Human Resource Development and Sustainable Growth

McCleery, Robert K

Abstract: Macroeconomic studies of total factor productivity in developing Asia yield results inconsistent with microeconomic observations of rising educational levels and increasingly sophisticated manufacturing products and processes. What can and do macroeconomic studies really tell us about growth prospects and challenges for Asian countries? Will Asia continue to emerge as an economic and political force? What role should or must human resource development play in sustainable growth?

1. Introduction

As the title of this paper indicates, I hope to bridge the considerable gap between two branches of the literature on economic growth and Asian development. The Asian financial crisis has alternatively been viewed as a temporary set-back on the path to a “Pacific Century” and an indictment of the Asian development approach. A full discussion of the causes and implications of the Asian Financial Crisis is outside the scope of this study.1 This paper merely looks at the real foundations for sustainable growth, assuming that important debt, liquidity, and financial market oversight issues are addressed. Conclusions will highlight both reasons for optimism and causes for concern.

The first branch of the literature tests for convergence, either in per capita income levels or in technological levels, between developing Asia and the OECD countries, or more generally between rich and poor countries globally. Some authors, most notably Paul Krugman (1994), have attempted to go beyond historical studies of convergence to draw implications about the future of growth in developing Asia. They ask the question, ‘is economic growth in developing Asia sustainable?’ Can the countries that experienced rapid growth in the 1980s and early 1990s regain those growth rates through this decade and beyond?

The second branch of the literature consists of assessments of the human resource development (HRD) status of Asian countries, and a few studies attempting to quantify the contribution of HRD (generally proxied by a single educational variable such as enrolment) to past growth. No real attempts have been made to go beyond historical assessments to look at the possible implications of today’s levels of HRD on future growth, or even to assess the quality and usefulness of the productivity estimates and the data on which they are based. Linking these two branches of literature would contribute to the quality and policy relevance of each.

This paper is motivated by the rapid rates of economic growth experienced by the Asian NIEs (Hong Kong, Korea, Singapore, and Taiwan), near NIEs (Malaysia and Thailand), and others (most notably China and Indonesia) in the last few decades, and the questions raised regarding the sustainability of growth both prior to, and in the wake of, the Asian financial crisis. After two to four decades of rapid economic growth, poverty reduction, and growing economic and political importance globally, these countries face severe challenges for the next decade. In some cases, the same economic, institutional and cultural factors that were seen as conducive to high growth previously (and perhaps worthy of emulation in other developing countries) are now seen as contributors to the crisis. Some western observers have gone as far as to imply that the Asian crisis (beginning with the end of Japan’s bubble economy through the prolonged economic and political difficulties in Indonesia) repudiates the last challenge to American-style capitalism and neoclassical economics.

2. Asia’s Growth Experience Revisited

Understanding and explaining past growth and evaluating the validity of criticisms of the sustainability of past growth is a prerequisite for estimating the long-run sustainable growth rate after the crisis. A full explanation of the rapid growth and the crisis is beyond the scope of this paper. Here, I will concentrate on showing that the critiques of Asian development did not anticipate the crisis, and taking a look at the role of human resource development (HRD) in past growth, the crisis, and future prospects.

It is clear that HRD has played a role in past growth and must continue to play a role if growth is to be sustained. But beyond that general agreement, the way different HRD elements fit together, how to understand and manage HRD-economy interactions, and how to evaluate HRD investments relative to other uses of scarce public funds are more problematic. I will use the abbreviation “HPAE” for “High Performing Asian Economies,” [namely Japan, Hong Kong, Singapore, Korea, Taiwan, Malaysia, Thailand, and Indonesia] following the terminology of the World Bank study The East Asian Miracle, for this set of countries that has experienced rapid and sustained growth over the past several decades. Various explanations for this rapid growth have been put forth, some stressing trade, others industrial policy, a few noting HRD levels and growth, and some emphasising factor accumulation. By conventional measures, rapid economic growth did take place, that growth was spread relatively evenly throughout the population, and most social development indicators, such as life expectancy, literacy, and nutrition have shown similar improvement. Other unifying characteristics of this growth include: rapid agricultural productivity growth, rapid growth of manufactured exports, high savings and investment rates, rapid urbanisation and structural change, and high levels and growth of human capital.

The high growth rates of GDP and GNP in the HPAE are well known, but the extent of the difference between their experience and that of other developing countries cannot be exaggerated. Table 1 shows per capita GNP and growth rates from 1980 to 1997. HPAE at every level of development substantially outperformed others at similar levels of development. High income Hong Kong and Singapore grew at 6.2 and 7.9 per cent, well in excess of the group as a whole, and China’s growth rate of 9.9 per cent is astonishing, relative to all low income countries. The levels of the two HRD variables (life expectancy and literacy) for the HPAE relative to countries at similar levels of development are impressive, but not as overwhelming. For instance, Malaysia’s high life expectancy is matched or exceeded by several Latin American and Eastern European countries with lower per capita income levels, and the same is true for the impressive adult literacy rates of China, Indonesia, and Korea.

The performance of the HPAE in industrialisation and export growth has been even more impressive than the growth of national income. It can be seen from Table 2 that the developing HPAE, with the exception of Hong Kong, have all moved from having somewhat small manufacturing output, compared with other countries with similar levels of per capital GNP. Korea, Indonesia, and Thailand have all undergone particularly sharp reductions in the importance of agriculture in GDP. Hong Kong and Japan are undergoing the transition from a manufacturing to a service-based economy. All of these economies have benefited from a sharp reduction in population growth, related to urbanisation, education, and income growth. The importance of these transitions in understanding current growth and projecting future growth is discussed in Section II.

Export growth has been rapid, particularly in the NIEs and in the latter (1980-93) period. The importance of export performance in economic growth can be inferred from the extremely high export/GDP ratios. But is reliance on foreign markets a sign of unsustainable growth? The short answer is that it depends on the relative stability of domestic and foreign markets. There is no reason to believe that the world market would be less stable than the domestic market. So, unlike a dependence on foreign capital inflows, which we have clearly seen as an element of unsustainable booms in Asia and Latin America, dependence on foreign markets through trade appears to be consistent with sustainable growth.

