83911 Safety Nets Social Protection & Labor Policy Note November 2013 | Number 16 Highlight Poverty Reduction and Since 2009 the World Bank has conducted Investing in People safety nets assessments The New Role of Safety Nets in Africa in a number of Sub- saharan African Experiences from 22 Countries countries. Recently a regional review Victoria Monchuk synthesized the finding of 22 such assessments Abstract to provide an overview of the current status Safety nets are on the rise in Africa, and beginning to evolve from scattered stand- alone programs into systems. Until recently, many African countries approached of safety nets in the social protection on an ad-hoc basis. But when the global crisis threatened recent region and presents progress in poverty reduction, safety nets increasingly began to be viewed as core lessons on how they instruments for poverty reduction in the region. Social protection programming has started to develop from emergency food aid programs to one-off interventions to can be strengthened to regular and predictable safety nets, such as targeted cash transfers and cash-for-work better tackle poverty programs. Some countries, such as Kenya, Mozambique, Rwanda and Tanzania, now and vulnerability. seek to consolidate programs into national systems. But as our review shows, there is still a long way to go. Safety Nets as Tools for Poverty Reduction and Investing in the Poor Over the past two decades, Africa’s strong economic growth has paved the way for poverty reduction. Between 1995 and 2008, the percentage of the African population living in poverty fell from 58 to 48 percent (World Bank 2011). Nevertheless, high poverty levels persist, especially in rural areas, and the gap between income groups is growing. In addition to chronic poverty, vulnerability is high because of frequent environmental and economic shocks. In the effort to increase the momentum toward sustainable poverty reduction, safety nets are an important tool in any country’s development strategy as they are targeted interventions providing regular and reliable support to poor households and help the poor invest in productive and capital-forming activities. Until the recent urgency to strengthen safety nets for the poorest in the face of the global crisis and repeated droughts, social protection has been implemented only on an ad hoc basis in Africa. Over the past few years, in the wake of the global economic 1 Policy Note: Safety Nets November 2013 | Number 16 and food and fuel price crises, a number of countries have started to coordinate their separate safety net programs into a Figure 1: More than half of the countries have national system. There is also momentum throughout the re- prepared or are preparing social protection gion to rationalize public spending to provide more adequate strategies (percent of countries) and targeted support to the poorest. Given the vast extent of poverty in Africa, safety nets cannot reach all of the poor. They need to focus on the extremely poor and on specific vulnerable groups for maximum effect and affordability—not only helping protect them but also providing a ladder out of 32% poverty in the longer term. 45% The Safety Net Experience in 22 African Countries 23% Safety nets have evolved differently across Africa in re- sponse to the specific political economy in each country. Hence, the policy frameworks, approaches, and institu- Social protection strategy exists and is operational tional arrangements that govern safety net systems are not Social protection strategy under preparation homogeneous across the continent. For instance, MICs in No social protection strategy exists southern Africa have strong government-led systems based on horizontal equity, whereas in fragile states and LICs, such Few African countries have well-planned safety net sys- as those in West Africa, the social protection agenda tends tems that can take a strategic approach to reducing poverty to be more donor influenced. Any measures to strengthen (Figure 2). Instead, a multitude of interventions exist that are 100 safety nets need to be designed in ways that take into ac- fragmented and together do not effectively target the poor. count these context-specific factors. In LICs, for example, in West Africa, safety nets are focused on80emergency relief. Few provide continuous support to the Despite this heterogeneity safety nets are taking hold as core poverty reduction instruments. More and more African large number of chronically poor, although such programs are countries are preparing social protection strategies to serve as 60 common in MICs (i.e. in South Africa and Botswana) more the foundation on which to build effective and efficient safety because of the social pension programs in those countries. net systems. Safety nets are also being placed higher on gov- The most common kinds of programs in Africa are school ernment agendas. The review shows that about three-fourths of 40 feeding programs, public works programs, in-kind emer- the countries studied include safety nets as a component of their gency programs and categorical transfer programs (Figure overall poverty reduction strategy and over half have prepared or 20National poverty-targeted cash transfers are not