The equally widespread belief that an earlier, lower population peak would require coercive “population control” is also incorrect. Population growth rates and average family size worldwide have fallen by roughly half over the past four decades, as modern contraception has become more accessible and popular. The average number of children born to each woman worldwide is not much higher than replacement fertility, an average that would eventually end population growth. Yet more than 40 percent of all pregnancies are unintended, with higher proportions in developed than in developing countries.
As these figures suggest, it might be possible to end and then reverse human population growth through a strategy aimed at elevating women’s status and increasing access to contraceptive services, so that essentially all births result from intended pregnancies. Preliminary calculations based on conservative assumptions suggest that global fertility would immediately move slightly below replacement levels, putting world population on a path toward an early peak followed by gradual decline. The success of such a strategy would have many other benefits, such as reducing disability and deaths among mothers and their children and freeing more women to earn money and participate actively in social affairs.
There are many barriers to a global movement to assure that almost all births result from intended pregnancies. Foremost among them are the views of certain religious and political leaders and economic thinkers. Better public understanding of the benefits of universal intended childbearing is needed to counteract these obstacles and bring such a vision closer to reality.
- Even though most women of reproductive age now use contraception, we are far from a world in which all births result from intended pregnancies. Based on survey data, approximately 40 percent of pregnancies are unintended in developing countries, and 47 percent in developed ones.
- More than one in five births worldwide result from pregnancies women did not wish to occur.
- An estimated 215 million women in developing countries have an unmet need for family planning: they are sexually active, don’t want to become pregnant, and yet for various reasons—including lack of access—are not using contraception.
- If all births resulted from women actively intending to conceive, fertility would immediately fall slightly below the replacement level; world population would peak within a few decades and subsequently decline.
- Assuring that all women are fully in control of the timing and frequency of childbearing is not expensive. Religious, cultural, and political opposition to contraception or the possibility of population decline is the key obstacle to such assurance. More research and a public better educated about sexuality and reproduction could engender a global social movement that would make possible a world of intended pregnancies and births.
Those who ponder humanity’s future in the twenty-first century generally take at face value demographic projections suggesting that the world population will reach something like 9 billion around 2050 and will then stabilize at about that level.1 The widespread belief that this 30 percent increase from today’s 6.9 billion people is inevitable undermines consideration of the role of population size in climate change, water scarcity, biodiversity loss, rising energy prices, and food security. Contributing to this is the related view that efforts to prevent population growth would require coercive government policies that constrain couples from having the children and the family sizes they want. While some analysts are confident that the world can feed, house, and otherwise support 9 billion or more people, others are less certain, and voices of caution about population growth are heard more often than in the past.2 A logical application of the precautionary principle in the face of current environmental problems would suggest that humanity could more easily accomplish these feats in an environmentally sustainable manner with a smaller population.
In a joint statement in 1993, representatives of 58 national scientific academies stressed the complexities of the population-environment relationship but nonetheless concluded, “As human numbers increase, the potential for irreversible changes of far-reaching magnitude also increases. … In our judgment, humanity’s ability to deal successfully with its social, economic, and environmental problems will require the achievement of zero population growth within the lifetime of our children.”3 In 2005, the United Nations’ Millennium Ecosystem Assessment identified population growth as a principal indirect driver of environmental change, along with economic growth and technological evolution.4
In October 2010, a group of US and European climate and demographic researchers published findings from an integrated assessment model calculating the impact of various population scenarios on fossil-fuel carbon dioxide emissions over the coming century. If world population peaked at close to 8 billion rather than 9 billion, along the lines described in a low-fertility demographic projection published by the UN Population Division, the model predicted there would be a significant emissions savings: about 5.1 billion tons of carbon dioxide by 2050 and 18.7 billion tons by century’s end.5
What if we could prove wrong the popular conviction that a future with 9 billion people and a growing population is inevitable? Suppose we could demonstrate that world population size might peak earlier and at a lower level if government policies aimed not at reproductive coercion but at individual reproductive freedom? Suppose such policies aimed to help all women and girls prevent unwanted pregnancies and conceive only when they want to bear a child? This article presents new data on births resulting from women’s active intentions to become pregnant. The hypothesis it probes may appear counterintuitive: if, starting at any moment, all pregnancies in the world resulted from each woman’s intent to give birth, human population would immediately shift course away from growth toward decline within a few decades.