Data on the composition of exports can also be found in Table 3. While the broad composition of exports has been relatively stable in the U.S. and Japan, we see sharp shifts in the export composition in developing Asia, reflecting changes in the overall production structure and comparative advantages of the economy. The growth in export share of machinery and equipment has been particularly strong, especially in Korea and Malaysia, and is indicative of the ability of the economies to competitively produce increasingly sophisticated manufactures.2

Educational attainment, measured both through enrolment rates (Table 4), education of the labour force (Table 5) and expenditure per student (Table 6), has also risen sharply. Rapid labour force growth has combined with rising school attendance to increase the educational attainment of the labour force, although data for cross-country comparisons are particularly weak. The paradox is that, despite the rise, the educational advantage in Asian countries is much less now than in the 1960s, in comparison with countries with similar levels of development.3 There are two simple explanations. First, although enrolment rates rose on pace with economic growth in the HPAE, they also rose rapidly in a number of slower-growing developing countries. The second explanation is that income per capita declined in a number of high human capital countries, most notably Central and Eastern Europe and the former Soviet republics. As the HPAE move rapidly up the development ladder, their competition gets tougher.

Figures 1 and 2 illustrate this point. Countries above the line have more extensive human resource development than average, for their level of economic development, and thus should (all else being equal) be relatively attractive locations for new foreign and domestic investment.4 Countries below the line have poor human resources for their level of development. Note that the largest positive “outliers” are Ukraine, Poland, Chile, Urugauy, and the Czech Republic. Only South Korea stands out, among the HPAE, as a country with an advantage over others at the same level of development today. If a logrithmic relationship between the two variables is postulated, the trend line fits the data better and Korea’s advantage evaporates, as shown in Figure 2.

Thus if Asia’s rapid growth over the past decades rested in large part on its advantage in HRD, relative to other countries at the same level of development, how concerned should we be if that HRD advantage has been substantially eroded? The answer would seem to be that we should be quite concerned, unless we have reason to believe that the link between HRD and economic growth has fundamentally changed. The next section discusses the theoretical link between HRD and economic growth

2.1 HRD and Economic Growth: Which Causes Which?

As indicated in the above discussion, HRD measures grew in tandem with economic growth. Of course, causality might run in either direction. While increased human capital often makes workers more productive, higher worker productivity typically leads to higher income, resulting in more human capital investment.5 Thus we can imagine a virtuous circle, starting either with an exogenous boost in income or human capital. In the former case, this induces additional public and private expenditures on education and training, nutrition, and health, which in turn make the labour force more productive and further increase both wage and capital income. Wage income growth promotes additional private HRD expenditure, while rising capital income promotes complementary investment and further economic growth. Both generate more tax revenues for governments, which in turn increase investments in both HRD and complementary physical and social infrastructure. One can similarly trace the impact of an exogenous increase in human capital to higher labour and capital productivity, increased income, investment and growth etc.

The two crucial links are those between income growth and investment in HRD, and between investment in HRD and labour productivity.6 Untaxed capital income, land or natural resource rents may not induce HRD spending, especially when the distribution of the asset generating the return is very uneven. Thus growth based on extensive private landholdings (agricultural plantations in Argentina and Brazil), mineral wealth (diamonds in Botswana, oil in the Middle East, etc.) will only spur HRD if the government effectively taxes and transfers. Social structures that limit the labour force participation of women and minorities naturally reduce the potential benefits to society from improvements in their health and human capital. Similarly, increased HRD in the form of education not appropriate to the level of development of the country or state of the economy may not boost labour productivity. Indeed, strong ties between income growth and private HRD were deliberately weakened by the government in Korea and Japan, as spots in higher education programmes were rationed so as not to outpace the economy’s ability to absorb the graduates. Alternatively, economic stagnation may lead to unemployment or underemployment of human resources. If either link is weak, the virtuous circle breaks down.

The performance of the HPAE, in contrast to other developing countries, clearly illustrates this point. In the rest of the developing world, we see examples of HRD growth that have not been accompanied by income growth in countries like the Philippines, Sri Lanka, Cuba, Jamaica, Costa Rica, and Zimbabwe. Economic stagnation or collapse took place in several Eastern European countries and former Soviet republics despite relatively high levels of HRD. Alternatively, rapid income growth has taken place in some economies despite low or stagnating levels of HRD such as Pakistan, Botswana, Nigeria, Oman, Saudi Arabia, and South Africa.

The first and second groups of countries, those for which economic growth underperformed relative to HRD levels, were characterised by either strong government intervention in markets, macroeconomic instability, or both. The last group consists of countries that grew as a result of natural resource exports, were characterised by a very uneven distribution of income, or both. The HPAE seem to have avoided both the stifling of initiative caused by a crushing tax burden and/or high government deficit and the economic and social disruption of economic leadership and wealth concentrated in the hands of a small elite. Of course, war, external shocks, and other factors beyond the control of policymakers have played a role in shaping the performance of these and all other developing countries.

Actually measuring the direct impact of HRD investments on the economy, a necessary step toward econometric tests of causality, is difficult.7 Physical infrastructure generates direct returns once it is built, lasting until it is taken out of service. HRD generates both direct and indirect results, beginning as much as fifteen or twenty years later (in the case of, for instance, pre-natal care), and lasting perhaps for the lifetime of an individual. The impact of HRD may also be hard to isolate due to data problems, model misspecification etc. High private and public rates of return to education found in microeconomic studies generally do not translate into strong impacts on growth in single or multi-country macroeconomic studies, due to the above problems.8 Cross sectional rather than time series regressions for a single country exchange one set of problems for another. As pointed out by Behrman (1993) and others, “First, the returns from human resource investments vary with a country’s stage of development…Second, the rates of return from human resource investments depend on a country’s economic and social conditions and development strategy.”9 Also, there is no reason to expect a consistent, high coefficient on one particular type of education across all developing countries.

2.2 HRD-Economy Interactions

How much interaction between the HRD provision sectors and the economy is necessary? How much is desirable? How much is even possible?

Clearly some feedback from the HRD needs of the economy to the HRD providing sectors is necessary and desirable. Information on job prospects, wages, etc. should be available to those deciding whether to advance to the next level of education or what track or specialisation to pursue. To the extent that different tracks or specialisations receive different levels of public subsidy, the decisions on whether or how much to expand opportunities in various subsidised disciplines should be made with an eye to both supply and demand from the economy. An example of an error that could perhaps have been avoided with better policy coordination is the rapid expansion of vocational education programmes and graduates in Thailand without a corresponding increase in demand for those graduates. The result was a combination of costly expansion of the vocational education programme and a sharp rise in unemployed vocational school graduates, from 13 per cent of total open unemployment in 1977 to 31 per cent in 1986, even as total open unemployment was rising by more than 10 per cent per year.10 Since vocational education is quite costly, and much of that cost is borne by the state, it appears that an expensive failure to coordinate educational policy to the requirements of the labour market occurred.

