In the past 50 years, family-planning programs have been heavily promoted across the developing world. A vast academic literature now tests both the intellectual rationale for these programs, as well as their impact on a wide range of demographic and economic outcomes. In recent years, the availability of new methods and new datasets from the developing world has intensified the academic research on these issues even though the support for family-programs themselves has diminished. This paper examines the economic and demographic literature on family planning programs and summarizes evidence of their impact on fertility as well as additional outcomes such as child mortality, investments in children’s human capital, the economic status of households and the macro-impacts on communities.
The goal is to provide policy-makers with an understanding of the strengths, limitations and points of agreement that emerge from this vast literature.
PRELIMINARY AND INCOMPLETE DRAFT
In the past 50 years, family planning (FP) programs have been heavily promoted across the developing world as a means to reduce fertility rates and promote economic development.
The central assumption behind such programs is that the decline in birth rates during the early stages of demographic transition can promote economic growth, reduce environmental pressures, reduce dependency ratios and strengthen a societies’ ability to invest in health and education (Coale, Hoover, and Press 1958). At the micro-level, it has been assumed that a decline in fertility would relieve women of the burden of repeated child-bearing and free up opportunities for them to increase schooling and participate in the labor-force.
A significant literature – shaped by economists and demographers – now tests these assumptions (Kelley and McGreevey 1994; Kelley 1995). Much of the literature however, remains either theoretical or focused on macro-correlations between variables such as fertility or population growth and indicators of development such as GDP growth or female education. The causal impact of declining fertility and/or the impacts of FP programs on fertility have proved to be difficult to find. One of the main challenges faced by researchers is that fertility decline is affected by a wide range of variables, including socioeconomic variables such as income, education (particularly female education) and female employment. Changes in these variables can affect the demand for FP, the structure of the programs, and their ultimate impact. There is also the issue of policy itself. FP programs are rarely rolled out randomly. Placement of programs in areas with distinct characteristics made it difficult to identify the precise policy driver of any observed change in behavior.
In recent years, the research has been enriched by the availability of new methods and new datasets from the developing world. This includes cross-sectional surveys such as the Demographic and Health Surveys (DHS), panel datasets such as the Family Life Surveys, and the use of random assignment evaluation methods that study causal relationships under careful scientific experimental structures. This paper examines this literature and summarizes evidence of the impact of FP programs on fertility as well as additional outcomes such as child mortality, investments in children’s human capital, the economic status of households and the macro-impacts on communities. We define an FP program as any organized effort to encourage couples to limit their family size, and space their births by using contraceptive information and services.
This includes legislative, regulatory, and programmatic efforts to supply contraceptives to a population as well as efforts to reduce the demand for children and/or increase the demand for contraception through information and/or social marketing campaigns. The paper is organized as follows: Section 2 provides a brief history of FP programs in the post WWII era and argues that FP programs have declined in priority after the ICPD conference in Cairo in 1994. Sections 3 and 4 provide an overview of two strands of the literature on FP programs: non-experimental studies that use cross-sectional or panel data to evaluate large-scale FP programs in states, countries or regions; and experimental studies that analyze random or pseudo-random pilot projects. Section 5 examines the literature on the cost-effectiveness of FP programs. Section 6 provides some perspectives that are likely to interest policy-makers.
PRELIMINARY AND INCOMPLETE DRAFT
2. Family planning programs: A short history
Family planning (FP) programs emerged after World War II. The world’s first major program was established in India in 1951 and was soon after followed by Pakistan, the Republic of Korea, and China. By 1975, about 74 developing countries had established them (Seltzer 2002; Cleland et al. 2006). Most programs fall into three general groups: (a) those that specifically aimed to curtail population growth through explicit policies such as promoting contraception and/or establishing incentives to have fewer children; (b) those that did not aim to curtail population growth, but promoted FP for other purposes; and (c) those with no explicit population policies but allowed outside donors to run programs that were mostly small in scale (Nortman and Hofstatter 1980; Nortman 1985). The first group was dominated by Asian countries, mainly East Asia and some South Asian countries (Mauldin, Berelson, and Sykes 1978; Lapham and Mauldin 1985; Mauldin and Ross 1991)1. In China and Vietnam for example, the governments formally announced in the early 1960s that couples should have no more than two or three children and began a wide-range of interventions that either directly or indirectly contributed towards this goal. Many governments provided citizens with incentives to meet these targets.
