EDUCATION, TECHNOLOGICAL CHANGE,
At one level, the importance of education to economic growth seems obvious. But many questions remain: how much education is needed, who should get it, who should deliver it, how should its delivery be organized, who should pay, who will benefit, and how is education related to other types of economic and social policy? In the past it has been difficult to answer these questions, because, surprisingly, analysts have had trouble explaining why and how education is related to growth and competitiveness. But the last decade has seen new progress in theoretical and empirical work on education and growth. This paper uses current economic thinking about the relationship between education, development, and growth as well as recent developments in educational reform in the U.S. to discuss educational policy in developing countries. One of the fundamental conclusions of that research is that the relationship between education and growth cannot be understood in the abstract. Education is not something that can be tacked onto the society and economy regardless of the surrounding conditions. Different conditions require different educational strategies.
The importance of education to technological change and productivity is at once obvious and opaque. We would be hard pressed to find someone in the industrialized or developing worlds who would deny the importance of education or human capital accumulation to the economic health of any country. But beneath this apparent unanimity, profound questions and disagreements lurk. How much education is needed, who should get it, who should deliver it, how should its delivery be organized, who should pay, who will benefit, and how is education related to other types of economic and social policy?
In the past it has been difficult to answer these questions because, surprisingly, analysts have had trouble explaining why and how education is related to growth and competitiveness. Does education promote productivity and growth because it prepares a technical and intellectual elite or does it strengthen the economy by reducing fertility and improving health? Some analysts in the U.S. argued that education promoted growth in a capitalist economy by creating a docile and obedient workforce.
But the last decade has seen new progress in theoretical and empirical work on education and growth. One of the fundamental conclusions of that research is that the relationship between education and growth cannot be understood in the abstract. Education is not something that can be tacked onto the society and economy regardless of the surrounding conditions. Different conditions require different educational strategies. This insight, as simple as it may seem, has had profound implications for educational strategies in both industrialized and developing countries. In developing countries, it has exposed the folly of educational systems copied from the U.S. or Europe. In developed countries it has suggested that an education system that was appropriate for the post-World War II economy, may no longer be adequate for current economic, political, and technological conditions. In the U.S., the problem is not that schools have deteriorated, as many politicians have argued, but rather that the economy has changed, leaving the schools behind.
And even more complex, new thinking about education and growth in economics suggests that technological and economic conditions may themselves influence the acquisitions of skills and the accumulation of human capital. (Human capital is endogenous in some of these models.) Thus policy makers cannot simply think of education as a tool which they can use to promote growth. Rather trade and commercial policy can influence human capital accumulation which can in turn influence growth and productivity. This line of thinking suggests that free trade, which most economists in the developed world believe would benefit developing countries, may, in some cases, slow the accumulation of knowledge and human capital.
The purpose of this paper is to use current economic thinking about the relationship between education, development and growth to discuss educational policy in developing countries. We first discuss recent developments in the economic theory that link growth and competitiveness to education. We then present the insights derived from the theory for educational and economic policy in developing countries. The subsequent section of the paper discusses relevant economic and educational changes now taking place in the U.S. We end with a summary and conclusions.
Economic Theory, Growth and Education
The promotion of education and training has been a staple of economic development policy for decades. And the importance of education to growth and education in the developed world has also long been recognized. Historians argue that early public education in the U.S. helped launch an era of U.S. economic hegemony. In 1983, the profoundly influential publication A Nation at Risk, (U.S. Commission on Excellence in Education 1983) blamed rising trade deficits and stagnant standards of living on a deteriorating U.S. education system. Seven years later, the Commission on Skills in the American Workforce (1990) argued that the country had a "choice" between two future paths. One involved high levels of education, high skills, and rising standards of living, while the other was based on low wages and low skills, and would lead to increasing inequality and a deterioration of the average standard of living. Business people and educators argued that the German and Japanese success in export markets was due to their apparently superior education systems.
Empirical evidence developed over many years has confirmed the importance of both education and technology to economic growth (see Mowery and Rosenberg 1989; Denison 1985; Benhabib and Spiegel 1992; and Mankiw, Romer and Weil 1992). And hundreds of studies have shown that individual earnings rise with education. On a cross country basis, a one-year increase in schooling augments wages by between 5 to 25 percent, after allowing for other factors (World Bank 1991, table 3.2).
