99518 MINISTRY OF POPULATION DEVELOPMENT AND SOCIAL PROTECTION REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA 2015 ОН FOREWORD The Ministry of Population Development and Social Protection has long been cooperating success- fully with the World Bank towards improving social welfare outcomes in Mongolia. With the support of the World Bank-financed Multisectoral Technical Assistance Project our collaborative efforts have been committed to undertaking some policy analysis and review of the existing welfare benefits, improving the welfare management information system,developing policy options for improving impacts of benefits and capacity building of the sector human resources. This report on Mongolia’s social welfare program design and beneficiary profiles provides valuable analysis for developing policy solutions for improving coverage and quality of social welfare services, effectively targeting the members of the households in need of social assistance, consolidating numer- ous benefits and increasing the effectiveness of resources used for welfare, especially at the times of economic challenges. The report will also be a great contr ibution to the process of developing social protection sector strategy. I would like to extend my appreciation to the World Bank social protection team and Ms. Junko Onishi for prepar ing this report and Mr. Jehan Arulpragasam for his guidance in addition to all those who have contr ibuted by providing their technical advice and comments. ERDENE SODNOMZUNDUI Member of Parliament Member of Cabinet Minister for Population Development and Social Protection FOREWORD Over the past several decades,Mongolia has transformed its economic and political system from one where most economic decisions were centrally made to one where outcomes are dr iven by markets. The results at the aggregate level have been clear: r ising production, better services, and higher living stand- ards for much of the population. At the same time,the vagar ies of the market can leave many vulnerable, susceptible to shocks,unable to share in the country’s increasing prosper ity. As many countr ies have done, Mongolia has sought to protect the vulnerable and ensure that the benefits of growth accrue to broader population through its system of social transfers. But how well are these transfers working? How could the system be changed to be even more effective? Before either of these questions can be answered,one must understand how those systems are working in practice: Who are the reci pients? Are the transfers effectively benefitting the poorest people? Using a detailed and comprehensive dataset, the authors of this report provide some answers. The findings are,in some ways,encouraging: Mongolia has an effective delivery mechanism for imple- menting social welfare programs and, compared to other countr ies around the world, Mongolia spends generously on social welfare. These facts suggest that the social welfare system as a whole has promise for achieving the objectives of helping the most vulnerable to avoid or escape poverty. The results also suggest,however,that some of the program features make it less effective than it could be. While the delivery mechanism is effective, and the aggregate transfers are substantial, many of the beneficiar ies are not those most in need. Only about 56% of total welfare spending goes to the poorest 40% of the population in the integrated household information database, while the top 40% received 28% of total welfare transfers. Mongolia’s current economic situation makes the efficiency of the social welfare system all the more important. Slowing growth is making the prospects of the poor and near poor increasingly precar i- ous—ensur ing the strength of the social safety net is more important than it has been for years. At the same time,falling commodity pr ices and slower growth of tax revenues serve to amplify the need for the programs to achieve their goals efficiently. As an institution devoted to the elimination of absolute poverty,the World Bank str ives to understand the dr ivers of poverty reduction and the factors that help ensure that prosper ity is broadly shared among the population. I hope that this review provides the evidence needed for Mongolia’s leaders to create a more effective and efficient social welfare system, one that puts Mongolia’s poor and vulnerable first. Resident Representative and Country Manager, World Bank JAMES ANDERSON REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA Contents Acknowledgements ..................................................................................................................................................... 9 Executive Summary ................................................................................................................................................... 11 1. Background ............................................................................................................................................................. 18 Data ........................................................................................................................................................................... 18 Limitations of the Study ........................................................................................................................................ 19 Structure of the report ........................................................................................................................................... 19 2. Social Welfare Programs .................................................................................................................................... 20 3. The Proxy Means Test and Data Source for the Analysis ........................................................................... 23 School attainment of adults ................................................................................................................................. 24 Disability ................................................................................................................................................................... 24 Work .......................................................................................................................................................................... 25 PMT score quintiles ............................................................................................................................................... 25 4 4. Key Findings............................................................................................................................................................ 27 Budgets and Expenditures on Social Welfare ................................................................................................... 27 Total Number of Beneficiaries ............................................................................................................................. 29 Coverage of Social Welfare Programs ............................................................................................................... 30 Equity in Benefit Distribution ................................................................................................................................ 31 Program Benefits Received by the Elderly......................................................................................................... 34 Program Benefits Received by the Disabled ..................................................................................................... 35 Program Benefits Received by the Poor ............................................................................................................ 36 5. Policy Implications and Recommendations.................................................................................................... 38 Consolidating Fragmented Programs ................................................................................................................ 38 Making Social Welfare Programs Poverty-Targeted ....................................................................................... 39 Increasing the Generosity and Coverage of Poverty Reduction Programs ................................................ 41 Strengthening Active Labor Market Programs that Serve as a Safety Net and Designing Social Welfare Programs that Promote Labor Force Participation .......................................................................... 41 Strengthening the Monitoring and Evaluation of Social Welfare Programs ............................................... 41 6. Conclusion ............................................................................................................................................................... 43 REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA References .................................................................................................................................................................... 43 ANNEX I: Review of Individual Programs ............................................................................................................... 44 ANNEX II: Inflation-Adjusted Expenditures for Transfers 2010 – 2013 ........................................................... 77 ANNEX III: Proportion of Beneficiaries in Each Social Welfare Quintile According to PMT Score.............. 78 List of Tables Table 1: Number of Individuals in the SW Admin/PMT Data by Aimag ........................................................... 23 Table 2: Highest Level of Education/Qualification Attained (18 years or older) ............................................. 24 Table 3: Individuals with Disabilities by Type of Disability .................................................................................. 24 Table 4: Individuals in the Database by Household PMT Score Quintile ......................................................... 25 Table 5: Average Household Size and Number of Children by Household PMT Score Quintile ................. 25 Table 6: Characteristics of the Household Head by Household PMT Score Quintile .................................... 26 Table 7: Beneficiaries and Expenditures by Program in 2013 (thousands MNT) .......................................... 27 Table 8: International Comparison: Percent of Total Benefits Received by the Poorest 20 Percent ......... 31 Table 9: Beneficiaries Receiving Social Welfare Pensions by Aimag ............................................................... 45 Table 10: Highest Level of Education Attained by Social Welfare Pensions Beneficiaries (18 years old and above) .......................................................................................................................... 45 Table 11: Social Welfare Pensions Beneficiaries with Disabilities, by Type of Disability.............................. 46 Table 12: Social Welfare Pensions Beneficiaries by PMT Score Quintile ........................................................ 46 Table 13: Beneficiaries Receiving Social Welfare Allowances by Aimag ......................................................... 48 Table 14: Highest Level of Education Attained by Beneficiaries of Social Welfare Allowances, Benefits, and Concessions and the Food Stamp Program ............................................................... 49 5 Table 15: Number of Individuals with Disabilities by Type of Disability among Beneficiaries of Social Welfare Allowances, Benefits, and Concessions and the Food Stamp Program ............. 49 Table 16: Beneficiaries of Social Welfare Transfers by PMT Score Quintile ................................................... 50 Table 17: Number of Individual Beneficiaries Receiving Social Welfare Services by Aimag ....................... 52 Table 18: Number of Individual Beneficiaries Receiving Social Welfare Services ......................................... 53 Table 19: Highest Level of Education Attained by Adult Users of Social Welfare Services ......................... 54 Table 20: Number of Individuals with Disabilities by Type of Disability among Beneficiaries of Social Welfare Services ............................................................................................................................ 54 Table 21: Number of Beneficiaries of Social Welfare Services by PMT Score Quintile ................................ 55 Table 22: Individual Beneficiaries Receiving Allowances and Assistance for the Elderly by Aimag .......... 57 Table 23: Highest Level of Education Attained by Recipients of the Allowances and Assistance for the Elderly ........................................................................................................................ 58 Table 24: Number of Individuals with Disabilities by Type of Disability among Beneficiaries of Allowances and Assistance for the Elderly .......................................................................................... 58 Table 25: Beneficiaries of Allowances and Assistance for the Elderly by PMT Score Quintile ................... 59 Table 26: Number of Individual Beneficiaries Receiving Benefits for the Elderly with State Merits and Average Number of Programs Received by Aimag ....................................................... 61 Table 27: Highest Level of Education Attained by Benefits for the Elderly with State Merits Beneficiaries ............................................................................................................................................... 61 Table 28: Individuals with Disabilities by Type of Disability among Beneficiaries of Benefits for the Elderly with State Merits ............................................................................................. 62 Table 29: Beneficiaries of Benefits for the Elderly with State Merits by Wealth Quintiles ............................ 62 REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA Table 30: Beneficiaries Receiving Allowances and Concessions for the Disabled by Aimag ..................... 65 Table 31: Highest Level of Education Attained by Recipients of the Allowances and Concessions for the Disabled ................................................................................................................. 66 Table 32: Individuals with Disabilities by Type of Disability among Beneficiaries of the Allowance and Concessions for the Disabled ..................................................................................... 66 Table 33: Beneficiaries of Allowances and Concessions for the Disabled by Household PMT score Quintile .................................................................................................................................... 67 Table 34: Beneficiaries Receiving Allowance for Pregnant and Lactating Mothers and Mother Heroes Benefits by Aimag ......................................................................................................... 68 Table 35: Highest Level of Education Attained by Beneficiaries of the Allowance for Pregnant and Lactating Mothers and Mother Heroes Benefits ....................................................... 69 Table 36: Individuals with Disabilities by Type of Disability among Beneficiaries of the Allowance for Pregnant and Lactating Mothers and Mother Heroes ............................................. 70 Table 37: Beneficiaries of Allowance for Pregnant and Lactating Mothers and Mother Heroes by PMT Score Quintile ................................................................................................................ 70 Table 38: Beneficiary Households Receiving the Food Stamp Program by Aimag ....................................... 71 Table 39: Highest Level of Education/Qualification Attained by the Household Head of Food Stamp Program Beneficiary Households .......................................................................................................... 72 Table 40: Household Heads of Food Stamp Program Beneficiary Households with Disabilities, by Type of Disability .................................................................................................................................. 73 Table 41: Beneficiary Households Receiving the Child Money Program by Aimag....................................... 74 Table 42: Number of Child Money Program Beneficiaries with Disabilities by Type of Disability .............. 75 Table 43: Child Money Program Beneficiaries by PMT Score Quintile ............................................................. 75 6 List of Figures Figure 1: Inflation-Adjusted Expenditures by Social Welfare Program Category 2010- 2013, Excluding the Child Money Program (taking 2010 as the base year) ............................................ 12 Figure 2: Percentage of Benefits Received by Individuals, According to Household PMT Score Quintiles and Source of Program Benefits .............................................................................. 13 Figure 3: Pro-Poor Programs: Percentage of Benefits Received by Individuals, According to Household PMT Score Quintiles by Program ..................................................................................... 13 Figure 4: Regressive Programs: Percentage of Benefits Received by Individuals, According to Household PMT Score Quintiles by Program ..................................................................................... 14 Figure 5: Expenditures by Social Welfare Program Category in 2013............................................................ 28 Figure 6: Inflation-Adjusted Expenditures by Social Welfare Program Category 2010- 2013, Excluding the Child Money Program (taking 2010 as the base year) ............................................ 28 Figure 7: Number of Beneficiaries by Social Welfare Program Category in 2013 ....................................... 29 Figure 8: Number of Beneficiaries by Social Welfare Program Category Excluding the Child Money Program 2010- 2013 .................................................................................................................. 30 Figure 9: Percentage of Individuals in the SW Admin/PMT Database Receiving a Social Welfare Program, and the Number of Program Benefits Received (excluding the Child Money Program) ............................................................................................................................ 30 Figure 10: Percentage of Benefits Received by Individuals According to Household PMT Score Quintiles and Source of Program Benefits .............................................................................. 32 Figure 11: Pro-Poor Programs - Percentage of Benefits Received by Individuals According to Household PMT Score Quintile, by Program ...................................................................................... 32 REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA Figure 12: Regressive Programs: Percentage of Benefits Received by Individuals According to Household PMT Score Quintile, by Program ...................................................................................... 33 Figure 13: Gender Distribution of Program Beneficiaries by Program ............................................................. 34 Figure 14: Proportion of Estimated Total Benefits Received by the Elderly, by Number of Programs Received and Share of Program ........................................................................................ 34 Figure 15: Proportion of Estimated Total Benefits Received by the Disabled 16 Years Old and Above, by Number of Programs Received and Share of Program ................................................. 35 Figure 16: Proportion of Estimated Total Benefits Received by the Disabled under 16 Years of Age, by Number of Programs Received and Share of Program .............................................................. 36 Figure 17: Proportion of Estimated Total Benefits Received by the Poor, by Number of Programs Received and Share of Program ........................................................................................ 37 Figure 18: Proportion of Food Stamp Program Beneficiaries among the First PMT Score Quintile by PMT Score ............................................................................................................................. 37 Figure 19: Trends in Inflation-Adjusted Annual Expenditures on Social Welfare Pensions (in 1,000,000 MNTs, using 2010 as the base year) ........................................................................... 44 Figure 20: Percentage of Individuals in Each Household PMT Score Quintile and the Proportion of Social Welfare Pensions Beneficiaries from Each Household PMT Score Quintile ............................................................................................................. 47 Figure 21: Trends in Inflation-Adjusted Annual Expenditures on Social Welfare Allowances (in 1,000,000 MNT, using 2010 as the base year) ............................................................................. 48 Figure 22: Percentage of Individuals in Each Household PMT Score Quintile and the Proportion of Social Welfare Allowance Beneficiaries from Each Household PMT Score Quintile .................................................................................................................................. 50 Figure 23: Trends in Inflation-Adjusted Annual Expenditures on Social Welfare Services (in 1,000,000 MNT using 2010 as the base year) .............................................................................. 52 7 Figure 24: Percentage of Individuals in Each Household PMT Score Quintile and the Proportion of Social Welfare Service Beneficiaries in Each Household PMT Score Quintile.... 55 Figure 25: Trends in Inflation-Adjusted Annual Expenditures on Allowances for the Elderly (in 1,000,000 MNT using 2010 as the base year) ................................................................. 56 Figure 26: Percentage of Individuals in Each Household PMT Score Quintiles and the Proportion of Beneficiaries of Allowances for the Elderly from Each Household PMT Score Quintile ............ 59 Figure 27: Trends in Inflation-Adjusted Annual Expenditures on Benefits for the Elderly with State Merits (in 1,000,000 MNT using 2010 as the base year) ....................................................... 60 Figure 28: Percentage of Individuals in Each Household PMT Score Quintile and the Proportion of Beneficiaries of the Benefits for the Elderly with State Merits by Household PMT Score Quintile .................................................................................................................................. 63 Figure 29: Trends in Inflation-Adjusted Annual Expenditures on Allowances and Assistance for the Disabled (in 1,000,000 MNT using 2010 as the base year) ................................................ 