common, 3). are preparing a social protection strategy (Figure 1). although some of the significant number of small programs are being expanded. For example, Rwanda’s Vision 2020 Although safety nets in Africa generally lack strong institu- 0 Umurenge Program. Chronically Temporary Categorically tional homes and coordinating bodies, examples of robust poor/chronically poor/affected vulnerable/special implementation arrangements exist. Responsibility for food insecure by shocks/repeatedly groups government safety net programs is generally spread over a Figure 2: Coordinated Systmes are Safety Net food insecure number of ministries, such as the ministries of social affairs, Uncommon (percent of countries) or other cross-sectoral ministries that often lack significant political power. Meanwhile, fragmented donor support often 100% leaves LICs with a host of isolated small programs that lack coordination or a political champion. For instance, both 14% 32% Liberia and Madagascar have more than five different public works programs, each operated by different agencies. The 80% 45% Ethiopia Productive Safety Net Program (PSNP) is, however, 50% 60% an example of how countries can create effective implemen- tation arrangements that span multiple ministries and pool 36% donor support in one basket. 23% 40% 20% Social protection strategy exists and is operational System of SSN programs exists 0% Social protection strategy under preparation System in progress No social protection strategy exists No system 2 Safety Nets | Human Development Network | The World Bank Lacking long-term, development-oriented safety nets, many evaluations are being undertaken. In the past most impact LICs and fragile states still react to crises by providing emer- evaluations have been for small donor pilots, such as Malawi’s gency relief (Figure 4). Very little information is available Zomba cash transfer program or Mali’s Bourse maman, but about the effectiveness of food distribution and emergency larger programs, such as in Ethiopia, Kenya, and Tanzania, are programs that are common in West Africa (for example, now benefiting from impact evaluations. in Benin, Burkina Faso, Cameroon, Mali, and Mauritania). Countries are increasingly looking to the positive experience of the risk-financing component of Ethiopia’s PSNP. Figure 3: Various Types of Safety Net Programs (percent of countries) 100% 95% 91% 82% 80% 77% 68% 59% 59% 60% 55% 40% 36% 20% 14% 100% 0% School PWPs Categorical General Other in-kind Emerg Fee Poverty 14% Micro- Social care feeding 32% transfers subsidies transfers programs waivers targeted CTs finance services 80% 45% 50% 60% Coverage of the poor and vulnerable by existing safety nets More monitoring data on safety net programs in Africa would help assess their effectiveness. In general, little is 36% is low, although growing in some countries. Taken together, 23% known about the effectiveness of safety nets in Africa. Lack each country’s safety nets cover only a very small share of the total number of poor people. For example, in Benin, the 40% of basic information systems is a crucial weakness. Many countries do not have administrative data on the number net coverage rate of all safety net programs is estimated to be 20% of beneficiaries reached by each program. Programs that only about 5 to 6 percent of the poor. The exception is uni- distribute food in response to emergencies, particularly lack versal social pensions programs common in southern Africa, Social protection strategy exists and is operational System of SSN programs exists 0% data. The impact of safety nets on poverty and welfare, where which cover a large share of the elderly population. However, Social protection strategy under preparation System in progress known, has generally been positive No social protection but strategy mixed. More impact exists the coverage of poverty-targeted No system programs in many MICs is still limited. Figure 4: Focus of Safety Nets Programs Targeted programs are still not widely available in Africa. (percent of countries) Poverty-targeted programs are rare and mainly practiced 100 in small pilot initiatives. Only 20 percent of the programs reviewed use some form of (proxy) means testing to target the poor (Table 1). In practice, safety nets in Africa use a wide 80 range of targeting mechanisms and often combine more than one. A key question is how well African safety nets are able to 60 identify and reach the poor and vulnerable, especially those in extreme poverty and vulnerability, given data and capac- ity constraints. Improving the extent to which safety nets can 40 reach the poor also depends on political viability. With better analysis, in part from safety net assessments, 20 several countries are on a path toward developing more ef- fective safety net systems. Thirt-six percent of the countries 0 analyzed are building a system whereas half need to make Chronically Temporary Categorically more progress (Figure 2). A number of countries are actively poor/chronically poor/affected vulnerable/special increasing the effectiveness and scale of existing programs food insecure by shocks/repeatedly groups food insecure (such as Tanzania’s TASAF, Ghana’s LEAP, and Kenya’s CT- 3 Policy Note: Safety