Similarly, the health care needs of a society change during the transition from a rural, agricultural society to one dominated by urban manufacturing and services. The former requires basic care through small-scale local units instead of large, centralised hospitals. Rather than the much-cited statistics on doctors and hospital beds per 1000 people, which are much more appropriate for urban, industrial societies, nurses or other health care professionals per 1000 people and measures of access, such as travel time and cost to the nearest health care facility may be more revealing of the health care situation. Traditional health care providers and medicines are also excluded from these statistics.

This is not to say that it is necessary or desirable to have HRD provision infrastructures that precisely mirror the needs of the labour force and the economy in general. The output of the educational system must serve the economy for perhaps 40 years. Similarly, the structure of today’s economy is generally not relevant to the primary education system. Learning to learn, rather than specific knowledge, is needed given the rapidly growing and evolving economies of the HPAE. Flexibility to respond to changing economic conditions is a desirable end in itself, as any effort to anticipate the future course of the economy 10 to 20 years in advance is doomed to failure.11 Where such decisions do not impose undue costs on others through government subsidies, students should have latitude to pursue their interests, both to maximise social welfare rather than just economic growth, and also on the grounds that such interest and dedication increase productivity. Efforts to channel students into socially productive occupations in Japan, Taiwan and South Korea by making permission to study abroad contingent on selecting a major in hard sciences like engineering may have seemed like an effective use of scare human resources and foreign exchange at the time.12 But it may have left these countries short of key personnel in finance, business, and administration today.

Examples abound of the potential forecasting errors in anticipating manpower requirement and the dangers of trying to manipulate the HRD system to hit a moving target. In the case of Malaysia, the Fifth Malaysia Plan projected HRD requirements for 1990, five years in advance. The projections underestimated professional and technical workers by more than 30 per cent, and administrative and managerial workers in 1990 were more than twice the number predicted! Root mean squared errors (a common measure of forecasting error) of the five previous forecasts since 1970 range from moderate (18 per cent) to bad (45 per cent).13

In Korea, a rapid relaxation of the quotas on college enrolment in 1981 led to huge increases, first in enrolment, then graduates, and ultimately unemployed graduates. From 1980 to 1985, college enrolment rates jumped from 16 to 36 per cent.14 By 1985 there were an approximately 110,000 unemployed college graduates, representing an unemployment rate among college graduates of 6.6 per cent. The situation seemed to be worsening dramatically, as more than 40,000 members of the 1986 graduating class (nearly 30 per cent overall, and more than 45 per cent of the female graduates) reported that they could not find a job.15 The authors of that 1990 study were rather worried about this situation and critical of the government’s haste in relaxing the quota on higher education enrolments, as illustrated in the following quote:

“However, this action in 1981 was taken too abruptly, and changes in the quota were made with little consideration of their potential impact upon the future labour market. In retrospect one may question the validity of such an enlargement of college enrolment by looking at the current level of the unemployment rate among college graduates…. The high level of unemployment among college graduates seems to have resulted from the structural imbalance between the demand in scientific and technical fields, on the one hand, and the excess supply in the area of non-technical and liberal arts, on the other.”16

In retrospect, it appears that the fears raised by the authors of the 1990 study were exaggerated, if not unfounded. Despite a continuing rise in college enrolment rates through 1990, from 36 to 38 per cent, the unemployment rate for college graduates fell to 4.5 per cent.17 Although that figure still represents a large number of people, it is not likely to be a problem for several reasons. First and foremost, the future of the Korean economy clearly lies less in standardised manufactures and more in services and technology and knowledge-intensive manufactures. Thus college educated workers, not just engineers, but graduates in business, economics, government, and other less technical fields, will have an increasing role in the economy. In addition to changes in the composition of output of the manufacturing sector, its share of the economy has remained constant since 1985, and may even have declined since 1990 (Table 2). secondly, there is evidence that market signals are working, through diffusion of information about job prospects and wages from graduates to students. Naturally the initial increase in enrolment consisted of some who had neither the inclination nor ability to compete in technical fields, and who did not realise the implications of the large cohort size for employment prospects. By 1990 the share of engineering students had edged back up toward the pre-boom level and there was an increased realisation that the diploma alone would be insufficient to guarantee economic security.

From a discussion of the theoretical and policy issues related to HRD, we now turn to the central question of the paper: is the rapid economic growth observed in Asian countries the result of technological progress, related to growing HRD levels and more effective technology transfer (working smarter), or the result of hard work and sacrifice, from high savings rates and a rapidly growing work force (working harder)? The answer to that question carries strong implications for the question heading Section 3 below.

3. Is HPAE Growth Sustainable?

We suspect that HRD, particularly human capital accumulation, is an integral part of technological progress. We have also seen the rapid industrialization and technological upgrading that has been taking place in the developing Asian countries. Naturally, we believe that as the products produced in the HPAE become increasingly advanced and competitive with the products produced in the OECD countries, the technological gap with the “West” is diminishing. Indeed, some of the manufacturing production taking place in the Asian NIEs involves factories, machinery, and management methods identical to those of the OECD-based parent firms. With this impressive modernisation, and the enviable track record of rapid growth through oil shocks and other changing international circumstances, few questioned the sustainability of Asian growth before the financial crisis beginning in the summer of 1997.

But at this point many questions exist. Does this crisis mark an end to rapid growth in developing Asia, as the crisis spreads throughout the region, bringing on economic stagnation such as Japan has experienced since the bursting of its economic bubble in the early 1990s? Or is it merely a brief downturn caused by a liquidity crisis, with growth rates quickly recovering to historical levels? Or will different countries face very different experiences in their attempts to restart growth?

Krugman’s now famous (or infamous) argument on the pitfalls of economic growth in the HPAEs runs as follows. Asian growth is characterised by very high levels of savings and domestic investment and rapid increase in the effective labour force, but with relatively low levels of technological progress. Thus, if growth is driven by increased inputs rather than by “catching up” with best-practice technologies, then it will be subject to decreasing marginal returns. The implication is that, for a fixed rate of domestic investment that exceeds the growth of the effective labour force, the rate of real GDP growth will decline over time. Unfortunately, at that point in his argument, he compares the “envy” of rapid Asian growth to the trepidation inspired by high reported growth rates of Russia and Eastern European satellites during the Cold War, claiming that both were ephemeral if not illusory and that there are no “lessons to be learned” from either. This apparent exaggeration for effect caused many to dismiss his ideas out of hand, rather than to examine and carefully refute those that are faulty. The last link in his logical argument, that an inevitable slowdown in growth in Asia threatens the emergence of a ‘Pacific Century’ in the 2000s, is also suspect. Let us first look at the obvious differences between the HPAE and the socialist countries as listed below, before considering the Krugman argument and the more rigorous studies on which his arguments are based.18

The Soviet economy was fundamentally different from the economies of the HPAE in several obvious and significant ways. Trade and other forms of economic openness figure significantly in the success of the HPAE. With the exception of Japan, all the HPAE rely on imported capital equipment. Exports and direct foreign investment figure prominently in manufacturing growth, the adoption and adaptation of new technologies, and in the generation of foreign exchange to relieve constraints to growth.

