Page 303

Nature Displaced:
Human Population Trends and Projections and their Meanings.

Richard P. Cincotta
Robert Engelman
Population Action International, 1120 19th Street, NW Ste. 550, Washington, DC 20036-3678

Unlike the great species extinctions of Earth's past, the one occurring today is less an episode than a process, whose full results will not be known for hundreds of years. Between the linked human-induced phenomena of global climate change and biodiversity loss, the planet could be passing into the equivalent of an entirely new geological epoch in just a few human generations. Or it could be that biodiversity loss will amount to little more than a manageable depletion, incurring regrettable scientific and economic losses but leaving the basic services provided by most major ecosystems largely intact.

The size and distribution of human population over the near and distant future will surely be a dominant factor in determining whether the loss of biodiversity that the world faces turns into merely a source of wistful regret for future generations, a planetary catastrophe, or something in between. Population growth enlarges the scale and extent of the human enterprise and hence inflates the likelihood that human activities will push native nonhuman populations and biotic communities past critical thresholds of tolerance and renewal.

Demands for housing (Mason 1996), food energy and arable land (Bongaarts 1994; Engelman and LeRoy 1995; Smil 1994), freshwater (Engelman and LeRoy 1993; Falkenmark and Widstrand 1992), and industrially fixed nitrogen (Howarth and others 1996; Smil 1991; Vitousek and others 1997) appear more sensitive to the growth of human population than to the growth of per capita income or even to recent changes in technological efficiency. Habitat conversion, historically the greatest threat to biodiversity, has been driven by these very demands—by housing needs, pressures to expand and intensify agriculture, and the quest to harness

Page 304

additional freshwater supplies. Climate change, the demise of commercial fish populations and coastal reefs, widespread soil degradation, and the re-emergence of infectious disease also reflect the strong influence of population dynamics and take a growing toll on biodiversity. These global changes threaten ecosystem function and raise the risk of future extinction. It thus makes sense to consider the prospects for human population growth.

In this article, we consider those prospects by examining the United Nations (UN) population projections—both how and what they project. The methods and meaning of UN projections are poorly understood by scientists outside the field of demography. And the recent misuses of the projections in the press have confounded the public.

Despite widespread perceptions to the contrary, there is nothing inevitable about most future human population growth. Our species now numbers 6 billion and is growing at a pace of just over 80 million per year. More than 95% of this growth is occurring in countries of the developing world. Most demographers expect human population at least to approach 8 billion in the next half-century. Beyond that expectation, however, no one can be certain that world population will ever rise to greater levels. There is equal uncertainty that population will stop growing at any particular time in the not too distant future.

We can be certain, however, that today women in most developing countries desire fewer children than their mothers or even their older sisters sought or had (Westoff 1991). Over the last 30 years, that trend, when supplemented with access to modern contraception and the information needed to use it safely and effectively, clearly has resulted in lower rates of childbearing in countries with traditionally high fertility (Robey and others 1994). In the future, changes could occur even more rapidly. Decisions made today will have an enormous influence on the demographic future. These decisions are likely to be among the most important that we can make to conserve as much as possible of the planet's remaining biodiversity.

Humanity's Place in Nature

Few scientists outside the field of ecology are aware of how ecologically unprecedented is the scale of human numbers—not just present numbers, but also those of the last several millennia. No other mammal of comparable body weight has ever attained anywhere near such abundance. By manipulating the qualities and quantities of other species through agriculture, Homo sapiens broke through the energy and nutrient constraints that limited it as a hunter-gatherer.

Statistical models relating the adult body weight of mammals to their observed abundance (Peters 1983, p 166–7) predict that the equilibrium density of mammalian species in their home ranges will vary according to the following relationships: DC = 15 W - 1.16 for carnivores and DH = 103 W - 0.93 for herbivores (grazers and browsers), where D is animal density expressed in individuals per square kilometer, and W is the adult body weight in kilograms. For a carnivorous mammal or herbivore the size of Homo sapiens (roughly 65 kg), these relationships predict 0.12 individual/km2, and 2.1 individuals/km2, respectively. The natural

Page 305

availability of preagriculture human diets, however, fell between carnivorous and herbivorous diets. In fact, we are still largely grain-, fruit-, and tuber-eating with a predilection for meat. A liberal estimate of the average density that our species would likely have attained without agriculture is around 1 individual/km2—similar to the density at which hunter-gatherers and nomadic pastoralists lived until relatively recently.

