SOCIAL PROTECTION AND AFRICA S PROGRESS TOWARDS ACHIEVING MDGs SOCIAL SAFETY

Transcription

1 SOCIAL PROTECTION AND AFRICA S PROGRESS TOWARDS ACHIEVING MDGs SOCIAL SAFETY Mina Baliamoune-Lutz, Professor and Kip Fellow of Economics, University of North Florida Paper prepared for the Conference on Experiences and lessons from social protection programmes across the developing world: what role for the EU?, organised by the European Report of Development in Paris, France, June, MOBILISING EUROPEAN RESEARCH FOR DEVELOPMENT POLICIES

2 SOCIAL PROTECTION AND AFRICA S PROGRESS TOWARDS ACHIEVING MDGs SOCIAL SAFETY ABSTRACT In this paper, I first assess Africa s progress towards achieving the first six Millennium Development Goals (MDGs), focusing on the main target indicators. Second, using recent African data on MDG indicators and an index of social protection, and after controlling for other relevant variables, I examine the relationship between social protection and various MDG indicators. I comment on the results of the analysis and provide a discussion of some areas in which African countries may need to place a greater emphasis on social protection. Mina Baliamoune-Lutz Professor and Kip Fellow of Economics, University of North Florida mbaliamo@unf.edu The views expressed in this paper are those of the authors, and should not be taken to be the views of the European Report on Development, of the European Commission or of the European Union Member States. 2

3 Unless things improve, it will take sub-saharan Africa until 2129 to achieve universal primary education, until 2147 to halve extreme poverty, and until 2165 to cut child mortality by two-thirds. (European Commission Report on Millennium Development Goals, 2004) Social protection is the foundation for the participation of the poorest in economic life on more favourable terms; and for their access to the health, education and other services that are critical for national economic, as well as social, development. This is a critical issue given the vulnerability of poor people, especially the bottom 20%, in the face of global market shocks and climate change. The MDGs need to recognise the impact of social protection on such vulnerability. (Andrew Shepherd 2008) 1. Introduction A recent draft discussion paper for EU-Africa MDGs partnership (European Commission 2009) states: Policy responses to the crisis in support of the MDGs in Africa might include, for example, stimulating a resumption of investment flows to Africa, developing country insurance mechanisms to reduce Africa's vulnerability to external shocks, and social protection for the most vulnerable populations. Indeed, social protection could have an effect on the progress of countries for each of the Millennium Development Goals (MDGs), but it is particularly critical to achieving meaningful progress in the first six MDGs. Sub-Saharan Africa (SSA) 1 is the region in which progress towards achieving most of the MDGs has generally been weak, and, given the recent world financial crisis and its potential impact on Africa s economies, the need for social safety-nets for the most vulnerable groups in society is more evident than ever before. In a recent paper, Anthony Hodges (2008) identifies five structural factors that need to be taken into account in efforts to strengthen social protection in West and Central Africa. These factors are the extent of poverty, the nature of inequity, supply-side weaknesses in basic social services, fiscal space, and governance/administrative constraints. It turns out that these factors also play an important role in progress towards achieving the MDGs. A study titled The Contribution of Social Protection to the Millennium Development Goals (World Bank 2003) emphasises the role of social protection, and maintains that: without appropriate social protection mechanisms, the MDG targets for 2015 will not be achieved. In fact, the case it makes is for a social protection perspective to infuse and inform all Bank operations, towards the end of achieving the MDGs. The African Development Bank 2009 MDG Report recognises the critical role of social protection in alleviating the negative effects of economic and financial crises, when it states: to deal with the immediate adverse effects of the crisis, African countries need welltargeted and flexible social protection measures. Such measures could include public works programs, food-for-work programmes, contributory pensions, and health insurance policies (African Development Bank, 2009). Social protection has more than one definition. The traditional definition views social protection as those policies that provide social safety-nets, social funds, labour market interventions, and social insurance (World Bank 2003). Social protection has also been defined as the set of interventions that assist poor individuals, households and communities 1 In this paper, Africa and sub-saharan Africa are used interchangeably. 3