65 Figure 30: Percentage of Individuals in Each Household PMT Score Quintile and the Proportion of Beneficiaries of the Allowances and Concessions for the Disabled from Each Household PMT Score Quintile ............................................................................................................. 67 Figure 31: Trends in Inflation-Adjusted Annual Expenditures on Allowances for Mothers (in 1,000,000 MNT using 2010 as the base year) .............................................................................. 68 Figure 32: Percentage of Individuals in Each Household PMT Score Quintile and the Proportion of Beneficiaries of the Allowance for Pregnant and Lactating Mothers and Mother Heroes in Each Household PMT Score Quintile .................................................................................71 Figure 33: Percentage of Individuals in Each Household PMT Score Quintile and the Proportion of Child Money Program Beneficiaries in Each Household PMT Score Quintile .... 76 REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA 8 REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA ACKNOWLEDGEMENTS This report was prepared by the Social Protection and Labor team in Mongolia, Junko Onishi and Tungalag Chuluun. Valuable comments and inputs were provided at various stages of the analysis by Munkhzul Lkhagvasuren (Ministry of Population Development and Social Protection). Jehan Arulpragasam pro- vided overall guidance. Comments were received from James Anderson, Tae Hyun Lee, Nithin Umapathi, Franziska Gassmann, and Erdene-Ochir Badarch. Funding for this report was provided by the Rapid Social Response Multi-Donor Trust Fund supported by the Russian Federation, Norway, the United Kingdom, Australia, and Sweden. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors and should not be attributed in any manner to the World Bank, to its affiliated organizations, or to mem- bers of its Board of Executive Directors, or the countries they represent. 9 REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA Definitions used in this note Allowance and Assistance Cash allowance and reimbursements for the disabled. for the Disabled Allowance and Assistance Cash allowance and reimbursements for the elderly. for the Elderly Allowance for Mothers and This includes benefits for pregnant and lactating mothers and cash Children benefits for mother heroes. Benefits for the Elderly with This includes discounts and concessions for those with state honors. State Merits Social Welfare Allowance/ This includes cash allowances for: caretakers; individual member Transfers of household requiring social welfare assistance; emergency and livelihood support. Disability Includes physical and mental disabilities. Although program eligibility for the disabled is based on documentation of certification of disability, the profiles of disability in this report are based on self- reported disability status by the household respondent in the proxy means test database. Pensionable age group Men and women above the retirement age, which is 60 years old for men and 55 years old for women. PMT score quintiles The PMT score was calculated per household. These household PMT scores were divided into quintiles. SW Admin/PMT data In this note, this refers to the social welfare program administrative 10 data linked with the PMT data, using the individual IDs. The analysis of this database is the basis of the beneficiary profiles in this note. Social Welfare Services These include community-based services and specialized/institutional care. Social Welfare Pensions The pensions provided to the elderly who do not receive pensions from the Social Insurance Fund, the orphaned, the disabled, and single parents with four or more children under 18. Social welfare program In this note, it refers to all programs other than the Child Money Program. Therefore, this includes Social Welfare Pensions, Social Welfare Transfers, Social Welfare Services, Benefits for the Elderly with State Merits, Benefits for Pregnant and Lactating Mothers, and the Food Stamp Program. REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA EXECUTIVE SUMMARY The Government of Mongolia, through the Ministry of Population Development and Social Protection (MPDSP) and the General Office of Social Welfare Services (GOSWS), implements 71 social welfare programs. While some programs are needs-oriented, others are merit-oriented. All programs are categorically targeted with one exception: the Food Stamp Program, which targets the poorest 5 percent of the population by applying the Proxy Means Test (PMT) methodology.1 To reduce fragmentation, the Government wishes to review and consolidate its social welfare programs into a more effective and efficient system. The objective of this note is to make policy recommendations based on a review of social welfare program budgets and design and on an assessment of coverage and characteristics of program beneficiaries. The review utilizes an analysis of PMT data and social welfare program administrative data. Social welfare programs are defined in the Social Welfare Law (2012) and other legislation detailing the eligibility criteria for beneficiaries. These programs can be categorized into seven program groups according to the way the programs are administered: Social Welfare Pensions Social Welfare Allowance Social Welfare Services, including Community-based Services and Specialized Care Services Social Welfare Service Allowance for the Elderly with State Merit Allowance for the Elderly Allowance for the Disabled Allowance for Mothers and Children In addition to these social welfare programs, MPDSP also executes the Child Money Program (CMP), aimed at redistributing mineral wealth to the next generation. The CMP is funded from the Human Development Fund, separately from the state budget for social welfare. The program provides 11 cash amounts of MNT 20,000 (approximately USD 11) per month to all children from 0 to 18 years old. Despite its objectives and the funding source being separate from the social welfare system, since it is implemented by MPDSP along with other social welfare programs, it is often (although not always) considered to be part of the social welfare system. Given the CMP’s large fiscal implications, it is considered critical to include it in this review. I. Budgets and Expenditures on Social Welfare and Coverage of Programs In 2013, the government spent 2.78 percent of GDP on cash transfers for social welfare including the CMP (1.37 percent excluding the CMP). This amount is relatively generous compared to the average 1.6 percent of GDP being spent by developing and emerging countries. It suggests that the government is committed to allocating adequate fiscal resources for the social welfare of the population. In 2013, the largest budget allocation for cash benefits by far was for the CMP, comprising approximately half of the total allocation for cash grants but also covering the largest number of beneficiaries. The second largest expenditure was for Social Welfare Pensions (MNT 83 billion, approximately USD 50 million2), followed by the Allowance for Mothers and Children (MNT 67 billion, approximately USD 40 million3). The combined amount for these two larger program groups was still less than two-thirds of the budgeted amount for the CMP in 2013. Excluding the CMP, the expenditures for all programs increased slightly between 2010 and 2013, taking inflation into account (Figure 1). The largest increase was the two-fold increase for Social Welfare Pensions. The expenditures on Benefits for Pregnant and Lactating Mothers increased about 40 percent between 2011 and 2013, although the budget for Mother Heroes benefits decreased about 45 percent. 1. The PMT data was updated in December 2013, with preliminary assessments indicating good quality of data. 2. Using the 1,661 MNT per USD real exchange rate as of December 2013. http://www.mongolbank.mn/eng/liststatistic.aspx?did=3 3. This is derived by adding the budget for the Pregnant and Lactating Mothers program and the benefits for the Mother Heroes program, using the 1,661 MNT per USD real exchange rate as of December 2013. http://www.mongolbank.mn/eng/liststatistic.aspx?did=3. REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA Figure 1: Inflation-Adjusted Expenditures by Social Welfare Program Category 2010- 2013, Excluding the Child Money Program (taking 2010 as the base year) Source: Data from GOSWS. In terms of coverage, the CMP was the largest program in 2013, providing benefits to one million beneficiaries. This is more than twice the size of the second largest program, the Allowance for Mothers and Children with 405,000 beneficiaries. The smallest program is the Benefits for the Elderly with State Merits, with only 4,000 beneficiaries in 2012. 12 Half (49.4 percent) of all individuals in the SW Admin/PMT database received at least one program benefit when the CMP is included. This coverage of the CMP is impressive considering that the poverty rate in Mongolia was 27 percent in 2012. Only a handful of developing and emerging countries have coverage of social welfare benefits higher than the poverty rate (World Bank, 2014). The analysis confirms the overwhelming coverage of the CMP, with 94 percent of children 0-18 years old receiving the program. Interestingly, children 0-18 years old in every quintile received the CMP. Coverage drops considerably, however, when the CMP is not included, with only 19 percent of the population receiving at least one program. Beyond the CMP, the vast majority of individuals (81 percent) did not receive any social welfare program benefits. Around 14 percent received one program, 3.4 percent received two programs, and just above 1 percent received 3 programs. Only 0.5 percent received four or more programs. Social welfare programs were generally pro-poor, with a larger share of the benefits going to the poorer segment of the population. According to estimates based on the SW Admin/PMT database, the largest share of total benefits was received by individuals in the lowest social welfare quintiles, estimated by PMT scores.4 The poorest 20 percent received 34 percent of total social welfare transfers (Figure 2).5 However, it is important to note that in European and Central Asian countries, the share of total benefits going to the poorest 20 percent reaches over 40 percent (e.g., Croatia and Kosovo). In the best-performing countries around the world, the poorest 20 percent receive over 50 percent of total benefits (e.g., Argentina, Panama, and Peru). 4. The analysis of share of benefits received assumes that those who were listed as program beneficiaries in the administrative data were indeed receiving the full benefit amounts. For cash benefits, the amounts were annualized according to program design (benefit amounts and payment frequency). For reimbursements, it is assumed that all beneficiaries received their full amount in the year of the analysis (for programs that are provided less frequently than once a year), and the average reimbursement amounts were estimated using the 2012 expenditure figures obtained from GOSWS. 5 “The State of the Social Safety Nets 2014” states that the benefit incidence of total benefits in Mongolia going to the poorest 20 per- cent is 22.5 percent, based on analysis of household survey data (HSES 2007). The benefit incidence of 34 percent given in this report is estimated based on analysis of PMT and SW Administrative data. REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA Figure 2: Percentage of Benefits Received by Individuals, According to Household PMT Score Quintiles and Source of Program Benefits Share of 35% 30% Mothers 25% Disability Merit 20% Elderly 15% SW Transfers 10% SW Pensions 5% Child Money 0% Q1 Q2 Q3 Q4 Q5 Source: SW/PMT data, World Bank staff estimates. The largest program that contributed to the share received by individuals in the lowest social welfare quintile was the CMP (16 percent). As shown in Figure 2 above, the next highest contributors were Social Welfare Pensions (9 percent), Social Welfare Transfers including the Food Stamp Program (4 percent), and the Allowance for Mothers and Children (3 percent). Although most programs were categorically targeted, some programs were more pro-poor than others. The pro-poor programs (with a larger share of program benefits received by quintile groups 13 with lower PMT scores) were Social Welfare Pensions, Social Welfare Transfers, CMP, the Allowance for the Disabled, and the Allowance for Mothers and Children (Figure 3). Figure 3: Pro-Poor Programs: Percentage of Benefits Received by Individuals, According to Household PMT Score Quintiles by Program6 Source: SW/PMT data, World Bank staff estimates. 6. In Figure 3, the first bar (% individuals in quintile) for each quintile depicts the share of individuals in that quintile. Therefore, program benefit shares higher than the first bar indicate that the quintile group received disproportionately more program benefits. If program shares are lower than the first bar, it means that the quintile group received disproportionately less program benefits. REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA In contrast, some programs were less pro-poor (regressive). These programs were: the Benefits for the Elderly with State Merits, and the Allowance for the Elderly. In particular, the proportion of Benefits for the Elderly with State Merits that went to the non-poor is stark, with the vast majority of beneficiaries coming from the highest three social welfare quintile groups according to PMT scores. The fact that the Benefits for the Elderly with State Merits were enjoyed by the non-poor is notable since the benefit amounts of this program were the most generous compared to other programs. Figure 4: Regressive Programs: Percentage of Benefits Received by Individuals, According to Household PMT Score Quintiles by Program7 Source: SW/PMT data, World Bank staff estimates. The gender distribution of beneficiaries differed across programs. The CMP and the Allowance 14 for the Disabled both had equal numbers of male and female beneficiaries. The beneficiaries of programs associated with motherhood—including many under the Social Welfare Transfers—were naturally women: beneficiaries of the Allowance for Mothers and Children were all female, and around 58 percent of Social Welfare Transfers beneficiaries were women. More women received the Allowance for the Elderly (69 percent) and Social Pensions (54 percent), most probably due to the lower age of eligibility (55 years old for women compared to 60 years old for men) and the longer life expectancy of women compared to men. The only exception was the Benefits for the Elderly with State Merits, which benefited men (82 percent) far more than women. II. Program Participation by Beneficiary Type The elderly: Around 70 percent of pension-aged individuals benefited from at least one social welfare program, not including social insurance pensions which are not part of this analysis. About 76 percent of total social welfare benefits received by the elderly went to those who benefited from multiple programs, which may be a strong indication of the fragmented nature of programs addressing the needs of the elderly. The disabled: Around 66 percent of the disabled received benefits from at least one social welfare program. Around 63 percent of the disabled over 16 years old benefited from at least one program, while 87 percent of the disabled under 16 years old benefited from at least one program. Among the disabled above 16 years of age, Social Welfare Pensions comprised the largest proportion (76 percent) of benefits that went to the disabled. Around 43 percent of the benefits were received by those benefiting from only one program (predominantly Social Welfare Pensions), while those benefiting from multiple programs received a larger share of their benefits from Social Welfare Transfers and the Allowance for the Disabled. 7. In Figure 4, the first bar (% individuals in quintile) for each quintile depicts the share of individuals in that quintile. Therefore, program benefit shares higher than the first bar indicate that the quintile group received disproportionately more program benefits. If program shares are lower than the first bar, it means that the quintile group received disproportionately less program benefits. REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA The disabled under 16 years of age are not eligible for Social Welfare Pensions, and 81 percent of all benefits to this group were provided by the CMP. Although only about one in five disabled children under age 16 received benefits from three or more programs, when they did, they received a combination of the CMP, Social Welfare Transfers, and the Allowance for the Disabled. The poor: Around 62 percent of individuals from households in the poorest quintile (according to PMT scores) were direct beneficiaries of at least one social welfare program, but the figure dropped to 29 percent when excluding the CMP. The majority of the poor received only one program—about 64 percent of total benefits going to the poorest quintile went to beneficiaries who received only one program, including the CMP. The largest share of benefits for the poor came mainly from the CMP. The Food Stamp Program was the only program targeted at the poor: all 15,118 household beneficiaries were among the poorest 2.7 percent (according to PMT scores). Among the bottom 2.7 percent, 85.6 percent of households were beneficiaries of the Food Stamp Program. III. Policy Recommendations A. Consolidating Fragmented Programs Three strategies are recommended in consolidating the existing social welfare programs: • Consolidate small programs by prioritizing program design based on the need to provide comprehensive support to the targeted population so beneficiaries do not need to access multiple programs. For example, although the disabled derive their benefits predominantly from Social Welfare Pensions, those who receive multiple programs are combining the Allowance for the Disabled and Social Welfare Transfers. This suggests that in order to better address the needs of the disabled, elements of these two programs may be combined into a more comprehensive program. In designing such a program, it would be important to take into account the different needs of male and female beneficiaries, in addition to the specific needs of the target group. • Strengthen programs that are needs-based and gradually phase out entitlement-based programs. • In the future, consider transferring health-related benefits to be covered by Health Insurance. 15 B. Making Social Welfare Programs Poverty-Targeted To contain costs of social welfare programs while making them more effective, programs could be made poverty-targeted. For example, benefit amounts for the richer quintiles could be reduced or eliminated, using the readily available PMT database. The CMP is expected to have a considerable poverty impact, but the level of program expenditures may be difficult to sustain. As discussed above, the program has wide coverage and a large benefit amount. At the same time, the total budget for transfers of this program alone was MNT 240 billion (approximately USD 130 million) in 2013, accounting for 51 percent of total transfers made through social welfare programs in Mongolia. Considering the fiscal deficit faced by the government, this level of expenditures on the CMP may be difficult to sustain. Taking these facts into account, possible options going forward include: • Continue the CMP as is as the program is effectively reaching the poor, if the fiscal environment allows for the program to continue; • Introduce program design features to the CMP such that the poor receive larger amounts, while maintaining universal coverage of the program. This could be achieved by using the PMT to identify children in poor households or through categorical means of providing larger amounts for children in households with many (three or more, for example) children. Altering program benefits for children in different types of households will make program administration more complex, and if poverty targeting is introduced, strong operational design and implementation plans would need to be prepared to address the potential exclusion of deserving children. Reallocation of benefits could be done in a cost-neutral way or in a way that reduces total program cost: o To maintain cost neutrality, the grant amounts reallocated to the poor would be equal to the reduction in grant amounts for the better-off. o To reduce program cost, the sum of the grant amounts reallocated to poorer children would be lower than the sum of grant amounts reduced for better-off children. REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA • Terminate the CMP and replace it with a new poverty-targeted cash transfer program. Termination of the CMP will give the government fiscal space to design a poverty-targeted welfare program, with clear poverty alleviation objectives. Two design options could be: o Implement the Poverty Benefit program stipulated in the Social Welfare Law (2012).8 The new program would need to provide significant levels of coverage and benefits to compensate for the loss of benefits experienced by poor households due to termination of the CMP, with a strong operational design to address the potential exclusion of poor households. The advantages are that the program would have a clear poverty alleviation objective, and depending on coverage and benefit amounts, it could lead to considerable budget savings. The disadvantages are that this will be a new program, and as with any new program, considerable operational preparation will be required. o Add a child benefit component to the new Poverty Benefit Program described above to maintain the focus on providing support to the next generation (as in the CMP). Again, it would be very important to ensure that this new Poverty Benefit Program (or at least the child component) would be large enough in terms of coverage and benefit amounts to compensate for termination of the CMP. If the child benefit component focuses on the poorest 40 percent of households, this would still cover half of the children currently covered by the CMP, as poorer households have more children. Adding a child benefit component to the Poverty Benefit Program would increase the complexity of program design as the transfer amounts per household would differ according to the number of children. Further simulation analysis is recommended to identify the most effective scenario for poverty impact. The HSES data could be used to determine the optimal benefit size and the methods and threshold for targeting. Introduction of design changes must be accompanied by strengthening of certain features of program operations, particularly if poverty targeting will be introduced: • Strong communication campaigns would be critical. 