But why and how does education promote growth and development? Most analysts agree increases in health, nutrition and higher labor force participation rates are important reasons why education fosters economic development. In 1890 Alfred Marshall wrote that "health, strength, physical, mental and moral ... are the basis of industrial wealth." But these notions have not been explicitly incorporated into economic theories of growth. For many years, developing countries emphasized the need to train a technological and scientific elite. More recently, many analysts have argued that skill and educational deficiencies among lower level workers directly involved with production and agriculture have been a fundamental block to economic development. Which educational policy is most efficient will clearly depend on the underlying reasons why education promotes growth and on the mechanisms and processes through which education is translated into development and increased productivity.
For most of the period since World War II, economic thinking about growth has been based on what has been referred to as the neoclassical benchmark model, which was developed by Solow (1956, 1957). This model assumed that all capital and labor were homogeneous, thus eliminating any consideration of differences in the quality of labor (and capital) that might arise from education, technology, or other factors. In this model, income levels were influenced only by population growth (which was considered exogenous) and the accumulation of physical capital (which resulted from savings). In the absence of constant, exogenous technological change, the model implied that the per capita growth rate of national income must approach zero in the long run. The policy implications of this model flowed directly from its basic assumptions--in order to raise per capita income, keep population growth to a minimum and raise the savings rate, which would raise the per capita rate of capital accumulation. Education was nowhere to be seen.
Thus the model was useless for analyzing the effects, to say nothing of the causes, of education. The assumption of homogenous labor excluded any consideration of differential effects of education on labor, and while many analysts believed that education might promote technological innovation, this model assumed that such innovation was exogenous (its causes were not under consideration in the model).
Moreover, the earliest empirical tests of the Solow model, so-called growth accounting exercises (Solow 1957; Abramowitz 1956; Denison 1961), suggested that most of the growth in output could not be explained by population growth and the accumulation of capital. Abramowitz labeled the unexplained "residual" responsible for most of the growth in output, "The Economists Index of Ignorance" (later it received the more neutral term the "Solow Residual"). Discouraged by the simplistic policy implications and the weak empirical support, analysts turned their attention to exploring the components of the Solow Residual to explain economic growth.1 Early attempts by Solow (1960), Kaldor and Mirrlees (1962) and many others sought to make the model more sophisticated by dropping the assumption of homogeneous capital. They recognized that at any moment, capital included both new and old vintages of equipment and that newer equipment embodied more advanced technology. These so-called vintage capital models enjoyed a period of popularity but the ultimately more successful theoretical extensions of the Solow model were based on dropping the assumption of homogeneous labor.
These investigations, which examined the determinants and importance of human capital investment, were initiated by Schultz (1960) and led to a rich literature on human capital and on-the-job training (Becker 1964; Mincer 1974; and Schultz 1961). By accounting for differences in human capital, economists were then able to account for a much larger share of economic growth.
But the growth accounting exercises were in effect empirical findings without an underlying theoretical explanation. More education was associated with more growth, but why? And there was no consideration of the causes of human capital accumulation (no model took such accumulation as endogenous). Surprisingly, given the consensus about the importance of education, education and endogenous human capital accumulation were not included into formal models until the 1980s. Findlay and Kierzkowski (1983) presented the first model which specifically included endogenous skill accumulation, and documented the importance of the stock of human capital in determining competitiveness, comparative advantage, and the pattern of trade. Subsequently Romer (1986) developed a growth model that explicitly included human capital although in this case human capital was exogenous.2 Nonetheless, it remained unclear how, and which kind of human capital contributed to economic growth.
The following years saw mounting empirical and theoretical evidence of the importance of technological change and human capital in competitiveness and growth. By the mid to late 1980s studies showed that the level of education (Mankiw, Romer and Weil 1992), the size of the educated work force (Romer 1986, 1989, 1990), the number of patents issued (Grossman and Helpman 1991; Judd 1985), and the size of privately and publicly funded research expenditures in the private and public sector influence not only a country's growth of income, but also its pattern and volume of trade. The approaches of the theoretical literature to explain exactly how human capital contributes to economic growth may be grouped into three rough categories, which are described below.
I) Education as a separate factor of production: One approach, developed by Romer (1986), Lucas (1988) and others, suggested that human capital, just like physical capital, can be viewed as a production input which can be accumulated. No explicit relation between human and physical capital and technological change was specified, however. In fact, in this analysis, human capital represented the effective, or average, technological knowledge of an economy, which could be accumulated in a separate education sector, without implied relation to the current standard of technology. The policy implications were that the competitiveness and growth rate of a country are closely tied to the share of its people receiving education and, most importantly the level of educational attainment.
