. 14
( 21)


Land reform in Mexico
As a result of the Mexican revolution (1910“17), which included widespread peasant revolts,
particularly in land areas where plantation-style estates abounded, land reform was a fait
accompli of the armed struggle. After the revolution, successive governments sought to
complete the land reform, essentially breaking up tillable holdings in excess of 200 hectares.
By 1976, 43 percent of land had been turned over to ejidos, which are village councils respon-
sible for distributing land to their members. Ejidos were prohibited from renting, selling, or
mortgaging the land. Unfortunately, the 28,000 ejidos, which provided land to 43 percent
Agriculture and development 383
of all the farming families in Mexico, held only 16 percent of the irrigated land. The large
landholders, a mere 2.5 percent of all landholders, were able to produce 40 percent of the
food on 20 percent of the land. They were able to do so because, despite land redistribution,
they had managed to hold on to and control the best land. They were able to finance irrigation
projects themselves or, in the more likely instance, to benefit from government-created irriga-
tion projects specifically aimed at large farms. And the large landholders had a near monopoly
on credit. For example, between 1956 and 1969, the private commercial farmers received 85
percent of all agricultural credits granted by financial intermediaries (de Janvry 1981: 215).
At first it appeared that the ejido system was a social, political, and economic success.
From 1938 to 1951, agricultural output leaped ahead at an annual rate of 4.3 percent. From
1951 to 1970, agricultural output growth exceeded 6 percent per annum. Then, however,
Mexican agricultural growth virtually stopped. From 1970 to 1976, agricultural output per
capita fell by more than 15 percent (de Janvry 1981: 217; Cypher 1990: 90). Grain imports
soared; between 1970 and 1979, they totaled 689,000 tons, and in the 1980“9 period, they
rose to 26 million tons (Russell 1994: 194). By 1996, 6 million tons of corn, more than
30 percent of national consumption, was imported from the US.
The great failure of small-scale Mexican agriculture following land reform was the result
of several factors. In the 1970s, when greater emphasis should have been placed on agricul-
ture because of the tightening of the “scissors” between land yields and population demands,
the Mexican state became increasingly involved in industrial development, neglecting agri-
culture. Second, with the onset of the oil boom (1976“82), the Mexican government took the
position that it would be more efficient to export oil, which commanded a high price at the
time, and import food.
When the oil boom collapsed, however, and the debt crisis ensued, Mexico adopted
neoliberal policies which reduced the size of governmental investments, particularly in the
agrarian sector, and drastically reduced the subsidized credits which had been allocated to
the smallholders, or ejidatarios. Subsidies which had been granted on fertilizers were virtu-
ally eliminated, and electricity prices were increased by 60 percent under the “get prices
right” or “real prices” doctrine of neoliberalism. In order to trim its budget, the government
lowered the buying price of corn and other staples, further squeezing the ejidatarios. Not
surprisingly, agricultural growth fell well below the rate of growth in population.
In 1992, Mexico instituted sweeping changes in agriculture, essentially allowing ejidatarios
to sell and rent their land and to use it as collateral for credit, while allowing corporations,
both domestic and foreign, to buy such properties. In essence, by the 1990s, the Mexican
effort at land reform had ended. Critics charge that these new trends will lead to a renewal of
land concentration, expelling as many as 10 million rural residents (largely drawn from the 2.8
million former ejido cultivators and their families) from the countryside into Mexico™s huge
urban areas, or into international migration and ending the ejido system.

Korea™s Saemaul Undong
Prior to Japanese colonial rule in the early twentieth century, the rural landholding nobility
in Korea, the yangban, had held both the land and the peasants in a vice-like grip for over
500 years. By the late nineteenth century, much of the Korean countryside had been swept
with unsuccessful peasant revolts and risings. Japanese rule brought some limitations on
the yangban, as some Japanese adopted Korean landholdings. For the peasants, however,
conditions and land concentration, as described in the following quotation, generally became
worse under colonialism.
384 The Process of Economic Development
In 1914, only 1.8 per cent of the households owned 51 per cent of the cultivated land.
Rents ranged from 50 to 60 per cent of the crop with tenants, who bore the costs of
production, left with 20 per cent or less of the final production. As in many Asian
countries in the colonial period, the change from sharecropping to fixed rents in rural
Korea meant in bad years there was no relief from starvation. Contracts were verbal and
could easily be manipulated or terminated by landlords and protestors faced possible
imprisonment by the colonial state.
(Douglas 1983: 192“3)

In 1953, in the aftermath of the Korean war, which had left 10 million Koreans homeless
in the devastated cities, a thorough-going land reform was instituted. Compensation paid to
the yangban was minimal, 150 percent of the value of the annual harvest, with full payment
spread over several years, an amount that was insufficient to compensate for the capitalized
value of the land. Land was then distributed to the peasants, with an upper limit of 3 hectares.
Once instituted, the land ownership pattern in South Korea has remained stable.
In the early years, Korean land reform had the appearance of a “title switching” program;
peasants were forced to sell their surplus staples to the government, which redistributed them
to the cities, with the price paid being so low that costs of production could not be covered.
At first, Korean farmers were forced to compete with food-aid imports which came into
Korea virtually free from the United States. Such difficult conditions in the countryside led to
a massive out-migration of farmers. Such a demographic shift, however, was largely accom-
plished without an expanding underclass, as the South Korean industrialization program, and
its emphasis on maintaining labor-intensive production via export substitution (see Chapter
10), helped to absorb the inflow of former agricultural workers. Still, the relative neglect of
agriculture, and the migration it fostered, was telling; in 1969, 29 percent of the dwellings in
Seoul were classified as slum/squatter dwellings, where many former peasants resided.

The neglect of agriculture created a poverty syndrome which had several dimensions.
Low rural incomes meant ineffective demand for agricultural inputs ¦ evidence for this
period shows widening disparities between rural and urban incomes there is evidence to
suggest that rural welfare, although not at the level of desperation of the 1950s, was not
advancing. Real rural incomes stayed nearly the same for the decade (1960“70), while
urban incomes doubled.
(Douglas 1983: 190“1)

Fortunately, government planners recognized the critical conditions in the agricultural
sector in the course of the 1960s, and reforms were introduced. For example, interest rates
above 20 percent per year on crop loans were declared illegal, and there was a general
expansion of irrigation facilities. In 1972, with the third five-year plan, the government
turned to an integrated strategy of agricultural development, the Saemaul Undong, which
has successfully moved land reform from title switching to a genuine program of rural
development. The strategies employed in South Korea contrast sharply with the growing
neglect of small-scale producers in Mexico. The third five-year plan dealt with the slow
growth in agricultural output “ only a 2.3 percent increase per year in 1967“71 “ via four
integrated strategies:

• the general diffusion of high yield varieties of seeds, the domestic production of fertilizers,
and greater application of pesticides;
Agriculture and development 385
• the mechanization of agriculture;
• state management of grain storage facilities; and
• a program of housing construction, rural electrification and feeder road construction.

During this period, the government allocated an extraordinary 28 percent of its budget to
agricultural development! As a consequence, the differential between rural and urban living
standards was greatly diminished. By the late 1970s, South Korea™s program of agrarian
development had been consolidated. In real terms, agricultural output increased by over
500 percent between 1970 and 1991. In 2004 Korea had a relatively large population of 48
million and a relatively modest population growth rate of 0.8 percent. With but limited land
for cultivation “ Mexico, for example, has nearly ten times the amount of land per capita “
Korea has continued to import grains. But with its tremendous strength in manufacturing
exports, Korea can afford to sustain heavy grain imports as it creates true dynamic compara-
tive advantage in production with greater value-added, while paying high wages.
While Korea™s experience with land reform and agricultural development has, of neces-
sity, taken place within certain physical limitations, and while it has not eliminated the need
to import cereals, it does demonstrate that a productive relationship between the state and
the rural population can be achieved, within the context of a strong program of land reform.
In Korea™s case, the successes achieved in agriculture have always depended upon state
intervention and a successful program of state-directed industrialization. Neoliberalism has
played virtually no role in South Korea™s success, nor has the program succeeded as a result
of an “agriculture first” or an agriculture-led policy. Instead there has evolved a balanced
policy of both agrarian and industrial development.

To the extent that export-led industrial growth has made rural development possible, the
use of such terms as “self reliant” to describe Korea™s rural development is extremely
misleading. The emphasis on private ownership has, in general, meant that increases in
production have been generated by state intervention into the market rather than through
an increase in a local corporate capacity to develop. It also suggests that the future of
Korea™s agricultural-cum-rural development will depend upon the ability to keep the
export engine of growth in high gear.
(Douglas 1983: 208)