16 • Introducing targeting, whether through PMT or other categorical means, is bound to introduce exclusion error (eligible beneficiaries being excluded). Therefore, an effective grievance redress mechanism would be essential. • If targeting will be conducted through application of the PMT, continued efforts would be needed to ensure that the PMT database is up-to-date and complete and addresses grievances from individuals regarding their registered level of welfare according to the PMT scores. C. Increasing the Generosity and Coverage of Poverty-Targeted Program(s) Currently, total estimated transfer amounts from all programs going to poor households are small. The average household transfer amount is approximately 12 percent of household consumption at the poverty line for a household of five members. Most countries provide social welfare benefits that represent 20-35 percent of the poorest households’ post-transfer consumption, although there is considerable variation worldwide. The most generous countries in the Europe and Central Asia region provide benefits that represent between 40-55 percent of the poorest households’ post-transfer consumption (World Bank, 2013). There is an urgent need to focus the attention of social welfare programs in Mongolia on ensuring that the poor are better protected from economic shocks. Increasing the generosity and coverage of programs to protect the poor could involve implementing the Poverty Benefit Program as described above or (in the short term) revamping the Food Stamp Program. D. Strengthening Active Labor Market Programs that Serve as a Safety Net for the Vulnerable To provide an effective safety net for vulnerable workers, it would be important to have effective 8. The direct translation of the program title in the Social Welfare Law from Mongolian is: “Benefit for members of HHs in need for social assistance” (Poverty benefit). REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA and robust Active Labor Market Programs. The current social welfare system provides no safety net support for those who are poor (but not as poor as the bottom 5 percent) or near-poor who do not meet the eligibility criteria for existing social welfare programs. Laborers with temporary jobs or seasonal jobs do not qualify for unemployment benefits due to program design, yet they are the ones who may be most affected by an economic adjustment or other external shocks. Similarly, for social welfare beneficiaries who are currently unemployed or under-employed, social welfare programs could be linked with co-responsibilities for able-bodied beneficiaries to engage in work/skills accumulation activities. Welfare program beneficiaries are entitled to benefits as long as they remain within the categorical eligibility criteria. While the government needs to increase the generosity and coverage of poverty-targeted programs, such programs should not create dependency but encourage beneficiaries to engage in labor. New generation work activation programs could be introduced for welfare beneficiaries who are unemployed, seasonally unemployed, or underemployed, applying the principles of “welfare-to-work” transition approaches proven to be effective in many middle-income countries. E. Strengthen Monitoring and Evaluation of Social Welfare Programs Stronger program monitoring and evaluation will enable the government to make more informed policy decisions. To date, no impact assessment of programs has been conducted, and the government has little understanding of program impact and how the programs can be improved. To help inform future decisions, the government could: • Request the addition of a specific module focused on social protection to the HSES questionnaire, or at a minimum request revision of the HSES questionnaire “other income” section to distinguish social welfare and social insurance benefits. • Fully utilize the social welfare management information system (MIS) when it is unified and put online. 17 • Conduct periodic reviews of programs by analyzing administrative data to inform decision- making. REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA 1. BACKGROUND The Government of Mongolia through the Ministry of Population Development and Social Protection (MPDSP) and the General Office of Social Welfare Services (GOSWS) implements 71 social welfare programs. These programs vary from housing for war veterans to social pensions for men 60 years old and above and women 55 years old and above. Some programs are needs-oriented, while others are merit-oriented. All programs are categorically targeted, with numerous and generous benefits particularly for individuals with special state merits, a heritage from the socialist period. The one exception is the Food Stamp Program, initiated as part of an Asian Development Bank (ADB) investment project, which targets the poorest 5 percent of the population by applying the Proxy Means Test (PMT) methodology.9 The Government wishes to review and consolidate its social welfare programs. Given the decentralized nature of current program data management, the MPDSP has little grasp of the profile of program beneficiaries: who the program beneficiaries are or how many programs they receive. The objective of this note is to review the budgets, design, eligibility, and benefit levels of social welfare programs; assess program coverage; describe the basic characteristics of program beneficiaries; and make policy recommendations. The current set of social welfare programs is based on the Social Welfare Law (SWL), enacted in 2005 and last amended in 2012. The Law regulates the issues related to social welfare and provides the main framework for four types of benefits: social welfare pensions, social welfare allowances, social welfare services, and social development services. The Law defines the categories of beneficiaries, benefits, and institutional and financial arrangements for each of the social welfare programs. The Social Welfare Law is planned to be amended in 2015, and it is hoped that this report can contribute some inputs to the amendment of the Law and the development of the Social Protection Sector Strategy. 18 Data The data for this study was provided by the MPDSP and GOSWS. Data on individual characteristics are from the 2013 PMT data collected by GOSWS. This data was merged by GOSWS with the program administrative beneficiary data, managed at the GOSWS district and aimag offices. Merging of these two databases was possible thanks to the unique national individual identification (ID) numbers recorded in both databases. The beneficiary data provides a snapshot of who the beneficiaries were as of June 30, 2014. For benefits provided less frequently (such as once every three years or five years), beneficiary status was recorded as if they had availed of/benefited from the service since June 30, 2009. Given the high population coverage of the PMT, the analysis assumes that the program coverage presented in this study is a proxy for national coverage of these programs. The 2013 PMT data collection set out to cover all the households in the country, in census style. The combined PMT and administrative data (hereafter referred to as the SW Admin/PMT data) contained a total of 2,636,370 individual records. However, 257,302 individual records from the administrative data could not be merged with the PMT data. These 257,302 individuals were not included in the analysis because without linking to the PMT data, there was no information on household or individual characteristics of these individuals that could be included in analyzing beneficiary profiles.10 Nonetheless, it is important to note that three-quarters of these individuals are recipients of the Child Money Program, which by definition suggests that they are children between 0-18 years of age. Information on expenditures on allowances and budgets for 2010, 2011, and 2012 and the budget summary data for 2010-2015 were obtained from GOSWS. These expenditures and budgets are broken down by program, listing the number of beneficiaries per program. These expenditures and 9. The PMT data was updated in December 2013, with preliminary assessments indicating good quality of data. 10. Although information on individual characteristics of these individuals is unavailable, the programs they received are known: around 3 percent (7,981 individuals) received SW pensions, none received SW transfers, around 0.3 percent (836 individuals) used SW services, 0.2 percent (474 individuals) received Benefits for the Elderly with State Merits, and none received the Food Stamp Program. REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA budgets only include the budgets for transfers and reimbursements provided to beneficiaries, not the operational/administrative costs of running these programs. Limitations of the Study Given the data issues described above (with 10 percent of individual records not linked to the PMT data), it is not evident how biased the data is when presenting overall program coverage. The total number of beneficiaries per program in the database does not neatly match with the number of beneficiaries reported in the GOSWS expenditures/budgets. However, the data does allow for presenting a good picture of the personal characteristics of program beneficiaries and program coverage overlap. Another limitation is that the administrative data that was linked to the PMT data only provides information on whether an individual received a specific program between July 2009 and June 2014 at least once. The data does not indicate how many times the individual received the same program, when the benefits were received, or how much was received. The estimated benefits presented in this note are based on how much individuals are supposed to receive as beneficiaries according to the program design. Similarly, for benefits that are provided less frequently than once a year, the estimation assumes that the benefit was provided in the year of the analysis. This estimation of individual benefits therefore represents the maximum amount an individual receives in a given year. For benefits and assistance that provide reimbursements (rather than a fixed cash amount), the average amount per beneficiary was based on 2012 expenditure data.11 Structure of the report The report is structured as follows. The report begins with a summary of programs reviewed, a description of the PMT targeting system, and the profile of individuals in the database. It then presents key findings from the review of budgets and the analysis of SW Admin/PMT data on program coverage and distributional equity of program benefits. The report concludes with a discussion of policy implications and recommendations that emerged from the key findings and the process of data 19 gathering and analysis. The Annexes include program descriptions based on the corresponding laws; the estimated benefit levels per individual; government expenditure/budget trends for each program since 2010; and information on social welfare program beneficiaries, by program grouping.12 11. The expenditures for 2012 were used since the more recent 2013 figures obtained by the authors were budget data, not the actual expenditures. 12. The program groupings include: Social Welfare Pensions, Social Welfare Allowance, Allowance and Assistance for the Elderly, Ben- efits for the Elderly with State Merit, Allowance and Assistance for the Disabled, and Allowance for Mothers and Children. Although the Food Stamp Program is included as a Social Welfare Allowance, the Food Stamp Program and Child Money Program are also reviewed separately. REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA 2. SOCIAL WELFARE PROGRAMS The Ministry of Population Development and Social Protection (MPDSP) currently executes 71 social welfare programs as well as the Child Money Program (CMP).13 The social welfare programs aim to provide minimum social welfare to eligible individuals, while the near-universal CMP aims to redistribute mineral wealth to the next generation. This section provides a brief description of the programs analyzed for this report as well as how they were categorized.14 The social welfare programs are defined in the Social Welfare Law (2012) and other legislation detailing the program eligibility criteria. All but one program are categorically targeted, providing a combination of cash transfers and reimbursements. The Food Stamp Program is the only poverty- targeted program and bases program eligibility on the PMT data. The social welfare programs can be categorized into seven program groups according to the way the programs are administered: Social Welfare Pensions (cash allowance) Social Welfare Allowance (cash allowance) Social Welfare Services, including Community-Based Services and Specialized Care Services (service provision) Social Welfare Service Allowance for the Elderly with State Merit (cash allowance and reimbursements) Allowance for the Elderly (cash allowance and reimbursements) Allowance for the Disabled (cash allowance and reimbursements) Allowance for Mothers and Children (cash allowance) 20 The programs and benefit amounts, as well as examples of types of benefits provided, are summarized below. Social Welfare Pensions provide a monthly cash allowance for citizens who are not entitled to a pension under social insurance legislation. These include: senior citizens15; vertically challenged (“dwarf individual”) persons 16 years of age and above; disabled persons 16 years old and above; children under 18 years old without foster parents; and single mothers 45 years of age and above or single fathers 50 years of age and above with four or more children under 18 years old. The monthly transfer amounts are MNT 115,000 (approximately USD 62). Social Welfare Allowances are provided for those who require social welfare assistance and emergency livelihood support, including pregnant women and mothers with infants and those who are taking care of others requiring care. The allowances vary in amount and frequency and include: The allowance for caregivers, which provides MNT 52,800 (approximately USD 29) per month; One-time assistance of MNT 1,200,000 (approximately USD 649) for young adults (18-24 years old) who became orphans before turning 18 years old or for households who lost their shelter due to a disaster or accident; One-time assistance of MNT 1,000,000 (approximately USD 541) for those with twins and MNT 3,000,000 (approximately USD 1,622) for those with triplets; • An annual allowance of MNT 120,000 (approximately USD 65) for single parents with three or more children under 14 years of age; • A quarterly allowance of MNT 60,000 (approximately USD 32) for individuals requiring permanent nursing care; and 13. The programs are listed in Annex II. 14. More detailed program descriptions are provided in Annex I. 15. Defined as men aged 60 years old and above, women aged 55 years old and above. REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA The Food Stamp Program, which is included in this group of programs for the purposes of this analysis. The Food Stamp Program merits attention because, as stated above, it is the only program to date that applies the PMT data for eligibility. It targets the bottom 5 percent of households in terms of welfare status, according to their PMT scores. The program provides food stamps to purchase food, equivalent to MNT 10,000 (approximately USD 5) per adult per month and MNT 5,000 (approximately USD 3) per child per month. Social Welfare Services include Community-Based Services and Specialized Care Services. Community-Based Services include services such as counseling, rehabilitation, temporary accommodation and care, day care service, and home-based care and services. These services are targeted at groups such as the elderly, the disabled, and victims of violence, citizens released from prison, or alcoholics and drug addicts. Specialized Care Services are institutional care services targeted at the single elderly, single disabled persons, and children in difficult circumstances. Social Welfare Service Allowance for the Elderly with State Merits has its legal basis in the “Law on Additional Benefits for Senior Citizens, who are the state hero, labor hero, and with the people’s honor and the state honor“(2008) which specifies financial benefits from the government for senior citizens with state merits (such as war veterans and those awarded Hero of Mongolia, Labor Hero, and other honors). The benefit types include cash assistance of MNT 200,000 (approximately USD 108) per month for veterans of war or surviving spouses and cash assistance of MNT 150,000 (approximately USD 81) per month to winners of the Mongolian State Prize, state nominees, Udarnik of the State, and senior veterans of the revolutionary struggle. In addition, reimbursements and vouchers for nursing homes and sanitariums (cash value of MNT 140,000, or approximately USD 76) and annual assistance for apartment and fuel costs (MNT 140,000) are also provided. Allowance for the Elderly provides assistance such as reimbursements for prosthetics, vouchers for nursing homes and sanatoriums, and reimbursements for transportation for use of health facilities and nursing homes. Most of this assistance is provided as reimbursements and in the form of vouchers. The reimbursements for prosthetics, given once every five years, are for body parts such as false teeth, 21 fingers, and toes as well as for hearing aids and spectacles. Allowance for the Disabled is for individuals with no capacity or limited capacity to participate in social relations like other persons due to body defect or mental, psychological, and perceptional disorders. Assistance to the disabled includes reimbursements for services such as: annual financial assistance to pay for apartment fees or fuel costs; reimbursements for rehabilitation and prosthetic correction; transport to and from school (or provision of a bus) for minors with disabilities and their guardian or caregiver; a one-way transportation fee and 50 percent of the rest and treatment fee at a sanitarium once a year; and land phone fee for the blind. Most of these allowances are reimbursements of fees and are not assigned a monetary value, except for: MNT 140,000 (approximately USD 76) for the annual apartment fee or fuel cost allowances, MNT 200,000 (approximately USD 108) annually for school transportation costs for children with disabilities and their guardians, and MNT 20,000 (approximately USD 11) monthly for the communication allowance for the blind. Prosthetics for specific body parts are covered as well as equipment such as wheelchairs, neck supporters, and walking sticks for the blind. Allowance for Mothers and Children includes two types of programs: allowances for pregnant and lactating mothers and allowances for the “Mother’s Glory” (also known as the Mother Heroes) program, which recognizes mothers who have given birth to four or more children. Once a Mother Hero, a mother receives benefits for the remainder of her life. The allowance for pregnant and lactating mothers is a monthly allowance of MNT 40,000 (approximately USD 22) provided for 12 months starting from the fifth month of pregnancy. Mother Heroes 1st medal recipients receive an annual benefit of MNT 200,000 (approximately USD 108), while 2nd medal recipients receive an annual benefit of MNT 100,000 (approximately USD 54). Although the allowance for pregnant and lactating mothers is provided only when a women is pregnant or lactating, the Mother Heroes benefits are provided for a lifetime once the person becomes an honoree. These two programs—particularly the Mother Heroes benefits—could be considered programs to promote population growth rather than to provide social welfare assistance for beneficiaries to meet minimum needs. REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA • In addition to these social welfare programs, MPDSP executes the Child Money Program (CMP). The CMP is regulated by the Law on the Human Development Fund for the purpose of distributing resources generated from mining revenues to citizens. As it is not regulated by the Social Welfare Law or related legislation, it is not strictly a social welfare program. The CMP is funded from the Human Development Fund, separately from the state budget for social welfare. It provides cash amounts of MNT 20,000 (approximately USD 11) per month to all children 0- 18 years old. Despite its objectives and the source of funding being separate from the social welfare system, since it is implemented by MPDSP along other social welfare programs, the CMP is often (although not always) considered to be part of the social welfare system. Given its size, the CMP has major fiscal implications and therefore is considered critical for inclusion in this review. In some sections of the report (particularly on overall budget trends), the CMP is analyzed separately since its disproportionately large size distorts the scale and budgets of social welfare programs, making analysis of other programs difficult. 22 REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA 3. THE PROXY MEANS TEST AND DATA SOURCE FOR THE ANALYSIS The first targeting methodology to identify household livelihood levels based on the PMT was approved by a joint order of the Chairman of the National Statistics Office (NSO) and the Minister for Social Welfare and Labor in 2010. The PMT formula included seven types of observable variables and a constant, identified by analyzing the 2007/08 Household Socio-Economic Survey (HSES). Targeting data was collected between 2010 and 2012, and the first PMT targeting database was applied to identify beneficiary households for the Food Stamp Program. The PMT formula was most recently revised in 2013. The second PMT formula differed from the first formula in three ways: (i) addition of new variables such as type of employment, land ownership, and disability; (ii) change in estimation of consumption from per capita estimation to use of per adult- equivalent scale; and (iii) reduction from four to three strata16 by combining all rural areas (Castro, 2013). The PMT targeting database is expected to be renewed every three years. Assessment of the 2013 data indicates that the quality is acceptable with very high coverage, good internal consistency, and external statistical validation17 with the 2010 census and 2012 HSES. The household survey data was collected between November and December of 2013 and covered 666,673 households. Compared to the 2010 census (adjusted by the estimated annual population growth to the population in 2013), the PMT data covered 86 percent of the population nationwide, with the lowest coverage of 80 percent in Ulaanbaatar and highest coverage of 95 percent in the Western Region (Asian Development Bank, 2014). 