Abramowitz who says we can be more than 100% ignorant in understanding the causes of economic growth because "what worries me [Abramowitz] is not what we don't know; it is the things we think we know that ain't so." ₁₇

When one reflects on the role of the World Bank and the IMF on this issue, the situation is even more alarming. It is not the things "we think we know that ain't so" that worries me, it is the things that people know which the Bank claims ain't so because they fly in the face of the Bank's treasured position. Nowhere is this better exemplified than in the Bank's Asian Miracle book.

To the question: "is innovation central to economic growth?" the answer on the basis of reality is an emphatic yes, although neoclassical economics, which is the "language of discourse" of much current development dialogue, ascribes growth to other factors.

Science and Technology Policy as a Development Tool

What is the justification or utility of pro-active science and technology policy as a development tool? When science and technology policy is defined broadly, as we have done above, the theoretical justification for it seems obvious. It seems easy to show the theoretical utility of having specific policies to govern science, engineering, and technology in public education, the research and development system, the SET in the production system, the consequences of technological change, and regional and international cooperation in science and technology issues. The various issues on which policy seems called for indicate the theoretical utility of the various policy initiatives. The underlying assumption of these initiatives is that market forces, on their own, are incapable of bringing about the desired changes that will aid technical change. The initiatives are designed to correct for market failures, externalities, and scale economies. However, even this theoretical validity has been questioned by neoclassical critiques on many grounds.

At one extreme it is argued that countries and states are better off if they get their macroeconomic fundamentals right: macroeconomic stability in monetary, fiscal, exchange-rate, and trade policies. When "market friendly" policies are in place, technological change or innovation will flow like manna from heaven on the basis of each country's comparative advantage. If the state must intervene, it must rely on "neutral" policies to prevent distortions in the allocative efficiency of market prices. This free market stance is aptly depicted by Lall: "firms have no learning costs, face no risk and uncertainty in absorbing new technologies, operate in isolation from other firms, can access costlessly all existing technologies and face no deficiencies in the factor markets in which they operate. They always make optimal choices as long as markets are not interfered with. (By assumption), any intervention distorts perfect markets . . . the best government is the one that provides stable macro-economic climate, essential "public goods" and stays out of the way of private enterprise." ₁₈

Needless to say, this view is far removed from reality, but it is the basis of neoliberal policy advice.

Although nobody will advocate bad macroeconomic policies, they are not sufficient to ensure technological progress. Stylized evidence from around the world indicates that substantial benefits have in fact accrued as a result of specific and targeted science and technology policy initiatives:

• Between 1915 and 1958 (the period generally characterized as the "Laissez faire" period of U.S. SET policy), the federal government forged a partnership among government, universities, and the aviation industry. It is generally accepted that the technological leadership of the U.S. aviation industry was crafted in that period, with landmark results, through planning, research, and management of the evolution of the industry.
• The invention of the semiconductor, which made several major breakthroughs possible, could not have resulted without U.S. government support for Bell Labs and the subsequent initiatives to spin off technological innovations.
• Neither shipbuilding nor steel making could have developed or survived first in Japan and then in South Korea without their aggressive SET policies. Both countries came to these industries without any comparative advantage, but created one through viable policies that now make them competitive against Scandinavian and U.S. producers.
• The emergence and growth of Korea's Chaebols—Samsung, Hyundai and Daewoo—can be traced to specific SET policy initiatives by the South Korean government, which has transformed them into major global players in the technology arena.
• If the doctrine of comparative advantage were followed, neither Denmark nor Holland could have created a refining dairy industry. Likewise, Japan, South Korea, and Taiwan could not have developed their steel and machinery industries without targeted S&T policies from their governments. ₁₉

On the basis of stylized information it seems that pro-active SET policies help to correct for the pervasive market failures that necessarily accompany technological change or innovation, thereby helping to reap the externalities of innovation and to create competitive advantage in activities that firms and countries would not engage in were they guided by market forces alone. SET policies are particularly useful for developing economies that wish to develop particular market niches in the global market, because "the main factors influencing differences in international competitiveness and growth across countries are technological competitiveness and the ability to compete on delivery." ₂₀ SET policies are required to create and sustain competitiveness, because competitiveness depends critically on technological innovation in the present global realities. Detailed case studies reveal that the ability to compete technologically depends on a conscious effort "to deepen the industrial process and widen industry's linkage with the science and technology infrastructure. ₂₁