The economies behind the “iron curtain” were exactly that – cut off from technological development, markets, and competition in the West. Not only were the flow of goods, investment, and technology restricted, but the movement of people was tightly controlled as well. In terms of actually measuring economic growth, such measures for the socialist countries relied, in the absence of market prices to use in aggregating production across sectors, on concepts such as Net Physical Product. Measurement of the value or quality of a product, as measured by the willingness and ability of consumers to purchase it, cannot be captured in such a system due to the lack of market prices. Allocation by rationing at low, controlled prices hid the poor quality of goods. Only in military and aerospace did the power of and feedback from the “consumers” result in some semblance of quality control. The markets of the HPAE were largely free; where protection of domestic markets did create opportunities for non-competitive behaviour by producers, administrative guidance, often in the form of export requirements, kept producers “honest.” Pressure to export, even more than pressure from imports, forced domestic producers to improve product quality, production techniques, and overall efficiency. The two cases are clearly not comparable.

Coming back to Krugman’s argument and Asian growth, the law of diminishing returns does not imply economic collapse, such as what befell the Eastern Bloc, merely a gradual slowdown in growth, for a given savings rate. And we should question the validity of the microeconomic concept of diminishing returns as applied to a national economy undergoing rapid structural change. New investment may not be adding to productive capacity in existing sectors, but being applied to the development and production of new products instead, thus diminishing returns clearly don’t apply. It is not even clear that the capital-labour ratio, in effective units, is even increasing in all Asian countries. In a growth accounting exercise for Korea, Pyo (1995) measures growth in physical capital at 9.8 per cent per annum, less than growth of human capital (10.3 per cent per year).

Krugman’s own logic does not imply crisis but rather a soft landing for Asia, perhaps just short of the advertised “Pacific Century.” But surely the Pacific Century depends not on productivity growth per se, but more on growth in industrial capacity, defence capabilities, and the willingness and ability to contribute to or lead political debate. All of these areas are growing apace with GDP, if not more rapidly.19

Before attempting to further refute Krugman’s claim, we must examine one additional piece of evidence in support of the claim of low total factor productivity in Asian development. Remember that we are considering macro measures of “pure” total factor productivity from the estimation of a one-sector production function. One hallmark of developing economies, particularly at fairly early stages of growth, is a rapid transformation of economic structure from agriculture to industry. This movement of labour from low to high productivity work, called the “shift factor,” can account for as much as one-third or more of total factor productivity (TFP) in industrialising countries experiencing rapid structural change.20 In some HPAE, this shift has been strong and continuing, from agriculture to labour-intensive then capital-intensive and the beginnings of high-tech industries. In this case, measured productivity growth for the developing HPAE overstates sustainable productivity growth within major industrial sectors. Clearly, once the shift of labour from agriculture to industry has lowered the share of labour in agriculture to a certain point, little additional productivity growth can be gained in that manner.

Two other factors help explain the remarkable growth in East Asia. A related “shift factor” is the move from rural to urban areas. To a great extent, this shift overlaps the shift from agriculture to industry, but emphasises the economies of scale to the provision of social infrastructure in the main cities. Agglomeration effects, related to the spread of technology, trade-related infrastructure, and other social infrastructure investments seem to have been significant. Secondly, and related to both urbanisation and rapid income growth, we have seen rapid demographic transition in East Asia. Participation ratios have increased as women, pushed by the reduced birth rate and pulled by higher educational attainment and greater employment opportunities, enter the labour force in unprecedented numbers. With the labour force growing significantly faster than the population, the increase in participation ratios alone explains 17-20 per cent of per capita growth in three of the ANIEs and a whopping 40 per cent of per capita growth in Singapore.21

The boost to savings, output, and employment from the demographic transition also clearly cannot be sustained indefinitely. Japan and Hong Kong already face a growing burden of caring for retirees, which will reduce future savings and income growth. Korea and Taiwan should reach this point in the next 10 to 20 years, with Southeast Asia benefiting from the demographic transition for considerably longer.22

In Southeast Asia, the shift factors are likely to be dominant in recent rapid growth figures, while the demographic transition will be increasingly important in the next few decades. Urbanisation is certainly progressing throughout Southeast Asia, although at different rates and levels.

3.1 The Numbers: Kim and IMU

What precisely do the productivity estimates tell us? How consistent and reliable are the estimates? The numbers on which Krugman bases his argument are generated by three main studies: Kim and Lau (1994), Young (1995), and World Bank (1993). Without presenting a detailed review of findings and methodologies here, the general result is that pure technological progress in at least some of the HPAE was lower that the average in the OECD countries. Several other studies exist, coming to different conclusions using different methodologies, countries, and time periods.23

Hsiao and Hsiao (1996) make precisely this point in their review of estimates of productivity growth in developing Asian countries. Unfortunately, they make no effort to distinguish between modern and old methodologies, giving, for instance, implicit equal weights to various studies from Williamson (1969) to Young (1995). But it is certainly not true that all the HPAE have lagged behind in all periods.

Each of the two “new methodologies” introduced by Young, and Kim and Lau make a legitimate contribution to the measurement of productivity gains, and makes a reasonable case that doing studies the “traditional way,” using methodologies little changed from the work of pioneers like Williamson (1969) and Dennison (1974), may result in serious errors. But each study makes a different “breakthrough,” thus each is guilty of a potentially serious error according to the logic of the other.

Kim and Lau (1994) make a case against the assumptions of constant returns to scale, neutrality of technological progress, and profit maximisation, uniformly made in all other studies. Yet they make a much weaker case for their alternative-the “meta-production function” approach. The meta-production function is tested and accepted, but we know that failing to reject a null hypothesis at the 0.01 significance level may or may not be a meaningful experiment, depending on the power of the test. And the above findings of non-constant returns to scale etc. are all conditioned on the meta-production function being the correct specification.