An Ethical Basis for Action to Slow Population Growth
What can societies that value democracy, self-determination, human rights, personal autonomy, and privacy do to include demographic change among strategies for environmental sustainability? An important answer may lie in a relatively untested set of principles adopted by almost all the world’s nations at a 1994 UN conference held in Cairo. The third of three once-a-decade governmental conferences on population and development, it produced a program of action that abandoned the strategy of “population control” by governments in favor of a focus on the health, rights, and well-being of women.6 An operating assumption of this program is that when women have access to the information and means that allow them to choose the timing of pregnancy, the intervals between births lengthen, average family size shrinks, and teen births become less frequent. All of these improve maternal and child survival and slow population growth.7
Experts disagree on how reproductive autonomy compares with other strategies in slowing that growth. Some assume economic growth is the most effective means, although birthrates rose along with prosperity in many countries after World War II and remain relatively high in several wealthy oil-exporting nations in which women have fewer rights and lower status than men.8 Moreover, some analysts argue that the arrow of causation operates more in the other direction, with low fertility stoking economic growth.9
There is a more robust and demonstrable correlation between female educational attainment and fertility. Worldwide, women with no schooling have an average of 4.5 children, while those who have spent at least a year or more in primary school have just three. Women who complete at least a year or two of secondary school have 1.9 children—well below replacement fertility rates. With one or two years of advanced education for women, average childbearing rates fall even further, to 1.7.10 On this basis alone, those interested in depressing population growth rates might want to focus on improving women’s educational attainment.
Questions remain about whether education alone can bring about declines in fertility without other supporting conditions, especially easy, affordable access to a range of contraceptive options. Similar uncertainties cloud understanding of exactly how improved child survival and the empowerment of women affect fertility. Improving both factors certainly contributes to later births and smaller families and is valuable regardless of its demographic impacts. But without clear data on the magnitude of these influences, interventions related to schooling, child survival, and women’s empowerment are rarely seen as core aspects of governmental population policy.
This brings us to family planning. Access to safe and reliable contraception has exploded since the mid-twentieth century. An estimated 55 percent of all heterosexually active women worldwide now use modern contraceptive methods, while an additional seven percent use less reliable traditional methods.11 As the use of birth control has spread, fertility has plummeted from a global average of five children per woman in 1950 to barely more than 2.5 today.1
Glossary of Terms
Total fertility rate refers to the average number of children a woman would bear over her lifetime if at each point in her reproductive age she had the number of live births typical of women at that age. Note that the total fertility rate differs from the population growth rate, which is the percentage by which a population grows each year, and from the birthrate, which is the number of live births each year per thousand people in the population. The global total fertility rate currently stands at 2.53 children per woman.
Replacement fertility rate refers to the total fertility rate in a population that, if held steady over time and absent net migration, would result in a nonchanging population. This rate is often mischaracterized as uniformly and precisely 2.1 children per woman, but not all children survive to reproductive age, and the proportion of those who do not varies over time and by population. For the world as a whole, with many low-income regions still experiencing high death rates among young people, the replacement fertility rate currently stands at 2.35 children per woman.28 Surprisingly, the gap between global total fertility and replacement fertility is now less than one-fifth of one birth.
Even achievement of global replacement fertility would not stop population growth for several decades, due to population momentum. This is the tendency of a population, influenced by its age structure, to continue its current growth dynamic even as fertility changes. Because there are so many young people of reproductive age in any population that has had above-replacement fertility for some time, for example, even low fertility can produce an overall number of births that statistically overwhelms deaths among the smaller cohorts of older individuals. It can take decades before subreplacement fertility actually halts growth. If total fertility falls well below replacement however, this momentum is weakened and a peak in population will come sooner, followed by a decline. These demographic phenomena are evident in Japan, with a total fertility rate of 1.3 children per woman and a population that has already peaked and is now slowly shrinking.16
While not necessarily sufficient to depress fertility on a population-wide basis, family planning is essential to the phenomenon. Women may begin sexual activity later in life and may resort to abortion to terminate unwanted pregnancies. But humanity’s average family size could not have plummeted simply because women had diplomas, contractual rights, or confidence that their children would survive. To have small families, heterosexually active women and their partners need safe and effective contraception—modern birth control.
