97949 1 Overview Myanmar Social Protection Notes Series Promoting resilience, equity and opportunity in Myanmar: The role of social protection1 M yanmar is a country in transition and faces the following main challenges: a) reducing the incidence of poverty and improv- ing human development outcomes, particularly for the poor and vulnerable; and b) addressing the insecurity of incomes and devel- oping mechanisms to reduce exposure to risks and ability to cope with ill-health, disasters, and other shocks. Social protection policies and programs can play an important role in Myanmar’s strategy for poverty reduction and people-centered de- velopment. Many countries have used a range of programs to achieve their development goals, particularly on poverty reduction and inclusive growth. An effective social protection system can help the poor and vul- nerable better manage the opportunities and risks arising from the ongo- ing reforms, and can help promote household and community resilience and social cohesion. However, the present social protection system does not adequately ad- dress the needs of its people. While social protection policies have re- cently been formulated, the process is fragmented across ministries and there are important gaps: • Despite recent increases, government spending on social assistance is only 0.02 percent of gross GDP and reaching only 0.1 percent of 1. This Note was prepared by Puja Vasudeva Dutta (World Bank) with inputs from Mariana Infante-Villarroel and Inge Stokkel (World Bank). Comments were received from Abdoulaye Seck, Carlo del Ninno, Christian Bodewig, Habib Rab, Jehan Arulpragasam, Niko- las Mynt, Philip O’Keefe, and Ruslan Yemtsov (World Bank). This Note provides an overview of the technical notes prepared in the Myanmar Social Protection Notes Series. Any comments and questions can be addressed to pdutta@worldbank.org. The team is grateful to the Ministry of Social Welfare, Relief and Resettlement and to the Ministry of Livestock, Fisheries and Rural Develop- ment for their inputs and facilitation of field trips for the entire assessment; and to the Ministry of Labor, Employment and Social Security, the Ministry of Finance and the Ministry of National Planning and Economic Development for their inputs and guidance. The team benefited from contributions and field support from WFP, ILO, UNOPS-LIFT, SC, HAI, ActionAid, IOM, MDRI, and several UN agencies and NGOs throughout the process. The team is grateful to the Rapid Social Response program and its five donors the Russian Federation, Sweden, Norway, the United Kingdom and Australia for funding this assessment. 2 Building Resilience, Equity and Opportunity in Myanmar: The Role of Social Protection the population. This is well below the global average: on average, low-income coun- tries spend 1.1 percent, while African countries spend 1.7 percent. To date, develop- ment partners (DPs) have largely been responsible for social assistance provision. This is slowly changing with the recent introduction of government pilot programs to ensure equitable and affordable access to health and education. Preliminary evi- dence indicates these are meeting the desired objectives. • Government spending on social security provisions for workers in the formal sector is about 0.6 percent of GDP, reaching just about 3 percent of the population. While coverage and benefit levels are low, the proposed reforms raise concerns of fiscal sus- tainability. The proposed provisions carry the risk of crowding out social assistance spending while further straining developing delivery systems. In the short to medium term, it will be important to start putting in place the core building blocks of an effective, responsive, and sustainable social protection system for Myanmar. These include the following: • A high-level policy coordination mechanism for policy coherence and sustainable financing. This is critical to avoid the risk of costly and unsustainable parallel systems being developed for social security provision. It is also important to support social assistance expansion by pooling government and DP resources, and by strengthening the link between social protection and poverty reduction. • A substantial ‘social assistance block’, with the progressive expansion of social as- sistance coverage transitioning from DP-led to government-led social assistance provision. This would include scaling up recently introduced pilot programs (such as stipends), as well as the introduction of new pilot programs that can address unmet needs. The most likely candidate for the latter is a PWP, given its dual objectives of employment generation and asset creation, and the positive experience of DPs in implementing PWPs in Myanmar. • A modest ‘social security block’ in the short to medium term, as the demographic outlook in Myanmar provides the space for a gradual expansion of social security coverage. A gradual transition will also allow the space to build systems and capac- ity for managing and investing funds well, to avoid compromising the longer-term finances of the system (especially pensions). In the long term, gradual integration and coverage expansion in a sustainable manner. Developing a social protection system can bring important payoffs to Myanmar’s de- velopment agenda. An effective social protection system will provide adequate coverage of beneficiaries in need of assistance, improve responsiveness of programs to accommo- date those made newly vulnerable because of systemic shocks, and effectively connect beneficiaries of social protection programs to additional productive opportunities. Overview 3 1. Overview A reformist administration in Myanmar has cial cohesion by ensuring poor and vulnerable embarked on rapid economic and political groups benefit from economic prosperity. change that has the promise to unlock the country’s tremendous development poten- This Note examines opportunities to develop tial. In 2011, Myanmar started to move from an effective, responsive, and sustainable social military rule to democratic governance, from protection system that contributes to Myan- armed conflict to peace, and from a centrally mar’s development goals. Section 2 examines controlled and closed economy to one that can the patterns of poverty and vulnerability in end poverty and promote inclusive growth. Myanmar; this analysis suggests there is indeed Against this backdrop of broader social and a role for social protection as a core component political reforms, the country’s development of the country’s strategy for poverty reduction path has shifted towards opening and integra- and people-centered development. tion and towards inclusion and empowerment (World Bank, 2015b). Section 3 provides evidence from international experience to identify specific opportunities National policy frameworks identify people- to promote resilience, equity, and opportunity centered development and poverty reduction in Myanmar through social protection policies as a central objective on the development and programs. agenda and suggest the use of social protec- tion to help achieve it. The long-term Nation- Section 4 provides a framework for building an al Comprehensive Development Plan (NCDP) effective and sustainable social protection sys- identifies promoting human development and tem. It then reviews current social protection- poverty reduction as one of its strategic objec- related policies and programs in Myanmar, as tives. The Framework for Economic and Social well as their institutional arrangements and fi- Reform (FESR) also identifies various social pro- nancing, using this framework. tection instruments as a necessary component of Myanmar’s reform process (Planning Com- Section 5 makes recommendations to develop mission, 2013). an effective social protection system for Myan- mar, with key considerations for the short to Many countries have used social protection medium term as well as the long term. to achieve broader development goals, par- ticularly on poverty reduction and inclusive Section 6 concludes with a summary of the growth. Globally, social protection programs main messages and considerations for the have been found to help households and com- short term. munities build resilience to disaster and other risks; contribute to equity by preventing and alleviating economic and social vulnerabilities; and create opportunity through building com- munity assets and enabling households to in- vest in their children’s health and education and in productive investments. Social protec- tion helps reduce poverty and increases so- 4 Building Resilience, Equity and Opportunity in Myanmar: The Role of Social Protection 2. A profile of risk and vulnerability in Myanmar: identifying policy priorities2 Myanmar is a country in transition with great regional diversity. It is still a relatively young country, with the highest share of its population at active working age. Its more pressing needs are the following: a) reducing the incidence of poverty and improving human devel- opment outcomes, with a particular emphasis on reaching the poor and vulnerable. The poor fare worse when it comes to health, nutrition, and education outcomes; and b) ad- dressing the insecurity of incomes and developing mechanisms to reduce exposure to risks and ability to cope with ill-health, disasters, and other shocks. 2.1 Overview Zone, and around the national average in the Delta and Hill Zones. Rakhine (in the Coastal Zone) and Ayeyarwaddy (in the Delta) both have high poverty incidence and a large share Despite recent gains, a substantial proportion of the poor. However, the relatively high pop- of the population suffers from food insecurity ulation density of the Dry Zone and the Delta and poverty. Several socioeconomic and hu- means they are home to nearly two-thirds of man development indicators suggest standards the poor in the country. Similar patterns can be of living in Myanmar lag behind those in oth- seen for other human development indicators er countries in the East Asia region. Although (World Bank, 2015b). many people in Myanmar face constraints in accessing basic services, these are exacerbated for the poor. Three characterizations of pover- ty emerge: a) the ‘traditional’ poor, who have limited education, work long hours for little 2.2 Risks and vulnerabilities pay in agriculture or allied activities, and lack along the lifecycle the training and skills necessary for jobs out- side agriculture; b) the ‘transition’ poor, many of whom have transitioned out of agriculture Individuals and households in Myanmar are and live in small towns and cities, but still work exposed to a variety of risks, increasing their in low-skill, casual employment; and c) the ‘ex- vulnerability to poverty and hardship. Several cluded’ poor, who typically do not have access risks disproportionately affect people during to opportunities for productive economic activ- certain parts of the lifecycle, while others affect ity, often by virtue of their location, ethnicity, everyone regardless of age. gender, or other factors (World Bank, 2015b). Myanmar is still a relatively young country, Myanmar is a culturally, ethnically, and geo- with the 20-24 age group being the most graphically diverse country, and there are sub- populous. Although Myanmar will begin age- stantial disparities in living standards across ing quickly in coming decades, the share of states and regions.3 Incidence of poverty is the working-age population is not expected to highest in the Coastal Zone, lowest in the Dry shrink until after 2040 (World Bank, 2015c). 2. See Note on ‘Risks and vulnerabilities along the lifecycle: Role for social protection in Myanmar’. 3. The four broad agro-ecological zones are as follows: a) Hill: Chin, Kachin, Kayah, Kayin, Shan, and Sagaing; b) Dry: Bago, Magwe, and Mandalay; c) Delta: Yangon and Ayeyarwaddy; and d) Coastal: Rakhine, Mon, and Tanintharyi. Overview 5 Despite improvements, child malnutrition in Myanmar remains high; young children from poorer households fare worse. In 2009/10, 23 percent of children were underweight, 35 per- cent were stunted, and 8 percent wasted (UNI- CEF and MNPED, 2010).4 As many as 33 per- cent of children from households in the poorest wealth quintile were underweight compared with only 14 percent of children from the rich- est quintile (see Figure 1). There are also signifi- cant differences in access to health care before and during delivery. Most children in Myanmar are enrolled in pri- mary school, but many children, especially those from poorer families, drop out of school in the transition between primary and second- ary school (see Figure 1). Only 29 percent of children from households in the poorest quin- tile were enrolled in secondary school, com- pared with 80 percent of those from the richest quintile. Financial constraints play an impor- tant role (Enlightened Myanmar Research and Figure 1: Children from poorer households fare worse Notes: Nutrition outcomes – % of children under 5 malnourished; access to health care – % of pregnant women who have given birth in the two years preceding the survey, by wealth quintile; enrollment and dropouts – % of official school- age children enrolled in the corresponding level of school in the year preceding the survey, by wealth quintile. Source: Nutrition and access to health care – UNICEF and MNPED (2010); education – staff calculations using 2009/10 Multiple Indicator Cluster Survey (MICS) data (UNICEF and MNPED, 2010). 4. These refer to anthropometric indicators for children: underweight is low weight-for-age, stunting is low height-for-age, and wast- ing is low weight-for-height, where ‘low’ is median minus two standard deviations, based on the World Health Organization (WHO) standard and using Multiple Indicator Cluster Survey (MICS) 2009/10 data (UNICEF and MNPED, 2010). 