The primary contribution of this line of research was that for the first time, allusions to the important external effects of private human capital accumulation were included into formal models. Society as a whole benefits more than the individual from that individual's education, thus left to their own choices, individuals would invest in less education than was socially optimal. This created a justification for public policy to "internalize" the externality, by subsidizing human capital accumulation.
Internalizing the externality was really just an elegant statement of the conventional public goods justifications for public funding of education in capitalist economies (Friedman 1962). Moreover, since the human capital in these models was included in a highly aggregated form, they were unable to generate insights concerning relative investments in primary, secondary, or tertiary education; how this education should relate to the rate of technological change; or the appropriate government role in subsidizing on-the-job training.
II) Learning by doing: Another avenue explored by the theoretical literature based its analysis of human capital on learning by doing. Once again, labor was assumed to be homogeneous, but serendipitous productivity increases were generated as higher volume of output caused production workers to move down the learning curve. Young (1991, 1993); Lucas (1988); Boldrin and Scheinkman (1988); and Stokey (1988), all showed that learning by doing exhibits crucial scale and spillover effects. The benefits of learning by doing were seen to be twofold. The first benefit was the traditional notion (Arrow 1962) that the more volume of a particular good produced, the further labor moved down the learning curve, and the greater the improvement in efficiency and productivity. Second, the more volume of a particular product produced, the more skill useful for the related technology was obtained, making it easier to learn about new, relatively similar, production processes. Increased output therefore led to lower unit costs (although unit costs fell at a decreasing rate), and to important knowledge spillovers, which facilitated the adoption of new technology. Competitiveness and the pattern of trade are then determined by the size of the market (the volume of output in a specific sector) and by the knowledge content of the sector in which learning occurs, leading to important implications for public and commercial policy which will be discussed later.
While learning by doing emphasizes the educational benefits of particular types of production, research also suggests that prior education also influences the effectiveness of learning by doing. Analyses of the relationship between education, training, and earnings show that schooling and learning on the job are complementary. Thus learning by doing will be more effective if it is built on at least a minimal foundation of schooling.
III) The mutual interaction of technology, human capital, and economic conditions: The third class of models, rather than viewing education as a simple input into the production process, or emphasizing serendipitous and costless learning by doing, is based on the idea that the invention and adoption of new technology, the accumulation of human capital, and economic conditions are all interdependent--they are endogenous to the model (Nelson and Phelps 1966; Romer 1990; Grossman and Helpman 1991; Eicher 1993).
One hypothesis explaining the empirically observed interaction between technological change and human capital (see, for example, Bartel and Lichtenberg 1987; Davis and Haltiwanger 1991; Mincer 1991) was first proposed by Nelson and Phelps (1966). Specifically, skilled workers are assumed to possess a "comparative advantage" with respect to inventing and adapting new technologies. Nelson and Phelps (1966) suggested that the introduction of a new technology radically transforms the production environment. Skilled workers differ from unskilled in their ability to function in this new environment, since skills enhance the ability to handle new demands created by the new technologies. Nelson and Phelps proceed to rank jobs according to the degree to which they require adaptation to change from unskilled (highly routinized) to highly skilled (involving the necessity "to learn to follow and to understand new technological developments" p. 69).
One implication of this reasoning is that if the technology in a job changes, the quality of skills required must also change. This implies that the first class of models discussed above, which allows for human capital accumulation independent of technological change, is incomplete.
The interaction between human capital accumulation and technological change also relates to the distinction between the determinants of the adoption of existing technologies versus the creation of new technologies. Learning-by-doing models focus on the cost of adopting a new technique and production process, while the third group of models recognize explicitly that skilled workers also invent the new technologies, which must subsequently be absorbed into production. This distinction between adoption and innovation turns out to be crucial in the discussion of policy implications of the various models.
The proposition that education promotes both adoption and creation of new technology has strong empirical support. Benhabib and Spiegel (1992) show that human capital explains economic growth better when modeled to facilitate the adoption of new technologies, as opposed to being just another input into the production function. Other empirical work by Bartel and Lichtenberg (1987), Mincer (1989, 1991), Davis and Haltiwanger (1991), Juhn, Murphy and Pierce (1993), Berman, Bound and Griliches (1993), and Bound and Johnson (1992) has shown a large degree of complementarily and reciprocity between technological change and human capital. These studies find that a higher rate of technological innovation and adoption increases the demand of skilled relative to unskilled labor.