Questions and exercises
1 In discussing appropriate technology for less-developed nations, we noted that given
the relative labor abundance and relatively lower wages in the less-developed nations
compared to the more developed economies, and the relatively higher price of capital, in
many situations the optimal combination of labor and capital that should be used in the
less-developed nations would be more labor-using and less capital-using compared to a
higher-wage developed country.
a In a graph with the quantity of capital, K, measured along one axis and the quantity of
labor, L, measured along the other, draw one convex-to-the-origin production isoquant,
representing, say, 1,000 units of output for every combination of L and K on the curve.
Also draw in an isocost line for a developed country which is tangent to the isoquant
at some point. (You will remember that the slope of the isocost line is determined by
the relative prices of K and L.) Note the quantity of K and L used on the axes.
386 The Process of Economic Development
Now, assuming that the price of a unit of K in the less-developed country is the same
as in the more developed, but the wage rate is lower, show that the optimal combina-
tion of K and L to produce 1,000 units of output in the less-developed nation would use
more labor and less capital than in the more developed economy. Prove, too, that the
less-developed country would be the lower cost producer of the 1,000 units.
b What difference does it make to the way you draw the isoquant if, now, we assume
that not all combinations of labor and capital are technologically feasible to produce
1,000 units of output? What will the isoquant look like if there are only two different
combinations of K and L available, for example? What choice of technique (i.e.
which combination of K and L) will the developed country and the less-developed
country producer select?
2 Advocates of the Green Revolution have argued that the technologies employed are
“scale neutral.” That is, seeds and fertilizers are easily divisible, and no appreciable
change in unit costs is involved in altering the quantities used. Therefore, they hypothe-
sized, the Green Revolution should benefit both poor and wealthy farmers alike, without
an appreciable relative advantage to one or the other in use. Contrast this technical view
of the impact of the Green Revolution with a more “institutional” view.
a Assuming that the unit cost of inputs are nearly “scale neutral,” why did the many
institutionalists predict that the Green Revolution would increase intraregional and
interregional income disparities rather than diminish them?
b What other considerations are there besides costs of the new inputs in deciding
whether to use a new seed, a new fertilizer, or any new technology in agriculture?
c Can the response of poorer and richer farmers differ? Why?
3 Available research suggests that when peasant farmers are impacted by deforestation
and desertification, women are particularly affected. Why is this so? How are women
affected adversely?
4 Some peasants are quite “risk-averse,” for reasons discussed in this chapter. Imagine a
group of poor pastoralists struggling with the effects of a famine and drought.
a Why might it be “rational” for them to actively, if unintentionally, contribute to the
acceleration of environmental degradation, and to their own famine, via overuse of
grazing land, if they are concerned with guarding against the exhaustion of their
animal herds? Is this situation an example of a market failure?
b What could be done to prevent overgrazing of land?
c Are these people poor because they overgraze the land with the animals, or are they
overgrazing the land because they are poor?
a Explain, using supply and demand curves, how a low “buy price” (i.e. a price below
market equilibrium) by government for a staple product, such as rice in India, may
lead profit-oriented staple crop producers to switch to other crops.
b On the same graph, show the effect of targeted income subsidies to low income
consumers on the quantity and price of rice traded in the market, assuming the “buy
price” program is abandoned.
c Discuss the pros and cons of subsidized prices for staple commodities versus
targeting income subsidies as strategies to help the poor to purchase staple food
Agriculture and development 387
a How is true land reform different from a redistribution of landholdings? What
political purposes might each have?
b What are the economic reasons for pursuing land reform? For a redistribution of
landholdings that falls short of full land reform?
7 How important are improvements in the productivity of the agricultural sector relative
to efforts to increase productivity in industry? Can a country become developed without
an industrial and agricultural “revolution” in the economic sense of the term?
a Are food imports necessarily an indication of weakness in the agricultural sector
of an economy? Under what conditions might food imports, rather than domestic
production, be desirable and economically rational?
b Under what conditions would rising food imports indicate a weakness in the overall
economic strategy? Do countries have to produce everything they consume?
c Looking back at Table 11.1, for which countries would you guess that falling food
production per person is an indication of problems and for which might such a
result not be a problem?

1 A vicious circle has overtaken Sub-Saharan Africa, where a combination of public health issues
(malaria and AIDS, in particular), water scarcity and lack of water infrastructure, inability to irri-
gate, soil compaction (sometimes from overgrazing), a decline in public sector resources devoted
to agricultural extension and R&D, desertification, political instability, and sometimes warfare and
rising input prices, especially for fertilizers, has stalled and nearly stopped forward momentum
in agricultural production. With a prevalence of subsistence farmers and very poor agricultural
laborers, the rural population cannot avail themselves of the plentiful food stocks available around
the world. The availability of foreign aid, analyzed in Chapter 17, is crucial. For further discussion
see Sachs (2005), especially chapters 3, 10, and 13.
2 Note, however, that declining per capita food production is not the same as declining per capita
food consumption. For some newly industrialized nations, agricultural and food production should,
with successful structural transformation, fall in relative terms compared to industrial and service
production. The poorer a country, however, and the lower its level of human development, the less
likely it is that the structural transformation has taken place.
3 Note that this term is intended to signify extreme dependence on a very limited range of
primary product exports, be they agriculture or raw material exports. It is not strictly intended
to mean that an economy is literally dependent on only one export for its foreign exchange
4 It perhaps goes without saying that, though this phenomenon was only “discovered” at that time,
this does not mean that the effects of commodity price swings had not been in operation for some
time. This instability of prices, and the macroeconomic consequences, had been a focus of much of
the critical concern of heterodox economists, as discussed in Chapter 6, and of policy-makers in the
less-developed world who had to periodically confront such crises.
5 How can a country prevent a temporary and exogenous export price increase from having such
adverse effects? One way is to sterilize the increased inflow of export revenues, to prevent an
increase in the currency™s exchange rate value vis-à-vis other currencies. This can be done if the
central bank of the country sells more of its own currency, thus buying up foreign exchange. This
will increase the supply of its own currency, to balance the increased demand for that currency by
non-residents resulting from the higher prices of the export. Effective sterilization also requires that
government, with the increased revenues it earns and increased foreign exchange reserves, exercise
restraint in spending these “savings.”
388 The Process of Economic Development
6 Technically, the expected losses from the gamble failing exceed the expected gains from the gamble
succeeding, where in this case, the gamble being considered is a change in peasant production
(such as a new technology, or a new seed, or a new fallowing technique) where the probabilities of
success and failure are still subject to some degree of uncertainty.