6 Building Resilience, Equity and Opportunity in Myanmar: The Role of Social Protection Figure 2: Most workers are employed in the informal sector Notes: % of individuals aged 15-59, by sector of employment and rural/urban location, and by employment status and consumption quintiles. Source: Staff calculations using 2009/10 Integrated Household Living Conditions Assessment (IHLCA) data (IHLCA Project Unit, 2010). World Bank, 2015). There is a high degree of when parents migrate for work. Older people in gender equality in human development out- Myanmar also tend to continue working after comes in Myanmar, with no significant differen- the age of 60: nearly a third continue working tials in nutrition outcomes or in net enrollment for another decade. Many older people also rates between girls and boys. report poor health and functional limitations, particularly among those aged 80 and above There is a large informal sector in Myanmar. (HAI, 2013). In the absence of mechanisms to More than 50 percent of the working popula- promote savings, income insecurity can be an tion is employed in agriculture and related ac- issue for older people whose family cannot sup- tivities (see Figure 2). In 2009/10, 71 percent of port them. workers aged 15-59 years were self-employed workers, unpaid family workers, or employed as casual workers. A large proportion of the population, particularly the poor, is dependent on seasonal employment in agriculture and 2.3 Other common sources of on insecure casual work in the informal urban risk and vulnerability economy. Underemployment and incidence of working poor is high, especially in the lean agricultural season (Enlightened Myanmar Re- A large share of households is vulnerable to search and World Bank, 2015). Skill levels are the danger of falling into poverty; volatility low, especially among the poor; over half of the of household incomes and exposure to risks working-age population has either no formal remain a serious concern. A large share of education or only some primary education. households are clustered very near the poverty line and may fall into poverty after exposure to Most older people in Myanmar live with their large, severe, and/or frequent shocks. The com- families and rely on their families for support. mon sources of risk and vulnerability that affect They also contribute in return, typically by people of all ages include ill-health and expo- providing care for grandchildren, particularly sure to natural hazards, disasters, and conflict. Overview 7 Despite substantial improvements, health out- severely damaged the ability of communities to comes in Myanmar are poor relative to other recover on their own; they lost their productive ASEAN countries. Myanmar faces a double assets, which exacerbated problems such as burden of disease, with widespread prevalence malnutrition and food insecurity, keeping them of both non-communicable diseases and infec- in a cycle of poverty (UNEP, 2009).7 tious diseases. These diseases and malnutrition contribute to disability and premature death. In recent years, ceasefire agreements with In addition, deficiencies associated with food ethnic insurgent groups have been signed but security, access to safe water and sanitation, decades of conflict has exacerbated issues of unsafe working conditions, and access to af- income and food insecurity. Localized conflicts fordable good quality health care dispropor- driven by ethnic tensions and fighting between tionately expose poor and vulnerable house- government forces and armed groups have dis- holds to illness, accident, and early deaths. placed over 200,000 people, primarily in west- The effects of an illness, accident, or death of ern and northern Myanmar (WFP, 2013). In the income-earner can be devastating to poor Rakhine state, over 125,000 people have been and vulnerable households in terms of the ex- displaced by inter-communal conflict since penses associated with health care and lost mid-2012. Recurrent outbreaks of religious vio- income. Out-of-pocket private health expendi- lence, primarily targeted at the country’s Mus- ture in Myanmar is estimated to have been as lim minorities, have been recently reported high as 60 percent of total health expenditure particularly in Rakhine (World Bank, 2015b). in 2012/13 (World Bank, 2015d).5 Poor house- Impoverished, with almost no job opportuni- holds often go into debt to treat illness and ac- ties, most of the internally displaced people cidents. In the event of death, funeral expenses (IDPS) are wholly reliant on humanitarian as- can be substantial. sistance. Since the start of the peace process in 2011, bilateral ceasefires have been agreed Myanmar suffers from disaster losses every with 14 of the 16 armed ethnic groups. While year caused by a range of hazards that have differences remain between the government high immediate and long-lasting impacts on and the armed groups and conflict continues in people, communities, and the economy. In some areas (e.g. Kachin state), there is a shared the past 25 years, Myanmar has suffered 32 di- commitment towards a peaceful solution. As saster events (seismic activity, epidemic, flood, of 30 March 2015, peace negotiators agreed to landslides, and storms) affecting more than 4 the draft text of a historic nationwide ceasefire million people and causing USD 4.7 billion in agreement, bringing the country closer to end- damages.6 Cyclone Nargis was the worst natu- ing decades of conflict. ral disaster in the history of Myanmar, generat- ing a total loss of 21 percent of the country’s 2007 gross domestic product (GDP) (ASEAN, 2008). Disasters disproportionately affect the 2.4 Coping with risk and poor (particularly women, children, the elderly, people with disabilities, migrants, and margin- vulnerability alized groups), who have less ability to cope with and recover from disasters. Environmen- Individuals and households in Myanmar often tal degradation and recurrent disasters further resort to potentially growth-reducing strate- exacerbate vulnerability. For example, Nargis gies in order to cope with risks. In particular, 5. Poor data quality imply these estimates may not be entirely accurate. 6. Emergency Events Database (EM-DAT): http://www.emdat.be 7. See Note on ‘Social protection for disaster risk management: Opportunities for Myanmar’. 8 Building Resilience, Equity and Opportunity in Myanmar: The Role of Social Protection poor and vulnerable households have limited dence from qualitative studies suggests these capacity and fewer options for mitigating risks vary, depending on the extent of intra-commu- and coping with shocks. The most common nity diversity, the degree and recentness of con- coping mechanisms, particularly among the flict-related impacts, and exposure to external poor, are adjustments to household labor sup- assistance, among other factors (Enlightened ply (e.g. working longer hours, taking children Myanmar Research and World Bank, 2015). In out of school and putting them to work, and the Dry Zone and Delta, for example, there are migrating for work), incurring debt (usually in several examples of mechanisms to support the the form of high-cost loans from the informal poor and vulnerable, such as funeral grants and sector), reducing household consumption, and grants to children from poor families to attend selling assets. school (Thu and Griffiths, 2014). Information on community-based social pro- tection in Myanmar is patchy. The limited evi- 3. Role for social protection in Myanmar8 Social protection policies and programs can play an important role in Myanmar’s strategy for poverty reduction and people-centered development. Many countries have used a broad range of social protection programs to achieve their develop- ment goals, particularly on poverty reduction and inclusive growth. These include cash transfers (CTs), school-feeding, public works programs (PWPs), social secu- rity and social welfare programs. The multiple and pressing needs identified in opportunity by promoting investments in pro- the previous section call for a strengthening ductive activities. See Box 1 for common types of Myanmar’s social protection policies and of social protection program that have contrib- programs. Global experience suggests a wide uted to these goals. range of programs that can contribute to the government’s poverty reduction, people-cen- Social protection programs are a growing ele- tered development, and national reconciliation ment of the development agenda, being pres- agenda at the national level. These can also ent in virtually every country worldwide. So- help tackle the development constraints fac- cial assistance programs alone cover nearly 1.9 ing communities and households in Myanmar. billion beneficiaries globally. Developing coun- At the community level, social protection pro- tries spend an average of 1-1.7 percent of GDP grams contribute to building productive assets on social assistance programs such as CTs and and community resilience and stimulate the lo- PWPs (see Box 2). Social assistance spending in cal economy. At the household level, they help East and South Asia ranges from 0.6 percent of households build resilience and avoid negative GDP in Vietnam to 2 percent in India (see Fig- coping mechanisms, promote equity by build- ure 3). There is little difference between up- ing and protecting human capital through in- per-middle-income countries and low-income vestments in health and education, and create countries in terms of spending on social as- 8. See Note on ‘Framework for the development of social protection systems: Lessons from international experience’. Overview 9 Box 1: Promoting resilience, equity, and opportunity through social protection Resilience to risks and shocks can be built through social security and social assistance programs that minimize the negative impact of economic shocks on individuals and families. Examples of these pro- grams include old-age pensions, unemployment and disability insurance, and scalable PWPs. Equity for the poor and vulnerable can be achieved through social assistance programs such as CTs, food transfers, and welfare services that help protect against destitution, promote equality of oppor- tunity, and address chronic poverty. Opportunity for all is achieved by promoting equitable and affordable access to nutrition, health, and education services for all, including the poor and vulnerable, and by helping men and women access more productive employment. Source: World Bank (2012a). sistance, illustrating the affordability of safety the government of Indonesia is undertaking a nets and the role they play at various stages of shift from spending on universal fuel subsidies development. Countries tend to increase their to targeted household and community trans- level of spending on social insurance programs fers that are well aligned with the government’s as they get richer. overall development agenda (World Bank, 2015a). 3.1 At the national level Social protection has contributed to facilitat- ing social stability and long-lasting reconcili- ation in many conflict-affected contexts. For Countries that have invested in building so- instance, Burundi introduced PWPs to support cial protection systems have seen significant reinsertion and ensure long-lasting reconcilia- reductions in poverty and inequality, trans- tion in conflict-affected communities. El Salva- lating into inclusive growth. For instance, dor’s Temporary Income Assistance Program Brazil’s Bolsa Família conditional cash transfer provided cash assistance and job training to (CCT) program contributed to halving extreme youth and women heads of households, who poverty (from 10 to 4 percent) and reducing live in areas characterized by high levels of so- inequality by 15 percent between 2003 and cial exclusion and high rates of social and gen- 2013. Mexico’s Oportunidades (now Prospera) der violence. The program has helped reduce CCT program reduced incidence of poverty in the occurrence of violence and rebuild the so- rural areas by 8 percent (Fiszbein et al., 2011; cial fabric in violent or at-risk communities (An- Wetzel, 2013; World Bank, 2015a). In Thailand, drews and Kryeziu, 2013). unconditional cash transfers (UCTs) for older people (also referred to as social pensions) re- Social protection can help people cope bet- duced poverty incidence among beneficiaries ter and recover faster from disasters and oth- by 30 percent.9 er crises. For instance, Ethiopia’s Productive Safety Net Program (PSNP), designed to create Governments have also used social protection a predictable safety net for chronically poor to embark on growth-enhancing reforms and households, has proven an effective means of facilitate economic adjustments. For instance, responding to shocks. In response to the 2008 9. http://datatopics.worldbank.org/aspire/ 10 Building Resilience, Equity and Opportunity in Myanmar: The Role of Social Protection Figure 3: Social assistance spending (% of GDP) Sources: Andrews (2009), World Bank (2011, 2015a); Note on ‘Inventory of social protection programs in Myanmar’. food, fuel, and financial crisis, the PNSP’s Risk support community development. In 2013, Financing Mechanism (see Section 5.2) enabled communities receiving World Food Programme the government to scale up the program and (WFP) interventions benefited from the cre- provide additional support to the over 4 mil- ation of 764 km of renovated road, 205 km of lion existing beneficiaries as well as temporary irrigation and contour trenches, 2,200 hectares support to an additional 1.5 million individuals of land development and agro-forestry, and 21 (World Bank, 2012b). In the Philippines, the schools, among other local assets. government used the Pantawid Pamilya CCT to channel donor funding for emergency response Social protection programs bring not only efforts when Cyclone Haiyan hit. benefits to direct beneficiaries but also posi- tive spill-over effects to the larger commu- nity, further supporting resilience and eco- nomic development. In Malawi, the welfare of 3.2 At the community level non-participating poor households improved, through transfers from program beneficiaries as an informal support mechanism and through Social protection programs support commu- increasing purchasing power in the community nity resilience and economic development (Alderman and Yemtsov, 2013). through building community infrastructure and promoting connectivity. PWPs in particu- Social protection programs support commu- lar not only create jobs for needy workers but nity cohesion by promoting the participation also create and maintain small-scale commu- of the poor in community decisions. Commu- nity infrastructure (see Box 2). In Myanmar, nities often play a major role in the planning, PWPs implemented by development partners implementation, and monitoring of PWPs. In- (DPs) have also created community assets that dia’s Mahatma Gandhi National Rural Employ- Overview 11 ment Guarantee Scheme (MGNREGS) provides Box 2: PWPs – building community resil- a central role to local governments in imple- ience through community infrastructure menting the program and to communities in and household food security planning and monitoring implementation. Ethiopia’s PSNP has brought together commu- PWPs, also known as workfare or labor-in- nities around the targeting process, promoting tensive rural works programs, can serve two community consultation to identify beneficiary objectives: creation of jobs for needy work- households (Subbarao et al., 2013). ers and creation and maintenance of small- scale community infrastructure. They pro- Social protection programs can also enhance vide temporary employment at low wages the poverty reduction impact of community- mainly to unskilled workers in rural areas. driven development (CDD) platforms. For in- Such temporary employment during the slack stance, Argentina’s Trabajar program was able season or during crisis will typically contrib- ute to consumption-smoothing and poverty to reach the poorest effectively (85 percent of alleviation. Public works can also be used to the beneficiaries fell in the bottom 20 percent promote social cohesion; they are now pres- of the income distribution) through a combina- ent in 94 countries. International experience tion of focusing in poorer areas, setting wages with PWPs includes a diverse range of design that encouraged the poor to self-select into the features. For instance, India’s MGNREGS guar- program, and encouraging pro-poor infrastruc- antees 100 days of employment each year to ture (potable water and social infrastructure, all rural households that demand work. Ethio- both labor-intensive and disproportionately pia’s PSNP, on the other hand, provides work benefiting the poor). to poor households in need. Source: Subbarao et al. (2013). 