The way forward for social safety nets in Burkina Faso The way forward for social safety nets in Burkina Faso Frieda Vandeninden, Rebekka Grun, and Amina Semlali Social Protection and Jobs Africa Region World Bank © 2019 International Bank for Reconstruction and Development/ The World Bank 1818 H Street, NW Washington, DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org The findings, interpretations, and conclusions expressed here do not necessarily reflect the views 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. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. The International Bank for Reconstruction and Development/The World Bank encourages dissemination of its work and will normally grant permission to reproduce portions of the work promptly. For permission to photocopy or reprint any part of this work, please send a request with complete information to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA, telephone 978-750-8400, fax 978-750-4470, http://www.copyright.com/. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA, fax 202-522-2422, e-mail pubrights@worldbank.org. Design/layout and editing. Nita Congress Photo credits. Cover: Kelley Lynch/GPE; chapter 2: Dominic Chavez/The Global Financing Facility; all others: Ollivier Girard/CIFOR. Creative Commons CREATIVE license: https://creativecommons.org/licenses/ by-nc-nd/2.0/legalcode contents acknowledgments vi foreword viii abbreviations x executive summary xii chapter 1: introduction xxiv Background 2 Purpose and structure 5 Conceptual framework 6 chapter 2: social protection needs: analysis of risks and vulnerabilities 10 Poverty and vulnerability trends 12 Trends in nonmonetary welfare indicators 18 Poverty along the life cycle 20 chapter 3: institutional framework and spending review 30 An emerging social protection strategy 32 Institutional and coordination arrangements 36 Social spending trends 41 Focus on social safety net spending: a multitude of small programs 50 chapter 4: social safety nets assessment: do the programs meet the needs? 54 Coverage, gaps, and overlaps 56 Targeting accuracy and cost-effectiveness 68 Sustainability 76 Monitoring and evaluation 79 chapter 5: recommendations and concluding remarks 80 Recommendations 82 Concluding remarks 92 appendix A: Burkina Faso’s main social safety net programs 93 appendix B: HEA in Burkina Faso 115 appendix C: PMT in Burkina Faso 117 appendix D: social safety net coverage by region and share of poor 119 appendix E: glossary of social protection terms 120 references 122 v acknowledgments T he team offers particular thanks to the government of Burkina Faso, especially the offi- cials working on social protection programs, as well as the partners and stakeholders supporting the strengthening of the nation’s social protection system. The report was produced under the overall leadership and guidance of Rebekka Grun, Team Leader and Senior Economist. The main authors of this report are Frieda Vandeninden, Econ- omist; Amina Semlali, Senior Human Development Specialist; and Rebekka Grun. Invaluable team members include Inés Rodriguez Caillava, Operations Analyst, who assisted with numerous aspects of the report; and Chantal Lewis, Consultant, who coordinated the field data collection. Lydie Anne Billey, Program Assistant, provided administrative sup- port; Bruno Bonansea, Cartographer, produced the report maps. Bram Vandermeulen, then a student at Maastricht University, helped develop different simulation scenarios on gas subsidy reform. The authors also acknowledge the valuable contributions by editor and graphic designer Nita Congress. The authors would particularly like to thank the local Social Protection team in Burkina Faso: Ayaba Gilberte Kêdoté, Social Protection Specialist, Cath- erine Marie Z. Compaore, Team Assistant, and Roland Berenger Béréhoudougou, Program Coordinator, Sahel Adaptive Social Protection. The team is very grateful for the help Karime Ganemtoré, Permanent Secretary at the National Council for Social Protection, and his team provided. The authors give special thanks to Aminiata Nana Tou, Director of Budget Planning, Ministry of Finance; Tamboura Nadine, Director of NGO Cooperation, Ministry of Finance; Stella Somé, Director, Ministry of Public Affairs; Soumaila Sakandé, Coordinator, High Intensity Labor and Public Works Project; Abdoulaye Sawadogo, Department of Scholarships, FONER; Sylvain Nkenkeu, Senior Social Inclusion Program Officer, UNICEF; Francis Oubda, Social Policy Spe- cialist, UNICEF; and Fanta Sedogo, Health and Social Protection Program Manager, European Union. The report benefited greatly from comments and input by peer reviewers Emma Mercedes Monsalve Montiel, Economist; and Inés Rodriguez Caillava, Operations Analyst; as well as from comments and input by Ruslan G. Yemtsov, Lead Economist; Roland Berenger vi Béréhoudougou, Program Coordinator; Mame Fatou Diagne, Senior Economist; Prospere R. vii acknowledgments Backiny-Yetna, Senior Economist, Damien de Walque, Senior Economist; Mariam Diop, Senior Operations Officer; Alexis Lucien Emmanuel Madelain, Senior Energy Specialist; and Michael Weber, Senior Economist. Moreover, preparation of this report was significantly aided by the guidance, advice, and support of many managers and colleagues within the World Bank, including Jehan Arulpragasam, Practice Manager; and Cheick Kanté, World Bank Country Manager, Burkina Faso. Finally, the report greatly benefited from the foundational work in the field of social safety nets in Burkina Faso, notably the 2011 social safety nets review authored by Cécile Cherrier, Carlo del Ninno, and Setareh Razmara. The authors are grateful for the assistance and support received, and apologize for any omis- sions. foreword S ocial safety nets—predictable cash transfers to poor households, often in exchange for children going to school or for regular health checkups—have become one of the most effective poverty reduction strategies, helping the poor and vulnerable cope with crises and shocks.  Each year, social safety net programs in developing countries lift an estimated 69 million people out of absolute poverty and some 97 million people from the bottom 20 per- cent—a substantial contribution in the global fight against poverty. Almost every developing country now has a set of social assistance programs and other social protection measures— conditional and unconditional cash transfers, school feeding programs, public works, and some form of health insurance and pension coverage for the elderly—interventions that, in principle, help poor households manage risks and invest in their livelihoods. For a long period, the majority of countries in Africa implemented social safety nets and social protection programs only on an ad hoc basis. However, in the wake of the global economic, food, and fuel price crises starting in 2008, policy makers in Africa increasingly viewed social safety nets as core instruments for reducing poverty and helping vulnerable households man- age risk. Life in today’s globalized world is indeed fraught with a complex mix of risks, and rapid climate change is among the most pressing. We are now seeing how social safety net programs can play a huge role in disaster preparedness and response after a crisis. When the Southern Africa region was hit by the worst drought in 35 years, cash transfers were the primary response to support the recovery for some 32 million people who were food insecure due to the drought.  In less than a decade, social safety nets in Sub-Saharan Africa have become a core part of development strategies. Throughout the continent, social safety nets have changed the lives of millions of vulnerable people for the better. What is more, Africa has indeed become a great innovator in social safety nets, pushing the frontiers in many areas. But the countries also face many challenges in their attempts to bring their social safety nets to scale and ensuring their sustainability. To support governments in their quest to improve the efficiency and effectiveness of safety nets in their countries, the World Bank’s Africa Region undertook social safety net or social protection assessments in a number of coun- tries in Sub-Saharan Africa, including Burkina Faso. These assessments analyze the status viii ix foreword of social protection programs and safety nets, their strengths and weaknesses, and identify areas for improvement—all with the aim of helping governments and donors to strengthen African safety net systems and social protection programs to protect and promote poor and vulnerable people. They were all carried out with the explicit aim of informing governments’ social protection policies and programs. The government of Burkina Faso has shown a strong interest in learning how to strengthen its social safety net provision to better support the country’s poorest and most vulnerable households. This book was produced in response to a request by its Ministry of Economy and Finance. I am particularly excited to share the main takeaway from this book: With a few focused courageous policy decisions, several of the issues of the Burkinabe social protection system can be converted into opportunities. Most importantly, with improved targeting, Burkina Faso could technically eradicate poverty! The size of the actual poverty gap equals 2.26 percent of gross domestic product, which is close to the actual spending on social safety nets. This would be achievable by targeting existing social safety net expenditures better. Reallocating regressive subsidies and programs that mainly benefit the rich would open up the needed fis- cal space. Finally, it is worth noting that this book is relevant beyond Burkina Faso. It offers important lessons from which a global audience can benefit. I hope you enjoy reading it as much as I have. Michal Rutkowski Senior Director and Head of Global Practice Social Protection and Jobs World Bank Group abbreviations CARFO Independent Pension Fund for Civil Servants (Caisse autonome de retraite des fonctionnaires) CIOSPB National Center for Information, Educational and Vocational Guidance and Scholarships (Centre national de l’information, de l’orientation scolaire et professionnelle, et des bourses) CNSS National Social Security Fund (Caisse nationale de sécurité sociale) CNPS National Council for Social Protection (Conseil National de Protection Social) CRS Catholic Relief Services ECD early childhood development EMC Continuous Multisectoral Survey (Enquete Multisectorielle Continue) FONER National Fund for Education and Research (Fonds National pour l’Éducation et la Recherche) GDP gross domestic product HEA household economy approach M&E monitoring and evaluation NGO nongovernmental organization PEJDC Youth Employment and Skills Development Project (Projet d’emploi des jeunes et de développement des compétences) PMT proxy means testing PTR-HIMO Labor-Intensive Roadwork Project (Projet Travaux Routiers à haute intensité de main d’œuvre) SAP Early Warning System (Système d’Alerte Précoce) SONAGESS National Society for the Management of Food Security (Société nationale de gestion des stocks de sécurité) x THIMO Labor-Intensive Public Works (Travaux a haute intensité de main d’œuvre) xi abbreviations UNICEF United Nations Children’s Fund WFP World Food Programme WHO World Health Organization All dollar amounts are U.S. dollars unless otherwise indicated. For currency conversions, the average 2018 exchange rate is used: $1 = CFAF 565.71. executive summary Introduction and purpose The government of Burkina Faso has a strong interest in strengthening its social safety nets provision to better support the country’s poorest and most vulnerable households. It has demonstrated this commitment through past investments in social pro- tection. Against a backdrop of limited public finances and budgetary constraints, it is critical to ensure that the resources allocated for social protection—and in particular for social safety nets— are cost-effective. This report responds to a request by the Burkinabe Ministry of Finance to „„ Enhance knowledge about the current state of social safety nets and assess their effectiveness in meeting the needs of the poor; „„ Inform a debate on feasible reform and policy options to make social safety nets in Burkina Faso more effective and of greater impact and able to contribute to a con- solidation of expenditure. Main challenges demanding renewed social safety nets Poverty, risk, and vulnerabilities The country’s sustained economic growth has not benefited the majority of Burkinabes. An annual gross domestic product (GDP) growth rate exceeding 6 percent, on average, from the early 2000s to 2014 did not translate into comparable per capita gains, which only increased 2.6 percent from 2006 to 2013. A high demographic growth rate is an obstacle for future poverty reduction. With 5.4 children per woman, Burkina Faso has one of the highest fertility rates in the world, offset- ting GDP growth and hampering rapid reductions in poverty. The population is projected to xii increase from an estimated 19.2 million in 2017 to 29 million in 2030. xiii executive summary Although the poverty headcount ratio has declined, the absolute number of poor has increased. Further, most of the population lives near the poverty line, making them more vulnerable in case of adverse events. As of 2014, 40.1 percent of Burkinabes were poor and lived on less than $1.90 a day. Using the international poverty line standard of $3.10 a day, Burkina Faso’s poverty rate increases to 75 percent. Burkina Faso increasingly faces adverse natural conditions and is highly vulner- able to exogenous shocks and regional insecurities. This high level of exposure to shock exacerbates the vulnerability of the poor in particular; it also increases the risk of the nonpoor falling into transitional poverty. Households hit by shock are often forced to adopt FIGURE ES.1  Population exposure to shock and negative coping strategies, such as selling assets, poverty rates which can lead to chronic poverty. Only 2.4 percent of all households affected by shock receive support from Percentage of population either the government or nongovernmental organizations, 80 Urban Rural All although several food programs have been implemented to mitigate the impact of shocks. 60 40 Ninety percent of the poor live in rural areas. Expo- sure to, and the impact of, shocks is greater in rural areas 20 (figure ES.1). Reasons for this include the fact that such areas tend not to be reached by services; and the population is 0 dependent on climate-sensitive agriculture. The increased vul- Exposure to ≥ 1 Poverty rate Transient poverty shock per year rate nerability of the rural population in turn exacerbates poverty. Disparities in access to public facilities remain SOURCE: Authors’ calculations based on EMC 2014. high in rural areas. Examples are provided in figure ES.2. FIGURE ES.2  Urban and rural population access to public facilities Percentage of population 100 Urban Rural 80 60 40 20 0 Secondary Roads Sanitation Public Primary school Food market Drinking water school center transport SOURCE: Authors’ calculations based on EMC 2014. NOTE: Access means within a 30-minute walk. xiv THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO Poverty along the life-cycle Looking at poverty headcounts by age groups and risks along the life cycle offers important insights on where public interventions should focus. Risks are not homogeneously distributed and are typically higher in earlier stages of life, with important long-term—and sometimes irreversible—consequences in later stages of life (figure ES.3). FIGURE ES.3  Key risks, the life cycle, and social protection in Burkina Faso RISKS NEEDS „„ Malnutrition: The percentage of underweight children „„ A comprehensive early childhood reached 19.2 in 2016 development strategy „„ Mortality: Under age 5 mortality rates remain very high „„ Structural programs to improve nutri- at 88.6 per 1,000 in 2016 tion, addressing supply and demand „„ Early childhood development: Policies are largely barriers lacking, and only 2% of children participate in early childhood education programs—one of the lowest rates EXPOSURE TO SHOCKS in the world „„ School dropout rates: 50% of primary school–age „„ Inclusive programs to promote education children and 57% of lower secondary age—which and address supply-side constraints increases to 69% and 77% for the poorest quintile „„ The provision of alternative education „„ Few teachers: There are 55 students on average per programs, such as training and intern- primary school class ships and “second chance education” „„ Literacy: 61% of youth are illiterate „„ High subsistence agriculture: 92% live off of farm- „„ Secure jobs in agricultural areas by ing and not from selling produce promoting mobility from one sector to „„ Unstable jobs: Work duration is only 6.5 months a year the other and encourage reconversion (5.8 for agriculture) through income-generating activities „„ Income precariousness: 35% of agricultural workers „„ Adaptive social safety nets that have a second job; 78% of agricultural workers face at address shocks from agricultural products least one natural hazard „„ Improve literacy and new qualification through training and subsidized work „„ Lack of stable income and pension: Only 2.8% have „„ Targeted social pensions pensions, and none in the bottom quintile have pensions „„ Improved access to health facilities „„ Lack of access to health care and illiteracy: 90% are illiterate and only 50% consult a health provider in case of health issues „„ Potentially weak family support xv executive summary Social protection expenditure trends Overall, social protection expenditure has increased at a steady pace over the past decade. However, spending trends have varied by category of social protection (figure ES.4). FIGURE ES.4  Social protection expenditures by category: 2015 Expenditures for energy subsidies have experi- enced a downward trend. Electricity subsidies, for Billion CFAF example, weighed down the national budget during the 4.5 4.0 last decade; in 2016, the government took major steps to 3.5 address this. 3.0 2.5 2.0 Expenditures for social safety nets increased over 1.5 1.0 the 2005–15 decade. These expenditures accounted for 0.5 0.3 percent of GDP in 2005 and rose to 2.3 percent in 2015, 0 Social General Social Labor mkt Total social indicating an increasing appetite for finding new methods safety nets subsidies insurance programs protection of protecting the poor (figure ES.5). On average, Burkina Faso outspends other Sub-Saharan countries on social SOURCES: Estimates based on CNPS 2015b, 2016b; SE-CNSA 2015, 2016. safety nets relative to GDP. Composition of social safety FIGURE ES.5  Social safety net expenditure total and as a percentage of GDP, 2000–15 net expenditures % GDP Million CFAF Burkina Faso’s social safety net programs are frag- 2.5 160 mented. The country offers are more than 100 small social 140 safety net programs, 80 of which account for less than 2.0 120 1 percent of total social safety net expenditure. 1.5 100 80 The Ministry of Social Action has the largest share of pro- 1.0 60 grams (45), but the smallest share of total expenditure 40 (3.4 percent). The Ministry of Economy and Finance has 0.5 20 the largest expenditure share (28 percent) and the smallest 0 0 program share (five). 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 Social safety net expenditures mainly go toward SOURCES: World Bank 2012 for 2005–10; estimates for 2014–15 are based on in-kind programs. Only 36 percent of expenditures are CNPS 2015b and 2016b, and SE-CNSA 2015 and 2016. provided in cash (table ES.1). NOTE: No social safety net estimates were produced for 2011–13 given the scar- city of data. The first CNPS annual report was produced in 2013, but the effort to track program expenditures was only at its beginning, and data were incom- Social safety net coverage plete and not comparable with previous estimates. Social safety net coverage is not in line with pov- erty: only 2.6 percent of the entire population benefits from safety net programs. For example, the fourth richest quintile (Q4) benefits more from all safety nets than does the absolute poorest quintile (Q1); scholarships benefit next to no poor (figure ES.6). xvi THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO TABLE ES.1  Social safety net spending on main programs: 2015 Program type Program name % of total SSN spending In kind School feeding/take home School canteens 13.2 Cash Public works, workfare, and direct job creation Cash for work (improve production) 11.1 In kind Other food/in kind/transfer Sponsorship program 7.3 In kind Nutritional (therapeutic, supplementary, and PLHIV) Malnutrition—prevention 7.0 In kind Nutritional (therapeutic, supplementary, and PLHIV) Malnutrition—severe 6.1 In kind Food distribution Targeted food subsidies 5.4 Cash Scholarship benefits School support (high school) 5.4 In kind Nutritional (therapeutic, supplementary and PLHIV) Malnutrition—moderate 4.8 Cash Scholarship benefits Scholarships (high school) 4.6 In kind Other food/in kind/transfer Subsidies of fertilizers 4.0 In kind Nutritional (therapeutic, supplementary, and PLHIV) HIV care program 3.5 In kind Other food/in kind/transfer Support to informal workers 2.9 In kind Other food/in kind/transfer Distribution of food for cattle 2.6 In kind Other food/in kind/transfer Support for rain production 2.5 In kind Emergency support (including refugees/returning migrants) Food distribution to household victims of shocks 2.5 and natural catastrophes In kind School supplies School supplies 2.3 16 programs 85.3 NOTE: PLHIV = people living with HIV; SSN = social safety net. FIGURE ES.6  Social safety net coverage by program and quintile Percent 4 Total Q1 Q2 Q3 Q4 Q5 3 2 1 0 All social Scholarships Government support Other transfers Support from Food distribution assistance (in case of shocks) NGOs SOURCE: Authors’ calculations based on EMC 2014. NOTE: NGO = nongovernmental organization. xvii executive summary Social safety net coverage is not in line with vul- FIGURE ES.7  Social safety net coverage by age group nerability across the life cycle. Children are the and poverty headcount ratio poorest and most vulnerable members of the population, yet only six of Burkina Faso’s main programs focus on the Percent Social safety net coverage Poverty rate 0–5 age group. Furthermore, only 2 percent of the country’s 45 children benefit from crucial early childhood development programs; this rate of coverage is the second lowest in the 30 world after Afghanistan (figure ES.7). 15 Social safety net coverage is not aligned with pov- erty across the country’s regions. This is illustrated in 0 age 0–5 age 6–17 age 18–34 age 35–59 age 60+ figure ES.8. SOURCE: Authors’ calculations based on EMC 2014. FIGURE ES.8   Social safety net coverage and poverty headcount ratio, by region a. Social safety net coverage b. Poverty headcount ratio 4°E 2°E 0° 2°E 4°E 2°E 0° 2°E IBRD 44748 | JANUARY 2019 IBRD 44745 | JANUARY 2019 This map was produced by the Cartography Unit of the World This map was produced by the Cartography Unit of the World MALI Bank Group. The boundaries, colors, denominations and any MALI Bank Group. The boundaries, colors, denominations and any other information shown on this map do not imply, on the part of other information shown on this map do not imply, on the part of the World Bank Group, any judgment on the legal status of any the World Bank Group, any judgment on the legal status of any territory, or any endorsement or acceptance of such boundaries. territory, or any endorsement or acceptance of such boundaries. SAHEL SAHEL 14°N NIGER 14°N 14°N NIGER 14°N NORTH CENTRAL NORTH NORD CENTRAL NORTH BOUCIE DU BOUCIE DU MOUHOUN PLATEAU MOUHOUN CENTRAL PLATEAU EAST CENTRAL CENTRAL CENTRAL EAST 12°N  12°N 12°N  12°N CENTRAL WEST CENTRAL WEST CENTRAL CENTRAL EAST SOUTH CENTRAL EAST CENTRAL HAUTS-BASSINS SOUTH HAUTS-BASSINS BENIN BENIN SOUTH TOGO TOGO SOUTH CASCADES WEST 0° 2°E CASCADES 0° 2°E WEST 10°N  GHANA BURKINA FASO 10°N  GHANA BURKINA FASO SOCIAL SAFETY NET COVERAGE POVERTY HEADCOUNT RATIO 0 20 40 60 80 100 Kilometers 0 20 40 60 80 100 Kilometers CÔTE D'IVOIRE 0 20 40 60 80 Miles CÔTE D'IVOIRE 20 0.28 6.62 0 40 60 80 Miles 9.68 70.3 4°E 2°E 4°E 2°E SOURCE: Authors’ calculations based on EMC 2014. Targeting Targeting is not well aligned with poverty. Programs mainly target beneficiaries on a geographical basis. Data show that the largest concentration of beneficiaries of cash transfers (34.7 percent) is in the Central region, with only 8 percent from the North region, 6 percent from the East, and 3 percent from Boucle du Mouhoun—which are the three poorest regions. Expenditures on “other transfers” are mainly directed toward the richest quin- tile (62 percent), with one-quarter accrued by the second poorest quintile. There is a distinct pattern in the distribution of beneficiaries: even though many beneficiaries are in the second poorest quintile, their transfers are small; while the large transfers are mainly col- lected by a few rich households (figure ES.9). xviii THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO FIGURE ES.9  Social safety net targeting, by share of benefits and beneficiaries a. Distribution of benefits b. Distribution of beneficiaries 100 100 Q5 80 Q4 80 Q3 60 Q2 60 Q1 40 40 20 20 0 0 Scholarships Other Scholarships Government Other Support from Food transfers support (in case of transfers NGOs distribution shocks) SOURCE: Authors’ calculations based on EMC 2014. NOTE: NGO = nongovernmental organization. Sustainability There are financing sustainability concerns with regard to Burkina Faso’s social safety nets. Only 50 percent of safety net programs are entirely financed by the govern- ment, which may undermine continued program predictability. Low execution rates are also a cause for concern. Sustainability is further challenged by expenditure delays, resulting in low execution rates (59 percent of the allocated social safety net budget, excluding education and health programs). These expenditure delays raise questions about governance. Subsidies By redirecting energy subsidy expenditures (from the rich) toward social safety nets (to the poor), poverty and inequality would be improved. One of the fundamen- tal problems with energy and fuel subsidies is that they are disproportionately concentrated in the hands of the richer segments of the population. But there is an increasing awareness that there are less expensive yet more effective methods of protecting the poor. The elimination of energy subsidies would not affect the poor, as the poorest barely consume any energy products (figure ES.10). Given the notable budget allo- cated to the sector (1.05 percent of GDP in 2015), savings from phasing out subsidies could enable a substantial increase in the social safety net budget. Simulations conducted for this review show that removing the gas subsidy alone would enable savings equivalent to 2.31 percent of GDP without affecting the poverty rate and the poverty gap. xix executive summary FIGURE ES.10  Share of household income spent on energy consumption, by decile a. Gas b. Electricity c. Fuel Percent Percent Percent 80 80 80 60 60 60 40 40 40 20 20 20 0 0 0 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 Decile (poorest to richest) Decile (poorest to richest) Decile (poorest to richest) SOURCES: Authors’ calculations based on EMC 2014; MICA-MEF 2014a, 2014b, and 2014c. Institutional arrangements Institutional arrangements and interministerial coordination mechanisms are the prerequisite for an effective social protection system to avoid overlap and duplication and to ensure a more coherent approach. A single beneficiary registry minimizes inclusion and exclusion errors, while facilitating transparency and providing governments and partners with a cen- tral mechanism to identify potential program beneficiaries. The initiation of a much-needed such registry in Burkina Faso to identify, accurately target, and keep track of beneficiaries began in 2015 with the launch of the World Bank–supported unconditional cash transfer program Burkin-Naong-Sa Ya. Notable efforts have been made to create a conducive institutional environment, such as the creation of new coordinating institutions since 2013 for improved cross-sectoral cooperation; and the weaving together of the many sectoral strategies into a single National Social Protection Strategy. Despite these efforts, Burkina Faso’s social protection system and approach remains highly fragmented and weak. Most programs continue to operate in silos, with little communication between sectors and few attempts to explore synergies. Moreover, despite the adoption of the National Social Protection Strategy, numerous additional strate- gies have been endorsed. The way forward Recommendation 1: Strengthen the governance of the social safety net system. Despite progress in reducing poverty and vulnerability in the past decade, Burkina Faso’s xx THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO daunting human capital challenges and poverty levels call for a systematic build-up of the social protection system and of social safety nets in particular. The system’s institutional gov- ernance needs to be strengthened. In this context, the following actions should be prioritized. Action 1.1: Create a framework for multisectoral coordination. „„ Identify ministerial comparative advantages and clearly establish key actors’ respon- sibilities. „„ Review the institutional set-up under the stewardship of the central government. „„ Use one overarching national social protection strategy. „„ Improve information and data collection systems and develop ministerial reporting mechanisms. „„ The National Council for Social Protection should assign internal groups with time- bound and specific output and outcome goals to ensure prioritization. Action 1.2: Monitor and evaluate program effectiveness and impact—and address accordingly. „„ Establish a robust system of monitoring and evaluation to facilitate informed decision making. „„ Evaluate the impact of the various small safety net programs. Action 1.3: Strengthen public expenditure management. „„ Strengthen budget formulation and improve management, execution, and auditing. Action 1.4: Include citizens in oversight of execution and delivery. „„ Increase the transparency of public spending to ensure the benefits of public spend- ing reach the poor. Recommendation 2: Improve program targeting of the poor and vulnerable and move away from regressive spending. More resources need to be allocated to programs that target the poorest and most vulnerable. To this end, either (1) some funding should be shifted from programs benefiting the richest populations to programs targeting the poor, or (2) the overall program budget should be increased. Because Burkina Faso’s revenue sources are unlikely to create further substantial fiscal space, it is more realistic to reallocate expendi- tures from less efficient programs before considering a budget increase. The following actions should be prioritized. Action 2.1: Disseminate the use of a proven targeting system. „„ Use an integrated approach for the assessment of socioeconomic needs and condi- tions when possible to effectively reach the poorest. xxi executive summary „„ Align targeting methods with government objectives. „„ Ensure interventions are aligned with areas that suffer from high poverty rates and low coverage. Action 2.2: Shift spending from universal subsidy programs toward programs tar- geting the poor. „„ Continue with already initiated energy subsidy reform initiatives—that is, reduce spending allocated to broad-based consumer subsidy programs. „„ Protect the poorest consumers by expanding well-targeted social safety nets. „„ Design specific measures to mitigate the potentially adverse impact of reform on the poor while gradually phasing out subsidies. „„ Communicate and engage with diverse stakeholders to avoid misperceptions and information asymmetry. Recommendation 3: Improve the coordination of social programs by building a social registry. As the social safety nets system in Burkina Faso involves many sectors and comprises a large set of programs, further investments in a social registry containing socioeconomic information on the intended population would provide a common gateway for multiple programs. Further, it would improve coordination across sectors, agencies, and programs while reducing program duplication and overlap. The following actions should be prioritized. Action 3.1: Adopt a harmonized questionnaire. „„ Identify opportunities for integration of intake and registration processes. Integration across different programs is an efficient way to reduce costs. „„ Develop a common harmonized intake and registration questionnaire that captures all variables used by the different programs. This could help to rapidly expand the national social registry. Action 3.2: Build a social registry as a common gateway for multiple social pro- grams. „„ Further develop Burkina Faso’s nascent social registry, as it would improve the effec- tiveness of the overall social safety net system. „„ Use the social registry as a tool for coordination on the assessment of needs and conditions to determine potential beneficiary eligibility for multiple programs. Recommendation 4: Use social protection expenditure to build human capital where it matters most, such as for early childhood development and literacy. Early childhood development is not now addressed by Burkina Faso’s social protection system. At the same time, the country’s human capital needs in terms of nutrition, early stim- ulation, and learning are staggering. Literacy—both during the critical school ages as well as xxii THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO in adulthood—ranks far behind regional comparators. Literacy is a basic ingredient for suc- cess in the formal labor market and for effective citizen participation in society. The following actions should be prioritized. Action 4.1: Link social assistance programs to human capital building along the life cycle. „„ Enable parental investment in children’s human capital through cash and in-kind transfers and accompanying measures such as training on appropriate health behav- ior and early education of children (“soft condition”). „„ Through conditional cash transfers, parents can be requested to demonstrate spe- cific behaviors in order to receive the transfer, such as school attendance. These so-called “hard conditions” are often more effective, although both soft and hard condition approaches are valuable. „„ Scale up flexible service infrastructure. For example, all public works/Labor-Intensive Youth Public Works Program (THIMO) activities should be supplemented with mobile child care as in the Youth Employment and Skills Development Project; this easy-to- scale structure is supported by the Ministry of National Education and Literacy and the Ministry of Youth, Training and Vocational Integration. Action 4.2: Promote basic skills training for adults and youth. The essential basic literacy program of Burkina Faso’s National Fund for Literacy and Non- formal Education (FONAENF) requires at least 400 hours of training. In leading up to such substantial investments, smaller steps could be taken: „„ Identify and make use of every occasion to train social protection beneficiaries in basic skills. For example, beneficiaries of public works/THIMO programs could enroll in some literacy training alongside public works. „„ Various modalities of delivering basic skills should be tested and subjected to rigor- ous impact evaluation. Recommendation 5: Improve the capacity of social safety net programs to respond to shocks (before, during, and after) through adaptive social protection. Given the degree to which Burkina Faso is subject to adverse natural events and the lack of or irregularity of rainfall with corresponding food insecurity issues, it is crucial for the country’s safety net system to include flexible elements that can be scaled up quickly. This requires improving the existing early warning system for food insecurity and other issues. The follow- ing actions should be given priority. Action 5.1: Develop and test shock-responsive benefits. Shock-responsive benefits need to be rolled out rapidly in shock-affected areas and meet the needs of vulnerable beneficiaries within a few weeks of crises. xxiii executive summary „„ Develop a solid early warning system to identify the geographical area that is affected. „„ Develop an efficient targeting system that can quickly identify the most vulnerable/ poorest households within the affected area (a standard proxy means test process would take too long). Action 5.2: Strengthen the early warning system. Strengthen the existing Burkinabe early warning system to identify likely food-insecure prov- inces. „„ Shorter-term harvest forecasts are needed so as to be able to predict triggers in suf- ficient time (10 months before the end of the next rainy season). „„ Hydro-meteorological data should be defined for a smaller geographic grid and be combined with population data to predict impact. „„ Seamless cooperation between regional and national authorities is critical in using and diffusing data for maximum benefit. Concluding remarks With a few focused, courageous policy decisions, several issues related to the Burkinabe social protection system can be converted into opportunities. „„ Fiscal space is currently lost through governance issues such as missing coordi- nation, resulting in an overlap of programs and insufficient budget execution. That space can be regained. „„ Fiscal space is similarly lost through spending on regressive programs such as sub- sidies and scholarships rather than efficiently targeting social expenditures to the poor. This fiscal space can also be regained. „„ By regaining the fiscal space as described in the above recommendations, the gov- ernment could cover the country’s poor with an effective, efficient safety net—and more. This end is achievable simply by realigning and better targeting existing safety net expenditures. Reallocating regressive subsidies and scholarships would open fiscal space beyond that. Existing social expenditures can achieve greater impact by linking payments to the utilization of human capital building services such as education and health. Besides offering long-term strategies to alleviate poverty and build human capital, social safety nets can channel a short-term response to shocks against food security. Burkina Faso possesses the requisite building blocks to provide shock-response safety nets. With some improvements, these can be used to help the country face acute climate-induced hardships on a regular basis, providing a transition from humanitarian assistance to national systems. chapter 1 introduction This chapter presents the background, purpose, and conceptual framework for this study; clarifies the basic definitions for social expenditure that are used in this report; and briefly describes the report’s structure. 2 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO Background Burkina Faso—the “land of honest men,” as the Republic of Upper Volta was renamed in 1984—is a landlocked country in West Africa, surrounded by Mali to the north, Niger to the east, Benin to the southeast, Togo and Ghana to the south, and Côte d’Ivoire to the south- west. With a land mass of about 274,200 square kilometers, Burkina Faso’s population was estimated at 19.2 million in 2017. It was colonized by France and gained its independence in 1960; French remains the official language for government and business. The country’s capital is Ouagadougou, which is the administrative, communications, cultural, and economic center of the nation. It is also Burkina Faso’s largest city, with a population of about 1.5 million as of 2006. Burkina Faso has experienced sustained economic growth over the past decade, primarily due to its main export commodities of cotton and gold, but this growth has not translated into comparable per capita gains. After a long period of eco- nomic stagnation, the nation recorded an annual gross domestic product (GDP) growth rate As a nation, Burkina Faso exceeding 6 percent on average during the early 2000s and up to 2014. Growth rebounded in 2016–17 after a slowdown caused by a decline in global commodity prices and the political has seen great progress crisis of 2014–15. Burkina Faso remains dependent on a narrow base of natural resources for in recent years in terms economic growth. Despite its relatively high growth rate, the economy has not created suffi- cient jobs for the rapidly growing workforce, 80 percent of which is employed in agriculture of structural reform, and earns an annual income of less than $350; non-agricultural informal jobs consist of pre- sound economic policies, carious or low-earning occupations (World Bank 2017). steady investment, and The GDP growth rate is offset by a high demographic growth rate. With 5.4 chil- the establishment of a dren per woman, Burkina Faso has one of the highest fertility rates in the world, according to stable macroeconomic the World Bank’s World Development Indicators database. As a result, Burkina Faso’s average annual per capita gross national income increased by only 2.6 percent between 2006 and environment. 2013—this is lower than the global and African average rates for the same period. At this rate, it would take approximately 25 years for the country to double its average per capita income, which is clearly insufficient to achieve rapid reductions in poverty. By comparison, at current rates, it will only take 7 years for Ethiopia to double its per capita income and 12 years for Rwanda (World Bank 2017). Moreover, Burkina Faso’s rapid population growth puts pressure on delivery of basic services. Growing at an annual rate of 3.1 percent, the population is pro- jected to increase from an estimated 19.2 million in 2017 to 29.0 million in 2030. The country’s prosperity has not benefited the majority of Burkinabes. Food insecurity remains high, and extreme poverty is rampant and largely a rural phenomenon. According to the World Bank’s Human Capital Index—which measures the amount of human capital a child can expect to attain by age 18, given the risks of poor health and poor edu- cation that prevail in its country—the nation is one of the least developed in the world, with a rank of 144 out of 157 countries (World Bank 2018b). Sixty-one percent of its youth are illiterate, as are 90 percent of its elderly. Overall, almost half the population lives in poverty; in some regions, poverty rates can exceed 70 percent, particularly in rural areas and the regions at the border with Mali. Approximately 90 percent of the poor live in rural areas. Even though 3 chapter 1. introduction a majority of the population depends on agriculture for its livelihood, agricultural productivity is low and below the country’s potential. Indicators of human development remain low, and a large part of the population lacks access to basic services.1 For example, the under-five mortality rate was 81.6 per 1,000 live births, compared to an average of 76.5 in Sub-Saharan Africa. The net school enrollment of children age 6–11 was 66 percent, compared to 79 per- cent in Sub-Saharan Africa. Acute malnutrition among children was 7.6 percent, and the incidence of stunting was 27.3 percent in 2016. Access to basic infrastructure is also limited. The nation is increasingly facing adverse natural conditions and is highly vul- nerable to exogenous shocks. Burkina Faso regularly suffers from natural hazards, and the frequency of shocks affects the stock of both human and physical assets. Two-thirds of households report that they suffer from shocks each year, mostly related to natural hazards (UNDP 2016). Besides natural hazards such as droughts, floods, locusts, wildfires, and wind, households face risks due to food insecurity and regional instability; the latter exacerbates the poor’s vulnerability to shortfalls in consumption and results in seasonal hikes in the incidence of poverty. Regional insecurities pose an additional threat to the country’s development. The human development challenges facing Burkina Faso call for social safety nets that promote poverty reduction, reduce vulnerability, and accelerate human capital accumulation. International evidence points to multiple benefits from investing in The likelihood of lifting poor effective and efficient social protection (box 1.1). Social protection policies and programs— Burkinabes out of poverty is especially noncontributory social safety nets, when correctly developed and appropriately funded—can directly address extreme poverty and help households manage shocks and limited, and a large portion invest in their children and their assets. International experience also suggests that such of the population is highly investments can be affordable, where they are well designed and targeted. There is thus a critical need to strengthen social safety nets and build a coordinated system to manage the vulnerable to falling into the country’s human development challenges. poverty trap. The government of Burkina Faso has adopted a comprehensive National Social Protection Strategy. The strategy’s overall aim is to fight poverty and exclusion, protect against exogenous shocks, and increase the resilience of the most vulnerable by establishing a minimum social protection floor for all. The 10-year strategy, which was adopted at the end of 2012 and is implemented through three-year action plans, has six objectives: „„ Improve social safety nets for the poorest and most vulnerable „„ Increase access to social services „„ Promote employment and access to minimum income opportunities „„ Extend social insurance coverage, including to informal workers „„ Improve governance „„ Reinforce capacities Burkina Faso uses several instruments to achieve these objectives, including public works, food and nutrition programs, cash transfers, fee waivers for health and education (scholar- ships), social care services, and government price subsidies. Burkina Faso ranked 185 out of 188 countries on the 2016 Human Development Index (UNDP 2016). 