Nets November 2013 | Number 16 OVC). In countries like Rwanda and Tanzania, more sustain- should be scaled up, whereas ineffective programs should be able programs are starting to appear, backed by influential gradually phased out. The allocation of spending on scattered ministries such as finance and planning. More countries are emergency programs shows that, typically, neither donors nor moving toward building safety net systems and programs governments have focused on safety nets for addressing long- that are flexible to respond to crises (i.e. Cameroon, Mali, term poverty. This situation is now starting to change. Ethiopia, Mozambique, Niger, and Senegal). Kenya, Mozambique, Rwanda, and Tanzania are moving to harmonize programs for enhanced efficiency and coverage. Safety nets in Africa use a wide range of Table 1:  targeting mechanisms (percent of programs) Safety nets should be built on the basis of strong operational tools and coordinating bodies to ensure effective implementation Targeting method Frequency and oversight. Basic operational tools, such as beneficiary regis- tries, targeting, payment and M&E systems provide a platform Multiple 57% that enables programs to deliver support effectively to targeted Geographic 49% groups. More work is needed to understand how food-based Self-targeted 32% programs should play a part in improved safety net systems in Community-based/validated 30% Africa. Categorical 26% These systems need to be built during stable times so that they PMT/means-tested 20% can respond quickly to crisis. Establishing such systems takes time. Most countries in Africa (including Benin, Cameroon, Universal (excluding subsidies) 12% Mauritania, and Sierra Leone) did not have safety nets capable of effectively responding to the recent global crises but had to Well-targeted safety nets are affordable in Africa, especially resort to inefficient and expensive universal handouts. if inefficient spending can be reduced and redirected to the The role of safety nets in the context of subsidy reform and poorest. Especially in LICs spending on safety nets is low use of mineral resource proceeds should be further explored. compared to other countries world-wide (Table 2). Careful political economy considerations are important ■■ In LICs, because poverty is high and government income when balancing tightly targeted programs with other invest- low, attracting donor funds to support safety nets will ments that can benefit a wider set of people and contribute continue to be vital in both the short and long run. With to improved social outcomes. As more African countries will the exception of universal programs such as old-age benefit from newfound mineral wealth, getting the balance benefits, donors finance a large share of safety nets in right between effectively targeting those funds to the poor- Africa—over 80 percent of in Burkina Faso, Liberia, and est through safety nets or other investments in social services Sierra Leone (Table 2). and building both a fiscally and politically sustainable social protection system will be especially important. ■■ In MICs, however, current public budgets are sufficient to provide adequate support to the poorest. For instance, in Cameroon, estimates indicate that it would cost only 0.5 Implementing the Vision - what can percent of gross domestic product to provide an adequate other countries learn? safety net to half the chronic poor. The following recommendations apply to countries that are clas- sified as “Early stage/no plans” because they have no solid plans ■■ General subsidies are costly mechanisms for redistribut- for a national safety net system and/or no adequate programs in ing income and often do not benefit the poor, as is true place (mainly LICs and fragile states but also some MICs): of the fuel subsidies in Cameroon, Mauritania, and Sierra Leone. Reducing poorly targeted programs and subsidies a. Develop and operationalize a safety net strategy and can make fiscal space for more effective and better-target- coordinate scattered donor support. ed safety nets. Likewise, well-performing safety nets can be important mitigating mechanisms to facilitate reform b. Build key organizational tools on which safety net of expensive general subsidies. should be based such as a basic monitoring system, identification and targeting mechanisms, and a payment system that can channel transfers to the targeted poor Moving Forward to Strengthen Safety and vulnerable groups. Nets in Africa c. Develop a few key safety net programs based on a careful Coordinating safety nets into a coherent system should be a analysis of the country’s needs. This small number of key priority. Within a given country, a small number of coor- safety net interventions should: (i) provide regular sup- dinated and well-functioning programs can effectively and port to the chronic and extreme poor and (ii) be able to feasibly meet the needs of the poorest, as happens in Rwanda. expand and contract to provide assistance to vulnerable Programs that are well targeted and serving the poor effectively 4 Safety Nets | Human Development Network | The World Bank Table 2: Cost and Financing of Safety Nets in Africa Spending on Safety Nets (percent of GDP; incl. government and donor spending) Percent of Pecent financed by Notes total Govern- Exluding Including general government ment general general subsides spending (excl. sub- Country subsides subsides only (excl. subsidies) sidies) Donors Years Benin 0.3 0.9 0.5 1.1 35 65 Ave. 2005-10 Botswana 3.7 3.7 0 9.5 100 0 Ave 2009/10-2012/13 Burkina Faso 0.6 1.3 0.7 <1.0 20 80 Ave. 2005-09 Cameroon 0.2 1.6 1.4 1.5 23 77 Ave. 2008-10 Ethiopia 1.2a 1.2a 0 . 