Lessons from history suggest that women have sought and employed contraceptives since ancient times to avoid unwanted pregnancy when circumstances were inauspicious for the 15 to 18 years of parental commitment a new birth entails. Egyptian papyri that date back 4,000 years describe pessaries, ancient precursors to the diaphragm, made of acacia oil and crocodile dung.12 Literature from Asia to North America documents herbs used for centuries as emmenagogues, substances that induce immediate menstruation and hence expel recently fertilized eggs. In the Mediterranean, in the ages of ancient Greece and Rome, a booming trade in the contraceptive, or possibly abortifacient, silphium helped drive its source, a wild giant fennel, into extinction. And an ecclesiastical court record from 1319 preserves the personal account of a young widow in southwestern France who provided details of her use of an herbal contraception during an extended affair with a priest.13
We know, too, that women and their partners historically have moderated their reproduction in response to their external environments, natural and economic. (Until modern times, these were generally the same thing.) In eighteenth- and nineteenth-century Sweden, for example, birthrates neatly tracked the price of grain crops with a roughly nine-month delay.14 The Japanese population during the eighteenth-century Tokugawa shogunate declined during several decades of food scarcity—until a government propaganda campaign against infanticide (the dominant method of family-size control at the time) pushed fertility well above replacement levels in the nineteenth century, restoring demographic growth.15
Similar responses of fertility to external circumstances are evident today. The high cost of housing in Japan is prominent among the reasons offered by young people for delaying marriage and childbearing.16 In the United States, a two percent decline in the country’s birthrate in 2008 was attributed largely to the deterioration of the economy.17
Implications of Personal Fertility-Management Aspirations
History and recent fertility phenomena thus suggest the likelihood that the interest in safely and effectively managing the timing of pregnancy and childbirth may be nearly universal among women. Lack of education, affluence, and equality may simply be barriers—along with others related to patriarchal, pronatalist, and even medical cultural norms—to existing aspirations to avoid unwanted pregnancies.18
Data exist for the likely demographic impact of establishing conditions worldwide that would facilitate women’s choices about the timing of pregnancy. According to the Guttmacher Institute, a US reproductive health-care research organization, an estimated 215 million women in developing countries have an “unmet need for family planning.”19 This applies to women who are sexually active and express the desire to avoid pregnancy yet are not using contraception. Estimates of their number derive from demographic and health surveys conducted in certain developing countries every few years.20 Many women in developed countries may be in the same circumstances, but data are insufficient in most cases to suggest their numbers.
In early 2010, researchers with the Futures Group in Washington, DC, estimated the demographic impact of meeting unmet family-planning demand in 99 developing countries and one developed one. The researchers excluded China, on the assumption that government population policies aimed at limiting most families to a single child rule out births from unintended pregnancies. And they supplemented their country list with the United States, the world’s most populous developed county and one for which there is some data suggesting the magnitude of unmet need.21 Using accepted models for the impact of rising contraceptive prevalence on birthrates, the researchers concluded that satisfying unmet need for contraception in these 100 countries—with a cumulative 2005 population of 4.3 billion—would produce a population of 6.3 billion in 2050. Under the United Nations’ medium projection, the countries’ population would be 400 million higher, at 6.7 billion. Average global fertility at midcentury would be 1.65 children per woman, well below the population replacement fertility level—and would continue to fall.
This conclusion, if backed up by further research, is momentous. By implication, simply providing safe and effective contraceptive options to all sexually active women who do not want to become pregnant would end and then reverse world population growth. The effect is independent of any further fertility reductions that might occur as a result of greater educational attainment for women, improved child survival, women’s empowerment, and general economic advancement.
To some experts the idea that simply facilitating women’s childbearing intentions would end population growth, without significant demand creation for family planning through cultural shifts and other means, goes against survey findings from many African and some Asian countries. These findings suggest that in parts of these continents women’s average desired family size is as high as six or seven children.20 Wouldn’t facilitating these women’s childbearing intentions undermine any hope of ending world population growth? Not necessarily. For one thing, women expressing such high desired family sizes are at most a relatively small proportion of the world’s population (albeit significant in Africa’s). But the more important point is that a high desired family size can easily coexist with high levels of unintended pregnancy that, if prevented, would result in significantly lower birthrates than if not prevented.