Adler, Emanuel. 1987. The Power of Ideology: The Quest for Technological Autonomy in Argentina
and Brazil. Berkeley: University of California Press.
Alston, Julian and Phillip Pardy. 2006. “Developing Country Perspectives on Agricultural R&D,” pp.
11“29 in Philip Pardy, Julian Alston and Rogley Piggott (eds.), Agricultural R&D in the Developing
World. Washington, D.C.: IFPRI.
Baker, Kathleen and Sarah Jewitt, 2007. “Evaluating 35 Years of Green Revolution Technology in
Indian Villages of Bulandshahr District, Western UP, North India,” Journal of Development Studies
43 (February): 312“29.
Barraclough, Solon (ed.). 1973. Agrarian Structure in Latin America. Lexington, MA: Lexington Books.
Barry, Tom (ed.). 1992. Mexico: A Country Guide. Albuquerque, NM: Inter-Hemispheric Resource Center.
Bliss, C.J. and N.H. Stern. 1982. Palanpur: The Economy of an Indian Village. Oxford: Clarendon Press.
Borras, Saturnino. 2003. “Questioning Market-Led Agrarian Reform: Experiences from Brazil,
Colombia and South Africa,” Journal of Agrarian Change 3 (July): 367“94.
””. 2005. “Can Redistributive Reform Be Achieved via Market-Based Voluntary Land Transfer
Schemes?,” Journal of Development Studies 41: (January) 2005: 90“134.
Bouis, Howarth. 2007. “The Potential of Genetically Modified Food Crops to Improve Human Nutri-
tion in Developing Countries,” Journal of Development Studies 43 (January): 79“96.
Byerlee, Derek. 1992. “The Political Economy of Third World Food Imports: The Case of Wheat,” pp.
305“26 in Charles Wilber and Kenneth Jameson (eds.), The Political Economy of Development and
Underdevelopment, 5th edn. New York: Random House.
Byres, Terrance. 2006. “Agricultural Development: Towards a Critique of the ˜New Neoclassical
Development Economics™ and of ˜Neoclassical Neo-Populism™,” pp. 222“48 in Jomo K.S. and Ben
Fine (eds.), The New Development Economics. London: Zed Books.
Cypher, James M. 1990. State and Capital in Mexico. London and Boulder, CO: Westview.
de Janvry, Alain. 1981. The Agrarian Question in Latin America. Baltimore, MD: Johns Hopkins Press.
Deininger, Klaus. 1999. “Making Negotiated Land Reform Work: Initial Experience from Colombia,
Brazil and South Africa,” World Development 27: 651“72.
Dixon, Chris. 1990. Rural Development in the Third World. London: Routledge.
Doolette, John and James Smyle. 1990. “Soil and Moisture Conservation Technologies,” in John
Doolette and James Smyle (eds.), Watershed Development in Asia. Washington, D.C.: World Bank
Technical Paper 127.
Douglas, Mike. 1983. “The Korean Saemaul Undong,” pp. 186“214 in David Lea and D.P. Chaudhri
(eds.), Rural Development and the State. London: Methuen.
Eicher, Carl and John Staatz. 1998. “Agricultural Ideas in Historical Perspective,” pp. 8“38 in Carl
Eicher and John Staatz (eds.) International Agricultural Development, 3rd edn. Baltimore MD:
Johns Hopkins Press.
FAO (United Nations Food and Agricultural Organization). 2006. FAONewsroom. http://www.fao.org/
””. 2005. “The Role of Transnational Corporations,” http://www.fao.org/docrep/005/4671e/.
Fogel, Robert. 1994. “Economic Growth, Population Growth and Physiology,” American Economic
Review 84: 369“95.
Ghosh, Jayati. 2004. “Income Inequality in India,” IDEA. Online at http://www.networkideas.org/
Goldman, Abe and Joyotee Smith. 1995. “Agricultural Transformations in India and Northern Nigeria:
Exploring the Nature of Green Revolutions,” World Development 23 (March): 243“63.
Agriculture and development 389
Graff, Gregory, David Roland-Holst and David Zilberman. 2006. “Agricultural Biotechnology and
Poverty Reduction in Low-Income Countries,” World Development 34 (August): 1430“45.
Griffin, Keith. 1974. The Political Economy of Agrarian Change. Cambridge, MA: Harvard University
””. 1989. Alternative Strategies for Economic Development. New York: St Martin™s Press.
Griffin, Keith, Azizur Khan and Amy Ickowitz. 2002. “Poverty and the Distribution of Land,” Journal
of Agrarian Change 2 (July): 279“330.
””. 2004. “In Defense of NeoClassical Populism,” Journal of Agrarian Change 4 (July): 361“86.
Hill, Polly. 1986. Development Economics on Trial. Cambridge: Cambridge University Press.
Jank, Marcos Sawaya, Maristela Franco, Andre Meloni and Paulo Faveret. 1999. “Concentration and
Internationalization of Brazilian Agribusiness Exporters,” International Food and Agribusiness
Management Review 2 (3/4): 359“74.
Kay, Cristóbal. 2006. “Rural Poverty and Development Strategies in Latin America,” Journal of
Agrarian Change 6 (October): 455“508.
Lea, David and D.P. Chaudhri (eds.). 1983. Rural Development and the State. London: Methuen.
Lipton, Michael. 1977. Why Poor People Stay Poor. Cambridge: Cambridge University Press.
””. 2007. “Plant Breeding and Poverty: Can Transgenic Seeds Replicate the ˜Green Revolution™ as a
Source of Gains for the Poor?,” Journal of Development Studies 43 (January): 31“62.
”” and Richard Longhurst. 1989. New Seeds and Poor People. Baltimore, MD: Johns Hopkins
Luce, edward. 2007. In Spite of the Gods: The Strange Rise of Modern India. New York:
Mellor, John. 1998a. “Agriculture on the Road to Industrialization,” pp. 136“54 in Carl Eicher and
John Staatz (eds.) International Agricultural Development, 3rd edn. Baltimore MD: Johns Hopkins
””. 1998b. “Foreign Aid and Agriculture-Led Development,” pp. 55“66 in Carl Eicher and John
Staatz (eds.) International Agricultural Development, 3rd edn. Baltimore MD: Johns Hopkins
McCalla, Alex. 1998. “Agriculture and Food Needs to 2025,” pp. 39“54 in Carl Eicher and John Staatz
(eds.) International Agricultural Development, 3rd edn. Baltimore MD: Johns Hopkins Press.
Morris, Michael and Derek Byerlee. 1998. “Maintaining Productivity in Post-Green Revolution Asian
Agriculture,” pp. 458“73 in Carl Eicher and John Staatz (eds.) International Agricultural Develop-
ment, 3rd edn. Baltimore MD: Johns Hopkins Press.
OECD-FAO (Organization for Economic Co-operation and Development “ Food and Agricultural
Organization). 2006. Agricultural Outlook 2006“2015. Paris: OECD.
Pingali, Prabhu. 1998. “Confronting the Ecological Consequences of the Rice Green Revolution in
Tropical Asia,” pp. 474“93 in Carl Eicher and John Staatz (eds.) International Agricultural Develop-
ment, 3rd edn. Baltimore MD: Johns Hopkins Press.
”” and Randy Stringer. 2003. “Food Security and Agriculture in Low Income Food Deficit Coun-
tries,” paper presented to the International Conference on “Agricultural Policy Reform and the
WTO,” Capri, Italy, June 23“26. Online at http://www.ecostat.unical.it/2003agtradeconf/Invited%20
Pinstrup-Anderson, Per. 2003. “Food Security and Trade, the Challenges Ahead,” paper presented to the
International Conference on “Agricultural Policy Reform and the WTO,” Capri, Italy, June 23“26.
Raney, Terri. 2006. “Economic Impact of Transgenic Crops in Developing Countries,” Current Opinion
in Biotechnology 17: 1“5.
Reardon, Thomas and Christopher Barrett. 2000. “Agroindustrialization, Globalization and Interna-
tional Development,” Agricultural Economics 23 (September): 195“205.
Rigg, Jonathan. 2006. “Land, Farming, Livelihoods and Poverty,” World Development 34 (January):
Romero, Simon. 2006. “Manitoba Journal: Bolivian Reforms Raise Anxiety on Menonite Frontier,”
New York Times (December 21). Online at http://select.nytimes.com/search.
390 The Process of Economic Development
Rostow, Walt W. 1960. The Stages of Economic Growth. Cambridge: Cambridge University Press.
Russell, Phillip. 1994. Mexico under Salinas. Austin, TX: Mexico Resource Center.
Sachs, Jeffrey. 2005. The End of Poverty. New York: Penguin Books.
Schejtman, Alexander. 1992. “The Peasant Economy: Internal Logic, Articulation and Persistence,”
pp. 276“304 in Charles Wilber and Kenneth Jameson (eds.), The Political Economy of Development
and Underdevelopment, 5th edn. New York: Random House.
Schultz, Theodore. 1964. Transforming Traditional Agriculture. New Haven CT: Yale University
Schiff, Maurice and Alberto Vald©s. 1998. “The Plundering of Agricultural Policy in Developing
Countries,” pp. 226“33 in Carl Eicher and John Staatz (eds.) International Agricultural Develop-
ment, 3rd edn. Baltimore MD: Johns Hopkins Press.
Shaban, Radwan Ali. 1987. “Testing Between Alternative Models of Sharecropping,” Journal of Polit-
ical Economy 95: 893“920.
Smith, R.E. 1972. The United States and Revolutionary Nationalism. Chicago: University of Chicago
Smith, Stephen. 1994. Case Studies in Economic Development. New York: Longman.
Stiglitz, Joseph. 1992. “The New Development Economics,” pp. 261“75 in Charles Wilber and Kenneth
Jameson (eds.), The Political Economy of Development and Underdevelopment, 5th edn. New York:
Random House.
Stringer, Randy. 2001. “Food Security in Developing Countries,” pp. 70“104 in B.N. Ghosh (ed.),
Contemporary Issues in Development Economics. London: Routledge.
Strom, Stephanie. 2006. “Two Foundations Join in Africa Agriculture Push,” New York Times
(September 13): A16.
Tomich, Thomas, Peter Kilby and Bruce Johnston. 1995. Transforming Agrarian Economies. Ithaca,
NY: Cornell University Press.
Trimmer, Peter. 1989. “The Agricultural Transformation,” pp. 358“61 in Gerald Meier (ed.), Leading
Issues in Development Economics. Oxford: Oxford University Press.
””. 1998. “The Role of Agriculture in Indonesia™s Development,” pp. 539“49 in Carl Eicher and
John Staatz (eds.) International Agricultural Development, 3rd edn. Baltimore MD: Johns Hopkins
UNCTAD. 1987. Handbook of International Trade and Development Statistics. Geneva: UNCTAD.
UN, Department of Economic and Social Affairs. 2000. World Economic and Social Survey 2000. New
York: United Nations.
UNDP (United Nations Development Programme). 1993. Human Development Report 1993. Oxford:
Oxford University Press.
””. 1994. Human Development Report 1994. Oxford: Oxford University Press.
””. 1995. Human Development Report 1995. Oxford: Oxford University Press.
””. 2001. Human Development Report 2001. Oxford: Oxford University Press.
””. 2006. Human Development Report 2006. Oxford: Oxford University Press.
WARDA. 2007. “NERICA Contributes to Record Rice Harvest in Africa,” May 3. Online at http://
World Bank. 1992. World Development Report 1992. Oxford: Oxford University Press.
””. 1994. World Development Report 1994. Oxford: Oxford University Press.
””. 2000/1. World Development Report 2001. Oxford: Oxford University Press.
””. 2002. World Development Report 2002. Oxford: Oxford University Press.
””. 2003. Reaching the World Poor “ A Renewed Strategy for Rural Development. Washington,
D.C.: World Bank.
””. 2007. World Development Indicators Database (April 2007). Online at http://devdata.worldbank.
12 Population, education, and human

after reading and studying this chapter, you should better understand:
• the connection between population growth rates and the level and growth rate of
income per capita;
• the importance and causes of the demographic transition and its effect on birth
and death rates;
• the direction of causation is from poverty to population growth and not the
• the key determinants of the fertility rate and population growth, particularly
income per person and the level of education of women;
• the idea of the “opportunity cost” of having children and the impact on fertility
and population growth;
• the importance of education and human capital accumulation to economic growth
and human development;
• the particular significance of scientists, engineers, and other technically-trained
workers who apply R&D to the development process;
• the role for government action to overcome market failure in the creation of
human capital.

The most malleable factor of production available to any economy is its population. It
is not stretching the truth to say that an economy™s labor force is its most significant
resource endowment, and it is one that can be made more productive over time. Our
consideration of endogenous growth theories in Chapter 8 and of the recent successes
of the “high performance East Asian economies” (HPAEs) in Chapters 9“10 have high-
lighted the importance of an educated labor force to economic growth. A better educated
and trained labor force increases what economists call “human capital.” Education is the
means by which a nation is able to appropriate from and share in the gains arising from
technological and knowledge advances at the world level by augmenting the economy™s
stock of human capital.
A properly educated labor force is absolutely necessary for sustained growth and for
achieving full human development. An economy that succeeds in avoiding all the other
pitfalls of developing societies considered in the next part of the book but which neglects
392 The Process of Economic Development
education will not succeed in developing as quickly or to such a high level as would be
possible with more and better human capital.
The accumulation of a productive stock of human capital is thus one of the fundamental
keys to development.1 There is often a lingering question, however, about how population
growth affects the level and pace of economic growth and development. So, prior to consid-
ering in more detail the importance of the human capital input to development, it is necessary
to briefly examine the nature of what, since the time of Thomas Malthus, has been called the
“population problem.”

a population problem?
The so-called population problem is based on an assumption that rapid population growth
can cause the total population to exceed a nation™s productive capacity so that real income per
person falls or rises unnecessarily slowly. Deep down, concerns about a population problem
is simply a reassertion of the Malthusian specter of population outstripping output growth
considered in Chapter 4. We can, perhaps, see why some might think that population growth
is the cause of poverty if we remember from Chapter 2 that

% ”income per person = % ”total income ’ % ”population. (12.1)

