Jeni Klugman
 BEc (Sydney), LLB (Sydney), BCL (Oxford), PhD (Australian National)
 Fellow, Women and Public Policy Program, Kennedy School, Harvard University, USA
This email address is being protected from spambots. You need JavaScript enabled to view it.

 Edition: AHRLJ Volume 14 No 2 2014
  Pages: 705 - 734
 Citation: (2014) 2 AHRLJ 705-734
 Download article in PDF


The Protocol to the African Charter on Human and Peoples’ Rights on the Rights of Women in Africa has been hailed for its efforts to promote women’s health and rights.. The Protocol has now been signed and ratified by approximately two-thirds of African Union member states, from the most populous and largest to the smallest countries on the continent. The Protocol envisages major steps to improve the status of women on the continent, from economic opportunities and food security through to marriage and the rights of widows. This article seeks to contribute to the emerging literature on gender, health and rights, by exploring how government commitments to the health mandates of the Women’s Protocol have transpired in practice, one decade after its enactment, with a focus on resource allocations. The article’s scope includes a review of why sexual and reproductive rights matter, intrinsically, as rights, and evidence about their instrumental importance for development. Available evidence about status and trends in women’s health in Africa is presented, highlighting some advances as well as major shortcomings. This is the important empirical background against which to explore the human rights obligations of African states on this front, in particular the right to sexual and reproductive health and the potential contribution of the African Women’s Protocol. New analysis is undertaken of the extent to which governments have responded to the Protocol’s specific mandates with respect to military spending and social development, which suggests some promising trends. The conclusions highlight the finding that resource allocations in favour of health have significantly improved in countries that have ratified the Protocol, while underlining the importance of appropriate indicators and monitoring, and actions to ensure state accountability.

Key words: gender; reproductive health and rights; government spending; Africa

1 Introduction

The Protocol to the African Charter on Human and Peoples’ Rights on the Rights of Women in Africa (African Women’s Protocol) has been hailed for its efforts to promote women’s health and rights.1 The Women’s Protocol has now been signed and ratified by approximately two-thirds of African Union (AU) member states.2 These include a diversity of countries, from the most populous (Nigeria) and largest (Democratic Republic of Congo) to the smallest (Djibouti). However, 18 states - including some of the poorest countries, namely, Ethiopia, Chad and Niger - have not ratified the Women’s Protocol at the time of writing.3 Of the states that have not made commitments to the Protocol, many are conflict- or crisis-affected, notably Somalia, Egypt and Sudan.4

The African Women’s Protocol envisages major steps to improve the status of women on the continent, from economic opportunities and food security to marriage and the rights of widows.5 The focus here is on health, where among the highlights are promoting and protecting access to safe abortion6 and committing to the elimination of harmful practices, such as female genital mutilation7 and early marriage.8 Article 14 explicitly recognises the right of women to control their fertility, to choose methods of contraception, and the right to adequate health facilities, as well as other foundations of reproductive health and rights. This is consistent with Sen's concept of development as a process of expanding freedoms and agency.9 The Women’s Protocol commits signatories to redirect military resources to social development, a monitorable policy action which is a major focus of the article.10

The article seeks to contribute to the emerging literature on gender, health and rights, by exploring how government commitments to the health mandates of the Women’s Protocol have transpired in practice, one decade after its enactment, with a focus on resource allocation. The structure is as follows. First, I review briefly why sexual and reproductive rights matter, intrinsically as rights, and provide evidence about their instrumental importance for development. I then look at available evidence about status and trends in women’s health in Africa, highlighting some advances as well as major shortcomings. This is important empirical background against which to explore the human rights obligations of African states on this front, in particular the right to sexual and reproductive health and the potential contribution of the African Women’s Protocol. I undertake a new analysis of the extent to which governments have responded to the Protocol’s specific mandates with respect to military spending and social development. The final section broadens the discussion to review the structural and individual level determinants and constraints that matter for sexual and reproductive health, beyond levels of spending. The conclusions highlight the finding that resource allocations in favour of health have significantly improved in countries that have ratified the Women’s Protocol, while underlining the importance of appropriate indicators and monitoring, and actions to ensure state accountability.

Table 1 sets out, by way of context, the status of AU ratification of the African Women’s Protocol, and the availability of published data on health and military spending. It reveals that data gaps on military spending are far more frequent than gaps on health spending: Indeed, for the 36 countries that have ratified the Protocol, 12 are missing data on military spending, whereas health spending is available for all but Zimbabwe. Throughout, we rely on the Stockholm International Peace Research Institute which maintains an open source data on military spending for 171 governments since 1988, which defines military spending as including spending on personnel, operations and maintenance and procurement.11 Interestingly, the data gaps are more frequent among non-ratifying states: We lack data on military spending for seven out of 18 states.

Table 1: Data availability of health and military expenditures by ratification status




















Burkina Faso









Central African Republic



Cape Verde






Côte d’Ivoire
























Democratic Republic of Congo






Equatorial Guinea









Sahrawi Arab Democratic Republic



The Gambia



Sierra Leone












São Tomé and Principe












South Sudan































































South Africa






















































Note: Data is deemed missing if unavailable for more than three years during the period 2003 to 2012.

Sources: SIPRI Military expenditure by country as percentage of gross domestic product (2003-2012); WHO General government expenditure on health as percentage of GDP (2003-2011)

2 Sexual and reproductive health and rights matter

The sexual and reproductive health status and outcomes of women matter for several reasons. Firstly, there is intrinsic value in sexual and reproductive health and rights. This means that women have the right to safe and satisfying sex lives, and that they have the ability to make choices and decisions about whether or not to have children, when to have them and how many to have.12 Several international human rights instruments, including the Universal Declaration of Human Rights (Universal Declaration),13 the International Covenant on Civil and Political Rights (ICCPR)14 and the Convention on the Elimination of All Forms of Discrimination Against Women (CEDAW)15 are among those recognising the right to life, the right to health and the right to non-discrimination.16 Sen has powerfully argued this intrinsic value in Development as freedom, and this is now an accepted element of mainstream development thinking, as illustrated by the World Bank’s 2012 World Development Report on Gender.

Rights advocates have long underlined the critical importance of women being able to determine their own decisions, including around sexuality and health.17 Feminism is concerned with equity and justice, and overcoming oppression. Gender inequality and discrimination lead to ill health, as well as other forms of disadvantage.18

At the same time, sexual and reproductive health has instrumental value. At the household level, poor health can have negative implications for household consumption and other forms of financial well-being. A growing body of evidence shows that poor maternal health, in particular, can create an economic burden for households.19 In Burkina Faso, women who had more acute obstetric complications incur more financial burdens, for example, they report more borrowing, slower debt repayment rates and sold more assets in the subsequent year.20 More recent work in Kenya found that across all income groups, the costs of a maternal death exceeded one-third of annual consumption expenditure, and this catastrophic shock often left the families impoverished.21

Programmes and services that help to reduce the costs associated with health care can help prevent families from accumulating debt, which can drive families deeper into poverty. Therefore, national economies can also benefit when fertility rates fall and both mothers and their children are healthy.22

Labour productivity suffers because of poor maternal health. In sub-Saharan Africa, women are more than 60 per cent of workers in agriculture, and they contribute the bulk of unpaid work in domestic labour and caregiving.23 In addition to negative impacts on earnings, maternal mortality and morbidity can penalise other family members, especially those who have to fill the gap left when women are less able to work, reducing their own capacity to pursue educational and economic opportunities outside the household.24

Poor maternal health can bring negative inter-generational effects. Evidence suggests that poor maternal health during pregnancy can adversely affect the developmental outcomes of children later on in life and, conversely, if mothers receive adequate health care and nutrition during pregnancy, their children subsequently have better health, education and labour market outcomes.25

Furthermore, preventing young adolescent pregnancy positively affects girls’ life opportunities. In developing countries, pregnancy-related causes are the largest contributor to the mortality of girls aged 15 to 19 - nearly 70 000 girls die each year.26 Starting sex early increases the exposure of young people to HIV or other sexually-transmitted infections.27 Studies have also highlighted adverse long-term consequences on children born to young mothers, including low cognitive test scores, poor behavioural outcomes, grade repetition and economic disadvantage.28

In this light, it is not surprising that investments in health can boost an individual’s lifetime income, promote economic growth and reduce poverty.29 Studies have shown that health investments are positively linked to economic growth in sub-Saharan Africa.30

The evidence, therefore, is clear that improved maternal health is both a means and an end to gender equality and women’s empowerment,31 as well as important for development more broadly. We now turn to examine the situation in Africa, before outlining the context in terms of human rights.