The new growth models point out another reason why education should be considered endogenous. Schooling itself is also influenced by the current level of technology and quality of skilled labor in teaching.
These interactions between human capital and technological change can be summarized by the following four critical allocation decisions: 1) what share of the population should obtain which skills and how much existing human capital of what quality should be allocated 2) to education, 3) to the invention of new technologies, and 4) to the absorption of innovations.
Thus we have come a long way from the simple notion that more education is better. This third class of models implies that decisions about how much and what type of human capital to accumulate and what resources should be devoted to invention and to absorption cannot be considered independently.
We have reviewed three classes of recent models of the relationship between education and growth. The first treats education as a distinct factor of production, the second is based on learning by doing, and the third focuses on the mutual interactions between human capital development, invention and adoption of technological change, and economic conditions. We argue that the second and third groups offer specific and important insights for education in both developing and industrialized countries.
Learning-by-doing models are primarily relevant to the adoption of existing technologies, thus they seem particularly important for developing countries, which in general can make great strides by adopting existing equipment and adjusting it to relevant conditions. But we have also argued that schooling and learning on the job are complementary. While learning by doing seems to be a costless by-product of production, its effectiveness in generating competitiveness is influenced by a base level of schooling. Without that base, learning will probably take place, but at a slower rate.
By providing that base education, a country accelerates the dynamic benefits derived from productivity increases resulting from learning by doing. This factor, combined with the increases in health and labor force participation associated with increases in basic education, suggest that, in countries with high levels of illiteracy, there are potentially high social and economic returns to increased investments in primary education. Empirical evidence supports this conclusion. Indeed, numerous studies have shown that the spread of primary education translates into higher agricultural and family enterprise productivity through better absorption of new information and faster adoption of advanced techniques (Welch 1970; Krueger 1991). In Peru, for example, it has been estimated that the return to an additional year of primary education for self-employed women in the textile sector is 33 percent (World Bank 1990).
Thus, on the most basic level, the empirical evidence suggests that during the early stages of the development process, primary education should receive the most resources to develop a critical level of basic skills. However, a cursory examination of government policies suggests that the empirical evidence concerning returns to primary education are too often neglected. For example, while Brazil spends 69 percent of its public education budget on primary education, only 9 percent is spent on secondary but 23 percent on tertiary education. In addition, only 23 percent of all elementary schools received text books in the first grade in the early 1980s. In Chile, Costa Rica, the Dominican Republic, and Uruguay, the top fifth of the income distribution receives more than 50 percent of the subsidies for tertiary education, the poorest fifth receives only 10 percent (World Bank 1990).
The learning-by-doing model suggests that there are important learning benefits to large volume production in strategic areas. This creates a link between general economic and trade policy and human resource development. By subsidizing output, countries can try to promote larger volume production in order to achieve resulting benefits of learning by doing.
But which sector should be subsidized? Given that technological innovations take place all the time, Young (1991) has shown that subsidizing the sector with high knowledge content to be more effective in creating comparative advantage. Such a policy would move production into more skill intensive goods, and prepare the labor force more effectively for the advent and introduction of a new technology. The country with the largest market and most aggressive policy for moving production into high tech sectors (which tend to offer more opportunities for learning) would thus possess a comparative advantage in high tech goods. Active commercial policy could steer the economy in the same direction, and could either create even larger markets and comparative advantage, or accelerate the speed with which a country catches up to more advanced nations.
One policy implication of this is that trade should be managed so as to increase the market size. Even if an imported good is cheaper today than a domestically produced good, relying on the import will reduce the domestic producers' market share and thwart its ability to benefit from economies of scale. Temporary protection may allow the comparatively disadvantaged industry to exploit economies of scale, lowering its per unit cost to such an extent that it may become an exporter of the good in the future. This argument is especially relevant to developing economies, as free trade in the presence of economies of scale may simply consolidate the position of the technological leader. This is an argument for so-called Strategic Trade Policy, which influences the terms of trade of a targeted industry in the "right" direction in order to establish a comparative advantage.3 This approach implies "picking winners" in international competition. Only industries with high growth potential and high learning potential ought to be targeted.
Basic education and learning from experience are, by themselves, not adequate to support extensive innovation of technology to achieve technological leadership. Innovation must involve tertiary education, not only because post-secondary institutions supply the technical and scientific personnel who can carry out innovations, but because significant amounts of innovation actually take place in university labs. This was the thinking behind the earlier popularity of manpower planning in developing countries, which emphasized the development of a cadre of high level technicians, engineers, and scientists.
