Catholic Medical Quarterly Volume 64(1) February 2014

Towards an understanding of sub-Saharan African fertility transition with particular reference to Kenya

Emeritus Professor Seamus Grimes,
Department of Geography,
National University of Ireland, Galway.

Abstract

The persistent relatively high levels of fertility and population growth in many countries in sub-Saharan Africa are considered by many policymakers as a major obstacle towards economic progress on the continent. This neomalthusian interpretation of demographic trends is, however, disputed by some scholars, who insist on the need to examine demographic change within the particular social and anthropological realities of the cultures involved, and also within the political and economic context, rather than focusing on the demographic variable in isolation. Many would argue that, in addition to the technical analysis of demographic data, interpretations of demographic trends are closely intertwined with philosophical values and can be subject to ideological influences particularly within the policy environment of the international population community. This paper will explore different perspectives on demographic change in the context of sub-Saharan Africa, paying particular attention to the case of Kenya. It will seek to understand the relatively high levels of fertility, mortality and population growth in sub-Saharan Africa, despite the many decades of population programmes focusing on raising levels of contraceptive prevalence. Having explored the different philosophical perspectives, attention will be focused on empirical trends in relation to the dynamics of population change in the region.

Key words

philosophical values, ideological influences, neomalthusian perspectives, mortality, fertility, population growth, population programmes

Introduction

Ever since the rapid, though relatively recent improvements in levels of life expectancy in the less developed regions of the world, there has been a widespread concern, particularly in the developed world with the ensuing rapid growth in population associated with relatively high levels of fertility and more recently with the on-going impact of population momentum (Eberstadt, 2001). Unlike the more developed regions of the world, which had already progressed through the final stage of demographic transition to low levels of mortality and to below replacement fertility in most cases, the poorer regions and particularly sub-Saharan Africa continued to be characterised by relatively high, though declining, fertility and mortality and population growth. Although that generalised picture has begun to change in more recent years, with an increasing number of less developed countries moving in the direction of lower fertility levels, albeit with continuing population growth because of the high proportion of women of child-bearing age, the policy response towards interpreting the significance of continuing high fertility levels has been strongly influenced by the neomalthusian ideology of the international population community, which views population growth as a barrier to economic development in poor countries, and is strongly committed to raising contraceptive prevalence (the proportion of women aged 15-45 currently using contraception) as a means to reduce fertility levels (Cleland et al., 2010). Despite various critiques of neomalthusian thinking in relation to demographic change and economic development, such thinking remains dominant in the population policy environment (Furedi, 1997). Despite the fact that a considerable volume of demographic research remains closely allied to and to some extent compromised by involvement in policy-oriented work, some population researchers have questioned the appropriateness and the scientific basis of isolating the demographic variable as the primary determinant of economic progress (Furedi, 1997, Gould 2009). The continued preoccupation with higher rates of population growth in the poorest regions of the world is in stark contrast with a growing awareness of the increasing significance in the developed world and beyond of below replacement fertility, declining population rates and ageing populations (Eberstadt, 2001). This has given rise to possible accusations of double standards, with the European Union pushing for fertility reduction in less developed countries while providing incentives at home to increase fertility levels (The Economist, 2012).

This paper will question the neomalthusian interpretation of population growth in less developed regions and argue in favour of the need to examine demographic change within the overall context of the cultures involved and the on-going desire to have relatively larger families than in more developed regions. It will question the dominant view of international population agencies which have consistently presented people in poor countries as being unthinkingly responsible for what they see as ‘excess fertility’. It will argue that this predominantly negative interpretation of family size in poor countries is to some extent related to poor quality policy formulation based on questionable demographic analysis which has been influenced by ideological considerations (Gould, 1995). Such thinking is reflected in the ingrained conviction, widely diffused in the richer North that population growth in the South is problematic and needs to be controlled, partly in fear of high rates of immigration from the South. The other ingrained conviction in the North is that Western fertility control technologies are a major element of progress and that their usage reflects even a ‘moral’ responsibility which each person must exercise in relation to the environment.