In China, couples with more than two children (or one child in some parts of China) were subjected to fines and penalties, though the enforcement of this program varies significantly (Short and Fengying 1998; Attane 2002). Permanent methods of contraception (mainly sterilization) were often provided for free.2 In Korea, Indonesia and Thailand, FP programs focused heavily on the expansion of usage of IUDs and other temporary methods in addition to permanent methods. In South Asia, the programs were less strong than in East Asia but large in scope. India for example, established a vast network of clinics that were to provide contraceptive services. In the 1960s, this was followed by a public health–based outreach program which emphasized education and awareness particularly in rural areas (Harkavy and Roy 1997). A common feature of almost all programs in this group was that they were generally led, funded and managed by domestic governments, and involved a broad range of ministries and mass organizations that focus on educating, promoting, and encouraging couples to use FP methods. Another common feature of programs in this group is that they were typically one component of broader development policies that aimed to increase access to health-care, education and industrialization.
These authors have developed quantitative measures of family planning program strength, or “effort” that are based on the number and quality of institutions that are involved with family planning programs. The measures of effort came from the belief that strong family planning programs must possess some essential features: (a) It should offer a full range of contraceptive methods and deliver them through several delivery systems, particularly in rural areas; (b) It should have a corps of full-time fieldworkers and educated the public about contraception; (c) Prominent leaders should issue frequent statements favoring the use of contraceptives; (d) The program should have a full-time director, placed well up in the government structure, and various ministries and private agencies should provide technical, logistical and financial assistance. More will be said about these criteria, and the studies that support them later in this paper.
In Vietnam, Bryant (1998: 246) writes that right before fertility declined, thousands of health workers were given basic training and sent to villages to promote use of mosquito nets, distribute locally made drugs, deliver babies, administer vaccinations, and carry out other standard primary health care functions.
PRELIMINARY AND INCOMPLETE DRAFT
The second group of countries was dominated by Latin America.3 Until about 1960, governments in this region remained strongly pro-natalist in their ideals (Mundigo 1996). This changed due to a concern about the high incidence of unsafe abortions in the mid-1960s. Abortions performed in unsanitary conditions by unqualified personnel were believed to contribute to maternal mortality and also resulted in large public expenditures as women with abortion-related complications sought care en masse from public hospitals (Mundigo 1996). To minimize disagreement with the Catholic Church however, FP programs in Latin America began as small private initiatives that were largely funded by international donors and NGOs. In most countries, particularly Brazil and Peru, these programs were ultimately incorporated into national public health programs. By the 1980s, countries in this group generally had broader goals than simply reducing fertility and/or the practice of unsafe abortion. They generally aimed at improving maternal and child health through greater birth spacing, access to pre- and post-natal care. Some Asian programs also fall into this category. Bangladesh is particularly noteworthy. Its national program, launched in 1976, aimed to provide women with a wide a range of contraceptive methods through home-visits by a network of locally recruited female-health care workers.
Sterilization was included in the package of options and in the first few years of the program, compensation was offered to those who chose the procedure (Cleland and Mauldin 1991). Yet the program remained largely voluntary and focused on maternal and child-health more broadly. A similar effort is seen in Iran, which launched its program in 1989. Free contraceptives were distributed through a network of village health workers, who also advised women on a broad range of maternal and child health issues. The third group of countries was almost entirely dominated by Sub-Saharan Africa. Some countries did establish programs early on. Kenya and Ghana for example, established FP programs in the late 1960s. Tanzania established a FP program in 1970. Senegal established an urban FP program in 1976 and a rural program in 1979. Much of Francophone Africa however, remained largely untouched by the wave of interest in FP programs throughout this period.