18 The 2013 PMT data served as the basis for the analysis of this note. The PMT data was merged with the social welfare program administrative data, using the unique national identification number assigned to each citizen of the country. The social welfare program administrative database contains 2,636,370 individual records. As discussed earlier, of these individual records, 257,302 individual 23 records (9.8 percent) were not linked/merged with the PMT records. The remainder of this section provides a description of the characteristics of the individuals included in the combined PMT and social welfare administrative data (SW Admin/PMT data). In the combined data, there were 662,319 households. The average age was 28.4 years old, with 49 percent males and 51 percent females. In terms of geographic distribution, around 40 percent of the individuals were from Ulaanbaatar (Table 1). The two largest aimags in terms of the population in the database were Khuvsgul and Uvurkhangai. The smallest aimags were Govisumber followed by Dundgovi. Table 1: Number of Individuals in the SW Admin/PMT Data by Aimag Geographical Aimag/city Number of individuals Percentage code 11 Ulaanbaatar city/Улаанбаатар 954,107 40.1 21 Dornod/Дорнод 63,334 2.7 22 Sukhbaatar/Сүхбаатар 51,438 2.2 23 Khentii/Хэнтий 60,581 2.6 41 Tuv/Төв 75,874 3.2 42 Govisumber/Говьсүмбэр 13,321 0.6 43 Selenge/Сэлэнгэ 86,403 3.6 44 Dornogovi/Дорноговь 53,639 2.3 16. From four strata (Ulaanbaatar, Aimag Centers, Soum Centers, and countryside) to three strata (Ulaanbaatar, Aimag Centers, and countryside). 17. There are some concerns regarding external statistical validity for a number of indicators, but these indicators are not included in the analysis in this note. 18. Compared to the HSES 2012 and 2010 census, the distributions of key indicators are consistent except for a few, including the num- bers of people employed in the formal sector and the ownership of some assets and land rights in rural areas. REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA 45 Darkhan-Uul/Дархан-Уул 80,551 3.4 46 Umnugovi/Өмнөговь 48,779 2.1 48 Dundgovi/Дундговь 37,322 1.6 61 Orkhon/Орхон 80,315 3.4 62 Uvurkhangai/Өвөрхангай 96,795 4.1 63 Bulgan/Булган 52,942 2.2 64 Bayankhongor/Баянхонгор 72,926 3.1 65 Arkhangai/Архангай 80,095 3.4 67 Khuvsgul/Хөвсгөл 116,027 4.9 81 Zavkhan/Завхан 63,248 2.7 82 Govi-Altai/Говь-Алтай 51,200 2.2 83 Bayan-Ulgii/Баян-Өлгий 89,676 3.8 84 Khovd/Ховд 75,707 3.2 85 Uvs/Увс 74,788 3.1 Total 2,379,068 100.0 Source: SW Admin/PMT data, World Bank staff estimates. School attainment of adults Among adults 18 years old and above, two-thirds had either completed secondary education or attained less than secondary school education. Around 4.8 percent had no education, while 8 percent had primary school education and 15.8 percent had some secondary education. About one-quarter of the adults had higher than technical/vocational school education. 24 Table 2: Highest Level of Education/Qualification Attained (18 years or older) Level of schooling Number of individuals Percentage 1 None 76,282 4.82 2 Primary 126,531 7.99 3 Secondary 250,542 15.82 4 Completed secondary 592,312 37.4 5 Technical/vocational 141,256 8.92 6 Higher 396,891 25.06 Total 1,583,830 100 Source: SW Admin/PMT data, World Bank staff estimates. Disability Around 4.3 percent of the individuals reported having some disability. Of those with disabilities, 90.5 percent had certificates of disability. The most common disability (excluding the “others” category) was physical impairment (reported by 23 percent of those with disabilities) followed by mental impairment (reported by 20 percent of those with disabilities) (Table 3). Table 3: Individuals with Disabilities by Type of Disability Type of disability Individuals with disability Percentage 1 Blind (Харааны) 10,853 10.6 2 Mute (Хэл ярианы) 3,503 3.4 3 Hearing-impaired (Сонсголын) 7,400 7.3 4 Physically impaired (Хөдөлгөөний) 22,928 22.5 REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA 5 Mentally impaired (Сэтгэц, оюуны) 20,624 20.2 6 Combined (Хавсарсан) 12,485 12.2 7 Others (Бусад) 24,201 23.7 Total 101,994 100.0 Source: SW Admin/PMT data, World Bank staff estimates. Work Among the working-age group,19 58.3 percent of the individuals reported to have worked the previous week. Among those who did not work in the previous week, only 14.3 percent worked during the previous year. PMT score quintiles Although the household PMT scores were divided into quintiles, the individuals from those households did not neatly divide into quintiles because households in the lower quintiles had larger household sizes. Households in the lowest quintile had 5.2 household members on average, while households in the highest quintile had only 3.6 members (Table 4 and Table 5). Table 4: Individuals in the Database by Household PMT Score Quintile PMT score quintiles Beneficiaries of at least Percentage 1- lowest (poor) to 5- highest (better off) one program 1 599,390 25.22 2 479,628 20.18 3 450,992 18.98 4 449,735 18.93 25 5 396,556 16.69 Total 2,376,301 100.00 Source: SW Admin/PMT data, World Bank staff estimates. Poorer households had many more children between 0-18 years old compared to better-off households (Table 5). On average, households in the poorest quintile had one more child than other quintiles. Differences in the number of household members from quintile to quintile were driven by differences in the number of children in those households. Table 5: Average Household Size and Number of Children by Household PMT Score Quintile PMT score quintiles Average number of Average number of children 0-18 1 - lowest (poor) to 5 - highest (better off) household members years old 1 5.15 3.90 2 4.33 2.88 3 4.12 2.71 4 4.01 2.68 5 3.56 2.27 Total 4.31 2.86 Source: SW Admin/PMT data, World Bank staff estimates. 19. Defined as being between 18 years old and retirement age (55 years old for women, 60 years old for men). REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA While the average age and gender of household heads were similar across wealth quintiles, the poorer households had much higher proportions of household heads who were disabled and who were less likely to have worked during the week of the interview (Table 6). This may be a reflection of the PMT formula, as households with disabled members were more likely to have lower PMT scores, thus being identified as poor. Table 6: Characteristics of the Household Head by Household PMT Score Quintile PMT score quintiles Average % male % disabled % worked this week age among working-aged 1 - lowest (poor) to 5 - highest (better off) 1 44.6 77% 15% 49% 2 45.8 77% 9% 57% 3 45.6 79% 6% 62% 4 44.3 81% 4% 68% 5 43.3 78% 2% 75% Total 44.7 78% 7% 62% Source: SW Admin/PMT data, World Bank staff estimates. 26 REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA 4. KEY FINDINGS Budgets and Expenditures on Social Welfare In 2013, the government spent 2.78 percent of GDP on cash transfers for social welfare including the CMP (1.37 percent excluding the CMP). This is relatively generous, considering that the average spending by developing and emerging countries is 1.6 percent of GDP (World Bank, 2014). It suggests that the government is committed to allocating adequate fiscal resources to ensure the social welfare of the population. In terms of trends in the total expenditure on benefits, the largest allocation by far for cash grants in 2013 was for the CMP.20 The CMP comprises approximately half of the total allocation for cash grants but also covers the largest number of beneficiaries (Table 7 and Figure 5). The second largest expenditure was Social Pensions (MNT 83 billion, approximately USD 50 million21) followed by the Allowance for Mothers and Children (MNT 67 billion, approximately USD 40 million22). These two programs combined were still less than two-thirds of the budgeted amount for the CMP in 2013 (Table 7). Table 7: Beneficiaries and Expenditures by Program in 2013 (thousands MNT) 2013 % of 2013 state Program Number of MNT total % of SW budget % of 2013 beneficiaries (nominal) expenditures (MNT 7,444.6 GDP billion) Social Welfare Pension 63,423 83,191,227 17.60% 1.12% 0.49% Pregnant and Lactating 94,882 38,895,382 8.20% 0.52% Mothers 0.23% 27 Social Welfare Allowance 42,260 23,442,090 5.00% 0.31% 0.14% Community-Based Services 24,104 5,566,419 1.20% 0.07% 0.03% Social Welfare Support for 36,411 15,906,364 3.40% 0.21% Caregivers 0.09% Allowance for the Disabled 38,743 6,526,413 1.40% 0.09% 0.04% Allowance for the Elderly 125,941 14,429,702 3.10% 0.19% 0.08% Food Stamp Program 97,919 10,975,451 2.30% 0.15% 0.06% Benefits for Mother Heroes 202,474 28,098,558 6.00% 0.38% 0.17% Benefits for the Elderly with 4,962,636 1.10% 0.07% State Merits 0.03% Specialized Care Services 100,000 0.00% 0.00% 0.00% Child Money Program 1,000,000 240,000,000 50.80% 3.22% (Approved Budget) 1.41% TOTAL 472,094,242 100.10% 6.34% 2.78% Source: GOSWS. 20. The expenditure figures are the total spent on transfers and reimbursements and do not include the operational cost of these pro- grams or the overhead costs. 21. Using the 1,661 MNT per USD real exchange rate as of December 2013. http://www.mongolbank.mn/eng/liststatistic.aspx?did=3 22. This is derived by adding the budget for the Pregnant and Lactating Mothers program and the benefits for the Mother Heroes program, using the 1,661 MNT per USD real exchange rate as of December 2013. http://www.mongolbank.mn/eng/liststatistic.aspx?did=3. REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA Figure 5: Expenditures by Social Welfare Program Category in 2013 Source: GOSWS. Excluding the CMP, the expenditures for all programs increased slightly between 2010 and 2013, taking inflation into account (Figure 6). The largest increase was the two-fold increase for Social Welfare Pensions. Expenditures on Benefits for Pregnant and Lactating Mothers increased about 40 28 percent between 2011 and 2013, while the budget for Mother Heroes benefits decreased about 45 percent. Figure 6: Inflation-Adjusted Expenditures by Social Welfare Program Category 2010- 2013, Excluding the Child Money Program (taking 2010 as the base year) Thousand MNT 70,000 60,000 50,000 40,000 30,000 20,000 10,000 - 2010 2011 2012 2013 Social Welfare Pension Social Welfare Allowance Community-Based Services Social Welfare Support for Caregivers Allowance for the Disabled Allowance for the Elderly Food Stamp Program Specialized Care Services Source: GOSWS. REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA Total Number of Beneficiaries In 2013, the CMP, the largest program in terms of coverage, had an estimated one million beneficiaries. This is more than twice the size of the second largest program, the Allowance for Mothers and Children, with 405,000 beneficiaries. The smallest program was the Benefits for the Elderly with State Merits, with only 4,000 beneficiaries in 2012 (Figure 7). Figure 7: Number of Beneficiaries by Social Welfare Program Category in 2013 1,000,000 800,000 600,000 400,000 200,000 - 29 Source: GOSWS. Analysis of the SW Admin/PMT data confirmed the overwhelming coverage of the CMP. Around 94 percent of children 0-18 years old received the program. Interestingly, even when broken down by quintile, 94 percent of children 0-18 years old in every quintile received the CMP. In contrast to the trends in expenditures, the numbers of beneficiaries of social welfare programs other than the CMP have largely remained constant since 2010. The only increase in number of beneficiaries was for the Allowance for Mothers and Children between 2010 and 2011. This increase was due to a rise in the number of beneficiaries of the Benefits for Mother Heroes from 2010 to 2011, following the softening of the eligibility criteria.23 23. The number of children to which mothers had to give birth to become a Level 1 Mother Hero was reduced from 7 to 6. For a Level 2 Mother Hero, the number was reduced from 5 to 4 children. REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA Figure 8: Number of Beneficiaries by Social Welfare Program Category Excluding the Child Money Program 2010- 2013 250,000 200,000 150,000 100,000 50,000 0 2010 2011 2012 2013 Social Welfare Pension Pregnant and Lacta ng Mothers Social Welfare Allowance Community-Based Services Social Welfare Support for Caregivers Allowance for the Disabled Allowance for the Elderly Food Stamp Program Bene ts for Mother Heroes Bene ts for the Elderly with State Merits Source: GOSWS. Coverage of Social Welfare Programs When the CMP is included, half (49.4%) of all individuals in the database received at least one program benefit. This coverage is very high, considering that the poverty rate in 2012 in Mongolia was 27 percent. Only a handful of developing and emerging countries have coverage of social welfare 30 benefits higher than the poverty rate (World Bank, 2014). The coverage drops considerably, however, when the CMP is excluded, with only 19 percent of the population receiving at least one program. Analysis of the SW Admin/PMT data indicates that beyond the CMP, the vast majority of individuals (81 percent) did not receive Social Welfare Program benefits (Figure 9). Around 14 percent received one program, 3.4 percent received two programs, and just above 1 percent of the individuals received three programs. Only 0.5 percent received four or more programs. Figure 9: Percentage of Individuals in the SW Admin/PMT Database Receiving a Social Welfare Program, and the Number of Program Benefits Received (excluding the Child Money Program) Source: SW/PMT data, World Bank staff estimates. REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA Among those who received at least one cash transfer program, the average amount of estimated cash transfers received annually was MNT 313,167 (approximately USD 179), including the CMP.24 An estimated average cumulative amount of MNT 219,197 (118 USD) was received by around 3.6 percent of individuals (approximately 86,000 individuals) who received reimbursements and vouchers since 2009. As a reference, the cash component of the transfers represented between 15-18 percent of the Minimal Living Standard in 2014, depending on where the beneficiary resided.25 Equity in Benefit Distribution According to estimates based on the SW Admin/PMT database, the largest share of total benefits was received by individuals in the lowest social welfare quintiles, estimated by the PMT score.26 The bottom 40 percent received 56 percent of total social welfare transfers, while the top 40 percent received 28 percent of total social welfare transfers. Individuals from households in the poorest quintile received 34 percent of total social welfare transfers, which is comparable to other countries in the region (Table 8).27 However, in European and Central Asian countries, the share of total benefits going to the poorest 20 percent reaches over 40 percent (e.g., Croatia and Kosovo). In the best-performing countries around the world, the poorest 20 percent receive over 50 percent of total benefits (e.g., Argentina, Panama, Peru). Table 8: International Comparison: Percent of Total Benefits Received by the Poorest 20 Percent Country Share of Total Benefits Received by the Poorest 20 Percent Mongolia * 34 percent Croatia 40.6 Kosovo 43.4 Argentina 54.0 Panama 52.5 Peru 56.4 Timor-Leste (2011) ** 26.5 percent 31 Vietnam (2008) ** 25.7 percent Thailand (2009) ** 20.5 percent * Source: SW/PMT data, World Bank staff estimates. ** “The State of Social Safety Nets 2014” (World Bank, 2014). The largest program contributing to the share received by individuals in the lowest social welfare quintile was the CMP (16 percent). As shown in Figure 10, the next highest contributors were Social Welfare Pensions (9 percent), Social Welfare Transfers including the Food Stamp Program (4 percent), and Allowances for Mothers and Children (3 percent). Individuals in the second-lowest social welfare quintile received 22 percent of total social welfare transfers. As in the case of the lowest quintile, the programs that contributed to the share of benefits received by those in the second-lowest quintile were: the CMP, Social Welfare Pensions, Social Welfare Transfers, and Allowances for Mothers and Children. 24. This assumes that for those who received benefits that are provided less frequently than once a year (for example, once every three years or once every five years), the benefit was provided in the year of the analysis. Therefore, the average amount of MNT 313,167 is the average of the maximum amount that an individual could receive from social welfare programs. 25. Minimum Living Standards in 2014 were: MNT 146,600 in Western Region; MNT 149,600 in Khangai region; MNT 149,700 in Central region; MNT 144,500 in Eastern Region, and MNT 169,000 in Ulaanbaatar. http://www.nso.mn/page/60 (accessed October 14, 2014) 26. The analysis of share of benefits received assumed that those who were listed as being a program beneficiary in the administrative data were indeed receiving the full benefit amounts. For cash benefits, the amounts were annualized according to the program design (benefit amounts and frequency of payment). For reimbursements, it assumed that all beneficiaries received their full amount in the year of the analysis (for benefits provided less frequently than once a year), and the average reimbursement amounts were estimated using the 2012 expenditure figures obtained from GOSWS. 27. “The State of the Social Safety Nets 2014” states that the benefit incidence of total benefits in Mongolia going to the poorest 20 per- cent is 22.5 percent, based on analysis of household survey data (HSES 2007). The benefit incidence of 34 percent given in this report is estimated based on analysis of PMT and SW Administrative data. REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA Figure 10: Percentage of Benefits Received by Individuals According to Household PMT Score Quintiles and Source of Program Benefits Share of 35% 30% Mothers 25% Disability 20% Merit Elderly 15% SW Transfers 10% SW Pensions 5% Child Money 0% Q1 Q2 Q3 Q4 Q5 Source: SW/PMT data, World Bank staff estimates. Although most programs were categorically targeted, some programs were more pro-poor than others. The pro-poor programs, with larger shares of program benefits received by lower PMT score quintile groups, were Social Welfare Pensions, Social Welfare Transfers, CMP, Allowance for the Disabled, and Allowances for Mothers and Children (Figure 11). Figure 11: Pro-Poor Programs - Percentage of Benefits Received by Individuals According to 32 Household PMT Score Quintile, by Program28 Source: SW/PMT data, World Bank staff estimates. In contrast, programs that were less pro-poor (regressive) were: Benefits for the Elderly with State Merits, and the Allowance for the Elderly program. In particular, although the program is not an anti-poverty program, the proportion of Benefits for the Elderly with State Merits that went to the non- poor is striking. The vast majority of beneficiaries were in the highest social welfare quintile according to their PMT scores. 28. In Figure 11, the first bar (% individuals in quintile) for each quintile depicts the share of individuals in that quintile. Therefore, program benefit shares higher than the first bar indicate that the quintile group received disproportionately more program benefits. If program shares are lower than the first bar, it means that the quintile group received disproportionately less program benefits. REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA The fact that Benefits for the Elderly with State Merits were enjoyed by the non-poor is important to note given that the benefit amounts of this program were the most generous compared to other programs. Estimating from the program expenditure and number of beneficiaries in 2013, the average beneficiary of this program received approximately MNT 1,320,000 (approximately USD 713) per year. This is 12 times the generosity level of the Food Stamp Program which exclusively benefits those in the lowest quintile, with an approximate annual benefit of MNT 112,000 (approximately USD 60). Figure 12: Regressive Programs: Percentage of Benefits Received by Individuals According to Household PMT Score Quintile, by Program29 33 Source: SW/PMT data, World Bank staff estimates. The gender distribution of beneficiaries differed across programs (Figure 13). The CMP and Allowance for the Disabled both had equal numbers of male and female beneficiaries. The beneficiaries of programs associated with motherhood—including many under the Social Welfare Transfers— were naturally women: beneficiaries of the Allowance for Mothers and Children were all female, and around 58 percent of Social Welfare Transfers beneficiaries were women. More women received the Allowance for the Elderly (69 percent) and Social Pensions (54 percent), probably due to the lower age of eligibility (55 years old for women compared to 60 years old for men) and the longer life expectancy of women compared to men. The only exception was the Benefits for the Elderly with State Merits, which benefited men (82 percent) far more than women. 