4 to reduce their vulnerability by managing risk better (World Bank 2001). Norton et al. (2001) define social protection as: the public actions taken in response to levels of vulnerability, risk and deprivation which are deemed socially unacceptable within a given polity or society. But it seems that a more encompassing definition is the one used by the ILO, which defines social protection as: the set of public measures that a society provides for its members to protect them against economic and social distress that would be caused by the absence or a substantial reduction of income from work as a result of various contingencies (sickness, maternity, employment injury, unemployment, invalidity, old age, and death of the breadwinner); the provision of health care; and, the provision of benefits for families with children. (Bonilla and Gruat 2003). Based upon this last definition, it is clear that social protection can have a major impact on most, if not all, indicators used to assess progress towards achieving the MDGs. As pointed out by Norton et al. (2001): Social protection thus deals with both the absolute deprivation and vulnerabilities of the poorest, and also with the need of the currently non-poor for security in the face of shocks and life-cycle events. This is, indeed, quite relevant to the case of Africa, where chronic poverty levels are very high and where countries that rely on primary commodity exports tend to experience recurrent shocks. Social protection, as is clear from the definition above, does not rely on a single instrument. Policy-makers (and donors) may often need to make a trade-off among the different instruments, taking their short- and long-term impact into consideration. For example, social protection that promotes gender equity (MDG 3) may sometimes lead to less social protection for young men, which may cause the rates of crime and conflict to increase (Bourguignon et al. 2008). On the other hand, if we consider the long-run impact of social protection instruments that promotes female health and education, we may be better off pursuing precisely this type of social protection because of the significant positive externalities on children s education and health, and on peace-building. Barrientos (2002) reviews the situation of women in informal employment in Latin America and examines crucial public-policy issues in extending social protection to this group. He finds that there is a major gender dimension to informality, and women in informal employment are often excluded from formal social protection programmes. This is also the case for most SSA countries. For Baliamoune-Lutz (2007) and Baliamoune-Lutz and McGillivray (2009), tradepromoting polices and globalisation may have adverse effects on gender equity. Thus, if social protection is extended to informal employment, it would have a significant positive impact on gender equity (MDG 3). Barrientos and Barrientos (2002) argue that globalisation has led to important changes in both the nature of work and to workers access to social protection, and that work in the globalised economy is becoming increasingly informalised. Informal workers face higher social risk and have less social protection relative to workers in the formal sector. Most African countries have significant informal employment and a large rural sector which is generally characterised by the pre-dominance of informal employment. Countries such as Ethiopia and Zimbabwe use social protection schemes that have an impact on poverty reduction among rural households (Bourguignon et al. 2008). African countries need feasible, equitable and sustainable social pensions and other retirement income transfers. Kakwani and Subbarao (2005) find that poverty rates among 4

5 the elderly are higher than in the general population in most of the SSA countries that they examine. In SSA, the average coverage rate of mandatory pension systems is 6 percent; with Mauritius having the highest coverage rate, at 50 percent, followed by Cape Verde, with 27 percent. But the coverage rate in the majority of SSA countries is below 5 percent, and tends to be concentrated on civil servants and employees in public and large private enterprises (Forteza et al. 2009). The self-employed and workers in the informal sector usually have no coverage. Clearly, there is a need to extend social protection to this sector in Africa. Recent studies show that social pensions have contributed to poverty reduction in several African countries. For example, Samson (2006) shows that, in South Africa, noncontributory social pension almost closes the poverty gap for older people living alone. Kaniki (2007) finds that, in Mauritius, which has had a universal social pension since 1950, poverty rates for older people living with more than one younger person were 30 percent lower than they would be without the universal pension. Good macro-economic policies and efficient and equity-enhancing institutions can be particularly helpful, by creating an environment that provides incentives for savings. Good labour institutions can alleviate the cost of social assistance and pensions, and minimise the burden on governments oftenlimited budgets (see Freeman 2010). The remainder of the paper proceeds as follows. In the next section, I assess Africa s progress towards achieving the first six MDGs, focusing on the main target indicators. In Section 3, using recent African data on MDG indicators and an indicator of social protection, and after controlling for other relevant variables, I examine the relationship between social protection (using the World Bank index for social protection and rating) 2 and various MDG indicators. Section 4 summarises the main points and provides a discussion of some important areas in which African countries may need to place a greater emphasis on social protection instruments that could enhance Africa s progress towards achieving the six MDGs. 2 A better measure of social protection is one that describes the different social protection programmes, such as the ones included in the Social Assistance in Developing Countries Database. Unfortunately, data on most African countries are missing. Another good way to study these issues is to conduct case studies. However, the goal of this paper is to focus on the cross-sectional macro level and see if we could get a broad (general) picture of the impact of social protection in Africa. 5