A 1920 French law that banned advertising and distribution of contraceptives continued to prevail. Across most of Africa, issues of population growth remained sensitive and highly politicized throughout the post-war period. Nigeria for example, adopted a national population council to study the issue of population growth but did not adopt any national policies to lower fertility (Caldwell and Caldwell, 1983). This was at least in part because census data that formed the basis of such decisions was regarded as too controversial. The results of the 1962 and 1973 census were actually nullified due to dispute and controversy over accuracy of the size of minority groups.
Even when they were adopted, African FP programs differed significantly from their counterparts in other countries in several key ways. First, the focus was almost entirely on temporary methods, since permanent methods were regarded as culturally unacceptable (J. C Caldwell and P. Caldwell 1987; J. C Caldwell and P. Caldwell 1988). The establishment of robust supply chains for temporary contraceptives
Only five Latin American countries fell in the first group – Mexico, Colombia, the Dominican Republic, El Salvador and Guatemala. Mexico is the largest among these. In 1974, access to family planning was declared as a constitutional right for all couples. In 1977, a national coordinating body was establish to expand the supply of contraception and a demographic target of population growth of no more than 2.5 percent per year by 1982 was declared. A wide variety of methods, including oral contraceptives as well as permanent sterilizations, were offered and the contraceptive prevalence rate doubled within a span of less than five years (Rodriguez-Barocio et al., 1980).
PRELIMINARY AND INCOMPLETE DRAFT
however, proved to be very challenging in the African context. The health-care infrastructure in this region was weak and burdened with a high demand for curative services. Moreover, most rural women resided far away from clinics or health centers (Caldwell and Caldwell 1992). Stringent eligibility criteria also made it difficult to reach women. In many cases, a woman’s access to contraception required the written consent of husband, proof of marital status or age, blood tests (for oral contraceptives), frequent follow-up visits, and non-evidence based requirements that she be menstruating at the time that she starts using certain methods such as IUDs or hormone-based systems (Campbell et al., 1996). Cultural preferences for high fertility often made women unwilling to be seen attending these clinics. Moreover, since consumers did not receive adequate information about contraceptives, side-effects were often misinterpreted and rumors were propagated. These factors combined to cause discouragement and discontinuation in the long-run (Campbell et al., 1996). Across Asia and Latin America, the spread of primary healthcare services, rapid increases in female schooling, the processes of socio-economic development and the use of marketing campaigns to promote awareness of FP programs may have alleviated some of these problems.
A second distinctive feature of African programs is that they were supported by a large number of international donors who rarely coordinated their actions with national governments or even between themselves. Since the weakness of domestic health infrastructure ruled out the establishment of “vertical” programs that packaged FP with primary health services, donors preferred to fund standalone programs that they could establish, manage and monitor themselves (Seltzer, 2002; Robinson and Ross, 2007; Mayhew, Walt, Lush and Cleland, 2005). The programs thus often remained small-scale. The goals used to evaluate the programs were often short-term in keeping with the demands of short budgetcycles. This approach stands in stark contrast to Asian and Latin American programs that were typically run by Ministries of Health and were backed by long-term budget commitments. Donor retreat
International interest in FP programs lost momentum in the early 1980s. The intellectual shift behind this is often referred to as “revisionist thinking” and refers to a retreat from Malthusian fears about the crippling effect of population growth on economic growth as well as the concern with the adequacy of supplies of food and natural resources (Kelley 1995; Kelley 2001). A wide range of factors fueled revisionist thought: the rapid pace of fertility decline in Asia, the success of the green revolution, the lack of convincing academic evidence for a negative relationship between population
growth and economic growth, etc. Economists emphasized that the long-run impact of population growth in economic development may not necessarily be negative. On the contrary, investments in human capital and innovation in growing populations can even have positive effects on growth and development outcomes (Simon and Lincoln 1977; Boserup 1981). Critics of FP programs used this literature to make the case that many FP programs in the developing world had been conceptualized and implemented with a false sense of urgency after World War II, without sufficient internal debate, deliberation and consensus (Kelley 1995; Kelley 2001).