29. In Figure 12, the first bar (% individuals in quintile) for each quintile depicts the share of individuals in that quintile. Therefore, program benefit shares higher than the first bar indicate that the quintile group received disproportionately more program benefits. If program shares are lower than the first bar, it means that the quintile group received disproportionately less program benefits. REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA Figure 13: Gender Distribution of Program Beneficiaries by Program Child Money Program 50.9% 49.1% Mothers 100.0% Allowance for the Disabled 50.6% 49.4% Allowance for those with Merit 81.9% 18.1% Allowance for the Elderly 31.1% 68.9% SW Tansfers 41.8% 58.3% SW Pensions 46.4% 53.6% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Male Female Program Benefits Received by the Elderly According to the SW Admin/PMT data, around 70 percent of pension-aged individuals benefited from at least one social welfare program, not including social insurance pensions which are not part of this analysis. On average, a pension-aged individual receiving at least one social welfare program received an estimated MNT 331,000 in cash allowances a year. 34 Figure 14 suggests that the elderly benefit from multiple programs and are combining benefits from different types of programs. Around 76 percent of total social welfare benefits received by the elderly went to those benefiting from multiple programs. Figure 14: Proportion of Estimated Total Benefits Received by the Elderly, by Number of Programs Received and Share of Program Share of 35% 30% Mothers 25% Disability Merit 20% Elderly 15% SW Transfers SW Pensions 10% Child Money 5% 0% 1 program 2 programs 3 programs 4 programs 5 programs Source: SW/PMT data, World Bank staff estimates. REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA Around 23 percent of social welfare transfers received by the elderly went to beneficiaries who received only one program. The Allowance for Mothers and Children (in this case Mother Heroes Program) and the Allowance for the Elderly were the two programs that provided benefits to those elderly who received only one program. This was similar for those who received two programs. The fact that the elderly were receiving multiple programs to meet their needs may be a strong indication of the fragmented nature of programs addressing the needs of the elderly. Figure 14 above shows that Assistance for the Elderly, Social Welfare Transfers, and Social Welfare Pensions were the three larger programs that were needs-based, which the elderly were combining to meet their needs. This suggests that a more comprehensive program for the elderly would be a program with elements from the Assistance for the Elderly and Social Welfare Transfers. Program Benefits Received by the Disabled According to the SW Admin/PMT data, 66 percent of the disabled received at least one social welfare program. Coverage of the disabled above 16 years old and those below 16 years old were quite distinct, so they are reviewed separately below. For adults older than 16 years old, 63 percent received at least one program, predominantly Social Welfare Pensions which comprised an estimated 76 percent of the benefits going to the disabled. Social Welfare Pensions provided among the most generous benefits compared to other programs benefiting the disabled. Around 43 percent of benefits were received by those receiving only one program (mainly Social Welfare Pensions), while those receiving multiple programs received a larger share of their benefits from Social Welfare Transfers and the Allowance for the Disabled (Figure 15). The fact that the disabled were combining multiple programs but with benefits coming predominantly from Social Welfare Pensions suggests that a comprehensive program for the disabled largely based on Social Welfare Pensions could replace programs that currently provide smaller shares of benefits. Figure 15: Proportion of Estimated Total Benefits Received by the Disabled 16 Years Old and Above, by Number of Programs Received and Share of Program 35 Share of 45% 40% 35% Mothers Disability 30% Merit 25% Elderly 20% SW Transfers 15% SW Pensions 10% Child Money 5% 0% 1 program 2 programs 3 programs 4 programs 5 programs Source: SW/PMT data, World Bank staff estimates. On average, a disabled person above 16 years old receiving at least one social welfare program received an estimated MNT 924,000 in cash allowances a year. They were the largest beneficiaries in terms of average benefit amount per individual due mainly to the size of the Social Welfare Pensions, after the beneficiaries with state merits and honors. REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA Among the disabled below 16 years of age, 87 percent received at least one program and received. Around 81 percent of the share of transfers to the disabled younger than age 16 was provided by the CMP. About one-fifth of the disabled younger than 16 years old received benefits from three or more programs, namely the CMP, Social Welfare Transfers, and Allowance for the Disabled (Figure 16). Those younger than 16 years old are not eligible for Social Welfare Pensions. Figure 16: Proportion of Estimated Total Benefits Received by the Disabled under 16 Years of Age, by Number of Programs Received and Share of Program Share of total 35% 30% Mothers Disability 25% Merit 20% Elders 15% SW Transfers SW Pensions 10% Child Money 5% 0% 1 Program 2 Programs 3 Programs 4 Programs 5 Programs Source: SW/PMT data, World Bank staff estimates. 36 Program Benefits Received by the Poor Estimates based on the SW Admin/PMT data indicate that 62 percent of individuals from households in the poorest quintile according to PMT score were direct beneficiaries of at least one social welfare program, but the figure drops to 29 percent when the CMP is excluded. On average, an individual in the poorest household quintile benefiting from at least one social welfare program received an estimated MNT 362,542 in cash allowance per year. Around 64 percent of total benefits going to the poorest quintile went to beneficiaries who received only one program, including the CMP (Figure 17). The majority of the poor only received one program. The largest share of benefits for the poor came predominantly from the CMP followed by Social Welfare Pensions, although the CMP was the single and largest benefit received by many individuals in the poorest quintile. REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA Figure 17: Proportion of Estimated Total Benefits Received by the Poor, by Number of Programs Received and Share of Program received by the poor 70% 60% Mothers 50% Disability Merit 40% Elderly 30% SW Transfers SW Pensions 20% Child Money Program 10% 0% 1 Program 2 Programs 3 Programs 4 Programs 5 Programs Source: SW/PMT data, World Bank staff estimates. The Food Stamp Program is the only program that specifically targets the poor. Around 2.3 percent of households in the SW Admin/PMT database (15,118 households out of 662,559 households) received the Food Stamp Program. All households belonged to the poorest quintile, with the exception of one household which belonged to the third quintile. Figure 18 shows the proportion of households in the poorest quintile who received the Food Stamp Program by PMT score. The PMT score for the poorest 5 percent was between 0 and 244, and as shown in Figure 18, all (but one, not shown in the graph) of the beneficiaries were among the poorest 5 percent. In fact, all the beneficiaries were among the poorest 2.7 percent according to PMT score. Among the bottom 2.7 percent, 85.6 percent 37 of households were beneficiaries of the Food Stamp Program.30 Figure 18: Proportion of Food Stamp Program Beneficiaries among the First PMT Score Quintile by PMT Score Share of households PMT score for the 1 0.9 0.8 0.7 0.6 0.5 0.4 PMT score for the poorest 0.3 0.2 0.1 0 105 118 131 144 157 170 183 196 209 222 235 248 261 274 287 300 313 326 339 352 1 14 27 40 53 66 79 92 Source: SW/PMT data, World Bank staff estimates. 30. There were 2,543 in the SW/PMT database who were below the 2.7 percentile but were not registered as Food Stamp Program beneficiary households. REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA 5. POLICY IMPLICATIONS AND RECOMMENDATIONS To better protect the poor and ensure the fiscal sustainability of social welfare programs, options for making social welfare programs more efficient and effective are sought urgently. Given the relatively generous distribution of social welfare programs, it is possible for the government to contain costs while making programs more effective in terms of protecting the poor from negative shocks and reducing poverty. This section discusses the recommended strategies, based on the review of social welfare programs. Consolidating Fragmented Programs Three strategies are recommended in consolidating the existing social welfare programs and are discussed in greater detail below: • Consolidate small programs that share similar objectives and are targeted to the same beneficiary types by prioritizing program design based on the need to provide comprehensive support to the targeted population through a more comprehensive program; • Strengthen programs that are needs-based and gradually phase out entitlement-based programs; and • In the future, consider transferring health-related benefits to be covered by Health Insurance. Consolidate small programs. As noted earlier, the MPDSP/GOSWS currently implements 71 programs, although the number of programs is partly derived by the way these programs are counted. While these programs can be grouped in a thematic way, the individual programs are very specific and narrowly defined. These fragmented programs lead to administrative burdens and complicated processes for applications and requesting claims. They also result in low benefit amounts and low beneficiary coverage. 38 Some programs have higher benefit amounts but very low coverage. The “benefits for citizens nursing children in their family who require protection because of being target of violence” provides MNT 52,800 (approximately USD 29) on a monthly basis. The “allowance to be provided to citizens taking care of single, disabled, elderly person who does not have any child or relative” provides the same benefit amounts. In 2012, these programs covered 365 and 340 beneficiaries, respectively. With these small programs, it is difficult to avoid benefit overlaps and inefficiencies. Consolidation of these smaller programs should focus on designing programs to be comprehensive enough to provide the necessary coverage and benefit support to meet the needs of the targeted population. As discussed earlier, the analysis revealed that beneficiaries—particularly the elderly—are combining multiple programs to meet their needs. Social Welfare Transfers and the Allowance for the Elderly are two needs-based programs that the elderly are accessing to combine benefits. A more detailed analysis and qualitative beneficiary feedback study is required to understand which specific elements of these programs should be combined to provide a more comprehensive program for the elderly. Similarly, in addition to Social Welfare Pensions, the disabled are accessing the Allowance for the Disabled and Social Welfare Transfers. In parallel to the review of benefits for the elderly, a more detailed review including a qualitative beneficiary feedback study is recommended to ensure that a more comprehensive program can be designed to meet the needs of the disabled appropriately. In consolidating these programs and designing comprehensive programs, it would be important to take into account the different needs of male and female beneficiaries within the specific target groups. Strengthen needs-based programs and consider exit strategies for entitlement-based programs. Programs such as Benefits for the Elderly with State Merit and Mother Heroes program are based on entitlements to benefits rather than need for social welfare. Entitlement-based programs are more regressive than programs designed to address specific needs. As discussed earlier, the most regressive program that provides allowances and assistance for the well-off is Benefits for the Elderly REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA with State Merit, which also has the most generous estimated benefit amounts by far. Given that the vast majority of the benefits are received by the better-off, these programs should be phased out, to be replaced by non-cash-based recognition of the honor or by a one-time award instead of lifetime monthly benefits. The Mother Heroes program and Benefits for the Elderly with State Merit together comprise about 7 percent of the social welfare transfer budget, which was about 0.45 percent of the 2013 state budget. If these programs are eventually phased out or replaced by a one-time cash award, the government will have considerable fiscal savings in the long run that could be reallocated to poverty-targeted or needs-based programs. Transfer specific health-related benefits to the Social Health Insurance Program. As described above, the Allowance for the Elderly and the Allowance for the Disabled provide reimbursements for prosthetics of specific body parts as well as equipment such as hearing aids and false teeth. In 2012 alone, MNT 5.3 billion (approximately USD 2.9 million) in nominal terms was spent for these programs together. The categories for each provision under the programs are extremely specific, although it is unclear how effectively these reimbursements are addressing the needs of beneficiaries. The reimbursements for prosthetics and concessions for sanatoriums, nursing homes, and institutional care may be better suited to being covered under social health insurance and hence addressed in a comprehensive manner. Making Social Welfare Programs Poverty-Targeted Social welfare programs can be made poverty-targeted by eliminating or reducing benefits for the richer quintiles, using the readily available PMT database. The most effective way of containing the costs of social welfare programs while making them more effective is to add poverty targeting to existing programs that are currently categorically targeted. The Ministry already owns an updated PMT database, which allows the government to identify the poor. This is a very powerful tool that is not readily available to many governments around the world, which allows the government to make any program poverty-targeted if desired. 39 The CMP is expected to have a considerable poverty impact, but the level of program expenditures may be difficult to sustain. As discussed above, the program has wide coverage—benefiting the poor as well as non-poor—and a large benefit amount (MNT 240,000 or approximately USD 130 per child per year). At the same time, the total budget for transfers of this program alone was MNT 240 billion (approximately USD 130 million) in 2013, accounting for 51 percent of total transfers made through social welfare programs in Mongolia. Considering the fiscal deficit faced by the government, this level of expenditures on the CMP may be difficult to sustain. Taking these facts into account, possible options going forward include: • Continue the CMP as is as the program is effectively reaching the poor, if the fiscal environment allows for the program to continue; • Introduce program design features to the CMP such that the poor receive larger amounts, while maintaining universal coverage of the program. This could be achieved by using the PMT to identify children in poor households or through categorical means of providing larger amounts for children in households with many (three or more, for example) children. Altering program benefits for children in different types of households will make program administration more complex, and if poverty targeting is introduced, strong operational design and implementation plans would need to be prepared to address the potential exclusion of deserving children. Reallocation of benefits could be done in a cost-neutral way or in a way that reduces total program cost: • To maintain cost neutrality, the grant amounts reallocated to the poor would be equal to the reduction in grant amounts for the better-off. • To reduce program cost, the sum of the grant amounts reallocated to poorer children would be lower than the sum of grant amounts reduced for better-off children. REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA Terminate the CMP and replace it with a new poverty-targeted cash transfer program. Termination of the CMP will give the government fiscal space to design a poverty-targeted welfare program, with clear poverty alleviation objectives. Two design options could be: o Implement the Poverty Benefit program stipulated in the Social Welfare Law (2012).31 The new program would need to provide significant levels of coverage and benefits in order to compensate for the loss of benefits experienced by poor households due to termination of the CMP, with a strong operational design to address the potential exclusion of poor households. The advantages are that the program would have a clear poverty alleviation objective, and depending on coverage and benefit amounts, it could lead to considerable budget savings. The disadvantages are that this will be a new program, and as with any new program, considerable operational preparation will be required. o Add a child benefit component to the new Poverty Benefit Program described above to maintain the focus on providing support to the next generation (as in the CMP). Again, it would be very important to ensure that this new Poverty Benefit Program (or at least the child component) would be large enough in terms of coverage and benefit amounts to compensate for termination of the CMP. If the child benefit component focuses on the poorest 40 percent of households, this would still cover half of the children currently covered by the CMP, since poorer household have many more children than the better off. Adding a child benefit component to the Poverty Benefit Program would increase the complexity of program design as the transfer amounts per household would differ according to the number of children. Further simulation analysis using the HSES data is recommended to identify the most effective scenario for poverty impact. Given the limitations of the HSES questionnaire design, the analysis does not allow for a clean evaluation of the program. However, a simulation exercise could use the HSES data to analyze the different scenarios described above and determine the optimal benefit size and the methods and threshold for targeting. 40 Introduction of design changes must be accompanied by strengthening of certain features of program operations to ensure that there is wide understanding and acceptance of the policy changes and that the intended beneficiaries will receive the programs. Strengthening of the following operational features is recommended, particularly if poverty targeting will be introduced: • Strong communication campaigns to widely communicate the objectives of the policy change as well as changes in program design including eligibility, program benefits, and how to report issues and grievances would be critical. • Introducing targeting, whether through PMT or other categorical means, is bound to introduce exclusion error (eligible beneficiaries being excluded), no matter how good the targeting methodology is considered to be. Therefore, an effective grievance redress mechanism would be essential. If targeting will be conducted through application of the PMT, it would be very important to ensure the effective parallel implementation of the “Updating the information stored in the database” section of Cabinet Decree 404 (December 30, 2014). This will help keep the PMT database up-to-date in terms of changes to household welfare status made in response to grievances regarding registered welfare status based on the PMT scores, as well as ensure that the database is complete in case households were missed during the assessment. It is also important to note that the option of altering benefit amounts by type of household would complicate administration of the program, so the MIS system would need to be prepared and designed accordingly. 31. The direct translation of the program title in the Social Welfare Law from Mongolian is: “Benefit for members of HHs in need for social assistance” (Poverty benefit). REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA Increasing the Generosity and Coverage of Poverty Reduction Programs As discussed earlier, the coverage of poor households by social welfare programs and the generosity of benefits received by those in the lowest quintile according to PMT score are low. When translated into per-household transfer amounts, the transfer amounts are clearly small. Around 94 percent of households in the poorest quintiles had a household member who receives some social transfers, with an average of MNT 867,443 received (including the CMP) per year per household. Considering that the 2012 per capita poverty line was MNT 118,668 and that the average household consisted of five members, this average household transfer amount was approximately 12 percent of household consumption at the poverty line for a five-member household. Most countries provide social welfare benefits that represent 20 to 35 percent of the poorest households’ post-transfer consumption, although there is significant variation worldwide. The most generous countries in the Europe and Central Asia region provide benefits that represent between 40-55 percent of the poorest households’ post-transfer consumption (World Bank, 2013). The generosity and coverage of programs to protect the poor could be increased by establishing a new program targeted to the poor or (in the short term) by revamping the Food Stamp Program. There is an urgent need to focus social welfare programs on ensuring that the poor are protected from economic shocks. The poor do not have a program that caters to their needs, and a large portion of their benefits are derived from the CMP, which equally benefits those in all social welfare quintiles. Strengthening Active Labor Market Programs that Serve as a Safety Net and Designing Social Welfare Programs that Promote Labor Force Participation Having effective and robust active labor market programs would provide an important safety net for vulnerable workers. A gap in the current social welfare system is that there is no safety net support for those who are poor (but not as poor as the bottom 5 percent) or near-poor and do not fit in any of the categories for social welfare program eligibility. Unemployment insurance only covers formal workers who have contributed for at least 24 months with 9 months of consecutive coverage. 