3 Status of sexual and reproductive health in Africa: Where do we stand?

Africa has made some tremendous progress in improving women’s and girls’ sexual and reproductive health over the past two decades - but some major gaps remain. Here we review key highlights.

Most notable among the advances has been the drop in maternal mortality rates, by over 40 per cent since 199032 and, while data constraints make it difficult to assess trends in contraceptive use, there are some success stories. For example, Demographic and Health Survey (DHS) data suggests that Namibia has increased the number of women who use contraception by 20 percentage points.33 In North Africa, Egypt has moved from a 38 per cent prevalence rate to almost 60 per cent, and Zimbabwe has the highest prevalence rate in Eastern Africa with 59 per cent in the 2010-2011 DHS.34 Concerted local efforts have led to the elimination of female genital mutilation in some communities, such as in the well-known case of Tostan using collective empowerment approaches to end the practice in Senegal.35

Yet, enormous challenges continue to affect women’s health in Africa. Most notably, maternal mortality remains unacceptably high, contraceptive prevalence is low in many countries, and child marriage and gender-based violence are both pervasive phenomena.

Sub-Saharan Africa is the region with the highest maternal mortality in the world. Currently, about 165 000 African women die from childbirth-related complications each year, and for every death, approximately 20 more suffer from maternal morbidities, such as obstetric fistula.36 Chad ranks the worst, with 1 100 deaths per 100 000 live births. African women have a one-in-39 lifetime risk of dying from pregnancy or childbirth-related complications, and in countries such as Liberia, Niger, Somalia and Sierra Leone, the risk is much higher: one in 25.37 The risk factors for maternal mortality and morbidity include a high number of lifetime pregnancies, closely-spaced births and adolescent fertility.38 African women average close to five children each, and gaps in services make births riskier: Fewer than 47 per cent of births had skilled birth attendants. 39

The availability and utilisation of contraception are integral components of sexual and reproductive health and rights, as we explore below. Yet, there are vast differences across regions and countries in the extent to which this is realised, as shown in Figure 1.40 Among women aged 15 to 49 in North America, Latin America and Europe, prevalence rates are 73 to 78 per cent; however, in sub-Saharan Africa, more than two-thirds of women do not use modern contraceptives. 41

In sub-Saharan Africa in 2009, about a quarter of women reported having an unmet need for family planning. This figure reflects no overall improvement since the 1990s. At the same time, there is a wide variation in contraceptive prevalence across the continent. Among the 44 AU countries with data since 2003, the highest coverage is observed in Algeria and Cape Verde (both at 61 per cent), whereas Sudan (8 per cent) and Chad (3 per cent) have the lowest rates of contraceptive coverage.42

In terms of sub-regions, as shown in Figure 2, Southern Africa leads, with contraceptive prevalence of 59 per cent, approaching the global average, while Central and Western Africa are both below 20 per cent.43 Rates in Eastern Africa range from Kenya at 45 per cent, to Burundi at 9 per cent. Several AU countries that have not ratified the African Women’s Protocol have low prevalence rates, including Eritrea, Chad and Sierra Leone.

Figure 1: World contraceptive prevalence by region - Per cent women ages 15-49, any method, 2009


Figure 2: Contraceptive prevalence in Africa - Per cent women ages 15-49, any method 2009


Source: United Nations, Department of Economic and Social Affairs. (2011) World contraceptive use 2011

Sub-Saharan Africa counts for a disproportionate share of abortion-related deaths that happen globally every year - some two-thirds of the almost 50 000 deaths. This is largely attributable to unsafe abortions, often in poor and rural areas, often associated with traditionally-restrictive regimes that criminalise abortion. Young women face particular challenges in accessing abortion care - they account for approximately 40 per cent, or as many as 3,2 million,44 unsafe abortions worldwide.45 Tunisia and South Africa have been the most progressive reformers on this front, with a positive impact on maternal mortality.46 In South Africa, the annual number of abortion-related deaths fell by 91 per cent after liberalisation of the abortion law.47 It is worth noting that such legislation has also been associated with lower overall fertility rates and higher female labour force participation.48

Overall, fertility rates in sub-Saharan Africa have declined slightly, moving from 5,6 births per woman in 2004 to 5,2 births in 2011.49 Vast differences still exist sub-regionally. For example, Southern Africa, with a rate of 2,6 births per woman, has approximately half the fertility level of Central, Western and Eastern Africa.50 North Africa also has relatively low rates between 2 and 3 births per woman and, at the country level, rates range from 7 (Niger) to 2,9 (South Africa) births per woman.51

In sub-Saharan Africa, almost 40 per cent of women aged 20 to 24 were married before they were 18 years old, and 12 per cent married before their fifteenth birthday. The practice is especially widespread in West and Central Africa. Niger has the highest rates in the world, with 75 per cent of girls marrying before their eighteenth birthday, and in Chad, Central African Republic and Eritrea, between 20 and 30 per cent of girls married before their fifteenth birthday.52

Child marriage can bring a range of negative health outcomes. Girls who marry between the ages of 10 to 14 are five times more likely to die in childbirth than women aged 20 to 24 years, with specific risks related to obstetric fistula and obstructed labour.53 Child marriage is also strongly associated with high fertility levels (three or more births), repeated childbirth in less than 24 months and multiple unwanted pregnancies, as well as higher risks of contracting HIV. 54

Gender-based violence is another challenge that is both major and pervasive.55 It occurs in various forms, including intimate partner violence, sexual violence, and female genital mutilation.56 For sub-Saharan Africa as a whole, the best available estimates suggest that 37 per cent of women have experienced intimate partner violence and 12 per cent have experienced non-partner sexual violence.57 A sub-regional disaggregation of intimate partner violence in sub-Saharan Africa reveals that Central Africa has the highest prevalence rate with 66 per cent, while Southern Africa has the lowest with 30 per cent58 and, while estimates specific to North Africa are not available, the regional figure for North Africa and the Middle East stands at 35 per cent. With respect to the prevalence rates for non-partner sexual violence, in sub-Saharan Africa, Central Africa has the highest rate again with 21 per cent, and West Africa emerges with the lowest rate of 9 per cent. Available data for North Africa and the Middle East suggests that the rate stands at 4,5 per cent.59

In some countries, rates are much higher: For example, over 80 per cent of women in Uganda have experienced intimate partner violence. The National Violence against Children Surveys suggest that about one-third of young women in Swaziland, Tanzania and Zimbabwe experienced sexual violence before their eighteenth birthday.60

Violence has enormous repercussions for survivors, their children, communities and societies. Alongside psychological trauma and physical injuries, survivors often suffer reduced earnings, both in the short and long terms. In Tanzania, Veema (2013) estimated that women experiencing intimate partner violence earned approximately 29 per cent less than women who did not, and this figure increased to 43 per cent if the level of violence was severe.61 The World Bank has reported that the costs of violence range between 1,5 to 4 per cent of GDP, often exceeding what governments spend on primary education, for example.62

4 Key provisions of the African Women’s Protocol

We now turn to examine the key provisions of the African Women’s Protocol in light of the established literature around human rights and health, and women’s health. The African Women’s Protocol marks a potentially-important advance, with major provisions on sexual and reproductive health and rights.