This view of human fertility is derived from utilitarian forms of thinking related to the growing secularisation of society and a highly individualistic evaluation of the human person. It has been accompanied by major changes in social mores reflected in increasing levels of marriage breakdown, divorce, and more widespread abortion. Contraceptive technology, which has facilitated the separation of procreation and sexual activity, is generally viewed as acceptable. It is also widely accepted in the North and particularly promoted by the international population community that in order to improve the lives of people in poor countries this view of human sexuality must become widely adopted in these regions, and that having facilitated an increase in life expectancy, population agencies must now help bring about a reduction in fertility levels. Much of this thinking has evolved within an unquestioning environment among many population scholars (Greenhalgh, 1996; Connelly, 2008). Little attention has been given to the need to understand and appreciate the rationale behind the decisions made by people in poor countries about their family size, which are frequently assumed to result from an absence of choice. Because patterns of fertility are not just the result of biological processes, but also involve social, cultural and political processes, some scholars are arguing that there are solid grounds for rethinking our notions of fertility and call for a more embodied analysis which include a wider range of realities which are of significance to particular cultures (Underhill-Sem, 2009, 8). Kalipeni and Mbugua (2005), for example, emphasise the reverential regard shared by the large majority of Africans, including the well-educated and those resident in urban areas, for supernatural forces which influence their behaviour, and suggest that an understanding of these influences can help develop more effective policies for the prevention of HIV/AIDS. Having sought to explain some ideological influences in population policy, this paper will proceed to examine in some detail the empirical demographic trends in sub-Saharan Africa and particularly Kenya, and will seek to interpret those trends in a manner that is more sensitive to the local populations.

Deconstructing the neomalthusian ideology

The more recent and contemporary versions of neomalthusian thinking have presented population control as a poverty-reduction measure, a ‘development policy’, an instrument of family planning, a precondition for improving the position of women and a necessary measure to save the environment (Furedi , 1997). Rather than population growth per se, Furedi argues that the central preoccupation of the Malthusian movement has been a fear that the wrong kind of people tend to have the highest fertility rates. This preoccupation with controlling other people’s fertility, particularly in the poorer regions of the world, has been a key driver underlying population programmes from the 1950s. From the mid-1960s onwards, the existence of demand for fewer births was taken for granted within the international population policy debate and unfortunately, for the next quarter of a century, population policy in the developing world became synonymous with family planning programmes (Demeny, 2011).

This underlying assumption has been more recently translated into the concept of the ‘unmet need’ for the means for controlling fertility, a concept which abstracts individuals from their social and cultural context and seeks to generalise the experience of societies where a small family size has become the norm to those with a preference for larger families (Furedi, 1997). ‘Unmet need for contraception’ is defined in a World Bank briefing note as ‘the proportion of currently married women who do not want any more children but are not using any form of family planning or currently married women who want to postpone their next birth for two years but are not using any form of family planning’ (Mills et al., 2010). This includes women who are currently pregnant, women who are breast-feeding, and women who find it difficult to become pregnant. This note acknowledges limitations of this concept and how it sometimes is incorrectly interpreted as evidence of lack of access to a source of contraceptive supplies. It acknowledges that among the many reasons why women do not use contraception even where it is readily available and of good quality are cultural and religious objections, objections from a spouse, or fear of side effects. Despite the fact that this concept of unmet need has been consistently exploited by the international population community as a justification for promoting population control programmes, it has been severely criticised as a meaningful indicator, with mainly an advocacy rather than an analytical value. Similar indicators such as ‘readiness, willingness and ability’ to use family planning, based on the Demographic and Health Surveys, have been used to measure progress or stagnation of the spread of contraception, although researchers acknowledge that such indicators are overly simplistic representations of complex issues (Cleland et al., 2010).

Methodological issues associated with Demographic and Health Survey (DHS) are referred to by Westoff and Cross (2006) in their examination of factors that may underlie the stalling of fertility transition in Kenya, noting the inconsistency of answers to questions, with many respondents stating they did not want their last child but that they did want more children. Possible explanations for such inconsistencies may be that the design of the questionnaire is unsuited to the situation of those being surveyed, with women struggling to explain circumstances that the questionnaire never envisaged (Timaeus and Moultrie, 2008). A remarkable study in predominantly Muslim rural Gambia in 1993 explained that the main objective of the small percentage of Gambian women who used Western contraception at that time was to achieve the longstanding goal of a two-year-minimum birth interval in order to ensure the survival of many children (Bledsoe et al., 1994). The methodological sensitivity adopted by the researchers in this study helped to reveal that the typical western assumptions about contraception use did not apply in the circumstances of these women, who were primarily concerned about the potential of fine-tuning birth intervals, while also having concerns about the possible detrimental effects of long-term contraceptive use.