12 Building Resilience, Equity and Opportunity in Myanmar: The Role of Social Protection 3.3 At the household level 2011). Comparisons of six similar school-feed- ing interventions implemented in five coun- tries (Bangladesh, Burkina Faso, India, Kenya, Many of the programs noted above also help and Uganda) found they deliver similar positive address the risks and vulnerabilities facing impacts on school attendance, ranging from households and individuals at different stages a 6- to a 20-percentage point increase (IEG, of the lifecycle as identified in the previous 2011). The Philippines Pantawid Pamilya CCT section (see Figure 4). For instance, school- has increased primary school enrollment by 4.5 feeding and CT programs support school-age percentage points, pushing levels near univer- children, whereas PWPs, pensions, and social sal enrollment, and reduced severe stunting of security programs support working-age and young children by 10 percentage points (World older people. Bank, 2014b). Programs such as CTs and school-feeding can PWPs and social security programs can sup- be critical to prevent the intergenerational port the working-age population to avoid transmission of poverty by ensuring children negative coping strategies by ensuring income from poor households have similar access security during shocks. PWPs can serve as un- to education and health services to the non- employment insurance in times of economic poor. UCTs in Ecuador and South Africa reduced downturns and high unemployment. House- child labor and increased children’s schooling, holds benefiting from Ethiopia’s PSNP for five health, and nutrition in recipient households years increased their food security by a month (Baird et al., 2010); CCTs in Colombia and Mex- (Subbarao et al., 2013). Social security pro- ico have improved the nutritional status of chil- grams such as health and disability insurance dren in beneficiary households (Fiszbein et al., help households manage sudden losses of in- Figure 4: Social protection along the lifecycle Overview 13 come and decreased out-of-pocket expenditure invest 14 percent of the transfer amount in as- when experiencing health shocks (World Bank, sets (mainly farm animals and land for agricul- 2014a). tural production). In Brazil, the social pension program led to high incidence of investment A range of social protection programs can in productive capital (Alderman and Yemtsov, help ensure decent living standards in old 2013). age. Practically all countries have mandatory pension schemes, at least for civil servants. Most have mandates that cover at least some private sector workers. Different routes have been taken to expand coverage to workers in 3.4 Evolution of social the informal sector, especially the poor and vul- protection provision nerable. For instance, social pensions in Thai- land and Chile reduced poverty incidence by 30 percent and 12 percent, respectively (World Social protection provision typically evolves Bank, ASPIRE).10 Community-based care is also gradually to increase coverage and adequacy helping older people with their needs beyond of support, as well as to address additional income insecurity. The Voluntary Home Care needs through increasingly sophisticated pro- Programme in Sri Lanka provides home-based gram design, institutional arrangements and support and basic health and nursing services delivery systems. Countries like the Philip- (HAI, 2005). pines have seen a gradual expansion of social protection programs, building on the platform For particularly vulnerable groups, such as provided by the flagship program, the Pantawid women and people with disability or chronic Pamilya Program. This started as pilot in 2008 illnesses, social protection programs can help but was quickly expanded and the delivery sys- improve welfare and promote inclusion. In- tem improved to respond to the consequences dia’s self-help groups for people with disabili- that the food, fuel and financial crises had on ties had a positive impact on people’s aware- poor households. In 2014 the program reached ness and access of government services, as well 4 million poor households and it now serves as an on income-generating opportunities. PWPs platform to roll out the National Household Tar- can be specifically designed to reach vulner- geting System for Poverty Reduction, known as able groups that may be disadvantaged in the Listahanan. In addition, the government used labor market. For instance, nearly half the em- Pantawid Pamilyang along with the National ployment generated in India’s MGNREGS is for Community Driven Development Program women; this addresses the issue of low labor (NCDDP) to channel donor funding for emer- market participation of women in rural areas. gency response efforts when Typhoon Yolanda (Haiyan) hit. The government is now working Protecting households against shocks through on the operational convergence of the Panta- social protection not only eases poverty mo- wid Pamilya and the NCDDP. mentarily but also enables growth by poor and near‐poor households to make productive investments. Ethiopia’s PSNP demonstrated that three in five beneficiaries avoided having to sell assets to buy food in times of drought (Devereux et al., 2008). Similarly, Mexico’s Oportunidades program helped beneficiaries 10. http://datatopics.worldbank.org/aspire/ 14 Building Resilience, Equity and Opportunity in Myanmar: The Role of Social Protection 4. Myanmar’s current social protection system Core building blocks of Myanmar’s social protection system need to be further developed in order to address the following main challenges: providing adequate coverage of beneficiaries in need of assistance, improving responsiveness of programs to accommodate those made newly vulnerable because of systemic shocks, and effectively connecting beneficiaries of so- cial protection programs to productive opportunities. Social protection policies have recently been formulated, emphasizing the role of social pro- tection in achieving Myanmar’s goals of poverty reduction and inclusive growth. However, the process is fragmented across ministries and the social protection system does not yet fully address the priorities identified in the previous sections. Despite recent increases, government spending on social assistance is only 0.02 percent of GDP – well below the global and regional average. Social assistance programs reach only 0.1 percent of the population. The government has recently introduced CT pilots to promote equitable and affordable access to education and health services. Preliminary evidence indi- cates these are meeting the desired objectives. In addition, DPs have extensive experience of implementing social assistance programs in Myanmar. These can provide valuable insights to the government in building the ‘social assistance block’ of the social protection system. The ‘social security block’ of the social protection system is comparatively more developed, but faces the challenge of low coverage and benefit adequacy, while the proposed reforms raise further concerns of fiscal sustainability. Government spending on social security provi- sions workers in the formal sector is about 0.6 percent of GDP, reaching just about 3 percent of the population. Multiple and generous provisions carry the risk of crowding out social as- sistance spending while further straining developing delivery systems. An effective and sustainable social protection 4.1 Overview system responds effectively to the needs of beneficiaries by adopting a harmonized ap- proach at the policy, program, and delivery Social protection policies are currently being levels (see Figure 5). While priorities and the formulated, but in parallel processes across means to address them will evolve over time, it different ministries. The Rural Development is important to gradually develop the building Strategic Framework (RDSF) and the Social blocks of a responsive social protection system. Protection Strategic Plan (SPSP) were devel- This section examines Myanmar’s current social oped by the Department of Rural Development protection system to understand the challeng- (DRD) and the Ministry of Social Welfare, Relief es and identify opportunities for strengthening and Resettlement (MSWRR), respectively, and policies, programs, and delivery systems in a endorsed in 2014. Both policy frameworks in- systematic manner. corporate social assistance as a poverty reduc- tion tool through a multi-sectoral approach. Overview 15 Figure 5: Three levels of engagement of social protection systems Source: World Bank (2012a). The roles and responsibilities for coordination, low, at 0.57 percent as a share of GDP in design, and delivery of social protection under 2014/15 (see Table 1). The bulk of this financ- these frameworks are still being articulated. es the government’s social security programs. Sectoral policies in the Ministries of Education Spending on social assistance (0.02 percent of (MOE) and Health (MOH) continue to guide el- GDP) is extremely low. This finances two recent ements of social assistance provision, such as pilot programs, including stipends for poor and demand-side approaches to universal health vulnerable students (MOE) and a maternal and coverage and education for all, whereas social child health voucher scheme (MCHVS) (MOH), welfare services fall under MSWRR. Though not as well as some small-scale social welfare pro- yet introduced, both the RDSF and the SPSP as- grams (MSWRR). sign the mandate for implementing PWPs to the DRD. Social security provisions are legislat- Coverage of these social protection programs ed, with civil service pensions falling under the is low, reaching just about 3.2 percent of the mandate of the Ministry of Finance (MOF) for population; social assistance programs reach pensions and other social security schemes for only 0.1 percent of the population. In contrast, formal private sector workers under the Social in other countries in the East Asia Pacific region, Security Board (SSB), Ministry of Labor, Employ- social protection programs reach 44 percent of ment and Social Security (MOLESS). the population on average, with social assis- tance reaching 39 percent of the population.11 Despite recent increases, total government spending on social protection remains very 11. Estimates on the coverage of social protection programs in East Asia and the Pacific can be found at http://datatopics.worldbank. org/aspire/region/east-asia-and-pacific 16 Building Resilience, Equity and Opportunity in Myanmar: The Role of Social Protection In this context, DPs have played an important cial assistance provision, reaching 3 percent of role in providing social assistance to vulnerable the population. WFP programs (food- and cash- populations. Originally motivated by the need for-work, early childhood nutrition programs, to provide emergency support and humanitar- school-feeding, emergency relief including to ian relief in times of crises, many DP programs IDPs) spent about MMK211 billion between have been evolving to provide development- 2013 and 2015 – that is, the annual budget for oriented support in recent years (see Figure these programs was more than 5 times higher 6). WFP and the Livelihoods and Food Security than government social assistance spending.12 Trust Fund (LIFT) are the biggest financers of so- Table 1 : Government and DP expenditure on social protection, 2014/15 (revised estimates) Government Total expendi- % govern- % GDP Estimated number of % of eligible population programs ture 2014/15 ment beneficiaries covered (age group) (MMK billion) expenditure Civil service pensions 362 1.65% 0.55% 843,000 1.6% (18.8%, old age) (MOF) Social security for formal 4 0.02% 0.01% 765,000 1.5% (2.4%, working age) sector workers (MOLESS) Social welfare programs 10 0.05% 0.02% 25,000 0.05% (all age groups) (Department of Social Welfare, MSWRRR) Stipends program (MOE) 3.1 0.014% 0.005% 37,000 0.07% (0.38%, school age) Total social protection 379.1 1.73% 0.57% 1,670,000 3.2% (all age groups) Total social assistance 13.1 0.06% 0.02% 62,000 0.12% (all age groups) Largest DP programs WFP (all SA programs) 211 (2013-2015) .. .. 1,113,743 2.2% (all age groups) LIFT (SP and other 76 (2010-2014) .. .. 576,000 (hh) 1.1% (all age groups) programs) Notes: The Budget Department, MOF, does not compile disaggregated figures of other social protection programs; bud- get figures for MCHVS and MOLESS migrant services are small and will not significantly alter the analysis. MSWRR expenditure includes expenditure on supply of social welfare services. WFP programs include nutrition, asset creation (cash and food for work), school-feeding, emergency relief (including for IDPs), and food assistance for HIV/AIDS and tuberculosis patients. WFP and LIFT budgets and coverage are for the period between 2013-2015 and 2010 and 2014, respectively. Exchange rate: USD1=MMK1,034.13. Source: Department of Budget and Department of Pensions (MOF), SSB (MOLESS), Department of Social Welfare (MS- WRR), and Department of Planning (MOE), WFP, and LIFT. Coverage figures based on age-disaggregated population estimates of the 2014 Census (GoM, 2015). 12. See Note on ‘Inventory of social protection programs in Myanmar’. Overview 17 Figure 6: Summary of social protection programs by life cycle category Early childhood School-age children Working-age Old age All age groups (equity and (equity and population (equity (equity and resilience) (opportunity, equity opportunity) opportunity) and resilience) and resilience) • Institutional care for • Stipends, scholar- • Social security for • Pensions for civil • Care for people with orphans (MSWRR) ships, supply of formal workers (SSB) servants (MOF), disabilities (MSWRR) • Support for triplet textbooks (MOE) • Employment services defence (MOD) • Welfare services for and greater sets of • School supplies and for migrant workers • Voluntary homes for people with leprosy newborns (MSWRR) access to education (MOLESS) the elderly (MSWRR) (MSWRR) • Maternal and child for vulnerable • Voluntary homes for • Pensions for over • Rice donations to health vouchers children (UNICEF) women (MSWRR 100 year olds homes (MSWRR) (MOH) • School-feeding • Food and cash for (President's Office) • Cash and in-kind • Provision of food (WFP) work (WFP, LIFT) • Support to old emergency support and micronutrient • Other livelihood people self-help after disasters supplements for programs for groups (MSWRR/ (MOBA, MSWRR) pregnant and vulnerable HAI) • Cash and in-kind lactating mothers, communities (LIFT support to internally fortified food for and partners, ILO) displaced people children (MOH/ (WFP and partners, WFP) • Migration centers (MOLESS/IOM, ILO) HAI) • Cash transfers (SC) Notes: Programs covering more than 30,000 beneficiaries are highlighted in red. Source: Note on ‘Inventory of social protection programs in Myanmar’. 