1  4 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO BOX 1.1  Reasons to invest in social protection Social protection is a powerful way to fight poverty and reduce inequality. Evidence shows that social protection directly reduces chronic poverty and vulnerability by enabling poor households to meet their basic consump- 1 tion needs; protect their assets; and achieve better health, nutrition, and education outcomes. Social protection programs also build households’ productive assets and expand their income-earning opportunities by improving their labor market skills and enabling them to engage in higher-risk, higher-return activities. Moreover, by supporting participation in productive activities and redistributing income to poor households, social protection can significantly contribute to reductions in inequality. Social protection contributes to economic growth. Social protection contributes to local economic devel- 2 opment by improving labor market functioning, stimulating local markets through cash transfers, and developing community infrastructure. By boosting aggregate demand and facilitating difficult economic reforms, social protection also contributes to broad economic growth. Social protection can reduce social tension and promote stability. Countries emerging from social 3 tension and conflict situations have used social protection to foster peace and to rebuild social capital. During periods of crises and difficult economic reforms, social protection can help safeguard social stability. Safety nets are a critical part of a government’s capacity to respond to shocks. The triple crises of 2008 and recurrent natural disasters have demonstrated the pivotal role safety nets play in mitigating the impact of 4 shocks on poor and vulnerable households. Across the world, countries with well-established safety nets were able to scale up these initiatives swiftly to provide assistance to a large number of vulnerable households. Governments can and need to put in place coherent safety net programs and establish the prerequisites for scaling them up seamlessly, such as robust early warn- ing systems and contingency plans, and ensure that these programs are coordinated with a well-functioning emergency system. Countries can benefit significantly by creating an integrated social protection system. Adopting a systems approach to social protection will reduce inefficiencies and ensure more equitable delivery of benefits 5 from safety nets, pensions, insurance, labor programs, and targeted service delivery. A systems approach draws attention to opportunities to harmonize and expand existing programs in a way that reduces fragmentation and duplication, while promoting linkages across programs to capitalize on synergies. Adopting basic administrative tools such as social registries can help pro- grams harness possible economies of scale. Social protection is affordable—and the costs of not having it are high. It would cost a country only 1–2 percent of its GDP to scale up its social protection programs to ensure national coverage of the poor. Although this is only a 6 portion of the financing required to operate a full-fledged social protection system, it highlights what low-income countries can achieve in the short term. Social protection is a precondition for sustainable growth and social inclusion, and the costs of not protecting poor families are very high—and are borne disproportionately by women and children, undermining the productivity of future generations. Social protection spending can be made more efficient by reallocating financing from inefficient subsidies and ad hoc emergency food aid to predictable safety nets. SOURCE: Adapted from World Bank 2012. 5 chapter 1. introduction Additional objectives of the strategy focus on the development and improvement of institu- tional arrangements, delivery platforms—including targeting tools and a single social registry of beneficiaries—and monitoring and evaluation instruments. New institutions have emerged since 2012, with the objective of reinforcing intersectoral cooperation and strengthening partnerships. The social protection system itself, however, is fragmented, consisting of programs with limited coverage, effi- ciency, and coherence. Burkina Faso has more than 100 small social safety net programs, 80 of which account for less than 1 percent of total social safety net expenditures. Although social protection expenditures have increased in the last decade, these expenditures have not always been effective. The government is working to find ways to improve the effectiveness and efficiency of its social protection programs and expenditures. To this end, discussions are ongoing on potential expansion of the nascent unified social registry of beneficiaries and on establishing a dedicated cross-sectoral budget fund for social protection activities. Purpose and structure Government commitment to social protection has been demonstrated by an increase in social protection expenditure and a surge of new programs. In a con- text of decreasing public finance and budget constraints, however, resources allocated need to be cost-effective in reducing poverty and vulnerability. Social safety net programs should be designed so they cover the various vulnerable groups and the risks they face to ensure that no one is left out. Targeting and other mechanisms that minimize the leakage of expen- ditures to those not in need—and hence mitigate the impact of social protection policies on public finance—should be included in program design. To date, no comprehensive and systematic review of all ongoing social safety net activities in Burkina Faso has been undertaken. While the National Council for Social Protection (CNPS) monitors and has reported on selected programs (e.g., CNPS 2015b, 2016b, 2017b), no comprehensive assessment has been launched. This report responds to the Burkinabe Ministry of Economy and Finance’s request to take stock of the country’s social safety net programs, evaluate their suitability and fitness, and identify options to streamline and reform them for better effectiveness—all of which is directly in line with the government’s commitment to improve overall social safety net provision in Burkina Faso. The purpose of this report is twofold: „„ Identify social safety net gaps by determining whether the availability of social safety nets is commensurate with the needs „„ Assess overall social safety net performance and help consolidate expenditure The report is divided into five chapters: „„ Chapter 1 presents the background, context, purpose, and overall approach of the report. 6 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO „„ Chapter 2 reviews the poverty profile of Burkina Faso in detail and looks at how the situation has evolved over the last decade, including a profile of chronic versus tran- sitory poverty. It highlights household vulnerability to shocks and risks, as well as the determinants/correlates of poverty and vulnerability in Burkina Faso. The objective of the chapter is to understand poverty—including risks and vulnerabilities—across the life cycle. It draws on the World Bank’s poverty and vulnerability analysis and sys- tematic country diagnostic (World Bank 2016a, 2017), complemented by calculations using data from the 2014 Continuous Multisectoral Survey. The chapter concludes with a summary of the needs for social protection over the life cycle. „„ Chapter 3 investigates Burkina Faso’s institutional framework and reviews the his- tory of its social protection strategy, based on national documents and legislation and the recent World Bank Africa-wide social safety net report (World Bank 2012). Administrative data were collected to evaluate how much the Burkinabe government, and its partners, invest in social protection. Expenditures are presented over time, by social protection category, with a focus on social safety nets. The chapter looks at the supply of social protection from different angles: institutional—how are social safety net activities organized?, financial—how much does the country spend?, and functional—what are the main types of programs and how do they work? „„ Chapter 4 provides an assessment of the supply of social protection with a focus on social safety nets by (1) looking at the population covered by the different pro- grams, (2) assessing the targeting effectiveness and cost-effectiveness of the system, (3) analyzing and discussing its sustainability, and (4) determining current practices in monitoring and evaluation. The chapter highlights coverage gaps with respect to needs along the life cycle, and discusses targeting methods and the geo- graphical adequacy of social safety nets; it draws on administrative data as well as on a series of indicators computed using survey data. „„ Chapter 5 provides conclusions and recommendations, including suggestions both on how to strengthen the social safety net system and how to consolidate programs. The appendixes contain supporting analysis and examples of good practices. Most notably, appendix A provides a detailed review of the main social safety net interventions in Burkina Faso, including their level of expenditure, beneficiaries, targeting, and institutional set-up. The aim is to provide an overview of how these programs operate and overlap. Various sources were used to compile this appendix, including national documents, donor reports, interviews, and online information. Conceptual framework The technical analysis undertaken for this report drew inspiration from the following concep- tual framework. There is empirical evidence for poverty traps. Bandiera et al. (2017) note, in connec- tion with a randomized control trial of asset transfers in Bangladesh, that 7 chapter 1. introduction there is a threshold level of capital such that individuals with initial capital below the thresh- old remain trapped in poverty whereas those above move beyond the poverty threshold and escape poverty. For individuals close to the poverty threshold the asset transfer is sufficient for them to successfully take on a new and more remunerative occupational activity (livestock “…the word ‘trap’… rearing) whereas for those further away they fail to do this and fall back into poverty where they suggests that there is a way remain reliant on itinerant wage labor. (Bandiera et al. 2017) out. Indeed, there is—but Transferring an asset—a chunk of capital—to poor people can lift them over the like many traps, escape poverty trap. In fact, reducing the cost of capital in any way—whether through lower interest rates, lower rental costs, lower depreciation, or lower insurance costs—could reduce or even from poverty often requires eliminate the poverty trap. some help from the outside.” Subsidizing capital building, by way of regular cash transfers, can also lift peo- (Smith 2006) ple over the poverty trap. Human capital, as in health and skills, needs income to be built and maintained. The following scenario and accompanying figures illustrate. Illustrating the poverty trap: Mariam’s story Mariam is a poor woman in the rural Central East region of Burkina Faso. She spends most of her day tending her small flock of goats and sheep. When she is not in the field, she prepares meals for her family and looks after her children. Pastoralism and agriculture, combined with climate-induced food insecurity, make for a risky business. Variation in income is costly to Mariam, who struggles to make ends meet for her children. There is no credit market to bridge harvest uncertainties with loans. Her family does not usually eat three meals a day. While they receive enough calories to survive nutritionally, their food intake is not sufficient for them to build up long-term strength. Mariam would normally not eat enough to be very productive, but with sufficient food, she would be. Figure 1.1 illustrates how income today (proportional to food today) translates into income tomorrow via personal productivity. Up and until a basic threshold of food today, productivity does not rise. After that threshold, it rises substantially. The 45-degree line in figure 1.1 shows how the house- hold progresses over time. Any income/food level today FIGURE 1.1  Income dynamics and the poverty trap between A and B would mean a lower income tomorrow than today. The household would always gradually fall Income tomorrow back to A. This is called a poverty trap. Between A and Translation of, B, the household always falls back. Once the house- e.g., nutrition hold exceeds income level B today, it will have a higher into productivity income tomorrow than today. Productivity pays off, income rises. Only above B will the household escape the poverty trap. A B Income today Income today depends on and builds capital: human, physical, financial, and other capital. With good skills and strength, a person can earn an adequate income SOURCE: Based on Banerjee and Duflo 2011. 8 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO FIGURE 1.2  Household income, capital, investment, today, eat appropriately, and build even more strength costs, and the poverty trap for tomorrow. Financial capital can be invested in machines or other work improvements and contribute to a higher harvest tomorrow. Investing capital today gen- erates income tomorrow. Figure 1.2 illustrates these relationships by making fig- ure 1.1 a bit more complex. The horizontal axis shows Safety net the amount of capital k per head; that is, total capital K as a share of labor L. The vertical axis shows income per head y (total income Y per labor L). The uppermost curve illustrates how k today translates into y tomorrow, with y as a function f(k). This is not a linear or monoto- nous relationship, but somewhat S-shaped. A share s of y, sf(k) can be saved and reinvested as i. This is the second curve in the figure. As it is a straight share of f(k), the overall S-shape is the same. SOURCE: Based on Solow-Swan growth model as shown in https://cruel.org/ econthought/essays/growth/neoclass/solowtrap.html. The straight line illustrates the cost of capital, which is linear. It would comprise interest on loans, depreciation on physical assets, as well as any risks, expressed through the equivalent cost of insurance. In reality, Mariam does not have insurance but bears the risks directly herself. When the saved/invested amount of income sf(k) exceeds the cost of capital, the overall capi- tal stock grows. This is the case for k below k1* and y below y1*: capital will grow until reaching k1*. It is also the case for k above k2* and y above y2, where capital will grow until it reaches k3*. It is not the case for any capital stock between k1* and k2*, and an income between the corresponding y1* and y2*. Households with a capital stock of this amount are in a poverty trap. They will always gradually fall back to k1*. A safety net is extreme poverty insurance: if you fall Escaping the poverty trap One way for Mariam to escape the poverty trap is to insure her income insecurity; that is, below a certain poverty line, to insure her against poverty and give her and her family enough security to escape the you are eligible. This is the poverty trap. If Mariam’s family’s per capita consumption falls below the poverty line, she could receive a regular cash transfer. If of a sufficient size, the regular cash transfer—which concept of a social safety she would receive irrespective of weather, harvest, or other risks—can give Mariam’s family net, the net you should enough security to escape the poverty trap. fall into when you fall into A safety net works like that. It is extreme poverty insurance. If a person falls below a poverty. certain poverty line, he or she is eligible to receive a regular cash transfer every few months. This is the concept of a social safety net, the net people should fall into when they fall into poverty. The Burkinabe program Burkin-Naong-Sa Ya, implemented by the Ministry of Women, National Solidarity and the Family, works in this way. It pays CFAF 30,000 every three months for two years, to poor women with children under the age of 15. In May 2018, Mariam 9 chapter 1. introduction received her first cash envelope from Burkin-Naong-Sa Ya. She and her family have been identified as poor and are eligible for the program. She plans to grow her flock of small rumi- nants to improve the livelihood of her household. Cash for work programs have a similar objective. They provide poor families with a basic benefit to escape poverty and to ensure regular basic consumption of all household members. Generally, cash transfers to vulnerable beneficiaries that are financed out of the government budget and do not depend on a contribution by the beneficiary are called social safety nets, or social assistance. Social assistance is often accompanied by programs to improve beneficiary pro- ductivity. Beneficiaries or potential beneficiaries are provided with skills training or helped Government has a strong to find a new or better job—or any employment at all. The idea behind these programs is social motivation to finance to improve the welfare of beneficiaries by increasing their human capital and their access to information about opportunities. In figure 1.2, this would be illustrated by the two upper social assistance, social curves lifting up and tilting a bit toward the left. These programs can also improve prospects insurance, and productivity- to escape the poverty trap. improving measures. There are also social programs that prevent families from falling into a poverty trap in the first place. These are social insurance programs. Health insurance protects households against catastrophic health expenditures that could push them into poverty. Pen- sions prevent the decline of productivity that all humans may face in old age by doing the same. Both of these programs normally depend on a prior contribution by the beneficiary. There is an economic rationale for governments to engage in financing social programs, especially safety nets. The private credit market fails for households in poverty traps: there are no products available and payable for poor families to smooth the uncertainties of their income through loans or insurance. So market failure is one motivation. Another important motivation is to work toward the equality of all citizens, as is often man- dated by a country’s constitution. Government, representing its voters, who care about equity and equality with their fellow citizens, has a strong social motivation to engage in financing social assistance, social insurance, and productivity-improving measures. chapter 2 social protection needs: analysis of risks and vulnerabilities This chapter identifies the major sources of vulnerability for households in Burkina Faso; the most vulnerable groups, which are the potential targets of social protection programs; and the main risks these groups face throughout the life cycle—which, if not addressed, can perpetuate intergenerational transmission of poverty. 12 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO Poverty and vulnerability trends Over the last decade, Burkina Faso has enjoyed sustained economic growth.1 However, due to demographic pressures, this has not translated into similar per capita gains. Burkina Faso has recorded strong macroeconomic indicators over the last decades, with an annual gross domestic product (GDP) growth of 6 percent between 2005 and 2013 (World Bank 2017). However, rapid population growth—led by one of the highest fertility rates in the world at 5.4 children per woman—has placed the country under pressure. Economic growth does not directly benefit the population, as demographic trends further increase the burden on limited social services and on the narrow formal labor market. Even though GDP growth has been consistently above the regional average, the growth per capita between 2006 to 2013 has been much lower. In fact, Burkina Faso has one of the lowest gross national incomes in Sub-Saharan Africa (figure 2.1) and across the world. FIGURE 2.1  Gross national income per capita in Sub-Saharan Africa, 2015 Current dollars 7,000 6,000 5,000 4,000 3,000 2,000 1,000 0 Burundi Central African Republic Malawi Liberia Niger Congo, Dem. Rep. Madagascar Togo Ethiopia Mozambique Sierra Leone Burkina Faso Rwanda Uganda Mali South Sudan Benin Zimbabwe Chad Tanzania LDCs Senegal Cameroon Kenya Côte d’Ivoire Ghana Zambia Sub-Saharan Africa Sudan Bhutan Congo, Rep. Nigeria Cabo Verde Swaziland Angola Low & middle income Namibia South Africa Botswana SOURCE: World Bank, World Development Indicators database, http://datatopics.worldbank.org/world-development-indicators/. NOTE: LDCs = least developed countries; LDC average is based on United Nations classification. The poverty headcount ratio has been declining over time, while the absolute number of poor people has increased. The most recent poverty data show that 40.1 percent of Burkinabes are This chapter draws on findings from the World Bank’s poverty and vulnerability analysis and systematic 1  country diagnostic (World Bank 2016a, 2017), complemented by author calculations using EMC 2014. 13 chapter 2. social protection needs: analysis of risks and vulnerabilities poor,2 which represents a decline of 13 percentage points FIGURE 2.2  Poverty headcount ratio and number of over 11 years. By contrast, the number of individuals living poor over time on less than CFAF 421 a day in 2014 (the national poverty line) actually increased, from 7 million in 2009 to 7.2 million Percent Millions of poor in 2014 (figure 2.2). The decline of poverty has not been 60 7.6 able to keep up with the rapid population growth—led by 7.5 50 high fertility rates coupled with decreasing mortality and an 7.4 40 7.3 increase in life expectancy. 7.2 30 Most of the population lives close to the poverty 7.1 line, making the headcount ratio sensitive to a 20 7.0 6.9 small increase in the poverty line (figure 2.3). There 10 6.8 is a high concentration of individuals around the national 0 6.7 poverty line, making them more vulnerable in case of 2003 2009 2014 adverse events. If the poverty line is increased to $3.10 a day (CFAF 663), as per the international standard, Burkina SOURCE: World Bank 2016a, figure 2.1. Faso’s poverty rate increases to accounting for almost 75 percent of the population. Additionally, if the actual poverty line increased by only 10 percent (from CFAF 421 to CFAF 463), 1.4 additional million Burkinabes would be classified as poor—the latter being therefore considered at the margin FIGURE 2.3  Poverty headcount using different poverty line per capita/consumption distribution Poverty headcount (%) 100 National Internat’l National Internat’l poverty poverty line poverty poverty line 80 line $1.90/day line + 10% $3.10/day Poverty 40.1 43.7 47.9 74.6 60 headcount Poverty gap 9.7 11.0 12.4 22.2 40 Poverty gap 3.3 3.9 4.6 1.9 20 Poverty line 155,530 160,471 168,883 261,821 696.3 (annual LCU) 426.8 0 No. of poor 7,071,806 7,712,990 8,455,057 13,174,412 400 800 1,000 1,400 1,800 2,000 2,400 Per capita daily consumption (in CFAF) SOURCES: Authors’ calculations based on EMC 2014 and World Bank 2017. NOTE: LCU = local currency unit. Poverty headcount ratio is computed using a consumption aggregate and national poverty line based 2  on the Living Standards Measurement Study 2014. Burkina Faso’s national poverty line in 2014 was CFAF 153,530 per capita per year or CFAF 421 per capita per day. This poverty line is very close to the international poverty line, which is $1.90 per day in 2011 purchasing power parity (PPP) dollars. Indeed, using the 2011 PPP and inflation between 2011 and 2014, the international poverty line corresponds to CFAF 426.8 per capita per day in Burkina Faso; and the $3.10 per day per capita poverty line corresponds to CFAF 626.3 in Burkina Faso in 2014 (World Bank 2017). 14 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO of poverty. Figure 2.3 also depicts the consumption distri- bution, which is mainly concentrated around the national FIGURE 2.4  Inclusive growth: ratio of richest/poorest poverty line. quintiles’ per capita consumption over time While declining poverty trends should be taken Percent cautiously given the rapid population expansion, 12 2003 2009 2014 economic growth in Burkina Faso has still been 10 inclusive. Over the period 2003–14, the growth rate in 8 annual per capita consumption has been more rapid for 6 the bottom 40 percent of households: their per capita con- sumption growth has been twice as high as for those in the 4 top 60 percent. This inclusive growth is also reflected by a 2 decline in the ratio of consumption of the wealthiest 20 per- 0 cent of the population versus the poorest 20 percent, which Urban Rural All declined from 7.8 to 5.3 over 2003–15 (World Bank 2017). Interestingly, the decline has been much more pronounced SOURCE: World Bank 2017. in urban areas than in rural areas (figure 2.4) given the rapid urbanization process. Inequality still remains of a concern, even though it has been declining. The standard measurement of inequality, the Gini index, decreased by a significant 7 percentage points during the period from 2003 to 2014 (figure 2.5). FIGURE 2.5  Gini inequality index over time With a Gini coefficient of 35 in 2014, Burkina Faso is per- forming relatively well compared to other countries in the Percent Sub-Saharan African region. Its Gini coefficient is below 50 2003 2009 2014 countries such as Cameroon (46 in 2014), Benin (43 in 40 2011), and Senegal (40 in 2012), but above Mauritania and 30 Niger (32 and 34, respectively, in 2014).3 The Gini coeffi- cient is also below the Sub-Saharan average of 45.1 and 20 below the global average for low-income countries of 40 10 (World Bank 2017). Inequality is more pronounced in urban 0 areas (Gini of 38.4) than rural (27.3), which may be due to a Urban Rural All general low level of consumption in rural areas. Inequality can, however, be underestimated, as it is computed on SOURCE: World Bank 2017. aggregate consumption, which may underestimate the liv- ing conditions of the richest.4 World Bank, World Development Indicators database (http://datatopics.worldbank.org/world- 3  development-indicators/), accessed August 2017. In Burkina Faso, as in most low-income countries, inequality is measured on the basis of consumption 4  surveys. Therefore, measures of inequality reflect the difference between the values of consumption for different income groups. However, consumption-based indexes are an imperfect proxy of income/ wealth differences across households for two reasons. First, savings behaviors are not captured by this measurement; rich households have higher savings capacity and greater existing assets than do poor households. Second, consumption surveys suffer from methodological weaknesses. For example, important spending categories are not included, such as spending in foreign countries and/or spending 15 chapter 2. social protection needs: analysis of risks and vulnerabilities The context in Burkina Faso remains one of high vulnerability to various kinds of shocks. The likelihood of being exposed to shock at some point during the year is high in Burkina Faso. Only 31 percent of the population faced no shock during 2014. Looking at the number of shocks faced TABLE 2.1  Number of shocks by poverty status (% of by those living in poverty (according to the national pov- population) erty line), about 78 percent faced at least one shock over the year, while 47 percent faced more than two shocks. Number of shocks Total population Poor Nonpoor The likelihood of facing no shock is much higher for the 0 31 22 37 nonpoor, with 37 percent facing no shock at all (table 2.1). 1 30 31 30 Still, 63 percent of the nonpoor will be confronted with at 2+ 39 47 33 least one shock during the year, making them more likely to % of population 100.0 40.1 59.9 transition into poverty. Among the 7 million poor people in SOURCE: Authors’ calculations based on EMC 2014. Burkina Faso, more than 5.5 million are confronted with at least one shock per year. The most common types of shocks in Burkina Faso are covariant shocks, affect- ing a community as a whole and generally leading to adverse coping strategies, in particular for the poor. Idiosyncratic (i.e., household-level shocks) are less frequent, with health and employment issues affecting 2.5 and 4.0 percent of the population respec- tively (figure 2.6). Looking at the population affected by all types of shocks, 19 percent are affected by crime issues such as theft and violence. The two predominant types of shocks are covariant, with 63 percent of the population being affected by natural hazards (including droughts, fire, and flood) and 29 percent suffering from price fluctuations (including drop in agricultural price, increase in agricultural inputs, increase in food prices). The poor are much more affected by these two types of shocks with a difference of 23 percentage points for nat- ural hazards and 14 percentage points for price shocks. Only 2.4 percent of households affected by a shock receive support from either the government or a nongovernmental organization (NGO). Instead, families are forced to sell their assets or reduce their food consumption—negative coping strategies that make it more difficult to escape the poverty trap. Most households are forced to adopt such negative coping strategies (only 34 percent of those affected by any type of shock did nothing): the most common strategy is the use of savings (76 percent) and selling household assets (49 percent). The poor are more likely to adopt negative coping mechanisms compared to the nonpoor, in particular when it comes to the sale of assets. Reduction in food consump- tion is also widespread (21 percent of the poor and 18 percent of nonpoor). Such negative coping strategies may have long-lasting impacts on both asset accumulation and human capital building, which is likely to lead to chronic poverty. The very small percentage of the on luxury goods. Also, extreme groups are not statistically well represented in a nationwide survey. The extremely poor do not have an official identify or permanent home and are therefore not included in the sample. At the other end, the number of extremely rich is not sufficiently high to be statistically representative. For these two reasons, income or asset inequality can be seriously underestimated by consumption surveys. Divergent trends also are possible, since consumption differences may decline over time, but income/assets may increase exponentially for the richest over the same period (World Bank 2017). 16 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO FIGURE 2.6  Exposure to shocks and coping strategies (in percent of total, poor, and nonpoor population) a. Exposure to shocks b. Coping strategy Natural hazard Use savings Sell assets Price Do nothing Support from family/friends Crime and safety Reduce food consumption Other Employment or migration Poor Nonpoor Other Employment related All New loan Health related Support from government/NGOs Reduce education/health Household issues Reduce nonfood consumption 0 20 40 60 80 0 20 40 60 80 % of population % of population SOURCE: Authors’ calculations based on EMC 2014. NOTE: Categories are not mutually exclusive. population receiving support from the government or an NGO reflects the lack of appropriate social protection BOX 2.1  Need for adaptive social protection instruments (box 2.1). T he concept of adaptive social protection has emerged in If taking shocks into account to estimate pov- recent years; it emphasizes better enabling social protection to erty, the poverty headcount would increase address the impacts of all kinds of shocks on households—includ- substantially. The likelihood of transitioning into pov- ing natural disasters and climate change, economic and financial erty at some point during the year is not represented by crises, conflict and displacement, among others. A nascent area, the poverty headcount, which potentially underestimates adaptive social protection has begun to crystallize around two inter- the rate of transient poor. A reasonable hypothesis con- related approaches: (1) building the resilience of those households siders that those at the margin of poverty (considered as most vulnerable to shocks, and (2) increasing the responsiveness of poor, with the national poverty line increased by 10 per- social protection programs to adapt to and meet changed needs on cent) would transition into poverty if they faced at least the ground after shocks have materialized. Preparedness measures one shock over the year. By this definition, 1.05 million for safety nets can be advanced even further through additional nonpoor Burkinabes (given the national poverty line) are investments to make programs more flexible and capable of at high risk for transitioning into poverty during the year, expanding to reach additional households. given that they are at the margin of poverty and face at least one shock over the period. Taking shocks into con- SOURCE: World Bank 2018d. sideration, the poverty headcount would increase from 40.1 to 46.0 percent of Burkina Faso’s population. Poverty and transient poverty are mainly pre- dominant in rural areas and vary widely by region. More than 90 percent of poor Burkinabes live in rural areas, illustrating a wide gap between rural and urban poverty: 17 chapter 2. social protection needs: analysis of risks and vulnerabilities 47.5 versus 13.7 percent, respectively.5 This difference is FIGURE 2.7  Population exposure to shock and poverty mainly due to a rapid decrease of urban poverty over the rates years, which stood at 24.6 percent in 2009. The difference between urban and rural areas is also a consequence of a Percentage of population higher exposure to shocks for those living in more remote 80 Urban Rural All areas. More than 75 percent of those living in rural areas 60 are exposed to at least one shock during the year, com- pared to 48 percent in urban areas. Similarly, the rate of 40 transient poor (at the margin of poverty and facing at least one shock) is twice as significant in rural areas as in urban 20 (figure 2.7). 0 The headcount poverty ratio varies significantly Exposure to ≥ 1 Poverty rate Transient poverty shock per year rate across regions, from less than 10 percent in the Central region to more than 70 percent in the North region. The SOURCE: Authors’ calculations based on EMC 2014. degree of urbanization in each region is directly related to its level of poverty, with the region with the highest degree of urbanization (the Central region, which includes Ouagadougou) recording the lowest pov- erty rate, just under 10 percent. In predominantly rural regions, such as Boucle du Mouhoun and the East and Central West regions, the headcount ratio reaches levels higher than 50 per- cent; it is 70 percent in the North region (figure 2.8). These four regions contain one-third of the total population but more than half of the poor population. Looking at poverty rates by province provides further insight into the incidence of poverty: within each region, some provinces are hardly affected by poverty and drive the results by province. While in the North region, all provinces have a poverty headcount close to 70 per- cent, there are larger disparities in the Boucle du Mouhoun and East regions. In Boucle du Mouhoun, the high poverty rates are driven by the province of Sourou (85.7 percent); other provinces have much lower rates, such as 46.4 percent in Mouhon. In the East region, pov- erty headcount varies from 27.1 percent in Kompienga to 80.2 percent in Komondjari. Some regions face poverty rates equivalent to the national average, such as the Southwest, but contain pockets of poverty in a specific province, such as Ioba (69.5 percent). In Sahel, the province of Yagha pulls down the average headcount ratio, with 16.6 percent, while the poverty rate in the Soum province as a whole is much closer to the national average at 32.7 percent. If the national poverty line used in Burkina Faso were revised to a level closer to the World Bank’s 5  extreme poverty line, the urban poverty rate would jump to 18 percent. If the urban poverty line were equivalent to the rate used in Niger, the rate of urban poverty would increase even more dramatically, to 24 percent. These figures show that measurement of the poverty rate is highly sensitive to basic assumptions, and that these rates should therefore be interpreted with caution, particularly when comparing the rates of different countries using varying basic assumptions. It is worth emphasizing, however, that the poverty gap between rural and urban areas in Burkina Faso would remain roughly the same, at about 35–40 percentage points, regardless of the assumptions on which the poverty line is based. A gap of this level is high, even by Sub-Saharan African standards. The only other countries in the region with comparable gaps are Cameroon, Niger, and, to a lesser extent, the Republic of Congo (World Bank 2017). 18 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO FIGURE 2.8  Poverty headcount ratio by region and province, 2014 a. Region b. Province 4°E 2°E 0° 2°E 4°E 2°E 0° 2°E IBRD 44745 | JANUARY 2019 IBRD 44746 | JANUARY 2019 This map was produced by the Cartography Unit of the World This map was produced by the Cartography Unit of the World MALI MALI Bank Group. The boundaries, colors, denominations and any Bank Group. The boundaries, colors, denominations and any other information shown on this map do not imply, on the part of other information shown on this map do not imply, on the part of the World Bank Group, any judgment on the legal status of any the World Bank Group, any judgment on the legal status of any OUDALAN territory, or any endorsement or acceptance of such boundaries. territory, or any endorsement or acceptance of such boundaries. SAHEL SOUM 14°N NIGER 14°N 14°N LOROUM SENO NIGER 14°N YATENGA BAM YAGHA NORD CENTRAL NORTH NAMENTENGA SOUROU ZONDOMA SANMATENGA GNAGNA KOSSI PASSORE NAYALA OUBRITENGA KOMANDJARI KOULWEOGO BOUCIE DU MOUHOUN PLATEAU CENTRAL CENTRAL MOUHOUN KADIOGO GANZOURGOU EAST BANWA BOULKIEMDE KOURITENGA TAPOA GOURMA 12°N  12°N 12°N  SANGULE BAZEGA TAPOA 12°N CENTRAL WEST KENEDOUGOU BALE CENTRAL EAST ZIRO BOULGOU CENTRAL ZOUNDWEOGO TUY KOULPELOGO SOUTH KOMPIENGA HAUTS-BASSINS HOUET SISSILI NAHOURI LOBA BENIN BOUGOURIBA BENIN TOGO LERABA TOGO SOUTH 0° 2°E 0° 2°E CASCADES KOMOE PONI WEST 10°N  GHANA BURKINA FASO 10°N  GHANA BURKINA FASO NOUMBIEL POVERTY HEADCOUNT RATIO POVERTY HEADCOUNT RATIO 0 20 40 60 80 100 Kilometers 0 20 40 60 80 100 Kilometers CÔTE D'IVOIRE CÔTE D'IVOIRE 0 20 40 60 80 Miles 0 20 40 60 80 Miles 9.68 70.3 7.22 85.72 4°E 2°E 4°E 2°E SOURCE: Authors’ calculations based on EMC 2014. Trends in nonmonetary welfare indicators Burkina Faso lags in terms of basic access to education and health and overall standard of living. As noted in chapter 1, it remains one of the least developed countries in the world, with a Human Development Index ranking of 185 out of 188 countries, and a Human Capital Index ranking of 144 out of 157 countries (UNDP 2016; World Bank 2018b). Although overall living conditions have improved since 2003, this is not the case for the poorest (figure 2.9). Overall, the greatest improvement is related to the rapid expansion of telephone service, with the percentage owning a mobile phone exceeding 80 percent in 2014 compared to less than 5 percent in 2003. The next two greatest areas of improvement pertain to having a dwelling with an improved floor and owning a motorcycle within the household. In 2014, more than 50 percent of households lived in a dwelling that had an improved floor (cement or tile instead of mud or soil), representing an increase of 20 per- centage points compared to 2003); a similar percentage reporting owning a motorcycle. The use of household electricity almost doubled over the period, reaching approximately 20 per- cent in 2014. Ownership of a television set followed a similar pattern (20 percent in 2014). The use of electricity or gas for cooking remains relatively low (5 percent in 2014 versus 1 percent in 2003). Access to safe water was relatively high 2003–14, at about 80 percent. Even though the situation improved overall, households in the poorest quintile still fall behind the national average. Only 32 percent of those in the bottom quintile live in a dwelling with improved floors, about 4 percent use electricity for lighting, 5 percent own a TV, and less than 19 chapter 2. social protection needs: analysis of risks and vulnerabilities FIGURE 2.9  Improvement in living conditions, 2003 and 2014, % of total population and poorest quintile Percent 90 2003 2014 (all) 2014 (poorest quintile) 80 70 60 50 40 30 20 10 0 Owning a mobile Owning a Dwelling with an Using electricity Owning a TV set Using electricity Using a source phone motorcycle improved floor for lighting or gas for of safe water cooking SOURCE: Authors’ calculations based on World Bank 2017. 1 percent use a clean source of energy for cooking. Except for the use of mobile phones, the living conditions of those in the poorest quintile are below average conditions in 2003. In terms of access to water and sanitation, the situation has improved in both rural and urban areas. Access to a source of safe water has been generally good in both rural and urban areas. Even though access to water coverage increased from 51 percent to 65 percent from 2003 to 2010, coverage reached 89 percent in urban areas for the same period—driving the national average to 80 percent in 2014. Access to sanitation improved in both rural and urban areas, rising from 0.8 to 12 percent and from 15 to 34 percent, respectively, between 2003 and 2010, but these levels are far below the Millennium Development Goal target of 75 per- cent coverage (World Bank 2017). Disparities in access to public facilities remain high between urban and rural areas (figure 2.10). In 2014, the gap between urban and rural areas reached more than 50 percentage points when it came to access to secondary schools and health centers. While 80 percent of those in urban centers may access a secondary school in less than 30 minutes, less than 20 percent of rural inhabitants have access to such a facility. Access to roads and public transport also lag behind in rural regions (46 and 33 percent, respectively). Access to primary schools, food markets, and sources of drinking water are slowly improving in rural areas but remain below urban areas’ access. 20 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO FIGURE 2.10  Urban and rural population access to public facilities Percentage of population 100 Urban Rural 80 60 40 20 0 Secondary Roads Sanitation Public Primary school Food market Drinking water school center transport SOURCE: Authors’ calculations based on EMC 2014. NOTE: Access means within a 30-minute walk. Poverty along the life cycle Looking at poverty headcounts by age group and risks along the life cycle offers important insights as to where public social protection interventions should focus. Each age group—early childhood, school-age children and youth, adults, and the elderly—experiences different risks and vulnerabilities, which should guide social protection policies. Poverty is highest among children; of the 7 million poor in Burkina Faso, 4.2 million are chil- dren. Almost 45 percent of preschool children (those below 6 years of age) and 44 percent of school-age children (age 6–17) live in poor households. While still very high, poverty among young adults (age 18–34) is about 33 percent; it is 37 percent for middle-aged adults (age 35–59). The elderly, those above the age of 60, face the highest poverty headcount ratio (41 percent) after children; the elderly represent the smallest share of all poor (about 2.2 per- cent of the population), which is in line with their small share in the total population. Among the 40.1 percent of the population that comprises the country’s poor, most are children: 14.5 percent of the population are poor school-age children and 9.3 are preschool children. The remaining poor are young and middle-aged adults, representing 7.9 percent of the popu- lation and 6.1 percent, respectively (figure 2.11). The poverty gap follows similar trends to the poverty headcount. Following is a discussion of the poverty profile in Burkina Faso across the life cycle, focusing on the major risks faced by the various age groups; these are summarized in figure .17 at the end of this chapter. 21 chapter 2. social protection needs: analysis of risks and vulnerabilities FIGURE 2.11  Poverty headcount, gap and composition of headcount by age category Percent Percent 50 50 Headcount Poverty gap 40 40 2.2 60+ 30 6.1 35–59 30 7.9 18–34 20 20 10 14.5 6–17 10 0 9.3 0–5 0–5 6–17 18–34 35–59 60+ Total 0 Age group SOURCE: Authors’ calculations based on EMC 2014. The early childhood years (age 0–5) Health outcomes for young children have improved in Burkina Faso largely because of better access to health care services. Health services have become more accessible on average due to a regular increase in public health expenditures as a share of the overall budget. The nation has seen a notable decline in both malnutrition and mortality indicators over time; however, they still remain high and far from the Sustainable Development Goal (SDG) targets. The main risks young children face are poor maternal and infant care (but see box 2.2 for an innovative program in this regard), low infant birth weight, debilitating and life-threatening diseases, and inadequate diet and a lack of early childhood stimulation—both of which will impair their development and may perpetuate poverty. The health of a child begins before birth and depends on the nutritional status and health of the mother. The nine months of pregnancy permanently influence the func- tion of critical organs and the wiring of the brain, shaping the rest of a child’s life. The World Health Organization (WHO) has established that all pregnant women should make at least four prenatal care visits to promote the birth of healthy babies. In Burkina Faso, only 47 percent of pregnant women received the recommended four prenatal care visits from skilled personnel. Moreover, by making reproductive health services available, health authorities can reduce the number of undesired pregnancies and help ensure that pregnancies are sufficiently spaced. Despite improvements, maternal mortality rates remain among the highest in the world, with an estimated 80 percent of deaths being preventable. WHO reports that 50 percent of maternal deaths occur due to obstetrical complications within the first 24 hours—causes that range from severe bleeding to obstructed labor, all of which have highly effective interventions. According to the most recent demographic and health surveys for 2003 and 2010, maternal mortality has decreased from a level of 440 for 100,000 births in 1998 to 341 for 100,000 in 2010. Infant mortality follows the same trend; it has declined from 22 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO 91 per 1,000 to 65 per 1,000. Nevertheless, these data BOX 2.2  Mobile child care in Burkina Faso: a show the consequences of poor health and inadequate novel approach to public works care during pregnancy, birth, and the first days after birth, and the urgent need to increase the coverage of institutional deliveries and essential newborn care. C hildren put to sleep on the ground, exposed to sun, wind, and rain near dangerous construction sites and with little or no adult supervision while their mothers work. Young children miss- A mother’s death results in vulnerable families; and their infants, if they survive childbirth, are more likely to die ing school while watching even younger siblings. Women on the before reaching their second birthday. Poverty is a key brink of complete exhaustion, caring for their children and carrying contributing factor in preventable maternal death, partic- out domestic chores during the day while taking on wage labor ularly for impoverished women living in rural areas who throughout the night. Women without child care, unable to engage face geographical obstacles to accessing health care. in wage labor, falling ever deeper into poverty. Moreover, women living in poverty are at additional risk for pregnancy complications due to the arduous task of Despite a growing recognition that inadequate child care is a main transporting heavy loads of firewood or other fuels. Just factor hindering women’s participation in wage labor, social pro- as poverty is a factor contributing to maternal death, tection programs rarely offer viable solutions. Public works provide maternal death in turn perpetuates the intergenerational an important lifeline for many, but pose a hazardous environment transmission of poverty. for children. In Burkina Faso, however, an innovative approach that builds on cross-sectoral collaboration and (re)using existing The nutritional status of children in Burkina knowledge is attracting increasing attention with its potentially Faso has improved, but child malnutrition transformative impact on public works. remains a major challenge. The wasting rate (under- weight with respect to height) of children under age five The Youth Employment and Skills Development Project offers decreased from 15.7 in 1999 to 7.6 in 2016, while severe temporary work opportunities to 46,000 youth. Through its public wasting indicators have sharply declined since 2006 works component, participants—most of whom are women—are reaching 1.4 in 2016 (see note to figure 2.12). Under- recruited for six months at a time. When the project team realized weight (weight by age) and stunting (height by age) also that many of the participants were bringing their children to the declined over the period 1999–2016, but still remain work sites, they came up with a game-changing solution: mobile high. In general, even though malnutrition and mortality child care that followed the women from work site to work site. indicators have declined over time, they still remain high, and far from the SDG targets. The initiative has freed up women’s time so they can work and thereby invested in children’s human capital while creating a whole Burkina Faso has one of the highest rates of „„  new stream of jobs for women. The pilot has shown that it is possi- underweight in the West Africa region. In 2016, ble, at an extremely low cost, to set up a sustainable and replicable 19.2 percent of children were underweight, com- system by smashing silos, learning from past projects, and working pared to 11 percent in Ghana (2014), 12.8 percent with existing services. in Senegal (2014), 15 percent in Liberia (2013), and 18 percent in Benin (2014). „„ The stunting rate remains well above the worldwide average. Worldwide average stunting was 22.9 percent; Burkina Faso’s stunting rate was 27.3 percent in 2016. According to UNICEF data, this was higher than other West African countries, including Ghana (18.8 percent), Senegal (20.5 percent in 2015), and the Gambia (25 percent). „„ The under-five mortality rate—at 88.6 per 1,000 in 2016—is still very far from the SDG target of 25 per 1,000. Moreover, again according to UNICEF data, Burkina 23 chapter 2. social protection needs: analysis of risks and vulnerabilities Faso does not perform well compared to other FIGURE 2.12  Child malnutrition estimates West African countries such as Senegal (47.2 per 1,000), Ghana (61.6), the Gambia (68.9), and Libe- Percent ria (69.9). 