Paul Demeny, long-time editor of Population and Development Review, which is published by the strongly neomalthusian Population Council, ridiculed the concept of unmet need when he asked: are we trying to suggest that ‘two billion people in the past 30 years were added to the world’s population because their parents were too stupid to figure out what to do’ (The Economist, 28 May, 1994). In comparison to the need for food, water, medical care and fuel, Pritchett (1994, 31) argued that the need for contraception was very small in poor countries. In criticizing the methodology of surveys which purport to measure ‘unmet need’, he suggests that this measure reflects not just women who want contraception but also ‘those women who require motivation to want what they are presumed to need, suggesting a usage consistent with a very broad or very paternalistic definition of ‘need’. He also suggests that a reduction in the focus of family planning programmes on population growth will allow more attention to be paid to other important areas such as child and maternal mortality, the timing of first births, and the prevention of sexually transmitted disease. In a similar vein Eberstadt (1997) argues for greater humility on the part of those pushing for fertility reduction in seeking to explain the current trend towards widespread below replacement fertility, because of the absence of broad, obvious, and identifiable socioeconomic thresholds or common preconditions for such decline. Despite the growing awareness among researchers that fertility levels are influenced by a wide range of factors, many countries continue to pursue population policies as if family planning were the only thing that mattered (Casterline and Sinding, 2000).

Despite the acknowledged failure of the Indian family planning programme, Mohan (2004) claims that neo-liberal prescriptions and ideals about population control continue to be applied with the same vigour despite the so-called ‘reproductive health’ approach to family planning. The widespread and uncritical acceptance in the population literature of these programmes needs to be re-examined in the context of major health and social issues such as the growing problem of distorted sex ratios and rapid ageing, particularly in Asia, and the widespread diffusion of a sterilization culture (Grimes, 1998; Connelly, 2008; Hvistendahl, 2011). Lest one is under any illusion that the promotion of the unmet need concept has disappeared into the background, the recent summit on family planning on July 11th, 2012 in London organized by the British Government and the Gates Foundation has made it clear that this is not the case (The Economist, July 14th 2012). This summit won promises of $4.6 billion from donors and developing countries to provide modern contraception (coils, pills, injectables, implants and condoms) to an extra 120 million women by 2020. It is useful to consider this level of funding in the context of total development aid to Africa, which amounted to $47.9 billion in 2010 (OECD 2012). It is interesting to note that within the social sector of aid, which constituted 39.1% of the total in 2010, population and reproductive health which made up 13% of the social sector in 2009, was the only sector experiencing consistent growth since 1997. The proportion of the social sector for health fell from almost 16% in 2005 to less than 12% in 2009. It is also useful to realize that population programmes comprised 35.3% of total aid from the US in 2010 compared to only 7.3% on health. The US was the largest contributor to aid in 2010, contributing $7.7bn or 16% of the total. In 1984 USAID decided to adopt an integrated approach, piggybacking HIV/AIDS programmes on existing family planning programmes, a decision that has been heavily criticized (Mosher, 2008).

Population in context

This isolation of the population variable from its social and cultural context has been used as a justification for the implementation of population programmes, based on the argument that such programmes are effective regardless of the context (Furedi, 1997). This has been strongly refuted by Pritchett (1994) who argues that the contraceptive approach towards fertility control has been a failure. Some African specialists have, somewhat controversially, argued that persistently high fertility levels in sub-Saharan Africa were cultural, with Caldwell and Caldwell (1990a, 199) suggesting that ‘unlike many traditional societies those in sub-Saharan Africa generally had no concept of having enough children’. In a separate paper, these authors argue that high fertility in Africa is associated with social and family patterns that have emerged over millennia, and that the lack of understanding of these patterns can ‘wreak havoc’ on the assumptions of western demographers (Caldwell and Caldwell, 1990b, 119). A more recent paper acknowledges some unique aspects of Africa in that it has been primarily rural with high fertility supported by pronatalist institutions such as patrilineal descent, patrilocal residence, inheritance and succession practices and hierarchical relations that have remained unchanged for generations (Makinwa-Adebusoye, 2001). Rather than being immutable, however, the author provides substantial evidence of a marked decline in fertility in Africa in the past three decades, suggesting the need for a paradigm shift in our thinking about fertility transition in the continent. She notes, however, that the ideation of family size preference is a concept that has been slow to be adopted in Africa and that the recent stalling of fertility decline and low contraceptive prevalence may be a rational response to pervasive poverty and insecurity, particularly about the survival of children. The obliviousness of policymakers to cultural norms and practices relating to child bearing, and the lack of recognition of significant differences within the population are identified as reasons for the failure of Nigeria’s population policy, which was influenced by outside organisations (Obono, 2003).
Population growth in poorer countries today have been much higher than historically in Europe, where population growth was ‘controlled’ by variations in the marriage age and the proportion of people unmarried, unlike the situation in Africa where the marriage age has been traditionally under 18 with fewer than 1% of women remaining unmarried (Bengtsson and Gunnarsson, 1994). Among the other factors they identify for the apparent lack of connection between fertility patterns and economics are the high level of insecurity regarding property rights, inequality of opportunity, and institutional structures protected by the state which deprives large sections of the population from participating in a process of sustained economic development. They suggest that in high-risk or chaotic societies particularly in Africa, that the security problem has been solved mainly on the social side by large families. Gould (2009) denies that the demographic factor is dominant in explaining economic stagnation in many African countries and emphasizes the need to consider factors such as capital, governance, and lower levels of technological knowledge and practice. Eberstadt (1997) also concludes that demographic forces may only be of secondary importance in overall economic performance, noting the successful prospering of nations despite apparent ‘population problems’, while Furedi (1997) argues that the focus on population policies can be an admission of failure by the state in developing countries to deal with the central problems of social organization.