If preagriculture humans at that density were to exploit every square kilometer of Earth's habitable terrestrial surface, about 130 million square kilometers (Hannah and others 1994), the world would support roughly 130 million people. According to one estimate, world population surpassed that number during the early years of the Roman Empire (Biraben 1979 reprinted in Livi-Bacci 1992; Cohen 1995). The United States alone surpassed it just before World War II (US Bureau of the Census 1995).

Demographics Then and Now

We know with reasonable certainty that Homo sapiens has expanded in numbers from at most a few tens of millions in prehistory to nearly 6 billion at the close of the 20th century. Most of these billions arrived in the 20th century, as the march of technology (especially in sanitation, immunization, and agriculture) allowed, for the first time, the vast majority of babies born to survive to become parents themselves. Some of the most rapid population growth during the 19th century occurred in the United States, where annual increases, roughly 2.5–3%, were as high then as in sub-Saharan Africa today. The consistently high 19th-century growth rates are a major reason that the United States is today the third most populous country in the world.

The result of the victory over infant and child death is evident in every region and major city. The planet sustains nearly half its humans in urban areas. Roughly three of every five people live in Asia. Each of the other major world regions is home to several hundred million people, but the populations of the continents are growing at markedly different paces: Europe, with about 730 million people, at a mere 0.2% per year (UN 1996a); North America (mostly the United States and Canada), with 300 million, at about 1.0% annually; Asia, with 3.5 billion, at about 1.5% per year; and the Latin American and Caribbean region, with about 485 million, at about 1.7% per year. Standing apart from the rest of the world demographically is Africa, with 708 million, where population growth has continued for decades at nearly 3% per year, falling slightly now to 2.7%. The average of all uneven rates of growth worldwide is equivalent to that of Asia, or about 1.5% per year.

Despite the ever-larger population base, world population growth is gradually slowing. The annual rate peaked at 2.1% in the late 1960s and has drifted down since. When a growth rate decreases, however, growth itself continues until the rate reaches zero. And substantial growth continues decades after fertility descends to replacement levels (slightly more than two children per woman) or even dips below. That effect, known as population momentum, is due to the long lag time between birth and reproductive maturity that characterizes our species. The

Page 306

lag allows past growth to continue to augment the absolute size of the reproductive segment of the population (women roughly 15–49 years old), thus supporting high numbers of births despite a decrease in fertility to replacement level.

As world population increases, more modest rates of growth can add larger annual increments to the population base. That has occurred although the highest rates of global population growth, estimated to have occurred around 1970, saw only about 72 million people added to world population each year. Current lower rates of growth are adding more than 80 million people per year. The global annual growth increment itself has declined since 1988 and could continue to decline—although by how much and for how long is unknown. A previous temporary decline during the middle 1970s, reflecting devastating effects of famine and political upheaval on the age structure of China's huge population (NRC 1984), illustrates how uncertain demographic projections can be.

During the 1970s and 1980s, human fertility in industrialized countries, which was already near replacement levels, declined once more. Nearly all the European countries fell below the roughly two-children-per-couple average that, in the absence of immigration, is necessary to replace each generation with the one that follows. The meaning of that trend for future population is potentially enormous.

Throughout the developing world, couples desire smaller families and later childbirths and they increasingly have the means to achieve the family size they seek. Several good examples can be gleaned from East Asia and Southeast Asia. During the middle 1960s, South Korea, Taiwan, Singapore, Thailand, and the former Hong Kong Territory began effective programs to lower infant mortality, establish easy access to family-planning services (ADB 1997; Tsui 1996), and increase primary-school enrollments and educational attainment (ADB 1997; Birdsall and Sabot 1993; Birdsall and others 1996; UNDP 1996; World Bank 1993). Thirty years later, average fertility in each of these Asian states is below two children per woman (the US average).

Other developing countries—including Mexico (3.1 children per woman), Brazil (2.4), Indonesia (2.9), Tunisia (3.3), and Sri Lanka (2.2)—are also experiencing downward trends in fertility (UN 1996b). A recent analysis of regional patterns of demographic change (Bongaarts and Watkins 1996) suggests that the first country in each developing region to begin its transition to lower fertility was endowed with relatively high indicators of social and economic progress, as measured by the UN's Human Development Index (UNDP 1996). In each case, however, fertility decline spread to nearby countries—probably via transfers of expertise, experience, and information at the government and local levels—despite the neighbors' lower scores for economic and social development.