However, certain questions raised by critics need to be addressed in this paper. First, critics argue that the impact of SET policies have not been measured quantitatively and have tended to be subjective. Second, it is argued that SET initiatives, at least in the U.S., have been more politically motivated than based on the objective evaluation of their true economic effect. Third, some critics have argued that SET policy initiatives have opportunity costs which are rarely considered in SET policy investments. Perhaps, they argue, the resources allocated to SET initiatives would have yielded greater social returns in other uses. Fourth, it is argued that the utility of SET policy initiatives depends on the context and that they should not be applied mechanically to all economies. We shall briefly examine these issues.

Measurement of SET Policy Impact

It is revealing that in spite of Solow's path-breaking paper, ₂₂ the economics profession, especially the neoclassical school, has persisted in treating technology as a residual, even though there is no logical reason for this. Among the inappropriate tools of neoclassical economics, the technology factor is not an important component of economic growth. However, in the current global reality, firms, industries, and nations create, sustain, or lose competitive advantage on the basis of their differential rate and pattern of technological innovation.

Much of the measurement that has been done has focused on R&Amp;D effort (budget, personnel), patents registered, and scientific publications as measures of input and output respectively. The discussion of the scope of SET policy in the second part of this paper shows how inadequate these measures are. In fact, only a small part of innovation results from R&Amp;D effort per se, and much of this results from tacit knowledge which is not patentable or codified.

In the absence of a more appropriate theory of technical change, the measurement of the impact of SET initiatives must be done on a case-by-case basis. Unfortunately, generalizations across countries, firms, industries, and sectors call for great caution. It is true, however, that these kinds of quantitative measurement have been few and far between. Vonortas has examined this issue in greater detail and suggests a broader approach to evaluation which combines, as necessary, case studies of returns to specific technologies, statistical appraisals at macro- and microeconomic levels, and other hybrid approaches based on primary data. In the case of the U.S., he suggests, this approach must be preceded by the establishment of a list of common criteria for the evaluation of federal R&Amp;D initiatives. ₂₃

Opportunity Cost of SET Policy Initiatives

The arguments in this cluster are many. It is argued (at least in the U.S.) that many SET policy initiatives add little or no value because the observed effects could have occurred without the initiative. The clustering of high-tech industries around Route 128 in Massachusetts and Silicon Valley were not the result of public SET policy initiatives but were due to "historical, cultural, and educational patterns, leaders and entrepreneurs, academic excellence and private initiative." ₂₄ Second, it is argued that even when there is value added, the opportunity costs may not justify the investment, since returns in alternative uses would have been higher. In situations of resource scarcity, SET policy spending may seriously compromise allocative efficiency.

This critique is constructed in the paradigm of marginal analysis which assumes many conditions that do not exist in the reality of innovation decisions. Such decisions are characterized by risk, uncertainty of outcome, and limited knowledge of technical options. Also, social returns do not always coincide with private commercial returns. Typically many returns from SET policy initiatives are delayed many years into the future. It becomes quite problematic to calculate opportunity costs on the basis of this marginalist approach.

Efficacy of Other Non-SET Initiatives

It is argued that other methods of public resource allocation could yield higher dividends than SET policy initiatives. A more general form of this critique is that macroeconomic stability in monetary, fiscal, foreign exchange, trade, and tariff policies could be more effective and less distortionary in their effects on allocative efficiency than targeted SET policy initiatives.

We have discussed this at length elsewhere and we need not repeat the arguments. In the context of SSA, pervasive market failures make this arguments untenable. More importantly, the current growth tragedy of SSA will be hard to resolve without more purposeful SET policy initiatives. SSA countries are currently uninformed spectators in the global technology game.

Contextual Characteristics of SET Policies

This criticism is valid and should warn policy makers against the indiscriminate application or adoption of SET policies developed in different contexts. However, all governments, especially in the developed economies, have some form of SET policy initiatives, although the particular initiatives vary among nations. Even in the case of the U.S., several researchers have shown that, in spite of the ideological stance of the federal government against pro-active SET policy, the reality seems different from the rhetoric. The federal government has tended to become more proactive over time, especially in the 1980s and 1990s.

