Nafziger Economic Development (4th ed)

356Part Three. Factors of Growth

(Lampley et al. 2002:10). Botswana, a democracy with a population of 1.6 million, had the highest rate of GDP per capita growth, 1965–2000. The country, with able presidential leadership, strengthened private property rights, in the interests of economic elites, especially those in the diamond industry (Acemoglu, Johnson, and Robinson 2002). However, the economic success of Botswana is being destroyed by the high incidence of HIV/AIDS, 39 percent among adults aged 15 to 49. In 2002, Botswana’s deaths per 1,000 children under age 5 with AIDS was 107 compared to 31 without AIDS (Lampley et al. 2002:16). In 2010, life expectancy is expected to be 27 years, compared to 74 without AIDS (Chapter 8). The rapid development of an extensive and well-maintained road network, with truckers, miners, construction workers, teachers, and nurses fanning across the country, together with the ignorance about the disease, taboo about acknowledging it, and discrimination against carriers, kindled the epidemic. Work time lost from absences, sicknesses, and deaths in the workforce has devastated the country (ibid., p. 3; Thurow 2002).

AIDS has perhaps caused more suffering and damage to the social fabric in already heavily burdened countries than any pathogen since the bubonic plague of the 14th century. But the devastation of AIDS varies worldwide. Most people living with AIDS in DCs who benefit from chemotherapy and antiretroviral drugs can resume normal life. In the poorest LDCs, with weakened health systems and lack of access to generic antiretroviral drugs, HIV, however, is still a death sentence. In Africa, only 1 percent of those adults with HIV/AIDS have access to lifesaving antiretroviral therapy (WHO 2003:43–46).

To be sure, UNAIDS, U.S. aid (mostly separate from the United Nations in 2004), private initiatives (for example, the Bill and Melinda Gates Foundation and the Bill Clinton Foundations), and the waiving of patent rights to expensive drugs by some Western companies may reduce the cost of AIDS treatment in poor countries. However, even if DCs and their companies permit these countries to buy cheaper generic drugs, the lack of an effective health delivery system in many may prevent widespread effective therapy. The cost and complexity of AZT (Azidothymidine) and other therapies limit their uses in poor countries (Lampley et al. 2002:23).

Whereas prevention is the highest priority, improving HIV treatment reduces the stigma and increases the incentive for people to seek counseling and testing. Preventive measures, such as education on safer sex, promotion of condom use, prevention and treatment of sexually transmitted diseases, and reduction of blood-borne transmission, are cost-effective, especially if targeted at people at particularly high risk of acquiring and transmitting HIV infection, such as sex workers, migrants, the military, truck drivers, and drug users who share needles (ibid., p. 27; World Bank 1993i:20, 100–105).16

16Dyson (2003:427–442) finds that, in sub-Saharan Africa, both the prevalence of HIV infections and the economies of concentration in combating infection are greater in urban areas. Gersovitz and Hammer (2004:1–27), assuming a ready-made dynamic model of disease transmission from epidemiology, examine optimal implementation of prevention and therapy of infectious diseases.

LDCs need integrated AIDS prevention and care, including correct and culturally appropriate information and existing prevention tools. Brazil has mandated universal and free access to HIV care, including testing, counseling, and distribution of generic, antiretroviral drugs, which stopped the epidemic from spreading further. Uganda reduced prevalence by an “ABC campaign” of abstinence, being faithful, and condom use, and empowering women to negotiate safer sexual patterns (WHO 2003:47–50).

Other world regions are not exempt. The World Bank Development News (“The Plague We Can’t Escape,” March 17, 2003) reports the concern by the U.S. government that, left unchecked, a second wave of HIV/AIDS will devastate the nations of Russia, China, India, Nigeria, and Ethiopia, so that by 2010, their combined numbers will dwarf those of the rest of sub-Saharan Africa. The governments of Nigeria and Ethiopia doubt that their health systems can cope with the epidemic without substantial debt writedowns or foreign aid. Failure to check AIDS growth could mean the collapse of social and political institutions in the five countries.17

Conclusion

1.Since the 19th century, the wage premium for skilled relative to unskilled labor has increased.