6 2. Overview of Africa s progress towards achieving the MDGs 2.1 Africa s progress towards achieving the first six MDGs MDG 1: The eradication of extreme poverty and hunger MDG 1 is about eradication of poverty and hunger by It involves three main targets and nine indicators. The first target consists of reducing by half the proportion of people living on less than a dollar a day. Progress in this area is measured upon the basis of three indicators: the proportion of populations below $1 (or $1.25) per day, the poverty-gap ratio, and the share of the poorest quintile in national consumption. The second target consists of achieving full and productive employment and decent work for all, including women and the young. Progress in this area is measured upon the basis of four indicators: the growth rate in GDP per person employed, the employment-to-population ratio, the proportion of employed people living below $1 (or $1.25) per day, and the proportion of ownaccount/self-employed and contributing family workers in total employment. The final target consists of reducing by half the proportion of people who suffer from hunger. Progress towards achieving this target is measured using the prevalence of underweight children under-five years of age and the proportion of the population below the minimum level of dietary energy consumption. Africa s progress towards achieving MDG 1, although relatively significant, remains insufficient to reduce the proportion of extreme poverty by half. Figure 1a shows the extent of poverty using poverty headcount in selected years. We note that, while Africa has managed to reduce poverty consistently over the period , it still has the highest poverty headcount ratio, and lags behind all other regions in this area. While poverty has been reduced at global level and at regional level (including SSA), poverty levels in SSA are still high. According to projections made by the World Bank (2010 Global Monitoring Report), the low-growth scenario (assuming that the recovery projected for the post-crisis trend will not take place in the medium-run) predicts that the percentage of people living under $1.25 a day at global level is 18.5% in 2015 and 16.3% in For SSA, the predicted levels are 43.8% and 39.9%, respectively. The projections in the lowgrowth scenario indicate that, in 2020, the world will have 1,053 billion people living on less that $1.25 a day, of which 438 million (about 41%) are in SSA. The projections under a high-growth scenario (assuming a relatively rapid economic recovery in 2010, with strong growth continuing into the future), indicate that the percentage of people living under $1.25 a day in SSA to be 38% and 32.8% in 2015 and 2020, respectively. Thus, under the most optimistic scenario, SSA will still have about one third of its population living in extreme poverty. The most recent crisis is also affecting other areas that are critical to progress in achieving the MDGs. The projections for other MDG progress indicators indicate a similar dismal outcome for under-five child mortality, primary school completion rates, and access to improved source of water. The recent (pre-crisis) high growth in many African countries had renewed hope of the possibility of speeding up progress towards achieving the MDGs, especially given that some studies have shown that poverty responds to growth when initial levels of poverty are high (the case of SSA). However, the latest financial crisis and the subsequent slowdown in the global economy growth rates may seriously affect the amounts of Foreign Direct Investment (FDI), trade and aid flowing to Africa and thus affect growth, which would cause a further slowdown in the rate of progress. 6

7 Figure 1a. Poverty headcount ratio (PPP, % of population) 3. 6 M id d le East & N o rt h A f rica Lat in America & C arib b ean East Asia & Pacif ic South Asia Sub - Saharan A f rica Figure 1b. Employment to- population ratio ( ) MENA East Asia&Pacific SSA LAC South Asia Figure 1c. Prevalence of underweight children under-five years of age Figure 1d. Prevalence of undernourishment (% of population) M EN A LA C East A sia & Pacif ic So ut h A sia SSA 5 0 M ENA LA C East A sia & Pacif ic So ut h A sia SSA Source of data: World Bank MDG database (2010). Figure 1b portrays the evolution of one of the indicators (employment-to-population ratio) associated with the second target. Interestingly, SSA had a higher ratio relative to Latin America during most of the 1990s but a lower ratio since the late 1990s. In fact, the only two regions where this ratio has continued to increase during the period are Latin America and the Middle East and North Africa (MENA) region. This indicator, however, does not give a true picture of the quality of employment. For example, the lower ratio in the MENA region may imply, particularly in the oil-rich states of the Middle East, that women do not participate significantly in the labour force, which does not necessarily mean that they have lower social protection than women do in regions in which this ratio is high. MDG 2: Achieve universal primary education The target for MDG 2 is to ensure that all boys and girls complete a full course of primary education. Progress in this area is measured by three indicators: net enrolment ratio of primary education, the proportion of pupils starting grade 1 who reach the last grade of primary education, and the rate of literacy of year-old men and women. Figures 2a and 2b show that SSA has achieved significant progress in these areas. However, because Africa started from low levels, it is still lagging behind all other regions, and many African countries need a particularly strong increase in the rate of progress in order to achieve universal primary education by

8 Figure 2a. Net enrolment ratio in primary education LAC M ENA Sout h Asia SSA Figure 2b. Literacy rate of year-olds East Asia & Pacific LAC MENA South Asia SSA Source of data: World Bank MDG database (2010). MDG 3: Promote gender equality and empower women The target for MDG 3 is to eliminate gender disparity in primary and secondary education preferably by 2005, and at all levels by Progress in this area is measured by three indicators: the ratio of girls to boys in primary, secondary and tertiary education, the share of women in wage employment in the non-agricultural sector, and the proportion of seats held by women in national parliaments. We now know that the elimination of gender disparity in primary and secondary education did not take place (by 2005) in three regions: MENA, South Asia and SSA. More importantly, it does not appear that SSA will reach gender equity in primary and secondary education (see Figure 3a) within the next five years (by 2015). SSA has achieved tremendous progress in the proportion of seats held by women in national parliaments, raising the ratio by about one third over the period (Figure 3b). There are some SSA countries where the share of African women in the total seats in national parliaments exceeds their share in OECD countries. For example, in 2009, in seven African countries (Rwanda, South Africa, Angola, Mozambique, Uganda, Burundi and Tanzania) the share of women was higher than 30%. In Rwanda and South Africa, women held 56.3% and 44.5% of the seats, respectively. These numbers are much higher than the numbers in the United States and most EU countries. Yet, other measures of gender inequality are still quite high in many African countries, including countries in which women hold a significant share of seats in national parliaments. For example the ratio (in %) of girls to boys in primary and secondary education (in 2008) is 84% in Burundi and 86% in Mozambique. 8