It is useful to begin with the broader international context. The adoption of the Vienna Declaration and Programme of Action at the United Nations (UN) World Conference on Human Rights in 1993 is widely regarded as a watershed moment in the women’s rights movement. This universally recognised, for the first time, that women’s rights are human rights, paving the way for their integration into human rights norms and practice.

There are major synergies between human rights and health, which are especially relevant here, given the commitments that states have made to the protection and promotion of women’s health. States have recognised girls’ and women’s rights to health in several international agreements, including the right to health, medical care treatment and public health. The links between human rights and health have been well analysed by Ngwena and Yamin, as well as Cook and Shaw and others, and various issues of the Journal on Health and Human Rights.63

Human rights instruments that are relevant to health include provisions in CEDAW, ratified by 188 states, and which calls on state parties in article 24 to ‘take all appropriate measures to eliminate discrimination against women in the field of healthcare in order to ensure, on a basis of equality of men and women, access to health care services’.

There are also relevant commitments in ICCPR on non-discrimination;64 article 12 of the International Covenant on Economic, Social and Cultural Rights ( ICESCR ) (the right to the highest attainable standard of health); and the Convention on the Rights of the Child (CRC). The right to sexual and reproductive health was recognised in 1994 by the Programme of Action of the International Conference on Population and Development (ICPD), and subsequently adopted by 179 governments.65

As recently well summarised by Shaw and Cook,66 the UN Human Rights Council has acknowledged that preventable maternal mortality and morbidity are human rights violations.67 The UN High Commissioner for Human Rights has adopted guidance on the application of human rights to reduce preventable maternal mortality and morbidity. These guidelines underline the importance of underlying structural and social determinants, points that we return to in the final section of this article.68

At the heart of treaties are state accountability and responsibility. State parties are required to take positive action to realise the rights enumerated. However, the extent to which the rights to health laid out in global and regional legal instruments have been translated into constitutions, legislation and enforcement and accountability mechanisms at the national level varies. A recent analysis by Assi et al at the World Policy Analysis Centre suggests that, if we distinguish between different enumerations of rights, with attention to their gender dimensions, we observe the following patterns:

  • A broad right to health is more frequently included in constitutions of CEDAW member states relative to individual rights in health. Overall, 62 per cent of constitutions guarantee any health rights, either specifically to women in aspirational terms or universally. Protections of health were more common in constitutions adopted after CEDAW ratification (87 compared to 41 per cent).
  • Among all CEDAW ratifiers, 47 per cent protect women’s rights to medical care treatment by guaranteeing the right in any of the following ways: to women explicitly (17 per cent of countries); to citizens universally and providing general protection against discrimination based on gender (14 per cent); to women in aspirational terms (8 per cent); or universally without mentioning gender (9 per cent.
  • Most of CEDAW state parties (84 per cent) do not mention the right to public health in their constitutions. Out of all constitutions from CEDAW-ratifying countries, 12 per cent protect women’s rights to public health explicitly or guarantee the right to citizens and provide general protection against discrimination based on gender.

Looking more specifically at Africa, we see that several countries have constitutions that guarantee medical care to women, including Egypt, Namibia, Mozambique and Tunisia, while a couple of others have more general provisions, including Kenya and South Africa. Most African constitutions, however, have no provisions relevant to women’s access to health care. Nearly all lack a constitutional guarantee to public health, the exceptions being Congo and Tunisia, while Ethiopia has an aspirational provision to this effect.69

A recent study by the World Health Organisation (WHO), the first of its kind, asks: What evidence is available to policy makers that human rights have helped to improve women’s and children’s health.70 Drawing on quantitative data mainly collected for other purposes, the authors conclude that there is plausible evidence that human rights contributed positively to health and health-related gains for women and children in the four countries, such as increased access to emergency obstetric care (Nepal), increased access to modern contraception (Brazil), reductions in early childhood mortality (Malawi), and increased vaccination coverage (Italy).

Landmark litigation is emerging on this front.71 In the first ever maternal death case to be decided by an international human rights body, the CEDAW Committee held Brazil responsible for the preventable maternal death of Alyne da Silva Pimentel Teixeira, a Brazilian of African descent, due to post-partum haemorrhage following the delivery of a 27 week-old stillborn fetus in a private health centre. As Shaw and Cook underline, this decision establishes as a matter of international law that governments have human rights obligations to guarantee that all women in their countries, regardless of income level or racial background, have access to timely, non-discriminatory and appropriate maternal health services in public and private health facilities. Similarly, historic decisions have been issued for Paraguay and India. These decisions mark the first time that courts of law have applied constitutional and human rights law to hold governments legally accountable for the preventable maternal death of women arising from gaps in the healthcare system.

Turning now to Africa, we begin with the African Charter on Human and Peoples’ Rights (African Charter), ratified by all 53 member states of the AU, which is also the parent treaty of the African Women’s Protocol. Article 2 enshrines the principle of non-discrimination, including on the grounds of sex, and article 18(3) calls on state parties to eliminate ‘every discrimination against women and also ensure the protection of the rights of the woman ... as stipulated in international declarations and conventions’.

Recognised shortcomings of the African Charter include the failure to explicitly define discrimination against women; lack of guarantees of the right to consent to, and equality in, marriage, and an emphasis on traditional values and practices that can impede the advancement of women’s rights in Africa. This led non-governmental organisations (NGOs) in Africa to advocate for the adoption of the African Women’s Protocol to strengthen the provisions on gender equality, which led to the wide-ranging Protocol.

The Women’s Protocol has been described as an important advance in the protection and the promotion of the rights of women in Africa, and innovative in key respects.72 It is the first international law instrument to call for an end to all forms of violence against women, whether in private or in public, including sexual harassment; to prohibit all forms of female genital mutilation; protect women’s rights to seek abortion under certain conditions; prohibit forced marriages; and specify 18 years as the minimum age of marriage. There is an expansive set of rights enshrined in the Women’s Protocol, extending to a broad range of economic and social rights, including rights to equal pay for equal work and to adequate and paid maternity leave in both the public and private sectors, that are beyond the scope of this article.73

The focus here is on article 14, which spells out a woman’s right to health. It covers sexual and reproductive health, which includes her right to control her fertility; to decide whether to have children, her number of children and their spacing; to choose any contraceptive methods; self-protection and to be protected against sexually-transmitted infections; to be informed of her and her partner’s health status; and to have family planning education. State parties are required to:74

  • provide adequate, affordable and accessible health services, including pre-natal, delivery and post-natal health and nutrition services for women during their pregnancy and while they are breast feeding;
  • provide family planning information and services, including to women in rural areas; and
  • authorise medical abortion in cases of sexual assault, rape, incest and where continued pregnancy endangers the health or life of the mother or the foetus.

This means that a state party violates the Women’s Protocol when it, for example, has family planning policies that do not respect the rights of women to choose the number or spacing of their children, or prohibits abortion in cases noted above. At the same time, however, it is important to note Ngwena’s critique that the Protocol does not go far enough to the extent that it largely medicalises the grounds of abortion and implicitly makes health care professionals the gatekeepers rather than the women themselves.75

It is useful to outline the key elements of the human rights architecture in Africa. This is done drawing on the valuable Guide to Using the Protocol to Advance the Rights of Women in Africa, published by Equality Now.