50 Stunting 40 Low birthweight due to inadequate food intake by the 30 Underweight mother may lead to premature death or poor development Wasting in the early years of a child’s life. A lack of breastfeeding, 20 low food intake in infants, and inadequate feeding practices 10 can lead to stunted development, illness, and early death. Severe wasting 0 1999 2003 2006 2010 2011 2012 2016 Preschool children are not only largely at risk of experiencing poverty, but also face a multitude of SOURCE: UNICEF 2017 malnutrition database, www.data.unicef.org/nutrition/ risks and barriers to building their human capital. malnutrition. Malnutrition and mortality remain high, as noted, as does NOTE: Malnutrition estimates are based on WHO growth standards for children early childhood malnutrition, which irreversibly affects brain 0–59 months. and physical development, reducing school performance. Early childhood development policies are critically lack- ing: the lack of stimulation and low levels of parental education can have lasting impacts on children’s cognitive development. Without receiving adequate stimulation in early childhood, children may enter school ill prepared and be more likely to have poor academic performance, to repeat grades, and to drop out of school compared to children whose cognitive skills and overall school readiness are higher upon primary school entry. These findings call for a pro- poor early childhood development approach where social protection instruments can help link families and parents to adequate services (e.g., food security, health, education or stimulation, prevention of household-based violence). Access to health care suffers from supply-side constraints, all of which greatly affect preschool children. Access to health centers is still limited in rural areas, where only 32 percent of the population are within a 30-minute walk to such facilities; the compa- rable figure among the urban population is 78 percent (figure 2.10). Burkina Faso also suffers from a lack of human capital within the health sector, with a ratio of 5 doctors and 41 nurses per 100,000 citizens. These ratios are twice as low as those in Côte d’Ivoire and three times lower than those in Zambia. According to the most recent survey data, 54.7 percent of sick poor did not visit any health center; this percentage drops to 33.5 percent for those in the richest quintile, indicating that the cost of transportation and distance may be an additional barrier for the poor (World Bank 2017). 24 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO School-age children and youth (age 6–17) Besides the high prevalence of poverty in this group, school-age children face the substantial risk of not attending school. Almost half of primary school–age children in Burkina Faso are not enrolled in school; 57 percent of those of lower secondary education are not enrolled. The primary out-of-school rate is particularly high compared to the regional average (25 percent for Sub-Saharan Africa and 30 percent for West and Central Africa), as well as to such com- parator countries as Benin (25 percent), Ghana (30 percent), and Senegal (36 percent). Most at risk for nonattendance are girls, children living in rural areas, and poor children. A gender gap in attendance is already notable at the primary school level, with difference of 4 percentage points between attendance by boys and girls. The difference in out-of-school rates is striking for rural and urban areas, with more than three times as many children in rural areas not in school compared to their urban peers: 55 and 17 percent, respectively, and over twice as many not attending lower secondary school (65 and 31 per- cent, respectively). At all quintiles of income distribution, nonattendance rates are higher for lower secondary school than for primary school, but the gap is noteworthy for children in the richest quintile as very few of them do not attend primary school (15 percent). The poorest quintiles are definitely more at risk of being out of school: 69 percent of the poorest children do not attend primary school, and the nonattendance rate approaches 77 percent among those of lower secondary school age (figure 2.13). There is a significantly strong and positive correlation between education and wage performance in Burkina Faso. Completing primary education significantly increases the probability of being a wage worker. Failure to complete primary education increases the probability of employment in agriculture by 5 percentage points; conversely, completion of primary school decreases the chances of staying in agriculture by 23 percent- age points (World Bank 2017). FIGURE 2.13  Out-of-school rates, primary and lower secondary education (2010) Percent 80 Primary Lower secondary 60 40 20 0 Male Female Urban Rural 1 2 3 4 5 Quintile (poorest to richest) SOURCE: UNICEF, data.unicef.org. 25 chapter 2. social protection needs: analysis of risks and vulnerabilities The literacy rate among youth is very low in Burkina Faso. Only 39 percent of youth Illiteracy is rampant in can read and write, which is much lower than the regional average (71 percent for Sub-Sa- Burkina Faso—61 percent haran Africa and 65 percent for West and Central Africa) and comparator countries such as Ghana (86 percent), Senegal (66 percent), and Benin (42 percent).6 The gender gap is rela- of youth are illiterate. tively high for literacy, with only 33 percent of women being literate compared to 47 percent for males. The level of attendance for secondary or tertiary school is also below the regional average. The number of children attending secondary or tertiary school as a per- centage of the total number of children of secondary school age is only 19 percent in Burkina Faso; in comparison, this figure is 44 percent for Benin, 42 percent for West and Central Africa, 38 percent for Senegal and for Sub-Saharan Africa overall, and 37 percent for Ghana. There is a clear bias with respect to wealth, as 45 percent of the wealthiest secondary school- age children in Burkina Faso are in secondary or tertiary education, compared to 5 percent for the bottom quintile. School-age children face a high risk of poverty as well as numerous obstacles to their education, which are likely to have a long-lasting impact on their work life and well-being. Poverty is predominant in rural areas where households derive their livelihoods mainly from subsistence agriculture. Completing secondary or tertiary education decreases the likelihood of agricultural employment by 23 and 29 percentage points, respec- tively. Additionally, workers with higher levels of educational attainment dominate jobs in services, while 99.5 percent of those employed in the agricultural sector have no education or are primary school dropouts (World Bank 2017). Addressing school dropout rates and improving literacy should be taken into account in the social protection strategy, along with improving supply-side constraints. Social protection policies should focus on improving demand—that is, the attendance rates for primary and secondary education—but also work to address supply constraints. Access to secondary education is limited, especially in rural areas, where only 20 percent of children live within 30 minutes of a secondary institution (figure 2.10). Another supply-side constraint is the recruitment of qualified teachers, which does not meet the pace of the rapidly growing student population. The average student/teacher ratio is 55 in Burkinabe primary schools; it is 47 in Benin and Niger, and 40 in Senegal. Moreover, working conditions for teachers are often precarious and unstable (World Bank 2017). Relevant social protection schemes can help compensate for the lack of appro- priate education and poor quality education. For example, social protection can focus on technical or vocation training, as well as on-the-job training programs. Such policies, however, are scarce in Burkina Faso. Where they exist (such as in training and job incentive programs), they focus mainly on urban youth who already have some level of educational attainment. These rates are for youth age 15–24, which somewhat overlaps the school-age classification used here. 6  26 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO Adults (age 18–59) Some of the main risks both young and middle-aged adults face is to live on subsistence agriculture, be inactive, and be employed in a low-paying job. Unemployment overall is low, with only 0.2 percent of the country’s population unemployed and 10.8 percent inactive. The main area of activity for both young (age 18–34) and mid- dle-aged (age 35–59) adults is agriculture; 64 and 69 percent, respectively, report this sector as their primary activity (figure 2.14). The other predominant activities are trade and industry. Interestingly, young adults tend to be less involved in agriculture than those of middle age (a 5 percentage point difference), and more involved in trade (3 additional percentage points); this indicates a slowly transitioning labor market. FIGURE 2.14  Sectors of activity for young and middle-aged adults Percent 80 Age 18–34 Age 35–59 60 40 20 0 Agriculture Education/ Food/hotel Construction Transport Other services Industry Inactive Trade health industry SOURCE: EMC 2014. The high proportion of those active in agriculture does not reflect the precar- iousness of their livelihood. Most of those active in this sector (92.5 percent) work in agriculture only for their own consumption, with just 7.5 percent actually selling their products (this is based on EMC 2014 calculations). Additionally, workers in the agricultural sector tend to be affected by natural hazards, with 76 percent facing at least one shock and an average of 1.7 shocks over a year (figure 2.15). Overall, there is no clear gender gap for adults in Burkina Faso, as the inactiv- ity rate for both men and women is around 11 percent. Women are, however, less engaged in agriculture (63 versus 68 percent), and more active in other services (7.6 versus 5.8 percent). Adults engaged in subsistence agriculture face the highest level of income instability (table 2.2). Most agricultural workers work fewer than six months a year, and only 22 percent of them work more than seven months. This contrasts greatly with activities 27 chapter 2. social protection needs: analysis of risks and vulnerabilities FIGURE 2.15  Exposure to shock by activity a. Workers affected by natural hazards b. Average number of shocks Percent Number 80 1.5 60 1.0 40 0.5 20 0 0 Food/hotel industry Other services Education/health Construction Transport Trade Industry Inactive Agriculture Food/hotel industry Other services Education/health Construction Transport Trade Industry Inactive Agriculture SOURCE: EMC 2014. such as those in health and education, and construction TABLE 2.2  Income composition across select and transport, in which about 70 percent of workers population groups work more than seven months a year. Additionally, those Income source All Rural Rural Q1 working in agriculture often need to find another job Farm 40.4 60.5 74.7 (35 percent of agricultural workers); only 13 percent of Wages from agriculture activity 0.4 0.5 1.1 those engaged in other type of activities hold a second Wages from non-agriculture activity 17.4 6.3 2.3 job (figure 2.16). The average number of months of work Self-employment 35.8 29.5 20.1 is only 5.8 in the agricultural sector, compared to 8.4 Private transfers 3.2 3.1 1.7 months for all other sectors (for a total average of 6.5 Other 2.8 0.7 0.1 months for all sectors). SOURCE: World Bank 2016a, table 4.2. FIGURE 2.16  Share of workers, by number of months worked within each activity type Percent Percent with other job < 3 months 4–6 months 7–11 months 80 40 70 35 60 30 50 25 40 20 30 15 20 10 10 5 0 0 Education/ Other services Construction Transport Trade Food/hotel Industry Inactive Agriculture health industry SOURCE: EMC 2014. 28 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO The elderly (age 60 and above) Even though the elderly account for only 5 percent of the total population in Burkina Faso, 41 percent of them live in poor households and face risks related to the absence of income replacement. Roughly 20 percent are inactive; while 71 percent are active in agriculture, uniquely in subsistence agriculture. Only 2.8 percent of the elderly receive an old-age pension and 0.4 a spousal survivor pension, but none in the poorest quintile are covered (based on calculations from EMC 2014 data). The elderly may also lack family support, as the average household size is smallest for this age category (9.7 versus 11.1 for children), and a significant number of the elderly live on their own (2.2 percent, compared to less than 1 percent for all other age categories). This is also the age group with the highest rate of illiteracy—90 per- TABLE 2.3  Share of population having a health issue cent. Access to health care is rather limited compared to over the last four months, by age group that for other age groups: only 50 percent of the elderly 0–5 6–17 18–34 35–59 ≥ 60 consult health professionals in case of health issues, compared to 60 percent for adults, even though the 8.8 8.7 9.3 9.4 10.7 elderly are more at risk (table 2.3 and figure 2.17). SOURCE: Authors’ calculations based on EMC 2014. 29 chapter 2. social protection needs: analysis of risks and vulnerabilities FIGURE 2.17  Key risks, the life cycle, and social protection in Burkina Faso RISKS NEEDS „„ Malnutrition: The percentage of underweight children „„ A comprehensive early childhood reached 19.2 in 2016 development strategy „„ Mortality: Under age 5 mortality rates remain very high „„ Structural programs to improve nutri- at 88.6 per 1,000 in 2016 tion, addressing supply and demand „„ Early childhood development: Policies are largely barriers lacking, and only 2% of children participate in early childhood education programs—one of the lowest rates EXPOSURE TO SHOCKS in the world „„ School dropout rates: 50% of primary school–age „„ Inclusive programs to promote education children and 57% of lower secondary age—which and address supply-side constraints increases to 69% and 77% for the poorest quintile „„ The provision of alternative education „„ Few teachers: There are 55 students on average per programs, such as training and intern- primary school class ships and “second chance education” „„ Literacy: 61% of youth are illiterate „„ High subsistence agriculture: 92% live off of farm- „„ Secure jobs in agricultural areas by ing and not from selling produce promoting mobility from one sector to „„ Unstable jobs: Work duration is only 6.5 months a year the other and encourage reconversion (5.8 for agriculture) through income-generating activities „„ Income precariousness: 35% of agricultural workers „„ Adaptive social safety nets that have a second job; 78% of agricultural workers face at address shocks from agricultural products least one natural hazard „„ Improve literacy and new qualification through training and subsidized work „„ Lack of stable income and pension: Only 2.8% have „„ Targeted social pensions pensions, and none in the bottom quintile have pensions „„ Improved access to health facilities „„ Lack of access to health care and illiteracy: 90% are illiterate and only 50% consult a health provider in case of health issues „„ Potentially weak family support chapter 3 institutional framework and spending review How is social protection articulated, and who are the actors involved in its provision? What are the current institutional arrangements and social protection strategies? These are the main topics discussed in this chapter, followed by a brief analysis of spending trends and composition as well as some benchmarking analysis. 32 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO An emerging social protection strategy This section provides a brief overview of the evolution of social protection and safety nets in Burkina Faso. Social protection is a recent development. Burkina Faso was long characterized by A TIMELINE OF THE an absence of a consolidated national social protection strategy, with limited attention given EVOLUTION OF SAFETY to social safety nets. In 2000, it became one of the first African countries to prepare a pov- NETS IN BURKINA erty reduction strategy paper; this served as the central framework for the government’s FASO economic and development policies from 2000 through 2010. However, the initial strategies focused mainly on supply-side interventions in the health and education sectors—such as the 2000: Burkina Faso is provision of social welfare services for vulnerable groups, income-generating activities, and one of the first African contributory social insurance schemes. Direct noncontributory transfers were only used on an nations to prepare a pov- ad hoc basis and were limited to a few vulnerable groups, such as vulnerable school/HIV-af- erty reduction strategy fected children and people with disabilities (World Bank 2012). Social safety nets nevertheless paper, although safety nets played a role in human capital development and crisis response strategies. receive limited attention. Government has taken one step forward, and two steps back. In 2007, the Ministry 2007: A national social of Labor and Social Security led a multisectoral process to develop a national social pro- protection policy is devel- tection policy with social safety nets at its core (Ministry of Labor and Social Security 2007). oped but not adopted. The With the development of this national social protection policy, the first steps toward a national government asks minis- social protection strategy were taken. The policy, however, was never adopted. Instead, the tries to develop their own government asked each ministry to develop its own sectoral strategy. sectoral strategies. The importance of social safety nets was publicly acknowledged in 2010. 2010: The first public Burkina Faso’s first economic and development policy framework to explicitly emphasize the document emphasizing the importance of social safety nets was the Strategy for Accelerated Growth and Sustainable importance of social safety Development (Government of Burkina Faso 2010). The strategy, which replaced the 2000 Pov- nets is produced: the erty Reduction Strategy, included social safety nets as a priority action under its second pillar. Strategy for Accelerated In addition to improvements in access to basic services, universal school enrollment, and Growth and Sustainable employment, the new strategy proposed that social safety nets be extended. Development. Also, Burkina Faso becomes a Crises trigger initiation of a comprehensive social protection strategy. At the end pilot country for the UN-led of 2000, Burkina Faso suffered several consecutive crises, including natural disasters, politi- Global Social Protection cal and social upheaval, and global economic crises—all of which triggered the initiation of a Floor Initiative. social protection strategy. The food, fuel, and financial crises that began in 2008 reverberated throughout Burkina Faso’s economy, causing a spike in food prices, job losses, poverty, and 2012: Burkina Faso’s first social and political tension and conflict. Since then, the government’s interest in social pro- National Social Protection tection has increased—as an economic stabilizer in times of crisis, and as a mean to alleviate Strategy is adopted; it poverty, manage risks, and reduce adverse coping strategies. consolidates sectoral strategies, with safety nets In 2010, Burkina Faso became one of the pilot countries for the United Nations’ (UN’s) Global at its core. Social Protection Floor Initiative. The aim was to ensure basic levels of social protection 33 chapter 3. institutional framework and spending review access to essential services and social transfers for the poor and vulnerable. The prime minis- ter requested the aid of international partners in building the foundation for a social protection floor; this also entailed discussion of a national social protection strategy. The National Social Protection Strategy was announced by national decree Decem- ber 31, 2012. By weaving multiple national sectoral strategies into one, the aim was to reinforce cross-sectoral coordination. The heavy droughts of 2012 led to a food and nutrition crisis of devastating proportions, compromising some of the human development accom- plishments in prior years; the situation was aggravated by regional fighting and an influx of refugees. In 2013, the government adopted the strategy, with social safety nets at its core. The strategy had two main objectives: (1) to develop adequate and sustainable protection mechanisms against idiosyncratic and exogenous shocks through the use of safety nets, and (2) to extend social insurance coverage to informal and agricultural sectors. The execution of the national strategy was planned as three-year action plans, using a multisectoral approach involving relevant ministries and institutions. As part of the strategy process, three-year action plans were developed to organize the activities from the various strategies and to facilitate implementation. The 2016–18 action plan (box 3.1) contains social protection programs from numerous ministries and partners, including UNICEF, the World Bank, and the World Food Programme. Monitoring mechanisms were put in place to assess action plan implementation. Further, annual reports have documented social protection progress on a yearly basis, and outlined and planned interventions for the year ahead. The National Council for Social Protec- tion (Conseil National pour la Protection Sociale, CNPS) was made responsible for collecting the sectoral strategies. Health and education services constitute the bulk of social protection programs. The National Strategy for Social Protection rests on six pillars, with most resources allocated toward two of these pillars, namely to increase access to basic social services and to improve access to adequate social transfers (CNPS 2016b). „„ The first pillar focuses on improving transfers to the poor and vulnerable (e.g., cash transfers, food distribution, general price subsidies) and accounts for 29.4 percent of Burkina Faso’s total social protection spending. „„ Most of the spending (56.8 percent) is allocated toward the second pillar—increasing access to social services for the poor and vulnerable, primarily education (46 percent of spending is related to education only). Approximately one-tenth of the spending is allocated toward programs that facilitate access to public and private social centers. „„ The third and fourth pillars—employment promotion and access to minimum income, and increasing social insurance coverage for formal/informal workers—account for less than 1 percent of spending (2015). „„ The final two pillars focus on improving institutional and operational capacities. 34 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO BOX 3.1  Burkina Faso’s National Social Protection Strategy Action Plan 2016–2018 B urkina Faso’s National Social Protection Strategy for the period 2013–22 was articulated around six pillars. Action plans have been devel- oped for three years at a time—with defined priority objectives under which several programs will be articulated. The objectives for the years 2016–18 follow. Improve social safety nets for the poorest and most vulnerable households 1 „„ Objective 1: Achieve food security „„ Objective 2: Protect against shocks Increase access to social services for vulnerable groups 2 „„ Objective 1: Improve access to education „„ Objective 2: Improve access to health care Promote employment and access to minimum income 3 „„ Objective 1: Facilitate labor market access and income-generating activities „„ Objective 2: Improve income security Extend social insurance coverage, including for informal workers 4 „„ Objective 1: Improve social insurance service providers „„ Objective 2: Set up a universal health insurance program „„ Objective 3: Include informal workers under social insurance programs Improve governance 5 „„ Objective 1: Organize a monitoring and evaluation system „„ Objective 2: Reinforce the legal and institutional social protection framework „„ Objective 3: Improve the management of and timely response to crisis/natural disasters Reinforce capacities 6 „„ Objective 1: Improve operational capacities „„ Objective 2: Improve institutional capacities „„ Objective 3: Improve competencies of social protection staff SOURCE: CNPS 2016a. Multiple strategies lead to fragmentation. Although coordination was facilitated through the adoption of the National Social Protection Strategy, the endorsement of numerous parallel sectoral strategies complicated matters. In addition to the National Social Protection Strategy, the government encouraged and endorsed additional strategies related to social protection, resulting in a fragmented approach, despite progress. Even though the National Social Protection Strategy consolidates information from several ministries and includes most of their action plans, gaps in coverage remain. Moreover, there are overlaps and potential 35 chapter 3. institutional framework and spending review duplication of effort between the various national strategies; further consolidation is needed. The following details the main co-existing strategies. „„ The National Plan to Support Vulnerable Populations against Food Insecurity and Malnutrition. The objective of this 2012 plan—spearheaded by the Secretariat of the National Council for Food Security (Secrétariat Exécutif du Conseil National de Sécurité Alimentaire, SE-CNSA) was to improve food security and address the country’s most pressing nutritional needs. A committee determined anticipated needs through analytic assessments and forecasts. Yearly action plans were elaborated by a technical committee, defining interventions and the roles of the various actors involved in food security. The areas of intervention for 2015–16 were (1) food security through food distributions, food subsidies, unconditional cash transfers, and cash for work; (2) support to agricultural production; and (3) children/ nutritional improvement (SE-CNSA 2015, 2016). There is a clear overlap with the National Social Protection Strategy, and some of the same programs can be found in both strategies, while other programs are not reported at all.1 „„ The National Plan for Preparation and Response to Catastrophes. The objective of this 2009 plan was to effectively respond to diverse natural and human- itarian crises across the country so as to be able to better anticipate shocks through yearly analysis of the occurrence of natural disasters, food deprivation, violence/ political shocks, and pandemics. It entailed the creation of the National Council for Emergency Relief and Rehabilitation (Conseil National de Secours d’Urgence et de Réhabilitation, CONASUR) in 2009. In 2014, the plan focused on natural hazards, food insecurity, and pandemics.2 „„ The National Resilience Priorities Program. In 2014, a consultation process involving the Global Alliance for Resilience Initiative in Sahel and West Africa led to the development of this program. The SE-CNSA is responsible for program moni- toring, evaluation, and overall coordination. The analysis and statistics divisions of the ministries responsible for agriculture, health, education, social action, and infra- structure collaborate directly with the SE-CNSA. The program for the period 2016–20 builds on analytical studies that prioritize actions and existing national plans/ strategies that address the identified priorities. Programs from the National Social Protection Strategy are only partially included.3 Food distribution and subsidies provided by the National Society for the Management of Food Security 1  (SONAGESS) and the National Council for Emergency Relief and Rehabilitation (CONASUR) are included in both documents, but cash transfers, workfare, agricultural subsidies, and nutrition programs are not included in the National Social Protection Strategy. Programs under CNSA have been added to the aggregate for social protection in 2015, but it is unclear if these are actual or budgeted expenditures. Agricultural subsidies are included in the National Social Protection Strategy Action Plan for 2016–18. Program spending is only available for 2013–14. 2  The partially included plans are the National Social Protection Policy, the National Strategy for the 3  Promotion and Protection of Persons with Disabilities, the National Policy for Social Action 2014–2023, the National Employment Policy, the National Education Policy and Technical and Vocational Training, ongoing policies and strategies for access to health services for vulnerable people, the National Health 36 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO „„ National Education Strategy. The objective of the 2012 strategy, adopted by the Ministry of National Education and Literacy (2012–21), was to address both demand- and supply-side constraints to education by (1) increasing access to formal education, (2) improving the quality of formal education, and (3) developing nonfor- mal education. While the majority of the activities under the plan focus on supply (training of teachers, increasing the number of schools and training centers, etc.), most of its demand-side activities (such as school feeding, school supply kits) are also included in the National Social Protection Strategy. „„ National Health Strategy. The National Health Strategy was adopted in 2011 and revised in 2014, and is executed by the Ministry of Health. The strategy focuses on eight strategic axes: improving governance in the health sector, improving quality of services, development of human resources, promotion of health and disease control, infrastructure development, promotion of research, improving monitoring and infor- mation systems, and increasing financial health care accessibility. Most demand-side activities are also included in the National Social Protection Strategy, such as fee waivers for health care (HIV-infected patients, pregnant women and their children, etc.) and the pilot for universal health insurance. „„ National Climate Change Adaptation Plan. The objective of this 2015 plan COORDINATING (which has, as of this writing, not yet been translated into an action plan), adopted by MECHANISMS the Ministry of Environment and Fishery Resources, is to better anticipate the impact 2010: A Joint Ministerial of climate change by increasing the resilience, adaptation, and protection of the Committee on Social most vulnerable through development projects. The link to social protection is clear, Protection was estab- since one of the medium-term objectives is to improve social protection for vulnera- lished, tasked with kicking ble communities/households in order to secure their means of livelihood. off efforts to develop a national strategy. 2013: A coordinating Institutional and coordination body—the National Council for Social Protec- arrangements tion—was created to help This section discusses Burkina Faso’s institutional set-up for social protection and its internal reduce ministerial frag- coordination mechanisms; it concludes with a brief discussion of key coordination challenges. mentation and consolidate interventions. Intra-minis- As programs grow, an effective coordination mechanism becomes critical to ensure terial entities were created institutional efficiency. As Beegle, Coudouel, and Monsalve (2018, 184) note: within each ministry to coordinate and operation- Institutions—defined as laws, policies, and strategies—shape human behavior and human alize from within. interaction and are therefore central to the delivery of social safety nets. Often understood or defined as the rules of the game, institutions shape all aspects of social safety nets, ranging 2016: The first benefi- from establishing the benefit eligibility criteria to the rules that govern the organization that ciary/social registry was introduced, covering Development Plan, the Basic Education Strategic Development Program, the National Nutrition Policy, 24,000 beneficiaries. the National Drinking Water Supply and Sanitation Program, the National Policy of Land Security in Rural Areas, the National Rural Sector Program, and the National Policy for Scientific and Technological Research. 37 chapter 3. institutional framework and spending review delivers the social safety net program (including its mandate and human resource policies) and the laws that govern the sector… Beegle, Coudouel, and Monsalve (2018, 26) observe that if social safety nets are to be ade- quately expanded, institutions must evolve along multiple parameters, including the anchoring in laws and policies, mechanisms for coordination and oversight, and arrangements for management and delivery. Small pilot interventions may show results and contribute to building political support for the expansion of social safety nets, but broadening coverage typically requires consolidation. Often…program management will be shifted to government ministries or agencies over time; program designs and processes will become standardized; staffing will be transferred to the civil service or outsourced; and more comprehensive rules for the overall safety net system will be formalized in policies, strategies, and laws. Poverty is multidimensional; thus, safety net TABLE 3.1  Institutions responsible for social interventions are spread across a wide range of protection programs and number of programs (2015) ministries. In Burkina Faso, approximately 20 minis- No. of Spending Institution programs (mil. CFAF) tries and various partners oversee the country’s more than 130 social protection programs (table 3.1). Three Ministry of Women, National Solidarity and Family 45 8,372.98 ministries handle the largest number of social protection Ministry of National Education and Literacy 22 25,868.57 programs (88): the Ministry of Women, National Solidar- Ministry of Youth, Training & Vocational Integration 15 6,193.17 ity and Family (formerly, the Ministry of Social Action and Ministry of Labor and Social Security 8 24,510.13 National Solidarity); the Ministry of National Education Ministry of Higher Education 7 15,773.81 and Literacy; and the Ministry of Youth, Training and Ministry of Agriculture and Water Resources 6 22,100.03 Vocational Integration. While the Ministry of Women, Ministry of Economy and Finance 5 69,501.94 National Solidarity and Family is in charge of the largest Ministry of Health 4 8,011.95 number of social protection programs (45), the bulk of Permanent Secretariat of NGOs 5 219.04 social protection expenditure falls under the Ministry of Ministry of Infrastructure and Transportation 4 — Economy and Finance, which oversees only 5 programs. International partners 4 28,139.18 However, these five account for 28 percent of Burkina Ministry of Animal Resources 3 4,388.68 Faso’s total social protection spending—the largest Technical Secretariat for Universal Health Insurance 3 17.82 ministerial share—and comprise large food programs Directorate of Nutrition 2 15,972.24 and universal subsidies. The 45 programs overseen by Ministry of Environment and Fisheries Resources 2 357.41 the Ministry of Women, National Solidarity and Family Ministry of Industry, Trade and Handicrafts 1 4,169.25 represent only 3.4 percent of social protection spending; Ministry of Territorial Development 1 1,332.00 the 22 programs of the Ministry of National Education SOURCE: See box 3.3. and Literacy account for 10.5 percent of spending (fig- NOTE: — = not available. ure 3.1). Numerous actors are involved in implementation of the Burkina Faso’s social protection programs under each responsible ministry. These actors are related in a variety of ways, either working independently with no interaction and no coordination (coex- istence), or linking with the state or cooperating, forming partnerships or contracting with each other. Both the public and private sectors, donors, and nongovernmental organizations (NGOs), civil society, and community-based organizations (e.g., religious) are involved in 38 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO FIGURE 3.1  Number of programs and social protection spending by institutions (2015) Share of total social protection spending Number of programs 30 50 25 40 20 30 15 20 10 10 5 0 0 Mi Int Mi Mi Mi Di Mi Co MA Mi Mi Mi Mi Mi Mi Pe Te Ot Mi re he c n. n. n. n. n. n. n. n. n. n. n. r n. ’l p m SS m h. cto pa r Ag Yo An Te Ec Na La Hi He Ind En Inf an ar Se N gh ss u r ra on bo .& im v. ra alt tn t’l en rit us th c. te ion er .& er & om r& al or h tS Ed ,T try Wa Un s of Fis ial Ed Re rn Tra .& ,T ec Int y& So iv. t Nu g& er h uc De so ra .N ’l ns er cia Lit He Fin de at tri Re ur v. ies GO p. Vo er alt lS ion tio ce ,e so a ac c s nc hI Re ec n s tc. . In ur y e ns ur s. ce teg ity . s . SOURCE: See box 3.3. NOTE: MASSN = Ministry of Social Action and National Solidarity, former name of the Ministry of Women, National Solidarity and Family. delivering safety net interventions. In Burkina Faso, community-based organizations play an important role in service delivery, while donors play a critical role in supporting programs. Major donors include the European Union, the International Labour Organization, UNICEF, and the World Bank. Since 2006, financing for social safety net programs has become increasingly donor-dependent. Donors predominantly focus on nutrition programs, overseeing in particular three large nutrition programs that represent 11.5 percent of total social protection spending (figure 3.1). The government has indicated that it would like to strengthen local-level provision. The government recognizes that existing social safety nets have a limited impact on the poorest and most vulnerable, due in large part to an inadequate institutional arrangement. Safety nets have not been used to their full potential in Burkina Faso. However, the government has undertaken several initiatives toward developing a more coherent social protection approach and a more efficient social safety net system. In 2010, a joint Ministerial Committee on Social Protection was established, which created opportunities for cross-ministerial dialogue on safety net design and implementation. The committee was placed under the chairmanship of the Permanent Secretary of the Ministry of Economy and Finance. It was this committee that was initially tasked with kicking off efforts to develop the country’s National Social Protection Strategy. 39 chapter 3. institutional framework and spending review Intra-ministerial coordination entities began to be introduced within each min- istry involved in social protection activities. Around 2013, internal coordination/ interministerial entities were created to improve (1) internal coordination within the respective ministry and (2) the operationalization of programs implemented by the ministry. Each ministry has an entity that focuses on social safety nets and another that focuses on social insurance activities. Burkina Faso’s journey toward more comprehensive social protection and social safety nets progressed another step with the creation of the National Council for Social Protection in 2013. CNPS was created to help with coordination across min- istries, reduce fragmentation, and facilitate consolidation of the nation’s social protection interventions. This interministerial council was made responsible for the execution of the National Social Protection Strategy. Its executive leadership, which provides strategic over- sight and defines policy direction, consists of the prime minister, the minister of Economy and Finance; the minister of Women, National Solidarity and Family; and the minister of Labor and Social Security. The executing body consolidates social protection interventions, produces annual progress reports, and monitors implementation. The CNPS Permanent Secretariat was made responsible for the collection, review, and consolidation of the national strate- gies—feeding into the three-year action plans—and for monitoring and evaluation of plan implementation. The Permanent Secretariat was also made responsible for the production of the annual social protection reports. Strengthening the coordination and efficiency of social safety net systems goes hand in hand with investments in integrated systems, such as a social registry. Since the social safety net system in Burkina Faso involves many sectors and comprises a large set of programs, investing in integrated systems would reduce cost-inefficiencies and duplication of effort. When implementing programs, it is critical to identify, accurately tar- get, and keep track of beneficiaries to ensure program efficiency. A registry of beneficiaries is a database with socioeconomic information on beneficiaries (both household units and individuals), such as address, age, sex, school age, etc.; and their program eligibility as well as benefits already received. This can be very useful in coordinating between and across programs as individuals may be eligible for various programs (Beegle, Coudouel, and Mon- salve 2018). A single registry of beneficiaries minimizes inclusion and exclusion errors, while facilitating transparency and providing governments and partners with a central mechanism to identify potential program beneficiaries. Such a registry is used in 26 African countries, and is being developed in 16 others—including Burkina Faso. Registry coverage ranges from 89 percent of the population in Rwanda and 52 percent in Lesotho to 0.3 and 0.1 percent in Mozambique and Zambia, respectively (Beegle, Coudouel, and Monsalve 2018). Discussions are on track for establishing an integrated social registry, but an integrated beneficiary registry has not yet been established.4 Beneficiary registries track information on beneficiaries of specific programs based on program 4  enrollment decisions. They underpin the information system for beneficiary operations management systems. Beneficiary registries contain information on beneficiaries (individuals, families, households). Social registries are information systems that support the processes of outreach, intake, and registration, and assessment of the needs and conditions to determine potential eligibility for social 40 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO In Burkina Faso, initiation of a much-needed social registry began in 2015 with the launch of the World BOX 3.2  Integrated social registries: examples Bank–supported unconditional cash transfer program Burkin-Naong-Sa Ya—which, as of October 2018, T he government of Senegal is building a registry for programs had registered 134,168 households and 908,537 that address chronic poverty and support vulnerable house- individuals, covering almost 5 percent of the popu- holds. In 2012, it created a social protection agency, housed in the lation. As noted in box 3.2, registry development in Office of the President, to lead formulation of a social protection the Republic of Congo and Mali also began with the strategy and the design and coordination of social safety net introduction of a large-scale cash transfer program. interventions. As a primary pillar of this effort, the government Progress has been made in Burkina Faso, although established a unified national registry, which, by 2017, included challenges remain. The CNPS facilitated a roundtable data on the 450,000 poorest households nationwide (around of government agencies and donors with an interest 30 percent of the population). The registry already serves as an in the social registry. It laid out a roadmap toward entry point for several targeted interventions, including Senegal’s an integrated social registry, and the first steps have main conditional cash transfer program and the subsidized health already been accomplished, notably the development insurance program; its use is expected to expand, particularly to of a jointly accepted, harmonized questionnaire. Some programs designed to respond to shocks. The registry is housed in technical constraints have been identified; these can a dedicated department, independent of the department in charge be overcome, provided the various stakeholders con- of implementation of national cash transfer programs. tinue to align their agendas. In the Republic of Congo and in Mali, cash transfer programs have Despite notable progress in creating a more steadily expanded since their launch in 2013, and the establishment conducive institutional environment, Burkina of an integrated social registry has been a key part of this process. Faso’s social protection approach remains The two registries were developed while the programs were being highly fragmented. Strong efforts have been conceived. In Mali, the creation of an integrated social registry made toward a more comprehensive and inclusive began with the establishment of the Jigisemejiri Cash Transfer Pro- social protection approach. However, the institu- gram. Its objectives are to reduce intake and registration costs, and tional arrangements and interministerial coordination facilitate the rapid expansion of programs to respond to shocks. In mechanisms—the prerequisite and foundation of an the Republic of Congo, a registry was developed through the condi- effective social protection system—remain weak. tional cash transfer program. Most programs continue to operate in silos, with little communication between sectors and few attempts to SOURCE: Beegle, Coudouel, and Monsalve 2018. explore synergies. While cross-sectoral coordination has been made somewhat easier with the introduction of the National Social Protection Strategy, government endorsement of numerous sector-specific plans has, as discussed above, resulted in a highly fragmented approach.5 There is a clear overlap between the various strategies and plans, and further consolidation is needed. programs. They contain and maintain information on all registered households, regardless of whether they eventually benefit from a social program. Thus, households in social registries are not identified as “beneficiaries” but as “registered households.” These plans include the Health Development National Plan, adopted by the Ministry of Health to improve 5  preventive care of specific vulnerable groups and access to care for poor households; the Education for All Strategy of the Ministry of National Education and Literacy, which provides public school fee waivers targeted to girls and school feeding programs, scholarships, and conditional cash transfers (World Bank 41 chapter 3. institutional framework and spending review Social spending trends Burkina Faso’s overall social protection spending has increased at a steady pace over the past decade, although spending trends have varied by social protection category. Expenditures increased for social safety nets as well as for social insurance, while there has been a notable decrease in general subsidies expenditure. Total estimated social protection expenditure in Burkina Faso in 2015 was 4.2 percent of gross domestic product (GDP), which was estimated at about $10.4 billion (CFAF 6.32 trillion).6 Estimated spending included for social insurance, social safety nets, labor market programs, and general subsidies. Box 3.3 details how these estimates were derived. FIGURE 3.2  Social protection expenditures by category: 2015 Social safety nets constituted the bulk of spend- ing in 2015. This category accounted for 2.3 percent Billion CFAF of GDP (CFAF 147 billion), an increase over 2005, when 4.5 4.0 social safety net spending accounted for 0.3 percent of 3.5 GDP (CFAF 8.3 billion). In 2015, the next largest categories 3.0 2.5 of social protection spending were, in descending order, 2.0 general subsidies at 1.1 percent of GDP and social insur- 1.5 1.0 ance at 0.8 percent. General subsidies have been on a 0.5 downward trend, particularly for the electricity sector. Labor 0 Social General Social Labor mkt Total social market programs constituted about 0.2 percent of GDP in safety nets subsidies insurance programs protection 2015; since they are a rather recent phenomenon in Burkina Faso, it is not possible to track them over time (figure 3.2). Source: See box 3.3. Social safety nets Since 2005, social safety net expenditures have increased at a steady pace. The government’s They accounted for 0.3 percent of GDP in 2005 and reached 1 percent (CFAF 41 billion) of appetite for improving the GDP in 2010 (World Bank 2012). From 2010 onward, there has been a sharp increase in social safety net expenditure, reaching 2.3 percent (CFAF 147 billion) of GDP in 2015 (figure 3.3). effectiveness of social Government appetite for more effective social safety nets has also been reflected in the intro- safety nets is large, and it duction and scale-up of new programs, such as Burkin-Naong-Sa Ya (box 3.4). is reflected in the growing Reporting mechanisms have improved. Until 2010, expenditure estimates covered investments and sharp targeted subsidized food sales, targeted food distribution, nutrition programs, and school feeding programs; but did not include food distribution and public works programs since their increase in social safety net spending in the past decade. 