This is a definition, that is, it is a mathematically true statement. This says that the change in
real income per person over time depends both on the growth rate of aggregate real income
(measured by GNI or GDP) minus the rate of population growth. Obviously the faster that
total income grows, with population growth held constant, the more rapid will income per
capita rise. The faster that population grows, the slower will be the expansion in income per
person, for any given rate of aggregate income growth. This, however, does not mean that
faster population growth causes slower growth in income per capita or that slower popula-
tion growth leads to a faster increase in income per person.
So statement 12.1 says nothing about whether population growth affects income per
person or vice versa. It simply states that both population growth and total income growth
are important for what happens to income per person. It does not imply anything about
causation. If population is growing 2 percent per year, total income must grow by at least 2
percent just to keep income per person constant. If total income only grows 1 percent, then
per capita income will decrease by 1 percent with 2 percent population growth. But this does
not mean that population growth causes income per person to decrease. Both the growth of
total income and the growth of population are important, and we need to try to uncover which
of the two affects the other.
We shall in fact argue that causation actually runs in the direction of income to population
growth and not vice versa.2 Very simply, an increase in average income in an economy leads
to a lower population growth rate, all else the same.
Let™s consider some initial evidence. In Table 12.1, look first at the broad groupings of
countries ranked as “low-income,” “middle-income,” and “high-income.” From these aver-
ages, it is clear that population growth rates are lowest for the high-income economies and
highest for the low-income economies for the different time periods shown. Of course, this
alone still does not tell us whether population growth affects income or vice versa. But it is
suggestive evidence.
Looking at individual nations, the population growth rate tends to be highest for low-
income economies and lowest for countries in the “high-income” grouping. The “middle-
Population, education, and human capital 393
Table 12.1 Actual population growth rates, by region and selected countries
Share of world populationa
Population growth rate, annual percent

1960“70 1970“80 1990“2004 1980 2000

2.3 2.1 36.3 40.6
Low-income economies
2.3 1.8 0.9
2.3 1.7
India 2.2
2.6 2.4
Pakistan 2.8
2.9 1.3
Somalia 2.8
Zimbabwe 2.6 3.0 1.4

2.5 1.9 1.1 45.8 44.9
Middle-income economies
Argentina 1.4 1.6 1.2
Côte d™Ivoirec 3.8 4.0 2.5
1.4 1.3
Jamaica 0.7
2.3 2.9 2.5
South Koreae 2.6 1.8 0.8

1.1 17.8 14.9
High-income economies 0.8 0.8
0.4 1.4 1.1
1.0 1.1
Japan 0.2
United Kingdom 0.6 0.1 0.3
United States 1.3 1.1 1.2

Less-developed regions
East Asia and Pacific 1.9 1.5 1.1 31.5 30.6
Latin America and Caribbean 2.4 1.6 8.1 8.5
Middle East and North Africa 2.9 3.0 3.9 4.9
South Asia 2.3 2.1 1.9 20.4 22.4
Sub-Saharan Africa 2.9 2.5 8.6 10.9

Sources: World Bank 1983: 184“5, Table 19; 1994: 210“11, Table 25; 2002: 48“51, Table 2.1; World Development
Indicators 2006: 46“8, Table 2.1.
a Total world population in 1980 was estimated as 4,429.3 million and 6,057.3 million in 2000.
b China was reclassified as “middle-income” in 1999.
c Côte d™Ivoire was reclassified as “low-income” in 1994.
d Senegal was reclassified as “low-income” in 1995.
e South Korea was reclassified as “high-income” in 2003.

income” countries might seem to show more variability, but two of these countries have been
reclassified as “low-income” (Côte d™Ivoire and Senegal) and one, Korea, has been advanced
to the “high-income” category. The pattern of population growth rates being lower for econo-
mies with higher incomes tends to be confirmed among individual countries.
Another important tendency can be seen by looking at what has happened to population
growth rates over time. Reading across the columns, population growth rates have been
falling over time, slowly for the “low-income” and “high-income” economies and more
rapidly for the “middle-income” countries. The same pattern is observed for individual econ-
omies and in all the regions at the bottom of the table. As we will argue, falling population
growth rates are closely associated with rising income levels, though with a lag.
To return to the question that headed this section, we argue that there is not a popula-
tion problem. If anything, there is an income problem. Rather than an increase in income
per person resulting in ever more rapid increases in population, as Malthus had argued, the
394 The Process of Economic Development
relation is actually the other way around. Income per person and population growth rates
are inversely related, not directly related, for reasons we shall consider more fully below.
The best way to reduce population growth rates, then, is for a country to increase economic
growth and achieve higher income per person. Population growth is a fundamental conse-
quence of the level of income and not the reverse. It is income that needs to be increased, and
then population growth naturally and inevitably declines.

The natural and the actual rate of population growth
What accounts for differences in population growth rates among countries and regions? To
explain this, we need to define both the natural rate of population growth and the actual rate
of population growth.
In a purely accounting sense, the natural rate of population growth can be defined as

pn = (CBR ’ CDR)/10 (12.2)

where pn is the annual natural rate of population growth; CBR is the crude birth rate, which
is the number of live births per 1,000 population; and CDR is the crude death rate, which is
the number of deaths per 1,000 population.3
Using statement 12.2 for Côte d™Ivoire, for example, in 2004, with a CBR of 37 and a
CDR of 17, the natural rate of population growth was pn = (37 ’ 17)/10 = 2.0 percent. Crude
birth and crude death rates are shown in Table 12.2 for the same countries and regions as
in the previous table. The listings of the countries within each grouping is now shown not
alphabetically, however, but from the country with the lowest income to the country with the
highest income per person within each grouping. This is done to help us see if there are any
patterns we can detect connecting crude birth rates (CBRs) and crude death rates (CDRs) to
average income levels.
The natural rate of population growth, pn, calculated in Table 12.2, often is quite different
from the actual population growth rates shown in Table 12.1 (note that these are averages
for longer periods than the per year calculations in Table 12.2 for the natural rate). For
example, in 1970, the pn for Côte d™Ivoire was 3.1 percent, while the actual annual popula-
tion growth rate over both the 1960s and the 1970s was well above this figure. The natural
rate of population growth thus understated the actual rate of population growth for Côte
d™Ivoire. Considering Somalia in 2000, just the opposite is true; the natural rate of popula-
tion growth overstates the actual trend of population. For Japan, the natural rate of popula-
tion growth predicts the actual population growth rate quite closely.
What accounts for the difference between the natural rate of population growth and the
actual? The actual rate of population growth shown in Table 12.1 depends not only on the
natural rate, pn, as a result of births and deaths. The actual population growth rate also takes
into consideration migration flows between nations. We can define the actual rate of popula-
tion growth, pa, as

pa = pn + m

where m is net migration: m = (immigrants/100 population) ’ (emigrants/100 popula-
tion) and pa, the actual rate of population growth is a percentage. Thus, the actual rate
of population growth is equal to the natural rate of population growth plus or minus net
Population, education, and human capital 395
Table 12.2 Crude birth rates, crude death rates, and the natural rate of population growtha
Crude birth rateb Crude death rateb pn , %

1970 1993 2004 1970 1993 2004 1970 1993 2004

39 29 14 10 11 2.5 1.8 1.8
Low-income economies
41 29 24 18 10 2.3 1.9 1.6
India 8
48 40 19 9 2.9 3.1
Pakistan 27 7 2.0
33 19 12 6 2.5 1.1 0.6
China 8 8
Zimbabwe 53 38 30 16 12 23 3.7 2.6 0.7
50 48c 45 24 17c 17 2.6 3.1c
Somalia 2.8

35 23 16 11 2.4 1.5 0.9
Middle-income economies 8 7
Côte d™Ivoire 51 49 37 15 17 3.1 3.4
20 2.0
47 43 36 16 11 2.5 2.5
Senegal 22 2.7
34 21 18 6 6 2.6 1.5 1.2
Jamaica 8
Argentina 23 18 9 1.4 1.2 1.0
20 8 8
Korea 30 16 9 9 6 5 2.1 1.0 0.4

18 13 12 10 9 0.4 0.4
High-income economies 8 0.8
15 16 11 9 1.1 0.6 0.9
Ireland 22 7
United Kingdom 16 13 12 12 11 10 0.4 0.2 0.2
United States 18 16 14 10 9 0.6
8 0.8 0.7
19 10 9 9 1.2
Japan 7 8 0.2 0.0

Less-developed regions
East Asia and Pacific 35 21 15 9 2.6 1.3
8 7 0.8
Latin Americaand Caribbean 36 26 21 10 6 2.6 1.9 1.5
Middle East and North Africa 45 33 25 16 6 2.9 2.6 1.9
South Asia 42 31 25 18 10 2.4 2.1 1.7
Sub-Saharan Africa 47 44 40 15 18 2.9
20 2.7 2.2

Sources: World Bank 1994: 212“13, Table 26; 1995: 212“13, Table 26; 2002: 48“51, Table 2.1; World Development
Indicators 2006: 46“8, Table 2.1
a Countries are ranked in terms of their 1993 GNI per capita, from lower to higher incomes within income groupings.
b Per 1,000 population.
c 1992.

For countries with little immigration or emigration, or in which these flows are relatively
balanced, the actual rate of population growth will be very similar to the natural rate. For
countries with a high level of emigration relative to immigration, that is, with more people
leaving the country than entering, the natural rate of population growth, pn, will overstate the
actual rate of population growth, pa, as in Somalia. In countries, like Côte d™Ivoire, where
net migration is positive, the actual population growth exceeds the natural rate, as population
inflows exceed population outflows.
But we still have not answered a critical question for understanding population growth
and the relation between income levels and population. We need to understand what it is
that explains differences in CBRs and CDRs among countries, and hence population growth
rates, and their trends over time. Why do CBRs and CDRs differ among nations?
To understand this, it is helpful to consider the issue within a somewhat longer time
frame. A look at the so-called demographic transition will illustrate the factors affecting
population growth rates via the impact over time of various forces at work on the levels of
the CBRs and the CDRs.
396 The Process of Economic Development
The demographic transition
If we were to go back a century or so and examine natural population growth rates, we would
find a quite different picture from that shown in Table 12.2. Crude birth and crude death rates
were both higher, so that most countries had natural rates of population growth in the neigh-
borhood of about 1 percent per year.4 High death rates tended to nearly cancel out relatively
high birth rates, so population growth rates were quite low by the standard of many LDCs
today. Births and deaths were in a sort of perverse balance, so that there was no population
explosion. Natural population growth rates at 2, 3 and 4 percent are primarily twentieth- and
twenty-first-century phenomena.5
What changed to make such historically high rates of population growth a reality in some
LDCs, while population growth remained low in richer economies?