The Evidence From Particular SET Policy Initiatives

The U.S. and Some Developed Countries

The objectives for particular SET policy initiatives in the U.S. and other developed countries, on the one hand, differ from those which gave rise to the pro-active SET policies of the newly industrializing countries (NICs) of East Asia, on the other hand. In the case of the U.S., one must also recognize the ideological reluctance of the federal government, until recently, to adopt a comprehensive set of government policies to promote the use of technology in improving manufacturing. Support has been peacemeal and ad hoc. In contrast, the U.S. state governments began to develop an extensive array of programs in the 1980s to encourage research collaboration between universities and industries, the commercialization of new technologies, the spin-off of technology-oriented firms, and the technological modernization of existing firms. In the 1990s, according to the Office of Technology Assessment (OTA), "as it became painfully obvious that one U.S. industry after another was losing technological leadership," the federal government became far more pro-active. We must note, however that the thrust of U.S. SET policy initiatives was to ensure the international competitiveness of U.S. industry and to increase innovation by forging a closer alliance between R&Amp;D in federal labs, universities, and independent labs. Several researchers have analyzed these federal and state initiatives. ₂₅ We shall not repeat the details here but merely highlight the essential components. The basic building blocks were:

1. Programs initiated to make the capabilities of federal labs, which were mainly defense-related, relevant to the civilian marketplace. Seven hundred and twenty-six laboratories received over 25 percent of the Federal R&Amp;D budget for this purpose. In 1992 this amounted to $22 billion.
2. Legislation to foster laboratory-industry cooperation (1986).
3. Congressionally initiated legislation to allow nonprofit institutions to take title to patents derived from government-sponsored research. It is estimated that Columbia University earned over $20 million from this in 1994 alone.
4. Authorization in 1988 for the National Institute of Standards and Technology (NIST) to launch innovation activities, including the funding of privately initiated pre-competitive technology, the founding of industrial extension programs similar to agricultural extension, and the establishment of manufacturing technology transfer centers.
5. Modification of antitrust legislation to allow companies to form industrial R&Amp;D consortia without fear of prosecution.
6. Conversion of the National Science Foundation into a mission-oriented agency lending special support to four strategic areas to promote international competitiveness. (These were manufacturing, advanced materials and processing, biotechnology, and high performance computing and communication.)
7. Additional Clinton initiatives, which included:
   • support for a national network of manufacturing extension centers;
   • legal fiscal and financial incentive provisions to improve the environment for innovation;
   • improvements to the educational system with bias for SET;
   • infrastructural investment to support the national system of innovation; and
   • heavier support for basic science, mathematics and engineering in U.S. education.

An objective evaluation of the U.S. Government initiative in SET policy must cover all of these programs. But because they are very recent and some have not even been implemented, little evidence of their impact seem available to this researcher. In particular it will be necessary to evaluate the centers in TABLE I.

The components of these various SET initiatives that have received some assessment are the venture capital for commercialization, research parks, and the high-tech business incubators. Florida et al. and Rodgers criticize the involvement of the federal government in subsidizing venture capital and conclude that government is not a good investor because it cannot close down non-performing ventures fast enough to save taxpayer dollars. Further, they argue, the government is incapable of designing portfolios that respond to particular market risks. Rather than supplementing private initiative, which has had an enviable international record, government subsidies threaten its existence. The authors all conclude that this initiative has doubtful economic impact in the context of the highly developed capital market of the U.S. ₂₆

Bania et al. examined the research parks initiatives and conclude that the successful ones (Route 128 in Boston and Silicon Valley) were not due to public policy; that at least half of research park projects were failures; that they typically did not create jobs but merely transferred them from one area to another; that they did not foster industry-university linkages where these did not exist; and that, because of their high attrition rate, they made doubtful contribution to economic growth. ₂₇

Normile looked at Japanese technopolises, which are similar to the U.S. science parks, and reached similar conclusions. There were too many of them in the case of Japan (140 in 1994), and they did not forge the required industry-university ties where these did not already exist. R&Amp;D funds tended to be spread out too thinly, with doubtful economic impact. Much of the created rentable space remains vacant. ₂₈

The incubator initiatives in the U.S. have also been evaluated and found to have very low survival rates, in some cases less than 50 percent. They showed poor rates of growth, and none of the incubator businesses grew into a large conglomerate.