2012); the National Strategy for Food Security of the Ministry of Agriculture and Water Resources, which provides nutrition programs and health care access for mothers and children; the National Plan for the Organization/Coordination of Emergency Relief/Rehabilitation of the Council for Emergency Relief and Rehabilitation and other decentralized institutions, which manages the national food security stock; and the National Policy for Social Action, whose main focus is on welfare services. Source for GDP data: International Monetary Fund World Economic Outlook database; accessed 6  October 2017. 42 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO BOX 3.3  Methodology note on social protection spending data collection S everal documents were used to estimate spending social protection spending aggregates over the years, as detailed below. The launch of the CNPS annual reports in 2013 was a key milestone in tracking and collecting expenditure information. Because spending data for 2011–13 were relatively scarce, it was decided not to include or present data for these years. Data for 2016 on social safety nets were incom- plete at the time of this report’s preparation, but are presented at the program level where available (see appendix A). Overall, the data collection process assembled data from over 143 social protection programs. Data were collected through interviews with major stakeholders and by reviewing the following documents and strategies: „„ Social protection programs: The annual CNPS report on social protection (CNPS 2016b) provides information on the programs included in the National Action Plan 2015–17 (CNPS 2015a), with disbursement activities for the year 2015. Most programs listed in the action plan were included, but expenditures that are not directed toward the population—such as spending on meetings and sets of activities related to capacity building and better governance—were excluded. Programs were reclassified to reflect the definition of social protection categories used here. Therefore, the CNPS social protection aggregate estimate is not comparable to that used here. Program descriptions from the annual report (CNPS 2016b) were complemented and supplemented with information from the action plan (CNPS 2015a), particularly with regard to program institutional arrangements. „„ Programs related to malnutrition, small unconditional cash transfers, cash for work, and other agricultural subsidies not included in CNPS 2016a: Nutrition plans (SE-CNSA 2015, 2016); however, because these plans only report on budgeted cost, and not on execution, the estimates provided here may overestimate social safety net spending for those programs, especially if execution rates are low. „„ Training program for young entrepreneurship and THIMO for Youth: Execution report of the Youth Employment and Skills Develop- ment Project (UCP 2016). „„ 2005–10 social safety net estimate: Derived from previous social safety net review (Cherrier, del Ninno, and Razmara 2011). „„ 2014 social safety net estimate: Based on CNPS annual report (CNPS 2015b), excluding some programs that do not fit the definition of social safety net. Because the data collection effort of this review focuses on 2015, 2014 data potentially underestimate total spending. „„ Electricity subsidies: Estimated using World Bank (2016c); because 2015 data are not available, 2016 data were used in the aggregate of the general subsidy. „„ Gas and fuel subsidies: Key informant interviews (through the CNPS). „„ Cotton subsidies: International Monetary Fund Article IV Consultation (IMF 2016) and World Bank Systematic Country Diagnostic (World Bank 2017). expenditures were not tracked at the time. Since the introduction of the annual CNPS reports in 2013, it has become easier to assess expenditure, as a larger set of social safety nets implemented by various actors is now taken into account. The 2015 spending estimate was based on many different sources and takes more programs into account than does the CNPS annual report (CNPS 2016b). 43 chapter 3. institutional framework and spending review On average, Burkina Faso outspends other Sub-Sa- FIGURE 3.3  Total social safety net expenditure and as haran countries on social safety nets relative to a percentage of GDP, 2000–15 GDP. While on average, Sub-Saharan Africa countries spend 1.7 percent of their GDP on social safety nets, % GDP Million CFAF Burkina Faso spends approximately 0.6 percentage points 2.5 160 more (figure 3.4). 140 2.0 120 Burkina Faso’s social safety net expenditure 1.5 100 is higher than the average for developing and 80 1.0 60 transition countries globally. Burkina Faso spends 2.3 percent of GDP on social safety nets, compared to an 40 0.5 average of 1.5 percent of GDP by developing and transition 20 countries on these programs. Spending by region is as fol- 0 0 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 lows (Beegle, Coudouel, and Monsalve 2018): SOURCES: World Bank 2012 for 2005–10; estimates for 2014–15 are based on „„ Europe and Central Asia—2.2 percent CNPS 2015b and 2016b, and SE-CNSA 2015 and 2016. „„ Sub-Saharan Africa—1.5 percent NOTE: No social safety net estimates were produced for 2011–13 given the scar- „„ Latin America and the Caribbean—1.5 percent city of data. The first CNPS annual report was produced in 2013, but the effort „„ East Asia and Pacific—1.1 percent to track program expenditures was only at its beginning, and data were incom- plete and not comparable with previous estimates. „„ Middle East and North Africa—1.0 percent „„ South Asia—0.9 percent That being said, countries differ substantially in absolute average per capita social safety net spending in terms of U.S. dollars, in purchasing power parity terms. For instance, Sub-Sa- haran African countries spend an average of $16 per citizen annually on social safety net programs, whereas countries in Latin America and the Caribbean spend an average of $158 per citizen annually—although both regions spend 1.5 percent of their GDP on social safety nets (Beegle, Coudouel, and Monsalve 2018). FIGURE 3.4  Social safety net spending in Africa Percent of GDP 10 8 6 4 2 0 Central African Republic South Sudan Lesotho Mauritius Namibia South Africa Liberia Burundi Benin Mauritania Cabo Verde Seychelles Angola Guinea-Bissau Burkina Faso Botswana Sierra Leone Guinea Malawi Niger Rwanda Madagascar Senegal Mozambique Ethiopia Uganda Congo, Dem. Rep. Sudan Tanzania Comoros Cameroon Mali Ghana Kenya Chad Nigeria Togo Zambia Gabon Congo, Rep. Côte d’Ivoire São Tomé and Príncipe Somalia SOURCE: Beegle, Coudouel, and Monsalve 2018. 44 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO BOX 3.4  Burkin-Naong-Sa Ya boosts rural incomes to break cycle of intergenerational poverty B urkin-Naong-Sa Ya—which in the Mooré dialect means “end of poverty in Burkina Faso”—is a social safety net project begun in 2014 with $55 million in funding from the World Bank. It was implemented by the government in the North, Central West, Central East, and East regions, where over 60 percent of the nation’s poor households are concentrated. The project will be scaled up to include first the Sahel and then the Boucle du Mouhoun regions. The program brings together a mix of interventions in health and education to create an adaptive system designed to improve the effectiveness of public policies on social protection, and to build resilience in poor and vulnerable households to climate change and other shocks. It also enables these households to diversify their sources of income. Over 114,500 fragile households, selected on the basis of proxy means testing, are currently receiving a quarterly transfer of CFAF 30,000 (around $60) for households with fewer than five children and CFAF 40,000 (around $80) for households with five or more children. As the project coordinator, Emile Zabsonré, explains: “Thanks to these transfers, project beneficiaries are able to provide three meals a day for their families, cover school fees and health expenses for their children, and invest in income-generating activities, such as small businesses, live- stock rearing, and horticulture.” To address poverty in all its dimensions, the project links the award of cash transfers to activities that promote better nutrition, health, and education outcomes. Awareness raising and monitoring sessions are targeted to women living in fragile households, organized in groups of 20–25 members. Each month, they receive counseling on health, hygiene, and nutrition, as well as on the cognitive development of their children. Project community facilitators pay regular house visits to monitor progress and check on children’s physical and cognitive development. Finally, they mobilize the village each month to discuss ways to improve living conditions for all residents. “Our main objective is to break the intergenerational transmis- sion of poverty through actions that promote the cognitive and affective development of children and better schooling,” notes Gilberte Kedote, the World Bank project manager. The program is implemented by the government of Burkina Faso and cofinanced by various donor partners, including the World Bank. In January 2017, the project received additional financing of $6 million from the Adaptive Social Protection Multidonor Trust Fund. This fund is managed by the World Bank and aims to enhance the access of poor and vulnerable populations of the Sahel to effective social adaptive pro- tection programs. It seeks to extend its reach into other regions to help lower the national poverty rate from 40.1 percent in 2014 to less than 35 percent by 2020. SOURCE: World Bank 2018a. Government subsidies Energy subsidies Burkina Faso historically relied on subsidies to lower the cost of fuel and food as the principal means for protecting the poor and sharing wealth. However, it has been increasingly recognized that there are less expensive, yet more effective, methods of protecting the poor. Subsidies often fail to deliver economic or social value equal to the amount spent on them. One of the fundamental problems with subsidies is that they benefit 45 chapter 3. institutional framework and spending review the nonpoor more than the poor; this is particularly true when it comes to fuel and energy subsidies. As Chuhan-Pole (2012) notes: Expenditure data for seven African countries show that the distribution of these subsidies is disproportionately concentrated in the hands of the rich. Richer households spend a larger amount on fuel products, and, consequently benefit more than poorer households from any universal subsidy on these products. On average the richest 20% receive over six times more in subsidy benefits than the poorest 20%. Subsidies can also distort consumption and economic activity in unproductive ways. A pre- liminary estimate of energy subsidies revealed that they constituted a notable 1.05 percent of GDP in 2015; gas subsidies accounted for 0.36 percent and fuel subsidies for 0.23 percent (see figure 3.1). These estimates do not reflect losses in consumption taxes due to the subsi- dies. This review is an opportunity to inform current government policy processes undertaken in the energy sector and in the budget in general. General subsidy expenditure High electricity subsidies have weighed down Burkina Faso’s national budget has been on a downward over the last decade. During the years 2010–13, CFAF 86 billion went to electricity sub- trend, and since 2016 the sidies only. These subsidies have been inefficient, cost-ineffective, and fragmented. Further, they have not been paid, or have been received with delay (CFAF 28 billion in arrears for government has taken major 2014–15), which has forced the National Electricity Company of Burkina Faso (SONABEL) to steps aimed at reducing rely on costly short-term debt from local banks (World Bank 2016c). the burden of electricity Overall, general subsidy expenditures have decreased, with electricity rep- subsidies—representing resenting the sharpest decrease—from CFAF 86 billion to CFAF 21 billion in a major shift in social 2015. Fuel subsidies decreased from CFAF 24.5 billion in 2010 to CFAF 14.2 in 2015. While spending on gas subsidies is only available since 2015, these expenditures also appear to be protection spending. declining, dropping from CFAF 22.9 billion in 2015 to CFAF 18.8 billion in 2016 (see figure 3.1). Since 2016, the government has taken major steps to lessen the burden of elec- tricity subsidies and improve the financial sustainability of the energy sector. Among other actions, the government has adopted a new pricing structure and introduced an annual electricity subsidy cap: $21 billion per year in 2016 and 2017, respectively. This represents a major shift in social protection spending—although there is no information available yet on how the generated savings were used. Recent documents indicate that elec- tricity did not decrease in 2017, and actually increased compared to 2016 (ARSE 2017). The government has aimed at reinforcing transparency and predictability in budget transfers to the sector, while gradually passing on fuel price variations to electricity tariffs. It also aims to diversify energy use and rely on cheaper sources. The government has indicated a keen interest in and willingness to move toward a more inclusive social protection system. It seeks policy recommendations on how to improve the social safety net system and redirect subsidy funding toward more effective safety net programs. There is an interest in further embarking on energy subsidy reform; the government may need some support in dealing with the complexity of this issue. 46 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO Agricultural subsidies The government paid CFAF 6.5 billion directly to cotton farmers in 2015; how- ever, its poverty alleviation impact is questionable since cotton farmers are typically better off than producers in other farming sectors. Since the cotton sec- tor is highly exposed and vulnerable to exogenous shocks—both adverse climate and price fluctuations—the government decided to introduce a cotton subsidy in 2007. No information is available on expenditure trends or policy reform. Food subsidies decreased from approxi- mately CFAF 3 billion over the years 2008–16 (ranging from CFAF 4.2 billion in 2008, peaking at CFAF 8.7 billion in 2013 as a result of the food crisis—and continued to decrease until reaching CFAF 1.5 billion in 2016. Social insurance Social insurance labor market penetration is difficult given the highly informal character of economic activities. Contributory programs remain out of reach for most Burkinabes, because a majority of workers are active in informal sectors and are thus unable to benefit from the social insurance for which only salaried employees (and their dependents) in formal sectors are eligible. Agricultural, domestic, and small enterprise workers are often excluded from social insurance. Social insurance programs are accumulated in special funds out of which benefits are paid according to specified rules so as to identify those participants Contributory programs who suffer the occurrence of relevant risk. remain out of reach for most Despite the expenditure increase and a doubling of public pension beneficiaries Burkinabes. from 2000 to 2016, only 69,000 individuals receive old-age pensions. In Burkina Faso, the number of elderly is estimated at more than 1.2 million. Formal private sector employees and their dependents are covered by the National Social Security Fund (Caisse nationale de sécurité sociale, CNSS). Fund challenges include high operating costs and a low transformation rate for social security contributions. CNSS is a public social insurance fund, under the technical supervision of the Ministry of Labor and Social Security; the Ministry of Economy and Finance provides financial supervision. CNSS provides employees with benefits such as pension allowance, disability allowance, survivors’ pension, occupational hazard and family and maternity allowances, as well as scholarships for children and allowances for medical checkups. It is managed by a tri- partite board and a director, who administers the program. The Independent Pension Fund for Civil Servants (Caisse autonome de retraite des fonctionnaires, CARFO) provides allowances and benefits to public servants and military officials. However, beneficiaries are not insured against the risk of sickness. Services include pension allowance, disability allowance, and allowances for legal depen- dents (pensions for survivors and orphans, etc.). There are also family allowances from the national budget. Pension expenditures have increased substantially over time. Expenditures on public pensions from CARFO almost tripled, rising from CFAF 8.4 billion in 2000 to CFAF 24.4 47 chapter 3. institutional framework and spending review billion in 2016, while the number of beneficiaries dou- TABLE 3.2  Pension expenditure [payouts] and bled (2000 data from World Bank 2012). Private pension beneficiaries, 2011–16 expenditures from CNSS also rapidly increased, dou- Spending (million CFAF) Beneficiaries bling from CFAF 12.0 billion in 2011 to CFAF 28.6 billion Year CARFO CNSS CARFO CNSS in 2016—and thus catching up with spending on pub- 2011 16,124.3 12,019.5 16,645 35,751 lic pensions (table 3.2). In 2015, public civil servant 2012 17,570.8 13,050.0 17,632 38,751 and private pension payments represented 0.36 and 2013 19,118.3 21,725.1 18,609 42,157 0.38 percent of GDP, respectively. 2014 20,661.4 25,602.0 19,552 45,496 2015 22,608.0 24,321.4 20,719 48,212 The general objective of these contributory 2016 24,444.1 28,551.4 21,777 n.a. programs is to minimize the negative impact of SOURCE: See box 3.3. economic shocks on individuals and families— so-called risk pooling. Program beneficiaries receive benefits or services in recognition of contributions based on earnings toward insurance schemes. The contribution for an insured public sector employee is ~5.5 percent of covered earnings, while the employer contribution is ~5.6 percent of covered payroll. If self-employed, the contribution is ~11 percent of declared earnings. The system was extended in 2006 to the self-employed, freelancers, and entre- preneurs through voluntary insurance. The voluntary insurance instituted by law in 2006 (Law 15-2006) made it possible to extend pensions to people practicing a professional activity in sectors not necessarily covered by compulsory insurance. Voluntary insurance is the incentive given to nonsalaried workers to join the CNSS. This gives the worker the right to social security guaranteed by the state. Under voluntary insurance, the self-employed are both employer and employee. The people targeted by voluntary insurance are essentially professionals (contractors, paralegals, transporters, hotel owners, etc.), informal sector actors (craftsmen, masons, welders, restorers, street vendors, etc.), and agro-silvo-pastoralists (cotton farmers, fish farmers, breeders, porridge farmers, fruit and vegetable sellers, etc.). Voluntary insurance also covers any person practicing a professional activity and not affiliated with the CNSS and/or the CARFO. By way of example, a worker who has a monthly salary of CFAF 40,000 will pay, under the old-age insurance, 5.5 percent of his salary or CFAF 2,200; the other 5.5 percent is the responsibility of his employer. A professional in the informal sector can be insured for the old-age pension with a monthly contribution of CFAF 4,400, or 11 percent (about $8) of his income given a monthly salary of CFAF 40,000 (about $80). In 2015, the government adopted the Universal Health Insurance Plan. The objec- tive was to cover, by pooling risks and resources, the cost of health and maternity care for the population. The health insurance scheme should benefit everyone, even the needy. Operation- alization of the plan has been initiated in the form of a pilot program begun in 2017 (box 3.5). 48 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO BOX 3.5  Pilot program for universal health insurance T he objective of the Pilot Program for Universal Health Insurance (Caisse nationale de l’assurance maladie universelle, CNAMU) is to attain universal health insurance for all people in Burkina Faso—including the many informal workers who are not covered by public or private pensions. In 2015 Burkina Faso adopted a universal health coverage scheme law with the intention of enabling all citizens, regardless of social class or income level, to receive quality care. This is a very important initiative and it could potentially turn into a national flagship program. The health fund will consist of worker contributions (which Program name Pilot Program for Universal Health Insurance will vary depending on household capacity); the fund will Category Social insurance, contributory program be complemented by a government subsidy to allow health Responsible agency Ministry of Health insurance coverage for the poorest. This set-up is referred Target group(s) Universal to as “solidarity risk sharing for health expenses.” Feasibility Year 2018 studies are under way to determine the appropriate mix of Location Nationwide contributions and subsidies to ensure universal coverage. No. of beneficiaries — A preliminary study by the NGO ASMADE estimated that Cost CFAF 50 million (estimate) the subsidy needs to be 8 percent of total program costs; SOURCE: CNPS 2017b. however, this seems rather low for regions with a high prev- alence of poverty and vulnerability. The basic health service and medication packages to be offered are yet to be determined. A strategic program plan is being prepared for the period 2017–26. It is critical that implementation mechanisms and capacities be strengthened at the community level to ensure proper coverage and imple- mentation. Willingness to contribute to the fund will depend on its effectiveness and actual impact. The health fund is under the responsibility of the Ministry of Health, while local community-based health funds will help with program implementation. However, recent studies show that the community health funds are still sparse, and face capacity challenges that could jeopardize program implementation. ASMADE is currently working with the Ministry of Health to increase the number of community funds. There is a willingness to provide free health services for the poorest, but this will depend on effective identification mechanisms. A single national registry of beneficiaries is critical for program success. In order to effectively identify poor households, accurately target and track beneficiaries and their program eligibility, further investment is needed in Burkina Faso’s nascent national registry of beneficiaries. This reg- istry would contain socioeconomic information for potential beneficiaries. Expansion of the registry is highly dependent on the capacity and resources of the social workers in charge of collecting household data—which in turn would require investment in the staff capacities of local social centers (which are currently suffering from understaffing and a lack of resources) and community health funds. The actual targeting methods for free health care have not yet been discussed. It should be noted that the government has also introduced measures to allow free medical care—through fee waivers—for some vulnerable categories of the population, namely, pregnant women and children age 0–5. This program, Free Basic Health Care Program for Pregnant Women + Children, is summarized in appendix A. 49 chapter 3. institutional framework and spending review Labor market programs Labor market programs are a relatively recent phenomenon in Burkina Faso. It is thus not possible to track them over time. While public works programs have a long history, measures to activate or provide unemployment support (active versus passive labor market measures) are much scarcer. Burkina Faso recently implemented active labor market mea- sures, including training programs for youth (provided since 2014) and job incentive programs. Recent developments have brought about an increase in labor market programs, expendi- tures on which have risen from accounting for 0.09 percent of GDP in 2014 to 0.17 percent in 2015, as new programs have been implemented. High spending levels do not necessarily equate to effective spending or maxi- mum impact. Numerous factors affect program impact on poverty and inequality, including program coverage, transfer levels, and beneficiary incidence (see chapter 4). Policy makers need to consider the interaction of these factors when designing policies to reduce poverty and inequality. Also, resources should be used in the most cost-effective and pro-poor man- ner possible. For example, subsidies usually consume an unjustifiably high share of public finances, with the bulk going toward rather inefficient fuel subsidies, which have the most unequal impact of subsidies, as the well-off tend to reap most of the benefits. Summary BOX 3.6  Social accountability Despite the significant increase in spending on social protection programs, many challenges remain. Burkina Faso has dedicated funding that has helped many escape extreme poverty, but significant S ocial accountability is an important approach in monitoring programs and evaluating whether they are delivering what they intended. Social accountability involves citizens’ oversight of gaps in coverage and benefit levels remain, and the risk of fragmentation and duplication is high. Numerous government activities, including management of public finances social safety net interventions do not equate to appropri- and service delivery. Tools include social auditing, community score ate coverage. The more programs and actors involved, cards, citizen score cards, and citizen satisfaction surveys. Social the greater the risk of a lack of coordination and inef- audits may go beyond oversight of program finances and procure- ficiency of programs. Much more needs to be done to ment to examine all aspects of a program, including level of access expand coverage to ensure that social safety net pro- to information, accountability, public involvement, program outputs, grams reach more of the nation’s poor and vulnerable and outcomes. In Ghana, for instance, report cards are being used populations, particularly given the frequent occurrence in the education sector to track teacher absenteeism. of natural and human-made crises (box 3.6). SOURCE: Adapted from World Bank 2016b, 106–7. 50 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO Focus on social safety net spending: a multitude of small programs Social safety nets in Burkina Faso are characterized by a multitude of small programs, mainly in-kind transfers. Overall, about 130 social protection programs were implemented in 2015, out of which 109 were social safety nets. Among the safety net pro- grams, 80 percent (86 programs) recorded expenditures lower than 1 percent of total social safety net expenditures. Sixteen social safety net programs account for 85 percent of total social safety net expenditures, while the remaining 15 percent is spread across 93 rather small programs. There is a large number of very small programs, which could indicate poten- tial fragmentation. The 16 largest social safety net programs are listed in table 3.3. In Burkina Faso, food and in-kind transfers have traditionally made up the bulk of social safety net programs. A 2011 review of social safety nets in Burkina Faso found that targeted food distribution and sales, nutrition, and school feeding programs—all food-based programs—accounted for 90 percent of total social safety net spending in 2005 and around 70 percent in 2009 (Cherrier, del Ninno, and Razmara 2011). In 2015, almost TABLE 3.3  Social safety net spending on main 16 programs: 2015 Program type Program description % of total SSN spending In kind School feeding/take home School canteens 13.2 Cash Public works, workfare, and direct job creation Cash for work (improve production) 11.1 In kind Other food/in kind/transfer Sponsorship program 7.3 In kind Nutritional (therapeutic, supplementary, and PLHIV) Malnutrition—prevention 7.0 In kind Nutritional (therapeutic, supplementary, and PLHIV) Malnutrition—severe 6.1 In kind Food distribution Targeted food subsidies 5.4 Cash Scholarship benefits School support (high school) 5.4 In kind Nutritional (therapeutic, supplementary and PLHIV) Malnutrition—moderate 4.8 Cash Scholarship benefits Scholarships (high school) 4.6 In kind Other food/in kind/transfer Subsidies of fertilizers 4.0 In kind Nutritional (therapeutic, supplementary, and PLHIV) HIV care program 3.5 In kind Other food/in kind/transfer Support to informal workers 2.9 In kind Other food/in kind/transfer Distribution of food for cattle 2.6 In kind Other food/in kind/transfer Support for rain production 2.5 In kind Emergency support (including refugees/returning migrants) Food distribution to household victims of shocks 2.5 and natural catastrophes In kind School supplies School supplies 2.3 16 programs 85.3 SOURCE: See box 3.3. NOTE: PLHIV = people living with HIV; SSN = social safety net. 51 chapter 3. institutional framework and spending review three-quarters of total social safety net expenditures was dedicated to implementation of FIGURE 3.5  Social in-kind/food transfers, while just over one-quarter (26.7 percent) went for in-cash social safety safety net expenditure nets (figure 3.5). composition: 2015 The main social safety net program categories are nutrition programs and other in-kind benefits, followed by scholarship and workfare programs. Approximately 26.7% 19 percent of social safety net spending is allocated toward in-kind transfers—mainly agri- in cash cultural input subsidies—and 18 percent is dedicated to nutritional programs, with another 73.3% in kind 13 percent going toward school feeding (figure 3.6). A relatively large share of social safety net spending is dedicated to two in-cash program categories: scholarship benefits and pub- lic works and workfare programs are allocated 12.0 and 12.8 percent, respectively, of total social safety expenditures. In contrast, poverty alleviation programs (cash transfers targeted toward the poor) account for only 1.8 percent of social safety SOURCE: See box 3.3. net expenditures. As table 3.3 shows, the two major social safety net programs by percentage of total social safety net FIGURE 3.6  Social safety net spending by program expenditure are a school canteen program (13.2 percent) and category: 2015 a cash-for-work program to improve production methods (11.1 percent). Scholarship benefits 12.0 Public works, etc. 12.8 In cash Most of Burkina Faso’s social safety net programs Poverty alleviation 1.8 are financed through the National Fund for Social Family and child allowance 0.0 Protection. This fund (Fond National de Protection Sociale, Emergency support 0.0 FNPS) was instituted in 2014. It consists of a combination of Other 19.1 national and external sources, including central and decentral- Nutritional programs 18.1 School feeding 13.3 ized public resources; contributions from the private sector; In kind 8.3 and support from NGOs, associations, and bilateral and Food distribution programs 7.7 multilateral partners. The collected data and program descrip- School supplies 4.0 tions indicate that about half of social safety net expenditure Emergency support 2.5 is nationally financed, with donors involved in 45 percent of Targeted subsidies 0.2 expenditure (figure 3.7). 0 5 10 15 20 Percent Appendix A provides a detailed review of the 23 SOURCE: See box 3.3. main social safety net programs in Burkina Faso. FIGURE 3.7  Social The 23 programs reviewed account for more than 90 percent safety net spending by of total social safety net spending in Burkina Faso in 2015. financing source, 2015 The information presented describes their design, targeting method, evolution of spending, 5.0% – not available and beneficiaries over time where and as available; this is summarized in table 3,4. This compilation provides the basis for determining the relevance of and gaps within the current system, and allows judgments to be made as to whether the programs meet their objectives 26.3% Mixed and—from a higher-level perspective—whether these objectives are aligned with the needs of 49.9% National Burkinabes. Chapter 4 is partially based on the findings of this program review. 18.8% Partners SOURCE: See box 3.3. 52 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO TABLE 3.4  Overview of the 23 main social safety net programs in Burkina Faso Annual Beneficiaries expenditure Funding Program title Implementing agency Target groups Targeting method Geographical area(s) (2015) (2015) sources Cash transfer programs THIMO for Youth Ministry of Youth, Out-of-school Geographical, Cities: Ouagadougou, 2,000 urban CFAF 450 World Bank (component of Youth Training & Vocational youth age categorical, self- Bobo-Dioulasso, million Employment and Skills Integration; 16–35 selection Ouahigouya, Manga, Development Project) implemented by PTR- Kaïn, Tangaye, Barga, HIMO (urban) and Rambo, Béré, Guiba; HELVETAS (rural) rural Cash for Work Program Ministry of Agriculture Vulnerable Geographical, Provinces: Banwa, 226,000 CFAF 16,272 WFP, ECHO, and Water Resources households community Nayala, Sourou, million government Kadiogo, Namentenga, Boulkiemdé, Gnagna, Kourwéo-go, Oubritenga, Oudalan, Noumbiel National Scholarship Ministry of University Categorical, self- Nationwide 6,400 CFAF 6,645 Government, for Public and Private Higher Education; students selection million national University Students implemented by CIOSPB University Student Ministry of 1st year Categorical, self- Nationwide 44,669 CFAF 7,817 National Financial Aid Program Higher Education; university selection million implemented by FONER students Unconditional Cash Technical Alliance Poor and n.a. Regions: Sahel and 2,770 CFAF 91.8 Oxfam, CRS Transfers to Poor in for Development vulnerable Central North households million Sahel and Central North Assistance (ATAD) households Burkin-Naong-Sa Ya MFSNF; implemented Women, Geographical, Regions: North, East, 17,063 947 World Bank by pilot committee households community Central East, Central with children, West, Sahel extremely poor, rural, people with disabilities Unconditional Cash SONAGESS; Poor and Geographical, Soum and Oudalan 27,000 CFAF 1,620 NGOs Transfers (Supplement implemented by NGOs vulnerable community (Sahel region) individuals million to Subsidy/Cereal households (budget) Distribution) In-kind transfer programs Education School Supplies Ministry of National Primary school Geographical 361 communes — CFAF 3,386 National Program (Cartable Education and Literacy; students million Minimum) implemented by NGOs Child Development Compassion Children and Poverty mapping, 11 regions 60,651 CFAF 10.4 Compassion through Sponsorship International youth poverty criteria billion International Program Government School Ministry of National Primary school Geographical Nationwide 2,696,000 CFAF 19,257 Mainly Feeding Program Education and Literacy; students million government, (Primary Education) implemented by CRS WFP, CRS Food Food Assistance for the MASSN, Vulnerable Geographical, Regions: Central, 300,000 CFAF 2,119 Government Vulnerable SP-CONASUR populations community Cascades individuals million via SONAGESS, and partners Targeted Food Subsidies Ministry of Agriculture Vulnerable pop- Geographical, Nationwide 656,137 CFAF 7,915 National and Water Resources, ulations community (target) million SONAGESS 53 chapter 3. institutional framework and spending review Annual Beneficiaries expenditure Funding Program title Implementing agency Target groups Targeting method Geographical area(s) (2015) (2015) sources Food Shops Ministry of Agriculture n.a. n.a. Varies, but mainly 150 stores CFAF 1,585 Ministry of (Boutiques témoins) and Water Resources, urban areas million Industry, SONAGESS Trade and Handicrafts; JICA; Oxfam Food Assistance to MASSN, Victims of Not clear, possibly Primarily the Sahel 405,429 CFAF 3,655 Nationally Vulnerable and Disaster SP-CONASUR natural hazards, HEA individuals; million financed, Affected Households people living 57,918 Ministry of on subsistence households Economy agriculture and Finance Nutrition and health fee waivers Malnutrition Prevention Ministry of Health, WFP, Children age Categorical, — 284,644 10,247 WFP, Program partners 6–23 months geographical (budget) European Commission Treatment of Severe Ministry of Health Children < 5 Categorical, All regions 149,510 8,970 European Acute Malnutrition suffering from geographical (budget) Commission, severe acute WFP, UNICEF malnutrition Moderately Acute Ministry of Health, WFP Children < 5 Categorical, 350,082 7,001 European Malnutrition Treatment suffering from geographical; Commission, moderately female-headed WFP, UNICEF acute households malnutrition; prioritized where pregnant/ possible lactating women ART (Antiretroviral Ministry of Health Persons with Categorical National 72,000 CFAF 5,092 National; Therapy) Treatment HIV million UNFPA Patient Program Free Basic Health Care Ministry of Health Pregnant Categorical National 3,337,755 CFAF 1,294 Ministry of Program for Pregnant women; demand driven (2016) million Health Women + Children children age 0–5 Agriculture Fertilizer Subsidy Ministry of Agriculture Vulnerable Community Nationwide 12,511 tons CFAF 5,878 Ministry of Program and Water Resources farming house- of fertilizer million Economy holds and Finance Seed Subsidy Program Ministry of Agriculture Vulnerable Community Rural areas, national 5,592 tons of CFAF 2,550 Ministry of and Water Resources populations, seeds million Economy female-headed and Finance households Rain-Fed Agriculture Ministry of Agriculture Vulnerable HEA and SAP Rural areas 174,093 CFAF 3,681 — Support Program and Water Resources farming (budget) households, female-headed households Livestock Feed Support Ministry of Animal and Vulnerable SAP; HEA Sahel, North, Central 76,000 tons CFAF 3,820 Ministry of Program Fisheries Resources; farming house- North, East; no charge of livestock million Economy SONAGESS holds for Soum Province feed and Finance; partners NOTE: CRS = Catholic Relief Services; DGPV = Directorate General of Crop Production; FONER = National Fund for Education and Research; JICA = Japan International Cooperation Agency; MASSN = Ministry of Social Action and National Solidarity (former name of MFSNF); MFSNF = Ministry of Women, National Solidarity and Family; PTR-HIMO = Labor-Intensive Roadwork Project; SAP = Early Warning System (Système d’Alerte Précoce); SONAGESS = National Society for the Management of Food Security; SP-CONASUR = Permanent Secretariat of the National Council for Emergency Relief and Rehabilitation; UNFPA = United Nations Population Fund; WFP = World Food Programme. chapter 4 social safety nets assessment: do the programs meet the needs? This chapter assesses the adequacy of Burkina Faso’s social safety net system and programs by examining the various programs’ coverage, targeting effectiveness, and cost-effectiveness. It also looks at some policy reform simulations, sustainability, and monitoring and evaluation practices. 56 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO Coverage, gaps, and overlaps Two methods are used to estimate how much of total social safety net expendi- tures actually reach the poor. Administrative data were collected as part of the present program review (see appendix A), which presents the main design features and expenditures of Burkina Faso’s 23 largest social safety net programs. The program review sheds light on the extent of coverage, as well as the targeting effectiveness, of some of the major programs. Administrative data on coverage of social protection programs are unfortunately limited in Burkina Faso. However, available coverage data are complemented by figures from the Overall, the ones that need household survey (Enquete Multisectorielle Continue, EMC). EMC 2014 provides an important them the most—the poorest baseline despite its limitations. Program adequacy is analyzed by looking at beneficiaries. quintile—benefit the least from social safety net Social insurance coverage While the focus here is on social safety nets—given their direct link to poverty programs, compared to all reduction objectives—the country’s significant level of expenditure on social other income groups. insurance (contributory social protection) deserves attention as well. The number of elderly (those above age 55) in Burkina Faso is estimated at more than 1.2 million; how- ever, administrative data report that only 69,000 individuals receive an old-age pension. While public and private pension expenditures have increased, the number of beneficiaries remains limited. Contributory programs remain out of reach for most Burkinabes. Limited access to contributory programs—including old-age, survivor, and disability pensions—is corroborated by survey data, with only 1.7 percent of the total population covered (directly or indirectly through a household member) by at least one contributory program according to EMC 2014. Access to social insurance is greatest in urban areas and among individuals in the richest quintile of per capita consumption (figure 4.1). Virtually no one from the poorest quintile has access (EMC 2014), compared to about 6 percent of those in the richest FIGURE 4.1  Access to social insurance by percentage of population age 55 and over a. By quintile b. By type of insurance Percent Survivor pensions 6 4 2 Old-age pensions 0 1 2 3 0 Total Q1 Q2 Q3 Q4 Q5 Percent SOURCE: Authors’ calculations based on EMC 2014. NOTE: Coverage figures include both direct and indirect beneficiaries. 