  • The African Commission, which began functioning in 1986, was established under the African Charter to protect and promote human and peoples’ rights and to monitor state compliance with the Charter and its Protocols. It comprises 11 independent commissioners who are elected by the Executive Council of the AU to serve for six-year terms on a part-time basis, with the possibility of a single re-election, and meets twice yearly for two-week sessions. During these sessions, the African Commission considers:
  • complaints (also referred to as communications) submitted by individuals, NGOs, institutions, lawyers and state parties;
  • periodic state party reports under article 62 of the African Charter on compliance; and
  • reports of human rights violations and promotional activities of the commissioners.
  • 2 Special Rapporteur on the Rights of Women in Africa

  • The African Commission has special mechanisms such as Special Rapporteurs and working groups to bolster its activities in the promotion of human and peoples’ rights. The Special Rapporteur on the Rights of Women in Africa was established in 1998, who is selected from among the commissioners and whose tasks include following up on the implementation of the African Women’s Protocol, by preparing reports on the situation of women’s rights in Africa, and proposing recommendations to be adopted by the African Commission. Complaints for violations of rights under the Women’s Protocol may be brought by individuals or approved NGOs (those with African Commission observer status in the case of the African Court and those with accreditation to the AU in the case of the African Court of Justice and Human Rights) directly to the relevant court as well, subject to provisions in the relevant court’s protocol. At the time of writing, the African Commission has yet to issue any decisions that directly address women’s rights. As Equality Now notes, there are however several determinations under the African Charter that are relevant to women’s rights claims, which may be useful to draw upon in making arguments for women’s rights.76
  • The African Court on Human and Peoples’ Rights was established to strengthen the protective mandate of the African Commission, and decided its first case in 2009.77

At the domestic level, provisions of the African Women’s Protocol can be used in complaints and arguments before the courts. After domestic remedies have been exhausted, practitioners can bring cases to the African Commission or the African Court, if applicable.

If the African Women’s Protocol has been ratified by the state but not duly incorporated into national law, practitioners can still call on courts to take judicial notice of the state’s obligations under the Protocol. This was successfully done in a 2008 case of teacher rape in Zambia, for example. The High Court of Zambia noted the government’s obligations under the Women’s Protocol, even though this had not been formally incorporated into domestic law. The plaintiff was awarded significant damages for pain and suffering and mental torture, as well as aggravated damages and medical expenses. The judge also directed the Ministry of Education to put regulations in place to protect students in school; and referred the case to the Director of Public Prosecutions to initiate a criminal case against the teacher, calling the failure to prosecute ‘a dereliction of duty’. Significantly, the judgment cited the full text of the relevant article of the Women’s Protocol. While judicial notice does not make the international instrument enforceable as law, it does mean that the decision can be cited as precedent in subsequent legal arguments.

Strong advocacy is critical to promote these rights at the local level. Women’s groups around the world have used international human rights commitments to lobby for progressive legal reform. In Morocco, for example, CEDAW was used as a platform to successfully push for equal citizenship rights and land rights for tribal women.78 Likewise, increased awareness can help to support and refresh dialogue on women’s rights and health. Equality Now has undertaken a campaign for the enactment of a law against female genital mutilation in Mali. This refers to Mali’s obligations under the African Women’s Protocol: In 2005, Mali ratified the African Women’s Protocol which in article 5(b) requires state parties to prohibit FGM through legislative measures backed by sanctions. This is in addition to article 1 of Mali’s Constitution, which sets forth the right of all citizens to integrity of person and guarantees the protection of all citizens from inhumane, cruel and degrading treatment, as do other similar international obligations of Mali.

Of special interest here is article 10(3) of the African Women’s Protocol, urging African governments to spend more on women’s development and less on the military. Article 10 is about the right to peace, and includes a specific provision, worded as follows (article 10(3)):

    States parties shall take the necessary measures to reduce military expenditure significantly in favour of spending on social development in general, and the promotion of women in particular.

Article 26 of the Women’s Protocol - related to implementation and monitoring - commits state parties to ensure ‘the implementation of this Protocol at national level’ and to ‘undertake to adopt all necessary measures and in particular shall provide budgetary and other resources for the full and effective implementation of the rights herein recognised’. It is also important to highlight the 2001 Abuja Declaration, where heads of state in Africa committed to ensuring that all necessary resources were made available and used effectively in the area of public health, and pledged to allocate 15 per cent of their annual national budgets to health.79

In this context, it is interesting and important to understand whether trends in military and health spending reflect these aspirations and commitments. Health is an important element of social development in general, was specifically cited in the Abuja Declaration, and is amenable to empirical investigation. We use the most comprehensive data available, from the Stockholm International Peace Research Institute, the World Bank, and the World Health Organisation.80 We look at spending for the period 2003 to 2011, complemented by more recent data where available. The earlier years provide a useful baseline, prior to the African Women’s Protocol coming into force.

5 How has spending on health changed one decade later?

We have noted that the African Women’s Protocol commits state parties to significantly reduce military expenditure in favour of spending on social development, and the promotion of women in particular, and therefore turn to examine whether there is empirical evidence of this shift. The focus is on health, given the centrality of sexual and reproductive health to human rights, and the major commitments to these rights in the Protocol. Reductions in military spending can help create the fiscal space needed to allow other priority expenditures, including health, to expand.81 The focus on health does not diminish the importance of other dimensions of gender equality but, for reasons of space and tractability, these issues are outside the scope of the article. Whilst it is impossible to prove causality between the changes in observed levels of spending on military and health, it is nonetheless interesting and important to document and understand the broad empirical trends.82

Over the decade under review, there is evidence of increased effort in health spending among members of the AU. Between 2003 and 2011, general government expenditure on health rose from 2,5 to 3,0 per cent of their Gross Domestic Product (GDP). This was supported by rising donor investments on African healthcare systems. In 12 African countries, external funding accounted for over 30 per cent of their total health spending in 2006, and exceeded 50 per cent in Mozambique and Madagascar.83

We undertook an analysis of spending levels and trends to compare countries that had, and had not, ratified the Women’s Protocol. Highlights are as follows:

  • In 2011, countries that ratified the African Women’s Protocol directed more of their GDP to health than countries that have not ratified the Protocol (3,3 per cent versus 2,5 per cent). 84
  • The largest efforts are evident in Lesotho and Malawi, which averaged 6 per cent from 2003 to 2011, both of which have ratified the Protocol.85
  • Looking at changes over time, Lesotho, Liberia and Rwanda achieved the largest gains, increasing on average by more than 3 percentage points, and all three countries have ratified the African Women’s Protocol.86
  • At the low end, the following countries directed less than 1,5 per cent of their GDP toward health care over the 2003 to 2011 period: Cameroon, Côte d’Ivoire, Guinea, Chad, Congo, Eritrea and Gabon.87 While many of these countries at the low end have in fact ratified the Women’s Protocol, the ‘non-ratifiers’, Chad and Eritrea, are the only countries that witnessed decreases in spending as a share of GDP for the period (by -1,7 and -0,2 percentage points, respectively).
  • The largest declines in spending as a share of GDP are seen among ‘non-ratifiers’: Chad, as previously mentioned, and São Tomé and Principe, which had a decrease of about 3,3 percentage points.88
  • Among North African countries with data (Algeria, Egypt, Libya and Tunisia), Tunisia leads in spending as a share of GDP averaging 3 per cent over the 2003 to 2011 period, while Libya spent the least with 1,9 per cent.89