The Newly Industrializing Countries of South East Asia

Evidence from particular SET policies in the newly industrializing countries (NICs) of Southeast Asia is better appreciated when placed in the context of the objectives for the initiatives. The purpose of the initiatives in these economies was the acquisition of technological capabilities. Unlike the U.S. and other advanced economies, where innovation originated predominantly from national labs, universities, and research centers apart from industry, technology in the NICs had to be acquired from external sources and adapted, assimilated, and improved. Ernst, Mytelka, and Ganiatsos define technological capabilities as knowledge and skills that firms need so that they can acquire, assimilate, use, adapt, change, and create technology. This "goes beyond engineering and technical knowhow and includes organizational knowhow as well as knowledge of behavioral patterns, for instance, of workers, suppliers and customers." ₂₉ They classify firm-level technological capability (FLTC) into six groups: production, investment, minor-change, strategic marketing, linkage, and major-change capabilities. Lall and Weiss also categories some capabilities as national-level capabilities (NLTC). ₃₀

Table 3 shows the interventionist stance of the eight countries. Hong Kong best approximates the innovation-without-science-policy stance, while Japan, South Korea, and Taiwan occupy the opposite pole in the interventionist position of the countries.

Lessons from East Asia's NIGs for SSA

An examination of these NICs revealed a number of factors common to the most successful economies:

1. They acquired the broadest range of technological capabilities. They also showed the highest growth rates and the best international competitiveness.
2. The national innovation system was properly connected with outside systems, and the SET policy, the market, and the financial system interacted readily with one another.
3. The government exercised overall leadership, not only in creating a macroeconomic and regulatory framework conducive to technological innovation, but also in taking the appropriate steps to enhance the various components of national-level technological capabilities.
4. The government took the initiative to create technological winners through targeted support for particular industries at specific times and played the leadership role in spearheading the technological development of particular industries.
5. There was a high degree of rapport among government, the system of innovation, and the corporate business sector, and the role of each was clearly defined, allowing each to play a role appropriate to the evolving circumstances of targeted industries.
6. here was no universally valid model for fostering innovation among the countries. Some depended on small firms while others depended on large conglomerates, such as the chaebols of South Korea. Some used direct foreign investment liberally, while others restricted such investment. While all relied on imported technology, the most successful countries were the ones that were selective in their choice and use of imported technology. Taiwan promoted strategic technologies by creating and promoting large state owned firms.
7. For the most successful countries, liberalization of the economy came after substantial advances had been made in the acquisition of technological capabilities for purposes other than gaining technological mastery. In particular, economies where infant industry protection was used in the most dynamic and selective manner were the most successful in gaining technological mastery.
8. The economy which was most noninterventionist and which approximates the neoclassical ideal of innovation without science policy—Hong Kong—was not only the least successful technologically, but was also the one in which a clear trend towards deindustrialization was most visible.
9. The most common factor of success was the implementation, within the overall pattern of resource allocation, of the appropriate volume and quality of investment in technical education and support for firm-level training and human development in targeted industries and sectors.
10. critical underlying factor of success in the most interventionist economies was the high managerial and technical competence of the bureaucracy—it was well trained, motivated, and well insulated from political pressures. It is argued that no SSA country has this kind of a competent bureaucracy. The implication is that the market failures which have been observed will be replaced by more serious government failures if the extent and scope of government intervention in East Asia is attempted in SSA. We would argue that SSA bureaucracies should be upgraded to identify, plan, manage, and monitor a few selective SET policy initiatives to begin to build technological capabilities in a few areas instead of leaving everything to market forces.
11. The state in each of the success cases was not an umpire or weak state but a strong, purposeful one that led the way to technological mastery, albeit through different strategies.