2.Despite destruction of physical capital during World War II, the economies of Germany and Japan grew rapidly in the postwar period because their labor forces – with high degrees of skill, experience, education, health, and discipline – remained intact.

3.Investment in human capital includes expenditures on education, training, research, and health, enhancing a people’s future productivity.

4.Economists who analyze the relative rates of returns to investment to primary education and secondary education in LDCs disagree on whether LDCs should put greater priority on primary education. Psacharopoulos and Woodhall, who find that the higher average returns are from primary education, argue for more emphasis on primary education. Knight, Sabot, and Hovey, however, question this emphasis in a study that shows that the marginal rates of returns to the cohort entering into the labor market were lower for primary education.

5.Although employers sometimes use secondary and university education as a screening device, they discover and pay wage premiums for literacy and numeracy, even in manual work.

6.Public expenditure per student for higher education in LDCs is about 10 times as high as for primary education.

7.In LDCs the expansion of primary education redistributes benefits from the rich to the poor, whereas the growth of secondary and higher education redistributes

17 New diseases can spread quickly in an interdependent world with a highly mobile population. In 2003, after early fear that SARS (severe acute respiratory syndrome) would spin out of control, the World Health Organization (2003) helped coordinate an effort for prompt detection and reporting, timely global alerts, travel recommendations and screening measures at airports, and collaboration by scientists and clinicians that may have provided lessons for dealing with future pandemics.

358 Part Three. Factors of Growth

income from the poor to the rich. In light of this pattern, LDCs may want to charge their richer citizens for the full cost of secondary and higher education.

8.In most LDCs, boys are sent to school far more often than girls. Yet a number of studies indicate that educating girls has a high payoff in improving nutrition, reducing fertility and child mortality, and increasing labor force productivity.

9.One planning method for producing specialized skills is to use input–output relationships to determine future demand for various types of high-level personnel. However, assuming a fixed input–output relationship does not recognize the possibility of substituting one skill for another or adapting output to skills array.

10.If wages are adjusted more closely to productivity, LDC educational planning is easier.

11.On-the-job training tends to balance demand for, and supply of, training. In addition extension agents and training at short-term vocational or technical institutions can help people improve their skills without appreciably disrupting production.

12.Distance learning through electronic media can dramatically reduce the cost of continuing education and secondary and higher education, including teacher training.

13.Some economists argue that LDCs do not lose from the brain drain, as the worker earns an income equal to his or her marginal product. However, we can question this analysis, because marginal product may exceed the wage, high-level skills increase the productivity of other production factors, and government highly subsidizes education in developing countries.

14.There is no evidence of a backward bending supply curve for labor unique to LDCs. Furthermore, the aggregate supply curve of labor in LDCs is clearly upward-sloping.

15.Although affluent Asians may be more likely to consider manual work degrading than affluent Westerners, these attitudes appear to be primarily related to the more abundant supply of cheap labor in Asia.

16.Poor nutrition and health reduce labor productivity. However, health has improved, and nutrition has probably not deteriorated in LDCs since the 1960s.

17.HIV infection and AIDS-related deaths have had a substantial adverse impact on economic growth in some developing countries, especially in Africa.

QUESTIONS TO DISCUSS

1.What impact has the increased premium on skilled labor had on attitudes toward and the relative wage for physical work in DCs?

2.Would you expect the returns to a dollar of investment in education to vary from those in industrial plant, machinery, and equipment? Would noneconomic educational benefits affect the decision under perfect competition to equalize the expected marginal rate of return per dollar in each investment?

3.Does empirical evidence on the rates of return to education show that LDCs should put a priority on primary education relative to secondary and university education? Would you expect the relative returns to primary education, on the one hand, and secondary and postsecondary education, on the other hand, to change as economic development takes place?

4.To what extent is education a screening device for jobs rather than a way of increasing productivity in LDCs? How might the importance of screening and enhancing productivity vary by educational and skill level, sector, or world region in the developing world?