The reason for this is not hard to understand: women’s individual reproductive decisions arrive at their desired family size, if at all, only cumulatively. Decisions about the desirability of pregnancy are made singly, in individual acts of sexual intercourse in which conception is possible. Whatever one’s hopes for an eventual number of children, pregnancy decisions occur in the context of current personal, economic, and social circumstances. Desired family size can be compared to house size and the number of cars owned. We may wish to have a large house and many cars, but our circumstances may not allow for us to have either without endangering our finances and well-being. We decide moment by moment whether working toward that goal makes sense for us. So it is with reproductive intentions; every step of a woman and her partner’s reproductive lives is governed by their immediate circumstances.
It seems likely that even in countries where women respond in health surveys that they desire six or seven children, they would end up with fewer, possibly many fewer, if at each step of their reproductive lives they were able to choose precisely when to become pregnant. In some developed countries with low fertility, women express a desire to have two children yet have closer to one on average. With the right partner, the right job, the right apartment, and the right economic and social-support systems, a woman in Japan, for example, might have the two children she desires. But with options to prevent or terminate pregnancies, many Japanese women have one child or none; the national average is 1.3.16
All of this suggests the value of developing and testing the hypothesis that meeting the needs of women and their partners for personal control of pregnancy could lead to the end of population growth. Physician and reproductive specialist Malcolm Potts has found that in all countries where women can choose from a range of contraceptive options, backed by access to safe and legal abortion services, total fertility rates are at or below replacement fertility levels.22
If these findings can be borne out consistently by additional research, those who worry about the impact of global population growth on environmental and social sustainability might usefully advocate for worldwide universal access to family-planning services. The need for such access is enshrined in the second target of the fifth UN Millennium Development Goal, which calls for developing countries to “achieve, by 2015, universal access to reproductive health.”23 This concept embraces more than family planning, including a holistic state of sexual and reproductive well-being that encompasses maternal and child health, prevention of AIDS and other sexually transmitted infections, access to safe abortion services (where these are legal), and post-abortion care.
A Thought Experiment with Data
The Futures Group study has not yet gained the widespread attention its findings merit. Among the reasons for this may be that the concept of “unmet need” for contraception is not widely understood among the public, news media, and policymakers. Moreover, because of lack of data the study excluded not only China, with a fifth of the world’s population, but dozens of other developing countries—and all the world’s industrialized countries other than the United States.
Newly available data on unintended pregnancy in many countries, assembled by the Guttmacher Institute, support an alternative research approach to the question of the demographic impacts of births that result from pregnancies women never sought or wanted to have. These data, based on a range of surveys worldwide, provide the basis for beginning to answer an intriguing and valuable question: What would happen to world population growth if every pregnancy worldwide, starting tomorrow, were the outcome of a woman’s active intention to become pregnant and bear and help raise a child? If no pregnancies were unintended, in other words, how many births would there be compared to current births, and how would this new birthrate affect the future of human population?
Averaged over the 73 countries for which data exist, and comprising 83 percent of the world’s births, just under ten percent of births result from pregnancies occurring among women who never wanted to have another child. Even under the most conservative scenario—extrapolated globally, with all births from pregnancies that are merely mistimed considered equivalent to births from intended pregnancies—a hypothetical world population in which women only become pregnant when they want to would reduce today’s global total fertility rate to 2.29 births per woman. That figure is slightly below today’s global replacement fertility rate—placing world population on a direct path toward future decline, albeit at a very slow pace given population momentum (and assuming neither future fertility decline nor improvement in mortality among young people). Under the less conservative assumption that one-quarter of births from mistimed pregnancies are equivalent to unwanted pregnancies, the total fertility rate sinks lower, to 2.22 births per woman—resulting in a somewhat faster track toward a human population peak, even with no future fertility decline. These calculations are, at best, first-order analyses of the impact on world population growth of an idealized scenario in which all births are the outcomes from intended pregnancies. As noted, they do not take into account the possibility that global fertility would continue its decline once all births resulted from intended pregnancies. More survey research and data on pregnancy intention among individual women in all countries would be needed to make a more robust determination of demographic impacts.
But the essence of research on this question remains hopeful—and little known: a successful global effort that assured all women the capacity to decide for themselves whether and when to become pregnant would also place world population on a path toward a reasonably imminent peak followed by slow demographic decrease. Additional efforts to see that women have the educational, economic, legal, and political opportunities they deserve would accelerate this transition.