Source UNDP, Human Development Report, 1995

Conclusions

Certain conclusions can be drawn on the issue of particular SET policy initiatives. These are:

1. While the increasing SET intervention in the U.S. and the other developed countries was motivated by a desire to maintain technological leadership, intervention in the NICs was driven by the objective of acquiring technological capability for global competition.
2. While specific SET initiatives in the U.S. followed extensive R&Amp;D activity, R&Amp;D activity in the NICs was directed toward technological deepening and toward facilitating the acquisition of technological capability that matched their particular competitive strategies. For some it was (as Ernst et al. show) the catching-up capabilities (Thailand and Indonesia), while for others it was a keeping-up strategy (Singapore, Malaysia), and for yet others it was a getting ahead strategy (Taiwan, South Korea, and Japan). ₃₁
3. The experience of Hong Kong seems to indicate that an open-door, noninterventionist SET policy leads to low technological capability and eventually to deindustrialization. Emerging evidence in SSA seems to indicate that neoliberal policies which do not assign any specific role to the promotion of technological capabilities lead to deindustrialization. ₃₂
4. Innovation outside government policy initiatives is possible and seems to have taken place in some countries, notably the U.S. However, even the U.S. seems to have become more pro-active in SET policy initiatives lately because of a growing realization that the most competitive firms internationally come from countries with highly SET-interventionist governments. Their pro-activity seems to be a clear recognition that U.S. firms will be disadvantaged without U.S. Government SET pro-activity.
5. In the case of SSA, the emerging evidence seems to indicate that the neo-liberal policies that they are forced to implement, which assign little or no place to SET policy initiatives, are unlikely to put them on the path of sustainable economic development.
6. Ironically, also, SET policies and institutions which predated the current wave of liberalization in SSA do not seem to have brought about significant innovation because the initiatives were government driven, largely divorced from the productive system, and not based on any clear vision, mission, or choice of a future role in the global economy. In most cases they were initiatives limited to the R&Amp;D system, without a realization of the broader scope of SET policies and the need to harmonize these with other macro- and microeconomic policies.

NOTES

Note 1: D. Ernst, T. Ganiatsos, and L.K. Mytelka, eds., Technological Capabilities and Export Success: Cases form Asia (Geneva: UNCTAD, 1995). Back.

Note 2: D.C. Mowery, Science and Technologyolicy in Interdependent Economies (Boston: Kluwer, 1994). Back.

Note 3: C. Weiss, Jr., "Scientific and Technological Responses to Structural Adjustment: Human Resources and Research Issues in Hungary, Turkey and Yugoslavia," Technology and Society 15 (1993) 281-299 Back.

Note 4: In the metaphor of Reinert and Dastol (1995), economists have been "unknowingly approaching the understanding of economic growth much in the same way one would peel an onion, ever finding a new issue, once the previous issue had been brought into view." Back.

Note 5: A. Wad, "Science and Technology Policy," in The Uncertain Quest: Science, Technology and Development, ed. J.1. Salomon et al. (United Nations University Press, 1994). See summary in his table 1, p. 352. Back.

Note 6: L.K. Mytelka and D. Ernst, "Catching Up, Keeping Up and Getting Ahead: The Korean Model under Pressure" in Technological Capabilities and Export Success: Cases form Asia, ed. D. Ernst et al. (Geneva: UNCTAD, 1995). Back.

Note 7: B. Ruivo, "'Phases' or 'Paradigms' of Science Policy," Science and Public Policy 21, no. 2 (1994): 157-G4 Back.

Note 8: Perez 1994 Back.

Note 9: R.K. Goel and R. Ram, "Research and Development Expenditure and Economic Growth: A Cross Country Study," Economic Development and Cultural Change 42, no. 2 (1994). Back.

Note 10: W. Easterly and R. Levine, Africa's Growth Tragedy: A Retrospective, 1960-89, International Bank for Reconstruction and Development Working Paper, 1995. Back.

Note 11: C. Cooper, "Are Innovation Studies in Industrialized Economies Relevant to Technology Policy in Developing Countries?" UNU/INTECH Working Paper no. 3 (1991). Back.

Note 12: Reinert and Daastol 1995 Back.

Note 13: M. Justman and M. Teubal, "A Structuralist Perspective on the Role of Technology in Economic Growth and Development," World Development 19, no. 9 (1991): 116783. Back.

Note 14: See also Cooper 1993. Back.

Note 15: Wangwe 1993. Back.

Note 16: S. Lall, "New Forms of State Intervention in Industry: Lessons from Experience," a paper presented at the University of Limburg, Maastricht, Nov. 1995. Back.

Note 17: M. Abramowitz, "Sources of Ignorance: Old and New" Journal of Economic History, March 1993). Back.