Again using the meta-production function, Kim and Lau (1994) find that their estimates of the capital augmentation parameter (which is related to but not the same as the rate of TFP growth), is statistically significant at the 0.01 level only for the five OECD countries in the sample, not for the four ANIEs they studied (Hong Kong, Singapore, South Korea, and Taiwan). This is another “non-finding,” which has no meaning unless the power of the test (still conditional on the correctness of the form of the production function, not to mention the quality of the data) is established. The fact that all technological progress is capital augmenting in their model is another source of concern, as it implies that even the modest rates of technological progress observed are dependent on the high investment levels. If those investment levels come down, as a result of income growth, changing priorities as a society, lower savings rates due to aging populations and an increased social security dependency burden, or Krugman’s presumed decline in the rate of return to capital, TFP may fall proportionately.

Little attention has been given to the fact that some of their technological progress coefficients estimated for the East Asian economies are actually higher than those obtained using traditional methodologies, due to the estimates of returns to scale parameters that are less than one and decreasing over time for all of the economies studied. The emphasis in both the paper and the subsequent debate has been on the lack of convergence with the U.S. in terms of technical efficiency, and what implications that might have for future growth in East Asia.24 Yet the picture one gets from their Table 7.1 and Figure 8.2 is strong convergence for Hong Kong, gradual convergence for Singapore, and gradual divergence for both of the larger ANIEs, Taiwan and Korea. When weighted by size, the small divergence factors (-0.003 in both cases) outweigh the larger convergence factors for the two city-states (+0.009 and +0.004, respectively), leading the authors to say that there is divergence with “East Asia.” Given the point noted above, that the estimation procedure was sufficiently imprecise as to not find the capital augmentation parameters for the ANIEs statistically significant, this difference of less than 0.002 is clearly not statistically significant. In fact, since the confidence intervals around the estimates of the capital augmentation parameters for the ANIEs includes zero, Kim and Lau (1994) cannot rule out rates of technical progress of twice the levels shown here, roughly twice the U.S. level!

In fairness to Kim and Lau (1994) they clearly realise the limitations of their study and the need for further work. In addition to the problem of potentially different rates of depreciation in the two groups of countries (if, as in Japan, depreciation is more rapid in the ANIEs due to product mix, technologies, etc., the relative technological progress figures are biased towards the OECD countries), the problem of omitting land from the production function is mentioned, and they recognise that “Another potentially omitted factor of production is R&D or knowledge capital, which must have been growing over time, especially in the G-5 countries.”25

They close with a short list of possible ways to, in my words, reconcile their results with what we know about East Asian growth and development. Most telling are two points:

(1) “…for the East Asian NICs, there has been, until very recently, very little investment in research and development, especially in basic research. Thus the indigenously generated improvements in technology must be quite scarce.”26

(2) “…their industries are, at least until recently, by and large not knowledge or technology intensive. Moreover, these industries typically employ mature technologies, the capital equipment for which, being mostly imported, has been fully priced (i.e. the acquisition as well as royalty costs fully reflect the possible efficiency gains and the amortisation of R&D and other developmental costs) in the international market.”27

Interpreted this way, the paper serves as a warning that to catch up rapidly to the OECD countries in technology will require improvements in the support for R&D and quality of the personnel involved, and further upgrading of industry to include the production of more knowledge-intensive and technology-intensive consumer and capital goods as growing domestic capacity allows.

In summary, Kim and Lau (1994) make a much stronger case for the standard methods being incorrect than for their method being correct. Depreciation and human capital issues are ignored. Output, capital, and labour are all aggregated to single figures, in contrast to Young ( 1995) discussed below. In my view, the authors establish only that rates of technological progress in the ANIEs that do not dwarf those in the OECD countries; in other words, the primary reason for the high growth of the ANIEs has not been a “catching up” technologically with Japan and the West.

3.2 The Numbers: Young

Young’s (1995) primary contribution is to deal much more carefully with the data and form of the production function. He also uses the trans-log specification, but disaggregates output by manufacturing and non-manufacturing, labour inputs by sex, age, and education, and capital by carefully deriving asset stocks and depreciation rates for five types of capital. He makes extensive use of published and unpublished Census and survey data from the ANIEs to construct these detailed series. His methodology also allows the estimation of TFP by sub-period, although the results for short sub-periods may be biased due to changing levels of capital utilisation or other quirks of the end-points chosen.

Young’s results indicate that his more detailed look at labour and capital inputs yield higher TFP estimates for three of the four countries. As noted above, the Kim and Lau (1994) estimates, with returns to scale parameters of less than one and falling over time, should (and do, for the G-5 countries) produce larger TFP estimates than does the standard methodology. But in three of the four cases, Young’s (1995) estimates are the same or larger: Hong Kong (2.4 per cent versus 2.4), Korea (1.6 versus 1.2), and Taiwan (2.4 versus 1.2). The case of Singapore yields wildly different estimates under the two methodologies (-0.7 per cent versus 1.9), indicating a potentially serious flaw in one or both studies.

Taking the two studies together, they show us that we, as economists, still have a long way to go before we can adequately understand and measure technological progress. It is important to note that the years since the two studies have not seen further breakthroughs or reconciliation.

4. Conclusions

In conclusion, we cannot completely refute the “Krugman Critique” of Asian growth. Convergence in technologies between the Asian NIEs and the OECD countries, in the technical definition used in these studies and to the extent that we can accurately measure it, seems to have been slower than most observers thought. Considering the favorable impact on productivity over this period of unsustainable structural and demographic changes only strengthens the critique. And if productivity growth is linked to savings and investment, as modelled by Kim and Lau (1994), a possible demographically or culturally-based savings slow-down in coming decades would affect income directly and productivity growth as well. Yet much of the flavour of Asian development, particularly the strong human resource development record, policy flexibility, and drive to excel, is not captured in these formal models. And the promise of the “Pacific Century” lies not so much on future productivity growth, but on the undeniable expansion of production in the region, and the expectation that the economic giants of the region will ultimately achieve commensurate political structure.

It would be nice to be able to say that HRD will provide the answer to enable productivity growth to be sustained of increased, even if investment rates decline with the maturation of the economies. But the truth is that we clearly do not understand enough about HRD, its impact on the economy, measurement of its rates of return, the dependency of such rates of returns on specific cultures, institutions, or even level of development in a given society to make such bold predictions. Our basic concepts of how to turn measures of HRD into fodder for econometric regressions are not sufficiently developed to allow us to measure with any certainty the contribution of HRD in its broadest definition to economic growth.

There are many problems and idiosyncrasies with the HRD data in individual countries that make cross-country or even time series comparisons dangerous. In some cases, the providers of data have a vested interest in the magnitude of the numbers, as funding may depend on enrolment, number of people served, or perceptions of the quality of the service. Improvements in data collection, distribution, and timely evaluations of HRD programmes are required not only for academic studies such as this, but more importantly, for agents in the economy/society to make effective decisions.