Toward a World of Intended Pregnancies and Wanted Children
Given the feasibility of such a transition, why isn’t it happening today? Why aren’t higher proportions of births the result of intended pregnancies? And what might we do to overcome the obstacles and actually bring that world about?
Popular as it is with women and couples, contraception remains a deeply sensitive issue for much of the public. Vehemently opposed by the Catholic Church and regarded with suspicion by many other Christian, Islamic, and even some Jewish religious leaders, open advocacy for contraceptive availability and use inevitably risks stoking religious opposition. Influence of the Catholic Church hierarchy has blocked efforts in the Philippines, for example, to include access to modern contraception in the country’s government health system.24 Opposition from the Holy See, which has permanent observer status within the UN system, led to silence on reproductive health in the UN Millennium Development Goals when they were forged in 2000—even though representatives of the world’s governments had pledged to achieve universal access to reproductive health by 2015 at the UN conference in Cairo in 1994.25 Only in 2007 was language aiming at that reproductive health access target added to Millennium Development Goal number five. Seven years of opportunities to achieve the target had been squandered.
Perhaps more destructive than religious opposition is a relative denigration in most cultures of concerns that lie principally in the sphere of women. Access to contraception is clearly one such concern, since women bear the babies and undergo most of the risks to life and health associated with reproduction. At least since the rise of agricultural, urban, and hierarchical societies, male interests in reproduction have differed markedly from those of women. Men are often anxious to produce a multitude of future heirs, soldiers, laborers, farmers, and followers. Women tend to be strategically concerned with the survival and well-being of each of their children.13 These gender differences are anything but ironclad, and in many cultures the gender gap in attitudes has narrowed in recent decades, especially as women’s status has risen relative to men’s. In other cultures, however, the gap not only remains wide, it demands the subjugation of women, sex, and reproduction to male needs.
Beyond male reproductive dominance lies the conviction among neoclassical economists that endless economic growth is possible and that it requires endless population growth. Politicians often measure their self-worth based on the size of their electorates. They happily side with economists on the idea that endless economic and demographic growth is both possible and desirable.
With all these factors in play, it is not surprising that the world’s governments are nowhere close to allocating the resources the Cairo conference had estimated would be needed for all women in developing countries to have reasonable access to decent family-planning services. This was roughly $18 billion for the year 2010, a third of which was to be contributed by industrialized-country governments (the 1993 dollars in the UN document are here converted to current dollars).6 Despite that commitment—and an increase in the population of reproductive-age people in developing countries, from 2.3 to 2.9 billion—actual expenditures from these governments on international family-planning assistance fell from $723 million in 1995 to $338 million in 2007.26,27 Assistance has changed little since the latter year.
A global social movement is needed to pressure policymakers and influential cultural and thought leaders to reverse this dismal trend. Raising $9 billion a year from wealthy governments that currently spend just a few hundred million on international family-planning assistance shouldn’t be as difficult as it is. As much money is allocated for a few days worth of military activities worldwide. A comparable or greater amount probably would be needed to assure that the vast majority of pregnancies in wealthy countries are intentional, but this sum has never been estimated. Significant investments in all countries in education on sexuality and reproduction are also needed, but what these should be is unknown as well.
Still, the point undoubtedly still holds: a world in which almost all births result from intended conceptions would not be prohibitively expensive or difficult, aside from cultural barriers, to bring about. Yet due to contraception’s sensitivity—complicated by a history pockmarked with episodes of contraceptive coercion in China, India, Peru, and a few other countries—environmentalists and advocates for women’s rights and health have never succeeded in forging an activist alliance capable of raising the modest sums needed for all to have access to family planning.
Several elements are needed if a global social movement to promote family planning and intentional pregnancy is ever to have its own birth. One is more research about the likely population and environmental outcomes of a world of fully intended pregnancies—and the policies, programs, and costs that could lead to such a world. Another is agreement that any such policies and programs must be based on reproductive rights rather than on coercion, and therefore on the intentions of women and their partners rather than on those of anyone else. And a third is the creativity to shape—or the courage to stand up to—the religious, economic, and other cultural forces that promote population growth and oppose the gender and reproductive health conditions that undermine it.
There is nothing fated about a world of 9 billion people—in 2050, or ever. While true control of population is beyond our aspirations and capacities, policy choices are available that will nudge our numbers closer to environmentally and socially sustainable levels. The choices are rooted in human development and human rights, specifically the right of all, and most directly of women, to decide for themselves when it is the right time to bring a new child into the world.