Note 18: Lall, "State Intervention." Back.

Note 19: R. Sinha, "Economic Reform in Developing Countries: Some Conceptual Issues," World Development23, no. 4 (1995): 563. Back.

Note 20: Mytelka and Ernst, "Catching Up." Back.

Note 21: Nelson and Rosenberg 1993. Back.

Note 22: Solow 1957. Back.

Note 23: N.S. Vonortas, "New Directions for U.S. Science and Technology Policy: The View from the R&Amp;D Assessment Front," Science and Public Policy 22, no. I (1995). Back.

Note 24: S. Raymond, "Listening to the Critics: Enlarging the Discussion of Policy for Science Based Development," draft paper prepared for conference on Policy for Science Based Development at the New York Academy of Sciences, April 1995, p. 8. Back.

Note 25: For example see 1. Feller, "American State Governments as Models for National Science Policy," Journal of Policy Analysis and Management 11, no. 2 (1992): 288-309; H. Etzkowitz, "Technology Centres and Industrial Policy: The Emergence of the Interventionist State in the USA," Science and Public Policy, 21, no. 2 (1994): 7988; M.M. Crow, "Science and Technology Policy in the U.S.: Trading in the 1950 Model, Science and Public Policy (1994); Skolnikoff 1995; and J.S. Tatum, "Science, Technology and Government: Re-examining the Relationship," Technology in Society 17, no. I (1995): 85-102. Back.

Note 26: R. Florida and D.G. Smith, JL, "Keep the Government out of Venture Capital," Issues in Science and Technology 9, no. 4 (1993); T.J. Rodgers, "Subsidies with Deathly Strings" in Forbes 152. no.2 (1993). Back.

Note 27:

27. N. Bania, R.W. Eberks, and M.S. Fogarty "Universities and the Startup of New Companies: Can We Generalize from Route 128 and Silicon Valley?" The Review of Economics and Statistics 25, no.4 (1993). Back.

Note 28: D. Normile, "Bright Science City Dreams Face Sober Realities," Science 266 (1994): 1176-77. Back.

Note 29: Ernst, Ganiatsos, and Mytelka, Technological Capabilities. Back.

Note 30: S. Lall, "Technological Capabilities and Industrialization," World Development 20, no. 2 (1993); C. Weiss, Jr., "Scientific and Technological Responses to Structural Adjustment: Human Resources and Research Issues in Hungary, Turkey and Yugoslavia," Technology and Society 15 (1993): 281-99. Back.

Note 31: Ernst, Ganiatsos, and Mytelka, Technological Capabilities. Back.

Note 32: See H. Stein, "Deindustrialization, Adjustment, the World Bank and the IMF in Africa," World Development 20, no. 1 (1992): 1833-49. Back.

References

Abramowitz, M. "Sources of Ignorance: Old and New" Journal of Economic History, (March 1993)

Aganbegyan, A.G. "A System Approach to Technical Development."International Journal of Technology Management 10, 1 ( 1995).

Bania, N., R.W. Eberks, and M.S. Fogarty. "Universities and the Startup of New Companies: Can We Generalize from Route 128 and Silicon Valley?" The Review of Economics and Statistics 25, 4 (1993).

Busseret, T.J., H.M. Cameron and L. Georghion. "What Difference Does it Make? Additionality in the Public Support for R&Amp;D in Large Firms." International Journal of Technology Management 10, 41516, (1995): 587-600.

Cooper, C. "Are Innovation Studies in Industrialized Economies Relevant to Technology Policy in Developing Countries" UNU/INTECH Working Paper 3 (1991).

Crow, M.M. "Science and Technology Policy in the U.S.: Trading in the 1950 Model." Science and Public Policy (1994).

Easterly, W., and R. Levine. "Africas Growth Tragedy: A Retrospective, 1960-89." International Bank for Reconstruction and Development Working Paper, 1993.

Ernst, D., T. Ganiatsos, and L.K. Mytelka, eds. Technological Capabilities and Export Success: Cases form Asia. Geneva: UNCTAD, 1995

Etzkowitz, H. "Technology Centres and Industrial Policy: The Emergence of the Interventionist State in the U.S.A." Science and Public Policy, 21, 2 (1994) 79-88.

Feller, I. "American State Governments as Models for National Science Policy." Journal of Policy Analysis and Management 11, 2 (1992): 288-309.