57 chapter 4. social safety nets assessment: do the programs meet the needs? quintile. Given the high level of informal agricultural work in rural areas, social insurance coverage is highest in urban areas; particularly in the Central region (7.3 percent of the total population covered), followed by Hauts-Bassins (3.5 percent). Looking at those age 55 and above, only 2.8 percent receive (directly or indirectly through a household member) an old-age pension; 0.4 percent receive a survivor pension (EMC 2014). Social safety net coverage by income In Burkina Faso, 40.1 percent of the population lives in poverty, yet social safety net coverage is alarmingly low. Only 2.9 percent of the entire population benefits from safety nets, which is well below the identified needs. Safety net coverage is defined as the share of the population receiving at least one benefit. According to EMC 2014 data, only 2.9 percent of the population benefits from safety nets such as scholarships, government support in case of shocks, NGO support, food distribution, and other transfers. (Coverage of other types of safety nets is analyzed using administrative data, as these are not captured in survey data; see box 4.1.) BOX 4.1  EMC survey data limitations D ata on coverage and performance of social protection programs are limited in Burkina Faso, althought the EMC provides an important baseline. Social protection programs are highly fragmented and often small in scale and coverage. It is not surprising that household survey data do not capture many beneficiaries of such programs. If coverage is low among the general population, there would be few obser- vations in a nationally representative survey, hence limiting the analysis possible with the data. Another shortcoming concerns in-kind transfers, as estimating the value of in-kind transfers may be problematic. In some cases, the amount of the benefit is not available, and the survey includes only a participatory dummy, limiting possible analysis only to coverage. The main cash transfers or benefits identified in the EMC 2014 survey No. of observations are old-age pensions, disability pensions, survivor benefits, scholar- Old-age pensions 152 ships, transfers from the government in case of shock, transfers from Survivor benefits 31 an NGO or religious organization in case of shock, and other transfers. Disability pensions 1 There are also 172 individual-level observations from people reporting Scholarships 52 living mainly on free food distribution (this only captures individuals Government support (in case of shocks) 346 Other transfers 25 who report living mainly from food distribution and can then exclude Support from NGOs 189 those who receive food distribution but estimate living mainly from Food distribution 172 other sources, such as income or own production). Additionally, food SOURCE: EMC 2014. distribution is only included as a participatory dummy, as no estimates/ transfer amounts are available. The small number of observations should be kept in mind when considering the results presented, especially for specific benefit types. The analysis should be taken as indicative and reflective only of the programs mentioned. Overall, 10,411 households were surveyed, representing 77,037 Individuals. 58 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO Each social safety net program has a different redistributive pattern when look- ing at coverage by quintile, but a common characteristic among them is a lack of focus on the poorest quintile. A positive signal, given the high prevalence of poverty in Burkina Faso, is that overall social safety net coverage is higher for the second quintile (4 percent) than for the two richest ones (3.1 and 3 percent, respectively), although coverage of the poorest quintile lags behind for each program (figure 4.2). Scholarship programs are mainly distributed to the two top richest quintiles, with no impact on poverty alleviation. Less than 1 percent of the total population ben- efits from scholarships. Simulations based on EMC 2014 indicate that, in absence of this program, poverty would not even increase—meaning that the program has virtually no effect on poverty. Government support in case of shocks indicates less leakage to nonpoor house- holds, even though the second richest quintile benefits almost as much as the poorest. In the richest quintile (Q5), 0.2 percent benefit from these programs; and in the second richest (Q4), 0.8 percent benefit. In contrast, in the poorest quintile (Q1), only 0.9 per- cent are covered by this type of safety net. Other social safety net transfers and support from NGOs have the lowest overall coverage, with 0.4 and 0.5 percent, respectively; and the second poorest quintile (Q2) benefits the most from these. A startlingly low 1.4 percent of the very poorest receive a social safety net transfer. FIGURE 4.2  Social safety net coverage by program and quintile, EMC 2014 Percent 4 Total Q1 Q2 Q3 Q4 Q5 3 2 1 0 All social Scholarships Government support Other transfers Support from Food distribution assistance (in case of shocks) NGOs SOURCE: Authors’ calculations based on EMC 2014. Social safety net coverage by age group Coverage is highest for young adults and relatively low for the elderly and young children. While the groups that face the highest headcount ratios are children (44 percent of whom are poor), few of them receive a social safety net transfer. Only 18.8 and 29.6 percent of beneficiaries are children age 0–5 and 6–17, respectively. Approximately 31 percent of ben- eficiaries are young adults, only 12 percent are middle-aged (35–59), and 9 percent are elderly 59 chapter 4. social safety nets assessment: do the programs meet the needs? (60 and up; figure 4.3). The low coverage of the groups that FIGURE 4.3  Social safety net coverage by age (percent are most at risk of poverty calls for a more detailed review of beneficiaries) and headcount ratio of social safety nets by age group, using administrative data. Percent Social safety net coverage Poverty rate 45 Children age 0–5 30 Children age 0–5 account for 21 percent of the country’s population, with 1.7 million of them liv- 15 ing in poverty. Despite this only 18.8 percent of children are covered by social safety nets (based on EMC 2014) 0 age 0–5 age 6–17 age 18–34 age 35–59 age 60+ (table 4.1). Of the 23 main social safety net programs, only 5 focus on this group: the national nutrition programs for prevention and treatment of moderate and severe malnu- SOURCE: Authors’ calculations based on EMC 2014. trition, the free basic health care program for pregnant women and children age 0–5, and treatment of HIV-in- fected mothers and children. These programs have TABLE 4.1  Coverage of main social safety net programs focusing on children age 0–5, 2015 relatively low beneficiary targets (the actual number of beneficiaries is not available for the nutrition programs) Program description Number Percenta or level of beneficiaries. The largest nutrition program Prevention of malnutrition b 284,644 7.41 (prevention) aimed to reach 284,644 children in 2015—a Treatment of moderate malnutrition b 70,503 1.84 figure well below the actual number of poor children. Treatment of severe malnutritionb 209,590 5.46 The largest programs (the prevention of malnutrition and Treatment of HIV children and mother 5,092 0.13 the treatment of severe malnutrition) planned to cover Free health care for mother and children 24,313 0.63 only 7.41 and 5.46 percent of children, respectively. Children 0–5 3,840,678 Poor children are also covered indirectly by the Bur- Poor children 1,695,322 kin-Naong-Sa Ya program. SOURCE: Appendix A. NOTE: Poor children are also covered indirectly by the conditional cash transfer Even though coverage remains low, disaggregation of program Burkin-Naong-Sa Ya, which targets poor households with children (see adult coverage below). beneficiaries by region, whenever possible, offers more a. Coverage is estimated by dividing the number of beneficiaries of each pro- insight. Nutrition programs are better aligned to poverty, gram by the total population of children. as the majority of beneficiaries are in the poorest regions b. Numbers are planned beneficiaries from January to June. (Boucle du Mouhoun, East and North). While overall coverage is low, the coverage of poor children might be higher, improving targeting effectiveness. Besides fee waivers and free health care programs, alternative options to improve children’s health status exist. Conditional cash transfers for health utilization (visits to health clinics for routine preventive care) for children age 0–5 have proven to be effective in increasing the numbers of visits made in Burkina Faso. The evaluation of a two- year pilot cash transfer program (the Nahouri Cash Transfers Pilot Project, running from 2008 to 2010) found that conditional cash transfers led to an increase of almost 50 percent in health care visits over both nonbeneficiaries and beneficiaries receiving unconditional cash transfers (Akresh, de Walque, and Kazianga 2012). Another impact evaluation (Akresh, de Walque, and Kazianga 2016) found that conditional cash transfers aimed at educational and health care behavior (school enrollment and preventive health checkups for young children) significantly 60 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO improved children’s education and health and household socioeconomic conditions. The eval- uation also found that conditional cash transfers outperformed unconditional. Burkina Faso is one of the worst-performing countries worldwide with regard to early childhood development (ECD) indicators (figure 4.4). ECD programs pro- vide children with early cognitive and language skills and are recognized as one of the most cost-effective investments in building human capital and promoting sustainable development (UNICEF 2014). ECD program attendance is practically 0 percent when looking at children in the lowest quintile—despite this being the group most in need of such programs. Attendance Access to ECD programs rates increase to 9 percent for the upper quintile. in Burkina Faso is the Besides formal programs, ECD can also take place within the family or house- lowest in the world after hold environment. However, Burkinabe children age 3–5 record some of the lowest levels Afghanistan, with only of support worldwide for learning within the home environment. For example, only 14 percent of children had interacted with an adult in the last three days in four or more activities that 2 percent of all children promote learning and school readiness. The child-adult learning interaction is much higher benefiting from ECD in some of the other very poor countries in the region (UNICEF 2014), including the Central programs. African Republic (74 percent), the Democratic Republic of the Congo (59 percent), Burundi (34 percent), and Mali (29 percent). Violent discipline is widespread in Burkina Faso. Almost 80 percent of children age 2–4 experience psychological aggression and/or physical punishment over the course of a month (UNICEF 2014). A growing body of research indicates that children who have expe- rienced physical punishment tend to exhibit more aggressive and antisocial behavior, and hence also developmental and educational difficulties (Durrant and Ensom 2012). FIGURE 4.4  Percentage of children 36–59 months old attending an ECD program in selected countries Percent 100 80 60 40 20 0 Korea, Dem. People’s Rep. Jamaica Barbados Belarus St. Lucia Thailand Uruguay Cuba Trinidad & Tobago Croatia Vietnam Moldova Brazil Ghana Mongolia Georgia Venezuela, RB Argentina Lebanon Guyana Ukraine Nepal Serbia Egypt, Arab Rep. Tunisia Nigeria Qatar Montenegro Albania Dominican Republic Malawi Morocco Kazakhstan Panama South Africa Colombia Congo São Tomé & Príncipe Suriname Belize Swaziland Oman Cameroon West Bank & Gaza Togo El Salvador Turkmenistan Lao PDR Myanmar Kyrgyzstan Sudan Macedonia, FYR Jordan Senegal Zimbabwe Iran, Islamic Rep. Uzbekistan Honduras Gambia, The Botswana Costa Rica Indonesia Algeria Cambodia Sierra Leone Mauritania Djibouti Bangladesh Bosnia & Herzegovina Guinea-Bissau Benin Mali Bhutan Syrian Arab Republic Congo, Dem. Rep. Tajikistan South Sudan Central African Republic Burundi Chad Iraq Yemen, Rep. Somalla Burkina Faso Afghanistan Côte d’Ivoire SOURCE: UNICEF, data.unicef.org. 61 chapter 4. social safety nets assessment: do the programs meet the needs? There is a clear gap in social safety nets focusing on children age 0–5, and there is scope to better link cash transfers with ECD activities. While there are few programs focusing on young children or ECD activities, there is room to improve the design of cash transfer programs and include an ECD component. Such initiatives have begun, nota- bly with the cash transfer program Burkin-Naong-Sa Ya, as beneficiaries also benefit from support measures to increase nutrition and ECD knowledge. Also, the THIMO program offers beneficiaries in the Manga and Bobo areas free nursery services, which include educational stimulation and health content. Children age 6–17 School-age children represent the largest share TABLE 4.2  Coverage of main social safety net of the population in Burkina Faso, making up programs focusing on children age 6–17, 2015 33 percent of the population and comprising 3.4 Program description Number Percenta million individuals. Given the country’s high poverty School feeding program 2,696,000 44.81 rate, there are about 2.7 million poor school-age children Sponsorship program 60,651 2.28 in Burkina Faso. Only three social safety net programs School supplies program n.a. were found in the review to be focused on school-age Children 6–17 6,015,918 children (table 4.2): a school feeding program, a school Poor children 2,658,950 supplies program, and an NGO sponsorship program. SOURCE: Appendix A. According to the CNPS (2016b), the school feeding NOTE: Poor children are also covered indirectly by the conditional cash transfer program provides meals to primary school students program Burkin-Naong-Sa Ya, which target poor households with children (see adult coverage below). and covers 2.7 million children; this corresponds to a a. Coverage is estimated by dividing the number of beneficiaries of each pro- high coverage of all children enrolled in primary school gram by the total number of children. (2.9 million). Because it covers virtually all primary school students, it also covers the nonpoor. Information on coverage of the school supply program is unavailable. The NGO sponsorship program covered about 60,651 school-age children, or 2.28 percent. The school feeding program is one of the largest programs in Burkina Faso. It accounts for 13 percent of total social safety net spending, and has proven to increase school attendance and reduce child labor. Impact evaluations of the school feeding program in the Sahel region (Kazianga, de Walque, and Alderman 2012) show that the program increased the enrollment rate by about 4 percentage points (4.7 for boys and 3.5 for girls) and decreased child labor—especially on-farm labor, which is more incompatible with school hours. The pro- gram also increased learning outcomes in basic mathematical computations compared to the control group. Other types of food-for-education programs can yield spillover effects on the entire family. Kazianga, de Walque, and Alderman (2012) show that in a similar setting (Sahel region), a take-home ration program generated at least a similar effect on enroll- ment and attendance. Kazianga, de Walque, and Alderman (2014) find that the take-home rations bring additional benefit to other household members compared to school feeding. Intra-household redistribution of rations improves the nutritional status of younger (below age 5) siblings. The authors also find that achieving similar results using cash transfers would be much costlier than the take-home rations. Take-home rations are provided in five provinces 62 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO (see appendix A) by the Catholic Relief Services. This initiative could be extended in other regions and reinforce the impact of the school feeding program—which could decrease exclu- sion errors, as take-home rations would create an additional incentive for school enrollment of poor children. Social safety net interventions focusing on secondary school children are lack- ing. Although barely 35 percent of those age 12–18 are in school (EMC 2014 data), there are no specific interventions in place aimed at discouraging dropping out of school. Scholarship programs cover only university students. There are labor market interventions that focus on out-of-school youth, age 16–25, notably the Youth Employment and Skills Development Project (Projet d’emploi des jeunes et de développement des compétences, PEJDC), and its vocational training component. However, program coverage is relatively low, and its age con- ditions are not likely to prevent early high school dropout. School enrollment and retention could be improved by other types of social safety nets than those currently in place, particularly conditional cash transfer programs. While both conditional and unconditional cash transfers relax household budget constraints and allow parents to invest more in human capital, an impact evaluation of cash transfers in Burkina Faso showed that conditional cash transfers have greater impact and are most cost-effective in improving the enrollment and education outcomes of children who are less likely to go to school or who are not already enrolled (Akresh, de Walque, and Kazianga 2013). This option could be considered to reinforce the impact of school feeding on primary school enrollment (focusing on marginal children) as well as on secondary school students. Adults Several safety net programs address the need to create more secure jobs. Burki- nabes face unstable jobs and, given the population growth, the labor market is under high pressure. Programs that aim to address this include the social safety net cash for work pro- gram, the THIMO for Youth program (component of PEJDC), as well as labor market programs that support the employment of graduates (part of PA/ISJD), the job creation project and the training program for young entrepreneurship (component of PEJDC). The cash-for-work pro- gram is one of the largest cash transfer programs in Burkina Faso; it reached about 226,000 beneficiaries in 2015. Although the public works programs are important, more opportunities should be created in rural areas. This is one of the purposes of THIMO, which is increasingly focusing on rural areas. While some social safety nets focus on encouraging the transition from the agricultural sector, most households live on subsistence agriculture. Subsistence agriculture can be greatly affected by natural hazards, thus making it necessary to secure production by providing adequate social safety net programs. More than 65 percent of adults work in agricultural activities, and most face job instability. Sixty percent work fewer than six months during the year, and 35 percent have a second job. Farming income remains the main component of their income, especially for those in the bottom quintile (representing about 75 percent of total income)—while almost 80 percent of those living from agriculture report being affected by at least one natural hazard a year. Four social safety net programs focus on 63 chapter 4. social safety nets assessment: do the programs meet the needs? securing farming income by providing subsidies for agricultural input: subsidy for seeds, for fertilizer, support for rain-fed production, and endowment of livestock feed. These programs focus on vulnerable population, and the quantity of subsidies/number of beneficiaries varies yearly based on the Early Warning System (Système d’Alerte Précoce, SAP) and HEA identifi- cation methods. The actual number of beneficiaries is unknown, however. Household income is low and often precarious, and people often end up in a poverty trap. The government and its partners have been paying more attention to provid- ing regular cash transfer programs to the most vulnerable as a way to increase their financial security and escape poverty. Unconditional cash transfer activities have been financed by the World Bank, Catholic Relief Services, and Oxfam. The World Bank program (Burkin-Na- ong-Sa Ya) has primarily focused on the country’s poorest regions (North and East), but was expanded to the Central West and Sahel regions in 2017. According to 2015 estimates, roughly 40,000 households received unconditional cash transfers, which is still well below the number of severe poor. In 2018, the Burkin-Naong-Sa Ya program expanded to an additional region, reaching 442,416 individuals overall, which represents a significant increase in cover- age of the poor, reaching almost 18 percent of them (table 4.3). The potential of conditional cash transfers related to health and education has been proven effective in Burkina Faso. The 2008–10 Nahouri Cash Transfers Pilot Project showed that conditional cash transfers improve overall household socioeconomic conditions, especially the education and health outcomes of children to which the conditions are tied (Akresh, de Walque, and Kazianga 2012, 2016). While an impact evaluation of the Burkin-Naong-Sa Ya program is not yet available, it is expected that adding hard conditions, similar to those of the Nahouri project (school attendance and utilization of preventive care for young children) could reinforce its impact, given the similarities in context. This would address directly some key human capital deficiencies highlighted in figure 2.17 pertaining to these ages, as well as address chronic poverty in adulthood. For the time being, the program has a soft condition favoring a range of human capital–building behaviors. Adding hard conditions TABLE 4.3  Coverage of main social safety net programs focusing on adults, 2015 Coverage of Coverage of poor Program name/description Target group Region individuals (%) adults (%)a Burkin-Naong-Sa Ya Poor/vulnerable house- North, East, Central East, 442,416 17.86 holds with children Central West, Sahel (2018) Unconditional cash transfer to poor in Sahel Poor/vulnerable Sahel, Central North 19,390b 0.78 and Central North regions Unconditional cash distribution operations Poor/vulnerable Sahel 189,000 7.62 (supplement to cereal distribution, in particular (target)b in Soum and Oudalan) Number of adults 18–34 4,282,934 Number of adults 35–59 2,899,087 Number of poor adults (18–59) 2,477,815 SOURCE: Appendix A. a. Coverage is estimated by dividing the number of beneficiaries of each program by the total population of poor adults age 18–59. b. Coverage is estimated from household coverage multiplied by an average household size of 7. 64 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO may be more difficult in regions that suffer from a lack of appropriate school and health infra- structure: the supply needs to be able to satisfy demand. The PEJDC addresses gender issues under the THIMO and the training program for young entrepreneurship. Unlike other labor market and safety net programs (except those targeting mothers), the PEJDC explicitly sets a target level for female beneficiaries. The elderly There are no social safety net programs that specifically focus on the elderly. As noted earlier, the elderly face a greater risk of falling into poverty then the rest of the popula- tion (41.2 percent versus 40.1 percent) and are more likely to face health issues, which also affects their ability to have a stable source of income. As most of the elderly live with their children and grandchildren, they indirectly benefit from other social safety nets, such as those focusing on vulnerability in case of shocks. The vulnerabilities of the elderly are acknowledged in a recent strategic plan outlined by the Ministry of Health dedicated to those age 60 and above, but no specific program has been put in place yet. Across the life cycle Vulnerability with respect to shocks is widespread across the whole life cycle. Burkina Faso is highly affected by shocks—especially natural hazards—with about 70 per- cent of the population exposed to at least one shock during the year. Almost 80 percent of the poor, who are predominantly living on subsistence agriculture, are affected by shock on a yearly basis—leading to adverse coping strategies and therefore diminishing the opportunity to escape poverty through asset accumulation. To mitigate the impact of shocks, several food distribution and subsidy (sale of food products at a subsidized price) programs have been implemented, mainly in regions at high risk of malnutrition as identified by the SAP (box 4.2). The programs typically run from May–August each year, and cover a relatively large number of individuals (compared to other social safety net programs). While food assistance is one of the oldest types of social safety net interventions in the country (having been in existence since the mid-1990s), the interventions were reinforced in 2014 with the introduction of social mea- sures; hence the recent increase in beneficiaries. In addition to food programs in vulnerable areas, these social measures included a new food subsidy program, Food Shops. Available mostly in urban areas, it is therefore not a response to natural hazard shocks, but rather to social tensions (table 4.4). ECD and literacy, and the elderly’s lack of income, stand out as risks that have not been addressed. There are clear gaps in terms of risk coverage along the life cycle of Burkinabes (table 4.5). Young children lack ECD programs; school-age children lack school retention/incentives programs, and have no program that allows them to improve literacy (in or outside of school; there are no programs for the elderly. Some other risks along the life cycle are addressed by social safety nets, but most likely not sufficiently given the scale of the program with respect to the number of individuals covered—that is, risks linked to high levels of labor market pressure, unstable jobs, and low incomes. While the risk of malnutrition and child mortality are addressed, information on nutrition programs is lacking and no conclusions 65 chapter 4. social safety nets assessment: do the programs meet the needs? BOX 4.2  Targeting methods in Burkina Faso Geographic targeting. Geographic targeting uses location to determine eligibility for benefits. That is, people who live in designated areas (particularly areas with high levels of poverty, food insecurity, malnutrition, or exposure to natural disasters) are identified as eligible, while those who live elsewhere are not. In Burkina Faso, geographical targeting is common and is based on the Early Warning System (Système d’Alerte Précoce, SAP), which identifies each year the zone with the highest risk of food insecurity. Categorical. Categorical methods include targeting individuals or households that belong to an easily identifiable and specific social or demographic group. Categorical targeting typically involves defining eligibility in terms of characteristics that are fairly easy to observe and dif- ficult to manipulate, such as age, sex, ethnicity, disability status, or land ownership. Age is a commonly used category in cash child allowances, school feeding programs, and social pensions. Poverty targeting methods that approximate a potential beneficiary’s poverty, welfare, or vulnerability status. These include various methods, which are often combined, such as the following: „„ Community targeting. Community-based targeting uses a group of community members or leaders, whose principal functions in the community are not related to the social protection program, to determine who in the community is eligible for program assistance. „„ Proxy means testing (PMT). PMT uses statistical inference to define the eligibility of a household. It approximates (proxies) household welfare by computing a score based on easy-to-observe and relevant characteristics to measure and predict poverty such as socioeconomic characteristics; household demographic structure; the location, quality, and sanitation of dwellings; ownership of assets and durable goods, which are highly correlated with household welfare levels. In Burkina Faso, several cash transfer programs use PMT target- ing methods, whose design (i.e., parameters included in the test) may differ. The design of the PMT for the Burkin-Naong-Sa Ya program is described in appendix C. „„ Household economic approach (HEA). Under this approach, households are targeted according to their risk of facing malnu- trition and their ability to face shocks. The HEA methods entails a yearly tracking of indicators such as crop production, rainfall prediction, as well as information on household characteristics and coping strategies. Self-targeting. Self-targeted programs are technically open to everyone, although they are designed in such a way that only those with a very high level of need will enroll. For example, in public works, wages are set low so that the program is only attractive to those who are will- ing to work at a low wage. SOURCE: Adapted from Beegle, Coudouel, and Monsalve 2018; del Ninno and Bradford 2015; and Schnitzer 2016. TABLE 4.4  Food assistance programs Program name Region Beneficiaries Food Assistance to Vulnerable Populations Mainly Central and Cascades 300,000 Targeted Food Subsidies Mainly Central North and Sahel 656,137 (target) Food Shops Urban areas n.a. Food Assistance to Vulnerable Populations and Disaster-Affected Households Mainly Sahel, Soum, and Oudalan 405,429 SOURCE: Appendix A. 66 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO TABLE 4.5  Alignment of safety net programs with risks and vulnerabilities across the life cycle Age group Risks Status Programs addressing the risks Malnutrition Partially addressed 3 nutrition programs (no information on actual number of beneficiaries) Young children Mortality Partially addressed Health fee waivers for mother/children (relatively low coverage) (age 0–5) Early childhood development Unaddressed School dropout (primary) Addressed School feeding, school supplies, and sponsorship programs School-age children School dropout (secondary) Partially addressed Reintegration of out-of-school girls (age 6–17) Literacy Unaddressed High subsistence agriculture Addressed Subsidy for seeds, for fertilizer, support to rained production and endowment of livestock feed (coverage unknown) Labor market pressure and Partially addressed Cash-for-work program, THIMO for Youth (component of PEJDC), as Adults unstable job well as labor market programs supporting installation of graduate (age 18–59) youth (part of PA/ISJD), job creation project, training program for young entrepreneurship (component of PEJDC) Low and precarious income Partially addressed 3 unconditional cash transfer programs No income source Unaddressed Elderly (age 60+) Health issues Partially addressed Sporadic free health services Across the life cycle High exposure to shocks Addressed 4 food distribution and subsidy programs NOTE: PEJDC = Youth Employment and Skills Development Project. can be drawn about the target level of beneficiaries. This indicates that efforts to improve reporting should be reinforced. There are also potential overlaps in the area of food distribu- tion and subsidies, as well as with unconditional cash transfer programs, and no information on their coordination mechanisms is available. As in many other countries in the region, social safety net coverage in Burkina Faso is inadequate and does not meet the needs of the people living in poverty. Only a small percentage of the poor in Burkina Faso are covered; this is not uncommon, as most African countries are characterized by limited coverage. However, countries such as Botswana, Namibia, and South Africa have managed to ensure that all the poor receive safety net benefits of some kind, given that their safety net coverage is higher than the poverty rate. There are improvements that can be made. For example, the targeting methods of some of the programs are potentially inefficient, not taking the welfare characteristics of the Burkinabes into account; this is detailed in the next section. Social safety net coverage by region Social safety net coverage is not aligned with the prevalence of poverty across regions, with the richest Central region benefiting the most. According to EMC 2014, the Central region has the largest coverage rate (figure 4.5); this is mainly due to schol- arship programs. The programs target postsecondary school students, and most higher education institutions are based in the capital, Ouagadougou. The Central North region has the second highest coverage at 5.9 percent, while the poverty rate is 46.8 percent. 67 chapter 4. social safety nets assessment: do the programs meet the needs? FIGURE 4.5  Social safety net coverage and poverty headcount ratio, by region a. Social safety net coverage b. Poverty headcount ratio 4°E 2°E 0° 2°E 4°E 2°E 0° 2°E IBRD 44748 | JANUARY 2019 IBRD 44745 | JANUARY 2019 This map was produced by the Cartography Unit of the World This map was produced by the Cartography Unit of the World MALI Bank Group. The boundaries, colors, denominations and any MALI Bank Group. The boundaries, colors, denominations and any other information shown on this map do not imply, on the part of other information shown on this map do not imply, on the part of the World Bank Group, any judgment on the legal status of any the World Bank Group, any judgment on the legal status of any territory, or any endorsement or acceptance of such boundaries. territory, or any endorsement or acceptance of such boundaries. SAHEL SAHEL 14°N NIGER 14°N 14°N NIGER 14°N NORTH CENTRAL NORTH NORD CENTRAL NORTH BOUCIE DU BOUCIE DU MOUHOUN PLATEAU MOUHOUN CENTRAL PLATEAU EAST CENTRAL CENTRAL CENTRAL EAST 12°N  12°N 12°N  12°N CENTRAL WEST CENTRAL WEST CENTRAL CENTRAL EAST SOUTH CENTRAL EAST CENTRAL HAUTS-BASSINS SOUTH HAUTS-BASSINS BENIN BENIN SOUTH TOGO TOGO SOUTH CASCADES WEST 0° 2°E CASCADES 0° 2°E WEST 10°N  GHANA BURKINA FASO 10°N  GHANA BURKINA FASO SOCIAL SAFETY NET COVERAGE POVERTY HEADCOUNT RATIO 0 20 40 60 80 100 Kilometers 0 20 40 60 80 100 Kilometers CÔTE D'IVOIRE 0 20 40 60 80 Miles CÔTE D'IVOIRE 20 0.28 6.62 0 40 60 80 Miles 9.68 70.3 4°E 2°E 4°E 2°E SOURCE: Authors’ calculations based on EMC 2014. The regions with the highest prevalence of poor populations have the lowest levels of safety net coverage. Thus, the North, Boucle du Mouhoun, and the East—with, respectively, 70.3, 59.5, and 51.6 percent poor populations—have limited coverage, especially Boucle du Mouhoun, where only 0.8 percent of the population receives a transfer. In general, coverage of social safety nets is not in line with poverty risk (see appendix D for detailed per- centages). Regional coverage of in-cash social safety nets is mainly led by government support in the Central North, Central East, and Sahel regions. In the Central region, high coverage is driven by the scholarship program, as noted above. Food distribution pro- gramming is mainly present in Boucle du Mouhoun, with 3.9 percent of individuals reporting living mainly from free foods. Coverage by region is directly linked to programs’ targeting accuracy (most benefits reach the poorest quintile), this varies according to the different methods used to reach the poor. Programs such as a scholarship initiative, which has a low targeting accuracy given its focus on university students who are mainly in the highest welfare quintile and living in cities with a postsecondary education institution, have low coverage in the poorest and rural regions (appendix D). Social safety net coverage in Burkina Faso remains limited, inadequate to the needs as reflected by the poverty rate, and one of the lowest rates in the region. Social safety net coverage in African countries is mostly characterized by signifi- cant undercoverage of the poor. However, countries with a similar—or even a lower—gross domestic product (GDP) per capita have a higher level of social protection (including of social insurance) coverage than does Burkina Faso at 4.3 percent: for example, Malawi (41.7 percent in 2013), Niger (20.1 percent in 2014), Ethiopia (13.2 percent in 2010), and the Democratic Republic of Congo (10 percent in 2012) (figure 4.6). The coverage gap in Burkina Faso is 68 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO FIGURE 4.6  Social protection coverage in African countries Percent 80 70 60 50 40 30 20 10 0 Mali 2009 Zambia 2010 Comoros 2004 Bhutan 2012 Chad 2011 Guinea 2012 Burkina Faso 2014 Cameroon 2014 Madagascar 2010 South Sudan 2009 Lesotho 2010 Nigeria 2015 Benin 2003 Mozambique 2008 Liberia 2014 Gambia 2010 Congo Republic 2005 Congo, Dem. Rep. of 2012 Tanzania 2014 Ethiopia 2010 Namibia 2009 Senegal 2011 Djibouti 2012 Niger 2014 Kenya 2005 Zimbabwe 2011 Cape Verde 2007 Sierra Leone 2011 Rwanda 2013 Côte d’Ivoire 2015 Timor Leste 2011 Malawi 2013 Mauritania 2014 Mauritius 2012 Swaziland 2009 Gambon 2005 South Africa 2010 Ghana 2012 Botswana 2009 SOURCE: ASPIRE (Atlas of Social Protection Indicators of Resilience and Equity) database. NOTE: The data here are for all forms of social protection, including pensions, which is generally relatively small compared to social safety nets. therefore important, as even assuming that social safety nets are targeting the poor and not leaking to the nonpoor, only a small percentage of the poor are potentially covered. The targeting methods of some programs are potentially inefficient, not taking the welfare charac- teristics of the Burkinabes into account, as detailed in the next section. Targeting accuracy and cost- effectiveness Targeting Extensive social safety net coverage does not guarantee a high impact on pov- erty reduction if transfers do not properly target those in need. In a context of low coverage as in Burkina Faso, it is crucial that the resources allocated to the social safety net reach those in need—the poor and vulnerable—and that inclusion error is minimized. The distribution of benefits is directly linked to the methods used to select beneficiaries—if any. 69 chapter 4. social safety nets assessment: do the programs meet the needs? Restricting eligibility to certain individuals and using one of the targeting methods outlined in box 4.2 can result in better targeting accuracy—but has the potential of high costs. It is there- fore important to put targeting and the consequent impact of poverty in perspective: does $1 spent reduce poverty by $1? Social safety net programs in Burkina Faso mainly target beneficiaries on a geographical basis, although this is often combined with an additional selection method (figure 4.7). Administrative data for the 23 main social safety net programs show that 14 programs use the geographical targeting method and combine it with either poverty, categorical, and/or self-targeting. Looking at expenditure, about a quarter of social safety net spending uses a combination of geographical and categorical targeting, including the nutrition program for children under age 5. Categorical targeting alone is used for the school feeding program, which accounts for 19 percent of total spending and has nationwide coverage. Nine programs—including the unconditional cash transfer programs for the poor and food assistance—use both geographical and poverty targeting, although the specific method used varies between HEA (see appendix B) and PMT (see appendix C), but most always combined with community targeting. Public works and scholarship programs imply an application pro- cess; hence they include a self-selection targeting mechanism (mixed with categorical and/ or geographical methods). This is in line with other countries in the West African region, where most safety net spending is categorically targeted (Beegle, Coudouel, and Monsalve 2018). FIGURE 4.7  Spending and number of social safety nets by targeting methods % of SSN spending Number of programs 25 25 20 20 15 15 10 10 5 5 0 0 Geographical Categorical Geographical Geographical Not Categorical & Poverty None Geographical & categorical & poverty & categorical available self-selection & poverty & self-selection SOURCE: Appendix A. Although the geographical selection of areas of intervention is not perfectly aligned with levels of poverty, it is in line with the overarching economic char- acteristic of the country. While it is not possible to analyze expenditures by geographical areas (since total program expenditures are often spent in multiple regions), the number of social safety net programs offers some insights (figure 4.8). 70 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO FIGURE 4.8  Number of social safety nets by region The two regions with relatively low poverty rates are host to a rather large set of programs. 4°E 2°E 0° 2°E Sahel has the most programs (7 out of 23), while poverty IBRD 44749 | JANUARY 2019 is relatively low (20.7 percent). Sahel is known as a frag- This map was produced by the Cartography Unit of the World MALI Bank Group. The boundaries, colors, denominations and any other information shown on this map do not imply, on the part of the World Bank Group, any judgment on the legal status of any territory, or any endorsement or acceptance of such boundaries. 14°N SAHEL NIGER 14°N ile region exposed to natural hazards, and several safety NORTH CENTRAL NORTH nets have been implemented to support production (and hence self-consumption) in the region, such as food BOUCIE DU MOUHOUN CENTRAL PLATEAU CENTRAL EAST assistance programs, unconditional cash transfers, and 12°N  12°N CENTRAL WEST CENTRAL SOUTH agricultural programs. The Central region has a relatively CENTRAL EAST large number of programs (4), including cash-for-work HAUTS-BASSINS BENIN CASCADES SOUTH 0° TOGO 2°E programs and food subsidies (Food Shops). The cash- WEST 10°N  GHANA BURKINA FASO PNUMBERS OF SSN for-work program is one of the largest cash transfer CÔTE D'IVOIRE 0 20 40 60 80 100 Kilometers programs in Burkina Faso and operates mainly in the 0 20 40 60 80 Miles 0 7 4°E 2°E capital, which is in line with the need created by rapid urbanization. SOURCE: Authors’ calculations based on EMC 2014. The two poorest regions could benefit from more safety net programs, in particular pro- grams targeting the poor. There is only one cash transfer program in the North region, and there are none in Boucle du Mouhoun. The geographical distribution of beneficiaries as revealed by EMC 2014 data also shows some misalignment with respect to poverty. Most beneficiaries (34.7 per- cent) of cash transfers live in the Central region; while only 8 percent come from the North region, 6 percent from the East, and 3 percent from Boucle du Mouhoun—the three poorest regions. Geographical coverage of the food distribution program shows that 72 percent of beneficiaries live in Boucle du Mouhoun, followed by 14 percent in Hauts-Bassins (figure 4.9). The misalignment of social safety nets with poverty by region reflects a greater focus on preventing food crises rather than tackling poverty alleviation. Geo- graphical targeting in Burkina Faso relies mainly on the SAP, which is often combined with the HEA poverty targeting method. These two methods have preliminarily been used to reduce vulnerabilities with respect to shocks and increase food security. Not surprisingly, these meth- ods lead to a greater focus on regions that face a high frequency of natural hazards, such as the Sahel. EMC survey data point to the need to improve targeting of social safety nets for the poor and vulnerable (figure 4.10). While the survey does not capture the ben- efits repartition of all social safety nets,1 it shows that benefits are primarily accrued by the top quintiles. More than 85 percent of scholarship benefits go to the top quintile, and less than 1 percent goes to the three bottom quintiles. Spending on other transfers is also largely directed toward the richest (62 percent of spending), with one-quarter accrued by the second 1 The amounts received for government support in case of shocks and support from NGOs are unknown, as only entitlements are included in the survey for these programs. Food distribution is also in the form of a dummy that accounts for those living mainly from the transfer. 71 chapter 4. social safety nets assessment: do the programs meet the needs? FIGURE 4.9  Distribution of social safety net beneficiaries a. Cash b. Food 4°E 2°E 0° 2°E 4°E 2°E 0° 2°E IBRD 44752 | JANUARY 2019 This map was produced by the Cartography Unit of the World IBRD 44744 | JANUARY 2019 MALI Bank Group. The boundaries, colors, denominations and any This map was produced by the Cartography Unit of the World other information shown on this map do not imply, on the part of MALI Bank Group. The boundaries, colors, denominations and any other information shown on this map do not imply, on the part of the World Bank Group, any judgment on the legal status of any the World Bank Group, any judgment on the legal status of any territory, or any endorsement or acceptance of such boundaries. territory, or any endorsement or acceptance of such boundaries. SAHEL SAHEL 14°N NIGER 14°N 14°N NIGER 14°N NORTH NORTH CENTRAL NORTH CENTRAL NORTH BOUCLE DU MOUHOUN BOUCIE DU PLATEAU PLATEAU CENTRAL MOUHOUN CENTRAL CENTRAL CENTRAL EAST EAST 12°N  12°N 12°N  12°N CENTRAL WEST CENTRAL SOUTH CENTRAL EAST CENTRAL WEST CENTRAL EAST CENTRAL SOUTH HAUTS-BASSINS HAUTS-BASSINS BENIN BENIN TOGO TOGO SOUTH 0° 2°E CASCADES WEST CASCADES SOUTH 0° 2°E WEST 10°N  GHANA BURKINA FASO 10°N  GHANA BURKINA FASO DISTRIBUTION OF BENEFICIARIES DISTRIBUTION OF BENEFICIARIES (SSN- NO FOOD) (FOOD DISTRIBUTION) 0 20 40 60 80 100 Kilometers 0 20 40 60 80 100 Kilometers CÔTE D’IVOIRE 0 20 40 60 80 Miles 0.4 34.7 CÔTE D'IVOIRE 0 20 40 60 80 Miles 4°E 2°E 0.0 72.0 z 4°E 2°E SOURCE: Authors’ calculations based on EMC 2014. FIGURE 4.10  Social safety net targeting, as a share of benefits and beneficiaries a. Distribution of benefits b. Distribution of beneficiaries 100 100 Q5 80 Q4 80 Q3 60 Q2 60 Q1 40 40 20 20 0 0 Scholarships Other Scholarships Government Other Support from Food transfers support (in case of transfers NGOs distribution shocks) SOURCE: Authors’ calculations based on EMC 2014. quintile. The distribution of beneficiaries of scholarships is in line with the distribution of ben- efits. Other transfers show a distinct pattern for distribution of beneficiaries, indicating that there are a lot of beneficiaries in the second quintile, but that the transfer received is small; large transfers are received by a few rich households. Looking at entitlements of government support in case of shocks and support from NGOs, about half of the beneficiaries are from the two bottom quintiles. About half of all food distribution beneficiaries are from the bottom two quintiles; the number of beneficiaries from the richest quintile is the lowest among the social safety nets. 72 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO Given the low coverage and poor targeting, the impact of social safety nets on With better targeting poverty is low; however, closing the poverty gap is technically feasible. In the absence of social safety nets, poverty would increase by just 1.2 percentage points.2 Closing Burkina Faso could the poverty gap would be financially feasible with better targeting. The size of the actual poverty technically eradicate gap equals 2.26 percent of GDP, which is close to actual spending on social safety nets.3 This indicates that with better targeting, Burkina Faso would technically be able to eradicate poverty. poverty. It would be financially feasible to close the poverty gap with Simulations: subsidies versus safety nets Another way to improve the effectiveness of social safety nets and enhance improved targeting. The pro-poor spending would be to shift expenditure away from transfers that are size of the actual poverty regressive. In the government’s overall social protection budget (4.22 percent of GDP), a high 1.05 percent is dedicated to general price subsidies on energy products. These subsidies gap equals 2.26 percent of are usually known to benefit only the richest portion of the population, as they are the ones GDP, which is close to actual who consume the most of these types of products. The simulation exercise that follows (see spending on safety nets. box 4.3 for details of the methodology) has two objectives: (1) to confirm that energy subsi- dies are regressive, and (2) to estimate what the impact on poverty would be if—for the same social protection budget—expenditure was shifted away from non–pro-poor subsidies (i.e., the case of gas) toward pro-poor social safety nets. Energy products often more expensive in poor areas Energy prices are not aligned with poverty: the price is higher in provinces that record a high poverty rate (figure 4.11). Gas subsidies are therefore not used as a pro-poor policy. The price of energy products varies across regions and provinces as a conse- quence of transportation costs, geographical location, and product availability (need to have this for fuel and electricity). For gas, the price per kg after transportation and taxes varies from CFAF 644.01 to CFAF 715.33. The subsidy amount also varies by region, from CFAF 214.74 to CFAF 515.5 (MICA-MEF 2014a). Subsidies could potentially make energy products more affordable in the poorest region; however, the postsubsidy price as estimated using EMC 2014 data shows that the price is not aligned with poverty: the price is higher in provinces that record a high poverty rate. The gas subsidy is therefore not used as a pro-poor policy. Energy consumption: subsidies do not benefit the poor Rich households spend more than poor households on energy products (fig- ure 4.12). The EMC provides data on fuel, gas, and electricity consumption, which allows the progressivity of energy subsidies to be estimated. The average amount spent by house- holds for each energy product increases depending on welfare distribution (welfare is based Based on a static simulation of the poverty rate in the absence of social safety nets that are quantified 2  in the EMC, i.e., scholarship and other transfers. Only entitlements are captured for support from the government and NGOs in case of shocks, as well as food transfers. Computed as the sum of the distance between actual consumption and the poverty line for each poor 3  household, based on the per capita national poverty line (CFAF 153,530 per year). 