What this translates into in terms of actual spending on the ground obviously depends on the size of income per capita. The overall trend is impressive with the public spending per capita on health more than doubling from US $19 in 2003 to US $49 in 2011.90 However, this is also where the relatively low levels of economic development in Africa are thrown into sharp relief - the average world spending per capita is now over US $600 dollars (US $615), as shown in Figure 3, thus demonstrating enormous disparities in income translating into the resources available for health. 91

Figure 3: Per capita government expenditure on health - by region, 2003-2011 (Current USD)

Chapter 2 18 2014-4

Source: World Health Organization. (2013). Global Health Observatory Data Repository


In the most recent year available (2011), public health spending per capita among signatories ranges from US $7 (Democratic Republic of Congo) to US $819 (Equatorial Guinea).92 At the high end for that year, alongside Equatorial Guinea are the Seychelles (US $404) and South Africa (US $329), all countries which have ratified the African Women’s Protocol. Among countries that spent the least in 2011, in addition to DRC, are Eritrea (US $7) and Burundi (US $7,6). Eritrea and Burundi are yet to ratify the Women’s Protocol. Per capita spending data for North Africa is unfortunately limited to Algeria, which in 2011 spent in the mid-range with US $182 per capita.93

Outliers are interesting to observe, particularly in terms of ratification. Over the entire period of 2003 to 2011, the highest spending countries at the sub-regional level are Botswana (Southern Africa), the Seychelles (East Africa), Cape Verde (West Africa) and Equatorial Guinea (Central Africa).94 Notably, all but Botswana have ratified the African Women’s Protocol. At the low end, Lesotho (Southern Africa), Burundi (East Africa), Guinea-Bissau (West Africa) and Chad (Central Africa) spent the least per capita on health in their respective sub-regions, and Chad and Burundi are yet to ratify the instrument. 95

Botswana is a noteworthy case. There have been major public investments in health and a recent World Bank Public Expenditure Review notes that almost all urban residents are within 15 kilometres of a primary health care facility, as are 80 per cent of rural residents, reflecting widespread healthcare provision.96 However, civil society organisations, in a report for Solidarity for African Women’s Rights on the Protocol, have highlighted worries about the lack of political will for ratification of the African Women’s Protocol.97 Given the forecast of the decline in mineral resource revenues, there are concerns about sustainability of and commitment to public spending on health.

Liberia increased spending efforts from 1,2 to 4,6 per cent of GDP between 2003 and 2011. A recent World Bank study highlights these gains, but finds that the distribution of spending is inequitable. The disparity between spending in the richest and poorest regions was more than six times and there are large differences in regions in the availability of skilled health personnel. Donor funding accounted for about 80 per cent of public health sector funding, which underlines potential issues of sustainability.98 Clearly, the outbreak of Ebola in West Africa in 2014 created enormous stress on the health system in Liberia and neighbouring countries.

In sum, then, we see encouraging trends in health spending and evidence of increased government effort in many signatory countries, including Equatorial Guinea, South Africa and Gabon, which have had the largest increases since 2003.99 We also see relatively significant efforts in non-signatory countries for which data are available, including Botswana and Mauritius.100 In North Africa, we see increases in Algeria, although the country is yet to ratify the Women’s Protocol.101

At the same time, it is important to underline that we have not reviewed the evidence about how this money is being used, whether it is being actually directed to programmes likely to benefit women’s and girls' health, nor anything about the links between such spending and outcomes. The share of social spending devoted to human priority concerns varies significantly across countries, and likely matters even more than how much they spend on social services overall.102 Improving the quality of services and availability of contraceptive methods can increase agency. Take-up of weak services is often low. A recent Egyptian study suggests that certain dimensions of quality are particularly important. These include privacy, both during medical examinations, and respect for patient confidentiality, facility cleanliness, friendly staff, shorter waiting times and limited staff turnover.103

The broad picture is nonetheless indicative of improved efforts and directions, without being definitive or conclusive. It is important to recognise that assessing whether the goal of equitable distribution of health-related resources has been met would require careful identification of the similarities and differences in the health needs of men and women in the local context, alongside an analysis of the obstacles that currently prevent men and women from realising their potential for good health.104 With these caveats in mind, we now turn to the evidence on military spending.

6 Trends in military spending

Publicly-available data about military spending is relatively scarce. The best-established and most renowned source is the Stockholm International Peace Research Institute (SIPRI), which is an independent international institute that researches conflict, armaments, arms control and disarmament. SIPRI figures reflect government reporting and if there is evidence about the unreliability of national data, estimates are made using government budget and expenditure accounts.105 Here we focus on the period 2003 to 2011, and include 2012 data when available.

Military spending as a share of GDP globally stands at about 2,5 per cent.106 In recent times, there have been divergent trends regionally, with military spending falling in North America, as well as Western and Central Europe witnessing reductions in 2012, in part due to austerity measures, and there were increases in North Africa, South America and East Asia.

In 2012, military expenditure averaged about 2 per cent as a share of GDP among AU countries with available data (35 out of 54 countries). At the country level, using available data, we see a large range. At the top end, South Sudan and Algeria spend 8,4 per cent and 4,5 per cent respectively on military as a share of GDP while, at the low end, we see Mauritius with 0,2 per cent and Ghana with 0,3 per cent.

In terms of ratification, in 2012 AU countries that have ratified the Maputo Protocol spent less as a share of GDP than ‘non-ratifiers’ (1,7 per cent versus 2,3 per cent of GDP). It is also notable that some data is missing for several countries, many of which are ‘non-ratifiers’, where the spending might be expected to be high, including Chad, Central African Republic, Somalia and Sudan.

As a share of GDP, military spending in the AU shrank by over one-fifth between 2003 and 2012, from an average of 2,4 to 1,9 per cent of GDP . This average trend inevitably conceals divergence at the country level: Burundi, Ethiopia, and Botswana reduced spending by 4,9, 2,0 and 1,8 percentage points, respectively, while the countries with the largest increases are Swaziland (1,5 percentage points), Libya (1,3 percentage points) and Algeria (1,2 percentage points).

Given gaps in the data, it is difficult to draw any conclusions about the relationship between (non)-ratification and military spending. Nevertheless, spending as a share of GDP is moving in welcome directions, namely, military spending is falling and health spending is increasing. This is summarised in Figure 4 overall, and separately for ratifying and non-ratifying countries.

Figure 4: Average public health and military expenditure in the African Union 2003-2011


Sources: Author estimates based on SIPRI. (2013). Military expenditure database and World Health


Comparative trends can be illustrated with selected country cases for the period 2003 to 2011:

  • Lesotho (5,7 per cent), Malawi (5,7 per cent) and Botswana (4,9 per cent) led in terms of share of GDP spent on health care. Among these countries, Lesotho and Botswana also cut military spending as a share of GDP by about 1 and 2 percentage points, respectively.107 Indeed, between 2010 and 2012, Lesotho reduced military spending by some 40 per cent falling as a share of GDP from 3,2 per cent of GDP in 2010 to 1,9 per cent.
  • While Swaziland increased military spending over the time period and has the largest increase over the 2003 to 2011 period as measured in percentage points, the government spends significantly more on health spending and recently ratified the African Women’s Protocol in 2012. 108

Exploring the health and military spending in countries that have not ratified the Women’s Protocol reveals a mixed picture. On the one hand, ‘non-ratifiers’ like Chad and Algeria spend more on military than on health. Yet, we also have the cases of other ‘non-ratifiers’, such as Sierra Leone, Botswana, Madagascar and Mauritius which direct a larger share of their GDP to health than to military purposes. Indeed, Mauritius’s military spending (only 0,2 per cent of GDP) is the lowest in the AU.