Florida, R. and D.G. Smith Jr. (1993). "Keep the Government out of Venture Capital" in Issues in Science and Technology 9, 4 (1993).

Forje, J.S. "Development and Technology Policy in Changing Society: Challenges for the African Political Structures." 1995.

Grabowski, R. "The Successful Developmental State: Where Does it Come From?" World Development 22, 3, (1994): 413 422.

Goel, R.K. and R. Ram. "Research and Development Expenditure and Economic Growth: Across Country Study." Economic Development and Cultural Change 42, 2. (1994).

Justman, M. and M. Teubal. "A Structuralist Perspective on the Role of Technology in Economic Growth and Development." World Development 19, 9 (1991) 1167-1183.

Kwon, J. "The East Asia Challenge to Neoclassical Orthodoxy." World Development 22, 4 (1994): 635-644.

Lall, S. "Technological capabilities and Industrialization." World Development 20, 2.(1993).

Lall, S. "The East Asian Miracle: Does the Bell Toll for Industrial Strategy?" World Development.22, 4 (1994): 645-654.

Lall, S. "New Forms of State Intervention in Industry: Lessons from Experience." Paper presented at the University of Limburg, Maastricht Nov. 1995.

Meyer-Krahmer, F. "Technology Policy Evaluation in Germany" International Journal of Technology Management 10, 41516 (1995).

Meyer-Stamer, J. "Micro-level Innovations and Competitiveness." World Development 23, 1, (1995): 143-148.

Mowery, D. Science and Technology Policy in Interdependent Economies. Boston: Kluwer, 1994.

Normile, D. "Bright Science City Dreams Face Sober Realities." Science 266 (1994): 1176-7

Mytelka, L.K. and D. Ernst. "Catching Up, Keeping Up and Getting Ahead: The Korean Model under Pressure," in Technological Capabilities and Export Success: Cases from Asiz, ed. D. Ernst et al. Geneva: UNCIAD, 1995.

Odhiambo, T. "Redefining Africa's Priorities for the 1990's: Industrial Development Ventures through Technology Parks." Keynote address to the Third Presidential Forum, Kampala, July, 1995.

Raymond, S. "Listening to the Critics: Enlarging the Discussion of Policy for Science-Based Development." Draft paper prepared for a conference on Policy for Science-Based Development. New York Academy of Sciences' Working Paper, April 1996.

Rodgers, T.J. "Subsidies with Deathly Strings," in Forbes 152, 2 (1993).

Ruivo B. "'Phases"' or "'paradigms"' of Science Policy." Science and Public Policy 21, 2, (1994): 157-164.

Russell, A. "Merging Technological Paradigms and the Knowledge Strucutre in International Political Economy." Science and Public Policy 22, 2 (1995): 106-116.

Shapiro, H. and L. TayloL "The State and Industrial Strategy." ?World Development 18, 6, (1990): 861-878.

Sinha, R. "Economic Reform in Developing Countries: Some Conceptual Issues." World Development23, 4 (1995): 557-575.

Stein, H. "Deindustrialization, Adjustment, the World Bank and the IMF in Africa." World Development 20, 1 (1992): 1833-1849.

Streeten, P. "Markets and States: Against Minimalism." World Development 21, 8, 1281-1298.

Tatum, J.S. "Science, Technology and Government: Re-examining the Relationship." Technology. in Society 17, 1 (1995): 85-102.

Thisen, J.K. "The Development and Utilization of Science and Technology in Productive Sectors: Case of Developing Africa." African Development 28, 4.(1993).

Vonortas, N.S. (1995). "New directions for U.S. science and technology policy: The view from the R&Amp;D assessment front" in Science and Public Policy Vol. 22, No. I.

Wad, A. "Science and Technology Policy." Salomon J-J et al. edition. The Uncertain Quest: Science, Technology and Development, United Nations University Press, 1994.

Wei, L. "International Technology Transfer and Development of Technological Capabilities: ATheoretical Framework." Technology in Society 17, 1 (1995): 103-120.

Weiss, C. JL "Scientific and Technological Responses to Structural Adjustment: Human Resources and Research Issues in Hungary, Turkey and Yugoslavia." Technology in Society 15, (1993): 281-299.