73 chapter 4. social safety nets assessment: do the programs meet the needs? BOX 4.3  Simulations methodology T he simulations presented here were performed using EMC 2014 data, as well as administrative data provided by various counterparts. EMC data were first used to compute the annual consumption of various energy products by household (by summing quarterly data over individ- uals in each household) and looking at how subsidy expenditures were distributed across the various welfare categories. Because the EMC data only provide information on subsidy expenditures and not on the quantity of energy products consumed, the EMC 2014 price data set was used to estimate the latter. This data set includes the prices of a large set of goods and services by quarter. Data are collected at the community level (not the household or individual level), and yearly prices of energy products were estimated using the average of quarterly prices in each province. The annual prices of energy products therefore represent postsubsidy prices. Not surprisingly, prices vary by province, as both market prices and subsidies vary across the 45 provinces. The actual quantity of gas consumed by households was estimated by dividing expenditure on energy products by each household, using postsubvention prices. The market price of gas (the presubsidy price) was taken from legal agreements between the Ministry of Infrastructure and the Ministry of Economy and Finance. Since market price is also influenced by transportation costs and country provision of gas, it was estimated at the market price at the provincial level to reflect this heterogeneity.1 The impact of the potential removal of subsidies was assessed in several ways (i.e., the removal of gas subsidies as market prices are not available for fuel; and since electricity is often jointly consumed among several households, it is challenging to evaluate impact at the household level): „„ By looking at the impact on household welfare: given the estimated quantities, it was estimated by how much their total welfare aggregate would have decreased if the price of gas increased. „„ By estimating how much the removal of subsidies would ease the government’s budget constraints: it was estimated that CFAF 13 billion was spent on gas subsidies in 2014 and considered different scenarios and alternative ways of using this budget rather than on subsidies.2 The legal agreement (MICA/MEC 2014a) provides the price by locality. Price by province has therefore been computed in several ways, according to the number of 1  localities by province. In total, there are 45 provinces over 13 regions. (1) If we have the price for one locality in a province, we assign that price to the entire province; (2) if we have prices for more than one locality, we assign the average of those prices to the province; (3) if we do not have prices for any locality in a province, we assign to that province the average of the (computed) prices for the other provinces within the same region (e.g., Central East average for Koulpelogo province and Central North average for Namentenga province, plus 9 other cases). Price varies slightly according to the quantity of the package. However, since we have no information on the quantities of gas that were bought (only household expenditure on gas), we take the average price of 2.75 kg, 6 kg, 10.8 kg, and 12.5 kg. In 2016, administrative data provided by the government on the price of the gas subsidy were CFAF 18.6. 2  on a consumption aggregate, the same that allows estimation of the official poverty rate). Even though poor households consumed more fuel than any other energy products, the aver- age annual spending of the poorest decile (CFAF 8,589) is comparatively much lower than that of the richest decile (CFAF 69,933). The share of households consuming each energy product reveals (independently of the amount consumed) that few poor households consume energy prod- ucts (figure 4.13). Virtually no household in the poorest decile consumes gas (only three 74 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO FIGURE 4.11  Observed price of gas in CFAF per kg and poverty headcount, by province a. Price of gas b. Poverty headcount 4°E 2°E 0° 2°E 4°E 2°E 0° 2°E IBRD 44161 | JANUARY 2019 IBRD 44746 | JANUARY 2019 This map was produced by the Cartography Unit of the World This map was produced by the Cartography Unit of the World MALI MALI Bank Group. The boundaries, colors, denominations and any Bank Group. The boundaries, colors, denominations and any other information shown on this map do not imply, on the part of other information shown on this map do not imply, on the part of the World Bank Group, any judgment on the legal status of any OUDALAN the World Bank Group, any judgment on the legal status of any OUDALAN territory, or any endorsement or acceptance of such boundaries. OUDALAN territory, or any endorsement or acceptance of such boundaries. SOUM SOUM 14°N LOROUM SENO NIGER 14°N 14°N LOROUM SENO NIGER 14°N YATENGA YATENGA BAM YAGHA BAM NAMENTENGA YAGHA SOUROU NAMENTENGA ZONDOMA SOUROU ZONDOMA SANMATENGA GNAGNA SANMATENGA KOSSI GNAGNA PASSORE KOSSI PASSORE NAYALA KOMANDJARI OUBRITENGA KOMANDJARI KOULWEOGO OUBRITENGA NAYALA KOULWEOGO BOULKIEMDE KADIOGO GANZOURGOU KADIOGO GANZOURGOU BANWA MOUHOUN SANGULE BANWA MOUHOUN BOULKIEMDE KOURITENGA OUDALAN KOURITENGA TAPOA GOURMA GOURMA TAPOA 12°N  BAZEGA 12°N 12°N  SANGULE BAZEGA TAPOA 12°N BALE BOULGOU KENEDOUGOU BALE KENEDOUGOU ZIRO ZOUNDWEOGO ZIRO BOULGOU TUY KOULPELOGO KOMPIENGA ZOUNDWEOGO TUY KOULPELOGO KOMPIENGA HOUET HOUET SISSILI NAHOURI SISSILI NAHOURI LOBA LOBA Observed price of gas, in FCFA (per KG). BOUGOURIBA BENIN BOUGOURIBA BENIN LERABA TOGO LERABA TOGO 0° 2°E KOMOE PONI PONI 0° 2°E KOMOE 10°N  GHANA BURKINA FASO 10°N  GHANA BURKINA FASO NOUMBIEL Observed price of gas, in FCFA (per KG). NOUMBIEL POVERTY HEADCOUNT RATIO 0 20 40 60 80 100 Kilometers 0 20 40 60 80 100 Kilometers CÔTE D'IVOIRE 0 20 40 60 80 Miles CÔTE D'IVOIRE 367 583 0 20 40 60 80 Miles 7.22 85.72 4°E 2°E 4°E 2°E SOURCE: Authors’ calculations based on EMC 2014. FIGURE 4.12  Household spending on energy consumption, by decile a. Gas b. Electricity c. Fuel Thousand CFAF Thousand CFAF Thousand CFAF 30 140 80 120 70 25 100 60 20 50 80 15 40 60 30 10 40 20 5 20 10 0 0 0 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 Decile (poorest to richest) Decile (poorest to richest) Decile (poorest to richest) SOURCES: Authors’ calculations based on EMC 2014; MICA-MEF 2014a, 2014b, and 2014c. observations), 1.8 percent consume electricity, and 33.1 consume fuel. In fact, fuel is the most common and accessible type of energy product in Burkina Faso. Removing subsidies and alternative policies Given that the distribution of energy consumption is regressive, removal of gas subsidies would barely affect the poor; rather, it would mainly affect the wel- fare of the richest quintile (figure 4.14). Focusing on gas,4 removing subsidies would Given the joint consumption of electricity among households and the nonavailability of a market price for 4  fuel. 75 chapter 4. social safety nets assessment: do the programs meet the needs? FIGURE 4.13  Share of household income spent on energy consumption, by decile a. Gas b. Electricity c. Fuel Percent Percent Percent 80 80 80 60 60 60 40 40 40 20 20 20 0 0 0 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 Decile (poorest to richest) Decile (poorest to richest) Decile (poorest to richest) SOURCES: Authors’ calculations based on EMC 2014; MICA-MEF 2014a, 2014b, and 2014c. generate total savings of CFAF 12.9 billion. This is almost FIGURE 4.14  Impact of subsidy removal on population the same amount as social safety net expenditures, the well-being equivalent of 2.31 percent of GDP (or CFAF 147 billion). The gas subsidy budget is equivalent to the total cost of the Billion CFAF school feeding program (CFAF 19 billion). Poverty would 0 not be affected, as post-reform simulated poverty is still −20 40.1 percent. −40 −60 Table 4.6 presents the impact on poverty and the inequality −80 of alternative uses of gas subsidy savings according to var- −100 ious scenarios, as follows: −120 „„ Scenario 1: Current situation: gas consumption −140 1 2 3 4 5 6 7 8 9 10 All is subsidized. Decile (poorest to richest) „„ Scenario 2: No subsidy; savings from removal are not spent. SOURCES: Authors’ calculations based on EMC 2014; MICA-MEF 2014a, 2014b, and 2014c. „„ Scenario 3: Savings (CFAF 12.9 billion) are divided equally among the poor; that is, all those with a per capita annual income less than CFAF 153,530. „„ Scenario 4: Savings are divided equally among households in the first decile; that is, those who are further away from poverty line, since the upper limit of decile 1 is CFAF 109,144.5, and the poverty line is CFAF 153,530. „„ Scenario 5: Same as Scenario 4, but with geographical targeting; that is, savings are divided equally among rural households in the first decile. „„ Scenario 6: Geographical targeting only and no means test; that is, savings are divided equally among all rural households, regardless of poverty status. 76 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO TABLE 4.6  Impact on poverty and inequality of various gas subsidy savings scenarios Scenario 1 Scenario 2 Scenario 3 Scenario 4 Scenario 5 Scenario 6 Scenario 7 Scenario 8 Total cost (CFAF) 13 billion 0 13 billion 13 billion 13 billion 13 billion 13 billion 13 billion Beneficiaries 97.244 0 7,375,787 2,699,906 147,066 4,077,434 169,461 96,164 Transfer amount (CFAF)** n.a. n.a. 1,757 4,800 88,126 3,179 76,480 134,774 Headcount (%) 40.1 40.1 39.2 40.1 39.4 40.1 39.6 39.9 Poverty gap 0.09666 0.09668 0.09215 0.09208 0.09347 0.09607 0.0957 0.09614 Poverty severity 0.03286 0.03286 0.0307 0.02924 0.03151 0.03261 0.03255 0.03272 Gini 0.35553 0.35402 0.35115 0.350 0.35171 0.35431 0.35326 0.35395 SOURCES: Authors’ calculations based on EMC 2014 and MICA-MEF 2014a, 2014b, and 2014c. NOTE: n.a. = not applicable. Headcount, poverty gap, and poverty severity are based on Foster–Greer–Thorbecke formula, respectively FGT0, FGT1, and FGT2. The total cost of each scenario is equal to the savings from phasing out gas subsidies. Darker shading indicates lowest values. „„ Scenario 7: Use targeting system of current social safety net beneficiaries (as cap- tured by EMC 2014) with state programs; that is, savings are divided equally among those who already receive government support. „„ Scenario 8: Use targeting system of current social safety net beneficiaries (as cap- tured by EMC 2014) with NGO programs; that is, savings are divided equally among those who receive support from NGOs. Sustainability The expansion of social safety nets has been accompanied by shifts in the design of programs. It has also often been associated with investments in systemic instru- ments, such as targeting systems, registries, and payment systems to strengthen the overall system and increase efficiency. Over the last five years, resources have been invested not only for transfers but also to build a more sustainable social safety net system. This includes efforts to gain political and public support. Social safety net systems need to have a solid institutional basis to be credible and sustainable; this requires a clear legal framework. This framework needs to be built on policy objectives that are widely supported among the population. In 2015, the government invested in different programs in view of improving governance and institutional capacities, and reinforcing the legal and institutional framework for social pro- tection. The activities included several formulations of staff involved in the provision of safety nets at the local and central levels, large-scale information sessions on safety nets and publi- cation of the National Social Protection Strategy in various national languages, and awareness campaigns (CNPS 2016b). Only half of safety net spending is financed entirely by the government, which may undermine its predictability. A large percentage (19 percent) of safety net spending is financed by international partners (mostly multilateral), while another 26 percent is financed 77 chapter 4. social safety nets assessment: do the programs meet the needs? using a combination of national and international funds. Dependency on external donors introduces uncertainty and unpredictability, and may lead to program termination. Given the large numbers of actors involved in the provision of safety nets in Burkina Faso, a beneficiary registry would reduce the cost of targeting ben- eficiaries and avoid duplication. A registry has been initiated under the cash transfer program Burkin-Naong-Sa Ya; but it remains very limited, with less than 0.8 percent of the population covered as of 2015 (figure 4.15). The registry has, however, been extended with the increase in program beneficiaries to 442,416 individuals, or more than 2 percent of the population, in 2018. FIGURE 4.15  Coverage of social registries by African countries Coverage (percent of population) 30 25 20 15 10 5 0 Côte d’Ivoire Mozambique Nigeria Benin Chad Zimbabwe Madagascar Ghana Mali São Tomé & Príncipe Burkina Faso Comoros Congo, Rep. Sierra Leone Swaziland Kenya Mauritania Rwanda Tanzania Mauritius South Africa Lesotho Senegal Gabon SOURCE: Beegle, Coudouel, and Monsalve 2018. Other elements of a sustainable system are missing, such as common deliv- ery platforms. As the number of programs grows, the risk of overlap and duplication is exacerbated by the absence of shared systems. Delivery platforms such as interoperable management information systems and shared payment systems allow administrative cost savings and facilitate planning (Beegle, Coudouel, and Monsalve 2018). These are, however, absent in Burkina Faso. Sustainability is further challenged by spending delays, resulting in a low execution rate. A necessary condition of a credible social safety net, and of the overall social protection system, is a commitment by the public and the government—especially if long-term effects are expected, such as allowing poor households to invest in human- and capital-building activities. Financial execution rates (actual expenditure as a share of budget) 78 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO are low, about 60 percent on average, with respect to budgeted amounts for cash and food social safety nets;5 they are 108 percent for education and health social safety nets. TABLE 4.7  Execution rates for selected programs Execution rates vary greatly with respect to the Execution rate nature of the program, financing sources, and Program (% of budget) program duration (table 4.7). The school feeding School feeding (2016) 100 program had a 100 percent execution rate in 2016, as Food assistance to vulnerable populations 247 is expected with a long-standing program. However, Burkin-Naong-Sa Ya 110 in 2015, some delays were reported due to contract- Unconditional cash transfer to poor in Sahel and Central 10 ing process issues affecting the delivery of food to North regions schools, leading to a low execution rate. Execution rates Subsidies for fertilizer 53 can also be affected by political changes, as with the SOURCE: CNPS 2016b. long-standing program of food assistance to vulnerable populations; its expenditure doubled due to the intro- duction of social measures in 2014 (box A.1). Additionally, execution rates may depend on external partners when they are involved in the financing and/or delivery of the transfer. The Burkin-Naong-Sa Ya program, for instance, saw increased spending as the initial number of beneficiaries had been decided by its financial partner (the World Bank). On the other hand, only 10 percent of planned expenditures were spent on unconditional cash transfers to the poor in the Sahel and Central North regions, given a lack of external funding. Other common reasons for low execution rates include contracting and acquisition issues, as illustrated by the low execution rate for the fertilizer subsidies program. Spending delays raise questions about governance. In a 2016 survey conducted by the national anti-corruption NGO REN-LAC, one-third of respondents cited having personally experienced corruption or witnessed acts of corruption, notably in the public procurement process. Such acts of corruption are likely to undermine the impact of social safety nets, pro- ducing delays and low execution rates. Transparency and accountability of resources used for social safety nets are a key determinant of their effectiveness. With low scores for government effectiveness and voice and accountability,6 there is a need in Burkina Faso for mechanisms that reinforce accountability and give a voice to civil society, such as a grievance redress mechanism. A recent World Bank initiative will support development of this process in both the social safety net and education sectors. Based on social safety net execution rates reported by the CNPS (2016b), and excluding programs from 5  the Executive Secretariat’s National Food Security Strategy (SE-SNSA), for which only budgeted amounts are available. These data are from the Worldwide Governance Indicators database (https://info.worldbank.org/ 6  governance/wgi/#home). Voice and accountability capture perceptions of the extent to which a country’s citizens are able to participate in selecting their government, as well as freedom of expression, freedom of association, and a free media. Government effectiveness captures perceptions of the quality of public services, the quality of the civil service and the degree of its independence from political pressures, the quality of policy formulation and implementation, and the credibility of the government’s commitment to such policies. 79 chapter 4. social safety nets assessment: do the programs meet the needs? Monitoring and BOX 4.4  Considerations in improving M&E evaluation Despite recent improvements, Burkina Faso still faces a gap in its monitoring and evalua- M onitoring and evaluation are essential strategic components of safety net program management. Information produced by M&E is vital in providing feedback to enhance program effec- tion (M&E) processes (box 4.4). A big step toward tiveness, making projects accountable to the public, and helping improving monitoring was the 2013 establishment of the government better allocate budget resources. Monitoring and eval- CNPS, which then enabled the gathering of evidence uation are separate yet complementary. Monitoring is a continuous to show progress made in program implementation. process of collecting and analyzing information to better understand The CNPS publishes annual reports of social protection how well a program is operating against expected outputs. Evalua- interventions that track spending, beneficiaries, and tion is an objective assessment of program effectiveness that uses physical and financial execution rates. However, the specialized methods to determine whether a program meets its latter data allow for just partial evaluation of the National objectives, to estimate its net results or impact, and/or to identify Social Protection Strategy, as it only measures the whether the benefits the program generates outweigh its costs. Pro- change between budgeted (as stated in the three-year grams with strong M&E systems benefit from feedback that allows action plans) and actual spending. Reasons for low/high for improvement in program productivity, effectiveness, and impact. execution rates are briefly reviewed, but there is no anal- Key considerations include the following: ysis of whether intended results were achieved. Better synthesis of the causes underlying unintended results „„ Devise a customized management information system that of the strategy would help in providing lessons and automatically produces disaggregated information presented in recommendations for improvement. The lack of M&E user-friendly ways to facilitate understanding of findings. processes at the program level is mainly responsible for the difficulty in formulating adequate yearly reviews. „„ Effective M&E systems require a strategic focus and manage- ment support more than they require costly investments in Improving the monitoring process would require information technology. better coordination among institutions and „„ Avoid temporary M&E systems decoupled from each other with dif- partners. The CNPS annual reports do not track all ferent reporting requirements without a sufficient in-house support programs. Collaboration with other institutions involved structure and without being linked to a program’s strategic plan. in social protection should be reinforced (e.g., to include Keep in mind potential synergies in scaling up programwide M&E nutrition programs and NGO programs) to enable better systems to sector and national levels. tracking of social protection interventions in the country. The most recent three-year action plans report budgeted „„ Ensure the M&E unit is independent and has sufficient authority interventions for 2017 for bilateral and multilateral part- and access to upper management. An M&E unit should also be ners as well as NGOs; this may indicate their willingness shielded from political influence. to track interventions in future annual reports. „„ Coordination and communication are critical, particularly for complex programs. Hold periodic follow-up meetings to ensure Few programs are designed with an M&E com- program goals are in sight. Data collectors and processors ponent, and no harmonized M&E system exists. should work within an enabling environment and have enough While the government has expressed interest in invest- incentives to manage data effectively. ing in M&E for social safety nets, only a few programs have an M&E system. While the CNPS monitors execu- SOURCE: World Bank n.d. tion of the national plan for social protection in its annual reports, there is no strategy for evaluating programs. chapter 5 recommendations and concluding remarks This final chapter offers some recommendations aimed at further increasing the impact, efficiency, and sustainability of social safety nets—as well as the overall social protection system in Burkina Faso. 82 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO Recommendations Recommendation 1: Strengthen the governance of the social safety net system Despite progress in reducing poverty and vulnerability in the past decade, Burkina Faso’s daunting human capital challenges and poverty levels call for a systematic build-up of the social protection system and of social safety nets in particular. The system’s institutional gov- ernance needs to be strengthened. In this context, the following actions should be prioritized: „„ Create a framework for multisectoral coordination „„ Monitor and evaluate program effectiveness and impact—and address accordingly „„ Strengthen public expenditure management „„ Include citizens in oversight of execution and delivery Action 1.1: Create a framework for multisectoral coordination Identify ministerial comparative advantages and clearly establish key actors’ responsibilities. The relative strengths of each relevant line ministry should be better har- nessed. Competition among existing services and institutions is harmful and delays delivering to beneficiaries. Review the institutional set-up under the stewardship of the central govern- ment. The prime minister or the Ministry of Economy and Finance should lead an institutional review that resolutely identifies redundancies and dysfunctionalities in agencies and bodies entrusted with social protection and proposes bold reform decisions to remove these. Exam- ples of successful interagency coordination, including down to the local level, include the Rwanda Vision 2020 Umurenge Programme public works component and the Tanzania Pro- ductive Social Safety Net. Use one overarching national social protection strategy. Further ministerial social protection–related action plans and strategies all need to derive from the overarching strategy and need to refer to each other according to the established responsibilities. Improve information and data collection systems and develop ministerial reporting mechanisms. The difficulty in obtaining basic information from the institutions responsible for the programs reviewed in this study provides a strong indication of the inade- quacy of existing information systems and the insufficiency of data collection on these topics. This is an important area to improve in preparation for a unified registry of beneficiaries (see Recommendation 3). Ensure prioritization. The National Council for Social Protection—Burkina Faso’s intersec- toral oversight body—should assign its internal groups time-bound and specific output and outcome goals to ensure prioritization. The council currently only meets once or twice a year 83 chapter 5. recommendations and concluding remarks and mainly focuses on information sharing, with the primary output being a list of programs and resources spent. Lack of prioritization is often due to a lack of expectation as to what should be produced. In contrast, successful coordination would involve leadership groups being assigned specific output goals. An example of an effective coordination structure is Ethiopia’s Safety Net Support Facility, which has provided training and backup on leadership, understanding terms of reference, preparing agendas, chairing meetings, dealing with nonat- tendance, and documenting action points. Action 1.2: Monitor and evaluate program effectiveness and impact—and address accordingly Establish a robust system of monitoring and evaluation to facilitate informed decision making. Burkina Faso’s government currently has little information on the effec- tiveness and impact of its social protection programs. As mentioned above, there is a general lack of accuracy and consistency in data collection by the responsible institutions. It is essential to systematically monitor and track program implementation so irregularities can be detected, and timely action taken to make improvements. Evaluations are needed to estimate impact and identify whether the benefits a program generates outweigh its costs, and so on, and thus enable informed decision making as to which programs to expand, rationalize, or consolidate. Evaluate the impact of the various small safety net programs. Even by regional standards, Burkina Faso has an extremely large number of small, and frequently apparently overlapping, social safety net programs. These programs need to be evaluated for reach, effectiveness, efficiency, and governance. Overlaps need to be addressed. Resources need to be removed from ineffective and regressive programs, and channeled to scale up effective programs. Such actions would reduce overall implementation costs and likely create fiscal space, while at the same time improving program impact. Moreover, gains could be made if interactions across social protection categories were exploited, in a move toward a more inte- grated social protection system. Action 1.3: Strengthen public expenditure management Strengthen budget formulation and improve management, execution, and auditing. Budget formulation needs to be strengthened: all major social safety net pro- grams should be shown explicitly in the budget. Management, execution, and auditing must be improved. A fragmented budget planning process discourages better use of available resources to achieve desired results. Action 1.4: Include citizens in oversight of execution and delivery Increase the transparency of public spending to ensure the benefits of public spending reach the poor. Civil society organizations, individual citizens, and the media should be able to monitor processes associated with social protection spending and service delivery as well as social worker deployment, presence, and performance. The agencies 84 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO responsible for program execution need grievance redress mechanisms, including nationwide hotlines, complaint processing and resolution, and publication of results. These mechanisms could be pooled across programs and agencies. Evidence of public spending governance in Burkina Faso is not encouraging, since at times, elites or interest groups reap the benefits of spending intended for the poor. Recommendation 2: Improve program targeting of the poor and vulnerable and move away from regressive spending More resources need to be allocated to programs that target the poorest and most vulnerable. To this end, either (1) some funding should be shifted from programs benefiting the richest populations to programs targeting the poor, or (2) the overall program budget should be increased. Because Burkina Faso’s revenue sources are unlikely to create further substantial fiscal space, it is more realistic to reallocate expenditures from less efficient programs before considering a budget increase. The following actions should be prioritized: „„ Disseminate the use of a proven targeting system „„ Shift spending from universal subsidy programs toward programs targeting poor Action 2.1: Disseminate the use of a proven targeting system Use an integrated approach for the assessment of socioeconomic needs and conditions when possible to effectively reach the poorest. Social safety nets in Burkina Faso use several methods to target the poorest. While it is acknowledged that pro- grams may have different eligibility criteria and thresholds to reflect their specific objectives, While PMT and HEA can the use of an integrated approach for assessing socioeconomic needs and conditions for the have higher costs, they also same target group (e.g., the chronic poor) should be considered. Most social safety nets use a combination of geographical and either categorical or poverty targeting. However, poverty have a higher targeting targeting varies by program and can include community targeting, proxy means testing (PMT), efficiency. and the household economy approach (HEA). While the categorical method is relatively less costly than poverty targeting, it is also less precise, leading to relatively high inclusion and exclusion errors. On the other hand, while the main poverty targeting methods—PMT and HEA—can have higher costs, they also have a higher targeting efficiency. Therefore, a PMT with HEA variables could be an option. The government should take the lead and engage in a discussion with all partners involved in social safety nets with the objective of reducing costly duplication across different methods. Align targeting methods with government objectives. The choice of PMT and HEA methods is not without consequence for the prioritization of beneficiaries, and there must be a clear understanding of how the methods may achieve government objectives (box 5.1). The more widespread use of HEA methods shifts the focus of social safety nets on food-insecure households. It might therefore fail to identify chronically poor households and bias priorities away from the poorest at the national level by focusing on zones with vulnerable livelihoods. 85 chapter 5. recommendations and concluding remarks In extreme cases, the overlap between beneficiaries for PMT and HEA methods may be low. BOX 5.1  Targeting efficiency of PMT and HEA Ensure interventions are aligned with areas that S suffer from high poverty rates and low cover- chnitzer (2016) studied the targeting efficiency of PMT and age. Analysis of administrative and survey data revealed HEA using data from Niger. The analysis revealed that when that regions with high poverty rates, such as the North measuring welfare using household consumption, PMT leads to and Boucle du Mouhoun, face significant undercoverage better targeting and of cash social safety nets compared to other regions. By limited inclusion error Percent addressing other constraints, such as reinforcing local compared to HEA. Among 30 Seasonal poor capacities and improving security in the border regions, PMT beneficiaries, 24 per- Chronic poor could help improve access to social safety nets for the 25 cent are poor (19 percent population. chronic poor and 5 per- 20 cent seasonal poor); HEA 15 Action 2.2: Shift spending from captures only 15 percent 10 universal subsidy programs toward of them. However, HEA programs targeting the poor better captures the sea- 5 sonal poor: 10 percent 0 Continue with already initiated energy subsidy PMT HEA versus only 5 percent with reform initiatives—that is, reduce spending PMT (see figure). allocated to broad-based consumer subsidy programs. If these efforts are accompanied by a shift While PMT better captures the poor, the study also shows that it toward spending on better targeted social safety net might fail to capture those vulnerable to shocks. HEA targeting programs, overall savings and greater impact on poverty includes a larger share of less-resilient beneficiaries (measured as reduction can likely be achieved. Since energy subsidies having a smaller household size, less livestock, and less diversified have been largely inefficient and have weighed down livelihoods). Burkina Faso’s economy, it makes sense to continue the reform efforts initiated. Freed-up expenditures should The study also points out several HEA shortcomings, including the be redirected toward more sustainable social protection lack of a theoretical model behind the weight used to determine measures that benefit the poor to a greater extent. In eligibility (a reliance on community, compared to a reliance on light of potential reform initiatives, the following should regression predictors of poverty for PMT) and the use of variables be taken into consideration: that are not easily verified (which creates measurement error). „„ Identify whom subsidies benefit. For Simulations suggest that a combined targeting tool—a permanent instance, CFAF 6.5 billion was paid directly to program that uses PMT which is scaled up in case of shock using cotton farmers in Burkina Faso in 2015; how- HEA—shows an overall better efficiency in targeting both chronic ever, since this category of farmers tends to and seasonal poor. be better off than producers in other farming sectors, the impact on poverty alleviation is SOURCE: Schnitzer 2016. questionable. Implementation procedures and beneficiary targeting methods could be revised so poor farmers would benefit more widely. „„ Assess fiscal costs of subsidies and the impacts of reform/price increases on households. Conduct fiscal cost estimates and distributional analysis of subsidies by income group, sector, and product; and simulations of the 86 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO impact of price increases and possible mitigation, as discussed in chapter 4 for gas subsidies. Gather consumer views and feedback to understand the impact of reform on them. There is often a lack of information and understanding on the cost of sub- sidies. „„ Assess the readiness of social safety nets to mitigate the impact of higher prices. Since the government would have to be able to rely on its safety net infrastructure to assist vulnerable groups affected by price increases due to subsidy removal, it is important to assess safety net readiness and areas where these would need to be strengthened. „„ Identify the impact of higher energy prices on firms and competi- tiveness. Determine whether a gradual transition may be warranted to mitigate employment risk. The removal of subsidies „„ Analyze the underlying political economy. Successful reform design is depen- must be accompanied by dent on an understanding of the political context of subsidy reform. Identify the political forces that created energy subsidies in the first place and the motivations efforts to protect the poor. behind them. Analyze the operations of special interest groups. This knowledge is The social safety net system important when building political buy-in for reform. must be strengthened „„ Learn from other country experiences. South-South learning is one such tech- nique—for example, as in Indonesia and Jordan. Although each country will have to beforehand, and measures customize its own process, a tremendous amount can be learned from how other designed that mitigate the countries have carried out subsidy reform. potentially adverse impact Protect the poorest consumers by expanding well-targeted social safety nets. of subsidy reform. Cutting expensive and poorly targeted subsidies must go hand in hand with credible policies to introduce better mechanisms for social protection. This is critical to the success of subsidy reform. The potentially negative impact on the poor of reducing any regressive programs must be mitigated by expanding the coverage of well-targeted programs. Design specific measures to mitigate the potentially adverse impact of reform on the poor while gradually phasing out subsidies. If subsidies are eliminated sud- denly and without strengthening social safety nets, it can have a very negative impact on the poor. Simply removing subsidies is not a viable option for reformers; instead, a majority of the most successful reform attempts have been a gradual process staggered over time. One option could be to begin by reducing expenditures for the energy subsidies that are mainly used by wealthier consumers before moving on to reducing expenditures for subsidies that mainly lower-income groups use and benefit from. Targeted compensation can be another way of protecting affected households from the impact of subsidy reform. „„ Targeted compensation. Examples include through the Burkin-Naong-Sa Ya cash transfer program. This can be effective even in the short term provided that targeting is accurate, a list is maintained of those in need, and there is a low-cost mechanism for transferring cash/benefits. 87 chapter 5. recommendations and concluding remarks „„ Target subsidies. A common way to target subsidies to poor households is to provide a larger subsidy or lower tax rate for kerosene—which is used more by poor households—than for other fuels. Another way would be to issue vouchers or smart cards that allow low-income consumers to purchase a limited amount of certain products at a reduced price. „„ Subsidize electricity connections. There is a need to find a way to increase access to electricity for low-income families by subsidizing the connections, while ensuring the reliability and quality of supply for those who already have access—and removing subsidies from households with larger connections. Efficient electricity services are associated with raising productivity and living standards, and access is often associated with improved educational outcomes. „„ Establish a public transport system. This is what Cameroon proposes to com- pensate for its envisaged subsidy removal, meaning that poorer people will not need to buy gas anymore. Communicate and engage with diverse stakeholders to avoid misperceptions and information asymmetry. Transparency and clarity around price adjustments is Communication is a critical important, as is explaining the costs and negative impacts of subsidies as well as the ben- element of effective energy efits reform could bring. Design a communications campaign prior to the start of the reform process. Increase the flow of information, and inform the public well in advance of any price subsidy reform, and to increases, while clearly explaining targeted actions to mitigate negative impacts. A coordi- make a credible offer to the nated communications and outreach effort (or informational reform) creates transparency and builds public trust and can reduce political risk and opposition based on misperceptions. It public that the removal of also can enhance the conditions for success (box 5.2). visible benefits will deliver new, yet currently invisible, Recommendation 3: Improve the gains. Reformers must find coordination of social programs by building a way not only to make that a social registry promise credible but also to As the social safety net system in Burkina Faso involves many sectors and comprises a large communicate to the public set of programs, further investments in a social registry containing socioeconomic information on the intended population would provide a common gateway for multiple programs. Further, what they are doing. it would improve coordination across sectors, agencies, and programs while reducing pro- gram duplication and overlap. The following actions should be prioritized: „„ Adopt a harmonized questionnaire „„ Build a social registry as a common gateway for multiple social programs Action 3.1: Adopt a harmonized questionnaire Identify opportunities for integration of intake and registration processes. Social safety net programs in Burkina Faso have their own intake and registration processes. These processes can be costly, so integrating them across different programs is an efficient way to reduce costs. When different programs require common information from similar population 88 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO BOX 5.2  Energy subsidy reform in Indonesia: need for political buy-in, trust building, and communication A lthough there is ample evidence that energy subsidies are one of the greatest economic distortions globally given their lack of sectoral efficiency, fiscal sustainability, and distributional/equity impact, this is not always enough to allow reforms to take hold. A key finding from The Political Economy of Energy Subsidy Reform (Inchauste and Victor 2017) is that the most successful reforms nearly always involve a large amount of political engineering, building political will, and ensuring buy-in: that is, active efforts by policy leaders to identify the political forces that created energy subsidies in the first place and then redirecting or inoculating those forces. Such political engineering has allowed govern- ments to avoid or blunt the political force of well-organized special interest groups. In some instances, a challenge of subsidy reform has been that, even though the general public may understand the gains of subsidy reform, they still oppose it as they think it is the only way they can benefit from their government. They simply don’t believe that the government will find other ways of compensating them. This finding points to the importance of laying the foundation beforehand. One of the most successful approaches in phasing out subsidies can be found in Indonesia. In 2015, fuel subsidies were removed in Indonesia, with virtually no dissent— since the previous government had built trust a head of time by putting in place an effective direct cash transfer system as the foundation of subsequent subsidy reform. Nineteen million people trusted the government and believed they would receive money, because they had done so before. Also, the government now offered its people something they truly wanted—free health care and education. It was often pointed out that, while fuel prices might rise, the money would be used to deliver these critical services. As a result, $15.6 billion—over 10 percent of total state expenditure—could be transferred to health, education, and infrastructure budgets. (It was recently announced that Indonesia’s fuel and electricity prices are to be frozen until the end of 2019, a presidential election year.) An additional ingredient in Indonesia’s success in reducing electricity subsidies was clear communication of the benefits of reform and an emphasis on protection of the poor; this helped shore up com- munity support. Source: Beaton, Lontoh, and Wai-Poi 2017. groups, it can be efficient to share these rather than collect the same information multiple times. Develop a common harmonized intake and registration questionnaire that cap- tures all variables used by the different programs. This could help to rapidly expand the national social registry. In the Burkina Faso context, the types of information collected by different programs are very similar. This even applies to programs that use different targeting methods, such as PMT and HEA. A harmonized questionnaire that captures all variables used by the different programs could serve as a building block for the establishment of a social registry and allow for rapid expansion. Such a harmonized questionnaire would also improve overall system efficiency by reducing data collection costs, strengthening coordination between partners, and enabling broader and faster responses to shocks as programs could be scaled up based on the expanded registry (Schnitzer 2016). 