41 Laborers in temporary or seasonal jobs do not qualify for unemployment benefits, yet they are the ones who may be most affected by economic adjustment or other external economic shocks. For social welfare beneficiaries who are currently unemployed or under-employed, social welfare programs could be used to promote engagement in work/skills accumulation activities. The current design of social welfare programs is categorically targeted, so most beneficiaries are entitled to benefits as long as they remain within the categorical eligibility criteria. While the generosity and coverage of poverty-targeted programs need to be increased, such programs should not create dependency but should encourage beneficiaries to engage in labor. New generation work activation programs could be introduced for beneficiaries who are unemployed, seasonally unemployed, or underemployed, applying the principles of “welfare-to-work” transition approaches proven to be effective in many middle-income countries. These programs could aim to link vulnerable individuals (including the poor, migrants, and those living in the gher district) with seasonal jobs and skills development programs. The linkages could be forged through co-responsibility for able-bodied welfare recipients to engage in work/skills accumulation activities. An assessment of existing employment promotion programs implemented by the Ministry of Labor could help inform the integration of effective activation policies into social welfare programs. Strengthening the Monitoring and Evaluation of Social Welfare Programs Stronger monitoring and evaluation of programs would enable the government to design effective social welfare programs in the future. No impact assessment of programs has been conducted to date, so the government has little understanding of program impacts and how programs can be improved. The government could consider: • Requesting the addition of a specific module focused on social protection to the HSES questionnaire, or at a minimum, requesting revision of the HSES questionnaire “other income” section to distinguish social welfare and social insurance benefits. REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA • Fully utilizing the social welfare management information system (MIS) when it is unified and put online. • Conducting periodic reviews of programs by analyzing administrative data to inform decision- making. Although the HSES module that includes social welfare programs could potentially be a good source of information, the categories of programs included in that section of the current HSES questionnaire do not allow for systematic analysis of program effectiveness and impact. The module that includes the data on social welfare programs is the “other income” section that asks household respondents about individual income from sources other than jobs. Unfortunately, the program categorization mixes contributory (insurance) programs and non-contributory (welfare) programs, and the programs are defined too broadly. Therefore, it is currently not possible to analyze the information in a meaningful manner. With some input from the MPDSP, the questionnaire may be revised for future surveys, which would serve as an important data source for monitoring and impact assessment of social welfare programs. It is important to ensure the accuracy of data entered at the local level and to build capacity to analyze the data for timely monitoring and reporting. The social welfare administrative data used for this analysis are all maintained at aimag and district-level welfare offices, with very little oversight ensuring the quality of data entered into the system. With the establishment of the online MIS (currently under development), the GOSWS will be able to access the data on a regular basis to monitor and report on program implementation. In addition, a general clean-up of program codes would help ensure better data management. With the development of social welfare programs over time, the coding used as program identifiers also proliferated. However, programs that have been discontinued continue to be on the code list. 42 REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA 6. CONCLUSION The above recommendations provide broad strategies for reforming social welfare programs but do not address the specific design or operational details on how to make individual programs more efficient and effective. Each of the recommendations, if adopted, would require careful consideration of program design and implementation details. The focus however, must remain as stated in the Social Welfare Law: to ensure and protect the minimum welfare of vulnerable groups. Finally, an area that has not been covered by the analysis for this report but merits attention is tax incidence and its potential contribution to wealth redistribution. Analysis is needed to understand the effect of the flat income tax rate on all income groups in Mongolia whether it dilutes the impacts of the generally progressive nature of social welfare programs. REFERENCES Asian Development Bank (2014). March 2014 Mission Report: Food and Nutrition Social Welfare Program and Project. Castro, M. C. (2013). Improving the Proxy Means Testing Models to Identify Households for Social Assistance in Mongolia (Final Report). GOSWS. Expenditures/Budgets Excel Sheets 2010 – 2013. Law of Mongolia (2012). Law on Social Welfare. Law of Mongolia (2005). Law on Social Security of Persons with Disability. Law of Mongolia (2005). Law on Social Security of Senior Citizen. Law on Additional Benefits for Senior Citizens, who are the state hero, labor hero, and with the 43 people’s honor and the state honor (2008) Ministry of Population Development and Social Protection (2013). Handbook on Social Welfare Legislation. World Bank (2013). A Diverse and Dynamic Region: Taking Stock of Social Assistance Performance in East Asia and the Pacific. World Bank (2014). The State of Social Safety Nets 2014. World Health Organization (2011). World Report on Disability. REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA ANNEX I: REVIEW OF INDIVIDUAL PROGRAMS Social Welfare Pensions According to Article 12 in the Law on Social Welfare (2012), Social Welfare Pensions are granted on a monthly basis for citizens who are not entitled to a pension under social insurance legislation. These include the following categories of people: • Men aged 60 years old and above, women aged 55 years old and above; • A vertically challenged (“dwarf individual”) person 16 years and above; • A disabled person 16 years old and above who has lost 50 percent or more of his/her capacity to work; • A child under 18 years old whose foster parent has died; and • A single mother aged 45 years old and above or a single father aged 50 years old and above with four or more children under 18 years old. The current monthly transfer amount of Social Welfare Pensions is MNT 115,000 (approximately USD 62), increased from MNT 103,600 (approximately USD 56) per month in January 2014. This monthly amount is between 68 to 79.5 percent of the Minimum Living Standard in 2014, meaning that with the income from Social Pensions alone, a one-person household would be considered to be living below the Minimum Living Standard. Expenditures on Social Welfare Pensions The budget for Social Welfare Pensions has increased steadily in the past three years, with total expenditure almost doubling between 2010 and 2012 in real terms. This trend is driven by the substantial increase in the allocation for the disabled (defined as those with labor capacity lost by 50-60 percent 44 and those with over 70 percent lost) as well for children who have lost their breadwinner (Figure 19). Figure 19: Trends in Inflation-Adjusted Annual Expenditures on Social Welfare Pensions (in 1,000,000 MNTs, using 2010 as the base ear) 25,000 Social pensions for Ederly or Disabled Disabled with lost labor 20,000 capacity by 50 to 60 percent Disabled with lost labor capacity by more than 70 percent 15,000 Ver cally challenged (dwarfs) over 16 Welfare bene ts for single 10,000 mother or father Children under 18 with no breadwinner 5,000 Elderly who do not qualify for Social Insurance Bene ts for reindeer herders - 2010 2011 2012 2013 (budget) Source: GOSWS. REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA Characteristics of Beneficiaries of Social Welfare Pensions In the SW Admin/PMT data, a total of 46,958 individuals received Social Welfare Pensions. This is about 2 percent of the individuals in the SW Admin/PMT data and does not include contributory pensions. Generally, beneficiaries only received one type of Social Welfare Pension but also received other social welfare program benefits. However, a small minority of individuals (199 individuals) received multiple types of Social Welfare Pension benefits. The vast majority (over 88 percent) of individuals who benefited from multiple pensions programs were those who received pensions for the disabled as well as pensions for caring for children who lost a breadwinner. Table 9: Beneficiaries Receiving Social Welfare Pensions by Aimag Geographical Number of individuals Aimag Percentage code receiving SW pensions 11 Ulaanbaatar/Улаанбаатар 14,797 31.5% 21 Dornod/Дорнод 1,732 3.7% 22 Sukhbaatar/Сүхбаатар 1,270 2.7% 23 Khentii/Хэнтий 1,536 3.3% 41 Tuv/Төв 1,569 3.3% 42 Govisumber/Говьсүмбэр 243 0.5% 43 Selenge/Сэлэнгэ 2,034 4.3% 44 Dornogovi/Дорноговь 1,041 2.2% 45 Darkhan-Uul/Дархан-Уул 1,370 2.9% 46 Umnugovi/Өмнөговь 1,013 2.2% 48 Dundgovi/Дундговь 792 1.7% 61 Orkhon/Орхон 1,399 3.0% 62 Uvurkhangai/Өвөрхангай 1,943 4.1% 45 63 Bulgan/Булган 1,016 2.2% 64 Bayankhongor/Баянхонгор 1,802 3.8% 65 Arkhangai/Архангай 1,858 4.0% 67 Khuvsgul/Хөвсгөл 2,809 6.0% 81 Zavkhan/Завхан 1,818 3.9% 82 Govi-Altai/Говь-Алтай 1,207 2.6% 83 Bayan-Ulgii/Баян-Өлгий 2,658 5.7% 84 Khovd/Ховд 1,466 3.1% 85 Uvs/Увс 1,585 3.4% Total 46,958 100.0% Source: SW/PMT data, World Bank staff estimates. School attainment Social Welfare Pensions beneficiaries 18 years old and above had lower school attainment than the general population, with over 86 percent of them having completed secondary education or less. The percentage of those with no education was over five times higher than that of the general population (4.82 percent) in the SW Admin/PMT database (Table 10). Table 10: Highest Level of Education Attained by Social Welfare Pensions Beneficiaries (18 years old and above) Level of schooling Number of beneficiaries Percentage 1 None 12,050 28.03 2 Primary 4,595 10.69 REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA 3 Secondary 7,762 18.05 4 Completed secondary 12,465 28.99 5 Technical/Vocational 2,082 4.84 6 Higher 4,040 9.4 Total 42,994 100 Source: SW/PMT data, World Bank staff estimates. Disability Around 62.3 percent of Social Welfare Pensions beneficiaries reported having disabilities. Around 33.2 percent of them suffered from mental impairments, followed by 20.7 percent suffering from physical impairments. Around 11 percent suffered from multiple disabilities (Table 11). Surprisingly, around 18 percent (6,065 individuals) of the 34,402 recipients of Social Welfare Pensions for the disabled reported not having disabilities, as indicated by the PMT data. Table 11: Social Welfare Pensions Beneficiaries with Disabilities, by Type of Disability Type of disability Number of individuals Percentage 1 Blind (Харааны) 2,437 8.34 2 Mute (Хэл ярианы) 1,440 4.93 3 Hearing-impaired (Сонсголын) 1,884 6.45 4 Physically impaired (Хөдөлгөөний) 6,043 20.68 5 Mentally impaired (Сэтгэц, оюуны) 9,694 33.17 6 Combined (Хавсарсан) 3,251 11.13 7 Others (Бусад) 4,473 15.31 46 Total 29,222 100 Source: SW/PMT data, World Bank staff estimates. Work Among working-age Social Welfare Pensions beneficiaries, 33.8 percent had worked in the past one week, while only 5.4 percent had worked in the last year. These percentages are considerably lower than those of the general population. PMT score quintile Possibly reflecting the high proportion of disabled among the beneficiaries,32 Social Welfare Pensions beneficiaries were overwhelmingly assessed as being among the poor according to their PMT scores. Around 48 percent of Social Welfare Pensions recipients were categorized as belonging to the poorest quintile, and 23.3 percent were in the second-lowest quintile (Table 12). The progressive and pro-poor nature of Social Welfare Pensions is evident when the proportion of the population in the database in the estimated wealth quintiles is compared to the proportion receiving Social Welfare Pensions (Figure 20). Table 12: Social Welfare Pensions Beneficiaries by PMT Score Quintile PMT score quintiles Number of Social Welfare Percentage 1 - lowest (poor) to 5 - highest (better off) Pensions beneficiaries 1 22,288 47.53 32. Since disability is one of the variables in the PMT formula, households with disabled household members are more likely to get a lower PMT score than a household with no disabled member. Therefore, households with disabled members are more likely to be identi- fied as being poor. REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA 2 10,986 23.43 3 6,891 14.69 4 4,492 9.58 5 2,239 4.77 Total 46,896 100 Source: SW/PMT data, World Bank staff estimates. Figure 20: Percentage of Individuals in Each Household PMT Score Quintile and the Proportion of Social Welfare Pensions Beneficiaries from Each Household PMT Score Quintile 47 Source: SW/PMT data, World Bank staff estimates. Social Welfare Allowances According to Article 13 of the Law on Social Welfare (2012), Social Welfare Allowances include the following: • Allowance for caregivers; • Allowance for individual household members requiring social welfare assistance; • Allowance for emergency and livelihood support; and • Allowance for pregnant women and mother with infants. The benefit amounts for Social Welfare Allowances vary in amount and frequency. The allowance for caregivers currently provides MNT 52,800 (approximately USD 29) per month.33 One-time assistance of MNT 1,200,000 (approximately USD 649) is provided to young adults (18-24 years old) who became orphans before turning 18 years old or to households who lost their shelter due to a disaster or accident. One-time assistance of MNT 1,000,000 (approximately USD 541) is provided to those with twins and MNT 3,000,000 (approximately USD 1,622) to those with triplets. An annual allowance of MNT 120,000 (approximately USD 65) is provided to a single parent with three or more children under 14 years old, and a quarterly allowance of MNT 60,000 (approximately USD 32) is provided to individuals requiring permanent nursing care. 33. This was increased from MNT 48,000 in January 2014. REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA Expenditures on Social Welfare Allowances From 2010 to 2012, overall expenditures for Social Welfare Allowances increased by 60 percent in real terms. The budget allocation for 2013 indicates that this increased further, with the total amount more than doubling the 2010 level (Figure 21). Figure 21: Trends in Inflation-Adjusted Annual Expenditures on Social Welfare Allowances (in 1,000,000 MNT, using 2010 as the base year) 18,000 16,000 Cash transfers for caregivers 14,000 12,000 Cash for those with twins or 10,000 triplets 8,000 Assistance for underaged 6,000 orphans or lost shelter due to a disaster/accident 4,000 Concession for kindergarten 2,000 - 2010 2011 2012 2013 (budget) 48 Source: GOSWS. Characteristics of Beneficiaries of Social Welfare Allowances According to the SW/PMT data, a total of 63,263 individuals received at least one Social Welfare Allowances program.34 The average age of recipients was 45.7 years old, and 58.2 percent of beneficiaries were female. The distribution of beneficiaries for Social Welfare Allowances and Social Welfare Benefits and Concessions was similar to the distribution of the population in the SW Admin/ PMT database. As for the Food Stamp Program, Khuvsgul, Arkhangai, and Uvurkhangai had relatively higher coverage, while Ulaanbaatar had lower coverage (Table 13). Table 13: Beneficiaries Receiving Social Welfare Allowances by Aimag (number of beneficiaries and percentage) Geo- Aimag Receiving Receiving Social Food Stamp Program graphical Social Welfare Welfare Allowance (# of household heads code Allowance for Caregivers receiving the program) 11 Ulaanbaatar/Улаанбаатар 11,646 32.7% 11,946 35.0% 2,715 18.0% 21 Dornod/Дорнод 1,018 2.9% 1,067 3.1% 1,603 10.6% 22 Sukhbaatar/Сүхбаатар 895 2.5% 993 2.9% 583 3.9% 23 Khentii/Хэнтий 1,082 3.0% 978 2.9% 893 5.9% 41 Tuv/Төв 741 2.1% 659 1.9% 271 1.8% 42 Govisumber/Говьсүмбэр 214 0.6% 203 0.6% 33 0.2% 43 Selenge/Сэлэнгэ 943 2.6% 1,060 3.1% 339 2.2% 44 Dornogovi/Дорноговь 896 2.5% 874 2.6% 231 1.5% 34 This includes those who received Social Welfare Allowance and/or Social Welfare Allowance for Caregivers and/or Food Stamp Program. REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA 45 Darkhan-Uul/Дархан-Уул 971 2.7% 1,126 3.3% 259 1.7% 46 Umnugovi/Өмнөговь 900 2.5% 930 2.7% 110 0.7% 48 Dundgovi/Дундговь 742 2.1% 533 1.6% 260 1.7% 61 Orkhon/Орхон 1,506 4.2% 1,456 4.3% 481 3.2% 62 Uvurkhangai/Өвөрхангай 1,832 5.1% 1,613 4.7% 1,130 7.5% 63 Bulgan/Булган 1,018 2.9% 918 2.7% 280 1.9% 64 Bayankhongor/Баянхонгор 1,361 3.8% 1,357 4.0% 545 3.6% 65 Arkhangai/Архангай 1,450 4.1% 1,374 4.0% 1,576 10.4% 67 Khuvsgul/Хөвсгөл 2,576 7.2% 2,594 7.6% 2,308 15.3% 81 Zavkhan/Завхан 943 2.6% 607 1.8% 285 1.9% 82 Govi-Altai/Говь-Алтай 677 1.9% 728 2.1% 198 1.3% 83 Bayan-Ulgii/Баян-Өлгий 1,411 4.0% 1,395 4.1% 399 2.6% 84 Khovd/Ховд 1,152 3.2% 461 1.3% 248 1.6% 85 Uvs/Увс 1,612 4.5% 1,294 3.8% 372 2.5% Total 35,586 100.0% 34,166 100.0% 15,119 100.0% Source: SW/PMT data, World Bank staff estimates. School attainment Beneficiaries of Social Welfare Allowances and household heads benefiting from the Food Stamp Program were less well-educated than the general population in the database (Table 14). Around 80 percent completed secondary schooling or attained less than secondary schooling (compared to 66 percent for the general population in the database). The proportion of those with no education was more than twice the level of the general population in the database. 49 Table 14: Highest Level of Education Attained by Beneficiaries of Social Welfare Allowances, Benefits, and Concessions and the Food Stamp Program Level of schooling Number of beneficiaries Percentage 1 None 9,794 13.8 2 Primary 12,164 17.14 3 Secondary 14,839 20.91 4 Completed secondary 19,909 28.05 5 Technical/Vocational 5,860 8.26 6 Higher 8,406 11.84 Total 70,972 100 Source: SW/PMT data, World Bank staff estimates. Disability One-quarter of Social Welfare Allowances, Benefits, and Concessions beneficiaries and household heads benefiting from the Food Stamp Program reported having disabilities. The most common disability was physical impairment (25.0 percent), followed by mental impairment (22.2 percent). Around 17 percent of those with disabilities suffered from multiple disabilities (Table 15). Table 15: Number of Individuals with Disabilities by Type of Disability among Beneficiaries of Social Welfare Allowances, Benefits, and Concessions and the Food Stamp Program Type of disability Individuals with disability Percentage 1 Blind (Харааны) 2,053 10.54 REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA 2 Mute (Хэл ярианы) 592 3.04 3 Hearing-impaired (Сонсголын) 584 3 4 Physically impaired (Хөдөлгөөний) 4,863 24.98 5 Mentally impaired (Сэтгэц, оюуны) 4,330 22.24 6 Combined (Хавсарсан) 3,347 17.19 7 Others (Бусад) 3,700 19 Total 19,469 100 Source: SW/PMT data, World Bank staff estimates. Work Labor participation among beneficiaries was low, with only 39.5 percent reporting to have worked in the past week. Among those who did not work in the past week, only 10.6 percent worked during the previous year. The proportion of long(er)-term unemployed adults among the beneficiaries with no work in the past year was high, close to double that of the general population in the database. PMT score quintiles Social Welfare Transfers are largely progressive, providing more benefits to the poor compared to the better-off. Half of those receiving Social Welfare Allowances, Benefits, and Concessions or the Food Stamp Program were categorized as being in the poorest quintile, according to their PMT scores (Table 16 and Figure 22). Table 16: Beneficiaries of Social Welfare Transfers by PMT Score Quintile PMT score quintiles Social Welfare Transfer beneficiaries Percentage 1 - lowest (poor) to 5 - highest (better off) (including Food Stamp Program) 50 1 38,278 49.88 2 15,039 19.6 3 10,874 14.17 4 8,140 10.61 5 4,402 5.74 Total 76,733 100 Source: SW/PMT data, World Bank staff estimates. Figure 22: Percentage of Individuals in Each Household PMT Score Quintile and the Proportion of Social Welfare Allowance Beneficiaries from Each Household PMT Score Quintile Source: SW/PMT data, World Bank staff estimates. REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA Social Welfare Services Articles 18 and 19 in the Social Welfare Law (January 19, 2012 version) specify the services provided under Community-Based Welfare Services and Specialized Care Services. Community-Based Welfare Services include the following: •Organized training for confidence building, self-reliance, working skills talent development; •Counseling; •Rehabilitation; •Temporary accommodation and care; •Day care service; •Home-based care and service; •Other social welfare services based on the needs of citizens and households; •Support to homeless citizens and their family members for socializing, civil registration, and temporary shelter; and Socialization of citizens and households requiring social welfare assistance, help in forming a community group, implementation of income generation projects, and life skills training. These services are targeted to the following ten beneficiary types: • Elderly; • Disabled; • Children in difficult conditions; • Victims of violence; • Citizens released from prison; • Alcohol- and drug-addicted persons; 51 • Citizens with incurable disease; • Homeless households and individuals; • Migrant households and individuals in need of social welfare assistance; and Single parents. Specialized care services include the following: • Specialized care services for the elderly; • Institutional care services for disabled persons; • Institutional care services for children (18 years of age and below) living in difficult conditions; and • Institutional care for triplets or more. These specialized care services are targeted to the following types of beneficiaries: • A single elderly person who cannot live independently and has no child and/or has a child who is elderly/disabled and thus cannot take care of him/her; • A single disabled citizen incapable of living independently who has no child and requires professional service and special conditions; • A child in a difficult situation, a child specified in paragraph 25.5 of the Family Law, and a disabled child 18 years old and below; and • Triplets, quadruplets, and so on who are four years old or younger, with parental permission to be based in a specialized home at the expense of the State. Expenditures on Social Welfare Services In absolute terms, the expenditures and budget for Social Welfare Services are the smallest compared to all other programs. Nonetheless, total expenditures on Social Welfare Services doubled in real terms between 2010 and 2012, heavily driven by rehabilitation services. Other than the expenditures REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA on rehabilitation services, the expenditures for all other Social Welfare Services programs have been flat in real terms since 2010 (Figure 23). Figure 23: Trends in Inflation-Adjusted Annual Expenditures on Social Welfare Services (in 1,000,000 MNT usig 2010 as the base year) Source: GOSWS. 