Revisionist intellectual thought was also reinforced by the voices of NGOs in international policy. These groups highlighted examples of FP programs that had not gone well in parts of China, India, Indonesia, Mexico, Peru, etc (for a summary, see Seltzer, 2002: 62—70). In India for example, the controversial
PRELIMINARY AND INCOMPLETE DRAFT
HITTS model (Health Department operated, incentive-based, target-oriented, time-bound, and sterilization-focused) was so unpopular that it contributed to the collapse of the Indian government in 1977 and prompted India to launch a significant critique of FP programs in international policy circles (Harkavay and Roy, 2007).4 Feminists argued that women in particular, paid a high price for population policies, for they had often been viewed as passive “targets” who needed to become “acceptors” of contraception (Dixon-Mueller, 1993). They demanded that policies recognize women as key agents in the process of reproduction and must empower them – through education, information and access to health services (that include but are not limited to FP) – to have establish control over their bodies. The impact of the rights-based approach was visible at the Vienna Conference on Human Rights in 1993 where there was an explicit recognition of the importance of reproductive rights and the need for national and international development policies to be built around these rights.
The biggest shift however, occurred in Cairo at the International Conference on Population and Development in 1994. The definition of reproductive rights was took center-stage and included not only issues of reproductive decision-making, but sexual health and female empowerment more generally.6 Delegates explicitly called for dropping demographic and FP program targets in favor of a broader policy agenda that included a range of reproductive and sexual health measures. FP thus became embedded into a broader set of policy-goals. In the words of the UNFPA,
“[The ICPD Programme of Action] places human rights and well-being explicitly at the centre of all population and sustainable development activities. The Programme of Action moves discussion beyond population numbers and demographic targets: its premise is that development objectives — including early stabilization of population growth — can be achieved only by basing policies and programmes on the human rights, the needs and aspirations of individual women and men. Human-centred development -in the sense of investing in people generally, and particularly in health, education and building equity and equality between the sexes — is seen as a firm basis for sustained economic growth and sustainable development (UNFPA, 1995:9).” The representatives of 179 governments agreed on the need to ensure universal access to reproductive health services by the year 2015. These governments also agreed to increase spending on population
According to Harkavy and Roy (2007), the government’s Department of Family Planning estimated that more than 20 million births were averted between 1956 and 1975. Calculations based on the number of births averted concluded that the annual birthrate fell from about 42 live births per 1,000 population in 1960–61 to about 38 in 1970–71 and about 35 in 1974–75.
The critique of FP programs also came from health advocates who argued that despite the strides made in the safety of technologies such as oral pills and injectibles, safety issues remain. They did not agree that the benefits of choice outweighed the risks and argued that contraceptive safety needed greater attention (Seltzer, 2002). Other health advocates argued that FP had absorbed far too much policy attention and development assistance,
neglecting other critically important issues.
This includes the right not to be alienated from their sexual or reproductive capacity and bodily integrity through coerced sex or marriage, denial of access to birth control, sterilization without informed consent, freedom from unsafe contraceptive methods, from unwanted pregnancies or coerced child bearing, from unwanted medical attention.
PRELIMINARY AND INCOMPLETE DRAFT and related programs. The needed resources were estimated at $17 billion a year by 2000, climbing to nearly $22 billion by 2015.7
This agenda was however met with resistance by several groups. First, there were those who challenged the coupling of a gender ideology with issues of reproductive health. In Jordan for examples, elites felt that the proposals regarding reproductive health were acceptable, but proposals that aimed to reduce gender-based violence, deliver sex-education to adolescents, spread information about STIs and the promote gender equality conflicted with existing cultural norms and could not be implemented quickly (Luke and Watkins, 2002; Seltzer, 2002).