0 100 2009 Kenya 0.8 0.8 0 1 29 71 2010 Lesotho 4.6 4.6 0 8 2010/11 Liberia 1.5 1.5 0 4.4 6 94 Ave. 2008-11 Madagascar 1.1 1.1 0 5 . . 2010 Mali 0.5 0.5 0.1 . 40 60 Ave. 2006-09 Mauritania 1.3 3.2 1.9 4.6 62 38 Ave. 2008-13 Mauritius 4.4 5.2 0.8 9 . . 2008/09 Mozambique 1.7 3.1 1.4 . 38 62 2010 Niger . . . 1.0-5.0 33 67 Ave. 2001-06 Rwanda 1.1 1.1 0 . . . 2010/11 Sierra Leone 3.5 5.6 2.1 13.1 15 85 2011 South Africa 3.5 . . . . . 2010 Swaziland 2.1 2.1 0 . . . 2010/11 Tanzania 0.3 0.3 0 1 . . 2011 Togo 0.5 1.3 0.8 1.8 25 75 Ave. 2008-10 Zambia 0.2 2.1 1.9 . 25 75 2010/11 Average 1.7 2.2 0.6 4.4 32 68 . Ave. LICs 1.1 1.7 0.6 3.7 27.5 72.5 . Ave. MICs 2.7 3.2 0.7 7 49.3 50.7 . Ave. ECA 1.8b 1.8b . . . . Latest 2008-10 Ave. LCR 1.1c 1.1c . . . . 2010 Ave. MNA 0.7 6.4d . . . . Latest Sources: Country safety net assessments, Silva et al. (2013), Woolard and Leibbrandt (2010), World Bank (2012b), World Bank (2012c) Notes: Numbers may not add up due to rounding errors. The spending data presented include donor financing except general budget support but excludes funding by the private sector. a Only includes PSNP and does not include spending on other safety net programs. b Government spending only, includes subsidies in very rare cases, where data is available, latest year 2008-11. c Year 2010 for 10 countries in Latin America and the Caribbean. d Latest year available for 11 countries in the Midde East and North Africa region. Spending includes general subsidies and ration cards. 5 Policy Note: Safety Nets November 2013 | Number 16 households in the case of emergencies or seasonal fluctua- The following recommendations apply to countries that are tions in consumption. classified as “Established” and that already have a national safety net/social protection system in place (mainly of MICs): The following recommendations apply to countries that are clas- sified as “Emerging” because their safety net systems are in the a. Strengthen the existing safety net and social protection process of being developed (mainly of LICs but some MICs): system to ensure that it is reaching the extreme poor. Even when countries have well-established programs, some gaps a. Continue to reform existing categorical, universal, and ad can remain with some members of the poorest not receiv- hoc emergency programs to make them more efficient tools ing sufficient support. for reducing poverty. For instance, social pension programs could be more cost effective if they were targeted only to b. Continue harmonizing and consolidating fragmented those elderly who are poor. Efforts to reallocate universal safety net programs. This may require policymakers to subsidies and expensive emergency programs towards bet- reduce the number of existing programs by assessing their ter targeted safety nets should continue. individual targeting effectiveness and impact vis-à-vis other interventions. b. Continue scaling up a few key well-targeted programs and continue harmonizing and consolidating programs. c. Continue strengthening targeting, M&E, grievance, and Unique beneficiary registration systems should be explored. payment systems. References Grosh, Margaret, Carlo del Ninno, Emil Tesliuc, and Azedine Ouerghi. 2008. For Protection and Promotion: The Design and Implementation of Effective Safety Nets. Washington, D.C.: World Bank. Silva, Joana, Victoria Levin, and Matteo Morgandi (2013). Inclusion and Resilience: The Way Forward for Social Safety Nets in the Middle East and North Africa. MENA Development Report. Washington, D.C.: The World Bank. Woolard, Ingrid and Murray Leibbrandt. (2010) “The Evolution and Impact of Unconditional Cash Transfers in South Africa.” Southern Africa Labour and Development Research Unit, University of Cape Town. World Bank. 2011. “Africa’s Future and the World Bank’s Support to It: Africa Regional Strategy.” World Bank, Washington, D.C. World Bank. 2012a. Managing Risk, Promoting Growth: Developing Systems for Social Protection in Africa—The World Bank’s Africa Social Protection Strategy, 2012–2022. Washington, D.C.: World Bank. World Bank, 2012b. “Europe and Central Asia Social Protection Database.” World Bank, 2012c. “Latin America and the Caribbean Social Protection Database.” World Bank. 2013. “Securing the Transformational Potential in Africa’s Mineral Resources.” PowerPoint presentation, World Bank, Washington, D.C., February. The findings, interpretations, and conclusions expressed herein are those of the author(s), and do not necessarily reflect the views of the International Bank for Reconstruction and Development/The World Bank and its affiliated organizations, or those of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. For more information, please visit www.worldbank.org/sp. 6 © 2013 International Bank for Reconstruction and Development / The World Bank