Overall, most countries are spending more on health as a share of national income than they are on their militaries. This is shown in Figure 4, where most countries are above the 45 degree line. Among countries that have ratified the Women’s Protocol, 23 of the 33 countries with data spend more on health and, of the ‘non-ratifiers’, 8 of out 12 with data spend more on health.

Figure 5: National public health and military expenditure: Contrasting Maputo Protocol signatories and non-signatories, average share of GDP expenditure (average for period 2003-2011)


Sources: SIPRI. (2013). Military expenditure database and World Health Organization. (2013). Global Health Observatory Data Repository

7 So what can health spending potentially achieve?

We know that, on the supply side, long distances and poor quality hinder equitable access to health services. Evidence from Burkina Faso illustrates the severity of the problem, with around half of women identifying distance as a primary barrier to health care access.109

An increasing number of examples from Africa illustrate how innovations supported by public spending can address supply side barriers. In Rwanda, results-based financing (RBF) is improving the delivery and quality of health services. Small financial incentives are given to service providers, conditional on their patients receiving maternal and child health services. After 23 months, institutional deliveries improved by almost 25 per cent.110 Another programme in Zimbabwe also showed promising results. Rural health clinics receive a subsidy conditional on the provision of a set of free health services to pregnant women and young children. After the first year, the number of women who had four or more prenatal appointments increased by 65 per cent.111 Likewise, in Nigeria coverage of institutional deliveries increased by 30 percentage points (to 39 per cent) and coverage of ante-natal care increased from 16 per cent to 77 per cent in the year to December 2012, following the introduction of an RBF approach.112

However, we know that spending alone is not enough. Measures to enhance accountability and transparency are needed, alongside changes that enable women to have a strong say in their own healthcare and broader decisions. We address each of these in turn.

Promising approaches to improving maternal health outcomes include participatory community mechanisms and ameliorating the quality of information available about services, to help make providers more accountable to users.113 In Uganda, for example, a social accountability mechanism which increased community participation as well as information about the health centre’s performance showed favourable outcomes. After one year, institutional delivery rates increased, and after three years, the rate of stillbirths dropped by almost one percentage point. Information about the quality and quantity of the centre’s services played a critical role in the functioning of this accountability mechanism; when the community lacked such information, this approach did not change health outcomes. 114

Budget monitoring can play an important role. In South Africa, the Eastern Cape Province has the worst health outcomes in the country. There is a high demand for public services, but limited delivery capacity, corruption and dysfunctional administration systems.115 The lack of accountability of managers and administrators, wasteful expenditure and poor planning were revealed in 10 consecutive audits. A civil society group launched the Public Service Accountability Monitor (PSAM), which is a multi-pronged campaign, including budget monitoring, access to information litigation, and community mobilisation. Major changes followed, including the replacement of the head of the Department of Health, the investigation of thousands of employees for misuse of funds and corruption, and the blacklisting of over 100 companies with potentially-fraudulent tenders. While these changes are not empirically linked to the campaign, the outcomes suggest that the increased public awareness and detailed documentation of the mismanagement of funds helped to provide the evidence base and increased the political will for change.116 This has promising implications for rights-based approaches focused on accountability.

Core to feminist approaches is the pivotal importance of disadvantage and the distribution of power.117 Social accountability mechanisms are shaped by surrounding power structures, social contexts and broader policy architectures.118 In cases where regimes are pluralistic and democratic and the rule of law is upheld, there is greater space for political and civil rights and voice.119 These factors help facilitate social accountability mechanisms. The legal and policy frameworks can help facilitate citizen participation and oversight. Finally, the capacity of civil society to organise and function, citizens’ relations with the government, and people's willingness to question authority are all important.120

This underlines the importance of gender inequality in the society more broadly, to health outcomes. Indeed, there is a clear correlation between maternal mortality rate and gender inequality, as illustrated in Figure 5, using the UNDP’s multidimensional index of gender inequality, which includes economic opportunities, education and health. Simply put, the more gender unequal the society, the more likely a woman is to die giving birth.

Figure 6: Maternal mortality rate and gender inequality


Source: Grépin and Klugman, 2013

Weak agency inhibits women's capacity to insist on safer sex practices. For example, in 10 countries around the world, more than half of the women report that they cannot ask their partner to use a condom. In Niger, this figure stands at 89 per cent.121 There is evidence of African women’s increasing preference for concealable contraceptive methods (injectables), rising from 6 to 20 per cent in sub-Saharan Africa. In some contexts it has been found that giving women ultimate control over family planning shows results. For example, unwanted births in Zambia were reduced only when women had individual control over contraceptive decision making.122 However, enabling women to access and use contraceptives in secret can also carry risks. For example, in some cases, women fear - or more actually, face -violence from their partners if the use of contraceptives without the spouse’s knowledge is discovered.123

8 Conclusions

The African Women’s Protocol is a landmark agreement in the promotion of the health and rights of women in Africa. The article has examined the extent to which the Women’s Protocol is associated with positive shifts in government resource commitments, in the context of the broader literature about women’s health and human rights. A human rights approach provides tools to hold governments legally accountable for their failure to address the preventable causes of maternal death and to enable good sexual and reproductive health - and this was combined with an empirical budget analysis to try to understand what is happening in practice.

Our review suggests that there has been some improvements in spending efforts since the Protocol came into force. Public spending is moving in the right direction. Indeed, we documented that in the AU, general government expenditure on health increased by some 20 per cent, a significant boost, and available data on military spending as a percentage of GDP reveals a fall of about 21 per cent in the AU over the decade.

Interestingly, the increasing health efforts were most marked among governments that had ratified the Women’s Protocol. However, there does not appear to be a strong correlation between ratification and the direction of trends in military spending.

The findings point to several key recommendations: first, the importance of openness and transparency, and ensuring that reliable budget data is published on a regular and timely basis; second, and relatedly, the value of budget monitoring by civil society and human rights advocates to support efforts to promote health and human rights, and considering the scope to bring cases to help establish jurisprudence in this important sphere. Third, as the cross-country findings are inevitably limited in depth, it is important to undertake further research and analysis using gender-responsive budgeting, building on the pioneering work of UN Women, UNDP and others.

At the same time, I underlined the importance of broader factors that drive sexual and reproductive health outcomes for women and girls. While spending on social development is critical, accountable and transparent structures and institutions are also crucial. More fundamentally, expanding the agency and choice of women and girls is fundamental. This is linked to broader structures of equality and inequality in society which affect multiple dimensions of wellbeing.

1. The African Women’s Protocol was adopted in 2003 and came into force in 2005.

2. At the time of writing, 36 out of 54 African states had ratified the Women’s Protocol; (accessed 17 October 2014).

3. Solidarity for African Women's Rights Report on the mapping exercise on the status of ratification of the Protocol on the rights of women in Africa (data sheet); Algeria, Botswana, Burundi, Central African Republic, Chad, Egypt, Eritrea, Ethiopia, Madagascar, Mauritius, Niger, Sahrawi Arab Democratic Republic, Sierra Leone, Somalia, São Tomé and Principe, Sudan, South Sudan, Tunisia (September 2012).

4. FIDH Women’s rights in Africa: Eighteen countries are yet to ratify the Maputo Protocol! (2013) (accessed 17 October 2014); World Bank List of fragile and conflict-affected states FY 2014 - Fragility, Conflict and Violence (2013) Resources/511777-1269623894864/HarmonizedlistoffragilestatesFY14.pdf (accessed 17 October 2014).

5. en_eng.pdf (accessed 17 October 2014).