A second challenge to the Cairo agenda came from religious groups felt that the expanded definitions of reproductive health and reproductive choice tacitly included abortion and more controversial methods of fertility reduction. Even though the Cairo agenda was carefully worded to not support abortion in any circumstances, many people believed that the two issues were too deeply related to be separated in practice (Seltzer, 2002).8 In 2001, with support from the Vatican, the United States publicly opposed abortion, once again implemented the “gag rule” and thereby withdrew association with all organizations that offer women abortion services as a part of their general effort to expand reproductive choices for women.9,10
A final challenge came from the sheer breadth and language of the agenda itself. Some have argued that the focus on sexual health and reproductive rights was so broad that it simply failed to gain traction in parliaments and chambers of government across the developed world (Glasier et al. 2006; Fathalla et al. 2006). While Cairo advocates emphasized the importance of rights, donors were most interested in arguments that demonstrate a clear economic return on investment (Fathalla et al. 2006). The loss of focus also led to a fragmentation of academic and policy research. Many turned their attention to new competing priorities, such as HIV (Glasier et al. 2006; Fathalla et al. 2006, Blanc and Tsui, 2005). A visible sign of just how divisive the Cairo agenda was comes from the UN’s Millennium Development Goals, agreed to by nearly all nations in 2000.
The only goals that were related to reproductive health were the reductions in maternal and child mortality. Reproductive choices and reproductive rights were completely sidelined. In fact, the Cairo goal of universal access to reproductive health services was possibly the only goal that had been agreed to through a series of global conferences that did not make the final list of eight Millennium Development Goals (United Nations, 2000). In 2007, these goals were modified to include “universal access to reproductive health” by 2015. Progress was to be measured by www.unfpa.org
The Program of Action stated that “in no case should abortion be promoted as a method of family planning,” and elsewhere that “in [such] circumstances in which abortion is not against the law, such abortion should be [made] safe.” This was intended to be a compromise between those who opposed abortion on all grounds and those governments and NGOs who permitted abortion in varying degrees. 9
This was one of President George W. Bush’s first acts in office in January, 2001. 10
A cap of $15 million was placed for foreign NGOs and multilateral organizations who could not certify that they will not support any abortion-related activities, even if they use their own funds for these activities four indicators: the contraceptive prevalence rate, the adolescent birth rate, antenatal care coverage, and the unmet need for FP (United Nations 2007; UNFPA 2011) Another sign of the damage from Cairo is seen in the international HIV policies. In the late 1990s, policymakers in the United States and indeed much of the world, were compelled to focus on the challenge of HIV. Rather than building services into FP programs however, donors chose to establish entirely new programs. One of the biggest examples is the establishment of President’s Emergency Plan for AIDS Relief (PEPFAR) under US President George W. Bush in 2003. This was one of the largest efforts in history to address the challenge of one disease. The only relationship between this program and FP was its attempt to encourage abstinence as a form of prevention of HIV.11 Some have argued that the focus on HIV and AIDS simply replaced the Cairo agenda, when they should have in fact simply reinforced and complemented it (Blanc and A. O Tsui 2005).
In summary, the global interest in FP programs has swung from extreme interest after World War II to disinterest at the turn of the century. The weakness of such programs is most pronounced in SubSaharan Africa. The rate of contraceptive prevalence remains only 26%, less than half of the world average, despite significant investments in treating sexually transmitted diseases such as HIV (WDI 2010). At the current time however, there appears to be a renewed interest in the role of FP, particularly in approaches that are broad-based, female-focused, voluntary and respectful of basic human rights. In the section ahead, we review the literature on the effectiveness and impact of FP programs with the goal of demonstrating that such programs can have impact on not just fertility but a variety of other aspects of women’s well-being.
3. What do we know about the impact of FP programs? Perspectives from the non-experimental approach
FP programs in the 1960s, 1970s and 1980s were mostly established in the absence of scientific evidence or agreement about program “best-practices”. This is mainly because detailed time-series data on economic as well as demographic variables was scarce at both the micro- and macro-level. Over time however, data was gathered and researchers began to test some of the fundamental assumptions underlying FP programs. The first wave of studies used a non-experimental approach, i.e. they evaluated the impact of FP programs involving using cross-sectional or panel data from a country, region, or set of regions to test the hypothesis that FP programs impacted contraceptive use or fertility. As more data became available, and FP programs were rolled out, these studies were updated and expanded. The research continues to evolve today, even though the interest in FP programs has declined among policy-makers.