The third category of models discussed above, has clearly pointed out the problems with this policy prescription. The education of a high-level scientist population must be in relation to the capabilities of a country to absorb the technological innovations, on the one hand, and the stock of scientists on the other. "Overinvestment" in tertiary education can be detrimental to a country's economic growth, if it stands in no relation to the country's technological capabilities, and thus leads to underemployment of high-skilled labor. An emphasis on tertiary education without an adequately educated mass population, for example, may lead to a glut of frustrated university students (see, for example, the Philippines).
Indeed, the manpower planning approach to train high-level personnel has not been effective, and most analysts now agree that the returns to expensive (usually publicly funded) tertiary education in developing countries are not large enough to warrant excessive effort (Psacharopoulos and Woodhall 1985). Even if developing countries can innovate technologically, an adequate base of education among the general population is still required to adopt those innovations. Without access to huge markets that would allow long learning curves, an uneducated workforce cannot make effective use of innovations, homegrown or not.
Furthermore, the Japanese experience especially has shown that much innovation takes place gradually as workers try to solve small problems. This process, which might be called "innovating by doing," can effectively use production workers if they have the skills and understanding to make a contribution. This source of innovation is not available with an uneducated workforce.
Any accumulation of cutting edge technological know how must be carried out in relation to the production and technological possibilities of a country (and vice versa). Studies have confirmed that an exceedingly aggressive policy of adopting and advancing technology without the appropriate level of human capital can slow growth (see Young's (1992) comparison of Singapore and Hong Kong). Young's empirical study has shown that an economy's attempts to leapfrog technologies, (i.e., adopt new technologies without having generated the prerequisite human capital) is not a strategy which utilizes resources efficiently.
U.S. Education Reform and Insights for Developing Countries
Recent economic theories that relate education to economic growth therefore have a variety of implications for economic, trade, and educational policy in developing countries. But recent developments in the industrialized world also provide important insights for educational planners.
Ironically, while theory has exposed the learning and educational benefits of large volume production for developing countries, many industrialized countries, especially the U.S., are trying to adjust their production strategies so that they can produce much smaller quantities of particular products efficiently and rapidly. This argument, developed by Piore and Sabel (1984) and many others, contrasts mass production to flexible production.
Mass production depends on large volume production of identical or similar items to recoup the fixed cost of automation and engineering and to maximize the time during which the workforce is operating near the optimal point on the learning curve. Mass production generally involves detailed planning and engineering by a cadre of highly educated technical and professional personnel. But these planners try to simplify or "dumb down" the jobs of production workers. The assembly line with highly fragmented and repetitive jobs is the paradigmatic example of mass production. Mass production does involve sophisticated equipment and production workers do need a minimal level of education, but these skill demands are not high and the large volume allows maximum operation of the learning curve.
Mass production was particularly successful in the U.S. with its gigantic internal market. Moreover, the American education system was well suited for the skill demands of the mass production approach. The U.S. has, by international standards, very high quality university and especially post-graduate education to prepare the scientific, technological, and professional leadership. But the quality of education received by the three quarters of the population that does not graduate from college or university is of much lower quality than the equivalent levels of education in many European and Asian countries, implying that U.S. workers in comparison to these foreign counterparts learned a larger share of the required skills on the job. Thus the logic of learning-by-doing models applied to the educational needs of production workers while the models that emphasized the interdependence of education and technological innovation were most relevant to the training of higher level personnel.
But economic and technological changes have undermined the basis of the mass production system. Much more intensive international competition and faster changes in products and technologies have greatly reduced opportunities for seemingly endless production of standardized goods using unchanging processes. The learning curve has much less time to operate. Certainly the need for high level technical personnel remains. Indeed it has intensified. But the low quality education of production workers has now become a greater liability for the U.S. system. Other industrialized countries that never had the luxury of the U.S. mass market were never as dependent on learning by doing based on high volume production. This may explain why their non-university education systems are stronger than the U.S. system.
Thus in the U.S., there is a growing realization that much of the workforce is not adequately educated. In past years, high school graduates could get reasonably well-paid unionized jobs that could support a moderate middle-class lifestyle. But in the last 15 years, the real earnings of high school graduates have fallen sharply (Levy and Murnane 1992; Katz and Murphy 1992; Bound and Johnson 1992).