In most developed countries, where there is access to affordable, effective contraception and safe abortion, women are more likely to have the number of children they want than are women in developing countries. Where these circumstances prevail and where childbearing and rearing are expensive or constrain economic mobility, total fertility consistently remains below the replacement level of slightly more than two children per woman (Potts 1996).

The other great trend shaping world population, aside from changes in fertility, is rapid mortality decline—or rapid increase in average life expectancy. Life

Page 307

expectancy began its climb in middle-18th-century Europe (figure 1). By the late 20th century, people in all corners of the world had longer life expectancies. The dominant influences are at both ends of the age spectrum: smaller proportions of children are dying in the first few years of life, and larger proportions of adults are surviving to old age.

Demographers assume that the mortality decline will continue, placing some further upward pressure on the pace of population growth. Falling mortality, however, could moderate worldwide as additional improvements in health care and nutrition become more difficult to achieve. In eastern Europe, mortality has actually risen in recent years; and in sub-Saharan Africa, the AIDS pandemic is reversing recent progress in infant mortality (US Bureau of the Census 1994). Both trends and the growing specter of emerging infectious diseases (Olshansky and others 1997) raise questions about the strength of the UN's assumption of continued mortality decline well into the 21st century.

The Project of Projecting

In projecting an image of the future, the challenge for demographers is to understand the complex and uneven trends in fertility, mortality, and migration and to consider to what extent they are likely to continue and—perhaps most critical—at what levels they might end. Given the hodgepodge of modern demographic trends, all that can be said with certainty about future trends and end points is that we cannot be certain. The UN Population Division, which produces the most widely cited tables of international population information, has addressed such uncertainty by computing every 2 years a three-piece set of population projections. The most recent series, published in 1996, projects populations for each of the UN's 185 member countries to 2050 (see UN 1996b).

Figure 1
Life expectancy in three developed countries (1750–2000) and three developing
countries (1950–2000). European life expectancies for years before 1950 from tables
compiled by Livi-Bacci (1992), citing various authors who have analyzed
historical records. Data from 1950 and beyond from current UN tables (UN 1996b).

Page 308

But scientists and journalists should take note: the UN projections are not statistically predictive. They are not estimates calculated from models of underlying behavioral relationships, nor are they the extrapolated curves with which biologists are most familiar. For that reason, the projections tend to be poorly understood and commonly misused.

The three pieces making up the UN's set of projections are its low, medium, and high variants generated for each country. Each variant differs from the other two in just a couple of key assumptions—its fertility end point and the path of fertility to that end point (see country examples in figure 2). When plugged into a model that generates births and eliminates the dead from each age group (and adds immigrants and subtracts emigrants where necessary), each variant traces a different population trajectory through the future.

To generate the medium variant's fertility curve, UN demographers use assessments of each country's situation and progress to make an educated guess of when each country will achieve replacement-level fertility. In each case, for projection purposes, this date is assumed to fall before midcentury. A fertility trajectory is then created that allows national fertility to fall—or, where needed, to rise—smoothly to its replacement-level end point. Once fertility arrives at this point, it is assumed to stay there indefinitely. By completing this exercise for fertility of each country (using standard mortality assumptions) and migration, adding all national populations for each year computed, the UN arrives at a continuous medium variant trajectory for world population.

The 1996 UN medium variant projects a global population of about 9.4 billion people around the middle of the 21st century, compared with the known 2.5 billion in 1950 and the 5.9 billion in 1998. If extended beyond 2050, as the UN does in its long-range projections (UN 1992; also see Haub and Yinger 1992; McNicoll 1992), population then grows fairly slowly, stabilizing at around 12 billion early in the 22nd century. The medium variant, however, is only one element of the projections.

To generate the other elements, the low and high variants, the UN adheres to the same model used to generate the medium variant but adjusts the fertility end point and the path of fertility to that point. In the low variant—a lower bound for plausible scenarios—each country's fertility end point is reset to achieve 1.6 children per woman before 2050 and held constant thereafter. To fix an upper bound of plausibility, the high variant applies the same schedule to settle at 2.6 children per woman.