Gould (1995) criticises ideological influences in Kenya’s population policies, which he suggests have overly influenced an approach that favours the contraception hypothesis toward fertility reduction over the childhood mortality and economic change hypotheses. He argues that Kenyan data on childhood mortality and total fertility rates suggest that the fertility transition in Kenya was triggered by improvements in childhood survival and that rather than focusing on a policy of family planning and contraceptive provision, the policy presumption might be expected to allocate resources as a first priority to childhood and maternal health programmes. He argues that data collection in the Contraceptive Prevalence Survey and both the Demographic and Health Surveys are designed to examine the validity of the policies rather than provide insights into the dynamics of population change, and suggests that analysts need to be aware of the biases in such surveys and to be sensitive to both the political and ideological contexts of African population analysis.

Despite the continued focus within the international population community on seeking ways to overcome resistance within the less developed world to fertility reduction, Eberstadt (1997) points out that in the absence of viable predictive theory for long-term forecasting, population science has no reliable framework for predicting future trends, and that demographers have been unable to forecast the onset of fertility decline or the trajectory of that decline in developing countries. An example of this is the case of South Africa where fertility transition began about 1965, followed by Kenya from the 1980s, which was not recognized by the demographic literature until the 1990s (Caldwell and Caldwell, 2002). In agreeing that there is no good theory with predictive power to explain how specific cultural, social, economic and political conditions affect fertility, Khan and Lutz (2007, 9) suggest the need for greater humility in relation the ability of demographers to anticipate major social changes. Noting that the great majority of the world’s population with below-replacement fertility (less than 2.1 children per woman) live in low-income countries, Eberstadt (2010, 55) argues that there are ‘few socioeconomic preconditions for rapid and pronounced fertility decline.’ On the other hand, Demeny (2011) has suggested a number of conditions associated with development which may influence fertility decline, including rising costs of education and health care, access of women to opportunities in the labor market, effectively enforcing school attendance, making child labour illegal, effective guarantees of property rights and the development of pension schemes.  

Empirical trends: placing Kenya in context

Having explored the ideological influences of neomalthusian thinking on population policies and also attempts to place the demographic component within its broader social and cultural context, attention will be turned towards placing Kenya’s recent demographic evolution within the broader context of sub-Saharan Africa. The continuing preoccupation with high fertility and rapid population growth has left the international population community poorly prepared to respond to the increasing spread of below replacement fertility, not only in the developed world, but increasingly in less developed regions, due to the aging of the world’s population and mortality crises associated with HIV/AIDS (Eberstadt, 2001). The historical experience reveals that a sudden drop in fertility rates is usually followed by decades of population growth because of ‘population momentum’ resulting from a high proportion of women in the reproductive ages in the population. While some demographers have argue that population momentum could account for half of the developing world’s population growth in this century, Blue and Espenshade (2011) suggest that current trends suggest a substantial global population contraction in the second half of this century because of declining momentum.