6. Art 14 African Women’s Protocol.

7. Art 5.

8. Art 6.

9. A Sen Development as freedom (1999).

10. Art 10(3) African Women’s Protocol.

11. For specific details, see database/definitions (accessed 17 October 2014).

12. United Nations Population Fund By choice not chance: Family planning, human rights and development (2012).

13. Adopted by the United Nations General Assembly in 1948.

14. Entry into force in 1976, and currently has 168 parties.

15. Adopted by the United Nations General Assembly in 1979, and came into force in 1981. Currently, 188 states have ratified. See (accessed 17 October 2014).

16. Other important declarations include the Vienna Conference on Human Rights (1993); the Programme of Action of the International Conference on Population and Development (1994); the Beijing Declaration and Platform for Action, adopted at the Fourth World Conference on Women (1995); the Southern African Development Community's Declaration on Gender and Development (1997); and the Addendum on the Eradication of All Forms of Violence Against Women and Children (1998), recognising that violence against women can constitute a violation of human rights.

17. See, eg, M Langford et al (eds) Millennium Development Goals and human rights: Past, present and future (2013); P Hunt & T Gray T (eds) Maternal mortality, human rights and accountability (2013) 129-143.

18. L Doyal What makes women sick: Gender and the political economy of health (1995).

19. K Grépin & J Klugman Closing the deadly gap between what we know and what we do: Investing in women’s reproductive health (2013) 3.

21. A Kes et al The economic burden of maternal mortality on households: Evidence from three sub-counties in rural Western Kenya (2014).

22. Department for International Development Choices for women: Planned pregnancies, safe births and healthy newborns (2010) 8.

23. World Bank Gender and development: World development report (2011).

24. Grépin & Klugman (n 19 above) 3. See also AE Yamin et al ‘Costs of inaction on maternal mortality: Qualitative evidence of the impacts of maternal deaths on living children in Tanzania’ (2013) PLoS ONE 8(8): e71674.

25. Grépin & Klugman (n 19 above) 4-5.

26. UNFPA State of the world’s population 2013: Motherhood in childhood (2013).

27. MJ Hindin & AO Fatusi International perspectives on sexual and reproductive health (2009) 35 58-62.

28. JP Azevedo et al Teenage pregnancy and opportunities in Latin America and the Caribbean on early child bearing, poverty and economic achievement (2014) 90B00PUBLIC0.txt (accessed 17 October 2014).

29. TP Schultz ‘Health and schooling investments in Africa’ (1999) 13 Journal of Economic Perspectives 67, as cited in JC Anyanwu & AE Erhijakpor Health expenditures and health outcomes in Africa (2007) 2

30. K Gyimah-Brempong & M Wilson ‘Health, human capital and economic growth in sub-Saharan and OECD countries’ (2004) 44 Quarterly Review of Economics and Finance 296, as cited in Anyanwu, & Erhijakpor (n 29 above) 2.

31. Department for International Development (2010) 8.

32. UNFPA East and Southern Africa Saving mothers’ lives (2013);jsessionid=D3E1DCE44D 68D0F4724915D55C9D5AA3.jahia01 (accessed 17 October 2014).

33. Current use of contraception among currently-married women; any method. Moving from 29% to 55,1%; USAID Stat Compiler , (1992 DHS), (2006-07 DHS).

34. Current use of contraception among currently married women; any method: (38%) 1988 DHS to 59% USAID Stat Compiler . (2005 DHS).

35. UNFPA Ending harmful practices through community-led social change in Senegal (2010); http: // (accessed 17 October 2014).

36. UNFPA (n 32 above).

37. UNFPA The health of mothers is improving: Campaign on accelerated reduction of maternal mortality in Africa (accessed 17 October 2014); Grépin & Klugman (n 19 above) 16.

38. UNFPA (n 32 above).

39. World Bank Fertility Rate, Total. Data Bank (accessed 17 October 2014); Grépin & Klugman (n 19 above) 8.

40. United Nations Department of Economic and Social Affairs World contraceptive use 2011 ceptive2011.htm (accessed 17 October 2014).

41. United Nations Department of Economic and Social Affairs (n 40 above); UNFPA (n 32 above).

42. Available data from 2004 to 2010 analysed. Figures from Algeria represent 2006 rates and figures from Cape Verde represent 2005 rates; Chad (2004) and Sudan (2006); United Nations Department of Economic and Social Affairs (n 40 above).

43. Calculated using data from United Nations Department of Economic and Social Affairs (n 40 above). Global figure 62,7% (2009); women aged 15-49, married or in a union using any method. Regional composition: Southern Africa: Botswana, Lesotho, Namibia, South Africa, Swaziland; Northern Africa: Algeria, Egypt, Libyan Arab Jamahiriya, Morocco, Sudan, Tunisia; Eastern Africa: Burundi, Comoros, Djibouti, Eritrea, Ethiopia, Kenya, Madagascar, Malawi, Mauritius, Mozambique, Réunion, Rwanda, Somalia, Uganda, United Republic of Tanzania, Zambia, Zimbabwe; Central Africa: Angola, Cameroon, Central African Republic, Chad, Congo, Democratic Republic of the Congo, Equatorial Guinea, Gabon, São Tomé and Principe; and Western Africa: Benin, Burkina Faso, Cape Verde, Côte d'Ivoire, The Gambia, Ghana, Guinea, Guinea-Bissau, Liberia, Mali, Mauritania, Niger, Nigeria, Senegal, Sierra Leone, Togo.

44. UNFPA (n 26 above).

45. I Shah & E Ahman ‘Age patterns of unsafe abortion in developing country regions’ (2004) 12 Reproductive Health Matters 9.

46. J Benson et al ‘Reductions in abortion-related mortality following policy reform: Evidence from Romania, South Africa and Bangladesh’ (2011) Reproductive Health 8(39).

47. R Jewkes & H Rees ‘Dramatic decline in abortion mortality due to the Choice on Termination of Pregnancy Act’ (2005) 95 South African Medical Journal 250.

48. DE Bloom et al ‘Fertility, female labour force participation and the demographic dividend’ (2009) 12 Journal of Population Economics 79-101.

49. World Bank Fertility Rate, Total (2013). Data Bank: (accessed 17 October 2014).

50. EL Madsen, The Wilson Center, New Security Beat Why has the demographic transition stalled in sub-Saharan Africa? (2013) (accessed 17 October 2014).

51. USAID Stat Compiler; total fertility rate for African sub-regions.

52. R Vogelstein Ending child marriage: How elevating the status of girls advances US foreign policy objectives (2013) 5.

53. A Raj et al ‘Prevalence of child marriage and its effects on fertility and fertility-control outcomes of young women in India: A cross-sectional, observational study’ (2009) Lancet 1883-1889. Fistulas can occur due to early sexual initiation, which in many cases occurs before a girl’s first period when she is a child bride.

54. As above.

55. For a global review, see J Klugman et al Voice and agency (2014) ch 3.

56. R Coomaraswamy Integration of human rights of women and the gender perspective: Violence against women, Report of the Special Rapporteur on Violence Against Women, Its Causes and Consequences, Ms Radhika Coomaraswamy: Addendum: Communications to and from governments (1999).

57. WHO Africa region: Botswana, Cameroon, Democratic Republic of the Congo, Ethiopia, Kenya, Lesotho, Liberia, Malawi, Mozambique, Namibia, Rwanda, South Africa, Swaziland, Uganda, United Republic of Tanzania, Zambia, Zimbabwe; WHO Global and regional estimates of violence against women (2013) Appendix 1; Lifetime prevalence of physical and/or sexual intimate partner violence among ever-partnered women. WHO Global and regional estimates of violence against women (2013) 17-19.

58. Prevalence of intimate partner violence - Central Africa: 66%; East Africa: 39%; Southern Africa: 30%; West Africa: 42% in WHO (n 57 above).