Those are not error limits. Instead, the low and high variants are distinct, but extreme, scenarios of demographic change applied to every country. The low variant mimics the behavior of many European and several East Asian populations that over the last 2 decades have dipped below replacement fertility (1.2–1.9 children per woman). The high variant mimics a number of Central American and South American countries that have momentarily stabilized at levels somewhat higher than replacement (Haub and Yinger 1992). For example, total fertility of both Uruguay and Argentina has fluctuated erratically below 3.5 children per woman for at least 50 years without ever having reached replacement levels. In

Page 309

Figure 2
Fertility curves for three countries (UN 1996b)—past data and projected high-, medium-, and low-variant
scenarios and corresponding national population projections (UN 1996c).

Page 310

Costa Rica and Chile, fertility declined rapidly during the 1970s but stalled at similar levels.

By generating low and high variants, the UN projections present an envelope of plausibility, suggesting that a range of population futures is possible. Those two scenarios project a 2050 world population between 7.9 billion and 11.9 billion (figure 3).

A Separate Demographic Reality

The demographic experience of the world suggests that total fertility is dynamic and highly responsive to the circumstances of women and couples. The UN series of projections, however, must necessarily remain mechanical and thus reproducible every 2 years. Perhaps the most mechanical feature is the UN's assumption of a stable fertility end point for each variant. But even the paths drawn to those end points often appear inconsistent with past data.

Figure 3
Past and projected world population from UN estimates and
projections (1996c) and annual increment of world population
growth (annual change in growth) derived from these data and
projections. The trough in the world-population growth increment
that began in the middle 1970s was caused by irregularities
in China's growth increment. China's irregular growth
was due to an age structure shaped by high mortality
(NRC 1984) and low fertility (Coale and Li 1987) that occurred
in the wake of famine and political upheaval during the
Great Leap Forward from 1958 to 1960.

Page 311

Three examples illustrate these points (see figure 2). In the case of Nigeria, where there is still little evidence of substantial demographic change, all three fertility variants seem highly speculative and contrived. For Japan's projected fertility, at least two of the variants seem difficult to reconcile with past trends. In the case of Colombia, however, high-, medium-, and low-variant fertility curves all seem similarly consistent with past data.

Although there are good reasons to expect fertility decline to continue where families are typically large, there is no particular reason to assume that fertility rates will settle between 2.0 and 2.1 or at 2.6 or 1.6 children per woman. In fact, Sweden, Luxembourg, the United Kingdom, and France—each below replacement-level fertility today—have been there before (UN 1996b; Livi-Bacci 1992), bobbing back up above replacement-level during national baby booms and moving downward during political and economic turmoil.

Recently, the UN low-variant population projection has been used by several analysts and journalists (Buchanan 1997; Eberstadt 1997; Wattenberg 1997) as evidence that UN demographers are predicting alarming declines in global population beginning as “soon” as 2040. That is either a gross misunderstanding or a misuse of the projections. The low variant does, in fact, trace a downward path after that date. But eventually it must, by its very nature. With each national population in the world fixed forever at 1.6 children per woman—about a half-child below the replacement level—there is ultimately nowhere for the calculated population to go but down. The high variant, just as artificially, forces the trajectory upward, and the medium variant ultimately forces stability. Clearly, it makes little sense to use one variant without reference to the others.

There is no good reason to assume that the below-replacement-level fertility experienced in some industrialized countries today will be sustained long enough to lead to a substantial net population decline in the long run. Fertility rates might well rise again if the direct costs or opportunity costs of childrearing decline or if larger families regain social approval. Nor does it make sense to assume that below-replacement-level fertility returns to and stabilizes exactly at replacement-level fertility. Realistically speaking, we do not know.

In practice, most journalists and analysts take the UN's “medium variant,” or middle trajectory, to be the most probable one, whether for national, regional, or global population figures. It is often expressed inaccurately as the “expected” population future. That hardly makes the medium projection the “most likely” scenario within the wide range of plausible paths described by the high and low variants. True, neither the high nor the low extreme could be properly considered as “likely”—they are extremes, after all—but neither is there any special center of gravity midway between them. In fact, the medium projection uses a reasonable, repeatable method that cuts a path through a future without surprises, one in which demographic change is gradual and limited.

A relative absence of demographic surprise, however, has not always been the rule. Until the 1950s, demographers most often underestimated population growth. The largest cause of failure among early population projections was that their authors missed the fact that mortality was falling at an increasing rate in the developing world. The country-by-country triumphs of sanitation, clean water

Page 312

supply, antibiotics, and vaccinations were the big surprise. Of less impact, but still important, were increases in fertility in the industrialized countries after World War II—the so-called baby boom—that were difficult to foretell.