As we look towards 2025, and remember that many 20th-century population forecasts proved quite inaccurate, it is expected that more people may be living in Africa than in all today’s more developed countries, with sub-Saharan Africa having a median age of around 20 years. It is only in the poorest regions of the world such as sub-Saharan Africa that demography continues to be shaped by population growth, with life expectancy increasing and child mortality declining. One of the remarkable consequences of differential rates of fertility decline in different regions of the world is that between today and 2030, it is estimated that China will experience a decline of around 100 million or 30% in the number of people 15-29 years, while over a third of the growth in this young manpower of 70 million people will take place in sub-Saharan Africa (Eberstadt, 2010). Africa could potentially benefit from a demographic dividend because of its young and rapidly growing workforce and declining dependency ratio, assuming that effective policies are implemented. McKinsey (2012) estimate that there is a potential to create between 54 million and 72 million more stable wage-paying jobs in Africa by 2020, which would raise the share of wage-paying jobs to between 32 and 36%.

Table 1 Demographic indicators in selected sub-Saharan African countries

Country	Maternal Mortality Rate 2010	Infant Mortality Rate 2012	% change < 5 mortality since 2005	Life expectancy at birth 2012	Total Fertility Rate 2012
Senegal	370	56.4	-9.9	60.1	4.69
Rwanda	340	64	-9.6	58.4	4.81
Kenya	360	52.3	-8.4	63	3.98
Uganda	310	62.4	-6.8	53.4	6.65
Ghana	350	47.2	-6.3	61.4	3.39
Zambia	440	66.6	-5.6	52.5	5.9
Mozambique	490	78.9	-5.5	52	5.4
Ethiopia	350	77.1	-5.4	56.5	5.97
Tanzania	460	66.9	-5.3	53.1	4.02
Madagascar	240	51.4	-5.1	64	4.96
Nigeria	630	91.5	-4.8	52	5.38
Benin	350	61.5	-4.8	60.2	5.22
Niger	590	112.2	-4	53.8	7.52
Mali	540	111.3	-3.6	53	6.35
Malawi	460	81	-2.8	52.3	5.35
Guinea	790	61	-1.4	58.6	5.04
Zimbabwe	570	28.2	0.5	51.8	3.61
Lesotho	620	55	0.7	51.8	2.89
Namibia	200	45.6	1.8	52.1	2.41
Liberia	770	72.7	1.8	57.4	5.02
Average	461.5	67.16	-4.22	55.87	4.928

Source: CIA Factbook 2012; Demographic and Health Surveys since 2005

Table 1 provides data on infant and child mortality, life expectancy and total fertility (TFR) for 20 countries in sub-Saharan Africa in 2012. It also provides the rate of change in under five mortality since 2005. In relation to mortality levels, some ground is being regained after a period of considerable deterioration in life expectancy during the 1990s mainly because of the devastating impact of HIV/AIDS and also other major diseases such as malaria in the case of some countries, and because of genocide in the case of Rwanda. In relation to all indicators, there is considerable variability between countries, and of considerable variability is also found within countries between urban and rural regions and between different income groups. There were sharp declines in mortality throughout Africa into the 1980s and 1990s related to expansion of vaccination programmes supported by foreign donors (Gould, 2009). The average maternal mortality rate (MMR) in Table 1 is 461.5, ranging from the relatively low 200 for Namibia to 770 for Liberia and 790 for Guinea. Death in childbirth or around childbirth is a significant risk in many areas of Africa. An analysis of maternal mortality from 1980 to 2008 for 181 countries has revealed substantial decline, which would have been much greater but for the effect of the HIV epidemic in eastern and southern Africa (Hogan et al., 2010). The rates provided in this paper are 586 for sub-Saharan Africa central (down from 770 in 2000), 508 for sub-Saharan east (down from 776 in 2000), 381 for sub-Saharan southern (up from 373 in 2000) and 629 for sub-Saharan west (down from 742 in 2000). In the case of sub-Saharan east and southern, the rates in 1980 (707 and 242) were considerably lower than in 2000 (776 and 373), because of the impact of HIV/AIDS. In contrast to these high rates for Africa, both Europe and North America had an MMR of 7.0 in 2008. Despite these huge differences, Hogan et al (2010) argue that based on the evidence of their study there is good reason for optimism that substantial decreases are possible over a short period. Among the reasons suggested for the decline are lower fertility rates, improvements in income per head, maternal educational attainment and in coverage of skilled birth attendance. Because of the lack of the most basic postpartum care, childbirth and labour complications can have fatal consequences for both mother and baby, and thus one of the most urgent needs is the provision of neonatal units that are easily accessible for poor communities. It is generally accepted that close to 80% of these deaths could be avoided particularly with skilled assistance during childbirth.