89 chapter 5. recommendations and concluding remarks Action 3.2: Build a social registry as a common gateway for multiple social programs Further develop Burkina Faso’s nascent social registry, as it would improve the effectiveness of the overall social safety net system. A social registry supports the processes of intake and registration and the assessment of socioeconomic needs and condi- tions. This enables program administrators to share resources for intake and registration and reduce duplication of effort and administrative costs. It can also simplify intake and registra- tion procedures for people, allowing them to access multiple benefits and services through a common application form so they do not have to provide the same information multiple times. Use the social registry as a tool for coordination on the assessment of needs and conditions to determine potential beneficiary eligibility for multiple pro- grams. Social registries can serve as a common gateway for coordinating registration and assessment of needs and conditions to determine potential eligibility for multiple programs. As the social safety net system in Burkina Faso involves many sectors and comprises a large set of programs, further investment in a social registry containing socioeconomic information on the intended population would provide a common gateway for multiple programs and increase coordination across sectors, agencies, and programs. Recommendation 4: Use social protection expenditure to build human capital where it matters most, such as for early childhood development and literacy Early childhood development is not now addressed by Burkina Faso’s social protection system. At the same time, the country’s human capital needs in terms of nutrition, early stimu- lation, and learning are staggering. The country’s literacy—both during the critical school ages as well as in adulthood—ranks far behind regional comparators. Literacy is a basic ingredient for success in the formal labor market and for effective citizen participation in society. The fol- lowing actions should be prioritized: „„ Link social assistance programs to human capital building along the life cycle „„ Promote basic skills training for adults and youth Action 4.1: Link social assistance programs to human capital building along the life cycle Enable parental investment in children’s human capital through cash and in-kind transfers. This investment should be encouraged; encouragement can be in the form of accompanying measures such as training parents on appropriate health behaviors and early education of children—a so-called “soft condition.” It can also be in the form of a “hard condi- tion,” whereby specific behaviors must be demonstrated in order to receive the transfers—for example, growth monitoring by a pediatrician or school attendance for older children. 90 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO According to the impact evaluations conducted on cash transfer pilots in Burkina Faso, hard conditions are more effective. Akresh, de Walque, and Kazianga (2012, 2013, 2016) show that while all cash transfers improved human capital, preventive health visits and school atten- dance were higher with a hard condition. Also, children traditionally at risk of discrimination, such as girls and lower-ability children, fared better under hard conditions than with uncondi- tional cash transfers. Building on this, the cash transfer program Burkin-Naong-Sa Ya should consider adding hard conditions for some key health behaviors in areas where the health and education infra- structure are sufficient for regular attendance. Regular health screening for children age 0–5, as well as schooling for children age 6–15 could be made compulsory. This would directly address key human capital deficiencies highlighted in chapter 2 pertaining to these ages. In regions where the health and education infrastructure are insufficient to condition social assistance on its utilization, soft conditions are appropriate. Currently, Burkin-Naong-Sa Ya uses a “moral contract” whereby beneficiaries make a com- mitment to good nutrition for all household members, school attendance by school-age children, health care attendance for sick household members, and civic participation. Scale up flexible service infrastructure. For instance, all public works/Labor-Intensive Youth Public Works Program (THIMO) activities should be supplemented with mobile child care as in the Youth Employment and Skills Development Project; this easy-to-scale structure is supported by the Ministry of National Education and Literacy and the Ministry of Youth, Training and Vocational Integration. It provides safe care and supervision for children age 0–5; basic health screening, stimulation, and learning activities for these children; an opportunity for mothers to work without interruption; an opportunity for older siblings, usually girls, to go to school instead of watching over smaller siblings; and a training opportunity in early child- hood education for selected THIMO beneficiaries. Action 4.2: Promote basic skills training for adults and youth Adult basic skills are seriously deficient in Burkina Faso, and more severely so among the poor. Basic skills include functional literacy and numeracy as well as basic computer skills. This issue is currently insufficiently addressed, but it could be integrated with social protection. Training in basic skills constitutes a substantial effort and is not easily combined with other interventions—but it can be done with appropriate, careful planning. The essential basic literacy program of Burkina Faso’s National Fund for Literacy and Nonformal Education (FONAENF) requires at least 400 hours of training. Given the sizable investments of time and resources (trainers, materials, venues) required, every occasion to train social pro- tection beneficiaries in basic skills should be identified and taken. „„ Identify and make use of every occasion to train social protection beneficiaries in basic skills. For example, beneficiaries of public works/THIMO programs could enroll in some literacy training alongside public works. 91 chapter 5. recommendations and concluding remarks „„ Various modalities of delivering basic skills should be tested and subjected to rigor- ous impact evaluation. Recommendation 5: Improve the capacity of social safety net programs to respond to shocks (before, during, and after) through adaptive social protection Given the degree to which Burkina Faso is subject to adverse natural events and the lack or irregularity of rainfall with corresponding food insecurity issues, it is crucial for the country’s safety net system to include flexible elements that can be scaled up quickly. This requires improving the existing early warning system for food insecurity and other issues. The follow- ing actions should be given priority: „„ Develop and test shock-responsive benefits „„ Strengthen the early warning system Action 5.1: Develop and test shock-responsive benefits Shock-responsive benefits need to be rolled out rapidly in shock-affected areas and meet the needs of vulnerable beneficiaries within a few weeks of crises. „„ Develop a solid early warning system to identify the geographical area that is affected. „„ Develop an efficient targeting system that can quickly identify the most vulnerable/ poorest households within the affected area (a standard PMT process would take too long). Burkin-Naong-Sa Ya has tested a shortened PMT process in the Boulkiemdé region to roll out a shock-response transfer. This process involved a campaign to invite households to self-identify as poor, a PMT survey with a shortened questionnaire for this reduced sample of self-targeted people, and community validation to add any forgotten poor households. The procedure took around one month, compared to a standard PMT survey and evaluation, which can take several months from survey to registration in Burkina Faso. The National Council for Social Protection has, together with UNICEF, developed a similar method of fast-process targeting. However, no data are currently available on its effectiveness. Action 5.2: Strengthen the early warning system Strengthen the existing Burkinabe early warning system to identify likely food-insecure provinces. 92 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO „„ Shorter-term harvest forecasts are needed so as to be able to predict triggers in suf- ficient time (10 months before the end of the next rainy season). „„ Hydro-meteorological data should be defined for a smaller geographic grid and be combined with population data to predict impact. „„ Seamless cooperation between regional and national authorities is critical in using and diffusing data for maximum benefit. Concluding remarks With a few focused, courageous policy decisions, several issues related to the Burkinabe social protection system can be converted into opportunities. Closing the poverty gap „„ Fiscal space is currently lost through governance issues such as absent coordi- nation, resulting in an overlap of programs and insufficient budget execution. That would be financially space can be regained. feasible. The size of the „„ Fiscal space is similarly lost through spending on regressive programs such as sub- actual poverty gap equals sidies and scholarships rather than efficiently targeting social expenditures to the 2.26 percent of gross poor. This fiscal space can also be regained. domestic product—which „„ By regaining the fiscal space as described in the above recommendations, the gov- ernment could cover the country’s poor with an effective, efficient safety net—and is close to actual spending more. on social safety nets. This end is achievable simply by realigning and better targeting existing safety net Thus, with better targeted expenditures. Reallocating regressive subsidies and scholarships would open fiscal space programs, Burkina Faso beyond that. Existing social expenditures can achieve greater impact by linking payments to the utilization of human capital–building services such as education and health. technically would be able to eradicate poverty. Besides offering long-term strategies to alleviate poverty and build human capital, social safety nets can channel a short-term response to shocks against food security. Burkina Faso possesses the requisite building blocks to provide shock-response safety nets. With some improvements, these can be used to help the country face acute climate-induced hardships on a regular basis, providing a transition from humanitarian assistance to national systems. appendix A Burkina Faso’s main social safety net programs A s part of the review process, information was collected on 143 social protection programs currently implemented in Burkina Faso, 123 of which are social safety net programs. Detailed descriptions of the 23 core social safety net programs can be found in this appendix—both cash transfer and in-kind transfer programs. Whenever possible, the program’s objectives, approach, target group, targeting method, benefits, cost, impact, operating institutions, and financing information are provided. The social safety net programs described in this appendix were selected with respect to their size (measured in terms of expenditure and coverage) and targeting mechanisms. Together, these 23 programs account for 94 percent of all social safety net expenditures in the coun- try. The 16 largest social safety net programs that target poor and vulnerable populations in Burkina Faso are presented, as well as 7 smaller safety net programs, as some of them may benefit from an exchange of best practices—or even a consolidation into larger programs. Consolidation would reduce overall implementation costs and create fiscal space, while improving program impact. However, further program evaluation is needed before consolidat- ing. Data and information on these social protection programs were collected using a variety of sources, including National Social Protection Strategy action plans and reports and documen- tation from institutions, donor agencies, and other organizations. Interviews were conducted with key stakeholders to collect additional data and information. 93 94 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO Cash transfer programs Descriptions of Burkina Faso’s main cash transfer programs are presented in this section: two labor-intensive public works programs, two scholarship programs, and three unconditional cash transfer programs. Labor-intensive public works programs Labor-intensive public works programs are usually temporary interventions that condition the cash transfer on participating in community projects/activities. They can be either cash-for- work or food-for-work. Social safety nets and labor market sectors are very much interconnected. It is often safety nets that enable education and health programs to reach the poor, providing the critical sec- toral underpinnings for a well-functioning labor market. Social safety net programs boost economic mobility by increasing resilience and safeguarding against chronic poverty and vulnerability to shocks, and by protecting and investing in human capital and promoting pro- ductive capacities. Access to jobs can increase capacity to self-insure and thereby reduce reliance on safety nets. Traditionally, social protection systems have brought together various instruments to help societies manage human-made and natural risks. Increasingly, the focus is on reducing the risks that people face rather than only managing them. However, less has been done to improve their capacity to self-insure by creating job opportunities. Further developing the links between the labor market sector and social safety nets could provide new opportunities to have a greater impact on poverty and vulnerability and improve standards of living. The aim is to better balance expenditures on programs that reduce risks and improve earnings opportunities through better jobs, as well as expenditures on transfers and other redistributive arrangements. The labor market in Burkina Faso is under high pressure given the country’s population growth. The need to create jobs is addressed by the Labor-Intensive Youth Public Works Program THIMO, a component of the World Bank–funded Youth Employment and Skills Development Project (Projet d’emploi des jeunes et de développement des compétences— PEJDC), and by the Cash for Work social safety net program, among others. The Cash for Work program is one of the largest cash transfer programs in Burkina Faso and reached about 38,000 beneficiaries in 2015, mainly in the Central and Plateau Central regions. Even though the poverty rate is relatively lower in these regions, the country’s rapid urbanization calls for job creation in cities. However, more opportunities should be created in rural areas as well, which is a planned focus of THIMO. These two labor-intensive public works programs are described below. 95 appendix A.  Burkina Faso’s main social safety net programs THIMO for Youth Labor-Intensive Public Works for Youth (Travaux à haute Program title THIMO for Youth (PEJDC component) intensité de main d’œuvre, THIMO) is a component of Category Labor-intensive public works program the Youth Employment and Skills Development Proj- Responsible agency Ministry of Youth, Training and Vocational ect. The objective is to provide immediate temporary Integration; operated by PTR-HIMO (urban) and employment for out-of-school youth (age 16–35). HELVETAS NGO (rural) Target group(s) Out-of-school, low-skilled youth age 16–35; women In urban areas, the program is implemented by the gov- Targeting method(s) Geographical, categorical, self-selection through ernment, while a nongovernmental organization (NGO) application (+lottery) is responsible for rural implementation. The ministry Geographical area(s) Urban (Ouagadougou, Bobo-Dioulasso, Ouahigouya, Manga, Kaïn, Tangaye, Barga, department responsible for all direct job creation pro- Rambo, Béré, Guiba); rural grams in urban areas is the Labor-Intensive Roadwork No. of beneficiaries Target 2014–18: 13,000, 30% women (urban) Project (Projet Travaux Routiers à haute intensité de and 15,000 (rural) main d’œuvre, PTR-HIMO). In six rural townships— Actual 2015: 2,000 (urban) 0 (rural); actual 2016: four in the North region (Kaïn, Tangaye, Barga, and 7,000 (urban) and 1,510 (rural) Rambo) and two in the Central South region (Béré and Average cash 2016: CFAF 37,000, monthly, over six-month transfer period Guiba)—implementation has been delegated to the NGO HELVETAS Intercooperation, whose work is mainly Annual expenditure 2015: CFAF 450 million (1.25% of social protec- tion expenditure in 2015) focused on reforestation. The type of work in urban 2016: CFAF 1,110 million (urban); CFAF 181 mil- areas varies from location to location, but includes gar- lion (rural) bage collection, gutter cleaning, road reconstruction, Funding sources World Bank and cleaning parks and public markets. An environmen- SOURCES: PEJDC (fiche synthetic); CNPS 2016b, 2017b. tal awareness campaign is also envisioned. Beneficiaries receive monthly cash payments over a six-month period, three vaccinations, and work injury insurance. In 2016, the cash transfer was CFAF 37,000 and team leaders received a CFAF 500 daily premium. A voluntary savings plan (less than 10 percent of a worker’s monthly salary), aimed at supporting income-generating activities, is also available. Female participation is encouraged through the creation of jobs attractive to women, and a percentage of jobs are reserved for women. In urban areas, females apply for these jobs to a greater extent than do men, while overall dropout rates are relatively low at 5 percent. Expenditures have increased notably, as has the number of beneficiaries in urban areas. However, due to a lack of appropriate road maintenance equipment, target goals were not fully met. The target level of beneficiaries in urban areas over the project duration (2014–18), is 13,000; in rural areas, it is 15,000. In 2015, only 2,000 participants of a target of 4,000 were recruited in urban areas,000 (CNPS 2016b). Despite an expenditure increase from 450 million in 2015 to 1,110 million in 2016, the program remains below target levels (62.5 percent finan- cial execution). Going forward, THIMO will increasingly focus on creating much-needed job opportunities in rural areas. While recruitment in rural areas began in 2016 in the Central and North regions, program execution has been delayed pending a decision as to the partner firm for public work realization (CNPS 2017b). 96 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO Cash for Work Program The objective of this program is to strengthen productive means. Through community work, for instance, the most vulnerable populations such as those affected by natural hazards are aided. Activities are implemented during the mitigation period, January–June (SE-CNSA 2014). Program title Cash for Work Program A cash-for-work component was incorporated into the Category Labor-intensive public works program National Plan to Support Vulnerable Populations against Responsible agency Ministry of Agriculture and Water Resources Food and Nutritional Insecurity in 2014. The compo- Target group(s) Vulnerable households nent is financed by a different partner each year. It also Targeting method(s) Geographical and community receives government funding. Geographical area(s) Numerous provinces (Banwa, Nayala, Sourou, Kadiogo, Namentenga, Boulkiemdé, Gnagna, Cash for Work uses a combination of geographic and Kourwéo-go, Oubritenga, Oudalan, Noumbiel) community targeting mechanisms. Focus groups are No. of beneficiaries 2014: 685,177; 2015: 226,000; 2016: 38,535 also used to collect information on potential beneficia- Average cash 2014: 50,000; 2015: 72,000 2016: 20,000 ries and household vulnerability. The combined targeting transfer monthly, over three-month period involves a three-step approach: (1) selection of regions Annual expenditure 2014: CFAF 29,970 million; 2015: CFAF 16,272 with high rates of malnutrition and high exposure to food million; 2016: CFAF 2,312 million insecurity; (2) selection of at-risk communities within Funding sources WFP, ECHO, government the selected regions based on a household economy SOURCES: SE-CNSA 2014, 2015, 2016. approach (HEA) targeting methodology; (3) community NOTE: Expenditure estimates are based on an average household size of seven. targeting, using focus groups, within each selected community to collect data (e.g., household size, income, livestock owned/agricultural land, if any) on potential beneficiaries. Beneficiary and financing needs are assessed on an annual basis. In recent years, expen- ditures and number of beneficiaries have rapidly decreased, while the actual wage transfer amount has fluctuated for unknown reasons. The large fluctuations in the amount transferred could reflect a change in tasks supported by the program. It could also indicate improved tar- geting, resulting in fewer beneficiary inclusion errors and lowered wages. Scholarship programs for higher education Scholarships can be an important means to subsidize less well-off students, so they can pursue higher education. However, it is critical to avoid potential inequity by ensuring that scholarships are not captured by the richest quintile. The higher education system in Burkina Faso is confronted with complex challenges. There are currently only four public universities and three polytechnic university centers, which are scheduled to become universities, in the country. Decisions have been made to expand uni- versity access through large amphitheaters, while pursuing a policy of recruiting volunteer teachers. This is done with little regard for quality despite the already existing challenges, including high failure and repetition rates. The government has a history of encouraging uni- versity attendance through its scholarship programs (World Bank 2018c). 97 appendix A.  Burkina Faso’s main social safety net programs The two government bodies that handle scholarships in Burkina Faso are the National Fund for Education and Research (FONER) and the National Center for Information, Educational and Vocational Guidance and Scholarships (CIOSPB). Most national student aid comes from FONER, which provides nonreimbursable grants to new secondary school graduates who are enrolling in any of the universities. It also provides loans for students during their last year of education. CIOSPB provides scholarships for which all new secondary school graduates can apply. Once the scholarship quota has been reached, an additional 300 scholarships are available for female applicants. The scholarship programs in Burkina Faso do not seem to implement household income requirements. Without this, it is difficult to assume that they are pro-poor, especially in the absence of socioeconomic data on beneficiaries. It will be hard for them to have an impact on poverty alleviation and social inclusion unless scholarship programs explicitly consider a household income maximum as a criterion for granting scholarships. The two main scholarship programs for higher education are described below. National Scholarship for Public and Private University Students The objective of this program is to encourage high- Program title National Scholarship for Public and Private University Students er-level education among Burkinabe youth. This program has offered university students different types Category Cash transfer program of scholarships since 2009: scholarships for new bach- Responsible agency Ministry of Higher Education, implemented by CIOSPB elor’s degree students under age 22; scholarships for Target group(s) University students students under age 23, entering their second year of Targeting method(s) Categorical and self-selection through application university; scholarships for students under age 26 enter- Geographical area(s) Nationwide ing their third or fourth year of university (maximum one No. of beneficiaries 2015: 6,400; 2016: 7,300 year repeated). Moreover, students who have a master’s Average cash CFAF 175,000, throughout one full school year degree can apply for specific loans. The number of ben- transfer eficiaries is only known for the 2015 (6,400) and 2016 Annual expenditure 2015: CFAF 6,645 million; 2016: CFAF 6,121 (7,300). million Funding sources Government, national The targeting method is self-selection. Students pursue SOURCES: CNPS 2015b, 2016b. the scholarship on their own by submitting an applica- tion to the implementing institution, CIOSPB. Recent data indicate that student income levels are not considered as part of the selection process—which could possibly mean that nonpoor students reap most of the benefits from the program. A maximum level of household income is not explicitly discussed in project documents, and the only selection criteria seem to be that the applying student be below a certain age and have above-average grades. A means to test income should be incorporated into the selection process. In addition, it is important to collect socioeconomic information— on geographic region of origin, degree pursued, sex, etc.—in order to assess who benefits from the and to avoid inclusion errors, among other things. The program needs to be made 98 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO more equitable, and it should be ensured that it benefits poor students, by considering house- hold income as a criterion for scholarship consideration. This is currently not the case. University Student Financial Aid Program Program title University Student Financial Aid Program The objective of this program is to further extend schol- Category Cash transfer program arships to Burkinabe youth beyond the beneficiaries Responsible agency Ministry of Higher Education, implemented by FONER reached by the National Scholarship Program. Target group(s) 1st year university students Targeting method(s) Categorical and self-selection through application Given the limited coverage of the National Scholarship Geographical area(s) Nationwide Program, this program was created to support and No. of beneficiaries 2015: 44,669 2016: 46,941 encourage additional students in their pursuit of univer- sity education. The program has been in effect since Average cash CFAF 130,000 throughout one full school year, transfer with possibility of extension 2014 and is overseen by the Ministry of Higher Education; Annual expenditure 2015: CFAF 7,817 million; 2016: CFAF 6,270 million FONER is in charge of execution. The program targets Funding sources Government, national first-year university students under the age of 23. SOURCES: CNPS 2015b, 2016b. Unconditional cash transfer programs Unconditional cash transfers can help protect families from falling deeper into poverty, allowing poor households choice and flexibility in allocating resources to meet the needs they find most pressing. In order to help households escape the poverty trap, the government of Burkina Faso and its partners have increasingly focused on supplementing the income of poor households through the provision of unconditional cash transfers. Three of the largest social safety net programs are described below. Unconditional Cash Transfers to Poor in Sahel and Central North The objective of this program is to provide the poor and vulnerable with a safety net, in order to help them avoid falling deeper into the poverty trap. The program provides unconditional cash transfers to Program title Unconditional Cash Transfers to Poor in Sahel and poor and vulnerable populations in the Sahel and Cen- Central North tral North regions. The program is financed by Oxfam Category Unconditional cash transfer program International and Catholic Relief Services (CRS); the Responsible agency Technical Alliance for Development Assistance Technical Alliance for Development Assistance (ATAD) Target group(s) Poor and vulnerable households NGO is responsible for program implementation. Targeting method(s) — Geographical area(s) Sahel and Central North regions In 2015, 2,770 households benefited from the program, No. of beneficiaries 2015: 2,770; 2016: 2,770 and the vast majority of beneficiaries were from the Sahel Average cash — region. Due to insufficient resources, the amount of the transfer transfer was reduced. Annual expenditure 2015: CFAF 91.8 million; 2016: CFAF 65 million (budget data) Program expenditure in 2015 was CFAF 91.8 million. The Funding sources Oxfam, Catholic Relief Services expenditure amount was reduced to CFAF 65 million in SOURCES: CNPS 2015b, 2016b. 99 appendix A.  Burkina Faso’s main social safety net programs 2016 due to insufficient funding resources (CNPS 2015b, 2016a). The 2015 expenditure repre- sented an execution rate of only 26 percent with respect to the planned budget. Burkin-Naong-Sa Ya The objective of this program is to reduce structural poverty in the most vulnerable regions through the provision of income support. It also aims to lay the foundation for a basic safety net system. Program title Burkin-Naong-Sa Ya While cash transfer programs have been present in Category Cash transfer Burkina Faso for several years, they were mostly ad hoc, Responsible agency Ministry of Women, National Solidarity and Family, urgent measures with short-term impact. However, in with World Bank technical and financial support, 2014, the government decided to try a new approach. implemented by a pilot committee With the help of the World Bank, the cash transfer pro- Target group(s) Women, households with children, extremely poor, rural, people with disabilities gram Burkin-Naong-Sa Ya was put in place. The aim Targeting method(s) Geographical and community was to provide poor households with young children Geographical area(s) Rural focus: 3 regions, 7 provinces, 42 com- income support over several years and thereby offer munes, 472 villages them some security and help boost their human capi- No. of beneficiaries 2015, 2016, 2017: 17,063; 2018: 81,558 (women tal—especially in terms of nutrition and early childhood receiving cash) development. Average cash CFAF 10,000 for households of ≤ 5 or less, transfer 13,000 for > 5, per semester, for 3 years The program’s second objective is to establish a basis Funding sources $56 million, World Bank; $50 million IDA loan, $6 for a coherent and coordinated social safety net system million DFID TF 2014–20 in Burkina Faso by building a management information SOURCES: CNPS 2015b, 2016b. system. It is also a starting point for a harmonized and integrated social safety net system, by building a common targeting tool that could be used in the long term, as well as a single national social registry. The program targets poor and vulnerable households with children under age 15. The targeting mechanism relies on an existing system, such as the Early Warning System (Système d’Alerte Précoce, SAP), and other geographical targeting tools to identify geographic areas affected by or vulnerable to food insecurity and other shocks.1 The selection of the zones is in tandem with implementation feasibility, which is constrained by the existence of a social action center, as social workers are key to the success of targeting and in providing transfer knowledge activi- ties. Identifying beneficiary household at the local level involves community targeting. The SAP is fed by several centralized sectoral information systems. A survey on agricultural security 1  is conducted every year on 1,000 people to determine the vulnerability level and caseload (World Bank 2012). 100 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO Unconditional Cash Transfers (Supplement to Subsidy/Cereal Distribution) The objective of this cash transfer program is to help Program title Unconditional Cash Transfers (Supplement to Subsidy/Cereal Distribution) households meet their main food needs. Category Unconditional cash transfer program The program serves as a supplement to the targeted Responsible agency SONAGESS; implemented by NGOs food subsidy program. While the targeted subsidy pro- Target group(s) Poor and vulnerable households gram is implemented in 13 provinces, the cash transfer Targeting method(s) Geographical and community focuses on only two of them—Soum and Oudalan, both Geographical area(s) Soum and Oudalan in the Sahel region. No. of beneficiaries 2015: 27,000; 2016: 18,269 (3,045 households) Average cash CFAF 60,000 per household, in 3 payments over In 2016, the cash transfer to vulnerable households transfer 3 months consisted of three payments of CFAF 20,000 (SE-CNSA Annual expenditure 2015: CFAF 1,620 million (budget) 2016). While the numbers of beneficiaries decreased Funding sources NGOs from 2015 to 2016, the estimated cost also decreased SOURCES: CNPS 2015b, 2016b. more than proportionally. In-kind transfer programs In-kind transfers consist of food rations, clothes, school supplies, shelter, fertilizers, seeds, agricultural tools or animals, and building materials, among others. They are a very common safety net instrument. Following is a review of the major in-kind transfer programs in Burkina Faso pertaining to education (two programs), food distribution/subsidies (four), agriculture (five), and health (six). Education programs School Supplies Program (Cartable Minimum) The objective of this program is to provide primary school students with basic school sup- plies. Program title School Supplies Program (Cartable Minimum) Category In-kind, education Addressing the challenge of decreasing distribution Responsible agency Ministry of National Education and Literacy, of school supplies, the School Supplies Program was implemented by NGOs launched by the Ministry of National Education and Target group(s) Primary school students Literacy, and in 2016, the Framework Association Targeting method(s) — of Consultants and NGOs Active in Basic Education Geographical area(s) 361 select communes across Burkina Faso (CCEB-BF) began gathering information on the distri- No. of beneficiaries — bution of basic school supplies in several communes Transfer School supplies across Burkina Faso. A decrease in the distribution of Annual expenditure 2015: CFAF 3,386 million; 2016: CFAF 3,871 million school supplies was observed, while the number of pri- Funding sources National mary students kept increasing, leaving many students SOURCES: CNPS 2015b, 2016b. without much-needed supplies. 101 appendix A.  Burkina Faso’s main social safety net programs The program targets 361 communes; no information is available on either the selection process or the number of beneficiaries. Expenditures have increased slightly, from CFAF 3,386 million in 2015 to CFAF 3,871 million in 2016. The CCEB is currently conducting a larger data collection effort in order to be better able to inform potential reforms. In addition to primary students receiving insufficient basic supplies, the CCEB has also observed delays in delivery, as well as low quality of the school supplies. All of this will be further researched through a more comprehensive data collection effort with the intention of ultimately informing policy decisions and educational reform. Child Development through Sponsorship Program The objective of this program is to increase children’s Program title Child Development through Sponsorship Program access to education while improving their health. Category In-kind transfer, educational In 2004, the NGO Compassion International introduced Responsible agency The NGO Compassion International (CI), link to national institutions the program, with the aim of improving children’s overall Target group(s) Children and youth well-being through additional educational, extracurricu- lar, and training opportunities—as well as improving their Targeting method(s) Based on poverty mapping by the National Institute of Statistics and Demography for choice nutritional status. Awareness and educational classes of localities and poverty criteria in choice of bene- are offered to parents a few times during the year. ficiaries (CI interview) Geographical area(s) 11 regions The Child Development through Sponsorship Program No. of beneficiaries 2008: 16,795; 2009: 18,424; 2010: 20,474; relies on individual private sponsorships of children. 2011: 26,722: 2012: 34,007; 2013: 44,078; Through the program, children and adolescents are able 2014: 50,698; 2015: 60,526; 2016: 68,243 to attend child development centers, while receiving a Transfer Education, one meal (two for the most vulnerable) over a 4-month period meal. Beneficiaries 11 years old or younger attend child Annual expenditure 2008: CFAF 2.4 billion; 2009: CFAF 2.0 billion; development centers for eight hours a week, usually 2010: CFAF 2.9 billion; 2011: CFAF 3.8 billion; on Thursdays when they do not have primary school. 2012: CFAF 5.0 billion; 2013: CFAF 6.9 billion; Beneficiaries age 12–14 usually attend twice a week 2014: CFAF; 2015: CFAF 10.4 billion; 2016: for a total of four hours; children above the age of 15 CFAF 12.9 billion attend for four hours, on Saturdays only. At the develop- Funding sources CI, which mainly relies on individual donations ment centers, children are provided with breakfast, and SOURCES: World Bank 2014 and CI website. the most vulnerable are also provided with take-home rations. Additional services are offered, including the following: „„ Vocational training. Training in areas such as mechanics, hairdressing, sewing, soap making, gardening, and animal breeding is available for adolescents. „„ Community services. Children also have the opportunity to participate in commu- nity services, such as planting trees and cleaning up at the child development center and in the community. „„ Parenting classes. Two to three times a year, parents are offered classes on hygiene, malaria prevention, reproductive health, and nutrition, as well as training in income-generating activities. 102 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO Food assistance/school feeding programs Food assistance programs are important to improve food security. These programs can pro- vide an important lifeline for the poor and vulnerable. For instance, they are one of the main instruments used to link safety nets to educational attainment objectives. They are also used to remedy nutritional deficiencies and improve overall health of schoolchildren while attracting the maximum number of children to school. Meals are generally provided in poor and food-in- secure areas. Five of the largest food programs in Burkina Faso are described below. Government School Feeding Program (Primary Education) The objective of this program is to improve children’s Program title Government School Feeding Program (Primary Education) health, nutrition, and learning. This nationwide school Category In-kind, school feeding feeding program is the largest social safety net in Responsible agency Ministry of National Education and Literacy, Burkina Faso, representing 13 percent of total social implemented by CRS safety net expenditure. The program primarily aims to Target group(s) Primary school students provide one meal per day to primary school students Targeting method(s) Geographical (of the complementary school feed- over a four-month period, out of the nine months of ing by CRS/WFP only) schooling. It is one of the country’s first safety net pro- Geographical area(s) Nationwide grams, initiated in the early 1960s by CRS. No. of beneficiaries 2015: 2,696,000; 2016: 2,906,000 Transfer Food transfer: one daily meal for 4 months; in The Ministry of National Education and Literacy is in areas with extreme vulnerability, one daily meal charge of the program, while CRS manages implemen- for 9 months plus take-home rations for girls tation. The program is mainly government financed; Annual expenditure 2015: CFAF 19,257 million; 2016: CFAF 20,045 million however, partners provide supplementary financing in Funding sources Mainly government, WFP, and CRS regions that are particularly affected by food insecurity SOURCES: CNPS 2014, 2015b, 2016b, 2017; World Bank 2012. and low literacy rates. CRS complements the government school feeding in five provinces (Bam, Sanmatenga, Namentenga, Gnagna, and Komondjari) through the provision of one meal a day during the five months of school that are not covered by the national program. CRS also provides take- home rations for girls and offers educational sessions on proper hygiene. There are no targeting mechanisms except geographical targeting of the complementary school feeding by CRS and the World Food Programme (WFP). The risk of excluding the poorest seems relatively high, given that school feeding programs may not reach the poorest. It is important to reduce exclusion errors while increasing program efficiency. Conditional cash transfers or additional take-home rations targeted to the poorest could be considered. The trend both in terms of expenditure and number of beneficiaries has been on the rise over the years. However, further analysis of the increase in beneficiaries is recommended, given the rapid population growth driven by high fertility rates, which outpaces the growth rate of beneficiaries. The data provided in table A.1 should be viewed with some caution, as the repartition between national and partner activities is unknown. In 2009, for example, CRS and the WFP contributed approximately 55 percent of expenditure (World Bank 2012). 103 appendix A.  Burkina Faso’s main social safety net programs TABLE A.1  Summary of Government School Feeding Program benefits 2008 2013 2014 2015 2016 Beneficiaries (children) 1,049,000* 2,427,681 — 2,696,000 2,906,000 Expenditure (million CFAF) 7,400* 19,944 19,774 19,257 20,045 Tons of food — — 43,944 — 35,920 SOURCES: CNPS 2014, 2015b, 2016b, 2017; asterisked items are estimates based on World Bank 2012. NOTE: — = not available. 2015 data are budget data given some delays in program delivery—after report publication. The program does not have a monitoring and evaluation component. Evaluations are con- ducted on an ad hoc basis by donors, and these evaluations mainly focus on donor-financed school canteens. Food Assistance for the Vulnerable The objective of this program is to help vulnerable pop- Program title Food Assistance for the Vulnerable ulations cope during times of food insecurity through the Category In-kind, school feeding provision of food transfers (cereals). Responsible agency Ministry of Social Action and National Solidarity; Permanent Secretariat of the National Emergency Since the mid-1990s, the National Society for the Man- Relief and Rehabilitation Council agement of Food Security (SONAGESS) has managed Target group(s) Vulnerable populations the national food stock and provides food assistance Targeting method(s) Geographical and community to tens of thousands of vulnerable people yearly. Food Geographical area(s) Central and Cascades regions (50%) distribution takes place on an ad hoc basis and varies No. of beneficiaries 2015: 300,000 greatly from year to year depending on needs as well as Transfer Food transfer, cereals (maize, mil, rice, sorghum) the remaining stock of food. The national stock consists Annual expenditure 2015: CFAF 2,119 million of 35,000 tons of staple cereals, mainly maize, mil, rice, Funding sources Government via SONAGESS, and partners, includ- and sorghum. Both the government and international ing Japan International Cooperation Agency partners finance and participate in replenishment of the SOURCES: CNPS 2016b; Oxfam 2015; Japan Ministry of Foreign Affairs, 2015. stock. Any region that has a food deficit above 5 per- cent is eligible. Charities submit requests for assistance (World Bank 2011). Vulnerable populations and communes at risk of malnutrition are identified through geograph- ical and community targeting. In 2011, the CNSA published the first operational plan, in which the geographical targeting is based on the SAP and the community targeting method is based on the HEA. In 2015, food was distributed to approximately 300,000 individuals—or 37,000 house- holds—half of which was distributed in two regions (Central and Cascades) with a program expenditure of CFAF 2,119 million (CNPS 2016b). In 2014, only 13,599 households received food transfers; the 2015 increase was possibly due to the government’s Social Measures ini- tiative and subsequent increases in budget allocation. The goal is to distribute 6,000 tons of cereals to half a million individuals. 104 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO Targeted Food Subsidies The objective of this program is to increase food security Program title Targeted Food Subsidies of vulnerable populations during the dry season. This Category In-kind, targeted subsidies is currently the largest food transfer program in place Responsible agency Ministry of Agriculture and Water Resources; SONAGESS in Burkina Faso. The Targeted Food Subsidies program Target group(s) Vulnerable populations was introduced in the early 2000s by the Ministry of Agriculture and Water Resources following the food Targeting method(s) Geographical and community crisis. While the Food Assistance for the Vulnerable Geographical area(s) National program provides free cereal transfers from the national No. of beneficiaries — food stock, this program is a retail program, offering Transfer Subsidized food (cereals) over 3 months; 2013: 8,032 tons; 2015, 35,043 tons; 2016: 37,229 tons cereals at a subsidized price. The aim is to increase food Annual expenditure 2013: CFAF 425 million; 2015: CFAF 7,915 mil- security of vulnerable and poor groups during the dry lion; 2016: CFAF 6,861 million period (May to August). Funding sources — Identification of beneficiaries is made through geograph- SOURCES: CNPS 2013b, 2015b, 2016b. No program information is available for 2014. ical and community targeting, using the HEA methods plan (CNSA 2011). In 2015, the program targeted more than 40 communes, mainly in the Central North and Sahel regions, with an objective of cov- ering 656,137 individuals, or 82,017 households (table A.2). The subsidized price of 50 Kg of cereals is CFAF 6,000. The increase in program expenditures partially stemmed from a willingness to rapidly scale up the program to increase its coverage of beneficiaries, but also due to unexpected costs related to storage during rainfalls. In 2015, 35,043 tons of cereal was sold at the subsidized TABLE A.2  Targeted Food Subsidies target population, by province, 2015 Region Province Communes Population Central East Kourittenga Gounghin, Andemtenga, Baskouré, Kando 10,072 Southwest Noumbiel Legmoin, Midebdo, Batié, Boussoukoula, Kpere 8,917 Namentenga Yalgo, Bouroum, Nabingo 79,515 Central North Sanmatenga As needed based on monitoring of the food and nutritional situation 161,535 Soum Nassoumbou, Diguel, Djibo Koutoukou Baraboulé, Tongomael 163,668 Seno Seytenga, Falangountou, Bani, Sampelga; Gorgaghi, Dori 71,044 Sahel Yagha Sebba, Titabé; Tankougandié, Mansila, Bondoré, Solhan 97,283 Oudalan Oursi, TinAkoff, Déou, Gorom-Gorom, Markoye 12,205 Yatenga Kain, Koumbri 3,702 Passoré Pilimpikou, Latodin, Samba, Kirsi 4,495 North Loroum Sollé, Banh 1,718 Zondoma Gourcy, Boussou 3,307 East Tapoa Namounou, Tansarga, Diapaga 38,676 Total 656,137 SOURCE: SP-CNSA 2015. 105 appendix A.  