9 Figure 3a. Ratio of girls to boys in primary and secondary education (%) Figure 3b. Proportion if seats held by w omen in national parliaments East Asia & Pacific MENA SSA 0 East Asia & Pacific LAC MENA South Asia SSA Source of data: World Bank MDG database (2010). MDG 4: Reduce child mortality The target for MDG 4 is to reduce by two thirds the mortality rate among children under five by 2005 from the 1990 level. Progress in this area is measured by three indicators: underfive mortality rate, infant mortality rate, and the proportion of 1 year-old children immunised against measles. Figure 4a shows that the average rate of immunisation against measles in SSA has increased by almost one third over the period ; from about 55 in 2000 to approximately 73% in In about half the countries, the rate in 2009 exceeded 80% and at least 10 countries (Seychelles, Mauritius, Cape Verde, Eritrea, Botswana, Swaziland, São Tomé and Príncipe, Rwanda, The Gambia and Kenya) had a rate in excess of 90% (in 2008). In contrast, Figure 4b indicates that Africa has a very significant lag in its progress towards achieving MDG 4. By 2008 (more than half way to the target year of 2015), Africa has, on average, reduced under-5 mortality rates only by about 13%. Figure 4a. Rate of immunization against measles Source of data: World Bank MDG database (2010). 9

10 MDG 5: Improve maternal health Two targets are associated with MDG 5. The first target is to reduce by three quarters maternal mortality rates and progress towards achieving it is assessed upon the basis of two indicators: the maternal mortality rate and the proportion of births attended by skilled health personnel. The second target is to achieve universal access to reproductive health by 2015, and progress in this area is measured upon the basis of contraceptive prevalence, adolescent birth rates, antenatal care coverage, and unmet need for family planning. Africa has a very significant lag in these areas. Figure 5a. Adolescent fertility rate Figure 5b. Contraceptive prevalence East Asia & Pacific LAC MENA South Asia SSA East Asia & Pacific LAC MENA South Asia SSA Source of data: World Bank MDG database (2010). In 2005, the maternal mortality rate (per 100,000 live births) was 900, which is at least 6 times the rate in East Asia and Latin America (LAC), 4.5 times the rate in MENA and almost twice the rate in South Asia (a rate of 500). Figures 5a and 5b indicate that Africa also significantly lags behind other regions in the rate of adolescent fertility and the prevalence of contraceptives. In both indicators, Africa s progress has been rather slow. MDG 6: Combat HIV/AIDS, malaria and other diseases The first target associated with MDG 6 is about halting and starting to reverse the spread of HIV/AIDS, and this target is assessed upon the basis of four indicators: HIV prevalence among the population aged years old, condom use, at least for high-risk sex, the proportion of population aged years old with comprehensive correct knowledge of HIV/AIDS, and the ratio of school attendance of orphans to school attendance of nonorphans aged years old. The second target calls for achieving (by 2010) universal access to treatment for HIV/AIDS for all those who need it. The indicator associated with this target is the proportion of population with advanced HIV infection with access to antiretroviral drugs. The last target is to halt and to begin to reverse the incidence of malaria and other major diseases. Five indicators are used to assess progress towards reaching this target: incidence and death rates associated with malaria, the proportion of children under 5 years of age sleeping under insecticide-treated bed-nets, the proportion of children under 5 years of age with fever who are treated with appropriate anti-malarial drugs, the incidence, prevalence and death rates associated with tuberculosis, and the proportion of tuberculosis cases detected and cured under directly observed short course treatment. 10