4.1 Suggestions for Further Research

It seems that both extremely careful analysis of the basic data and advanced econometric estimation of complex production functions are necessary to derive precise estimates of growth in total factor productivity. No study as of yet is immune from criticism in at least one area. Yet it is clear from the range of estimates presented in recent studies that growth in the Asian NIEs has not been led by TFP growth, despite the rapid modernisation of production methods and machinery. Whether or not convergence in pure productivity is taking place between the ANIEs and the OECD countries can still be debated, but it is clear that any such convergence is much less rapid than the convergence in GDP or GDP per capita. The reason why these results are not inconsistent with our observations of the increasingly modern machinery and methods of production in East Asia is that the measures of pure productivity growth currently in fashion are extremely narrow. Capital and labour are measured in “efficiency units,” so that increases in capital productivity (as reflected in prices) and human capital show up as increased factor supply, rather than productivity growth.

Secondly, we need to view the HRD requirements for growth in a dynamic fashion, as something that evolves with the economy. East and Southeast Asia were extremely well served by high levels of primary school enrolment and basic literacy in their first phases of industrialisation and growth (the 1950s and 1960s for Japan, 1960/65 to 1980 in Taiwan/Korea, and the 1980s and 1990s for Southeast Asia). Part of the benefit is that the provision of human capital is a very human capital-intensive process. It may be that some other developing countries were caught in a low human capital trap, where additional resources devoted to HRD yielded low returns because of the initial low levels.28 The demographic transition from high to low fertility rates was important in raising labour force participation rates, increasing savings through lower dependency ratios, and allowing increasing per capita expenditures on health care and per pupil expenditure on education without straining government budgets. The demographic transition, urbanisation, and industrialisation all have an impact on the ideal HRD provision systems and priorities for the future. To some extent, the experiences of neighbouring countries at higher levels of income in the “flying geese formation” or higher on the product development ladder may prove beneficial to countries like Malaysia. However, it is easy to use a simple analogy like the flying geese to understate the differences among the individual East and Southeast Asian countries, and thus understate the need to provide solutions based on the strengths and weaknesses of each individual country.

Thirdly, an evaluation of HRD policies must be done in a general equilibrium context. HRD policies and practices work, or fail to work, within a configuration of other factors, which are seldom fully integrated in the analysis. Economic, social, and political structures all jointly determine not only what will work, but also what is feasible to try. One simple example might be that social and cultural pluralism, which has inhibited growth with equity in many countries in Africa and even, to some extent, in the United States, seems to have been transformed into an asset in the context of the economic, social, and political structure of Malaysia.

Fourthly, there is a limit on what the state can do. No government can effectively forecast labour demand even 4-5 years ahead in these rapidly changing economies, and for most educational decisions, 10-40 years is the relevant time frame for calculating relative returns. In deciding where to place scarce resources, a government should follow two basic principles: invest not for the highest return, but where the difference between the private and social returns are the highest, and reward merit, while preserving equality of opportunity. Equality of opportunity, however, does not mean that anyone who wants to should get a subsidised college education!

Apart from the uncertainties of predicting future manpower requirements, it is obvious that development, at a minimum, means change. It means the obsolescence of old skills and the appearance of new opportunities. This is not a painless process; some will be left behind by progress. Educational contexts and styles of instruction differ in creating a capacity to accept or adapt to change. They differ in the extent to which imagination or conformity are recognised and rewarded. We are, at this point still far from understanding what type of education is most consistent with individual or economic development.29 But education, training, and human resources in general will only be more vital to growth in the new, globalised, information economy of tomorrow.

4.2 What Lies Ahead for Developing Asia?

To conclude, let us consider the impact of openness and the international environment on continuing economic growth in developing Asia. Clearly rapid growth has been sustained by the diffusion of technologies through direct foreign investment and imports of improved capital and intermediate goods. Openness to new technologies and products, plus the human resources required to understand them sufficiently to use them, and in some cases duplicate or even improve on methods or products to conform to local conditions, is one of the reasons for the relative success of developing Asia in the 1970s, 1980s, and 1990s. What may happen to reduce these advantages in the future?

First, the advantage held by the Asian developing countries in terms of openness to trade and foreign investment flows is shrinking. “Ironically, many governments (notably in Latin America) ended up implementing policies that went far beyond what the East Asian governments themselves had adopted since the 1960s…. Mexico, Bolivia, and Argentina, to cite some of the more distinguished examples, have undertaken more trade and financial liberalisation within five years than the East Asian countries have managed in three decades.”30 Political stability, long a hallmark of Asian growth, is either threatened (Indonesia, China and Taiwan) or revealed as perpetuating an inappropriate pattern of business-government relations (Thailand, Japan, and South Korea). With each instance of rocky transition, any impending political transition, such as that in Malaysia, is increasingly questioned by investors.

Secondly, despite the rapid advances made, the relative human capital advantage enjoyed in the 1960-70 period over countries with similar levels of income has eroded with rapid income growth.31 It is important to note that this is not a fault of policy, it is an unavoidable cost of success. Upgrading the quality of the labour force in line with rapidly rising income levels is made even more difficult in situations where possibilities for immigration and emigration exist. However, as wages rise in developing Asia, Latin American countries with similar levels of openness and human capital may prove to be equally attractive locations for investment, provided they maintain their economic and political stability. In the next ten years, the growth advantage some HPAE may have is likely to rest less on superior human capital and outward-oriented economic policies and more on infrastructure and manufacturing sector linkages than in the past.

Thirdly, intra-regional trade among the HPAE, particularly among ASEAN countries, increased rapidly in the 1990s. This “contagion effect” has been noted by other authors, and creates a “virtuous circle” of investment, trade, and economic growth…provided the region’s economies keep growing.32 With Japan already experiencing economic weakness and thus being a drag to growth in other Asian trading partners, the “virtuous circle” turned “vicious” with the crisis in Thailand, Indonesia, South Korea, Malaysia, and the Philippines, dragging down even the best managed economies in the region. The survivors were countries less highly linked to the region and less open, in terms of trade and finance (China and Vietnam). It was hoped that ASEAN economic cooperation would be a buffer to strong external shocks while attracting foreign investment with its large and growing combined market, but the poor prognosis for Indonesia now casts a pall over its smaller neighbours.