Strictly speaking, no population growth—not even tomorrow's—is really certain. Obviously, in the unlikely event that a nuclear war breaks out or a comet hits, all demographic bets are off. But leaving aside those unlikely possibilities, it is sobering to remind ourselves that infectious disease, war, and economic disruption still strongly influence population dynamics of individual nations and could do so more forcefully in the future. Clearly, words like inevitable and certain overstate the case.

More important, such language lulls observers into a conviction that no action in the present can influence the near demographic future. With much of the developing world exposed to television and computers in the span of a mere decade or two, a revolution in fertility patterns cannot be ruled out either. The likelihood of such changes is discounted in projections, perhaps reasonably, but such assumptions receive no discussion when the projection results are released to the public.

Nature's Place in Humanity

Humanity is pushing our planet across a series of important environmental thresholds at a time when our institutions—even in democratic societies—seem disinclined to take such threats seriously. This is the case whether the need is to secure the future or to help those whose well-being is most threatened today (Cincotta and Engelman 1997). What nonhuman genetic endowment shall we strive to preserve for future generations? Although the question is largely ethical and biological, we can be sure that demographics and economics will ultimately provide history with much or most of the answer (Morowitz 1991).

Among the complex factors that drive these changes in our ecology, human population growth is arguably the most easily addressed. Ten years ago, that statement would have seemed absurd. Five years ago, it might have been considered bravado. Now demographers tell us that the cessation of human population growth is within reach during the 21st century. We stand on that century's door-step.

How, then, should scientists view and represent the prospects for world population? Certainly not in terms of any inevitable figure. In peering into the future, it is useful to consider the UN population projections—the entire range described by the variants, not just medium variants—as a reasonably sound basis for describing a demographic future without substantial surprises.

We must loosen the grip that the medium projections have on the limited attention of policy-makers and the public. We need at least to bring attention to the range of growth suggested by the low and high projections for the next century and beyond. And, despite its necessarily artificial quality, we should hold forward the realistic hope offered by the low variant. Population growth might well slow further within the next few decades. Population size might peak before adding more than 2 billion people to our current numbers. And the world might

Page 313

someday experience a degree of population decrease before attaining relative long-term stability.

If the scenario becomes real, it will not be the product of “population control” or coercive government family-size targets. Rather, such a world will grow out of consistently pursued development initiatives that focus largely on the capacity of women to manage their own lives, especially their reproductive options. Such initiatives slow population growth while serving more immediate human needs. And in slowing and eventually easing to a halt the growth of human population, such a strategy can help to ensure that nature and its myriad ecosystems and species do not recede forever from their rightful place on the planet.

References

ADB [Asian Development Bank]. 1997. Emerging Asia. Manila Phillipines: Asian Development Bank.

Biraben J–N. 1979. Essai sur l'Evolution du Nombre des Hommes. Population (Paris) 34(1):13–25.

Birdsall N, Bruns B, Sabot RH. 1996. Education in Brazil: playing a bad hand badly. In: Birdsall N, Sabot RH (ed). Opportunity foregone. Washington DC: Inter-American Development Bank. p 7–47.

Birdsall N, Sabot RH. 1993. Virtuous circles: human capital growth and equity in East Asia. World Bank Working Paper, Policy Research Department. Washington DC: World Bank.

Bongaarts J. 1994. Can the growing human population feed itself? Sci Amer 270(3):35–42.

Bongaarts J, Watkins SC. 1996. Social interactions and contemporary fertility transitions. Pop Devel Rev 22(4):639–82.

Buchanan P. 1997. Demographic decline of west. New York Times, 17 Nov. p A23.

Cincotta RP, Engelman R. 1997. Economics and rapid change: the influence of population growth. PAI Occas Pap No 3. Washington DC: PAI.

Coale AJ, Li CS. 1987. Basic data on fertility in the provinces of China, 1940–1982. Pap East-West Pop Inst, No 104. Honolulu HI: East-West Center Program on Population.

Cohen JE. 1995. How many people can the earth support? New York NY: WW Norton.

Eberstadt N. 1997. The population implosion. The Wall Street Journal, 16 Oct. p A22.