5.How does LDC government investment in educational expansion affect income distribution?

6.What are some of the ways that an LDC can increase its rate of returns to investment in secondary and higher education?

7.What advice would you give to the top official in the Department of Education in an LDC who is designing a long-run program of education, training, and extension in his country? (You may either focus on LDCs in general, a particular LDC world region, or a particular LDC.)

8.How might government wage policies contribute to unemployment and underutilization of labor among the educated?

9.How can LDCs reduce the brain drain?

10.How do you explain cross-national differences in labor productivity?

11.Why does Africa have such a high incidence of HIV/AIDS?

12.What investments can an LDC government make to increase the health and reduce the incidence of disease in the LDC? Examine these investments in the light of the opportunity cost for other investments, including those in social capital.

GUIDE TO READINGS

The World Bank (2004i) examines the relationship between health and educational services, on the one hand, and economic development, on the other (http://econ.worldbank.org/wdr/wdr2004/text-30023/).

For health and education data, see World Bank (2003c) and U.N. Development Program (2003).

See World Health Organization’s World Health Report (2003) on health and mortality, including AIDS, and Lampley, Wigley, Carr, and Collymore (2002) and UNAIDS (2004) on the HIV/AIDS pandemic. Arrow (2004:20–21) examines the challenge of stopping or slowing the spread of drug-resistant strains of malaria. Deaton (2003a:113–158) finds that poverty but not inequality reduces life expectancy.

Kremer (2002:67–90) analyzes small pharmaceutical markets in LDCs and their implication for market failure in providing drugs for widespread LDC diseases such

360Part Three. Factors of Growth

as malaria, diarrhoeal diseases, and hookworm. Because pharmaceutical R&D is a global public good, each country has an incentive to free ride on research financed by other countries.

Schultz (1963), Schultz (1961:1–17), and Becker (1975) are major works on investment in human capital. Behrman and Rosenzweig (1994) assess LDC data on education and the labor force. Boissiere,` Knight, and Sabot (1985:1016–1030); Knight and Sabot (1990); Knight, Sabot, and Hovey (1992:192–205); Psacharopoulos and Woodhall (1985); and Psacharopoulos (1985, 1994) analyze returns to education. Harmon, Oosterbeek, and Walker (2003:147–200) survey studies of the microeconomic returns to education, showing that investment in education has an unambiguously positive effect on individual earnings. How large the aggregate effect is depends on whether you use neoclassical or new growth theories. Glewwe (2002:436–482) analyzes how educational policies affect educational outcomes and skills in LDCs. Shultz (1988) analyzes economic returns to education. Strauss and Thomas (1995) discuss human resources: modeling of household decisions. Jiminez (1995) looks at human and physical infrastructure investment decisions.

Dasgupta (1993) focuses on the relationship among nutrition, energy requirements, and worker productivity, with the implications of malnutrition for incentives, institutional reform, and income redistribution. He uses the findings of nutritionists on the link between food needs and work capacity to reconstruct modern resource allocation theory. Mahler (1980:66–77); Berg (1987); Scrimshaw and Taylor (1980:78– 88) analyze nutrition, health, and economic development. Weisbrod and Helminiak (1977:505–522) measure the impact of parasitic diseases on labor productivity.

See World Bank Group in Africa (2000) on electronic media in LDC education. Nair (1969:453–456) presents a skillful argument against Myrdal’s discussion

(1968:Vol. 2) of Asian attitudes toward manual work.

The effect of childhood environment on the creativity and self-reliance of the labor force is discussed by McClelland (1961) and Hagen (1962). See also Chapter 12.