Taking these three points together, are there reasons to believe that the erosion of competitive advantage in Asia might lead to a slow-down in growth due to a redirection of investment to Latin America or elsewhere?33 Perhaps. Mexico has quickly recovered from its crisis of 1995 (with much more substantial assistance from the US than was provided to Asia), and remains an attractive investment location. President George W. Bush is known to place a high priority on expanding US trade with Latin America, and may push the FTAA (Free Trade Agreement of the Americas) through the US congress. The EU is preparing to expand, with Hungary, Poland, and the Czech Republic being likely candidates for new infrastructure investments and greater trade with the current EU members. Asian countries are not likely to dominate the developing country growth tables in the future, but that may be due as much to improved growth prospects in Central and Eastern Europe and Latin America as to slower growth in Asia.

One final potential problem is the world trading system in general. The past thirty years have been a time of unprecedented expansion in trade and elimination of trade barriers. Transportation and communication revolutions have reduced the costs of moving goods, people, and ideas around the globe, and peace prevailed. The optimist might say that people have learned that it is better to trade than fight, and that reducing trade barriers and spreading technology have clearly benefited all, and that we should expect this situation to continue.

We all hope that this is true, but the pessimist would point to the increasing military buildup in Asia, tensions on the Korean peninsula, separatists and unrest in Indonesia, shaky US-China relations, Taiwan, and the South China Sea. The sustained period of globalisation and prosperity experienced since World War II or the Korean War is already unusual in human history. Another thirty years would be almost unprecedented. As we have seen, along with benefits of openness comes greater vulnerability to price shocks, disruption of trade etc. And along with wealth comes the temptation for others to take it. Although openness, acknowledging its dangers, is clearly the best policy for the long term, in the future, it need not be the unqualified success that it was in the past.

One does not have to be a sociologist to understand that people take pride in the aspects of their systems that work well, and perhaps discount the most effective parts of their rival’s systems. I think that both sides of the debate have been guilty of some exaggerations. ProAsia voices are guilty of making implicit projections based on past growth, without fully considering what that implies for the economy, the environment, and society. Detractors such as Krugman are guilty of underestimating the flexibility of the Asian economies, as well as the benefits of high levels of human capital, openness, and regional dynamism and cooperation. In portraying the difference between East and West as one between authoritarianism versus personal liberty, with the former system better at mobilising resources and the latter better at fostering innovation, Krugman is guilty of the very sin he preaches against in his paper-generalising “Asia” from the diverse economic and social systems of the individual high-growth countries. Instead, it seems far more likely that the growth rates of individual economies in Asia will diverge. Some, most notably Korea, Taiwan, and China, seem poised to regain “miraculous” rates of growth for some time, given their stage of development, human resources, and other factors. Others, like Japan, Singapore, and Hong Kong, seem likely to experience a gradual slow-down in economic growth, but not necessarily in social welfare.

1 See, for example, Montes (1998), Goldstein (1998), World Bank (1998).

2 The World Bank’s World Development Report and World Development Indicators database dropped this series, so it was not possible to update it.

3 In the East Asian Miracle (World Bank: 1993: 45) the results of a regression by Behrman and Schneider of per capita income on primary enrolment rates are presented. Indonesia, Thailand, Malaysia, South Korea, Singapore, and Hong Kong all displayed positive residuals in the regression using 1965 data, indicating that these countries had primary enrolment ratios from 8 to 29 percentage points higher than the trend line, or average relationship, given their income levels.

4 HDI-Social indicates the UN Human Development Index, without GDP per capita.

5 Other human development increases, such as health care for the elderly and terminally ill, and even some facets of a liberal arts education, are actually consumption expenditures.

6 The third link, that between labour productivity and income, seems to always hold, at least in the long run.

7 Clarete and Orani ( 1995).

8 Krueger and Lindahl (1999).

9 Behrman (1993: 187).

10 Sussangkarn(1990).

11 This is true not just for rapidly growing and changing developing countries, but even for the United States. The U.S. Bureau of Labor Statistics predicted growing shortages of college graduates in the 1970s and a surplus in the 1980s; neither prediction was correct. Their 1996 prediction of a surplus by the year 2000 was rightly treated with suspicion, if not outright derision, by the private sector. see Barron’s, July 1, 1996.

12 ADB (1990).

13 Lee, Ng and Yeoh (1995), compiled from Malaysian development plans, various years.

14 Choo and Cheong ( 1995).

15 Kim, Kim and Ihm in ADB (1990).

16 Op. cit., pp. 150-151.

17 Choo and Cheong (1995).

18 Kim and Lau (1994); Young (1995)

19 Defence spending has been growing about 20 percent per year in the region, to the dismay of many observers; see the Far Easstern Economic Review (1995). Japan, China, and ASEAN have been taken larger and more assertive roles in international organisations and political debates.

20 Reynolds (1978).

21 Young (1995: 3).

22 Mason (1997).

23 Jorgenson (1995); Kawai (1995); Liang (1995).

24 Note that one other prominent point in this literature is that the percentage of growth attributable to technological progress is much lower in East Asia. This is clearly not relevant to the future of Asian growth, unless one shows that technological progress is directly and positively linked to factor accumulation rates.

25 Kim and Lau (1994), p. 257.

26 Op. cit., p.265 (author’s emphasis).

27 Op. cit., p.265.

28 Low estimated returns to additional years of schooling in Africa may reflect the poor quality of teachers and schools.

29 For a more detailed discussion of these issues, see the paper by Lee (2000).

30 Rodrik (1996: 17-18); Edwards (1993).

31 See Tables 1 and 5, and Figure 2. Some exceptions include Korea’s high secondary and tertiary enrolments, and primary school enrolments in Indonesia and China. However, note also the incompleteness and non-comparability of data across countries, owing to differences in definitions, measurement error, and quality of education.

32 Petri (1993).

33 Perhaps to the high human capital countries of Eastern Europe as well, once supportive legal systems, macroeconomic and political stability, and appropriate institutions have been demonstrated?

References

Asian Development Bank. 1990. Human Resource Policy and Economic Development: Selected Country Studies. Manila: ADB.

Balassa, Bela and Associates, 1971, The Structure of Protection in Developing Countries. Baltimore: The Johns Hopkins University Press.

Barron’s, July 1. 1996.

Behrman, Jere R. 1993. Investing in human resources. In Economic and Social Progress in Latin America, 1993 Report, pp. 187-255. Washington D.C.: The Inter-American Development Bank.

Bello, Walden and Stephanie Rosenfeld. 1990. Dragons in Distress: Asia’s Miracle Economies in Crisis, London: Penguin.

Bhagwati, Jagdish. 1986. Rethinking Trade Strategy. In Development Strategies Reconsidered ed. J. Lewis and V. Kallab. Overseas Development Council, New Brunswick, New Jersey: Transaction Books.

Choo, Hak Chung and Kimoon Cheong. 1995. Human resource development and economic progress: The experience of Korea. Mineo, Korean Development Institute.