One response to the educational crisis for those students who do not go on to four year colleges has been the reform of the content and curriculum of secondary school. This involves a shift from a didactic pedagogy based on the transmission of information from teacher to student, to what is referred to as a student-centered approach which emphasizes inquiry and discovery on the part of the student.4 Rather than having teachers lecture to students, a student-centered strategy is based on group projects with open-ended or ambiguous outcomes which facilitate the understanding and application of underlying concepts. There is also a strong current in this reform movement that emphasizes the value of guided internships or apprenticeships in which students have an opportunity to apply school-learned concepts in realistic settings.
A fundamental notion that underlies these school reform movements is that the U.S. is moving into a more dynamic and competitive economy in which learning by doing based on primary or low-quality secondary education is no longer adequate. This puts a greater burden on the education system to produce graduates who can operate in more ambiguous, faster-changing, and less structured environments. Thus we are expecting many more workers to be effective in the types of activities previously carried out by college graduates. These added educational objectives can either be accomplished by sending more students to university or by reforming secondary schools (certainly the most effective strategy will involve a combination of these two approaches). The third class of models discussed above previously focused our attention on tertiary education; the current reforms in the U.S. suggest that some of the objectives that we look for in tertiary education may also be achieved in secondary schools.
What does this imply for developing countries? One possibility is that these countries focus their efforts on capturing those markets that allow high volume production. Certainly such markets still exist and possess significant growth potential in developing countries. The high volume production will allow workers to learn, but if countries focus on traditional labor intensive industries such as apparel and shoes, there will be few benefits to knowledge spillovers. Thus developing countries must target high volume industries with good potential for learning. The policy implications of this approach involve a refocus on primary education, avoidance of overinvestment in tertiary education, and appropriate trade and commercial policy to promote large volume production in specific sectors in order to exploit the learning curve.
But secondary school reform may offer a middle path that would allow developing countries to lay the groundwork to compete in more sophisticated markets without necessarily increasing their emphasis on costly tertiary education. This would still require a solid base of primary education, but it would put less reliance for human capital development on the learning curve generated by high volume production.
Summary and Conclusion
The competitiveness of a firm, an industry, and a nation is related to the mix of primary, secondary, and tertiary education and how that interacts with the level of development and the state of technology. The theories that we have reviewed suggest that a government may play a crucial role in enhancing and allocating the stock of human capital in the economy. Primary education enhances nutrition and health, increases the rate of return in the traditional sectors, and facilitates learning. It is also the foundation for moving to a stage of development where, with expanded secondary education, more and more techniques can be adapted from a technological leader. Comparative advantage, technological leadership and economic growth are most aggressively influenced by a nation's or firm's ability to absorb and advance new technologies, which requires the mastery of previous technologies, a highly skilled workforce, and a large enough stock of scientists and engineers. Here traditionally, the interaction between technological change and accumulation of human capital in the tertiary sector are central. We have argued that trends in school reform in the U.S. suggest that secondary schools could also play a large role in this last objective.
Public policy must have a national and an international focus. First, the government must establish reasonable priorities for the distribution of funds based on the rates of return, and stages of development. On the most basic level, it makes little sense to attempt to be a high-tech exporter and to subsidize heavily tertiary education when primary education is still insufficient. The empirical and theoretical literature suggests that the correct mix of educational and industrial subsidies, at each state of development, is crucial.
The above analysis suggests that strong emphasis ought to be placed on the accumulation of human capital, but not without relation to the existing industrial and technological state of development. Important and cheap advances in human capital can be achieved through learning by doing and targeting both high growth and high-tech industries, if the movement along an industry's learning curve can be accelerated, especially if it occurs in high knowledge content sectors. The theoretical literature suggests that these benefits may be achieved through commercial policy, adjustments in the terms of trade, and subsidies for targeted industries. We have also suggested that appropriate education and job training are needed as a foundation for more effective learning by doing.
As a first step, developing countries should try to capture mass markets in industrialized as well as developed countries. The products that can serve these markets have less demanding skill requirements, they can take advantage of the lower wage levels in developing countries, and the process of production itself generates human capital development through the learning curve. But taking this approach in effect simply follows that path taken by industrialized countries--mastery of standardized products and eventual concentration on more sophisticated and varied goods and services. And the benefits of this approach will be minimal if the products that are produced have little positive spillover effects. Moreover, unlike the industrialized countries, today's developing countries have the experience of those industrialized countries from which they may benefit. While it is unrealistic for developing countries to compete directly in the most advanced markets, improved education does offer an opportunity to build the foundations of economies that are less dependent on basic standardized items. But the key to this is not to try to copy the tertiary systems of the industrialized world, although a solid tertiary system is certainly necessary. Better opportunities lie in a variety of alternative policies including a strong emphasis on primary education, an increase in the quantity and quality of secondary education, and new departures in trade and commercial policies.