The demographic transition and the now-developed economies
What changed was the spread and speeding-up in some countries of the structural transfor-
mation toward industrialization and capitalist expansion (examined in Chapters 9“10) and
higher levels of income and development, which contributed both to falling birth rates in the
nations experiencing accelerated development and to a worldwide trend toward lower death
rates in nearly every nation, rich and poor. For nations on the road to industrialization, espe-
cially Western Europe, the United States, Canada, and Japan, both crude birth rates and crude
death rates fell relatively rapidly, so that their natural population growth rates remained close
to 1 percent. In these countries, there was no increase in natural population growth rates.
In these now-developed nations, rapid economic growth resulted in improvements in
living standards, incomes, and education which led people to choose to have fewer children,
thus significantly reducing birth rates. At the same time, these economies experienced falling
death rates, as improved economic conditions contributed to better health and longevity.6
When modern medicine set out to tackle major public health hazards, like malaria, measles,
smallpox, cholera, typhoid, diphtheria, poor sanitation, and so on, death rates fell even more
rapidly, especially after 1900 or so. The systematic understanding, production, and use of
antibiotics beginning in the early twentieth century also contributed to lower CDRs.
With these decreases in both the CBR and the CDR, the now-developed nations were in
the process of completing what is called the demographic transition. This occurs when both
the crude birth rate and the crude death rate have fallen to a level equal to less than 20 per
1,000 of population. As a result of this transition, population growth rates typically are at, or
even below, 1 percent per annum.

The demographic transition and the LDCs
In the less-developed nations, however, where the structural transformation toward more
productive industrialization and capitalist development was either absent, very primitive,
or dualistic (especially prior to decolonization in the late 1940s but also after), crude birth
continued to be significantly higher than in the now-developed nations. However, crude
death rates decreased in the LDC economies after 1900 or so, often to levels equal to or even
below those attained in the developed nations.7
The fall in death rates in the LDCs was not wholly the result of economic, social, and
health improvements taking place within these nations themselves. Instead, the sharp drop in
worldwide crude death rates, especially after 1945, was the consequence of the public good
Population, education, and human capital 397
characteristics of and positive externalities associated with the great strides in public health
and sanitation measures (water and sewage), immunization for childhood diseases, pest
control, and similar measures that had originated in the developed world. However, because
of the positive spillover benefits of many of these measures, the gains were transferred to and
available in much of the less-developed world as well.
However, with CBRs remaining high, the effect of a period of rapidly decreasing CDRs
was to open a growing breach in the less-developed nations between their birth rates, which
fell only slowly as incomes increased slowly, and their rapidly declining death rates. This
asymmetry led to an inevitable ratcheting upward in the rate of population growth in the
less-developed nations as the gap between slowly falling but still relatively high CBRs and
rapidly falling but quite low CDRs widened.
You will remember from statement 12.2 above that the natural population growth rate is the
difference between the CBR and the CDR (divided by 10 to state the result as a percentage
rate). Thus, if the CDR decreases quickly, as happened in most LDCs around the 1940s, the
result is an increase in the population growth rate, no matter what the CBR was. And that is
what happened. As a result, LDCs all of a sudden found themselves facing a higher popu-
lation growth rate due to the positive spread effects of public health measures pursued by
the developed countries that reduced worldwide crude death rates but which did nothing to
reduce CBRs in the LDCs.
Most of the less-developed countries have thus passed through but one-half of the demo-
graphic transition. Death rates have fallen significantly and are more in line with levels asso-
ciated with higher incomes per person in more developed economies, but they did so not
because of what happened internally in those economies but as a result of public health meas-
ures at the world level that reduced CDRs everywhere. As a result, in nearly every country
of the world, CDRs have passed through the 20 per 1,000 population threshold. Birth rates,
however, have remained relatively high in the less-developed economies, reflecting their
relatively low income levels, though CBRs have been falling as incomes have increased over
time. It is a cruel twist of fate that visits on the less-developed nations some of the best of the
developed world “ low CDRs “ and the worst of the less-developed world “ relatively high
CBRs “ simultaneously.

Graphing the demographic transition
Figure 12.1 is a stylistic representation of the phases of the demographic transition and the
changes over time in crude birth rates and crude death rates and their effect on population
In Phase I, prior to the Industrial Revolution and the spread of capitalist methods of produc-
tion within the now-developed world, both birth and death rates were high for all countries,
so population growth was relatively slow everywhere. The gap between the CBR line and the
CDR line graphically measures the rate of population growth.
In Phase II, death rates began to decline in the more-developed nations due to both
the effects of higher incomes that improved health care, but also as a consequence of
worldwide health measures which brought mortality rates down for all countries, regard-
less of income level, as discussed already. In Phase II, the developed nations experienced
declining birth rates due to rising income levels at the same time that their death rates were
decreasing, so population growth did not accelerate. The gap between the CBR line and
the CDR line remained relatively stable as both CBRs and CDRs decreased at somewhat
the same pace.
398 The Process of Economic Development
Birth and death

Birth rate

Birth rate, less-developed

Death rate

Birth rate, developed

Phase I Phase II Phase III Time

Figure 12.1 The demographic transition.

In the less-developed nations, on the other hand, in Phase II, crude birth rates remain high.
They do decrease, but only very moderately as incomes slowly rise. As a result, population
growth rates actually increased, as can be very clearly seen from the widening gap between
the birth and death rates lines. Again, this was due to the worldwide benefits of health meas-
ures that reduced CDRs in poor and rich nations alike, not from anything endogenously
occurring within the LDCs themselves.
It is only in Phase III, which has not yet been attained for most less-developed nations, when
their birth rates also decline below 20 per 1000 population that they also will complete the demo-
graphic transition. For those less-developed nations which fail to make the necessary structural
transformations considered in previous chapters and which have lagging per capita income levels,
population growth rates remain high as these economies are mired in Phase II of the demographic
transition with high CBRs (> 20) and low CDRs (< 20). The result of this imbalance is a relatively
high population growth rate. This is the situation facing countries like Pakistan, Côte d™Ivoire and
Senegal shown in Table 12.2.
Some countries in Sub-Saharan Africa are facing an even graver situation because of the
HIV/AIDS crisis. Zimbabwe suffered a nearly two-fold increase in its CDR between 1993
and 2004 (Table 12.2) as a result of the effects of HIV/AIDS. Having passed through half
the demographic transition (with CDRs < 20), by 2004 CDRs were again above 20 per 1000
population, as Zimbabwe™s transition moved back in time. As a result of this perverse trend,
the population growth rate declined from 2.6 percent to 0.7 percent. However, unlike lower
Population, education, and human capital 399
population growth rates attained in the “high-income” economies as a result of lower birth
and death rates, what we might call a “good” low population growth rate, Zimbabwe™s lower
population growth rate represents a retrogression to the high CBR, high CDR demographic
pattern characteristic of a pre-industrial world economy and, within the current context, is
a “bad” low population growth rate. Focus 12.1 looks at this demographic, economic and
human tragedy of HIV/AIDS in more detail.

After 1980, life expectancy began to fall in Sub-Saharan Africa. Why? Lots of reasons:
malnutrition, malaria, prolonged droughts, isolation from the world economy, geography,
and failed economic policies. And then there is HIV/AIDS. Consider the following data.

1980 1990 2000 2005 HIV


Africa 48.1 12.9 49.3 16.6 46.7 17.3 47.2 17.3 5.8
Zimbabwe 59.3 10.3 58.6 9.7 39.8 20.6 37.3 22.9 20.1
Kenya 57.8 11.4 57.9 10.2 48.4 14.7 49.0 14.5 6.1
World 62.6 10.4 65.9 9.2 66.7 8.9 67.6 8.6 0.9