11 Figure 6a. Malaria and HIV prevalence HIV prevalence rate, Malaria mortality Adult years (%) Source of data: World Bank MDG database (2010). Figure 6a shows that progress towards halting the spread of HIV over the period has been consistent, since the rate has not, on average, increased. On the other hand, Africa s progress towards halting and reducing deaths from malaria has been inconsistent. The rate of malaria deaths peaked in 2003 (exceeded 20%), went down in 2004, and then went up again in More recently, however, Africa seems to maintain a steady reduction in the rate of malaria mortality. In addition, Figure 6b indicates that progress in halting and reversing the incidence of TB has been quite slow. In summary, on average, Africa has achieved little progress compared to other regions, but looking at averages can be highly misleading. In order to provide a fair overview of Africa s progress, we need to emphasise four points. First, as pointed out by many recent studies and reports, the statistics show Africa lagging, but Africa, as a region, had a much lower starting-point in almost every indicator used to assess progress towards achieving the MDGs. In fact, for many indicators, the change compared to the initial levels (in 2000) has been very strong in absolute terms, and Africa s achievements become quite strong when we factor in the low levels of the starting-points or take into account the stagnation experienced from the 1970s to the early 1990s (see Global Monitoring report, 2010). Second, averages hide huge disparities across African countries. There are, indeed, some countries that have achieved highly significant progress while other have not progressed at all in several areas. As an illustration, Table 1 highlights the large disparities in progress among the best (a reduction of 30% or higher) and the worst performers (a reduction of 10% or less) in under-five mortality rates over the period Similarly, Figure 7 highlights that there are large differences among countries in anti-retroviral therapy coverage (an indicator used to assess progress towards achieving MDG 6). Progress measured by most other indicators also reflects significant disparity among countries. Country heterogeneity in MDG performance can often be explained by initial conditions, institutional environment and/or the presence of conflict. In addition, most fragile states (with many making weak progress) are in SSA and the large number of fragile states heavily affects the averages for the various indicators causing Africa to appear as severely lagging behind other regions Figure 6b. TB incidence and detection Incidence of TB TB detection by DOTS

12 Table 1: Mortality rate, under-five (per 1,000) Change (%) Best performance Botswana Namibia Rwanda Malawi Eritrea Swaziland Ghana Cape Verde Worst performance Guinea-Bissau Nigeria Burkina Faso Comoros South Africa Angola Zimbabwe Burundi Sudan Central African Republic Mauritania São Tomé and Príncipe Congo, Dem. Rep Somalia Kenya Chad Congo, Rep Source: World Bank MDG database (2010). 12

13 Figure 7. Antiretroviral therapy coverage (% of people with advanced HIV infection) Namibia Botsw ana Rw anda Senegal Benin Zambia Sw aziland Gabon Mali Kenya Malaw i Burkina Faso Uganda Tanzania Equatorial Guinea Ethiopia South Africa Cote d'ivoire Guinea Nigeria Lesotho Cameroon Angola Mozambique Congo, Dem. Rep. Mauritania Burundi Mauritius Central African Republic Sierra Leone Guinea-Bissau Togo Gambia, The Zimbabw e Liberia Congo, Rep. Ghana Eritrea Chad Niger Madagascar Somalia Sudan Source of data: World Bank MDG database (2010). 13

14 Table 2: Correlations among MDG indicators in Africa Employmentto-population ratio GDP growth per person employed Primary completion rate Enrolment in primary education (net ratio) Female held seats in national Parliament (% of total seats) Girls/boys in primary & secondary education maternal mortality rate Infant mortality rate Measles immunization Under-5 mortality rate TB incidence GDP growth per person employed Primary completion rate Enrolment in primary education Female held seats in national parliament Girls/boys in primary & secondary education maternal mortality rate Infant mortality rate Measles immunization (%) Under-5 mortality rate TB incidence TB prevalence Correlations are based on panel data for the period covering 40 SSA countries. Source of data: World Bank MDG database (2010) 14

15 Third, we observe weak correlations among the various indicators, so that some countries may be achieving significant progress in some indicators and little progress in others. As stated by Bourguignon et al. (2008), [t]he correlation between GDP per capita growth and non-income MDGs is practically zero thus confirming the limited linkage found between those indicators and poverty reduction. This limited correlation reflects some relative independence of the policy instruments governing progress towards MDGs from economic growth, as well as substantive differences in country policies and circumstances that may affect the relationship between these policies. The correlation among some MDGs can also be non-linear. For example, the correlation between education and health and growth may be weak (or even negative), become positive and strong within a medium range, and then become weak again at high levels of education and health (human capital). Table 2 shows that correlations among a large group of MDG indicators in Africa are, for the most part, weak. The strong correlations are between education and health indicators, especially those measuring gender equality and health. Finally, the distributional aspects of progress towards achieving the MDGs must be taken into account, otherwise many countries may end up with MDGs achieved in urban areas alone or exclusively among certain groups (those with access to health and education, for example), while rural areas and large groups with little or no access to social services would lag further behind. Unfortunately, the regional averages as well as the country averages do not provide information on these distributional aspects. These four points highlight the role that social protection can play in helping countries that lag behind to pick up a higher pace and to ensure that the distributional aspects of progress towards achieving the MDGs are taken into consideration by extending safety-nets to those who need them the most. 3. The contribution of social protection to progress towards achieving the MDGs 3.1 Social protection and MDG indicators Table 2 shows how various social protection interventions contribute to progress in achieving the first six MDGs. It is clear that, sometimes, different social protection interventions are needed to achieve different MDGs, while, in other times, the same intervention contributes to several MDGs. For example, safety-nets, social funds and maternity (and paternity) leave contribute to most MDGs, whereas pensions seem to contribute (at least directly) mainly to MDG 1. We also note that the same set of social protection interventions contribute to most indicators for MDG 4, 5 and 6, perhaps because of the high direct and indirect association among these indicators. A World Bank study (World Bank 2003) estimates the impact of social protection on poverty reduction through reducing transient poverty, lowering vulnerability, and reducing the need for income smoothing, and found that social protection interventions could reduce the total poverty head-count rate by 5 to 10%, which is a significant contribution to MDG 1. According to World Bank (2003), a recent estimate of poverty with and without social protection programs in the Kyrgyz Republic shows that among social protection beneficiaries, the extreme poverty headcount would have increased by 24%, the gap by 42% and the severity of poverty by 57%, if the beneficiaries had not had access to social protection. 15