Engelman R, LeRoy P. 1993. Sustaining water: population and the future of renewable water supplies. Washington DC: PAI.

Engelman R, LeRoy P. 1995. Conserving land: population and sustainable food production. Washington DC: PAI.

Falkenmark M, Widstrand C. 1992. Population and water resources: a delicate balance. Washington DC: Population Reference Bur.

Hannah L, Lohse D, Hutchinson C, Carr JL, Lankerani A. 1994. A preliminary inventory of human disturbance of world ecosystems. Ambio 23(4–5):246–50.

Haub C, Yinger N. 1992. The UN long-range population projections: what they tell us. Washington DC: Population Reference Bur.

Howarth R W, Billen G, Swaney D, Townsend A, Jaworski N, Lajtha K, Downing JA, Elmgren R, Caraco N, Jordan T, Berendse F, Freney J, Kudeyarov V, Murdoch P, Zhu Z-L. 1996. Regional nitrogen budgets and riverine N & P fluxes for the drainages to the North Atlantic Ocean: natural and human influences. Biogeochemistry 35:181–226.

Livi-Bacci M. 1992. A concise history of world population. Cambridge MA: Blackwell.

Mason A. 1996. Population, housing, and the economy. In: Ahlburg DA, Kelley AC, Mason KO (eds). The impact of population growth on well-being in developing countries. Berlin Germany: Springer, p 175–360.

McNicoll G. 1992. The United Nations' long-range population projections. Pop Devel Rev 18(2):333–40.

Morowitz HJ. 1991. Balancing species preservation and economic considerations. Science 253:752–4.

Page 314

NRC [National Research Council]. 1984. Rapid population change in China, 1952–1982. Report No 17. Washington DC: National Acad Pr.

Olshansky SJ, Carnes B, Rogers RG, Smith L. 1997. Infectious diseases—new and ancient threats to world health. Washington DC: Population Reference Bur.

Peter RH. 1983. The ecological implications of body size. Cambridge UK: Cambridge Univ Pr.

Potts M. 1997. Sex and the birth rate: human biology, demographic change, and access to fertility regulation methods. Pop Devel Rev 23(1):1–39.

Robey B, Rutstein SO, Morris L. 1993. Fertility decline in developing countries. Sci Amer. 269:60–7.

Smil V. 1991. Population growth and nitrogen: an exploration of a critical existential link. Pop Dev Rev 17(4):569–601.

Smil V. 1994. How many people can the Earth feed? Pop and Dev Rev 20(2): 255–92.

Tsui AO. 1996. Family planning programs in Asia: approaching a half-century of effort. Asia Pop Res Rep No 8. Honolulu HI: East-West Center Program on Population.

UN [United Nations]. 1992. Long-range world population projections: two centuries of population growth, 1950–2150. New York NY: UN Department of International Economic and Social Affairs.

UN [United Nations]. 1996a. World population 1996, wall chart. New York NY: UN Department for Economic and Social Information and Policy Analysis, Population Division.

UN [United Nations]. 1996b. World population prospects: the 1996 revision. New York NY: UN.

UN [United Nations]. 1996c. World population prospects: the 1996 revision, computer diskettes. New York NY: UN. (Note: the diskettes feature annual population data, rather than in 5-year intervals as appears in the printed publication).

UNDP [United Nations Development Programme]. 1996. Human development report, 1996. New York NY: UN.

US Bureau of the Census. 1995. Census of population and housing. CPH-2-1, Population and Housing Unit Counts, Table 16. Washington DC: US Dept of Commerce, (http://www.census.gov/population/www/ censusdata/ pop-hc.html). (last accessed : January 5, 1998)

US Bureau of the Census. 1994. The Impact of HIV/AIDS on world population. Washington DC: US Government Printing Office.

Vitousek PM, Aber JD, Howarth RW, Likens GE, Matson PA, Schindler DW, Schlesinger WH, Tilman D. 1997. Human alteration of the global nitrogen cycle: sources and consequences. Ecol Appl 7(3):737–50.

Wattenberg BJ. 1997. The population explosion is over. New York Times Magazine, 23 Nov. p 60–3.

Westoff CF. 1991. Reproductive preferences: a comparative view. Demographic and Health Surveys Comparative Studies No 3. Columbia MD: Institute for Resource Development/Macro Systems.

World Bank. 1993. The East Asian miracle: economic growth and public policy. Oxford UK: Oxford Univ Pr.