While life expectancy (LE) in the world has been moving upward and crude death rates
(CDR) down, in Africa as a whole and in Zimbabwe and Kenya as two concrete exam-
ples, the trends have been in the opposite direction. The downturn in life expectancy in
Zimbabwe has been especially sharp, falling from nearly 60 years in 1980 to less than
40 years in 2005. As noted earlier in the chapter, Zimbabwe has suffered a more than
two-fold increase in its CDR, an unprecedented reversal of the demographic transition.
The major explanation for this human tragedy is the high incidence of HIV/AIDS, which
was in the 20 percent range for adults in 2005. (These figures may be revised downward,
but the relative magnitude of the HIV/AIDS crisis is all too real, even if the incidence rate
is actually somewhat lower.)
The reasons for the higher incidence of HIV/AIDS in Africa than in other parts of the
world are complicated, too complicated to discuss here (for a short but very useful
overview, see Sachs 2005: Chapter 10). But the need to stop the disease in its tracks is
undeniable, on both human and economic grounds. The costs to countries and families
of HIV/AIDS treatment is immense. And it™s not just the direct costs of the antiretroviral
drugs that save lives and give people hope. It is also the opportunity costs of providing
health care (that is, what if other expenditures are sacrificed to care for HIV/AIDS patients
that might have been dedicated to economic development?) and the impact on worker
productivity for the current generation, but also for future generations.
How are future generations harmed? Because many children are being orphaned and
their education and futures short-circuited. The impact on economic growth and devel-
opment in those countries will reverberate for some time. More needs to be done, not
just by the countries affected, but by the international community. Remember, Goal 6 of
the Millennium Development Goals is to “have halted by 2015 and begun to reverse the
spread of HIV/AIDS.” One of the exercises at the end of the chapter asks you to see how
the world is doing on this goal.
Sources: Sachs 2005: Chapter 10; World Bank,
World Development Indicators Online
400 The Process of Economic Development
Determinants of the crude birth rate: understanding the dynamics
of population growth
Let™s return to Table 12.2. First, look at the crude birth rate for, say, 1993. Read down that
column. What do you observe? As we scan down, the CBR tends to decrease, it is true, which
is the inverse relation between the CBR and income that we noted earlier. That is one worthy
observation. But we also note the relatively large variance in the CBR from a high of 49 in
Côte d™Ivoire to a low of 10 in Japan in 1993.
Now read down the crude death rate column for 1993. There does not seem to be any
systematic relation between the level of income of a country and the CDR, certainly nothing
as clear-cut as for the CBR. Further, the crude death rate tends to be dispersed within a
relatively narrow band (a high of 17 in Senegal and a low of 6 in Jamaica) as a result of the
spread of world public health measures like vaccinations and mosquito control that have
lowered death rates worldwide.
The difference in the patterns of the crude birth rate and the crude death rate is because
each country™s CBR is determined by family-specific national determinants particular to each
individual economy, while the CDR is affected to a substantial degree by world health meas-
ures and their positive external effects, though national health policies do have an impact.
For crude birth rates, the most important country-specific determinants have been found to
be (a) family income and (b) the education level of women.8
There are, of course, country-specific cultural and religious factors which impinge on birth
rates in particular nations and regions, but even when these are factored in, the evolution of
family income and women™s education in each country are the most important determinants
of birth rates over time. Reading down the CBR column in Table 12.2 for 2004 confirms the
general tendency for the CBR to decline with higher income, as discussed earlier. Reading
across the columns from 1970 to 2004 for each individual country and for all regions, it can
also be observed that the CBR declined, too. Since average incomes rose in most countries
over this period of time (more about this in the next section), this is additional evidence for
the inverse relation between income and CBRs.
The CBR in China has decreased rapidly, such that the population growth rate is more
like what one might find in a high-income economy. Why is that? Easy. China™s “one-child
policy” instituted in 1979 enforces large fines on additional children in urban area, thus
increasing their cost and deterring many families from having more children.

The CBR and fertility rates
To review, population growth rates depend on the evolution of CBRs and CDRs in any partic-
ular country. A key determinant of CBRs is the average level of income in each economy.
Another significant factor affecting CBRs has been found to be the level of education of
women, and not simply because women become more aware of birth control. The issue is
rather more complicated and is economic at root, as is discussed more fully below. But keep
in mind that both higher average income levels and higher education levels for women are
the key variables that affect CBRs in every economy, and they do so via fertility rates. We
thus need to briefly consider this connection between the CBR and the fertility rate.
The total number of children that the average woman is expected to have is called the
fertility rate. The fertility rate tends to be lower in regions with higher incomes and higher in
regions with lower incomes, as can be seen quite clearly in the top part of Table 12.3. Since
the CBR depends upon what happens to fertility rates over time, the inverse relation between
Population, education, and human capital 401
Table 12.3 Fertility rates, income, and women™s education
Total fertility ratea Ratio of girls to boysb

per capita ($)c 1970 1990 2004 1970 1990 2005

All regions
580 (4.2%) 5.9 4.4 3.7 59
Low-income economies n.a. 87.8
2,241 (6.2%) 5.1 2.6 2.1 84 “
Middle-income economies n.a.
35,131 (7.3%) 2.4 1.8 1.7 94 96 99.9
High-income economies

Less-developed regions
East Asia and Pacific 1,627 (7.5%) 5.7 2.4 89 99.0
2.0 n.a.
Latin America and Caribbean 4,008 (5.5%) 5.3 3.4 2.5 95 100 101.4
Middle East and North Africa 2,640 (2.4%) 6.8 4.8 3.0 53 93.5
South Asia 684 (4.9%) 6.0 4.1 3.1 51 71 88.4
Sub-Saharan Africa 745 (3.2%) 6.6 6.1 5.3 58 86.5

Sources: World Bank, World Development Indicators 2002; World Development Indicators 2006; World Develop-
ment Indicators Online.
a Total fertility is the number of children that would be expected to be born alive to the “average” woman if she
were to live to the end of her child-bearing years and have children according to age-specific fertility rates.
b The number of females per 100 males in primary and secondary school.
c 2005. Number in parentheses after GNI per capita is the average growth rate of GNI per capita, 1970“2004
(1979“2004 for the Middle East and North Africa).

fertility rates and income also manifests itself in the relation between average income in an
economy and the crude birth rate as changing fertility rates affect the number of children
women choose to have.
Within all regions over the period 1970 to 2004, the fertility rate decreased. Those regions
that have shown the fastest growth in income per capita and that have reached higher levels
of income per capita, as in Latin America and the Caribbean and in East Asia, have demon-
strated the most rapid decline in fertility and birth rates within the less-developed world (see
Birdsall 1988). All regions have experienced declining fertility rates as income per capita has
increased; Table 12.3 shows average income in 2005 and the annual percentage change in
GNI per capita in every region since 1970.
Table 12.3 also shows the number of females per 100 males attending primary and
secondary school. Higher ratios in 1970, indicating a reduction of the education gender gap,
are associated with lower fertility rates in 1990 and 2004, underscoring the importance of
women™s education as a means of reducing fertility and population growth rates.9 Women™s
education increases opportunities for earning income and is a key factor in leading women to
choose to have fewer children. Note that in Latin America and the Caribbean there are more
girls than boys attending primary and secondary school.
It will be noted that the Middle East and North Africa still have a relatively high fertility rate
and population growth rate compared to other less-developed regions, despite a narrowing
of the education gender gap and a relatively high income level per capita compared to both
South Asia and East Asia, which have lower fertility rates.10 This is likely to be the result
of distinct cultural factors, perhaps partly reflecting the role of women in Islamic societies,
which have prevented as rapid a reduction in fertility and population growth that the impact
of more female education and rising income would incline us to predict. The same is true for
the higher fertility rates in Latin America compared to some lower income regions.
402 The Process of Economic Development
Nonetheless, abstracting from these regional and country-specific factors that can make
fertility rates higher or lower for cultural or religious reasons, the fact remains that over
time, as average incomes and women™s education rise within any economy or region, fertility
rates do fall. Higher incomes and more education for girls result in a smaller number of
children born per woman, and hence a lower birth rate, irrespective of the cultural, religious,
or other specific factors. That this has occurred in the Middle East and Latin America can be
confirmed by looking at what happened to fertility rates in both regions from 1970 to 2004.11
Over time, fertility decreased with rising income and education levels.
Family planning, the availability of and knowledge about contraception methods, social,
moral, and legal views on and access to abortion services, and a whole range of other factors
can result in fertility and population growth rates that are higher or lower than might be
expected given income per capita or the level of female education in any particular country.
Such interventions can help to speed the process of reducing CBRs so that countries pass
through the demographic transition threshold more quickly. However, these measures are
unlikely to be substitutes for economic growth and the expansion of educational opportuni-
ties for women as the most effective means to reduce fertility rates and crude birth rates.
As family income and women™s education rises, then, there is an unmistakable tendency
for birth rates to fall, as was confirmed in Table 12.2, because fertility rates decrease.12 This
means that average family incomes will tend to rise over time as per capita income rises,
assuming no strongly adverse changes in the income distribution, since family size will fall
over time.13 This is almost certain to raise family well-being.

The role of children in less-developed and developed economies
Though Table 12.3 suggests that fertility and hence birth rates tend to be inversely related to
the level of per capita income and the education of women, the explanation for such a link
needs to be clarified. At root, the reason is primarily economic.
We can formalize the relation between fertility rates and their determinants as

F = f(y, e, S) (12.4)

where F is the total fertility rate, y is per capita income, e is the number of females per 100
males in primary school, and S measures the social, cultural and political factors specific to
each country that also may affect fertility. It is expected that the first derivatives, fy, fe < 0,
that is, fertility is negatively related to income and education levels, while the relationship
between S and fertility (fS) is indeterminate. The reason for expecting an inverse relation
between fertility and income is predicated on the different roles children perform in low-
versus high-income economies. Let™s examine this assumption.
In economies at lower levels of per capita income, children play multiple roles for their
parents. even at young ages, children can provide an additional source of labor on family
farms thus contributing to total family income and welfare. Within the household, older girls
can care for younger siblings and help with the cooking, cleaning, fetching of wood and
water, and other simple but essential tasks that contribute to the family™s overall well-being.
Just as importantly, children often are a form of insurance for poor parents in their old
age, ensuring that they will be provided for. This effect is strongest in countries lacking
broadly-based social security and old-age pension systems. It is for this reason that the
so-called extended family, in which grandparents, parents, children, and perhaps aunts
and uncles and cousins live in the same house or in very close proximity to one another,
Population, education, and human capital 403
is so much more common in the less-developed world than in more developed economies.
Without an old-age social security system in place, poor elderly parents often may be
forced to live with their children and grandchildren as a means of survival, and the natural-
ness of such arrangements is deeply inculcated in all members of the family as a reciprocal
relationship. It should be remembered, nonetheless, that such arrangements are born out
of necessity and are at root a result of economic necessity as much as they may seem to be
desired social arrangements.
Another link between higher fertility rates and low incomes is found in infant and child
mortality rates shown in Table 12.4. Families in poor economies are aware that the proba-
bility of losing a child either at birth or at a young age is high, but they do not know precisely
how many, if any, of their children will fail to reach maturity. According to the table, in Sub-
Saharan Africa, the probability of mortality of a child before age five was nearly 1 in 5 in
2004, and this probability is higher the poorer the family. Thus, to the extent that children are
at least partly a form of old-age insurance and given that the risk of losing a child is highest
in Sub-Saharan Africa among all the less-developed regions, we would expect fertility rates
to be higher there than other regions because of the greater risk of loss. And that is precisely
what Table 12.3 confirms.
These multiple roles of children, and the perceived risks of mortality, thus lead parents
in poorer nations to “choose” to have more children for income and social security reasons.
Children are an economic resource for poor families. This does not reflect on how much
parents love their children, but simply represents an economic perspective on how families
“choose™ the number of children to have. Children are a valuable resource to their families in
poor families and poor economies.
There is another aspect of this “choosing” which comes into play in both low- and high-
income economies. This has to do with the opportunity costs of bearing children. Let™s consider
this by considering why parents choose to have children in high-income economies.
What is the “role” of children in richer economies? How does it differ? At higher levels
of per capita income, fertility declines, because the cost of having children rises in terms
both of the care required and educational and other direct expenses. But there also is another