16 Table 3: The contribution of various social protection interventions to the MDGs Social protection interventions Pensions Unemployment benefits Safety nets Fee waivers for health and education Transfers targeted to children or youth Social funds Labour market policies Disability insurance (DI) and social insurance (SI) Support to informal workers (skills and access to markets ) MDG 1 MDG 2 MDG 3 MDG 4, 5 and 6 reduce poverty among the old and disabled reduce transient poverty among the unemployed. smooth/raise income, thereby directly affecting poverty allow families to obtain services without impoverishing themselves can help reduce current as well as intergenerational transmission of poverty empower the extreme poor help build social infrastructure can create environment for job creation, productivity and wage growth. DI smooth income for those temporarily or permanently unable to work. enhances their productivity. contribute to access of girls to schools through attention to appropriate sanitation, security, etc. in the construction of buildings Construction and rehabilitation of schools Facilitate direct provision of education services, which is an essential component towards achieving UPE encouraging women s labour force participation create incentives for female education. addressing child labour can increase school enrolment SI smoothes household income (protects from shocks) and thereby increases children s school enrolment and lowers child labour. Girls enrolment may be especially sensitive to income or price, and safety-nets can help. contribute to access of girls to schools through attention to appropriate sanitation, security, etc. in the construction of buildings Safety-nets for orphans and vulnerable children can help cope with impacts of diseases, and protect them from HIV Construction & re-habilitation an essential component of social funds provides infrastructure in the form of health centres. Many social funds focus on building awareness about HIV/AIDS and rehabilitation of patients. Direct provision of health services is an essential component towards reducing child mortality and improving maternal health and social funds facilitate this. 16

17 Policies against discrimination cash/food) transfer programmes Child allowances or other assistance Access of disabled students to schools through appropriate infrastructure Community driven development Maternity (and paternity) leave Ensure that women, ethnic minorities and the disabled have equal access to employment If conditioned on school enrolment, create incentives for families to send children (esp. girls) to school provide income and consumption smoothing for parents and caregivers. Increases their enrolment, participation and inclusion and social funds can facilitate this through their focus on building construction and empowerment strategies. Ensure that women, have equal access to employment Can empower women and enhance their visibility in the public arena Encourages more women to enter and stay in the workforce. Can reward use of preventive health care Improves maternal health. Health insurance Makes medical treatment more accessible Sick leave Protects income during periods of illness and enables treatment. Workers compensation and occupational safety policies Source: Adapted from The Contribution of Social Protection and the Millennium Development Goals (World Bank, 2003). protect the occupational health (and Productivity) of workers and guard against work-related disability. 17

18 As the details in Table 3 show, social protection can also contribute significantly to achieving the other MDGs. Social protection programmes increase school (male and female) enrolments, improve health, and reduce child and maternal mortality, thus contributing to all other MDGs reviewed in this study. A 2003 World Bank study reports the following: Community-oriented [social funds] contribute greatly to the creation of social infrastructure, leading in turn to better supply of schools, clinics and other community assets, as has been the case in the Bolivia Social Investment Fund, and other funds in Central America, Peru, Ethiopia, Malawi, Armenia and Angola. This emphasis on infrastructure has had benefits in terms of access to education and health-care, and provides the opportunity to build accessible structures that allow for the inclusion of disabled people. Impact evaluations show that social fund investments resulted in increased enrolment rates in social fund communities in Armenia, Nicaragua, and Zambia. World Bank (2003, pp ) 3.2 Social protection and MDG indicators: Empirical analysis of African data The co-efficients of the correlation reported in Table 4 indicate that the linear association between most MDG indicators and social protection is rather weak. The proxy for social protection is the indicator of social protection and labour rating included in the CPIA policies for social inclusion (from the World Bank World Development Indicators & Global Development Finance database). The World Bank Group defines this indicator as follows: social protection and labour assess government policies in social protection and labour market regulations, which reduce the risk of becoming poor, assist those who are poor to manage further risks better, and ensure a minimal level of welfare to all people. I then try to estimate the correlation between MDG indicators and social contributions (as a percentage of total government revenue) in order to assess whether higher social contributions have a significant and positive contribution to the MDGs. Social contributions include social security contributions by employees, employers, and selfemployed individuals, and other contributions whose source cannot be determined, as well as the actual or imputed contributions to social insurance schemes operated by governments. We note that the linear association between social protection and social contributions is very strong (0.91), while the correlations between social contributions and the indicators for MDGs 1, 2, 3 and 5 is weak (0.3 or less) or has the opposite sign (social contributions reduce progress). Social contributions have a significant and positive correlation with all three indicators of MDG 6. Thus, the co-efficients of simple linear correlation do not provide strong evidence for the contribution (or at least the association) of social protection on progress towards achieving the MDGs. In theory, and given the earlier discussion, we should expect social protection to have a positive contribution to the MDGs (see Table 3). However, there are factors that may affect the effectiveness of social protection. First, the World Bank index of social protection does not take into consideration the aspect of equity, including gender equity, in providing social protection, nor does it account for government efficiency in mobilising revenues. Second, the relationships between social protection and MDG indicators may not necessarily be linear. In some cases, the contributions of social protections may require a minimum level (threshold) in order to have a significant impact, while, in other cases, there may be diminishing returns to social protection. Third, the interplay of social protection and other government policies could impact on the effectiveness of social protection. 18