Most of the research in this area faces two key challenges. The first is the challenge of measurement. Given that FP programs are heterogeneous in goals, quality of services, delivery systems and implementation strategies, researchers must construct a measure of program strength and not rely 11
The following conditions were imposed in the United States Leadership against HIV/AIDS, Tuberculosis, and Malaria Act of 2003: 20% of funds were to be spent on prevention, 15% to be spent on palliative care, and starting in 2006, at least 55% were to be spent on treatment, at least 10% be spent on orphans and vulnerable children, and at least 33% of appropriated prevention funds be spent on abstinence-until-marriage programs.
Amin, S. 1995. Socioeconomic change and the demand for children in rural Bangladesh. Population Council, Research Division. Angeles, G., Jason Dietrich, David Guilkey, Dominic Mancini, Thomas Mroz, Amy Tsui, and Feng Yu Zhang. 2001. A Meta-analysis of the Impact of Family Planning Programs on Fertility Preferences, Contraceptive Method Choice and Fertility. Measure Evaluation Project. Carolina Population Center.
Attane, I. 2002. China’s family planning policy: An overview of its past and future. Studies in Family Planning 33, no. 1: 103–113. Bauman, K. E. 1997. The effectiveness of family planning programs evaluated with true experimental designs. American journal of public health 87, no. 4: 666.
Becker, G. S, and H. G Lewis. 1973. On the Interaction between the Quantity and Quality of Children. The Journal of Political Economy 81, no. 2: 279–288.
Bertrand, J. T, M. E McBride, N. Mangani, N. C Baughman, and M. Kinuani. 1993. Community-based distribution of contraceptives in Zaire. International Family Planning Perspectives: 84–91. Binka, F. N, A. Nazzar, and J. F Phillips. 1995. The Navrongo community health and family planning project. Studies in Family Planning 26, no. 3: 121–139.
Birdsall, N., and D. T Jamison. 1983. Income and other factors influencing fertility in China. Population and Development Review 9, no. 4: 651–675.
Blanc, A. K, and A. O Tsui. 2005. The dilemma of past success: Insiders’ views on the future of the international family planning movement. Studies in Family Planning 36, no. 4: 263–276.
Bongaarts, J. 1987. Does family planning reduce infant mortality rates? Population and Development Review 13, no. 2: 323–334. ———. 1994. The impact of population policies: Comment. Population and Development Review 20, no. 3: 616–620. Bongaarts, J., T. Büttner, G. Heilig, and F. Pelletier. 2008.
Has the HIV epidemic peaked? Population and Development Review 34, no. 2: 199–224.
Bongaarts, J., and S. Greenhalgh. 1985. An alternative to the one-child policy in China. Population and Development Review 11, no. 4: 585–617.
Bongaarts, J., W. P Mauldin, and J. F Phillips. 1990. The demographic impact of family planning programs. Studies in Family Planning 21, no. 6: 299–310.
Boserup, E. 1981. Population and technological change: A study of long-term trends. University of Chicago Press. Caldwell, J. C, and P. Caldwell. 1987. The cultural context of high fertility in sub-Saharan Africa. Population and development review 13, no. 3: 409–437.
———. 1988. Is the Asian family planning program model suited to Africa? Studies in Family Planning 19, no. 1: 19–28. Caldwell, J. C, Barkat-e Khuda, Bruce Caldwell, Indrani Pieris, and Pat Caldwell. 1999. The Bangladesh fertility decline: an interpretation. Population and Development Review 25, no. 1: 67–84. Chowdhury, M. E, R. Botlero, M. Koblinsky, S. K Saha, G. Dieltiens, and C. Ronsmans. 2007. Determinants of reduction in maternal