52 Characteristics of Beneficiaries of Social Welfare Services Assuming that there were no technical issues with the administrative data for Social Welfare Services, according to the SW Admin/PMT data, there were only 6,455 users35 of Social Welfare Services nationwide (Table 17). This is approximately 0.3 percent of the population in the database. The very low utilization of Social Welfare Services is most likely due to the fact that these programs are still new and emerging. Among the recorded Social Welfare Services users, the average age was 53.7 years old, with 63.7 percent being female and 36.3 percent being male. Table 17: Number of Individual Beneficiaries Receiving Social Welfare Services by Aimag Geo- Aimag Community-Based Welfare Specialized Care Services graphical Services code individuals percentage individuals percentage 11 Ulaanbaatar/Улаанбаатар 2,917 45.19 1 0.91 21 Dornod/Дорнод 5 0.08 22 Sukhbaatar/Сүхбаатар 78 1.21 23 Khentii/Хэнтий 7 0.11 -- -- 41 Tuv/Төв 154 2.39 -- -- 42 Govisumber/Говьсүмбэр 29 0.45 -- -- 43 Selenge/Сэлэнгэ 1,784 27.64 -- -- 44 Dornogovi/Дорноговь 11 0.17 -- -- 45 Darkhan-Uul/Дархан-Уул 570 8.83 -- -- 35. Some of the 6,455 individuals used the services multiple times. REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA 46 Umnugovi/Өмнөговь 2 0.03 -- -- 48 Dundgovi/Дундговь 50 0.77 -- -- 61 Orkhon/Орхон 192 2.97 -- -- 62 Uvurkhangai/Өвөрхангай 4 0.06 -- -- 63 Bulgan/Булган 10 0.15 -- -- 64 Bayankhongor/Баянхонгор 10 0.15 -- -- 65 Arkhangai/Архангай 6 0.09 -- -- 67 Khuvsgul/Хөвсгөл 164 2.54 -- -- 81 Zavkhan/Завхан 3 0.05 -- -- 82 Govi-Altai/Говь-Алтай 5 0.08 -- -- 83 Bayan-Ulgii/Баян-Өлгий -- -- 84 Khovd/Ховд 453 7.02 109 99.09 85 Uvs/Увс 1 0.02 Total 6,455 100 110 100 Source: SW/PMT data, World Bank staff estimates. In general, the numbers of service users/beneficiaries are very small. In particular, “organizing people with similar needs” and “training on nursing and caring service skills” have small numbers of beneficiaries recorded in the administrative data. As for specialized care, nursing services for the disabled and for disabled children under age 16 both have small numbers of beneficiaries recorded in the administrative data (Table 18). Table 18: Number of Individual Beneficiaries Receiving Social Welfare Services Community-based care Number of beneficiaries (same individual using multiple services are 53 counted according to the number of services received) Advice and training for elderly, disabled, orphans, victims of violence 957 Assistance in income generation for elderly, disabled, orphans, victims of violence 1053 Rehabilitation services 4209 Temporary shelter 165 Home nursing and welfare services 475 Organizing people with similar needs 30 Training on nursing and caring service skills 7 Specialized care Specialized nursing services for the elderly 100 Specialized nursing services for the disabled 24 Specialized nursing services for (disabled) children under 16 18 Source: SW/PMT data, World Bank staff estimates. School attainment Users of Social Welfare Services were less well-educated than the general population in the database. More people had no education (7.5 percent among the users, compared to 4.8 percent among the general population) or only primary education (12.4 percent compared to 8.0 percent). There were also less Social Welfare Service users who completed higher education (15 percent) compared to the general population in the database (25 percent) (Table 19). REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA Table 19: Highest Level of Education Attained by Adult Users of Social Welfare Services Level of schooling Number of individuals Percentage 1 None 437 7.48 2 Primary 727 12.44 3 Secondary 996 17.04 4 Completed secondary 1,718 29.39 5 Technical/Vocational 900 15.4 6 Higher 1,068 18.27 Total 5,846 100 Source: SW/PMT data, World Bank staff estimates. Disability Around 29 percent of Social Welfare Services users reported having disabilities. Both physical (24.8 percent) and mental (23.2 percent) disabilities were common among the users (Table 20). Table 20: Number of Individuals with Disabilities by Type of Disability among Beneficiaries of Social Welfare Services Type of disability Individuals with disability Percentage 1 Blind (Харааны) 164 8.87 2 Mute (Хэл ярианы) 57 3.08 3 Hearing-impaired (Сонсголын) 143 7.74 4 Physically impaired (Хөдөлгөөний) 458 24.78 5 Mentally impaired (Сэтгэц, оюуны) 429 23.21 54 6 Combined (Хавсарсан) 234 12.66 7 Others (Бусад) 363 19.64 Total 1,848 100 Source: SW/PMT data, World Bank staff estimates. Work Among working-age users of Social Welfare Services, labor participation was low. Around 29 percent reported working the previous week, and among those who did not work during the previous week, 5.7 percent reported working during the previous year. PMT score quintiles Social Welfare Services are neither progressive nor regressive and seem to have even coverage across the wealth quintiles (Figure 24). Around 27 percent of the users were identified as being in the least well-off category, while around 11 percent were categorized as being among the well-off, according to their PMT scores (Table 21). REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA Figure 24: Percentage of Individuals in Each Household PMT Score Quintile and the Proportion of Social Welfare Service Beneficiaries in Each Household PMT Score Quintile Source: SW/PMT data, World Bank staff estimates. Table 21: Number of Beneficiaries of Social Welfare Services by PMT Score Quintile 55 PMT score quintiles Social Welfare Service Percentage 1 - lowest (poor) to 5 - highest (better off) beneficiaries 1 1,727 26.80 2 1,413 21.93 3 1,311 20.34 4 1,269 19.69 5 724 11.24 Total 6,444 100 Source: SW/PMT data, World Bank staff estimates. Assistance to the Elderly Article 5 of the Law on Social Security of Senior Citizen (2005) identifies the following assistance and allowances to be provided to senior citizens: Reimbursement every five years for costs of prosthetic correction of feet, hands, and teeth and orthopedic ear trumpet or eye patch purchased or made domestically; Allowance for reimbursement of voucher to stay in nursing homes and sanatoriums for the elderly; Annual financial assistance to pay the apartment fee if a senior citizen does not have a child to take care of him/her, or it is determined that his/her legal caretaker is not able to provide the needed support and assistance, or to buy fuel for a senior citizen who is a long-time blood donor (referred to as an “honored donor”) if he/she lives in an apartment without centralized heating or in a gher; One-way transportation fee and 50 percent of rest or treatment fee once a year for an honored REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA donor senior citizen or senior citizen in unavoidable need of medical treatment and care in a domestic sanatorium and nursing center; Financial assistance equal to 75 percent of funeral expenses specified in the Social Welfare legislation if the single elderly is not entitled to a funeral pension under the Social Welfare legislation; Free public transportation (except taxi) in the capital city and aimag centers regardless of territorial jurisdiction; Services provided at a priority at organizations of trade, transportation, communication, health, and daily public service; and Annual reimbursement of a one-way transportation fee if the senior citizen resides in the countryside 1,000 kilometers or more from the capital city and is hospitalized or has medical examinations in the capital city according to a decision by the physicians’ commission of the local medical center. Most of the allowances and assistance above are provided as reimbursements, except for the annual assistance for apartment fee which provides MNT 140,000 (approximately USD 76). Expenditures on Allowances and Assistance for the Elderly Although there has been some increase in the expenditures for Orthopedic Facilities for the Elderly, the expenditures for Allowances and Assistance for the Elderly have remained constant in real terms since 2010 (Figure 25). Figure 25: Trends in Inflation-Adjusted Annual Expenditures on Allowances for the Elderly (in 1,000,000 MNT using 2010 as the bas year) 56 Source: GOSWS. Characteristics of Beneficiaries of Allowances and Assistance for the Elderly According to the SW Admin/PMT data, a total of 83,708 individuals benefited from the Allowances and Assistance for the Elderly. The beneficiaries were slightly concentrated in Ulaanbaatar relative to the distribution of the population in the SW Admin/PMT data (Table 22). REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA Table 22: Individual Beneficiaries Receiving Allowances and Assistance for the Elderly by Aimag Assistance and Allowances for the Elderly Geographical code Aimag Number of individuals Percentage 11 Ulaanbaatar/Улаанбаатар 32,885 39.29 21 Dornod/Дорнод 675 0.81 22 Sukhbaatar/Сүхбаатар 1,339 1.6 23 Khentii/Хэнтий 2,166 2.59 41 Tuv/Төв 2,802 3.35 42 Govisumber/Говьсүмбэр 457 0.55 43 Selenge/Сэлэнгэ 3,097 3.7 44 Dornogovi/Дорноговь 1,387 1.66 45 Darkhan-Uul/Дархан-Уул 5,674 6.78 46 Umnugovi/Өмнөговь 1,327 1.59 48 Dudgovi/Дундговь 1,858 2.22 61 Orkhon/Орхон 4,300 5.14 62 Uvurkhangai/Өвөрхангай 2,919 3.49 63 Bulgan/Булган 1,814 2.17 64 Bayankhongor/Баянхонгор 1,662 1.99 65 Arkhungai/Архангай 2,112 2.52 67 Khuvsgul/Хөвсгөл 3,327 3.97 81 Zavkhan/Завхан 3,075 3.67 82 Govi-Altai/Говь-Алтай 1,378 1.65 83 Bayan-Ulgii/Баян-Өлгий 3,192 3.81 57 84 Khovd/Ховд 3,585 4.28 85 Uvs/Увс 2,677 3.2 Total 83,708 100 Source: SW/PMT data, World Bank staff estimates. Among the beneficiaries of Allowances and Assistance for the Elderly, the average age was 67.5 years old, with 68.9 percent being women and 31.1 percent being men. The higher proportion of women is most likely attributable to the fact that women are considered senior citizens at age 55 while men are considered senior citizens at age 60 (according to the Law on Social Security for Senior Citizens), in addition to the longer life expectancy of women. School attainment The elderly receiving Allowances and Assistance were less well-educated than the general population in the database. More people had no education (almost 9 percent among beneficiaries compared to 4.8 percent among the general population) or only primary education (21.3 percent compared to 8.0 percent). However, there were more beneficiaries of this program who had completed technical or vocational education (15.2 percent) compared to the general population in the database (8.9 percent) (Table 23). REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA Table 23: Highest Level of Education Attained by Recipients of the Allowances and Assistance for the Elderly Level of schooling Number of individuals Percentage 1 None 7,510 8.97 2 Primary 17,799 21.27 3 Secondary 11,280 13.48 4 Completed secondary 16,713 19.97 5 Technical/Vocational 12,701 15.18 6 Higher 17,690 21.14 Total 83,693 100 Source: SW/PMT data, World Bank staff estimates. Disability Around 6.8 percent of recipients of the Allowances and Assistance for the Elderly had disabilities. Physical disabilities (28.3 percent) were most common, most likely because these disabilities require prosthetic and orthopedic treatments (Table 24). Table 24: Number of Individuals with Disabilities by Type of Disability among Beneficiaries of Allowances and Assistance for the Elderly Type of disability Individuals with disability Percentage 1 Blind (Харааны) 1,014 17.9 2 Mute (Хэл ярианы) 124 2.19 3 Hearing-impaired (Сонсголын) 375 6.62 58 4 Physically impaired (Хөдөлгөөний) 1,600 28.25 5 Mentally impaired (Сэтгэц, оюуны) 459 8.1 6 Combined (Хавсарсан) 916 16.17 7 Others (Бусад) 1,176 20.76 Total 5,664 100 Source: SW/PMT data, World Bank staff estimates. PMT score quintiles Allowances and Assistance for the Elderly are slightly regressive and seem to benefit the higher wealth quintiles more than the lower quintiles (Figure 26). Only 15 percent of the beneficiaries were from the poorest quintile (Table 25). REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA Figure 26: Percentage of Individuals in Each Household PMT Score Quintiles and the Proportion of Beneficiaries of Allowances for the Elderly from Each Household PMT Score Quintile Source: SW/PMT data, World Bank staff estimates. Table 25: Beneficiaries of Allowances and Assistance for the Elderly by PMT Score Quintile PMT score quintiles Assistance for the Elderly Percentage 1 - lowest (poor) to 5 - highest (better off) beneficiaries 1 12,348 14.77 2 17,720 21.19 3 19,180 22.94 59 4 18,701 22.36 5 15,675 18.74 Total 83,624 100 Source: SW/PMT data, World Bank staff estimates. Benefits for the Elderly with State Merits Article 5 on assistance and allowances for senior citizens in the Law on Social Security of Senior Citizen (2005) includes additional financial benefits from the government for senior citizens with state merits (such as war veterans and those awarded Hero of Mongolia, Labor Hero, and other honors). The benefit types include: Monthly financial assistance to a veteran of war or surviving spouse of a citizen who died fulfilling his duties in the war, and o Monthly financial assistance to a winner of the Mongolian State Prize, State Honor, State Udarnik, and senior veteran of the revolutionary struggle, including: o Reimbursement of a two-way transportation fee for those residing over 1,000 kilometers from the capital and traveling for medical examination in the capital city once a year; o Reimbursement once a year of the voucher and two-way transportation for domestic sanatoriums and nursing homes; and o Financial assistance once a year for payment of an apartment fee or for fuel for senior citizens living in a gher or an apartment without centralized heating. The cash assistance for war veterans is MNT 200,000 (approximately USD 108) a month. Cash assistance for the winner of the Mongolian State Prize, state nominee, and State Udarnik and REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA senior veteran of the revolutionary struggle is MNT 150,000 (approximately USD 81) a month. The reimbursements for voucher and two-way transportation for sanatoriums and nursing homes are assigned a cash value of MNT 140,000 (approximately USD 76). The annual assistance for apartment and fuel costs is also MNT 140,000. Expenditures on Benefits for the Elderly with State Merits Perhaps due to the decreasing number of those eligible for these benefits (particularly war veterans), total expenditures for Benefits for the Elderly with State Merits have been declining in real terms since 2010. According to the 2013 budget, these total expenditures were expected to decline further. However, the expenditures on reimbursements for transport and housing for the Elderly with State Merits have remained constant over the years (Figure 27). Figure 27: Trends in Inflation-Adjusted Annual Expenditures on Benefits for the Elderly with State Merits (in 1,000,000 MNT using 2010 as the base year) 3,000 Cash assistance to State and Labor Heroes, and war veterans 2,500 and spouses Cash assistance to elders with state honors and prize winners 2,000 Transport for elders with honors and heroes Transport to local sanatorium 60 1,500 for elders with honors and heroes Housing or rewood for elders 1,000 with honor and heroes Fixed bene ts for State Heroes, Labor Heroes, People's honors 500 seniors Fixed bene ts for State Honors Seniors from State - 2010 2011 2012 2013 (budget) Source: GOSWS. Characteristics of Beneficiaries of Benefits for the Elderly with State Merits Perhaps not surprisingly, only 1,864 individuals were recorded in the administrative data as being recipients of the Benefits for the Elderly with State Merits. This was less than 0.1 percent of the population. The highest concentration of beneficiaries was in Ulaanbaatar. On average, the beneficiaries received three services/benefits, of which two services/benefits were specific to the elderly with state merits. The average age of beneficiaries of the state merits program was high at 76 years. Around 82 percent of the beneficiaries were male, compared to only 18 percent female. Table 26: Number of Individual Beneficiaries Receiving Benefits for the Elderly with State Merits and Average Number of Programs Received by Aimag REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA Geographical code Aimag Number of beneficiaries Percentage 11 Ulaanbaatar/Улаанбаатар 1,253 67.22 21 Dornod/Дорнод 19 1.02 22 Sukhbaatar/Сүхбаатар 11 0.59 23 Khentii/Хэнтий 27 1.45 41 uv/Төв 45 2.41 42 Govisumber/Говьсүмбэр 9 0.48 43 Selenge/Сэлэнгэ 40 2.15 44 Dornogovi/Дорноговь 22 1.18 45 Darkhan-Uul/Дархан-Уул 47 2.52 46 Umnugovi/Өмнөговь 23 1.23 48 Dudgovi/Дундговь 36 1.93 61 Orkhon/Орхон 45 2.41 62 Uvurkhangai/Өвөрхангай 27 1.45 63 Bulgan/Булган 24 1.29 64 Bayankhongor/Баянхонгор 13 0.7 65 Arkhungai/Архангай 25 1.34 67 Khuvsgul/Хөвсгөл 34 1.82 81 Zavkhan/Завхан 30 1.61 82 Govi-Altai/Говь-Алтай 24 1.29 83 Bayan-Ulgii/Баян-Өлгий 44 2.36 61 84 Khovd/Ховд 45 2.41 85 Uvs/Увс 21 1.13 Total 1,864 100 Source: SW/PMT data, World Bank staff estimates. School attainment School attainment of those receiving Benefits for the Elderly with State Merits was extremely high. Over 60 percent of the beneficiaries had higher than vocational/technical training. Given the high average age of the beneficiaries, the high levels of education may reflect the fact that the beneficiaries belonged to a privileged sector of the population (Table 27). Table 27: Highest Level of Education Attained by Benefits for the Elderly with State Merits Beneficiaries Level of schooling Number of individuals Percentage 1 None 119 6.39 2 Primary 247 13.26 3 Secondary 68 3.65 4 Completed secondary 178 9.55 5 Technical/Vocational 121 6.49 6 Higher 1,130 60.65 Total 1,863 100 Source: SW/PMT data, World Bank staff estimates. REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA Disability Around 4 percent of beneficiaries reported having disabilities, similar to the general population in the database. Physical disabilities (32 percent) and blindness (26 percent) were among the most common disabilities, with one in five beneficiaries with disabilities suffering from multiple disabilities (Table 28). Table 28: Individuals with Disabilities by Type of Disability among Beneficiaries of Benefits for the Elderly with State Merits Type of disability Individuals with disability Percentage 1 Blind (Харааны) 19 25.68 2 Mute (Хэл ярианы) 1 1.35 3 Hearing-impaired (Сонсголын) 2 2.7 4 Physically impaired (Хөдөлгөөний) 24 32.43 5 Mentally impaired (Сэтгэц, оюуны) 1 1.35 6 Combined (Хавсарсан) 16 21.62 7 Others (Бусад) 11 14.86 Total 74 100 Source: SW/PMT data, World Bank staff estimates. PMT score quintiles The PMT score reveals the regressive nature of Benefits for the Elderly with State Merits and the good social standing of the beneficiaries. Around 74 percent of the beneficiaries fell within the highest 40 percent of PMT scores (Table 29 and Figure 28). 62 Table 29: Beneficiaries of Benefits for the Elderly with State Merits by Wealth Quintiles PMT score quintiles Benefits for the Elderly with State Percentage 1 - lowest (poor) to 5 - highest (better off) Merits beneficiaries 1 86 4.62 2 150 8.06 3 247 13.27 4 447 24.01 5 932 50.05 Total 1,862 100 Source: SW/PMT data, World Bank staff estimates. REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA Figure 28: Percentage of Individuals in Each Household PMT Score Quintile and the Proportion of Beneficiaries of the Benefits for the Elderly with State Merits by Household PMT Score Quintile Source: SW/PMT data, World Bank staff estimates. Assistance and Concessions for the Disabled According to the Law on Social Security and a Person with Disability (2005), a person with disability is defined as an individual with no capacity or limited capacity to participate in social relations like other persons for more than 12 months due to body defect or mental, psychological, and perceptional disorders. 63 Article 5 of the Law describes the allowances and assistance to the disabled as: • Annual financial assistance to pay apartment fees for fully blind, fully dumb-deaf persons, dwarf individuals, and persons with disabilities who have fully lost their labor capacity and are in need of permanent care and to buy fuel if they live in an apartment without centralized heating or in a gher; • For minors with disability (up to 18 years of age), reimbursement once in two years for 100 percent of the cost of prosthetic correction in the country; • For persons with disability who are not entitled to receive allowances for rehabilitation and prosthetic correction due to industrial accident and occupational diseases from the Health Insurance Fund, reimbursement once in three years for 100 percent of the cost of prosthetic correction in the country; • For minors with disability (up to 18 years of age) and persons with disability who are not entitled to receive allowances or benefits for rehabilitation and prosthetic correction due to industrial accident and occupational diseases from the Health Insurance Fund, reimbursement once in three years for 100 percent of purchased or made special care instruments like orthopedic tools, wheelchairs, and other equipment made in the country; • Annual assistance for two-way transport to/from school and kindergarten for a minor with disability and his/her guardian or caregiver, or provision of a bus for them; • Reimbursement of one-way transportation fee and 50 percent of rest and treatment fee once a year in the domestic sanatorium and nursing centers for a minor with disability, his/her guardian or caregiver, a person with disability who lost 50 percent or more of his/her labor capacity and is not entitled to receive allowances for rehabilitation and prosthetic correction due to industrial accident and occupational diseases from the Health Insurance Fund; • Discount in the price of kindergarten food for a minor with disability and one child of a person who fully lost his/her labor capacity; • Discount in communication expenses for a blind adult; • Reimbursement once a year for a one-way transportation fee if a person with disability who resides REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA 1,000 kilometers away from the capital city will receive medical treatment or undergo medical tests in the capital city according to a decision of the specialist doctors’ commission of the local aimag clinic; • Free postal delivery of documents written by a blind person such as Braille letters, postcards, and parcels up to 10 kilograms and equipment, devices, or facilities for blind persons in the country; • Reimbursement once a year for 75 percent of a two-way transportation fee if a blind person needs to travel from the aimag to the capital city and from the capital city to the aimag to receive medical treatment or care at a domestic nursing home and sanatorium based on the conclusion of a medical institution; • Reimbursement once a year for 50 percent of a summer camp fee if a minor with disability stays in a summer camp; • Financial assistance equal to 75 percent of funeral expenditure from the social insurance fund for single disabled persons or minors with disability who are not entitled to receive a funeral pension under social insurance legislation; • One-time financial assistance equal to the pension during the period a person with disability competes at the Olympics or a continental or world-class competition and wins a gold, silver, or bronze medal; • Free public transportation facilities (except taxis) for persons with disability in the capital city and provincial centers notwithstanding the territorial jurisdiction; and • Priority services for persons with disability by trade, transportation, communication, health and public service organizations. Most of the allowances above are reimbursements of fees and are not assigned a monetary value, with the exception of: MNT 140,000 (approximately USD 76) for the annual apartment fee or fuel cost allowances; MNT 200,000 (approximately USD 108) annually for transportation costs for school for children with disability and their guardians; and MNT 20,000 (approximately USD 11) monthly for the communication allowance for the blind. 