19 Table 4: Correlations between social protection and MDG indicators Social protection and labour rating Social contributions (% of revenue) Social contributions (% of revenue) 0.91 Goal 1 Employment-to-population ratio GDP growth per person employed Undernourished (% of population) Goal 2 Primary completion rate Net enrolment ratio in primary education Goal 3 Seat held by women in national parliament Ratios of girls to boys in primary & secondary education MDG 4 Proportion of 1 year-old children immunised against measles Infant mortality rate Under-5 mortality rate MDG 5 Maternal mortality rate Adolescent birth rate MDG 6 TB incidence TB prevalence TB detection TB death Proportion of TB cases detected and cured under directly observed treatment short course HIV prevalence Malaria mortality rate Source: World Bank MDG database (2010) and World Bank World Development Indicators & Global Development Finance database (2010). I perform multiple regressions in which I control for several relevant policy variables and try to assess the contribution of social protection to progress towards achieving the MDGs. The following added-variable plots (avplots), which graph the partial correlation between social protection and various MDG indicators, are obtained from these regressions. Figure 8 indicates that social protection (SP) seems to have a positive (and linear) relationship with the employment-to-population ratio. On the other hand, the relationship between social protection and the indicators of MDG 2 (net enrolment ratio in primary education and primary completion rate) is non-linear and has a U shape (Figure 9). This suggests that, for social protection to have a positive contribution to progress towards achieving MDG 2, we need to have high levels of social protection. Similarly, the relationship - portrayed in Figure 10 - between social protection and the indicator of MDG 3 (the ratio of girls to boys in primary and secondary education, and the proportion of seats held by 19

20 women in national parliament) has a U shape, becoming positive only after social protection reaches a critical (high) level. However, once I remove outliers such as Rwanda, the relationship between social protection and the proportion of seats held by women in national parliament becomes statistically insignificant (results not shown). Figure 11 shows that social protection is associated with lower infant mortality and under-5 mortality rates. However, while the relationship of social protection with infant mortality is linear, its relationship with under-five mortality rates is non-linear, suggesting that social protection reduces under-five mortality rates only once it reaches high levels. Likewise, the relationship between social protection and the indicators of MDG 5 (maternal mortality rates and adolescent birth rates) has an inverted-u shape (Figure 12), suggesting that a significant reduction in maternal mortality rates and adolescent birth may require high levels of social protection. Finally, Figure 13 indicates that the relationship between social protection and the six indicators of MDG 6 is negative and linear. Overall, social protection seems to contribute to (or at least is positively associated with) Africa s progress towards achieving the first six MDGs. 20

21 Figure 8: Social protection and MDG1 e( employ-to-population ratio X ) e(sp X ) coef = , se = , t = 2.01 Figure 9: Social protection and MDG2 e ( p r i m a r y s c h o o l c o m p le ti o n r a te X ) e( SP_squared X ) coef = , se = , t = 3.32 e ( N e t e n r o lm e n t r a tio in p r im a r y e d u c a ti o n X ) e(sp_squared X ) coef = , se = , t =

22 Figure 10. Social protection and MDG3 e(r atio o f girls to b oys in prim ary an d seco nda ry X ) e(sp_squared X ) coef = , se = , t = 2.74 e( S ea ts he ld b y w om en in n a tion al p arlia m e nt X ) e( SP_squared X ) coef = , se = , t = 2.3 Figure 11: Social protection and MDG 4 e (m o rta lity ra te, in fa n t X ) e(sp X ) coef = , se = , t = e (m o rta lity ra te, u n d e r-5 X ) e(sp_squared X ) coef = , se = , t =