Table 12.4 Infant and child mortality rates
Under-five mortality rateb
Infant mortality ratea

1970 1990 2004 1970 1990 2004

All regions
134 94 79 209 147 122
Low-income economies
79 43 30 125 57 37
Middle-income economies
21 9 6 26 11
High-income economies 7

Less-developed regions
East Asia and Pacific 79 43 329 126 59 37
Latin America and Caribbean 84 43 124 54 31
Middle East and North Africa 134 60 44 81 55
South Asia 138 86 66 209 129 92
Sub-Saharan Africa 138 111 100 185 168

Sources: World Bank, World Development Indicators 2002 and 2006: Table 2.19.
a The number of children who die before age one, per 1,000 live births.
b The probability of a newborn baby dying before reaching age five, per 1,000 live births.
404 The Process of Economic Development
important cost to having children, and this is their opportunity cost to their parents, but
especially for the mother. At higher levels of income, the time spent caring for, feeding,
washing, taking children to and from school, dancing, and sports practices, and so on has a
higher value than at lower levels of per capita income in terms of forgone income. This is
true particularly for working mothers who have to take time off from paid employment to
nurse sick children, pick them up from school, get them to the dentist™s office, or whatever
it is that takes them away from the workplace. Simple economic analysis would suggest that
the number of children “demanded” would decrease as income rises since the (opportunity)
cost, or implicit price, of an additional child is greater at higher income levels.
Women remain the primary care-givers of children in every country regardless of shifting
responsibilities of men vis-à-vis women, and women thus bear the burden of this opportunity
cost (see Focus 12.2 for another view of the role of women within the household). As their
education level rises, women™s income-earning power increases and their relative contribu-
tion to family income grows. As a consequence, women begin to choose to have fewer chil-
dren because of the rising opportunity cost, as additional children mean a larger amount of
forgone income (see Sachs 2005: 64“5 on Iran™s experience).
At higher income levels, the lost income from children is greater than at lower income
levels, so the number of children a family chooses to have is inversely related to family
income. As women™s and family income rises, there are fewer children naturally “chosen”
for good economic reasons. At lower income levels, families “choose” to have more children
since the cost (which is the implicit price) of each is less in terms of forgone income. This
also explains why women who do not work outside the home, even in richer economies, may
choose to have more children, since the “price” of an additional child in terms of forgone
income is lower. It is a simple price and quantity-demanded decision based on the logic of
the downward-sloping demand curve you learned in basic economics.
In richer countries, parents do not have children as either investments in their futures, as a
substitute for social security, or as potential workers. As one of our more perceptive students
once put it, in richer economies (and in richer families in poor economies) children might
be better seen as a source of “entertainment” for their parents. They are doted on, dressed in
designer clothes, groomed from infancy for pre-schools and beyond, and generally treated as
objects of consumption rather than as investments, as is the case in poorer economies.
As objects of consumption who require that spending be undertaken, children are expen-
sive and are more expensive the higher the income of the family having them. The preva-
lence of the “nuclear” family “ mother and father and one or two children “ is thus the
norm in economies with higher average income as the economic rationale for the extended
family tends to dissipate. Children are less likely to be the source of income for their elderly
parents, as developed economies have created old-age security systems and many workers
have pension plans. As people age, fewer are dependent on family members for their living
arrangements and well-being.
There is another reason why families in both poor and richer nations may choose to have
children: to carry on the family name. Since family names typically are passed to the next
generation via male children, this results in a bias in favor of male over female babies. Nobel
Prize-winning economist Amartya Sen has written passionately about how this predilection
has led to abortions and undernourishment of female children through what he calls “natality
inequality.” In a famous study of Asia, he writes of the “one hundred million” missing women
(Sen 1990). In part, the abnormally low number of women to men in India and Pakistan and
in other countries in Asia and Africa “ roughly 93 to 100 rather than the more typical average
of 98 “ is the result of the “role” of male children in transmitting the family name to the next
Population, education, and human capital 405

It is not only fertility rates that decline with higher levels of education for women. Health
indicators, and hence family welfare, also improve.
In Africa, studies have shown that a 10 percent increase in female literacy rates reduced
child death rates by 10 percent; by comparison, changes in men™s literacy rates had no
effect on child mortality rates. In Thailand, it was found that women with some primary
education were 30 percent more likely to know how to treat diarrhea in their children “
using homemade salt and sugar solutions or other oral rehydration methods we consid-
ered in Chapter 1 “ than were mothers with no education, thus reducing infant deaths.
In studies in Indonesia, Kenya, Morocco, and Peru, the strong positive correlation of
years of women™s education and a reduction in child mortality rates was confirmed again
and again. In all instances, women™s education was more important than men™s education
in reducing the death rate of children.
Further, income earned by women is spent differently within the household. In Jamaica,
a household study found that women spend more of their income on goods for their
children than do men “ and less on alcohol. “In Guatemala, it takes fifteen times more
spending to achieve a given improvement in child nutrition when income is earned by the
father than when it is earned by the mother.”
What do these results say about the role of men™s and women™s incomes and education
levels within the household? What explains these differential effects? Why is the marginal
contribution of men™s income and education to household welfare less than women™s?
These are worthy subjects of discussion and research.
Source: World Bank 1993a: 41“3

generation. This results in girls not receiving the same level of medical care or nutrition as
boys, deficiencies that often are carried through life and contribute to higher mortality rates
than would be expected for females relative to males.

Human capital accumulation: augmenting initial endowments
Beginning in the 1960s, economists began to seriously study labor not just as a homoge-
neous factor of production, L, but as a differentiated and moldable input to production,
that is, as human capital. This suggested that nations could invest in people via education,
work training, on-the-job training, nutrition, health care, sanitation, and so on to increase
the quality of the employed labor force, just as investment could take place in increasing
the quantity and quality of physical capital via technological innovation. The dissatisfaction
with the inability of neoclassical growth theories to fully explain the sources of economic
growth by the accumulation of more physical capital and by the growth in the labor force led
economists to consider more fully education, training, and technology, which might account
for more rapid growth in some economies than in others.
Our consideration of endogenous growth theories in Chapter 8 and the successes of the
high-performance Asian economies (HPAEs) confirmed the importance of a well-trained
population and labor force within the development process. It is certainly not the only factor
in successful development, but it is important. Human capital accumulation is not sufficient
to guarantee success (think of the former Soviet Union), but it certainly would seem to be
necessary for success (for a critical view see Easterly 2001). In this section, we examine in
greater detail the contribution that human capital accumulation, especially schooling, can
have on the prospects for economic growth and human development.
406 The Process of Economic Development
The contribution of education to development
In the World Bank™s study (1993: 52“3, Table 1.9) of the HPAEs, they found that enrolments
in primary education in 1960 predicted the following proportion of economic growth over
the period 1960“85.
% of total
predicted growth
Hong Kong
South Korea 67
Thailand 87

These are astounding and instructive results. The level of primary education was far and
away the most important contributor to the predicted growth rates of the HPAEs and Japan.
Further, the accumulated and improving human capital stock of the HPAEs contributed to
their ability to be able to adopt, adapt and endogenize the ever-expanding pool of “best prac-
tice” technological knowledge being created at the world level. As a consequence, in most of
these economies, because of the qualities and relatively high level of training of the popula-
tion and labor force, levels of productivity and of “technical efficiency change” were able to
grow quickly (look back at Table 8.4 for confirmation of this). In other words, most of the
HPAEs were able to move closer over time to the ever-shifting-outward production possibili-
ties frontier of “international best technological practice” by becoming more technologically
efficient as their growth rates of what economists call total factor productivity (TFP) have
exceeded the rate of exogenous, best practice technological change. We will be examining
TFP in much more detail in the next chapter.
Table 12.5 provides some data on the level and pace of human capital accumulation for
various countries. Years of schooling is used as a proxy measure for the human capital stock.
This is a typical assumption made by economists that presumes that better educated workers
are more productive workers. A higher average level of education is taken to be an indicator
of a higher level of human capital accumulation.
The data in the table are quite suggestive. Countries were more likely to be middle- or
high-income in 1993 if their 1970 level of primary education coverage was close to universal,
that is, 100 percent. Among the less-developed regions, East Asia and the Pacific had the
second-highest stock of human capital in 1970 as measured by primary school enrolments.
Further, the mean value of schooling, shown in the last column of the Table 12.5, was higher
at higher levels of income and at higher levels of human development, as measured by the
HDI (see UNDP 1994: 138“9, Table 5). For example, while Indonesia is ranked among the
low-income economies, it is ranked as having “medium human development” according to
the HDI. South Korea, an upper-middle-income nation by the World Bank™s income rankings
at the time, was ranked thirty-second among all nations on the HDI index and was among the
countries with “high human development.” The level of schooling in both these economies
is one important reason for these results.
Note that the “average years of school” measure is the mean value for adults aged twenty-
five or more. To the extent that more years of school on average is a measure of greater
human capital accumulation and potential for growth, this trend augurs well, by itself, for the
Population, education, and human capital 407
Table 12.5 education and human capital accumulation
Primarya Secondaryb Tertiaryc Primary Average
student “ years of
teacher ratio

1970 2004 1970 2004 1970 2004 1970 1998 2004

66 100 18 46 3 9 43 42 43
Low-income economies
54 106 51 46 59 54 2.6
Bangladesh n.a. 2 7
107 24 52 5 11 41 41 5.1
India 78 72
116 16 62 3 16 29 5.0
Indonesia 80 22 20
Kenya 62 111 9 48 1 1 34 40 4.2

94 111 75 3 24 30
Middle-income economies 28 22 22
39 76 9 19 1 5 45 49 43 2.6
106 109 23 79 5 46
Mexico 22 27 27 7.2
119 93 46 84 5 19 47 31 30 5.3
Korea 103 105 42 91 89 57 31 30 10.8

100 100 105 67 26 17 16
High-income economies 77 27
101 101 65 105 53 57 23 18 17 12.2
100 100 102 18 52 26 21 10.8
Japan 87 20
United Kingdom 104 101 73 170 14 63 23 19 17 11.7
United States 100 84 95 47 83 15 15 12.4
88 27

Less-developed regions
East Asia and Pacific 90 113 24 69 1 17 30 23 22 n.a.
Latin America and 107 121 6 26 33 25
28 87 28 n.a.
104 24 67 4 23 34 24 24
Middle east and 70 n.a.
North Africa
South Asia 71 103 23 49 4 10 42 66 42 n.a.
Sub-Saharan Africa 51 93 6 30 1 5 43 40 49 n.a.