59. WHO (n 57 above) 47-48.

60. See United Republic of Tanzania, UNICEF, CDC, Muhimbili University Violence against children in Tanzania: Findings from a national survey 2009 (2011) (accessed 17 October 2014).

61. S Vyas Estimating the association between women’s earnings and partner violence: Evidence from the 2008-9 Tanzania National Partner Survey (2013).

62. Klugman et al (n 55 above).

63 See D Shaw & C Ngwena (eds) Health and human rights (2007); D Shaw & RJ Cook ‘Applying human rights to improve access to reproductive health services’ (2012) 119 International Journal of Gynaecology and Obstetrics S55-S59; AE Yamin ‘Toward transformative accountability: Applying a rights-based approach to fulfill

63. maternal health obligations’ (2010) 7 SUR International Journal on Human Rights; F Bustreo & P Hunt Women’s and children’s health: Evidence of impact of human rights (2013).

65. International Conference on Population and Development (ICPD) Programme of Action (accessed 17 October 2014). Note that some of the governments adopted ICPD with certain reservations.

66. Shaw & Cook (n 63 above).

67. United Nations Human Rights Council ‘Preventable maternal mortality and morbidity and human rights’ 18th session, 23 September 2011 A/HRC/18/L.8.

68. Yamin (n 63 above).

69. T-M Assi et al Are states parties to CEDAW living up to their commitments to women and girls? A globally comparative review of laws and policies (2014).

70. Bustreo & Hunt (n 63 above).

71. See AE Yamin & S Gloppen (eds) Litigating health rights: Can courts bring more justice to health? (2011); Shaw & Ngwena (n 63 above) Part IV.

72. See, eg, the foreword to the Guide to Using the Protocol to Advance the Rights of Women in Africa, published by Equality Now, by Soyata Maïga, Commissioner/Special Rapporteur on the Rights of Women in Africa.

73. See World Bank Gender at work (2014), for a recent review of the status of economic opportunities globally, including in Africa; J Klugman & S Twigg Gender at work in Africa: Legal constraints and opportunities for reform (forthcoming) African Journal of International and Comparative Law.

74. For the full text, see proto_women_eng.pdf (accessed 17 October 2014).

75. C Ngwena ‘Using human rights to combat unsafe abortion: What needs to be done?’ (2013) (accessed 17 October 2014).

76. See 52-68 for examples from the African and other regional human rights systems, including the European system.

77. When a new African Court of Justice and Human Rights comes into force (after the necessary ratifications), these will be merged.

78. The national citizenship law was reformed in 2004; a 2012 Ministerial Circular recognises tribal women’s rights and affirms full equality for men and women under CEDAW and the Constitution; S Riza Women’s collective action in the Middle East and North Africa (2013).

79. LG Sambo et al Health financing in Africa: 2000 to 2009 data analysis (2013) 2.

80. SIPRI Military expenditure database (2013); World Bank’s Data Bank; WHO Global health expenditure database (2013).

81. S Seguino Financing for gender equality: Reframing and prioritizing public expenditures to promote gender equality (2013) (accessed 17 October 2014).

82. The interested reader is also referred to key literature on gender budgeting, namely, UNDP Gender responsive budgeting manual for trainers (2005).

83. Global Fund to Fight AIDS, Tuberculosis and Malaria Trends in development assistance and domestic financing for health in implementing countries (2010) 7.

84. WHO (n 80 above).

85. As above.

86. Largest increases in GDP spending from 2003 to 2011. Lesotho increased spending from 4,03% to 9,08%; Liberia increased spending from 1,19% to 4,63%; and Rwanda increased spending from 3,34% to 6,13%.

87. Cameroon: 1,2%; Côte D’Ivoire: 1,3%; Guinea: 1,3%; Chad: 1,4%; Congo: 1,4%; Eritrea: 1,4%; Gabon: 1,4%. WHO (n 80 above).

88. WHO (n 80 above).

89. As above.

90. Per capita government expenditure on health at average exchange rate (USD) - (accessed 17 October 2014).

91. WHO Global Health Observatory Data Repository, health financing: health expenditure per capita by WHO region (2013) (accessed 17 October 2014).

92. n 90 above.

93. WHO (n 80 above).

94. Botswana (273); Seychelles (371); Cape Verde (97); and Equatorial Guinea (345).

95. Lesotho (47); Burundi (5, 7); Guinea Bissau (7, 12); Chad (7,8).

96. World Bank Botswana public expenditure review (2010) 92.

97. Solidarity for African Women's Rights. Report on the mapping exercise on the status of ratification of the Protocol on the Rights of Women in Africa (2012) 3.

98. World Bank Africa Region Liberia public expenditure review: Human development (2012) 47.

99. Changes from 2003 to 2011: Equatorial Guinea (83 to 818); South Africa (125 to 329); Gabon (67 to 192); WHO (n 80 above).

100. Changes from 2003 to 2011: Botswana (180 to 263); Mauritius (100 to 205); WHO (n 80 above).

101. Changes from 2003 to 2011: Algeria (63 to 182); WHO (n 80 above).

102. United Nations Development Program Financing human development (1991) 39.

103. T Rabie et al Transforming family planning outlook and practice in Egypt: A rights-based approach (2013).

104. L Doyal ‘Gender equity in health: Debates and dilemmas’ (2000) 51 Social Science and Medicine:931-939.

105. SIPRI Sources and methods (2013) (accessed 17 Octo-ber 2014).

106. All military spending figures in this article are based on SIPRI Yearbook 2013 and Military expenditure database: Military expenditure by country as percentage of gross domestic product, 1988-2012 (2013) (accessed 17 October 2014).

107. In the 2003 to 2012 period. Lesotho’s military spending fell from 2,8% to 1,9% and Botswana moved from 3,7% to 1,9%.

108. WHO (n 80 above).

109. Grépin & Klugman (n 19 above) 8.

110. (accessed 17 October 2014).

111. ‘In rural Zimbabwe, no fees, better care for women and children’ (2012) (accessed 17 October 2014).

112. Health Results Innovation Trust Fund (HRITF) Initial Results 17 April 2013.

113. D Berlan D & J Shiffman ‘Holding health providers in developing countries accountable to consumers: A synthesis of relevant scholarship’ (2012) 27 Health Policy and Planning 271-280.

114. M Bjorkman-Nyqvist et al ‘Information is power: Experimental evidence of the long run impact of community-based monitoring’ Working Paper.

115. International Budget Partnership When opportunity beckons: The impact of the public service accountability monitor’s work on improving health budgets in South Africa (2011) 1 ing-health-budgets-in-south-africa/ (accessed 17 October 2014).

116. International Budget Partnership (n 115 above) summary 2-3.

117. WA Rogers ‘Feminism and public health ethics’ (2006) 32 Journal of Medical Ethics 351.

118. A George ‘Using accountability to improve reproductive health care’ (2003) 11 Reproductive Health Matters 161-170; Z Bhutta et al ‘Community participation: Lessons for maternal, new-born, and child health’ (2008) The Lancet; Grépin & Klugman (n 19 above) 11.

119. M McNeil & C Malena Demanding good governance: Lessons from social accountability initiatives in Africa (2010) 186.

120. McNeil & Malena (n 119 above) 186.

121. World Bank Sexual and reproductive health and family formation, voice, agency and participation series (forthcoming) (2013) 9.

122. N Ashraf et al Household bargaining and excess fertility: An experimental study in Zambia (2010).

123. AA Bawah et al ‘Women’s fears and men’s anxieties: The impact of family planning on gender relations in Northern Ghana’ (1999) 30 Studies in Family Planning 54-66.