4.2 Social assistance provision introduction of CT pilots to promote equitable and affordable access to education and health in Myanmar13 services. These include the following: • MOE programs, including stipends for poor Government spending on social assistance and vulnerable students, stipends to all is low by international standards: Myanmar primary school students, scholarships, and spends 0.02 percent of GDP; on average, low- textbooks. MOE’s stipend program for poor income countries spend 1.1 percent, while Af- and vulnerable students has two modalities rican countries spend 1.7 percent (World Bank, of implementation: a) the ‘national program’ 2015a). Government social assistance pro- has been in operation since 2009/10 and is grams have so far been few and small in scale, based on MOE’s current guidelines for imple- and thus are far from providing a comprehen- mentation, covering all townships at a small sive safety net in the country. These include scale and prioritizing orphans; b) the ‘pilot’ largely several small programs implemented by program tests changes to the design and MSWRR to support groups such as vulnerable implementation of the national program – women, orphans, older people, and persons namely, criteria to prioritize townships and with disabilities with CTs, institutional care, and students, conditions, and increased benefit vocational training. levels. These programs aim to encourage school enrollment and prevent dropouts, In recent years, spending on social assistance especially among the poor and vulnerable. has increased, albeit from a low base, with the For instance, the stipends pilot sets different 13. See Notes on ‘The experience of public works programs in Myanmar’ and ‘The experience of cash transfers in Myanmar’ for a detailed analysis. 18 Building Resilience, Equity and Opportunity in Myanmar: The Role of Social Protection benefit levels for primary, middle, and high • Cash and food transfers for disaster and hu- school students to reflect how much more manitarian response aim to provide short- to costly it is for families to send children to medium-term relief to affected households, school as they grow older and to incentiv- assist with household food and income se- ize families to continue to send children to curity, particularly during the lean season, school at critical transition points (i.e. pri- and contribute to the dignity and income se- mary to middle school and middle to high curity of vulnerable groups. In recent years, school).14 with the increasing focus on medium-to long-term development objectives, CCTs to • MOH’s pilot MCHVS promotes access to promote maternal and child health have also health services for poor pregnant women been piloted. For example, a maternal and who cannot afford to access health servic- child health pilot CCT was introduced by Save es and who live in remote areas. The pro- the Children (SC) in Rakhine in 2014 and may gram provides vouchers for mothers to cover be scaled up. transport costs to MOH health facilities as well as other costs for accessing antenatal • C/FFW programs have been developed to care (ANC), delivery assisted by skilled birth respond to the specific local context. For attendants, and immunization of their chil- instance, C/FFW programs in the Dry Zone dren. focus on ensuring seasonal food and liveli- hood security (especially during the agri- These pilot programs form a critical compo- cultural lean season), CFW programs in the nent of Myanmar’s evolving social protection Delta focus mainly on post-disaster recovery, system. They address crucial development chal- and C/FFW programs in conflict-affected ar- lenges of malnutrtion and school dropouts, es- eas in the Border States respond to multiple pecially among children from poor households. needs of communities, such as post-disaster These programs are consistent with the grow- recovery after Cyclone Giri (Rakhine) and rec- ing global spread of CTs that leverage human onciliation and social cohesion (Tanintharyi). development outcomes and can be scaled up In 2014, WFP’s C/FFW programs helped ad- to cover a larger proportion of the population.15 dress food security issues for 225,500 ben- eficiaries with 45 days of employment on av- In this context, the extensive experience of erage. LIFT supported several CFW schemes DPs in implementing CTs, PWPs, and other that reached more than 172,000 households social assistance programs in Myanmar pro- until mid-2014, providing on average 25 days vides valuable insights to the government in of household food security. Recent LIFT-sup- scaling up coverage of social assistance. The ported PWPs as part of the Tat Lan Program largest DP-implemented programs are by WFP in Rakhine provided a higher number of days and LIFT. A number of other DPs and NGOs are (40-90). These also present an interesting also actively engaged in social assistance provi- model combining CFW with complemen- sion. At present, these programs cover selected tary CTs for those unable to work, ensuring townships in states and regions in the border broader community support. areas, the Dry Zone, the Delta, and Rakhine. There is a range of programs used to address • Various other programs, such as school- specific needs, including the following: feeding, provision of school supplies, and language enrichment programs, comple- 14. See Note on ‘Risk and vulnerability along the life cycle in Myanmar: The role of social protection’. 15. See Note on ‘Framework for development of social protection systems’. Overview 19 ment government efforts to facilitate access dence of impact on promoting income and food to education for vulnerable groups (food- security for beneficiaries, reducing reliance on insecure children, out-of-school adolescents, negative coping mechanisms. For instance, and ethnic minorities). WFP monitoring data show CFW beneficiaries spend on average 70 percent of the cash on While evidence on impact is patchy, qualita- purchasing food and 17 percent on accessing tive assessments indicate the recent govern- health and education services. At the commu- ment pilot programs and DP programs are nity level, these programs have created local meeting the desired objectives. With respect assets that support community development to CTs, there is some evidence of increased abil- and enhanced community resilience against di- ity to purchase food and increased utilization of sasters (LIFT, 2013; WFP monitoring data). health and education services, although gov- ernment programs are in relatively early stages of implementation. For instance, MCHVS ben- eficiaries have increased their utilization of ma- ternal and child health services (WHO, 2014). 4.3 Social security provision in The stipends program so far has encouraged Myanmar16 parents to get more involved in their children’s education, to improve attendance, and to cover education costs such as stationery, uniforms, Though Myanmar’s current social security sys- and transport with the stipend (SC, 2015). With tem faces the threefold challenge of coverage, respect to DP-implemented PWPs, there is evi- adequacy, and sustainability, the demographic 16. See Note on ‘Strengthening social security provision in Myanmar’ for a detailed analysis of the current system and proposed re- forms. 20 Building Resilience, Equity and Opportunity in Myanmar: The Role of Social Protection outlook in Myanmar allows for a gradual ex- efits. In addition, it increased the benefit levels pansion. In doing so, Myanmar can take ad- of existing provisions and ramped up the total vantage of the experience of its neighbors and mandatory contribution rate from the current 4 avoid costly mistakes. percent to 13 percent. Moreover, it allows for a voluntary contribution of 25 percent by worker Although the pension and SSB programs rep- towards housing benefits (GoM, 2012). At pres- resent the bulk of social protection spending ent, workers in the informal sector are not cov- in the country, they cover just about 3 percent ered, but the Law also introduces possible pro- of the population. Benefit levels are low and gressive extension of social security provisions do not provide adequate income security for to the informal sector. workers and pensioners. At present, only civil servants, military and political personnel, and The new Social Security Law was implemented employees of state economic enterprises (SEEs) in 2014, but in a phased manner. As of April are covered by an old-age pension scheme. The 2015, only the existing branches of social secu- current pension scheme is a non-contributory rity were covered. Cash benefits levels for exist- and unfunded defined benefit (DB) plan based ing provisions have been raised and collection on a formula that provides a pension worth 50 has started of additional contributions from percent of the final salary for a worker who employers for the work injury fund. The total spends 35 years in the civil service.17 Civil ser- contribution currently is 5 percent. vants are also provided benefits in case of in- validity and work injury, death, or termination of employment. At present, there is no pension scheme operating for workers in the private Recent reforms sector. The government is considering reforms to Workers in the formal private sector are cov- the civil service pension scheme, including ered by a contributory social security scheme changes to the parameters of the existing DB and provided medical care and five types of scheme as well as more systemic reforms that cash benefits for sickness, maternity/pater- will change the underlying architecture of the nity, death (funeral grant), and work injury. scheme. The MOF proposal covers two phases. These provisions were first introduced under First is a move from an unfunded to a partially the 1954 Social Security Act. The SSB (MOLESS) funded DB scheme, with the government put- administers these schemes through a large net- ting aside 10 percent of the wage bill for civil work of SSB township offices and medical facili- servants each year. The accumulated savings ties. would be earmarked to pay future pension benefits in the existing DB pension scheme. In 2012, the government adopted a new Social Second is a traditional defined contribution Security Law, which extended social security (DC) scheme. In late 2014, an alternative pro- provisions for formal sector workers in several posal was put forward by an inter-ministerial ways. The Law introduced additional branches committee set up by the President’s Office. This of social security – that is, family benefit, su- proposes establishing a Myanmar Provident perannuation benefits (i.e. old-age savings Fund (MPF) and diverting 5 or 10 percent of the scheme), disability and survivors’ pensions, civil servant’s wage to two individual accounts; unemployment insurance, and housing ben- one for retirement and one for pre-retirement 17. Defined benefit (DB) is a guarantee by the pension agency that a benefit based on a prescribed formula will be paid. The benefit is generally expressed as a percentage of the salary. Defined contribution (DC) is a pension plan in which the periodic contribution is prescribed and the benefit depends on the accumulated contributions and the investment return. Overview 21 withdrawals for education, health, and housing it difficult to estimate costs. Of particular con- (GoM, 2014a, 2014b). cern is the provision on housing benefits. Glob- al experience with publicly funded housing in In principle, the move from an unfunded to a social security schemes has been mixed at best partially funded scheme can help the govern- and suggests such programs can present major ment reduce what is likely to be a rising pen- governance challenges and absorb substantial sion bill by setting aside funds earmarked for resources that compromise the ability of funds this purpose. However, the challenges of a new to deliver on their other benefit commitments. defined contribution scheme for civil servants It would be advisable to remove housing enti- or an MPF are daunting at the present time. For tlements from the benefit package at this stage both reforms, a number of conditions would or, at a minimum, to tighten the regulations to have to be met in order to achieve their objec- prevent cost escalation and malfeasance. tives (see below). In particular, using multiple instruments to provide pensions for the same In addition, concerns related to financing, gov- group (i.e. civil servants) should be avoided. ernance arrangements, and investment policy These proposals for parametric changes to are of relevance to the proposed civil service the existing DB scheme, introduction of a DC reforms and the Social Security Law: scheme, and/or setting up the MPF need to be carefully considered to avoid the risk of duplica- • Financing: While a solid contribution base is tion. Not only would this strain administrative important, high contribution rates can have capacity,18 but also the additional budgetary adverse labor market and fiscal implications. outlay for these multiple provisions would likely Gradual accumulation of funds would be crowd out other social spending. desirable in Myanmar’s context, implying lower contribution rates and, in the case of The 2012 Social Security Law aims to expand the pension reforms, initial contribution only coverage and provide adequate benefits, but from new civil servants. It is also important the financial sustainability of the proposed to avoid, or at least to minimize, budgetary benefits is in question. The SSB has indicated contributions for the proposed contribu- its intent to revise the regulations and amend tory pension scheme and subsidies to the the law based on a more rigorous analysis and SSB schemes that may have the unintended costing of the proposed provisions. The discus- effect of crowding out other areas of social sion below highlights the main areas that would spending. benefit from a review. • Managing and investing the funds: There The overall package of services and benefits are serious concerns about the uncon- is generous for a country at Myanmar’s level strained and under-regulated investment of development and there are significant un- policy in the 2012 Law.19 Given the dearth of certainties on the financial obligation implied. liquid and tradable assets in Myanmar and The value of several benefits, including pen- the lack of experience in this area, it would sions and health services and benefits, is left be prudent to restrict investments. One op- open-ended in the Law and regulations, making tion would be a special non-marketable gov- 18. Trying to pursue short-term social outcomes (education, health, and housing) and long-term income security (pensions) through the MPF will require a much higher degree of administrative capacity and record-keeping than a pure retirement account and adds to administrative costs. 