362 Part Three. Factors of Growth

Scope of the Chapter

We analyze the relationships among capital, technology, and economic development. Nafziger’s supplement (2006b) discusses ways to increase the rate of capital formation, including free market inducement for household and business savings, capital imports (see also Chapter 15), exploiting idle resources (often overrated), increasing absorptive capacity (the ability to profitably utilize additional capital), moral suasion (development as if sacrificing for war), improving the tax system, emphasizing direct taxes, taxes on luxuries, developing financial intermediaries, increasing investment opportunities, redistributing income, local financing of social investment, and inflationary financing. The first part of this chapter summarizes studies that review how capital formation and technical progress contribute to economic growth and indicates why the contribution of the two differs in DCs and LDCs. Second, we identify the composition of the growth in total factor productivity, usually labeled technical progress – the residual factor in growth, the increased worker productivity arising from factors other than increases in capital per worker-hour. Third, we consider technical change as a prolonged process of learning by doing. The fourth section asks whether growth can be attributed only to increases in measured inputs. Fifth, we define research, development, and invention, and their relationship to technical change. Sixth, we examine the impact of information and communications technology (ICT) on economic growth, and how promising ICT is for LDCs.

The seventh section looks at the investment criterion, maximum labor absorption, and some of its inadequacies. Next, we discuss social benefit–cost analysis, the planner’s most comprehensive gauge for investment choice. We raise questions about the appropriate discount rate and how to treat risk and uncertainty. Subsequently, we indicate how private profitability must be adjusted for externalities, income distribution, indivisibilities, monopoly, savings effects, and factor price distortions to obtain social profitability. Our final section analyzes and assesses shadow prices, a way of adjusting prices more closely to social costs and benefits.

Capital Formation and Technical Progress as Sources of Growth

In the 1950s, U.N. economists considered capital shortage the major limitation to LDC economic growth. By capital, they meant tools, machinery, plant, equipment, inventory stocks, and so on, but not human capital.

On the basis of 19thand 20th-century Western growth, however, the British economic historian Sir Alex Cairncross, writing in 1955, questioned whether capital’s role was central to economic growth. To be sure, he agreed with U.N. economists that capital and income grow at about the same rate. But he felt that capital increases do not explain economic growth – that, in fact, the reverse was true: The amount of capital responds to increases in its demand, which depends on economic growth (Cairncross 1955:235–248).

Since 1955, econometricians have tried to resolve this controversy with studies measuring how factor growth affects output growth. The studies’ primary concern has been to determine the relative importance of the two major sources of economic growth – capital formation and technical progress.

Initial attempts at statistical measurement in the West and Japan in the late 1950s and 1960s indicated that capital per worker-hour explained 5–33 percent of growth in output per worker-hour. Scholars usually attributed the residual, 67–95 percent, to technical progress (Abramovitz 1956:5–23; Solow 1957:312–320; and Bruton 1967:1099–1116). A development economist used this evidence to argue that capital formation has been stressed too much and technical progress too little (Hagen 1980:201–203).

Studies of LDCs, which gain substantially from imitation or modification of DC technology, contradict findings based on DC data. These studies indicate that the contribution of capital per worker to growth, even among fast-growing East Asian NICs, is 50–90 percent, whereas that of the residual is only 10–50 percent (Maddison 1970; Robinson 1971:391–408; Young 1995:641–680; see also Jorgenson 1995; Chenery, Robinson, and Syrquin 1986).1

For command economies Russia–Soviet Union, pre-1989 Eastern Europe, and pre1976 China, the residual is even smaller than for the third-world countries of Asia, Africa, and Latin America. Virtually all growth in these command economies was attributed to increases in capital and other inputs, and only a tiny fraction to technical innovation, a combination that contributed to the Soviet collapse in 1991 (see Chapter 19).

The aggregate models used in studies of the sources of economic growth in developed and developing countries are rough tools. Yet these studies point in the same general direction. First, the major source of growth per worker in developing countries is capital per worker; increased productivity of each unit of capital per worker is of less significance. Second, the major source of growth per worker in developed countries is increased productivity, with increases in capital per worker being relatively unimportant. Accordingly, capital accumulation appears to have been more important and technical progress less important as a source of growth in developing countries than in developed countries.