* Thomas Bailey is the Director of the Institute on Education and the Economy and an Associate Professor in the Department of Economics, Education, Philosophy and Social Sciences at Teachers College, Columbia University. He holds a Ph.D. in labor economics from MIT. He has served as a consultant to many public agencies and foundations including the U.S. Department of Labor, the U.S. Department of Education, the U.S. Congress Office of Technology Assessment, the Alfred P. Sloan Foundation, the William T. Grant Foundation, and several state and local economic development and educational agencies. He has authored or co-authored books on the employment and training of immigrants and the extent and effects of on-the-job training including his most recent book The Double Helix of Education and the Economy which was written with Sue Berryman. Theo Eicher is an Assistant Professor in the Department of Economics at the University of Washington, Seattle. He holds a Ph.D. in Economics from Columbia University. His fields of interest include technology, human capital and trade growth and development.
1. It should be noted that the subsequent developments in trade theory, and the subsequent policy implication, which emphasized the importance of free trade and condemned distortions of prices caused by trade restrictions, were based on the neoclassical benchmark model. With the advent of the new growth theory in the 1980s, which included in the analysis factors such as economics of scale, human capital, learning-by-doing spillovers, and endogenous technological change, economists started to consider the role that strategic trade policy might play in promoting economic development.
2. See Berryman and Bailey (1992) for a full discussion of this approach and the broad policy that is aimed at achieving it.
3. Often this argument is confused with the classic infant industry protection argument, which neither assumes increasing returns to scale, nor knowledge spillovers.
4. See Berryman and Bailey (1992) for a full discussion of this approach and the broad policy that is aimed at achieving.
Abramowitz, M. 1956. "Resources and Output Trends in the U.S. since 1870." Occasional Paper 52 NBER, New York.
Arrow, Kenneth J. 1962. "The Economic Implications of Learning by Doing." Review of Economic Studies 29: 155-173.
Bartel, P. Ann, and Frank R. Lichtenberg. February 1987. "The Comparative Advantage of Educated Workers in Implementing New Technology." Review of Economics and Statistics 69: 1-11.
Becker, Gary S. 1964. Human Capital. New York: Columbia University Press.
Benhabib, Jess, and Mark Spiegel. October 1992. "The Role of Human Capital in Economic Development." C. V. Starr Center for Applied Economics, New York University, Economic Research Report RR# 92-46.
Berman, Eli, John Bound, and Zvi Griliches. January 1993. "Changes in the Demand for Skilled Labor within U.S. Manufacturing Industries." NBER Working Paper No. 4255.
Berryman, Sue E., and Thomas R. Bailey. 1992. The Double Helix of Education and the Economy. New York: Institute on Education and the Economy, Teachers College, Columbia University.
Boldrin, Michele, and Jose A. Scheinkman. 1988. "Learning by Doing, International Trade and Growth: A Note." The Economy as an Evolving Complex System Reading. Ed. Philip W. Anderson, Kenneth J. Arrow and David Pines. Reading, MA: Addison Wesley.
Bound, John, and George Johnson. June 1992. "Changes in the Structure of Wages in the 1980s." American Economic Review 82: 371 393.
Commission on the Skills in the American Workforce. 1990. America's Choice: High Skills or Low Wages! Syracuse, NY: National Center on Education and the Economy.
Davis, Steve, and John Haltiwanger. 1991. "Wage Dispersion within U.S. Manufacturing Plants, 1963-86." Brookings Papers of Economic Activity, Special Issue: 115-180.
Denison, Edward F. 1961. Sources of Economic Growth in the U.S. New York: Committee of Economic Development.
---. 1974. Accounting for U.S. Economic Growth: 1929-1969. Washington, DC: Brookings Institution.
---. 1985. Trends in American Economic Growth: 1929-1982. Washington, DC: Brookings Institution.
Eicher, Theo S. 1993. "Interaction Between Endogenous Technological Change and Endogenous Human Capital." Columbia University, mimeo.