Sources: World Bank, World Development Indicators 2002 and 2006: Tables 2.10 and 2.11.
a Total (“gross”) enrolment in primary school as a percent of primary school-age children (often 6“11 years). If
this is greater than 100, it indicates that some children younger and/or older than the standard age are enrolled in
primary school.
b Calculated the same as in the previous note, but as a percent of secondary school-age children (often 12“17 years).
c Determined by dividing the number of students enrolled in all post-secondary education (college, university,
technical, etc.) by the population aged 20“24.
d Most recent year, for population aged 15 and above.

less-developed nations approaching universal primary school coverage. Of course, quantity
is not enough; the quality of the education received is fundamental too, though this is inher-
ently more difficult to measure.
The “primary student“teacher ratio” in Table 12.5, which measures the number of students
per teacher in primary schools, can be interpreted as one indicator of the quality of education.
The ratio is positively related both to the mean years of schooling and to the level of income
of the country. It suggests that those countries most focused on enriching and enlarging their
stock of human capital will be those which strive to increase both the average number of
years of schooling, what we can call human capital broadening, and those which improve
the quality of schooling, what we can call human capital deepening, which is measured here
408 The Process of Economic Development
by smaller class size. The assumption is that in smaller classes teachers can give more
individual attention to students and that “more learning” is likely to take place.
Look at Bangladesh™s record. From 1970, human capital broadening definitely took place at
the primary school level so that by 2004 universal education had become the norm. However,
the primary student“teacher ratio suggests that it is likely that the quality of schooling
suffered as there was an average of fifty-nine children per teacher in 1998 compared to forty-
six in 1970. Yes, Bangladesh was educating more students, but was it done at the sacrifice of
quality as students were shoved into ever more over-crowded classrooms without the atten-
tion from teachers to make the learning experience of real value? It certainly looks like that
might be the case.
The situation was even worse in India over the same period as the average number of
students in each primary school classroom dramatically increased to seventy-two in 1998.
And if one thinks of the other complementary inputs to education “ books, pencils, paper,
desks “ is it likely that either Pakistan™s or India™s over-burdened teachers had sufficient
numbers of these for the larger number of students they were trying to teach?
By 2004, however, human capital deepening had begun to take place in both Bangladesh
and India, as average class size per teacher fell. In India the decrease from 1998 to 2004 was
quite dramatic. Nonetheless, average class size in these and in the low-income economies
remained, on average, about twice or more that of the middle-income and high-income
Korea, on the other hand, has continuously reduced the number of students per teacher
over the entire period, suggesting that there was constant improvement, or at least no dete-
rioration, in the quality of education in each classroom on average, at least as measured by
the student-to-teacher ratio.
So while universal primary and secondary education and human capital broadening
should be important goals for countries wishing to increase the level of development, it is
also important that sufficient financial and other resources be allocated so that the quality
of such education, that is what we call human capital deepening, is not compromised
in the interests simply of increasing the coverage ratio of the percentage of students in
school. Though this is not noted in Table 12.5, it is also of importance to maintain balance
in the pursuit of universal primary school coverage, keeping in mind the need to address
and reduce the gender education gap and the rural“urban education gap in designing human
capital broadening and deepening programmes.
Goal 3 of the Millennium Development Goals from Chapter 1 was to “promote gender
equality and empower women” and specifically to “eliminate gender disparity in primary and
secondary education preferably by 2005 and in all levels of education no later than 2015.”
Our tables do not look at these disparities in detail, but a good out-of-class exercise would be
to go to the World Bank website and look at the data on the so-called gender gap in primary,
secondary, and tertiary education. Is the gap closing, as the MDGs prescribe?

Human capital accumulation and market failure: what is
the role of government?
The evidence on the importance of education as a specific form of human capital to accel-
erating economic growth rates seems incontrovertible (but again, see the views of Easterly
2001).14 Nations with a larger stock of human capital tend to grow faster than those with less
human capital, and they are able to reach higher levels of income per person. The endog-
enous growth models reviewed in Chapter 8 are unambiguous on this, and much of the vast
Population, education, and human capital 409
economics literature on human capital is dedicated to exploring the “value” of education to
individuals and to society (see the classic articles by Mincer (1958) and Schultz (1960)).
One of the important reasons why growth rates do not necessarily fall with increased human
capital accumulation, that is, one possible reason for the absence of diminishing social returns
to human capital investment, is that human capital accumulation creates substantial positive
externalities as the exogenous growth models argue.
An important issue, however, is what role does government have in the provision of educa-
tion? If a better educated labor force is so important to economic growth and human welfare,
should government intervene to augment that stock of human capital? Or should these deci-
sions be left to individuals? Let™s consider this matter in the following way.
When an individual undertakes schooling, that person is, of course, likely to earn more
income over his or her lifetime as a consequence of the higher level of skills and knowledge
attained that increase their individual productivity. There is an abundance of sophisticated
economics research in labor economics that demonstrates the individual or private benefits
to receiving additional years of education. In fact, it is the positive return to more schooling
that is typically the personal motivation for undertaking more years of education in the first
place (either that or parents that insist on university study no matter what!).
However, besides the private benefits of increased income received by individuals who
have themselves received more schooling, there also are social benefits that accrue to the
economy as a whole as the result of those many individual decisions to accumulate more
human capital. Such social benefits can include: new knowledge created by more educated
individuals that adds to the well-being of others through new products, new medicines, safer
production processes, and so on; more efficient workers who thus reduce the costs of produc-
tion and prices of products and services paid by all consumers; a more educated workforce
which may be more inclined toward democratic processes and tolerance of differences; and
more educated workers interact with other people, so that the level of productivity of all
workers rises through the synergistic effects that endogenous growth theories hypothesize,
thus increasing the incomes of all involved, not just those who individually sacrificed to
undertake additional schooling.
What this means is that the social benefits of education exceed the private benefits of educa-
tion as a result of the existence of positive external effects that accrue to society at large and to
individuals who did not undertake additional schooling themselves. We can write this as:

social benefits = private benefits + positive external effects of
additional schooling (12.5)

There is a large literature in economics dealing with this and similar situations where
there are benefits accruing to third parties, that is, when there are positive externalities to
some action, in this case, to undertake additional years of schooling. From the research, we
know that the choices of individuals as to the level of education to pursue will result in less
education on average being undertaken than is socially desirable. This is a classic example
of market failure in the presence of positive externalities that requires some sort of govern-
ment intervention to bring private and social benefits in line with one another for the overall
social good. In effect, government intervention is necessary to reach the most efficient level
of average education in society that maximizes the net social benefits from education.
Figure 12.2 illustrates this divergence between the privately chosen level of education and
that which would be socially optimal. The curve labeled MPC = MSC = S assumes that the
marginal private costs of education (MPC) to individuals are equal to the marginal social
410 The Process of Economic Development





E E Average years of
P S education

Figure 12.2 The private optimum and the social optimum level of education.

costs (MSC) of providing education, and that these can be viewed as the supply curve (S) of
education, measured in terms of years of schooling on the horizontal axis. In other words,
the S curve shows the costs of providing education at each level of schooling, irrespective of
who receives the education or how that level of education is paid for.
The cost of schooling and the benefits are measured on the vertical axis. The private market
demand curve for education by individuals is shown by the curve MPB, which measures
the marginal private benefits of additional years of schooling accruing as increased income
and other benefits to those individuals who actually undertake the schooling. If individuals
choose how much education to attain based on their own private decisions alone, the average
quantity of education which is optimal for them to choose is Ep, where the supply curve, S,
crosses the individual market demand curve, MPB. This is nothing more than the textbook
demand and supply equilibrium learned in first-year economics now applied to the specific
case of the (private) demand and supply curves for years of schooling.
As noted already, however, some of the benefits of schooling are received indirectly by
others who are not themselves actually “consuming” additional years of education. These
spin-off or third-party benefits can be in the form of higher incomes, improved product
quality, lower prices, new goods and services, increased social cohesiveness, a higher level
of technological development, and so on accruing to others beyond those who actually have
undertaken additional schooling. Thus the marginal social benefits (MSB) of education curve,
which can be thought of as society™s total demand curve for education, lies outside and to the
right of the MPB curve, the private demand curve for education.
Population, education, and human capital 411
The gap between the MPB and the MSB curve measures the value of the positive
externality to society of any particular level of schooling, that is, from statement 12.5 above,
it is equal to

positive externality = social benefits ’ private benefits (12.6)

at any particular level of education. This gap widens as more years of education are accumu-
lated, reflecting the presumption that learning-by-doing, association effects, and other indi-
rect gains from increased education levels generate even more beneficial effects to society
at higher versus lower average levels of education, as the research considered in Chapter 8


. 14
( 21)