19. The Law and regulations place few limits on the nature of investments the social security funds permit. Some of the permissible options are not commonly allowed in global social security schemes. For instance, investments in real estate and loans to third parties and in economic enterprises are generally best avoided altogether. These open possibilities for risky investment and poten- tially malfeasance at the cost of members and with a potential for budgetary bail-out affecting the wider population. 22 Building Resilience, Equity and Opportunity in Myanmar: The Role of Social Protection ernment bond issued expressly for the provi- identification in SSB schemes; the Myanmar dent fund and linked to bank deposit rates, Economic Bank (MEB) issues e-pension cards to preferable to investing in illiquid assets. Ap- pensioners in the civil service pension scheme. propriate governance arrangements will also need to be worked out. Enrollment and verification of eligibility 4.4 Social protection delivery systems20 Government and DP social assistance pro- grams typically focus resources in selected areas and on specific groups. For social assis- The delivery of social protection programs tance programs, the methods used to identify comprises the following key processes, as il- program beneficiaries vary. Most programs use lustrated in Figure 7: communications and geographic targeting as a first step, reflecting outreach, beneficiary identification, eligibility the regional diversity of needs (see Section 2). and enrollment, transactions (in-kind or cash Note that the exclusive use of geographic tar- benefit transfers/contribution collections, etc.), geting needs to be carefully assessed keeping grievance redress, and monitoring and evalu- in mind political and social factors, particularly ation (M&E). We focus on four key processes in conflict-affected areas. In particular, conflict- below, summarizing experience from govern- affected areas where non-state actors are ac- ment- and DP-led social protection provision in tively providing services can generate dynamics Myanmar. that influence the way identifying poor areas and poor people is perceived. Identification Figure 7 : Social protection program delivery National identification exists in Myanmar but processes not all beneficiaries of social protection pro- grams have it. Myanmar has a national identi- fication system called the National Registration Card or Citizenship Scrutiny Card (CSC). All citi- zens of Myanmar are entitled to a national ID card from the age of 10. However, only 69 per- cent of those over 10 years of age possess one, with coverage particularly patchy in rural areas (GoM, 2015) and among vulnerable groups such as IDPs, migrants, and ethnic minorities. Beneficiaries who have a CSC use it as proof of ID to enroll in social protection programs; alternative ID mechanisms are used for those who do not have it (e.g. household list/book or statement from village administrators). The SSB at union level issues smartcards for beneficiary 20. See Note on ‘Institutional landscape for implementation and financing of social protection programs: Towards effective service delivery in Myanmar’. Overview 23 Most social protection programs typically rely Transactions (including benefit pay- on a combination of methods. These include ments, reimbursement, and contri- geographic targeting in the first stage, com- bined with some form of household or indi- bution collection) vidual targeting in a second stage. This can take the form of categorical targeting (i.e. all indi- Social assistance programs in Myanmar use viduals in specific groups) or reliance of poverty various payment modalities, such as cash, near and vulnerability criteria or a combination of cash (vouchers), and in kind (usually food).22 the two. In determining eligibility criteria, pro- Payment of benefits is usually made through grams could chose to cover all individuals be- direct distribution to beneficiaries. In the case longing to selected groups in a small number of social assistance programs, this is typically of communities or only the poorest and most the responsibility of local service providers. For vulnerable but spread in many more communi- instance, for MOE’s stipend program, school ties, given a fixed budget and level of benefits. heads collect funds at the township level and Government pilot programs follow the second distribute cash to beneficiaries at the school. option as these would be eventually scaled up For the MCHVS, the MOH management agency nationwide.21 For instance, the MOE stipends provides reimbursement to the voucher distrib- program uses a simplified poverty scorecard utors, and health care providers and midwives (i.e. a set of indicators that are strongly corre- collect funds at health facilities and reimburse lated with poverty) and community validation beneficiaries. While this arrangement is the of the beneficiary lists (GoM, 2014c). Eligibility currently the most feasible option, transpar- verification and beneficiary enrollment typical- ency can be increased by separating these roles ly happens at the local level, with communities of supply of services, program implementation, playing an important role. and payment distribution. Social security con- tributions are collected in cash from employers In contrast, social security programs enroll all and cash benefits are paid to workers at the individuals within groups, regardless of loca- SSB township office. In the case of the civil ser- tion, as defined by the legislation governing vice pensions scheme, however, payments are these schemes. SSB township offices are re- made electronically by a third party – MEB – to sponsible for the registration of companies and pensioner bank accounts. workers. Civil servants are required to apply at the administrative unit of his or her depart- Record-keeping, management infor- ment; authorization is centralized at the union mation systems (MIS), and monitor- level. ing Systems for record-keeping and information management are largely paper-based and im- plemented locally. Individual-level information on beneficiaries remains at the facility (school 21. An example from the stipends program illustrates this trade-off between broad vs. narrow eligibility criteria. For instance, if all 8 million grade 1-11 students were to be provided stipends at the current level of benefits, an estimated budget of MMK500 billion would be required. The budget in 2014 was MMK3 billion; even with the doubling to MMK7 billion in 2015, the current outlay is only about 1.5 percent of the estimated budget required for categorical targeting of all school-age children. 22. The experience of social protection implementation in Myanmar suggests cash-based support is feasible in most contexts in the country. There are some circumstances where food-based benefits are more appropriate, such as in remote areas or areas with insecurity and/or social tensions. It is important for new programs to assess local contexts to determine which transfer modality is the most appropriate. See Notes on ‘The experience of cash transfers in Myanmar’ and ‘Developing scalable and transparent ben- efit payment systems in Myanmar’. 24 Building Resilience, Equity and Opportunity in Myanmar: The Role of Social Protection or health clinic) or township level; the union tion system to support program implementa- level has access only to aggregate indicators, tion and monitoring. Systems for internal and making tracking of beneficiaries difficult. Some external monitoring are also gradually being improvements are evident already, with recent strengthened. For instance, the stipends pilot initiatives towards automation. For instance, program has set up a system of process moni- MEB has implemented an e-pension database toring and spot-checks by third parties, with that issues e-pension cards to eligible pension- systematic institutional arrangements to facili- ers, lowering the number of visits to the bank tate the feedback of monitoring information to branch from two visits to one. The SSB is cur- policy-makers. rently piloting a new computerized informa- 5. Opportunities and challenges in building an effective social protection system for Myanmar Developing a social protection system can bring important payoffs to a country's develop- ment agenda. For Myanmar, this will mean putting the core building blocks in place to ad- dress the risks and vulnerabilities facing its people. At the program level, this would imply the following: a) In the short to medium term, pro- gressive development and expansion of social assistance coverage through government sys- tems. This would include scaling up recently introduced pilot programs (such as stipends), as well as the introduction of new pilot programs that can address unmet needs. The most likely candidate for the latter is a PWP, given its dual objectives of employment generation and asset creation, and the positive experience of DPs in implementing PWPs in Myanmar; b) in the long term, gradual integration and coverage expansion of the pension and SSB schemes in a sustainable manner. Taking this system-oriented approach in Myanmar will need the involvement of several insti- tutions and require coordination within government, rather than introducing several unco- ordinated programs supported by DPs. 5.1 Overview tization and sequencing, consistent with fiscal and implementation constraints. The following criteria were used to arrive at the short to me- Developing a social protection system can bring dium term recommendations: important payoffs to Myanmar’s development agenda. The analysis of poverty and vulner- • Building on existing government policies and ability identified a wide range of priorities for programs, and drawing on relevant experi- social protection policies and programs; many ence from Myanmar and globally: The RDSF of these needs are not adequately addressed and the SPSP provide the policy frameworks at present. This section provides some recom- for expansion of social assistance provision in mendations for developing an effective social Myanmar and clarify institutional mandates protection system in Myanmar, based on priori- to do so. In this context, building capacity Overview 25 and systems through existing government Based on these criteria, the following recom- programs and platforms is desirable. The ex- mendations emerge as the most important in tensive experience of DPs in implementing putting in place the core building blocks of an social assistance in Myanmar also provides effective, responsive, and sustainable social useful insights. Similarly, global experience protection system for Myanmar in the short to provides useful lessons for the social security medium term. The government’s existing pro- policy reforms being proposed. grams provide the basis to do so, although the opportunities and challenges vary across social • Finding the right balance between address- security provision and social assistance provi- ing multiple needs and fiscal sustainability: sion (see Figure 8). Myanmar’s multiple and pressing needs call for strengthening social protection delivery Setting up a high-level policy coordination on several fronts. However, fiscal sustain- mechanism can facilitate evidence-based and ability calls for the identification of the most systematic policy development, harmonize efficient and effective interventions to both programs and delivery systems, and develop address the needs of those most in need and mechanisms to pool government and DP re- to help lay a foundation for delivery systems sources sustainably. This is particularly critical for future coverage expansion. for ensuring coherence in the current policy de- velopments in social security schemes and miti- • Finding the right balance between compre- gating the risk of parallel systems for subgroups hensive programs and operational complex- of the population that would be difficult to rec- ity: The proposed ‘social assistance building oncile in the future. In addition, inefficiencies block’ (see below) would ideally consist of arising from this component of Myanmar’s so- programs that address multiple constraints, cial protection system could crowd out resourc- e.g., CTs that support both poverty reduc- es for the expansion of social assistance cover- tion and human capital development and age. For social assistance, such a mechanism PWPs that address income insecurity and lo- would provide an opportunity to expand cover- cal asset creation. At the same time, it will be age by pooling government and DP resources. important for the government to start with simple models at the pilot stage while deliv- Rebalancing the program mix to increase the ery systems are being developed and tested share of social assistance spending is an im- as part of government systems. Expanding mediate priority for Myanmar. The current coverage and scope of existing programs and social protection system (see Section 4) places introducing new programs and entitlements greater emphasis on social security provisions, could come in the medium and long term. whereas the priorities identified make a strong Figure 8: Putting in place the core building blocks in short to medium term High-level policy coordination and sustainable financing Social assistance programs New social assistance programs Social security programs (stipends, MCHVS; currently (PWPs; currently DP-led) (civil service pensions and SSB schemes) in pilot stage) • Address seasonal unemployment and • Help save for old age and protect • Address school dropouts and income insecurity against sudden drops in income malnutrion • MOE, MOH • Pilot implementation through DRD • MOF, SSB (potentially through NCDDP) 26 Building Resilience, Equity and Opportunity in Myanmar: The Role of Social Protection case for substantially greater investments in 5.2 Policy coherence and social assistance programs. Both the RDSF and the SPSP emphasize the need to transition to sustainable financing government-led social assistance provision in the short to medium term. There are several el- There is a strong case for expansion of social ements that can strengthen this transition: assistance coverage in the short to medium term; this will require a coherent action plan, • First, the recent pilot programs (such as sti- increased government budget, and effective pends) that help promote equitable and af- mechanisms for pooling government and DP fordable access to services provide a prom- resources. In this context, it would be more ising basis for large-scale national programs. efficient and fiscally sustainable