Clarete, Ramon and Prudencia Orani. 1995. Policy imperatives for human resource development in the Philippines. Mimeo, Korean Development Institute.

Colclough, Christopher. 1996. Education and the market: which parts of the neoliberal solution are correct? World Development 24(4): 589-610.

Dennison, E.F. 1974. Accounting for United States Economic Growth, 1929 to 1969. Washington D.C.: The Brookings Institution.

Dollar, David. 1992. Outward oriented developing economies really do grow more rapidly: evidence from 95 LDCs, 1976-1985. Economic Development and Cultural Change 40(3).

Easterly, William. 1993. How much do distortions affect growth? Journal of Monetary Economics 32:

Edwards, Sebastian. 1993. Openness, trade liberalization, and growth in developing countries. Journal of Economic Literature 31: 1358-1393.

Fry, Maxwell J. 1988. Money, Interest, and Banking in Economic Development. Baltimore, Maryland: The Johns Hopkins University Press.

Fukuda, Shinichi and Hideki Toya. 1994. The role of human capital accumulation for economic growth in East Asian countries. Seoul Journal of Economics 7(1): 213-232.

Glewwe, Paul and Dennis de Tray. 1989. The poor in Latin America during adjustment. LSMS Working Paper No. 56, Washington, D.C.: The World Bank.

Goldstein, Morris. 1998. The Asian Financial Crisis: Causes, Cures, and Systemic Implications. Washington D.C.: Institute for International Studies, Policy Analysis in International Economics #55 (available on-line at http://www.iie.com/publications/publication.cfm?pub_id=22)

Hsiao, Frank and Mei-Chu Hsiao. 1996. Miracle or myth in the Asian NICs’ growth-The irony of the numbers. Mimeo presented at the Seventh Biennial Conference on U.S.-Asia Economic Relations, Kuala Lumpur, Malaysia, 14-19 May, 1996.

Hughes, Helen. 1993. East Asia: Is there an East Asian model? Economics Division Working Papers, Research School of Pacific Studies, Australian National University.

Japan Times Weekly. 1992. Japanese are Getting Richer, but Lifestyles are Still Poor. 25-31, May.

Johnson, Chalmers. 1982. MITI and the Japanese Miracle: The Growth of Industrial Policy. Stanford, California: Stanford University Press.

Jorgensen, Dale. 1995. International Comparisons of Economic Growth. Productivity. Vol. 2. Cambridge: MIT Press.

Kawai, Hiroki. 1995. International comparisons of economic growth. (In Japanese). In Trade Liberalization and Economic Development ed. Shinjiro Urata. Tokyo: Institute for Developing Economies.

Kim, Jong-II and Lawrence J. Lau. 1994. The sources of economic growth of the East Asian newly industrialized countries. Journal of the Japanese and International Economies 8: 235-271.

Kim, Sookon, Jae-Won Kim, and Chonsun Ihm. 1990. Republic of Korea. In Human Resource Policy and Economic Development, pp. 117-188. Manila: Asian Development Bank.

Krueger, Alan B., and Mikael Lindahl. 1999. Education for growth in Sweden and the world. Swedish Economic Policy Review 6(2): 289-339.

Krugman, Paul. 1994.The myth of Asia’s miracle. Foreign Affairs 73(6): 62-78.

Lee, Kiong Hock. 2000. Reforming education for the global information economy. Malaysian Journal of Economic Studies 37(1&2): 53-91.

Lee, Kiong Hock, Sor Tho Ng, and Kok Kheng Yeoh. 1995. Economic policies, human resource development and economic development in Malaysia. Mimeo, Korean Development Institute.

Liang, Chi-Yuan. 1995. Productivity growth in Asian NIEs: a case study of the Repubic of China, 1961-1993. APO Productivity Journal (Winter): 17-40.

Mason, Andrew et al. 1992. The Intergenerational distribution of resources and income in Japan. In The Family, the Market, and the State in Aging Societies, ed. John Ermisch and Naohiro Ogawa. Cambridge: Oxford University Press.

Mason, Andrew. 1997. Will Population Change Sustain the ‘Asian Economic Miracle’? Honolulu: East-West Center Asia Pacific Issues #33.

McKinnon, Ronald. 1973. Money and Capital in Economic Development. Washington, D.C.: Brookings Institution.

Montes, Manuel. 1998. The Currency Crisis in Southeast Asia. Singapore: Institute for Southeast Asian Studies.

Nasim, Anjum and Asya Akhlaque. Human Resource Development and Economic Policy: The Case of Pakistan. Mimeo, Korean Development Institute.

Petri, Peter. 1993. Common foundations of East Asian success. In Lessons of East Asia: A Country Studies Approach ed. D. Leipzinger, Washington, D.C.: World Bank.

Purcell, Randall B. (ed.). 1989. The Newly Industrializing Countries in the World Economy. Boulder: Lynne Rienner Publishers.

Pyo, H.K. 1995. A time-series test of the endogenous growth model with human capital. In Growth Theories in Light of the East Asian Experience ed. T. Ito and A. Krueger. Chicago: University of Chicago Press.

Reynolds, Clark W. 1978. A shift-share analysis of total factor productivity convergence in Mexico. Mimeo, IIASA, Austria.

Rodrik, Dani. 1996. Understanding economic policy reform. Journal of Economic Literature 34(1): 9-41.

Romer, Paul. 1990. Endogenous technological change. Journal of Political Economy 98(2):

Sahn, D.E., P. Dornash and S. Younger. 1996. Exchange rate, fiscal and agricultural policies in Africa: does adjustment hurt the poor? World Development 24(4): 719-747.

Stiglitz, Joseph and Andrew Weiss. 1981. Credit rationing in markets with imperfect information. American Economic Review (June): 393-410.

Sussangkarn, Chalongphob. 1990. Thailand. In Human Resource Policy and Economic Development, pp. 283-353. Manila: Asian Development Bank.

Wade, Robert. 1990. Governing the Market. Princeton, New Jersey: Princeton University Press.

Williamson, J.G. 1969. Dimensions of postwar Philippine economic progress. Quarterly Journal of Economics 83(1): 93-109.

World Bank. 1993. The East Asian Miracle. New York: Oxford University Press.

World Bank. 1998. East Asia: The Road to Recovery. Washington D. C.: The World Bank.

Young, Alwyn. 1995. The tyranny of the numbers: confronting the realities of the East Asian growth experience. Quarterly Journal of Economics 641-680.

Robert K. McCLEERY,*

Monterey Institute of International Studies

Is moderate drinking good for you?

Strategic decision processes: critical review and future directions