The average figure for infant mortality in 2012 is 67.16 deaths per 1000 live births, ranging from the very low figure of 28.2 in Zimbabwe to almost four times that level in Niger and Mali (Table 1). Malaria mortality affects children mostly and most diseases in Africa including malaria affect people living in the poorest areas with little access to healthcare. One of the most heartening developments, however, in recent years has been the very significant and relatively rapid decline in under-five deaths. The most significant cause of death to sub-Saharan African children is not from AIDS but rather includes malaria, diarrhea, pneumonia, other infectious diseases and preterm birth complications (Zuberi and Thomas, 2012). While not typically presented as direct causes of death, stunting, severe wasting and vitamin deficiencies are related to infectious diseases and are a significant risk factor of under-five mortality (Black et al., 2010). The first 12 countries listed in Table 1 experienced falls of over 4.4% a year since 2005, with Senegal, Rwanda and Kenya having falls of more than 8% a year. What took 25 years to bring about in India has taken place in Rwanda and Senegal within a five-year period, providing some hope that infant mortality could be significantly reduced throughout Africa within a relatively short period (Demombynes and Trommlerova, 2012). Gains in infant mortality are seen as one element in a wider set of improvements in human welfare in 17 African countries, among which are more democratic and accountable government, the end of the debt crisis, new technologies creating opportunities for business and political accountability and a new generation of policymakers, activists and business leaders (Radelet, 2010). Although not yet widely appreciated, this huge drop in child mortality in Africa should be seen as a major step forward in the development process, and in terms of both epidemiological and demographic transitions.

Related to high mortality rates in Africa generally, and particularly to high maternal mortality has been the devastating epidemic of HIV/AIDS, which has significantly reduced gains in life expectancy in recent decades. HIV prevalence rates for pregnant women in sub-Saharan Africa are 10-15% compared with 0.15% in the US (Oster, 2005). Rates in Africa vary from under 1.0% in Madagascar, Senegal and The Gambia to above 20% in Botswana, Zimbabwe, South Africa, Swaziland and Lesotho. Rates of untreated Sexually Transmitted Infections (STIs) are estimated to be around 11.9% in Africa compared with 1.9% in the US and 2% in Western Europe, which suggests that HIV transmission rates are likely to be higher in sub-Saharan Africa than in the West. For many women living in poverty, using their sexuality can constitute a form of survival, thus poverty becomes both a cause as well as an effect of rising HIV/AIDS prevalence rates (Gould, 2005). Others have argued, however, that a combination of the prevalence of polygyny and various degrees of marital instability together with inadequate health systems have contributed to facilitating high rates of HIV infection (Caldwell and Caldwell, 1990b). It is important also to appreciate how high HIV/AIDS-related morbidity and mortality of family and waged labour have exacerbated problems of food security, thus increasing malnutrition and vulnerability to infection (Poku, 2004). Yet, Oster (2005) notes that recent improvements in data collection have resulted in revisions downwards in a number of countries, particularly Kenya from the previous UNAIDS figure of 15% to 6.7%. Rather than being based on the total population, the UN’s estimates were based on tests of pregnant women attending prenatal clinics, who would constitute the group with the highest rates of HIV infection.

Rather than the narrow focus of prevention programmes to date, Kalipeni and Mbugua (2005) argue that a more comprehensive approach such as those which have been successfully implemented in Senegal and Uganda is needed, which incorporate the traditional beliefs and values which have long influenced community life in sub-Saharan Africa. In those African countries reporting a reduction, there has also been a reduction in the number of men and women reporting more than one sex partner over the course of a year, and this may suggest that part of the failure of the global response of the AIDS bureaucracy to date has been based on the experience of the disease in the US and Europe (Hanley and De Irala, 2010).

A major corollary of reductions in mortality levels are the gains in life expectancy, which as can be seen in Table 1 have been quite modest to date, partly because of the deterioration in the 1980s and 1990s associated with HIV/AIDS in Africa. With an average of only 55.8 years, for the group of countries in Table 1, the range of values from a low of 51.8 for both Lesotho and Zimbabwe (with HIV prevalence of 24.03 and 18.36) to the slightly higher values of 60 years or more for Senegal, Kenya, Ghana, Madagascar and Benin. Apart from Kenya with 6.87, the remaining countries have HIV rates from 0.1 for Madagascar to 2.31 for Ghana.