Findlay, Ronald, and Henryk Kierzkowski. December 1993. "International Trade and Human Capital." Journal of Political Economy 91: 957-978.
Friedman, M. 1962. Capitalism and Freedom. Chicago, Illinois: University of Chicago Press.
Grossman, Gene M., and Elhanan Helpman. 1991. Innovation and Growth in the Global Economy. Cambridge, MA: MIT Press.
Judd, Kenneth L. May 1985. "On the Performance of Patents." Econometrica 53: 567-585.
Juhn, Chinhui, Kevin M. Murphy, and Brooks Pierce. June 1993. "Wage Inequality and the Rise in Returns to Skill." Journal of Political Economy 101: 410-442.
Kaldor, Nicholas, and James A. Mirrlees. 1962. "A New Model of Economic Growth." American Economic Review: 117-192.
Katz, Lawrence F., and Kevin M. Murphy. June 1992. "Changes in Relative Wages, 1963-1987: Supply and Demand Factors." Quarterly Journal of Economics 107: 35-78.
Krueger, Alan B. August 1991. "How Computers Have Changed the Wage Structure." Princeton University, mimeo.
Levy, Frank, and Robert Murnane. September 1992. "U.S. Earnings Levels and Earning Inequality." Journal of Economic Literature 30: 1333-1382.
Lucas, Robert E. February 1988. "On the Mechanics of Economic Development." Journal of Monetary Economics 22: 2-42.
Mankiw, N. Gregory, David Romer, and David N. Weil. May 1992. "A Contribution to the Empirics of Economic Growth." Quarterly Journal of Economics 107: 407-437.
Marshall, Alfred. 1946. Principles of Economics. 8th ed. London: Macmillan.
Mincer, Jacob. 1974. Schooling Experience and Earnings. New York: Columbia University Press.
---. 1989. Labor Market Effects of Human Capital and of Its Adjustment to Technological Change. New York: Institute on Education and the Economy, Teachers College, Columbia University.
---. January 1991. "Human Capital, Technology, and the Wage Structure." NBER Working Paper No. 3581.
Mowery, David C., and Nathan Rosenberg. 1989. Technology and the Pursuit of Economic Growth. New York: Cambridge University Press.
Nelson, Richard R., and Edmund S. Phelps. May 1966. "Investment in Humans, Technological Diffusion and Economic Growth." American Economic Review 56.2: 69-82.
Piore, Michael J., and Charles F. Sabel. 1984. The Second Industrial Divide: Possibilities for Prosperity. Basic Books.
Psacharopoulos, George, and Maureen Woodhall. 1985. Economic Development: An Analysis of Investment Choices. Washington, DC: World Bank.
Romer, Paul M. October 1986. "Increasing Returns and Long Run Growth." Journal of Political Economy 94: 1102-1137.
---. November 1989. "Human Capital and Growth: Theory and Evidence." NBER Working Paper, No. 3173.
---. October 1990. "Endogenous Technological Change." Journal of Political Economy 98: S71-S102.
Schultz, T. Paul. August 1960. "Capital Formation and Education." Journal of Political Economy 68: 571-583.
---. March 1961. "Investment in Human Capital." American Economic Review 51: 1-17.
Solow, Robert M. February 1956. "A Contribution to the Theory of Economic Growth." Quarterly Journal of Economics 70.1: 65-94.
---. August 1957. "Technological Change and the Aggregate Production Function." Review of Economics and Statistics 39.3: 312-320.
---. 1960. "Investment in Technological Progress." Mathematical Methods in the Social Sciences, 1959. Ed. K. J. Arrow, S. Karlin and P. Suppes. Stanford: Stanford University Press.
Stokey, Nancy L. 1988. "Learning by Doing and the Introduction of New Goods." Journal of Political Economy 96: 701-17.
U.S. Commission on Excellence in Education. 1983. A Nation at Risk. Washington, DC: GPO.
Welch, Finis. January 1970. "Education in Production." Journal of Political Economy 78: 35-59.
World Bank. 1990. World Development Report. New York: Oxford Univ. Press. 79, 81.
---. 1991. World Development Report. New York: Oxford Univ. Press. 57.
Young, Alwyn. June 1991. "Learning by Doing and the Dynamic Effects of International Trade." Quarterly Journal of Economics 106: 369-405.
---. 1992. "A Tale of Two Cities." NBER Macroeconomic Annual. Cambridge: MIT Press.
---. June 1993. "Invention and Bounded Learning by Doing." Journal of Political Economy 101: 443-473.