to build on The short-term priority will be to strengthen existing programs (e.g. additional benefits and delivery systems and carefully evaluate and coverage for maternal and child health through monitor implementation. In the medium an expanded/modified platform of the MCHVS run, these programs can scale up and expand and MOH; for school-age children through sti- coverage to a larger share of the population. pends and MOE) and platforms (e.g. the NCDDP for introducing a new pilot PWP; see below). • Second, it will be important to pilot a modi- DPs can also help co-finance government pro- fied PWP to address seasonal unemployment grams, thereby facilitating harmonization and and income and food insecurity, while also providing an opportunity to ensure institution- building community infrastructure and resil- al and financial sustainability in social protec- ience to disaster risk. This can be designed tion delivery. Programs such as the stipends and piloted in the context of existing CDD and the NCDDP have already facilitated pooling platforms (such as the National Community- resources from various DPs to supplement the Driven Development Project (NCDDP)) to government budget. In the long term, this can provide locally relevant support at commu- be taken a step further to develop scalable pro- nity level. grams for more effective emergency response. Ethiopia’s PNSP demonstrates how an ad hoc • Third, strengthening systems and ensuring emergency response can be transformed into a fiscal sustainability should be a priority in comprehensive, sustainable, and scalable social civil service pension and SSB reforms before protection program by providing a platform to implementing new provisions that may strain pool DP resources (see Box 3). It also provides current systems and crowd out fiscal space an example of gradual strengthening of social for the above. The demographic outlook in protection provision through the strengthening Myanmar provides the space for such a grad- and expansion of a core program.23 ual expansion of social security coverage. With respect to social security provision, the The following sections provide the rationale for immediate priority in the short term is to as- these short-term recommendations and dis- sess the fiscal implications of the proposed pol- cuss how these building blocks can gradually icy reforms and develop appropriate coordina- evolve into a robust social protection system. tion mechanisms to avoid multiple schemes. This coordination goes beyond social protec- tion. For instance, the pension policy reforms need to be aligned with the government’s over- 23. Trying to pursue short-term social outcomes (education, health, and housing) and long-term income security (pensions) through the MPF will require a much higher degree of administrative capacity and record-keeping than a pure retirement account and adds to administrative costs. Overview 27 Box 3: Ethiopia’s PSNP – from ad hoc emergency response to comprehensive, sustainable, and scalable social protection provision Embedded in the government’s strategy and policy for food se- Scalability of PSNP curity and eradication of extreme poverty, the PNSP represent- ed a pivotal shift from annual emergency food aid appeals to a The financing mechanisms allow the program to scale up to planned approach to food security and predictable drought risk also provide assistance to risk-prone households when neces- management. sary. The RFM ensured financial commitments from donors were put in place before any crises occurred. In 2008, in re- The PSNP provides long-term support to chronically food-inse- sponse to increasing food prices and the failure of the belg cure rural households to help them cope with shocks, reduce rains, the government used the RFM to provide additional disaster risk through asset creation and rehabilitation of their transfers to 4.43 million existing PSNP beneficiaries and 1.5 mil- natural environment, and build household resilience. It con- lion individuals who had not previously participated but who sists of CFW projects to improve environmental assets (60 per- were negatively affected by the crisis. The RFM was again used cent of projects in soil and water conservation); unconditional in 2009 to provide additional transfers to 6.4 million PSNP par- food/cash transfers for the poorest 10 percent unable to work; ticipants. an insurance for work scheme in a particularly disaster-prone area; a contingency budget to respond to sudden emergency needs; and the Risk Financing Mechanism (RFM) for use when the contingency fund is insufficient. Evolution of PSNP The program has evolved significantly over the past decade and has become the backbone of social protection provision in the country. Phase I (2005-2006, USD70 million) focused on transi- tion from emergency relief to a productive and development- oriented safety net. Phase II (2007-2009, USD200 million) sig- nificantly expanded program coverage and focused on strengthening program procedures and delivery systems, in- creasing the productivity of public works, and developing more efficient financing instruments for risk management to ensure The PSNP’s long-term support has helped reverse the trend of predictable and timely responses to shocks. Phase III (2010- deteriorating livelihoods despite climatic shocks. Timely and pre- 2015, USD850 million) continues to consolidate program per- dictable assistance has enabled households to manage risk more formance and maximize long-term impacts on food security by effectively and avoid negative coping strategies and food insecu- ensuring effective coordination with other critical interven- rity. Thanks to economic growth and programs like the PSNP, 2.5 tions. The program’s next phase will further strengthen sys- million people have been lifted out of poverty since 2005 and the tems for social protection and DRM. It will be implemented in share of the population below the poverty line fell from 38.7 per- 411 districts, reaching up to 10 million food-insecure people cent in 2004/05 to 29.6 percent in 2010/11. per year, and has a total budget of approximately USD3.6 bil- lion from the government and 11 development partners. Source: World Bank (2014d). all plans for civil service human resource man- system could crowd out resources for the ex- agement. Similarly, the SSB reforms, especially pansion of social assistance coverage. with regard to health care services, need to be considered in the context of the government’s Implementing Myanmar’s poverty reduction broader efforts to ensure universal health cov- and development agenda will be greatly facili- erage. Tight coordination and planning are im- tated by an overarching coordination mecha- portant to avoid inefficiencies in the delivery nism that goes beyond ministerial mandates. system and in the use of public resources. In This will require strong political leadership at the absence of such mechanisms, there is a risk the union level, particularly in strengthening of parallel systems for subgroups of the popula- the links between social protection and rural tion that would be difficult to reconcile in the development. Indonesia’s model of high-level future. In addition, inefficiencies arising from policy coordination that links social protection this component of Myanmar’s social protection to poverty reduction objectives and provides 28 Building Resilience, Equity and Opportunity in Myanmar: The Role of Social Protection Box 4: Indonesia’s National Team for Accelerated Poverty Reduction (TNP2K) Indonesia has a comprehensive strategy for accelerating pov- TNP2K coordinates all four components of Indonesia’s com- erty reduction that is coordinated centrally by TNP2K. TNP2K prehensive acceleration of poverty reduction strategy: a) was established under the Vice-President’s Office through a household-based social protection programs, including a CCT Presidential Decree in 2010. It is a central body that is respon- (PKH), a UCT, in-kind assistance (e.g. Raskin), and support to sible for a) developing poverty reduction policies and pro- vulnerable groups, such as the disabled, senior citizens, or- grams; b) synergizing poverty reduction programs among min- phans, etc.; b) CDD programs such as PNPM Mandiri; c) liveli- istries/agencies through synchronization, harmonization, and hood promotion programs for small and micro enterprises; integration; and c) supervising the implementation of poverty and d) programs to increase access to basic services and im- reduction programs and activities. TNP2K provides high-value prove the quality of life for the poor. expert input into national discussions on policy design and implementation. It is assisted by the Control Working Group, Source: www.tnp2k.go.id/en/acceleration-policies/strategy- which coordinates and controls the implementation of poverty to-accelerate-poverty-reduction/overview-of-the-accelera- reduction programs. Apart from providing expert advice to tion-strategy/ policy design, TNP2K is responsible for delivering funds where budgeting of line ministries is insufficient and filling staffing gaps. strategic direction is relevant for Myanmar (see run. These are recently introduced programs, Box 4). and implementation systems are being gradu- ally developed. As a result, these programs are currently implemented by service providers Subnational roles and financing (schools, health clinics, midwives), with some support and oversight from township officials. Wide disparities in social protection priori- In the long run, separation of these roles of ties across Myanmar’s 14 states and regions supply of services, program implementation, suggests a national social protection system and payment distribution would be desirable. would need to balance a coordinated national Global experience indicates there are econo- approach with sufficient flexibility for locally mies of scale in setting up centralized delivery relevant programs. A more prominent role for systems for CTs within ministerial structures; region/state governments and local-level struc- more sophisticated coordination mechanisms tures in social protection programming, financ- are required for multi-sectoral CTs that address ing, and delivery would lead to an effective and multiple objectives. sustainable social protection system that ad- dresses local priorities and increases account- However, programs that address issues of ability to citizens. food and income security need to be strongly rooted in local needs and preferences, the lo- At present, social protection programs are fi- cal socioeconomic context, and existing com- nanced by the union-level social budget and munity-based mechanisms for support. In the implemented through ministerial structures. medium term, there are opportunities for fi- This approach is appropriate in the case of so- nancing social protection through intergovern- cial security programs, as pooling risks of dis- mental transfers for states and regional govern- ability, death, and longevity is more effective ments (such as the Poverty Reduction Fund) for and there are significant economies of scale locally relevant programs. Going one step fur- in administration. This approach is also appro- ther, even greater discretion can be provided priate for the pilot social assistance programs to local communities by providing community – stipends and the MCHVS – that aim to pro- block grants that incorporate some elements of mote access to services, at least in the short social protection (see Section 5.4). Overview 29 5.3 Rebalancing the program The following priorities emerge in the short to medium term:24 mix to address social protection priorities Scaling up existing cash transfers to invest in the health, nutrition, and In light of needs identified in Section 2, what is the appropriate mix of social protection poli- education of young children cies and programs for Myanmar? The current social protection system (see Section 4) places Investing in the health, nutrition, and educa- greater emphasis on social security provisions, tion of children can have significant long-term whereas the priorities identified make a strong benefits, for the children themselves as well as case for substantially greater investments in so- the economy in terms of a skilled, healthier, cial assistance programs. There is thus an imbal- and productive labor force. The first 1,000 days ance in the mix of social protection policies and of a child’s life are crucial in developing a sound programs, which needs to be corrected. Global foundation for better health, higher cognitive experience as described in Section 3 suggests a skills, lifelong learning, and future earning abil- number of program options to address Myan- ity. Equitable access to good-quality education mar’s needs over time. The emerging priorities is essential in building human capital and re- and program options are prioritized based on ducing the risk of current and future poverty. the criteria noted above. 24. See Notes on ‘Inventory of social protection programs in Myanmar’, ‘The experience of cash transfers in Myanmar’, and ‘The expe- rience of public works programs in Myanmar’ for further details on the listed priorities. 