The final column in Table 1 shows an average Total Fertility Rate (TFR) of 4.92 children per woman for this group of countries, ranging from the very low 2.41 for Namibia, a country with very low maternal mortality, and Lesotho with 2.89, which has high maternal mortality, to Niger with 7.52, Uganda with 6.65 and Mali with 6.35. TFRs in Kenya, Ghana and Zimbabwe are already less than four children per woman, indicating that fertility transition in sub-Saharan Africa is well under way. Makinwa-Adebusoye (2001) argues that the view that sub-Saharan Africa is somehow unique in its support for large family size is no longer be valid, with moderate fertility decline occurring during a period of severe economic recession. Gould (2009) points out that while the factors responsible for mortality decline are primarily structural and largely beyond the control of individuals, such as improvements in livelihood, income and health care, fertility is more related to human behavior and cultural choice.

Despite the significant reductions in fertility evident in Table 1, a recent study by Cleland et al (2010) was motivated by a concern that uncontrolled population growth in sub-Saharan Africa will hinder the attainment of development and health goals. The authors note a deep-seated resistance to the use of modern contraception, with the proportion of women wanting to postpone or cease child-bearing in 2004 at 47%, which differed little from the 46% in 1991. They also point out that the rate of change appeared to be happening faster in eastern Africa than in the west, but that the pace of increase in contraceptive use in eastern Africa had slowed from an annual increase of 2.7% in the 1990s to 1.45% thereafter. According to the 2011Uganda Demographic Health Survey (UDHS) 70% of currently married women 15-49 were not using contraception, and this is reflected in the relatively high TFR of 6.65. Among the main reasons given why users had discontinued use in the previous five years were health concerns or side effects (32%), followed by a desire to become pregnant (25%) and pregnancy (14%) (Uganda Bureau of Statistics, 2012).

Referring to the rapid acceptance of contraception in some African countries, notably Kenya, Zimbabwe, Botswana and South Africa, partly related to the fact that these countries were relatively prosperous in late colonial and the early independence period, Gould (2009) suggests that the persistence of high fertility in Africa is related to a strong pro-natalist culture, the dependence on land by many communities, the fact that children are more likely to be seen as hands to work rather than mouths to feed, and that newly born infants are welcomed regardless of their economic status. In seeking to explain earlier resistance to fertility change in sub-Saharan Africa, Caldwell and Caldwell (2002) suggest that apart from poor family planning programmes and the skepticism of local politicians about such programmes initially, important factors included communal ownership of land, the value of child labour enhanced by deference to the elders and strong traditional beliefs in fertility formed in eras of very high child mortality. In their view the key factors which have influenced the shift to lower fertility levels in the past 40 years have been economic policies since the mid-1980s, strongly influenced by IMF recommendations, which resulted in children becoming a greater economic burden to parents, and also the increasing tendency for children’s expenses to be borne by parents than being shared more widely across a range of relatives. In pointing out that the contraceptive prevalence rate in Kenya rose rapidly from an estimated 9% in 1979 to an estimated 33% in 1993, Gould (2005) highlights the assumption underlying the evolution of population policy that Kenya’s experience in fertility decline was no different from what happened elsewhere, and that the primary means for reducing fertility was raising the contraceptive prevalence rate. He notes, however, that despite implying a strong relationship between the rising contraceptive prevalence rate and declining fertility, the Fertility and Health Survey reports did not conclude that one had been the consequence of the other.