Burkina Faso’s main social safety net programs BOX A.1  Social Measures program F ollowing political instability, the government implemented a Social Measures program in September 2013, with the objective of improving living conditions in Burkina Faso. Social measures provide additional financing to a number of social protection programs, such as food dis- tribution, THIMO public works and scholarships. The social measures are articulated around seven axes: Revalorization of the remuneration of workers, such as 1 „„ Increasing the housing allowance of mainly civil servants „„ Changing the income tax schedule (from nine to seven installments), reducing rates for lower tranches and exemption for 1st tranches (representing a budget cost of CFAF 3 billion) Consolidation of social safety nets to favor vulnerable groups including 2 „„ Food support for indigent people in Burkina Faso’s 45 provinces „„ Support improvements in living conditions of orphans and vulnerable children in 45 provinces, including the distribution of school supplies „„ Support to the elderly and elderly with disabilities through the provision of health care Job and income creation, including 3 „„ Support to the Cash for Work program „„ Support to the THIMO public works program „„ Training for beneficiaries of public work programs Strengthening national funds, including 4 „„ Support job-creating projects through provision of funds to the Youth Initiatives Support Fund (FAIJ), the Informal Sector Sup- port Fund (FASI), and the Employment Promotion Support Fund (FAPE) Special measures related to prices of high consumption products, including 5 „„ Opening of food shops „„ Price control measures (e.g., to reduce illegal sale of fuel) (not in place anymore) 6 Measures in favor of universities „„ Infrastructure improvements „„ Provision of scholarships 7 Communication (information dissemination regarding the social measures) SOURCE: Adapted from Ministry of Economy and Finance 2014. 106 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO price; this was less than budgeted (50,000 tons) but the expenditure—CFAF 7,915 million— ended up being more than what had been budgeted (CFAF 1,500 million). Recurrent problems include difficulties in reaching certain communes, and logistical, distribution, and transporta- tion challenges. Food Shops (Boutiques témoins) The objective of this program is to improve access to Program title Food Shops (Boutiques témoins) food by opening food shops and selling food at a low- Category In-kind, school feeding er-than-average cost (subsidized). The aim is to improve Responsible agency Ministry of Agriculture and Water Resources; the overall well-being of the population, while managing SONAGESS social tensions. Target group(s) — Targeting method(s) — The Food Shops program was initiated in 2013 as part Geographical area(s) Varies, but mainly urban areas of the Social Measures initiative. The aim was to manage No. of beneficiaries — social tensions stemming from recent political instability Transfer Subsidized food throughout the year by opening and managing food shops in which subsi- Annual expenditure 2015: CFAF 1,585 million; 2016: CFAF 1,638 million dized food products are sold. The food sold in the shops Funding sources Ministry of Industry, Trade and Handicrafts; Japan comes from the intervention stock—another food stock International Cooperation Agency; Oxfam managed by SONAGESS, which is dedicated to emer- SOURCES: CNPS 2014, 2015b, 2016c; Oxfam 2015; JICA 2017; Ministry of Econ- gency intervention in case of shock. The program led to omy and Finance 2014. the direct creation of 1,272 new jobs (Ministry of Econ- omy and Finance 2014). In 2015, 150 shops were in operation at a cost of CFAF 1,585 million; in 2016, 148 shops were in operation, mainly in urban areas, at a cost of CFAF 1,638 million. No new shops were opened in 2015 and 2016, so the cost reflects operating costs. Program expenditure was higher in 2013 (CFAF 11,908 million) and 2014 (CFAF 11,000 million) due to the high fixed costs of setting up stores. Whether the poorest are reached in an efficient matter is questionable. Shops are sparse in remote areas and customers are required to present a national identification card, which may be a constraint for the poorest. Identification is required to ensure transparency and avoid abuse of the purchase of subsidized retail food. Even though the shops are open to all, Oxfam reports that vulnerable populations living in remote areas are not able to reach the stores to the same extent due to their location. A major program challenge, according to an Oxfam 2015 report, is the lack of targeting and legal precision in describing poor and vulnerable groups. The report suggests that this may partially be due to the program being guided by political interests and not part of a coor- dinated strategy. Oxfam recommendations include: (1) further investigate the impact the measure could have on grain producers and traders in order to avoid adverse effects on the cereal sector; (2) review program effectiveness and the resources allocated to it; (3) con- sider establishing selection criteria for the target population; (4) strengthen links between the national reserves and small producers, as well as between the national reserves and local food reserves (Oxfam 2015). 107 appendix A.  Burkina Faso’s main social safety net programs Food Assistance to Vulnerable and Disaster Affected Households The objective of this program is to support vulnerable Program title Food Assistance to Vulnerable and Disaster populations affected by disasters through the provision Affected Households of food transfers. The program is an emergency measure Category In-kind, emergency measure/feeding that aims to provide support (food transfers) to popula- Responsible agency Ministry of Social Action and National Solidarity; implemented by SP-CONASUR, with help of NGOs tions affected by natural disasters, mainly people who Target group(s) Victims of natural hazards, people reliant on sub- rely on subsistence agriculture. Food distribution targets sistence agriculture populations whose agricultural production is affected by Targeting method(s) Not clear, possibly HEA drought, bird attacks, etc. The transfer consists mainly Geographical area(s) Disaster-affected areas of cereals distributed over a three-month period without No. of beneficiaries 2015: 405,429 individuals (57,918 households) any conditions. On average, individuals should receive 2016: 99,493 400 grams of cereal daily (distribution occurrence is not Transfer Food (cereals), over 3 months known). Annual expenditure 2015: CFAF 3,655 million Funding sources Nationally financed, Ministry of Economy and Beneficiaries vary from year to year according to natu- Finance ral disasters. In 2015, two-thirds of beneficiaries were SOURCES: CNPS 2016b; SE-CNSA 2015 and 2016. from the Sahel region, mainly the Soum and Oudalan provinces. In 2016, beneficiaries were mainly from the Oudalan province (communes Oursi, Déou, and Tin Akoff) (SE-CNSA 2015 and 2016). It is not clear what targeting methods are used, if any. While the 2015 CNSA plan did not mention targeting mechanisms, the 2016 CNSA plan stated that beneficiaries are identified using the HEA criteria.2 Program expenditure in 2015 was CFAF 3,655 million, and the number of beneficiaries was 405,429—or 57,918 households. The initial plan was to reach 25,000 individuals. In 2016, however, the number decreased to 99,493 individuals, half of whom were refugees from Mali. Nutrition programs/health fee waivers Fee waivers and targeted subsidies typically subsidize health services or provide access to low-priced food staples to the poor. While progress has been made to decrease malnutrition and food insecurity in Burkina Faso, further efforts by the government and its partners are still needed. Global acute malnutrition is 7.6 percent and chronic malnutrition is 27.3 percent, according to the 2016 Standardized Monitoring and Assessment of Relief and Transitions (SMART) nutrition survey. A total of 132,899 individuals are currently in a state of severe food insecurity. Projections indicate 620,394 people are at risk for severe food insecurity during the lean season in 2018 (Harmonized Framework, November 2017, WFP 2017). Five of the larger nutrition and health fee waiver programs in Burkina Faso are summarized below. The first three fall under Burkina Faso’s Plan to Support Vulnerable Populations against The program is part of the CNSA (SE-CNSA 2015) action plan and is also tracked in the yearly 2  monitoring report of the CNPS (2016b). 108 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO Food Insecurity and Malnutrition, a large-scale initiative of the Ministry of Health’s nutrition department—with support from the WFP and additional international partners, and financial support from the European Commission. Malnutrition Prevention Program The objective of this program is to develop resilience Program title Malnutrition Prevention Program and prevent malnutrition in communities that have not Category In-kind, nutrition program fully recovered from the 2012 food and nutrition crisis. Responsible agency Ministry of Health Nutrition Department, WFP, plus The prevention program consists of blanket supplemen- international partners, such as UNICEF; partners vary from year to year tary feeding for all children age 6–23 months during the lean season to prevent new occurrences of malnutrition Target group(s) Children age 6–23 months and to avoid moderate malnutrition deteriorating into Targeting method(s) Categorical and geographical acute malnutrition. Geographical area(s) — No. of beneficiaries 2016: 1,819,445 children The program usually spans a three-month period (June– Transfer Food transfer, for a period of 3–5 months September); however, in 2016 it was extended to cover Annual expenditure 2016: CFAF 4,093 million six months (April–September). Families can purchase Funding sources WFP, European Commission supplementary feeding, such as locally produced micro- SOURCES: CNPS 2015b, 2016b; WFP 2017. nutrient-enriched flour (WFP 2015, 2017; EC 2016). The program is carried out by the WFP in coordination with the Ministry of Health, Department of Nutrition, and related regional offices. The estimated cost per child per day in 2016 was $0.50, for a ration of 200 grams of SuperCereal Plus. Addi- tionally, take-home rations of SuperCereal Plus are provided to caretakers (156,000) of severely malnourished children who are hospitalized at therapeutic feeding centers in the three regions with the highest levels of malnutrition. During the lean season (June–September), children age 6–23 months in TABLE A.3  Malnutrition Prevention Program estimated cost by region, 2016 the regions with the highest levels of malnutrition (East, North, and Sahel) are provided with rations of Super- Regions Cost (million CFAF) Cereal Plus to prevent acute malnutrition. These are Central West 554 so-called fortified blended foods—a cereal fortified with Southwest 293 micronutrients (WFP 2015, 2017; EC 2016). East 648 Sahel 472 In 2016, expenditure for the malnutrition prevention pro- Central North 594 gram was an estimated CFAF 4,094 million (table A.3), North 572 and the number of beneficiaries age 6–23 months was Central South 290 1,819,445. The cost by region varies from CFAF 290 Boucle du Mouhoun 671 million in Central South to CFAF 671 million in Boucle du Total 4,094 Mouhoun (SE-CNSA 2015 and 2016). SOURCE: SE-CNSA (2016) 109 appendix A.  Burkina Faso’s main social safety net programs Treatment of Severe Acute Malnutrition The objective of this program is to improve retention of children with severe acute malnutrition at in-patient Program title Treatment of Severe Acute Malnutrition treatment centers by providing food rations to care- Category In-kind, fee waiver takers. This program aims to reduce acute malnutrition Responsible agency Ministry of Health Nutrition Department and increase food security by strengthening the capac- Target group(s) Children < 5 suffering from severe acute malnu- ity of households to withstand shocks, by improving trition retention of children with severe acute malnutrition. The Targeting method(s) Categorical and geographical, based on SMART nutrition survey; whenever possible, female- program focuses on children under age five who suf- headed households are prioritized fer from severe acute malnutrition, according to World Geographical area(s) See table A.4 Health Organization (WHO) standards. Targeting and No. of beneficiaries See table A.4 distribution for the activities are carried out by the WFP Transfer Food transfer, for a period of 3–5 months + in coordination with the Ministry of Health’s nutrition training department and regional offices. Annual expenditure See table A.4 Funding sources European Commission, WFP, UNICEF The program uses a geographical targeting method SOURCES: CNPS 2015b, 2016b; EC 2016; WFP 2015, 2017. focusing on the regions with a high malnutrition rate based on the 2012 SMART nutrition survey. Focus areas include the Boucle du Mouhoun and Sahel regions, where the number of children with severe malnutrition is the largest (SE-CNSA 2016). Female-headed households are prioritized (WFP 2015). Additionally, take-home rations (incentives rations) are given to 21,000 caretakers of children receiving in-patient care at therapeutic feeding centers for severe acute malnutrition (WFP 2017). Awareness training is provided in targeted communities to promote behavioral change. Health workers and local NGO staff are trained in nutrition, hygiene, and sanitary care prac- tices. Further summary information about the program is in table A.4. TABLE A.4  Treatment of Severe Acute Malnutrition estimated program beneficiaries and costs, 2016 Number of children with severe acute malnutrition Estimated program cost Region Jan.–June July–Sept. Oct.–Dec. Jan.–June July–Sept. Oct.–Dec. Total cost Central West 5,722 2,435 4,017 343 146 241 730 Southwest 6,230 2,651 4,374 374 159 262 795 East 9,200 3,915 6,459 552 235 388 1,174 Sahel 10,958 4,663 7,694 657 280 462 1,399 Central North 8,818 3,752 6,191 529 225 371 1,126 North 9,600 4,085 6,740 576 245 404 1,226 Central South 5,245 2,232 3,683 315 134 221 670 Boucle du Mouhoun 14,732 6,269 10,344 884 376 621 1,881 Total 7,0503 30,001 49,502 420 1,800 2,970 9,000 SOURCE: SE-CNSA (2016) 110 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO Moderately Acute Malnutrition Treatment The objective of this program is to improve the nutrition Program title Moderately Acute Malnutrition Treatment status of moderately acutely malnourished children Category In-kind, fee waiver under age five and malnourished pregnant and lactating Responsible agency Ministry of Health Nutrition Department, WFP women through the treatment of moderate acute malnu- Target group(s) Children < 5 suffering from moderately acute malnutrition and pregnant/lactating women trition. Targeting method(s) Categorical and geographical based on SMART nutrition survey; whenever possible, female- This program aims to reduce moderate acute malnu- headed households were prioritized trition and increase food security by strengthening the Geographical area(s) See table A.5 capacity of households to withstand shocks—mainly No. of beneficiaries See table A.5 by improving the nutrition status of moderately acutely Transfer Food transfer, for a period of 3–5 months malnourished children and malnourished pregnant and Annual expenditure See table A.5 lactating women. Enrolled child beneficiaries receive Funding sources European Commission, WFP, UNICEF Plumpy’Sup, a lipid-based peanut nutrient supplement, SOURCES: CNPS 2015b, 2016b; WFP 2015, 2017. while women receive SuperCereal, a micronutrient-forti- fied food and vegetable oil. Targeting and distribution for the activities are carried out by the WFP in coordination with the Ministry of Health’s Department of Nutrition and related regional offices. Treatment of moderately acute malnutrition is provided to 348,100 children under age five and 124,000 pregnant and nursing mothers in the seven regions with the highest levels of malnu- trition and food insecurity (table A.5). Nutrition interventions include targeted supplementary feeding for the treatment of moderately acute malnutrition in children and malnourished preg- nant and lactating women (WFP 2017). Treatment for moderately acute malnutrition takes place in regions where the prevalence of global acute malnutrition is above or close to the 10 percent serious WHO threshold, as determined by the 2012 SMART nutrition survey. Awareness training is provided in targeted communities to promote behavioral change. Health workers and local NGO staff are trained in nutrition, hygiene, and sanitary care practices. Additionally, through Food for Assets, the WFP supports the government by working with TABLE A.5  Moderately Acute Malnutrition Treatment estimated program beneficiaries and costs, 2016 Number of children with moderate acute malnutrition Estimated cost Region Jan.–June July–Sept. Oct.–Dec. Jan.–June July–Sept. Oct.–Dec. Total cost Central West 21,294 8,691 13,472 426 174 269 869 Southwest 7,608 3,105 4,813 152 62 96 311 East 22,385 9,137 14,162 448 183 283 914 Sahel 21,786 8,892 13,783 436 178 276 889 Central North 20,292 8,283 12,838 406 166 257 828 North 18,877 7,705 11,942 378 154 239 770 Central South 10,816 4,415 6,843 216 88 137 441 Boucle du Mouhoun 23,194 9,467 14,674 464 189 293 947 Total 209,590 85,547 132,598 2,925 1,194 1,851 5,969 SOURCE: SE-CNSA (2016) 111 appendix A.  Burkina Faso’s main social safety net programs vulnerable populations preventively to create community and household assets that increase food security and help develop resilience against climatic shocks. These activities take place in six regions. Participants received $2.40 per day. The number of working days was reduced from 20 to 15 in order to reach more beneficiaries. The rate is slightly below the daily agri- cultural wage in order to prevent adverse effects on local labor markets. Cash transfers were distributed during the lean season (WFP 2017). Further summary information about the pro- gram is in table A.5. ART (Antiretroviral Therapy) Treatment Patient Program The objective of this program is to provide free antiretro- Program title ART (Antiretroviral Therapy) Treatment Patient viral treatment to people infected with HIV. The program Program is a fee waiver, providing free antiretroviral therapy for Category In-kind, fee waiver people infected with HIV. It falls under the responsibil- Responsible agency Ministry of Health ity of the Ministry of Health and is financed by national Target group(s) Persons with HIV funds as well as through grants from the United Nations Targeting method(s) Categorical Population Fund (UNFPA). Geographical area(s) — No. of beneficiaries 2014: 46,623; 2015: 72,000; 2016: 56,089 Free Basic Health Care Program for Transfer Antiretroviral therapy treatment Pregnant Women and Children Annual expenditure 2014: 5,092 million; 2015: CFAF 5,092 million; 2016: 10,580 Since the government is still in the process of establish- Funding sources National funds and UNFPA grants ing universal health insurance, a free waiver program SOURCES: CNPS 2015b, 2016b. for vulnerable groups was set up in the meantime. The focus is on pregnant women and children up to five years of age. In 2015, the state financed prenatal consultations as well as medication for children affected by malaria in 63 districts (CNPS 2015b). All pregnant women and children affected by malaria were treated. The number of benefi- ciaries is not available, however. Program title Free Basic Health Care Program for Pregnant Women and Children The program cost was higher than expected (financial Category In-kind, free health waiver execution rate of 150 percent)—CFAF 1,294 million in Responsible agency Ministry of Health 2015, as prenatal consultation costs were not taken into Target group(s) Pregnant women; children age 0–5 account in the budget. Targeting method(s) Categorical demand driven Geographical area(s) — At a 2016 Council of Ministers meeting, it was decided No. of beneficiaries — that the Ministry of Health would receive more funding. Transfer Free health care As a result, five additional activities could be added Annual expenditure 2015: CFAF 1,294 million to the program at a cost of CFAF 6.011 million in total Funding sources Ministry of Health (table A.6). SOURCES: CNPS 2015b, 2016b. 112 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO TABLE A.6  Free Basic Health Care Program for Pregnant Women and Children beneficiaries and cost, 2016 Free health care Beneficiaries (women) Cost (million CFAF) Ectopic deliveries in health facilities 656,079 3,000 Delivery of obstructed labor in health facilities 75,107 1,567 Newborn medical emergency care 24,313 63 Cesarean section and laparotomy 14,820 1,294 Prophylactic care for pregnant women 2,567,436 87 Total 3,337,755 6,011 SOURCE: CNPS 2016b. Agricultural programs Low productivity in the agricultural sector, and hence high prevalence of subsistence agri- culture, is addressed by social safety net programs that aim to boost production: subsidy for seeds (the largest agriculture program in terms of expenditure), for fertilizer, support to rain-fed production and endowment of livestock feed. Coverage in terms of numbers of bene- ficiaries is unknown. Fertilizer Subsidy Program This large program has been implemented nationwide Program title Fertilizer Subsidy Program since 2011. It subsidized 11,810 tons of fertilizers at a Category In-kind transfer, agricultural cost of CFAF 5,878 million in 2015. In 2016, the cost of Responsible agency Ministry of Agriculture and Water Resources/ SECNSA/Directorate General of Crop Production the program declined to CFAF 3,800 while the amount Target group(s) Vulnerable farming households of subsidized fertilizers increased to 12,665 tons. In Targeting method(s) Community 2015 and 2016, the level of financial execution was below 55 percent due to fertilizer procurement difficul- Geographical area(s) Nationwide ties (CNPS 2016b, 2017b). While the physical program No. of beneficiaries — execution is relatively good (target of 15,000 tons in Transfer Fertilizers 2015 and 2016), the weak financial execution may be Annual expenditure 2015: CFAF 5,878 million; 2016: CFAF 3,800 million explained by the fact that there are no additional pro- Funding sources Ministry of Economy and Finance gram components, such as monitoring and evaluation. SOURCES: CNPS 2015b, 2016b. The government has, however, expressed an interest in evaluating the program; so far, resources have not been allocated toward this (CNPS 2016a). The program uses community targeting methods, involving the municipalities, village devel- opment committees, and producer organizations, to determine eligible households. However, eligibility criteria are not clear, and the linkage to targeting methods used by other programs is unknown. TABLE A.7  Tons of subsidized fertilizers, 2011–16 The quantity of subsidized fertilizer has largely fluctu- 2011 2012 2013 2014 2015 2016 ated over the years, with a peak in 2014 and the lowest 15,073 18,725 13,176 21,612 12,511 12,665 levels in 2015 and 2016 (table A.7). The amount of fertil- SOURCES: Chantal report; CNPS 2016b, 2017b. izers distributed corresponds to the fertilizer needed for 113 appendix A.  Burkina Faso’s main social safety net programs approximately 2 hectares of land, which is a relatively small amount, but in line with the size of the land lots of the poorest households. Seed Subsidy Program The objective of this program is to safeguard vulnerable Program title Seed Subsidy Program farming populations through the provision of subsidized Category In-kind transfer, agricultural improved seeds. Since 2011, the Ministry of Agriculture Responsible agency Ministry of Agriculture and Water Resources and Water Resources has subsidized genetically modi- Target group(s) Vulnerable populations, female-headed households fied seeds, also called improved seeds, to the benefit of Targeting method(s) Community vulnerable households. In 2016, the goal was to provide Geographical area(s) Rural areas, national 10,000 tons of seeds at a subsidized rate; however only No. of beneficiaries — about 4,140 tons were distributed (41 percent physical Transfer Genetically modified/improved seeds execution rate). The amount of seeds distributed has Annual expenditure 2016: CFAF 2,738 million gradually declined, reaching its lowest level in 2016 Funding sources Ministry of Economy and Finance (table A.8). SOURCE: SE-CNSA 2014, 2015 Due to a lack of financial resources, expenditure in 2016 was below budgeted figures at CFAF 2,734 million, TABLE A.8  Seed Subsidy Program distribution which corresponds to a 20 percent financial execution 2011 2012 2013 2014 2015 2016 rate (CNPS 2017b). Tons 14,228 5,889 8,916 9,191 5,147 4,140 Eligible households are identified through community Million CFAF n.a. n.a. n.a. n.a. 2,550 2,734 targeting methods involving municipalities, village com- SOURCES: Chantal report; CNPS 2016b, 2017b. mittees, and producer organizations. However, eligibility criteria are not clear, and links to targeting methods used by other programs are unknown. The program states that it has a focus on supporting female-headed households. Since there is a lack of data on beneficiary households and on breakdown by region, sex, etc., it is not possible to assess who and how many benefit from the program. Rain-Fed Agriculture Support Program Program title Rain-Fed Agriculture Support Program Category In-kind transfer, agricultural The objective of this program is to increase the resil- Responsible agency Ministry of Agriculture and Water Resources/ ience of vulnerable populations through the protection Directorate General of Crop Production; Food and of their livelihoods. The program distributes agricultural Agriculture Organization inputs, as well as food to vulnerable households— Target group(s) Vulnerable farming households, female-headed households mainly cereal and cowpea seeds. Vulnerable households Targeting method(s) HEA and SAP in areas at risk of food crises receive a bag of 15 kg of Geographical area(s) Rural areas cereal each. In addition, female heads of households No. of beneficiaries — receive 7 kg of cowpea throughout the season. The total amount of cowpea distributed totals 48.5 tons. Transfer Agricultural products (e.g., cowpea seeds) and foods (cereals) Annual expenditure Estimated cost 2015: CFAF 1,245 million Funding sources — SOURCE: SE-CNSA 2014, 2015 114 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO In all, 478,745 people are at risk of food crisis, and there is a need to increase distributions; 1,104.6 tons of seed are required to meet the need. Livestock Feed Support Program The objective of this program is to help secure vul- Program title Livestock Feed Support Program nerable farming households against food insecurity Category In-kind transfers, agricultural and malnutrition through the provision of food for their Responsible agency Ministry of Animal and Fisheries Resources, livestock. By distributing cattle feed to vulnerable Directorate General of Crop Production; SONAGESS households, this program secures them against food Target group(s) Vulnerable farming households insecurity. Targeting method(s) SAP (for municipalities); HEA (for households) Geographical area(s) Rural areas In 2015, the following products were distributed free of No. of beneficiaries — charge for the poorest and subsidized for others: Transfer Livestock food (subsidized + free) approximately a total of 76,000 tons in 2015 15,000 tons of agro-industrial by-products for „„  Annual expenditure October–June 2015: CFAF 3,820 million; April– 332,357 cattle (10 percent of the total number of June 2016: CFAF 777 million animals exposed) Funding sources Ministry of Economy and Finance and partners. Details on partners could not be found. 11,000 tons of bran for ~720,000 sheep (20 percent ​​ „„  SOURCE: SE-CNSA 2015. of the total number of animals exposed) 50,000 tons of roughage for ruminants (SE-CNSA „„  2015) (free of charge in the province of Soum) TABLE A.9  Livestock Feed Support Program (tons) Annual expenditure in 2015 was approximately Province SPAI Bran Roughage CFAF 3,820 million. This was higher compared to previ- Bam 530 679 1,767 ous years, due to a temporary extension of the annual Gnagna 2,378 1,100 7,928 program duration period—seven months instead of the usual three (SE-CNSA 2014 and 2015). Estimated Loroum 462 501 1,540 program expenditure for 2016 was CFAF 777 million Namentenga 881 956 2,938 (according to the 2016 plan); as in previous years, the Oudalan 1274 839 4,248 duration period was three months. Passoré 425 671 1,416 Sanmatenga 891 1,443 2,972 The quantities and products delivered by province in Séno 4,594 1,534 15,316 2015 are detailed in table A.9, amounting to approxi- Soum 1,485 1,159 4,952 mately 76,000 tons of food. In 2016, quantities decreased Yagha 1,051 369 3,505 as a reflection of the shorter duration of the program. Yatenga 861 1,238 2,872 Zondoma 118 350 394 Total 14,956 10,846 49,853 SOURCE: SE-CNSA 2015. appendix B HEA in Burkina Faso I n October 2013, the government of Burkina Faso mandated the Ministry of Agriculture and Food Security to establish a national list of vulnerable people in the country, focusing mostly on food insecurity. A technical committee—comprised of nongovernmental orga- nizations (Action Against Hunger, Oxfam), United Nations agencies (Food and Agriculture Organization of the United Nations, World Food Programme), and government entities—devel- oped a household economy approach (HEA) targeting methodology, which was endorsed by the government in December 2013 with the objective of establishing a nationwide list of vul- nerable people. An HEA assumes that households living in a particular zone are relatively heterogeneous in terms of their access to food and income. Different households are more or less able to use local options to obtain food and income, such as access to capital and goods, education, job opportunities in the household, and so on. To capture these variations, the HEA character- izes typical households in each livelihood zone according to four socioeconomic categories: very poor, poor, medium, and rich. As the socioeconomic categories are defined with local informants, poverty or wealth are thus related to local standards, constituting a community approach process. In practice, HEA targeting relies on community knowledge and perceptions to identify the characteristics of the four socioeconomic groups in order to classify households based on unique criteria for the livelihood zones by community rather than nationally. (figure B.1) This targeting in Burkina Faso comprises several steps, some of which entail other targeting meth- ods: „„ Division of the country into different livelihood zones in which households will be compared „„ Initial village general assembly to provide information, adapt ranking criteria, and set up committees (targeting, validation, complaint) „„ Pretargeting of households 115 116 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO FIGURE B.1  Livelihood zones as determined by HEA methods in Burkina Faso South tubers and cereals Southwest fruits, cotton, and cereals West cotton and cereals West cereals and remittances Central Plateau cereals and market gardening Ouagadougou peri-urban North and East livestock and cereals North transhuman pastoralism and millet Southeast cereals, livestock, forestry and fauna Province capital Province National road Country border SOURCE: HEA Sahel, https://hea-sahel.org/burkina-faso/r-profil-hea. „„ Listing of beneficiaries (in Excel format) „„ Verification of data through home visits „„ Validation of the final list Program eligibility using HEA is based on whether a household is classified as deprived or not in various vulnerability areas, including food (number of daily meals, quality/diversification of diet, number of months of coverage of cereal needs), possession of productive and nonpro- ductive assets (land, nature and size of herd, agricultural equipment, means of transport, etc.), and sociodemographic characteristics (household size, age, etc.). Thresholds differ across villages, as the purpose of the method is to reflect local characteristics. The list of potential beneficiaries is validated by the local community. appendix C PMT in Burkina Faso P roxy means test (PMT) targeting has been used around the world and for various types of programs (conditional cash transfers, public employment programs, vocational train- ing, food stamps, etc.). Instead of measuring expenditures (which are hard to capture), PMT generates a proxy for household welfare using easy-to-verify household characteristics, such as socioeconomic characteristics, household demographic structure, location, qual- ity and sanitation of dwellings, and ownership of assets and durable goods (del Ninno and Bradford 2015). It uses statistical inference to define the variables and weights to be used to generate as eligibility criteria—that is, the model predicts household welfare using survey data. To be relevant, the characteristics must be highly correlated with the poverty level of households, easily measurable, and difficult to manipulate by household. The following steps are entailed in PMT implementation: „„ Identification of the determinants of poverty in the country (at the local level) based on representative and reliable household data. The data should include relevant vari- ables to measure and predict poverty with a regression analysis. „„ Calculation of the PMT score based on the best predictors of poverty and their weights derived from the previous regression. „„ Definition of a cut-off PMT score that gives the poverty threshold under which a household will be eligible for the program. The threshold may either be absolute and aligned with international standards of poverty or relative to a particular population for which a certain share will be selected. The cash transfer program Burkin-Naong-Sa Ya uses a PMT formula based on EMC 2010 data. Beyond categorical targeting using easily observable characteristics (focusing on preg- nant women and children age 0–15), the PMT of the Burkin-Naong-Sa Ya program aims to further restrict eligibility to those in need. The method was first tested in 2014 in Gourcy and Diébougou. 117 118 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO The variables used in the PMT formula include household characteristics, housing conditions, and assets, which are long-term determinants and correlates of poverty along multiple dimen- sions. The formula generates a score that allows ranking of households by level of poverty. Every household with a mother and children under age five are therefore assigned a PMT score, which determines its eligibility for the program.1 The list of identified beneficiary house- holds is then validated during the community validation process. In 2018, more than 130,000 households were surveyed: „„ North (Yatenga-Zondoma): 23,237 „„ North (Loroum-Passoré): 10,066 „„ East: 41,783 „„ Central East: 52,492 „„ Central West: 5,561 The PMT score must be lower than −0.60414285714286. 1  appendix D social safety net coverage by region and share of poor Social safety net Government support in Region coveragea case of shocks Scholarships Food distribution Share of poorb North 2.5 0.9 0.0 0.0 70.3 Boucle du Mouhoun 0.8 0.5 0.0 3.9 59.5 Central West 3.3 3.1 0.1 0.0 51.6 East 1.9 0.8 0.5 0.2 49.8 Central North 5.9 2.1 0.4 0.0 46.8 Plateau Central 0.3 0.1 0.2 0.1 45.8 Southwest 0.7 0.7 0.0 0.3 41.1 Central South 0.3 0.3 0.0 0.0 40.3 Central East 0.9 0.7 0.0 0.1 36.5 Hauts-Bassins 1.1 0.5 0.0 0.7 34.2 Cascades 0.3 0.0 0.0 0.2 22.5 Sahel 3.4 1.1 0.0 0.0 20.7 Central 6.6 0.3 5.5 0.2 9.7 SOURCE: Authors’ calculations based on EMC 2014. a. Darker shading indicates greater social safety net coverage. b. Darker shading indicates larger share of the poor. 119 appendix E glossary of social protection terms Social safety net.1 Social safety nets, also called social assistance, refer to noncontributory transfer programs targeted in some way to the poor or vulnerable. Safety nets aim to increase households’ consumption—either directly or through substitution effects—of basic commod- ities and essential services. They are designed to provide regular and predictable support to poor families. A distinctive feature is their noncontributory character; beneficiaries do not have to pay or contribute financially to receive benefits. Social safety nets can be classified into the following categories: „„ Cash transfers (conditional and unconditional), such as family benefits, scholarships, emergency cash support, and social pensions „„ In-kind transfers, such as school feeding, food distribution, and school supply programs „„ Fee waivers, to reduce the cost of access to education, health, or transport ser- vices „„ Social care services, which include institutional or community-based services such as family support (child and foster care, etc.), and support to people with dis- abilities and the elderly (home visits, centers, etc.) Social insurance. Social insurance programs are contributory with the overall objective of minimizing the negative impact of economic shocks on individuals and families. Participants receive benefits or services in exchange for contributions to a social insurance scheme. They include publicly provided or mandated insurance plans against old age, disability, death of the main household provider, and sickness. The definitions presented here are based on Beegle, Coudouel, and Monsalve (2018); Grosh et al. (2008); 1  120 and World Bank (2015, 2018d). 121 appendix E.  glossary of social protection terms Labor market programs. The objective of these programs is to improve the ability of households to provide for themselves through work via the development of fair labor policies. These policies can either be active, addressing job search assistance, training, apprentice- ships, and job subsidies; or passive, covering unemployment and retirement assistance. Government price subsidies. Government subsidies keep prices for consumers below market levels, and thus benefit households through lower prices. Government price subsidies are universal in the sense that all consumers have access to the same commodities at the same price. Risks and vulnerabilities. These terms are sometimes used interchangeably because people or households that are at risk are usually considered to be vulnerable and vice versa. Risk is the probability of a loss or injury. Its potential negative impact can occur as the result of some present process or future event. In the context of poverty analysis, the World Bank defines vulnerability as the probability or risk of being in poverty today or of falling into deeper poverty in the future. Vulnerability is a key dimension of welfare, since a risk of large changes in income can prompt households to reduce their investments in productive assets and human capital. Because vulnerability can influence household behavior and coping strategies, it is an important factor to consider when devising poverty reduction policies. For example, the fear of bad weather conditions or of being expelled from the land they cultivate can deter vulnerable households from investing in riskier but higher-productivity crops and thus reduce their capacity to generate income. references Akresh, Richard, de Walque, Damien, and Harounan Kazianga. 2012. “Alternative Cash Transfer Delivery Mechanisms: Impacts on Routine Preventative Health Clinic Visits in Burkina Faso.” Policy Research Working Paper No. 5958. World Bank, Washington, DC. —. 2013. “Cash Transfers and Child Schooling Evidence from a Randomized Evaluation of the Role of Conditionality.” Policy Research Working Paper No. 6340. World Bank, Washington, DC. —. 2016. “Evidence from a Randomized Evaluation of the Household Welfare Impacts of Conditional and Unconditional Cash Transfers Given to Mothers or Fathers.” Policy Research Working Paper No. 7730. World Bank, Washington, DC. ARSE (Autorité de Régulation du Sous-secteur de l’electricité). 2017. “Rapport annuel d’activités 2017 - ARSE.” ARSE, Ouagadougou. Bandiera, Oriana, Robin Burgess, Narayan Das, Selim Gulesci, Imran Rasul, and Munshi Sulaiman. 2017. “Labor Markets and Poverty in Village Economies.” Quarterly Journal of Economics 132 (2): 811–70. Banerjee, Abhijit V., and Esther Duflo. 2011. “Challenges of World Poverty.” MIT OpenCourseWare 14.73. Beaton, Christopher, Lucky Lontoh, and Matthew Wai-Poi. 2017. “Chapter 4. Indonesia: Pricing Reforms, Social Assistance, and the Importance of Perceptions.” In Gabriela Inchauste and David G. Victor, eds., The Political Economy of Energy Subsidy Reform. Directions in Development. Washington, DC: World Bank. Beegle, Kathleen G., Aline Coudouel, and Emma Monsalve, eds. 2018. Realizing the Full Potential of Social Safety Nets in Africa. Washington, DC: World Bank. Cherrier, Cécile, Carlo del Ninno, and Setareh Razmara. 2011. “Burkina Faso: Social Safety Nets.” Social Protection & Labor Discussion Paper No. 1403. World Bank, Washington, DC. Chuhan-Pole, Punam. 2012. “Who Benefits from Fuel Price Subsidies?” World Bank blog, Africa Can End Poverty. April 18. 122 123 references CNPS (National Council for Social Protection). 2015a. “Plan d’actions opérationnel 2015–2017 de la politique nationale de protection sociale.” CNPS, Ouagadougou. —. 2015b. “Rapport bilan 2014 de mise en œuvre de la politique nationale de protection sociale.” CNPS, Ouagadougou. —. 2016a. “Plan d’actions opérationnel 2016–2018 de la politique nationale de protection sociale.” CNPS, Ouagadougou. —. 2016b. “Rapport bilan 2015 de mise en œuvre de la politique nationale de protection sociale.” CNPS, Ouagadougou. —. 2017a. “Plan d’actions opérationnel 2017–2019 de la politique nationale de protection sociale.” CNPS, Ouagadougou. —. 2017b. “Rapport bilan 2016 de mise en œuvre de la politique nationale de protection sociale.” CNPS, Ouagadougou. del Ninno, Carlo, and Mills Bradford, eds. 2015. Safety Nets in Africa: Effective Mechanisms to Reach the Poor and Most Vulnerable. Washington, DC: World Bank. Durrant, J., and R. Ensom. 2012. “Physical Punishment of Children: Lessons from 20 Years of Research.” Canadian Medical Association Journal 184 (12): 1373–77. EMC (Continuous Multisectoral Survey). 2014. “Burkina Faso Continuous Multisectoral Survey 2014.” Global Health Data Exchange. FAO (Food and Agriculture Organization of the United Nations). 2015. “Revue des politiques, stratégies et programmes nationaux visant à renforcer la sécurité alimentaire et réduire la pauvreté rurale en rapport avec la protection sociale.” Government of Burkina Faso. 2010. “Stratégie de croissance accélérée et de développement durable (SCADD) 2011–2015. Canevas de rédaction.” Ouagadougou. Grosh, Margaret, Carlo del Ninno, Emil Tesliuc, and Azedine Ouerghi. 2008. For Protection & Promotion: The Design and Implementation of Effective Safety Nets. Washington, DC: World Bank. IMF (International Monetary Fund). 2016. “Burkina Faso.” IMF Country Report No.16/390. IMF, Washington, DC. Inchauste, Gabriela, and David G. Victor, eds. 2017. The Political Economy of Energy Subsidy Reform. Directions in Development. Washington, DC: World Bank. Kazianga, H., de Walque, D., and H. Alderman. 2012. “Educational and Child Labour Impacts of Two Food-for-Education Schemes: Evidence from a Randomised Trial in Rural Burkina Faso.” Journal of African Economies. —. 2014. “School Feeding Programs, Intrahousehold Allocation and the Nutrition of Siblings: Evidence from a Randomized Trial in Rural Burkina Faso.” Journal of Development Economics 106: 15–34. 124 THE WAY FORWARD FOR SOCIAL SAFETY NETS IN BURKINA FASO MICA-MEF (Ministry of Industry, Trade and Handicrafts and Ministry of Economy and Finance). 2014a. “Arrêté n°2014-012 conjoint MICA/MEF portant énumération des éléments de la structure de prix du Gaz butane et fixation de ses prix.” MICA-MEF, Ouagadougou. —. 2014b. “Arrêté N°2014-.006/MICA/MEF, portant énumération des éléments de la structure de prix du Gaz Butane de ses prix.” MICA-MEF, Ouagadougou. —. 2014c. “Arrêté N°2014-.009/MICA/MEF, portant énumération des éléments de la structure de prix du Gaz Butane de ses prix.” MICA-MEF, Ouagadougou. Ministry of Economy and Finance. 2014. “Communication introductive: Synthèse du bilan de mise en Œuvre des mesures sociales.” Concertations nationales sur les mesures sociales. Ministry of Economy and Finance, Ouagadougou. Ministry of Environment and Fishery Resources. 2015. “Burkina Faso National Climate Change Adaptation Plan.” Ministry of Environment and Fishery Resources, Ouagadougou. Ministry of Labor and Social Security. 2007. “Politique nationale de protection sociale.” Ministry of Labor and Social Security, Ouagadougou. SE-CNSA (Executive Secretariat of the National Council for Food Security). 2015. “Plan de soutien aux populations vulnérables a l’insécurité alimentaire et nutritionnelle.” SE-CNSA, Ouagadougou. —. 2016. “Plan de soutien aux populations vulnérables à l’insécurité alimentaire et à la malnutrition.” SE-CNSA, Ouagadougou. Schnitzer, Pascale. 2016. “How to Target Households in Adaptive Social Protection Systems? Relative Efficiency of Proxy Means Test and Household Economy Analysis in Niger.” Social Protection and Labor Discussion Paper No. 1612. World Bank, Washington, DC. Smith, Stephen C. 2006. “Poverty Traps and Global Development.” The Globalist. UCP (Unité de Coordination de Projet). 2016. “Projet Emplois des Jeunes et Développement des Competences—Bilan d’exécution au 31 décembre 2016.” Ouagadougou. UNDP (United Nations Development Programme). 2016. Human Development Report 2016: Human Development for Everyone. New York: UNDP. UNICEF. 2014. “Early Childhood Development: A Statistical Snapshot. Building Better Brains and Sustainable Outcomes for Children.” UNICEF Early Childhood Development Section, New York. World Bank. 2012. Managing Risk, Promoting Growth: Developing Systems for Social Protection in Africa. The World Bank’s Africa Social Protection Strategy 2012–2022. Washington, DC: World Bank. —. 2014. “Reducing Poverty and Investing in People: The New Role of Safety Nets in Africa.” Washington, DC. 125 references —. 2015. The State of Social Safety Nets 2015. Washington, DC: World Bank. —. 2016a. “Burkina Faso: Poverty, Vulnerability, and Income Source.” Report No. 115122. World Bank, Washington, DC. —. 2016b. “Ghana: Social Protection Assessment and Public Expenditure Review.” World Bank, Washington, DC. —. 2016c. “Program Document for a Proposed Grant and Credit to Burkina Faso for the First Energy and Fiscal Management Development Policy Financing.” World Bank, Washington, DC. —. 2017. Burkina Faso: Priorities for Poverty Reduction and Shared Prosperity: Systematic Country Diagnostic. Washington, DC: World Bank. —. 2018a. “Burkina Faso, Boosting Rural Incomes to Break the Cycle of Intergenerational Poverty.” Feature Story October 15. World Bank, Washington, DC. —. 2018b. “Burkina Faso.” Human Capital Index Brief. World Bank, Washington, DC. —. 2018c. “Burkina Faso Higher Education Support Project Project Information Document/Integrated Safeguards Data Sheet (PID/ISDS).” Washington, DC. —. 2018d. The State of Social Safety Nets 2018. Washington, DC: World Bank. —. n.d. “Safety Nets How To: Monitoring and Evaluation.” Fact sheet. World Bank, Washington, DC.