30 Building Resilience, Equity and Opportunity in Myanmar: The Role of Social Protection Social protection programs such as CTs (some- Addressing seasonal unemployment times combined with behavioral interven- and income and food insecurity, and tions) can help address malnutrition, school dropouts, and other development challenges. building resilient communities The experience of other countries and the pi- lots in Myanmar suggest CTs can reduce finan- cial and other barriers that constrain access to PWPs can be effective poverty reduction services, especially for the poor and vulnerable. mechanisms that protect households, help These can complement efforts to improve the build resilient communities, and promote so- supply and quality of education and health cial cohesion in Myanmar. The RDSF and the services. Together, these policies would help SPSP identify PWPs as key instruments to ad- reduce intergenerational poverty and promote dress the vulnerabilities of the rural popula- better human development outcomes and in- tion in Myanmar. They can be used to address clusive growth. issues of seasonal unemployment and food and income insecurity faced by landless and land-poor casual workers and farmers in ru- In Myanmar, the recently introduced pilot ral Myanmar, while at the same time building programs – MOE’s stipends and MOH’s MCH- community assets. By helping stabilize incomes VS – aim to promote access to education and at source, these programs can also contribute health care for children and pregnant women to reducing distress migration. PWPs can con- from poor households, and can provide the tribute to emergency response, post-disaster basis for large-scale national programs. The recovery, and building community resilience. government has made commitments to ensur- Lastly, they can play a particularly important ing education for all and providing universal role in national reconciliation efforts by provid- health coverage. Meeting these goals will re- ing practical approaches to peace-building and quire a combination of quality service provision social cohesion among communities affected and well-designed CTs to ensure equitable and by conflict. affordable access to services (see also Planning Commission, 2013). The short-term priority for The experience of the PWP models imple- both programs will be to strengthen delivery mented by DPs in Myanmar can provide the systems and carefully evaluate and monitor basis for developing a more stable safety net, implementation in order to continually improve financed and implemented by government. program design and delivery. In the short to me- However, given the regional diversity across dium run, these programs could gradually scale Myanmar’s agro-climatic zones, pilot PWP pro- up and expand coverage to a larger share of the grams would need to be designed to address population in need. In the long run, Myanmar specific local needs, based on detailed feasibil- can draw lessons from other countries, such as ity assessments. A key design question will re- Indonesia, that have built on the experience of late to setting an appropriate balance between implementing school stipends to design more objectives of asset creation and employment complex, multi-sectoral (health and education) generation in terms of the nature and labor- CCT models. intensity of projects. A pilot PWP can test mechanisms to inform design and implementation arrangements for a national program to respond to desired objectives. In terms of design, the pilot would need to determine eligibility (i.e. whether the pilot would allow people to self-select into Overview 31 the program or focus on the poorest and most households as part of CDD platforms. In ad- vulnerable) and appropriate mechanisms for dition, the effort made by the program to beneficiary selection, timing, and benefit lev- meaningfully include women in project com- els (e.g. that minimize distortions to local labor mittees presents an opportunity to achieve markets yet provide an attractive alternative to gender-sensitive community choices as part migration), and the nature of community infra- of an expanded menu. Finally, as the program structure assets. Two implementation modali- expands to conflict-affected areas, PWPs can ties can be explored: help with reconciliation efforts by including ex-combatants and ethnic minorities in com- • A national/regional PWP pilot model based munity development efforts. on current DP/NGO models. In current PWPs, project management structures do not explicitly make the linkage with govern- ment structures. There is potential to ex- The following additional priorities emerge for plicitly link PWPs to village and village tract the medium to long term:26 structures and build capacity beyond single- project implementation, and also to support the delivery of quality infrastructure through Mitigating and coping with disaster PWP models by linking with technical entities risk in government (e.g. DRD and other township- level officials). In the medium to long term, greater coordina- tion between social protection and DRM policy • A second option is to introduce key ele- frameworks would be required. In particular, a ments of traditional PWP models in com- flexible and scalable social protection program munity-based models of local infrastructure (such as Ethiopia’s PNSP, see Box 3) that can be development.25 This could be achieved by activated by early warning triggers would help strengthening the emphasis on pro-poor reduce the economic impact of disasters on investments and resilience-building in in- households. This would enable better risk-cop- frastructure development programs such as ing during response and early recovery, while the NCDDP. For instance, the NCDDP already also reducing disaster risk and building house- works through village tract administrators hold and community resilience in Myanmar. and supports engagement and capacity- building in the delivery of basic infrastruc- ture. It also generates important employ- Expanding social security coverage ment opportunities for local communities, with 80 percent of assets built directly by to the rest of the population communities. This platform can be extended to provide social protection. For instance, An additional priority for the long term would including features of PWPs such as labor- be to strengthen and expand existing pro- intensive approaches and encouraging the grams that help people save for old age and participation of the poor, including of wom- protect against sudden drops in income and en, through poverty targeting could provide welfare owing to health and other shocks. Cur- a safety net for income- and food-insecure rent social security programs cover just about 25. See Note on ‘Social protection delivery through community-driven development platforms: International experience and key con- siderations for Myanmar’. 26. See Notes on ‘Social protection for disaster risk management’ and ‘Strengthening social security provision in Myanmar’ for more details. 32 Building Resilience, Equity and Opportunity in Myanmar: The Role of Social Protection 3 percent of the population, and the vast ma- tain age, sometimes with additional poverty jority of workers employed in the informal sec- criteria) or a household-based CT that includes tor have little income security. Pensioners and older people as beneficiaries because they live workers in Myanmar will come to expect more in poor households rather than because they government support in coming years, reflecting are above a certain age. For instance, MSWRR likely changes in family support for older peo- is planning a social pension pilot program for ple and high out-of-pocket health expenditure. older people (aged above 90 years). A key con- The government needs to develop a long-term sideration would be fiscal affordability of these roadmap for expanding social security coverage options and interactions with elements of the based on a careful assessment of feasible op- social protection system, including civil service tions, including modeling of the fiscal implica- pensions, SSB schemes, and social assistance tions of alternative options. programs. A key consideration for Myanmar will be ad- dressing the three aspects of coverage, ad- equacy, and sustainability through an inte- 5.4 Strengthening delivery grated national strategy. Introducing multiple schemes using different delivery systems can systems create inequities and increase administra- tive costs. In the long run, a national pension scheme can in principle be introduced. The pri- The implementation of the policies and pro- vate pension scheme under the SSB has not yet grams described above has several implica- been defined, nor is implementation planned in tions for the gradual evolution of delivery the short term. Ideally, a fully integrated system systems in Myanmar. First, delivery systems for that harmonizes scheme design, institutional the civil service pensions and SSB schemes will arrangements, and delivery systems will be need to be strengthened through process sim- achieved. At a minimum, even if design param- plification and automation (see Box 5). Second, eters cannot be harmonized, there are likely to delivery systems for pilot CTs will need to be be important synergies as well as advantages further developed to support scale-up of these from interoperability between the two public programs. Third, for all government programs, and private (parallel) delivery systems. A grow- these systems will evolve gradually as changes ing number of countries globally and in the in the external macro environment (e.g. finan- region (e.g. Vietnam and Hong Kong) have ret- cial and telecommunication networks, regula- roactively integrated partially or fully their civil tory and supervisory frameworks, etc.) pro- service and private sector pension and other vide opportunities for further improvements. social security schemes. Doing so retroactively Fourth, these evolving program-specific deliv- can be both difficult and costly. ery systems need to be developed in a coor- dinated manner that facilitates harmonization In the long run, the challenge will lie in ex- in the medium to long run. While the relevant panding the coverage of the existing social sectoral ministries would guide the evolution of security programs to workers in Myanmar’s program-specific processes, an overarching co- large informal sector, particularly the poor ordination mechanism can provide a platform and vulnerable. For Myanmar, the most fea- for sharing information and lessons learned sible option is likely through a non-contributory and facilitate the development of harmonized transfer rather than an expansion of the cur- systems. rent payroll tax-financed system. This could take the form of social pensions (i.e. budget- Effective delivery systems ensure eligible ben- financed transfers to older people above a cer- eficiaries receive the correct amount of ben- Overview 33 Box 5: Strengthening social security delivery For pensions and SSB schemes, the main priority in the short to medium term would be to build on existing initiatives to fur- ther simplify and automate processes. In the long run, these delivery systems need gradually to become more client-centered through further simplification, with greater transparency, and with minimal compliance burden. Conversely, some of the reforms under consideration cannot be implemented without the right processes and systems. In the medium to long term, substantial improvements will be required. For example, introducing a new DC scheme rests on the abil- ity to track individual civil servants’ salaries, contributions, and pensions, so that contributions and accumulated benefits can be tied to individual accounts. As a result, investments will be required to upgrade the human resource MIS, an activity that has implications beyond pensions. Capacity for managing and regulating pensions and SSB funds will also need to be gradually developed. Source: ILO and MDRI (2015), Note on Strengthening social security provision in Myanmar. efits at the right time and with the right fre- ID card number) for program enrollment and quency, while minimizing the cost to both the payments can help track beneficiaries across program and beneficiaries (Grosh et al., 2008). programs. Myanmar can take advantage of the innovative ideas, the emerging emphasis on a coordinated • Selecting program beneficiaries can gradu- social protection provision, and the technologi- ally become more evidence-based, effective, cal solutions that have helped many countries and systematic, with a more harmonized ap- strengthen their social protection delivery sys- proach across programs. Given current data tems. and other constraints, a phased approach will be needed. In the short to medium term, These delivery system challenges differ be- a common platform for geographic target- tween social security programs and social as- ing will need to be developed within the sistance pilot programs. For the former, the government. However, the exclusive use of short-term priority is simplifying procedures geographic targeting needs to be carefully and reducing the burden on pensioners and assessed keeping in mind political and social SSB affiliates to enroll and receive their ben- factors, particularly in conflict-affected areas. efits (see Box 5). For the latter, the immediate Most social protection programs will typical- priority is to establish these systems, which will ly rely on a combination of methods, includ- eventually form the basis for large-scale nation- ing geographic targeting in the first stage, al programs as the pilots scale up. combined with some form of household or individual targeting in a second stage. In the medium to long term, household targeting There is scope to strengthen all stages of the systems will need to be further strengthened, delivery systems for government social protec- initially in the context of specific programs tion programs, as follows: but gradually evolving into an integrated sys- tem for reaching the poor and vulnerable. • Beneficiaries require identification when enrolling for the program and for receiv- • In the short to medium term, government ing benefits. Current systems for identifying social protection programs will likely con- beneficiaries rely on local functionaries, giv- tinue to deliver payments in cash. Scaling- en gaps in the coverage of national ID cards, up would require systems to be in place to particularly among the rural population and ensure accountability, transparency, and se- among the poorest and most vulnerable. Us- curity of funds. In the medium to long term, ing a unique personal identifier (such as an there is potential for a gradual transition to 34 Building Resilience, Equity and Opportunity in Myanmar: The Role of Social Protection more electronic payment systems, once the Increased role for subnational and financial and telecommunications markets other stakeholders develop and adequate legal and regulatory frameworks are in place. This will make the The roles of three key subnational stakehold- delivery of payments cheaper, quicker, safer, ers could be strengthened. First, township of- and more transparent. In addition, increas- fices already perform a number of often-un- ing the connectivity of the people of Myan- funded delivery functions. Increasing technical mar, especially the poor and vulnerable, to support to these offices and ensuring functions financial services will not only accelerate are adequately funded will be an important poverty reduction but also promote inclusive priority for the respective sectoral ministries economic growth. (MOE, MOH, and SSB). Second, the General Administration Department (GAD) is already • The emerging systems for communication an important stakeholder at local level; build- and outreach, grievance redress, record- ing on its technical functions but ensuring sup- keeping, and M&E will also need to be fur- port to the redefinition of its relationship with ther developed. Providing public information local communities around service delivery and having grievance and redress systems can support the institutional-strengthening of in place can improve transparency and cre- ministries and subnational units. Third, a mini- ates confidence in the program. Better re- mal but increasingly important form of sub- cord-keeping of beneficiary information and national government, village tract and village MISs also help improve accountability and levels, exists but so far has a limited role in the transparency. These provide opportunities implementation of social protection programs; to partner with DPs, non-governmental or- strengthening these structures to support so- ganizations (NGOs), and community-based cial protection delivery can promote