Kenya’s trajectory in maternal mortality since 1994 is revealing in that the current figure of 360, a major improvement on 590 in 1998, is only a small reduction, however, on the 1994 figure of 365, which suggests a significant deterioration associated with HIV/AIDS in the 1990s (Table 2). The average figure of 488 in 2008 masks the fact that rates of up to 1000 were found in some provinces. Data from Kenya’s Demographic and Health Surveys also reveal a reduction in childbearing among 15-19 year olds from 23% in 2003 to 18% in 2008/2009 (Kenya National Bureau of Statistics, 2010). A significant reduction occurred in Kenya’s infant mortality from 184 in 1948 to 66 in 1989, but again it rose to 77.3 in 1999. The much lower average of 52 in 2008 masks a range of values from 95 in Nyanza Province to 39 in the Eastern Province (Kenya Demographic and Health Survey 2008-2009). Among these more general factors, one particular factor which accounts for half the drop in Kenya’s infant mortality rate is the increased use of treated bednets in the malaria zone where 40% of the population lives, from 8% of households in 2003 to 60% in 2008 (Demombynes and Trommlerova, 2012). Data for Kenya’s under five years mortality reveal a similar trend once again, with significant progress from 219 in 1962 to 91 in 1989 and then a reversal to 116 in 1999, which is most likely related to mother-to-child transmission of HIV, and then a recovery again to 74 in 2012. Although still low by developed world standards, Kenya’s current life expectancy from birth of 62 is a major improvement from 45.2 years as recently as 2003, but this recent recovery reveals the extent the which ground was being recovered from the 1989 figure of 60 years, which had been lost due to the impact of HIV/AIDS (Table 2). Finally, Kenya’s TFR oscillated from 6 to 6.8 from 1948 to the early 1960s, before increasing to 7.8 in the late 1960s and to the late 1970s and then reducing gradually to the current estimate of 3.98 children per woman (Institute of Economic Affairs, 2010).

Table 2 Demographic indicators for Kenya

Maternal Mortality Rates per 100,000 live births
1994	365
1998	590
2003	414
2008	488
2012	360
Infant Mortality Rates per 1000 live births
1948	184
1989	66
1999	77
2008	52
2012	47
Under 5 years mortality per 1000 live births
1962	219
1989	91
1999	116
2008	74
Life expectancy
1948	35
1989	60
1999	57
2003	45.2
2008	59
2012	62
Total Fertility Rate
1948	6
1969	7.8
1999	5
2012	3.98

Source: NCPD (2011) Facts and Figures on Population and Development 2011; Institute of Economic Affairs, 2010.

Conclusion

Neomalthusian ideological perspectives continue to widely influence policymakers and particularly the international population community in their evaluation of the effects of population growth in less developed regions, with many concluding that it constitutes a major inhibitor of economic progress. Some scholars, however, argue that the neomalthusian perspective on development issues is related to the tendency to isolate the demographic component of society from the wider social and cultural context in which population change takes place Such neomalthusian perspectives have had and continue to have a very significant influence on population policy, which are mainly directed at poor countries with strong support from international agencies. A key assumption underlying such policy is that fertility levels must be reduced rapidly and that the main driver for bringing this about is a reduction in the so-called ‘unmet need’ for family planning by raising levels of contraceptive prevalence and overcoming whatever social and cultural influences which might contribute to resisting these changes.

This narrowly based evaluation of development issues and its concomitant pragmatic policy conclusion is questioned by scholars who argue in favour of an approach based on a deeper understanding of the anthropology of the societies involved and of their rationale for decisions made about family size. The paper examines recent empirical trends within sub-Saharan Africa in order to contextualize the emerging fertility transition within a broader context. The data reveal considerable reductions in total fertility levels in many parts of sub-Saharan Africa, with wide variability in relation to reductions in both fertility and mortality throughout the region. Indicators of both maternal and infant mortality rates reveal rapid rates of decline within a very short period, which some suggest provide a strong basis for further significant improvements. A broader examination of demographic trends in sub-Saharan Africa points to the fact that perhaps around half of the current population growth may be related to the population momentum associated with the age structure of the population. It also raises the possibility of this part of the world, which is expected to experience around one-third of the young workforce population growth between now and 2030, benefiting from a demographic dividend, if effective economic policies are put in place.

The recent progress in reducing child and maternal mortality and increasing life expectancy, however, must be contextualized within the reality of ongoing major social and political challenges throughout sub-Saharan Africa. The impact of HIV/AIDS, which shows some signs of amelioration in certain regions, has been quite devastating, and continues to impact negatively on societies. The more detailed examination of Kenya’s recent demographic trajectory is revealing in that much of the recent progress is about regaining lost ground, particularly to the HIV/AIDS epidemic during the 1990s. Kenya has also achieved considerable success in reducing the impact of malaria on infant mortality in its malaria zone. All of these improvements in the demographic indicators have taken place within the broader context of advances in areas such as political transparency, improvements in security levels, and in economic participation. While such important developments are hugely encouraging, it is vital that political leaders adopt a more pro-active approach towards focusing on these broader development challenges facing sub-Saharan Africa, rather than becoming too narrowly focused on socially engineering family size.

Funding

This research received no specific grant from any finding agency in the public, commercial or not-for-profit sectors.