In 1965, the Nobel prize winner John R. Hicks argued that econometric studies of growth sources in Western countries understate capital formation’s contribution to growth. Because many significant advances in knowledge are embodied in new capital, its separation from technical progress may lead to underestimating its contribution. Furthermore, accumulation of new capital is frequently offset by a decrease

1 However, Devarajan, Easterly, and Pack (2003:547–572) find that Africa’s slow growth, 1960–94, does not result from low investment rates. Indeed, output per worker fell, whereas capital per worker increased, indicating a reduction in TFP. The major culprits in this reduction may be decreasing capacity utilization (bottlenecks from factories experiencing shortage of imported inputs and the fact that many factories operate with one shift) and the reallocation of labor to lower productivity sectors in manufacturing.

364Part Three. Factors of Growth

in value in old capital, partly from obsolescence. Thus, Hicks contended, it is very wrong to give the impression to a LDC, having relatively small amounts of old capital, that capital accumulation is a matter of minor importance.2 Econometric studies of LDCs done since 1965 seem to confirm Hicks’s point. The rates of capital growth in developing countries (as well as Israel, which received substantial inflows of funds in the 1950s) were rapid enough to offset some of the understatement of capital in the production function. Developing countries concerned about rapid economic growth ignore capital accumulation at their peril.

Indeed, the World Bank’s (2004f:44) decomposition of GDP growth indicates that in LDCs, capital contributed more than productivity to GDP growth, 1990–2000. However, largely as a result of productivity growth from greater trade, productivity is expected to contribute more to GDP growth in LDCs than will capital, 2005–15 (Figure 11-1). Growth and productivity increases in East Asia and the Pacific are expected to remain strong through 2015. Keller (2004:752–782) indicates the effect of international R&D spillovers, foreign direct investment (especially by multinational corporations), and learning through exports and imports on productivity growth (see Chapter 17 on how increased globalization and international trade contribute to increased productivity growth).

Components of the Residual

Studies of Western growth find that the residual is a major contributor to economic growth. However, to label this residual technical knowledge without explaining it is to neglect a major cause of economic growth. Critics in the early 1960s objected to elevating a statistical residual to the engine of growth, thus converting ignorance into knowledge (Balogh and Streeten 1963:99–107). Accordingly, in recent years some economists have labeled this residual total factor productivity (TFP) rather than technical progress.

What does this residual include? Edward F. Denison (1967) studied the contribution that 23 separate sources made to growth rates in 9 Western countries for the period from 1950 to 1962. His estimates, particularly of labor quality, are based on reasonable and clearly stated judgments rather than on econometric exercises. Thus, he assumes that three-fifths of the earnings differentials among workers of the same age, geographical area, and family economic background are the result of education. Moreover, individuals in the labor force having the same years of school and hours in school per year obtained at the same age are considered to have an equivalent education, no matter when or where they received their education. Furthermore, 70 percent of the decrease in average hours worked is assumed to be offset by productivity increases. The reader aware of Denison’s assumptions and approximations can benefit from his work; he is careful in handling empirical data, stating assumptions, and in investigating how results may be sensitive to variations in assumptions.

2In contrast, Denison (1988:39) would remove all contributions of advances in knowledge from capital accumulation and attribute these improvements to technical progress.

FIGURE 11-1. Productivity Will Contribute More to GDP Growth through 2015 Than Will Capital or Labor. Source: World Bank 2004f:44.

In all DCs, growth in national income per worker is attributed more to increases in output per unit of input (the residual factor) than to increases in inputs, labor, capital, and land. Denison indicates that sources for increased output per worker for the United States or Northwestern Europe include advances in knowledge, economies of scale, improved allocation of resources, reduction in the age of capital, and decreases in the time lag in applying knowledge. Other empirical studies have included organizational improvements, increased education and training, learning by experience, and (linked to it) product variety (D. Addison 2003:4).

Western and Japanese total factor productivity growth slowed down after 1973. Martin Neil Baily and Robert J. Gordon (1986:347–420) show that 10 percent of the slowdown in annual productivity growth in the United States from 1948–73 to 1973–87 results from the understatement in productivity gains in the manufacture of