account- organizations (CBOs) that have been active in ability to citizens and locally relevant schemes. social protection provision in Myanmar. The role of communities in raising awareness of DPs, NGOs, and CBOs can also support the programs and services, validating beneficiary transition to government-led social assistance selection, local-level oversight, and monitor- in several ways. This includes for example, ing will remain central. partnership with government as technical re- source persons or in specific stages of delivery, These program-specific systems should ideally such as grievance and redress mechanisms, develop in a coordinated manner, such that monitoring, and communication and outreach common tools and platforms emerge to sup- activities. The private sector can also play an port various programs and harmonization is important role, particularly in paying benefits feasible in the long run. The number and scale and upgrading program information systems. of government programs and services are ex- In all cases, support needs to be framed within pected to continue to expand. The introduction government-led priorities and programs, with of different delivery mechanisms can lead to decision-making processes ultimately falling in unnecessary and costly duplication of efforts on government hands. the part of program implementers and poten- tial confusion on the part of communities and Finally, the role of non-state actors in previ- potential beneficiaries. ously conflict-affected areas will need to be articulated with subnational leadership. Non- state systems will continue to operate, at least in the short term (Jolliffe, 2014), thus the initial focus can be on establishing institutional ar- Overview 35 rangements that a) build on existing founda- tions for service delivery that minimize disrup- Box 6: Synergies between household- tion in provision, support a smooth transition, based and area-based poverty reduction and avoid reigniting grievances; and b) recog- programs nize the need for locally adapted approaches to service delivery. In addition, social protection Global experience suggests increasing synergies between area-based poverty reduction programs and social assistance programs can address the needs of particular- programs is proving effective to enhance the effectiveness of ly vulnerable groups such as conflict-affected both approaches in achieving poverty reduction. Poverty re- communities and ex-combatants. duction strategies in several countries identify area-based programs along household-based social protection programs as necessary instruments to reach the poor. In practice, how- ever, both approaches have been implemented separately as institutional mandates often belong to different ministries. Piloting social protection provision through CDD platforms27 China has integrated social protection programs in area- based development programs since 1986. Impressive results in poverty reduction have been achieved, but, by design, the The RDSF provides an opportunity to explore poor outside the designated ‘poor’ villages have not been covered. From this perspective, the government’s new rural operational synergies between poverty re- poverty strategy points the way to greater convergence over duction and social protection programs; so- time of area- and household-based programs, including so- cial assistance, social services, and anti-poverty programs. cial protection programs such as PWPs could be delivered through CDD platforms. Embed- Countries like Indonesia have pioneered an even more inte- ding social protection programs as part of CDD grated approach, by making both CDD programs (such as PNPM Mandiri and Generasi) and household-based pro- platforms can be an effective approach while grams (e.g. PKH) follow centrally defined objectives of pov- delivery systems are still evolving. This is espe- erty reduction, with discretion on how to achieve this locally cially relevant for PWPs, as noted above, where in the case of PNPM Generasi. This has unleashed synergies at local level that promote more targeted and locally relevant a gradual transition from DP-led provision to solutions to social protection provision. a more sustainable government-led system is Source: Note on ‘Framework for the development of social needed. Providing options for testing locally protection systems’. relevant social protection models can facili- tate the emergence of responsive schemes and can build on local institutions and community- based mechanisms for collective action and so- In addition, using labor-intensive approaches cial protection (see Box 6). (where wage rates are used as a mechanism to increase the participation of the poor and vul- In this context, DRD could explore the feasibil- nerable) can directly provide income security to ity of embedding social protection programs beneficiaries. In addition, interventions for vul- in the NCDDP platform to enhance poverty nerable groups (e.g. older people with no fam- reduction impact. Options to do so include ily support) or to cope with specific shocks (e.g. strengthening the emphasis on pro-poor in- funeral grants) beyond pro-poor infrastructure vestments in infrastructure development and could be considered based on local priorities offering communities an expanded menu of op- and delivery capacity. tions that includes social protection interven- tions. For instance, pro-poor asset creation can be encouraged through active participation of poor households in community decision-mak- ing processes and a pro-poor menu of projects. 27. See Note on ‘Social protection delivery through community-driven development platforms: International experience and key considerations for Myanmar’. 36 Building Resilience, Equity and Opportunity in Myanmar: The Role of Social Protection 6. Conclusions ing complementarities between community development programs and social protection programs for comprehensive poverty reduc- Countries that have invested in building social tion; and c) ensuring social security provision is protection systems have seen significant re- sustainable and delivers promised benefits to ductions in poverty and inequality, translating workers and older people. into inclusive growth. Global experience sug- gests that these programs can facilitate com- In the short to medium term, it will be impor- munity cohesion and support peace building tant to start putting in place the core building efforts. Having scalable social protection pro- blocks of an effective, responsive, and sustain- grams and delivery systems in place can help able social protection system for Myanmar. people cope better and recover faster from The government’s existing programs provide disasters and other crises. Programs like PWPs the basis to do so, although the opportunities support community resilience and economic and challenges vary across social security provi- development through building community in- sion and social assistance provision. These in- frastructure and promoting connectivity. These clude: also help address the risks and vulnerabilities • At the policy level: Setting up a high-level facing households and individuals at different policy coordination mechanism that can stages of the lifecycle: school-feeding and CT avoid the risk of costly and unsustainable programs support school-age children, whereas parallel systems being developed and devel- PWPs, pensions, and social security programs op mechanisms to pool government and DP support working-age and older people. resources sustainably; • At the program level: Rebalancing the pro- Myanmar’s national policy frameworks in- gram mix to increase the share of social as- clude social protection as a core component of sistance spending, building on government its people-centered development and poverty current pilot programs (such as stipends) reduction strategy. While a broad set of policies and introducing a PWP pilot to build the will be needed to make households and com- foundation of large-scale national programs; munities more resilient to current and future • At the delivery level: a) Establish and gradu- poverty, as the RDSF and the SPSP emphasize, ally strengthen delivery systems for social these efforts would need to be complemented assistance delivery through government sys- by social protection programs to guard against tems, including through platforms such as vulnerability and increase access to opportuni- the NCDDP; b) simplify processes and auto- ties for all. mate systems for the more established social security programs. An effective social protection system can help the poor and vulnerable in Myanmar better Developing a social protection system can manage the opportunities and risks arising bring important payoffs to Myanmar’s de- from the ongoing reforms, and can help pro- velopment agenda. In the long run, an effec- mote household and community resilience tive social protection system for Myanmar will and social cohesion. Myanmar can avoid mis- provide adequate coverage of beneficiaries in takes other countries have made and that have need of assistance, improve responsiveness of prompted years of fragmented and ineffective programs to accommodate those made newly social protection provision elsewhere, by a) vulnerable because of systemic shocks, and ef- building a comprehensive system with a focus fectively connect beneficiaries of social protec- on pro-poor investments right from the start tion programs to additional productive oppor- through expanding social assistance; b) explor- tunities. Overview 37 References Alderman, H. and Yemtsov, R. (2013) ‘Produc- GoM (Government of Myanmar) (2014a) tive Role of Safety Nets’. Social Protection Dis- ‘Myanmar Civil Service Pension Reform’. Nay- cussion Paper 1203. Background Paper for the pyidaw: Pensions Department, MOF. World Bank 2012-2022 Social Protection and Labor Strategy. Washington, DC: World Bank. GoM (Government of Myanmar) (2014b) ‘Pro- posal to Set Up the Myanmar Provident Fund’. Andrews, C. (2009) ‘Levels and Patterns of Naypyidaw: GoM. Safety Net Spending in Developing and Transi- tion Countries’. Safety Nets Primer 30, January. GoM (Government of Myanmar) (2014c) ‘Rules Washington, DC: World Bank. and Regulations for Civil Servants’. 26 May. Nay- pyidaw: GoM. Andrews, C. and Kryeziu, A. (2013) ‘Public Works and the Jobs Agenda: Pathways for So- GoM (Government of Myanmar) (2015) ‘The cial Cohesion?’ Background Paper for the World 2014 Myanmar Population and Housing Cen- Development Report 2013. Washington, DC: sus: The Union Report’. Census Report Volume World Bank. 2. Naypyidaw: GoM. ASEAN (Association of South-East Asian Na- Grosh, M., el Ninno, C., Tesliuc, E., and Ouerghi, tions) (2008) ‘Post-Nargis Joint Assessment’. A. (2008) ‘For Protection and Promotion. The Bangkok: ASEAN. Design and Implementation of Effective Safety Nets’. Washington, DC: World Bank. Baird, S., McIntosh, C., and Ozler, B. (2010) ‘Cash or Condition? Evidence from a Cash HAI (HelpAge International, 2005) Ageways: Transfer Experiment’. Policy Research Working Practical Issues in Ageing and Development, Is- Paper 5259. Washington, DC: World Bank. sue 66, December. Devereux, S., Sabates-Wheeler, R., Slater, R., HAI (HelpAge International) (2013) ‘The Situa- Tefera, M., Brown, T., and Tesholme, A. (2008) tion of Older Persons in Myanmar: Results from ‘Ethiopia’s Productive Safety Net Programme the 2012 Survey of Older Persons’. Yangon: HAI. (PSNP): 2008 Assessment Report’. Brighton: IDS. IEG (Independent Evaluation Group) (2011) ‘So- cial Safety Nets: An Evaluation of World Bank Enlightened Myanmar Research and World Support, 2000–2010’. Washington, DC: IEG. Bank (2015) ‘Qualitative Social and Economic Monitoring (QSEM) of Livelihoods in Myanmar, IHLCA (Integrated Household Living Condi- Round 5’. Report commissioned by LIFT. tions Survey) Project Technical Unit (2010) ‘In- tegrated Household Living Conditions Survey Fiszbein, A., Schady, N. et al. (2011) ‘Condition- 2009/10’. Naypyidaw: GoM. al Cash Transfers: Reducing Past and Present Poverty’. Washington, DC: World Bank. ILO (International Labour Organization) and MDRI (Myanmar Development Resource Insti- GoM (Government of Myanmar) (2012) ‘Social tute) (2015) ‘Evaluation of Social Security Board Security Law’. Naypyidaw: GoM. Operations’. Technical Report. Yangon: ILO and MDRI. 38 Building Resilience, Equity and Opportunity in Myanmar: The Role of Social Protection Jolliffe, K. (2014) ‘Ethnic Conflict and Social Ser- World Bank (2011) ‘Social Protection for a vices in Myanmar’s Contested Regions’. Yangon: Changing India: Volume II’. New Delhi: World The Asia Foundation. Bank. LIFT (Livelihoods and Food Security Trust Fund) World Bank (2012a) ‘Social Protection Strategy (2013) ‘Annual Report 2013’. Yangon: LIFT. 2012-2022’. Washington, DC: World Bank. Planning Commission (2013) ‘Framework for World Bank (2012b) ‘Affordability and Financ- Economic and Social Reform’. Naypyidaw: Gov- ing of Social Protection Systems’. Africa Social ernment of Myanmar. Protection Policy Brief. Washington, DC: World Bank. SC (Save the Children) (2015) ‘Qualitative As- sessment: Consolidated Report Myanmar World Bank (2014a) ‘Going Universal: How 24 School Grant & Stipend Programme Phase 1 & Countries Are Implementing Universal Health 2’. Yangon: SC. Coverage from the Bottom Up’. Washington, DC: World Bank. Subbarao, K., del Ninno, C., Andrews, C., and Rodríguez-Alas, C. (2013) ‘Public Works as a World Bank (2014b) ‘Philippines: Are Cash Safety Net: Design, Evidence and Implementa- Transfer Programs Effective?’ From Evidence tion’. Washington, DC: World Bank. to Policy, Social Protection’. Washington, DC: World Bank. Thu, E.E. and Griffiths, M. (2012) ‘Strengthen- ing Community Based Social Protection Prac- World Bank (2014c) ‘Project Appraisal Docu- tices for Child Protection’. Yangon: SPPRG. ment on a Proposed Credit in the Amount of SDR 391.9 Million (USD 600 Million Equivalent) UNEP (UN Environment Programme) (2009) to the Federal Democratic Republic of Ethiopia Learning from Cyclone Nargis – Investing in the for a Productive Safety Nets Project 4’. Wash- Environment for Livelihoods and Disaster Risk ington, DC: World Bank. Reduction’. Geneva: UNEP. World Bank (2015a) ‘The State of Social Safety UNICEF (UN Children’s Fund) and MNPED (Min- Nets’. Washington, DC: World Bank. istry of National Planning and Economic De- velopment) (2010) ‘Multiple Indicator Cluster World Bank (2015b) ‘Systematic Country Diag- Survey 2009-2010’. Naypyidaw: UNICEF and nostic’. Yangon: World Bank. MNPED. World Bank (2015c) ‘Live Long and Prosper: Wetzel, D. (2013) ‘Bolsa Família: Brazil’s Quiet Ageing in East Asia and the Pacific’. Washington, Revolution’. Washington, DC: World Bank. DC: World Bank. WHO (World Health Organization) (2014) ‘Mid- World Bank (2015d) ‘Myanmar Public Expendi- Term Review of Maternal and Health Voucher ture Review’. Yangon: World Bank. Scheme’. Yangon: WHO. WFP (World Food Programme) (2013) 'Relief Assistance'. WFP factsheet July 2013