23 Figure 12: Social protection and MDG 5 e(m a te rna l m ortality rates X ) e(sp_squared X ) coef = , se = , t = e(adolescent birth rate X ) e(sp_squared X ) coef = , se = , t =

24 Figure 13: Social protection and MDG 6 e (T B, in c id e n c e X ) e(tb, prevalence X ) e(sp X ) coef = , se = , t = e(sp X ) coef = , se = , t = e(tb, detection X ) e(sp X ) coef = , se = , t = e(tb death rate X ) e(sp X ) coef = , se = , t = -2.5 e( HIV prevalene X ) e( SP X ) coef = , se = , t = e( Malaria, mortality rate X ) e(sp X ) coef = , se = , t =

25 As stated earlier, there may be major factors affecting the effectiveness of social protection. The list of such factors could be long, and may require econometric estimations to determine which factors have the most impact, but it may also be useful to try to illustrate this point using graphs depicting how the interplay between social protection and some of these factors may influence its effectiveness. The results (plots) are based upon the same regressions from which the plots above were derived. Figure 14: MDG indicators and other government policies e ( p rim a ry s c h o o l c o m p le tio n ra te X ) e (N e t e n ro lm e n t ra tio in p rim a ry e d u c a tio n X ) e( human resource building, policies X ) coef = , se = , t = e(policies for social inclusion and equity X ) coef = , se = , t = 4.55 e(ratio of grils to boys in primary and secondary education X ) e(sp X social inclusion X ) coef = , se = , t = 5.38 The top two graphs in Figure 14 portray the effect of policies for human resource building (an index assessing the quality of national policies and public and private sector service delivery that affect the access to, and quality of, health and education services, including prevention and treatment of HIV/AIDS, tuberculosis, and malaria) and policies for 25

26 social inclusion and equity (an index assessing the quality of policies for social inclusion and equity cluster which includes gender equality, equity of public resource use, building human resources, social protection and labour, and policies and institutions for environmental sustainability) on MDG 2 indicators. The graphs suggest that these policies enhance the effectiveness of social protection. Similarly, policies for social inclusion and equity have a positive impact on the effectiveness of social protection in increasing the ratio of girls to boys in primary education. 4. Summary and discussion In this paper, I first reviewed Africa s progress towards achieving the first six MDGs, by focusing on the main target indicators. Second, I examined the relationship between social protection and various MDG indicators. The foregoing analysis suggests that social protection can be critical to achieving the MDGs. However, its influence on the various MDG indicators can often be non-linear and indirect, through its impact on other factors that, in turn, affect MDG indicators. Indeed, progress towards achieving MDGs may depend on other government polices that are not necessarily MDG-focused polices. As argued by Bourguignon et al. (2008), [d]ue to the diversity of MDGs and their apparent sector-specificity, there is a serious risk that the domestic policy environment and its specific features are not given sufficient consideration at the point of delivery. In a recent report on Achieving the Millennium Development Goals in Africa, Recommendations of the MDG Africa (June 2008), the Steering Group (chaired by the United Nations Secretary General) called for a number of key actions by sector, including support for African governments to launch a Green Revolution to double crop yields, with a focus on providing smallholder farmers with temporary subsidies for fertiliser and better seeds. The Report cites Malawi s agricultural support programme which provides an impressive example of the rapid gains that can be made with improved inputs. The Report makes specific recommendation for all MDGs but the health sector (especially maternal mortality) receives a special emphasis. The report states the following: Africa as a whole is off track to meeting the MDGs on reducing child mortality, improving maternal health and combating infectious disease (i.e., MDGs 4, 5 and 6)... Nevertheless, support for rapid scale-up of proven interventions as well as critically needed investments in basic healthcare systems remains insufficient. In most African countries the basic health infrastructure, human resources, equipment and supplies are inadequate to provide essential maternal, child and reproductive health services, and to control and treat infectious diseases. Malaria and other vector-borne diseases that can be controlled and treated continue to take millions of lives throughout Africa and are spreading to more parts of the continent due to rising temperatures caused by climate change. Interestingly, the preceding analysis suggests that social protection can be very effective in promoting health and reducing infectious diseases. It is also important to assess the role which aid and aid donors should play in extending social protection programmes. The EU, in particular, as the primary aid donor, seems to be a natural candidate for serious debate with SSA countries, and action on social protection and its role in speeding up progress towards achieving the MDGs. A special emphasis should be placed on the design and delivery of social protection to Africa s fragile states due to their weak institutions and government capacity (see Baliamoune-Lutz, 2009b, and Bourguignon et al. 2008). Fragile states tend to have low levels of social cohesion, often re-inforced by lack of social inclusion policies. Social inclusion can have a significant impact on progress towards achieving the MDGs and can influence the effectiveness of 26