- Demographic data other than those on pregnancy intentions are taken from World Population Prospects: The 2008 Revision Population Database. UN Population Division (online). http://www.un.org/esa/population/publications/wpp2008/wpp2008_highlights....
- Facing the consequences. The Economist 85–88 (November 27, 2010).
- “Science summit” on world population: A joint statement by 58 of the world’s scientific academies. Population and Development Review 20, 233–238 (2004).
- Reid, WVR et al. Ecosystems and Human Well-Being: Synthesis. Millennium Ecosystem Assessment (Island Press, Washington, DC, 2005).
- O’Neill, BC et al. Global demographic trends and future carbon emissions. Proceedings of the National Academy of Sciences 107, 17521–17526 (2010).
- Report of the ICPD. UN document A/CONF.171/13 (online) (1994). www.un.org/popin/icpd/conference/offeng/poa.html.
- The Health Benefits of Family Planning (World Health Organization, Geneva, 1995).
- Sutton, M. The wrong numbers: The perils of ignoring demography in East Asia. Asia Pacific Bulletin 85 (2010).
- Longman, P. Environmental consequences of low fertility rates. Newgeography.com blog (online) (2010). www.newgeography.com/content/001824-environmental-consequences-low-ferti....
- Lutz, W. Personal communication as reported in Engelman, R. Population and sustainability. Scientific American Earth 3.0, 22–29 (summer 2009).
- Haub, C. 2010 World Population Data Sheet (Population Reference Bureau, Washington, DC, 2010).
- Riddle, JM. Contraception and Abortion from the Ancient World to the Renaissance (Harvard University Press, Cambridge, MA, 1992).
- Engelman, R. More: Population, Nature, and What Women Want (Island Press, Washington, DC, 2008).
- Bengtsson, T & Dribe, M. Deliberate control in a natural fertility population: Southern Sweden, 1766–1864. Demography 43, 727–746 (2006).
- Harris, M & Ross, E. Death, Sex, and Fertility: Population Regulation in Preindustrial and Developing Societies (Columbia University Press, New York, 1987).
- Westley, SB, Choe, MK & Retherford, RD. Very Low Fertility in Asia: Is There a Problem? Can It Be Solved? Asia Pacific Issues 94 (East-West Center, Honolulu, 2010).
- Stein, R. U.S. birthrate drops 2 percent in 2008. Washington Post (April 7, 2010).
- Campbell, M, Sahin-Hodoglugil, NN & Potts, M. Barriers to fertility regulation: A review of the literature. Studies in Family Planning 37, 87–98 (2006).
- Singh, S, Darroch, JE, Ashford, LS & Vlassoff, M. Adding It Up: The Costs and Benefits of Investing in Family Planning and Maternal and Newborn Health (Guttmacher Institute, New York, 2009).
- Demographic and Health Surveys, performed by ICF Macro for the US Agency for International Development (online).www.measuredhs.com.
- Moreland, S, Smith, E & Sharma, S. World Population Prospects and Unmet Need for Family Planning (Futures Group, Washington, DC, 2010).
- Potts, M. Sex and the birth rate: Human biology, demographic change, and access to fertility-regulation methods. Population and Development Review 23, 1–39 (1997).
- UN Department of Economic and Social Affairs. The Millennium Development Goals Report 2010 (United Nations, New York, 2010).
- Philippines: Battle over reproductive health bill intensifies. IRIN News Service (online) (November 17, 2010). www.irinnews.org/Report.aspx?ReportID=91108.
- Crossette, B. Reproductive health and the Millennium Development Goals: The missing link. Studies in Family Planning 36, 71–76 (March 2005).
- Engelman, R. Facing a Changing World: Women, Population and Climate Change. State of World Population 2009 (UN Population Fund, New York, 2009).
- Engelman, R. Population growth steady in recent years. The Worldwatch Institute. Vital Signs (online). http://vitalsigns.worldwatch.org/vs-trend/population-growth-steady-recen....
- Engelman, R & Leahy, E. Replacement Fertility: Not constant, Not 2.1, but Varying with the Survival of Girls and Young Women. PAI Working Paper 1(4) (Population Action International, Washington, DC, 2006) (online). www.populationaction.org/Publications/Working_Papers/Replacement_Fertili....