64 Expenditures on Allowances and Assistance for the Disabled Although expenditures on housing assistance and reimbursements for orthopedic facilities for the disabled appear to have dipped in 2012, total expenditures for these programs for the disabled have been constant in real terms over the years (Figure 29). REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA Figure 29: Trends in Inflation-Adjusted Annual Expenditures on Allowances and Assistance for the Disabled (in 1,000,000 MNT using 2010 as he base year) Source: GOSWS. 65 A total of 33,897 individuals received the Allowances and Concessions for the disabled (Table 30). The average age of beneficiaries was 44.7 years old, with almost equal numbers of males and females. Table 30: Beneficiaries Receiving Allowances and Concessions for the Disabled by Aimag Geographical code Aimag Allowances for the Disabled Individuals Percentage 11 Ulaanbaatar/Улаанбаатар 13,585 40.08 21 Dornod/Дорнод 558 1.65 22 Sukhbaatar/Сүхбаатар 637 1.88 23 Khentii/Хэнтий 733 2.16 41 Tuv/Төв 588 1.73 42 Govisumber/Говьсүмбэр 420 1.24 43 Selenge/Сэлэнгэ 1,837 5.42 44 Dornogovi/Дорноговь 850 2.51 45 Darkhan-Uul/Дархан-Уул 1,114 3.29 46 Umnugovi/Өмнөговь 428 1.26 48 Dudgovi/Дундговь 693 2.04 61 Orkhon/Орхон 1,192 3.52 62 Uvurkhangai/Өвөрхангай 909 2.68 63 Bulgan/Булган 437 1.29 64 Bayankhongor/Баянхонгор 784 2.31 65 Arkhungai/Архангай 710 2.09 67 Khuvsgul/Хөвсгөл 1,732 5.11 REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA 81 Zavkhan/Завхан 1,715 5.06 82 Govi-Altai/Говь-Алтай 479 1.41 83 Bayan-Ulgii/Баян-Өлгий 1,619 4.78 84 Khovd/Ховд 1,352 3.99 85 Uvs/Увс 1,525 4.5 Total 33,897 100 Source: SW/PMT data, World Bank staff estimates. School attainment Probably due to disabilities, the level of educational achievement for beneficiaries of the Allowances and Concessions for the Disabled was low. The percentage of those with no education was about four times higher (19.8 percent) than the general population in the SW Admin/PMT database (4.8 percent). However, there were more beneficiaries of this program who completed technical/vocational education (10.4 percent) compared to the general population in the SW Admin/PMT database (8.9 percent), although not for higher education (Table 31). Table 31: Highest Level of Education Attained by Recipients of the Allowances and Concessions for the Disabled Level of schooling Number of individuals Percentage 1 None 6,392 19.82 2 Primary 3,422 10.61 3 Secondary 5,675 17.6 4 Completed secondary 9,428 29.24 66 5 Technical/Vocational 3,356 10.41 6 Higher 3,973 12.32 Total 32,246 100 Source: SW/PMT data, World Bank staff estimates. Disability Interestingly, only 76 percent of beneficiaries of Allowances and Concessions for the Disabled reported having disabilities in the PMT Survey. Mental disability (22.7 percent), physical disability (20.5 percent), and blindness (10.4 percent) were the most common disabilities reported (Table 32). Table 32: Individuals with Disabilities by Type of Disability among Beneficiaries of the Allowance and Concessions for the Disabled Type of disability Individuals with disability Percentage 1 Blind (Харааны) 2,675 10.35 2 Mute (Хэл ярианы) 1,411 5.46 3 Hearing-impaired (Сонсголын) 1,845 7.14 4 Physically impaired (Хөдөлгөөний) 5,302 20.52 5 Mentally impaired (Сэтгэц, оюуны) 5,883 22.77 6 Combined (Хавсарсан) 3,703 14.33 7 Others (Бусад) 5,023 19.44 Total 25,842 100 Source: SW/PMT data, World Bank staff estimates. REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA PMT score quintiles Allowances and Concessions for the Disabled are progressive, with a large proportion of beneficiaries in the poorest wealth quintiles (Table 33 and Figure 30). However, it is important to note that disability was given considerable weight in the PMT formula, so the fact that there is a higher proportion of disabled beneficiaries in poorer quintiles may be driven by the design of the formula. Table 33: Beneficiaries of Allowances and Concessions for the Disabled by Household PMT score Quintile PMT score quintiles Allowances and Assistance for the Percentage 1 - lowest (poor) to 5 - highest (better off) Disabled beneficiaries 1 12,909 38.15 2 8,032 23.73 3 6,039 17.85 4 4,515 13.34 5 2,346 6.93 Total 33,841 100 Source: SW/PMT data, World Bank staff estimates. Figure 30: Percentage of Individuals in Each Household PMT Score Quintile and the Proportion of Beneficiaries of the Allowances and Concessions for the Disabled from Each Household PMT Score Quintile 67 Source: SW/PMT data, World Bank staff estimates. Allowances for Mothers and Children Two types of programs are in this category: allowances for the “Mother’s Glory” (also known as Mother Heroes) and the Allowance for Pregnant and Lactating Mothers. The Mother Heroes program, described in Article 13 of the Law on Social Welfare (2012), recognizes mothers who have given birth to four or more children. The 1st rank medal of Mother Heroes is for mothers who have given birth to more than six children, and the 2nd rank medal is for mothers who have given birth to four children. The Allowance for Pregnant and Lactating Mothers, also in the Article of the same law, is provided monthly for 12 months starting from the fifth month of pregnancy. Under the Mother Heroes program, 1st medal recipients receive an annual benefit of MNT 200,000 REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA (approximately USD 108), while 2nd medal recipients receive an annual benefit of MNT 100,000 (approximately USD 54). The Allowance for Pregnant and Lactating Mothers is a monthly allowance of MNT 40,000 (approximately USD 22). Although the Allowance for Pregnant and Lactating Mothers is provided only when the mothers are pregnant or lactating, the Mother Heroes benefits are provided for a lifetime once the mother becomes an honoree. This lifetime benefit puts immense fiscal pressure on the government. These two programs—particularly the Mother Heroes benefits—should be considered programs to promote population growth rather than to provide social welfare assistance for beneficiaries to meet minimum needs. Expenditures on Allowances for Mothers Total expenditures on Allowances for Mothers have increased slightly since 2010 in real terms. Although expenditures for Mother Heroes have declined, the Allowance for Pregnant and Lactating Mothers has increased (Figure 31). The numbers of beneficiaries in both programs have remained stable over the years. Figure 31: Trends in Inflation-Adjusted Annual Expenditures on Allowances for Mothers (in 1,000,000 MNT using 2010 as the base year) 30,000 25,000 20,000 Pregnant and lacta ng 15,000 mothers 10,000 Mother Heroes I & II 68 5,000 - 2010 2011 2012 2013 (budget) Source: SW/PMT data, World Bank staff estimates. Characteristics of Beneficiaries of Allowances for Mothers A total of 235,059 women received either the Allowance for Pregnant and Lactating Mothers or Mother Heroes benefits. The proportion of beneficiaries seems proportionally distributed across aimags relative to the population in the SW Admin/PMT database (Table 34). The average age of beneficiaries was 48.7 years old. Around 2 percent of beneficiaries were 15 years old or younger, and over half of Mother Heroes benefit recipients were 50 years old or older. Table 34: Beneficiaries Receiving Allowance for Pregnant and Lactating Mothers and Mother Heroes Benefits by Aimag Geographical code Aimag Allowance for Pregnant and Lactating Mothers and Mother Heroes Benefits Individuals Percentage 11 Ulaanbaatar/Улаанбаатар 79,798 33.95 21 Dornod/Дорнод 6,992 2.97 22 Sukhbaatar/Сүхбаатар 5,695 2.42 23 Khentii/Хэнтий 6,669 2.84 41 Tuv/Төв 8,735 3.72 REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA Geographical code Aimag Allowance for Pregnant and Lactating Mothers and Mother Heroes Benefits 42 Govisumber/Говьсүмбэр 1,429 0.61 43 Selenge/Сэлэнгэ 8,780 3.74 44 Dornogovi/Дорноговь 5,427 2.31 45 Darkhan-Uul/Дархан-Уул 7,241 3.08 46 Umnugovi/Өмнөговь 5,297 2.25 48 Dudgovi/Дундговь 4,683 1.99 61 Orkhon/Орхон 6,787 2.89 62 Uvurkhangai/Өвөрхангай 10,153 4.32 63 Bulgan/Булган 5,845 2.49 64 Bayankhongor/Баянхонгор 8,055 3.43 65 Arkhungai/Архангай 8,617 3.67 67 Khuvsgul/Хөвсгөл 12,300 5.23 81 Zavkhan/Завхан 6,818 2.9 82 Govi-Altai/Говь-Алтай 5,409 2.3 83 Bayan-Ulgii/Баян-Өлгий 11,547 4.91 84 Khovd/Ховд 9,253 3.94 85 Uvs/Увс 9,529 4.05 Total 235,059 100 Source: SW/PMT data, World Bank staff estimates. 69 School attainment Beneficiaries of the Allowance for Pregnant and Lactating Mothers were slightly less well-educated than the general population in the database. In particular, the proportions with no education, primary education, and some secondary-level education were higher than those in the general population in the database (Table 35). Table 35: Highest Level of Education Attained by Beneficiaries of the Allowance for Pregnant and Lactating Mothers and Mother Heroes Benefits Level of schooling Number of individuals Percentage 1 None 14,192 6.05 2 Primary 33,847 14.43 3 Secondary 43,744 18.65 4 Completed secondary 72,403 30.86 5 Technical/Vocational 28,031 11.95 6 Higher 42,371 18.06 Total 234,588 100 Source: SW/PMT data, World Bank staff estimates. Disability Around 4.6 percent of beneficiaries of the Allowance for Pregnant and Lactating Mothers and Mother Heroes have disabilities, which is almost the same as the rate for the population in the SW Admin/PMT database (4.5 percent). Among those with disabilities, around 23.6 percent suffer from REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA physical impairments. Compared to the general population in the database, less mental impairments were reported among this group (Table 36). Table 36: Individuals with Disabilities by Type of Disability among Beneficiaries of the Allowance for Pregnant and Lactating Mothers and Mother Heroes Type of disability Individuals with disability Percentage 1 Blind (Харааны) 1,315 12.2 2 Mute (Хэл ярианы) 202 1.87 3 Hearing-impaired (Сонсголын) 873 8.1 4 Physically impaired (Хөдөлгөөний) 2,541 23.57 5 Mentally impaired (Сэтгэц, оюуны) 1,249 11.59 6 Combined (Хавсарсан) 1,426 13.23 7 Others (Бусад) 3,173 29.44 Total 10,779 100 Source: SW/PMT data, World Bank staff estimates. Work Among working-age beneficiaries of the Allowance for Pregnant and Lactating Mothers and the Mother Heroes program, 46.1 percent reported to have worked during the previous week. Among those who did not, 8.13 percent worked during the previous year. These rates of labor participation were lower than that of working-age women overall in the SW Admin/PMT database. Broken down by program, 45 percent of the working-age mothers receiving the Allowance for Pregnant and Lactating Mothers reported to have worked during the previous week, and among those 70 who did not, 15.0 percent worked during the previous year. Among the working-age mothers who were receiving Mother Heroes benefits, 46.7 percent worked during the previous week, while among those who did not, only 5 percent worked during the previous year. This suggests that although the short- term unemployed mothers receiving these programs were similar, working-age women receiving the Mother Heroes benefits were more likely not to be working in the long term. PMT score quintiles The Allowance for Mothers and Children is slightly pro-poor. Around 28 percent of the beneficiaries were categorized as belonging to the lowest quintile according to PMT scores, with the proportions declining among the better-off. Among those categorized as belonging to the highest quintile, only 12 percent were identified as benefiting from this program (Table 37 and Figure 32). Table 37: Beneficiaries of Allowance for Pregnant and Lactating Mothers and Mother Heroes by PMT Score Quintile PMT score quintiles Allowance for Pregnant and Percentage 1 - lowest (poor) to 5 - highest (better off) Lactating Mothers and Mother Heroes beneficiaries 1 65,683 27.97 2 53,433 22.76 3 47,509 20.23 4 40,247 17.14 5 27,923 11.89 Total 234,795 100 Source: SW/PMT data, World Bank staff estimates. REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA Figure 32: Percentage of Individuals in Each Household PMT Score Quintile and the Proportion of Beneficiaries of the Allowance for Pregnant and Lactating Mothers and Mother Heroes in Each Household PMT Score Quintile Source: SW/PMT data, World Bank staff estimates. The Food Stamp Program The legal basis of the Food Stamp Program is Article 22 on Food and Nutrition Support Services in 71 the Law on Social Welfare (2012). It states that Food and Nutrition Support Services will be provided to households and citizens requiring social welfare assistance, selected from households living lower than the current poverty line, and to homeless and wandering citizens. The program provides food stamps to purchase food, equivalent to MNT 10,000 (approximately USD 5.40) per adult per month and MNT 5,000 (approximately USD 2.70) per child per month. Characteristics of Beneficiaries of the Food Stamp Program According to the SW Admin/PMT database, 15,119 households benefited from the Food Stamp Program, comprising 2.3 percent of households in the database. Coverage of the Food Stamp Program was high in Khuvsgul, Arkhungai, and Dornod aimags (Table 38). In relative terms, the number of Food Stamp Program beneficiaries was low in Ulaanbaatar, considering its share of the population. The average age of Food Stamp Program beneficiary household heads was 44.9 years old (ranging from 15 to 98 years old), and 74 percent of them were male. Table 38: Beneficiary Households Receiving the Food Stamp Program by Aimag Geographical code Aimag Receiving Food Stamp Program Households Percentage 11 Ulaanbaatar/Улаанбаатар 2,715 18.0% 21 Dornod/Дорнод 1,603 10.6% 22 Sukhbaatar/Сүхбаатар 583 3.9% 23 Khentii/Хэнтий 893 5.9% 41 Tuv/Төв 271 1.8% 42 Govisumber/Говьсүмбэр 33 0.2% 43 Selenge/Сэлэнгэ 339 2.2% REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA 44 Dornogovi/Дорноговь 231 1.5% 45 Darkhan-Uul/Дархан-Уул 259 1.7% 46 Umnugovi/Өмнөговь 110 0.7% 48 Dudgovi/Дундговь 260 1.7% 61 Orkhon/Орхон 481 3.2% 62 Uvurkhangai/Өвөрхангай 1,130 7.5% 63 Bulgan/Булган 280 1.9% 64 Bayankhongor/Баянхонгор 545 3.6% 65 Arkhungai/Архангай 1,576 10.4% 67 Khuvsgul/Хөвсгөл 2,308 15.3% 81 Zavkhan/Завхан 285 1.9% 82 Govi-Altai/Говь-Алтай 198 1.3% 83 Bayan-Ulgii/Баян-Өлгий 399 2.6% 84 Khovd/Ховд 248 1.6% 85 Uvs/Увс 372 2.5% Total 15,119 100.0% Source: SW/PMT data, World Bank staff estimates. The average estimated amount received by households from the program was MNT 529,141 (approximately USD 286) per year. Considering that this program is targeted to the very poor (the bottom 5 percent of the wealth distribution), the amount is not very generous, particularly given the large family sizes (5.8 members on average) of these beneficiary households. This means that on average, the estimated benefit size was MNT 91,231 (approximately USD 49) per person per year. 72 School Attainment by Household Head Since education level of the household head is among the variables included in the PMT formula, it is not surprising that 94 percent of Food Stamp Program recipient household heads had completed secondary schooling or less. Compared to the fact that two-thirds of the general population had completed secondary schooling or less, household heads of the Food Stamp Program had considerably lower school attainment levels (Table 39). Table 39: Highest Level of Education/Qualification Attained by the Household Head of Food Stamp Program Beneficiary Households Level of schooling Number of household heads Percentage 1 None 2,414 15.97 2 Primary 3,210 21.23 3 Secondary 4,976 32.92 4 Completed secondary 3,653 24.16 5 Technical, Vocational 754 4.99 6 Higher 110 0.73 Total 15,117 100 Source: SW/PMT data, World Bank staff estimates. Disability Among the beneficiaries of the Food Stamp Program, the proportion of households with household heads with disabilities was 20.9 percent. This is considerably higher than the rate of disability of household heads in the general population in the database (7.2 percent). The most common disabilities reported were physical impairment (25.9 percent) followed by mental impairment (17.9 percent). REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA Table 40: Household Heads of Food Stamp Program Beneficiary Households with Disabilities, by Type of Disability Type of disability Individuals with disability Percentage 1 Blind (Харааны) 458 14.53 2 Mute (Хэл ярианы) 69 2.19 3 Hearing-impaired (Сонсголын) 241 7.65 4 Physically impaired (Хөдөлгөөний) 815 25.86 5 Mentally impaired (Сэтгэц, оюуны) 564 17.89 6 Combined (Хавсарсан) 373 11.83 7 Others (Бусад) 632 20.05 Total 3,152 100 Source: SW/PMT data, World Bank staff estimates. Work Among the 15,119 households receiving the Food Stamp Program, 10.8 percent of the household heads were not of working age (either younger than 18 years old or older than retirement age). One male beneficiary household head reported being 15 years old and one female household head reported being 16 years old, while the rest were above retirement age.36 Among working-age household heads, around 36.7 percent reported to have worked the previous week. Among those who did not work during the previous week, only 16.7 percent reported having worked during the previous year. Labor participation of the working-age household heads receiving the Food Stamp Program was considerably lower than that of the general population. PMT score quintiles 73 Except for one beneficiary household with a PMT score in the third quintile, all other households were in the poorest quintile of all households in the PMT database. Child Money Program The Child Money Program (CMP), funded from the Human Development Fund, provides MNT 20,000 (approximately USD 11) per month to all children from 0 to 18 years old. On average, beneficiary households had an average of two children between 0 to 18 years old. On average, these beneficiary households received MNT 480,000 (approximately USD 260) annually. The maximum number of children was 9 (in 29 households). Characteristics of Beneficiaries of the Child Money Program A total of 785,819 children (0 to 18 years of age) in the database were beneficiaries of the CMP, and the program had good coverage throughout all the aimags, proportionally distributed relative to the population in each aimag. The average age of CMP beneficiaries was 8 years old, with 50.9 percent of beneficiaries being male. 36 . The average age of Food Stamp Program beneficiary household heads who were not in the working-age group was 68.5 years old for male and 65.2 years old for female. REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA Table 41: Beneficiary Households Receiving the Child Money Program by Aimag Geographical code Aimag Receiving Child Money Program Individuals Percentage 11 Ulaanbaatar/Улаанбаатар 292,165 37.2% 21 Dornod/Дорнод 21,220 2.7% 22 Sukhbaatar/Сүхбаатар 17,161 2.2% 23 Khentii/Хэнтий 21,123 2.7% 41 Tuv/Төв 24,938 3.2% 42 Govisumber/Говьсүмбэр 4,729 0.6% 43 Selenge/Сэлэнгэ 28,374 3.6% 44 Dornogovi/Дорноговь 18,498 2.4% 45 Darkhan-Uul/Дархан-Уул 25,795 3.3% 46 Umnugovi/Өмнөговь 16,904 2.2% 48 Dudgovi/Дундговь 12,599 1.6% 61 Orkhon/Орхон 25,160 3.2% 62 Uvurkhangai/Өвөрхангай 33,830 4.3% 63 Bulgan/Булган 17,184 2.2% 64 Bayankhongor/Баянхонгор 26,089 3.3% 65 Arkhungai/Архангай 27,856 3.5% 67 Khuvsgul/Хөвсгөл 39,980 5.1% 81 Zavkhan/Завхан 22,126 2.8% 82 Govi-Altai/Говь-Алтай 18,248 2.3% 74 83 Bayan-Ulgii/Баян-Өлгий 34,693 4.4% 84 Khovd/Ховд 28,658 3.7% 85 Uvs/Увс 28,489 3.6% Total 785,819 100.0% Source: SW/PMT data, World Bank staff estimates. Education Of the 785,081 CMP beneficiaries, 74 percent reported attending kindergarten, school, or university. Broken down by age group, about 65.2 percent of the children between 3-5 years old were attending kindergarten. The vast majority (95 percent) of the children between 6-11 years old were attending school. The vast majority (97 percent) of 12-15 year olds were attending school, with 95 percent of those attending school. Around 90 percent of the 16-18 year olds were attending school, with 68 percent of them still attending school, 6 percent attending technical or vocational school, and 4.3 percent attending university. Disability The prevalence of disability among CMP beneficiaries was low at only 1.2 percent. Similar to the general population in the PMT database, the most common disability was physical impairment (22 percent) followed by mental impairment (20 percent). REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA Table 42: Number of Child Money Program Beneficiaries with Disabilities by Type of Disability Type of disability Individuals with disability Percentage 1 Blind (Харааны) 892 10.16 2 Mute (Хэл ярианы) 783 8.92 3 Hearing-impaired (Сонсголын) 461 5.25 4 Physically impaired (Хөдөлгөөний) 1,948 22.18 5 Mentally impaired (Сэтгэц, оюуны) 1,761 20.05 6 Combined (Хавсарсан) 1,440 16.4 7 Others (Бусад) 1,496 17.04 Total 8,781 100 Source: SW/PMT data, World Bank staff estimates. Work Among CMP beneficiaries who were 16 years old and above, 8 percent reported to have worked during the previous week. Among those who did not work during the previous week, only 2.3 percent reported to have worked the previous year. PMT score quintiles When CMP beneficiaries are broken down by PMT score quintile, it is clear that more children from poorer households received the CMP. Of the CMP beneficiaries, 30.4 percent belonged to households with the lowest PMT scores. The proportion of CMP beneficiaries was lower among households with higher PMT scores, and only 14 percent of children benefiting from the program were from the highest PMT score quintile (Table 43 and Figure 33). This is probably due to the fact that poorer households had more children between 0-18 years old compared to households that were better-off economically. 75 Table 43: Child Money Program Beneficiaries by PMT Score Quintile PMT score quintiles Number of beneficiaries Percentage 1 - lowest (poor) to 5 - highest (better off) 1 238,862 30.43 2 161,299 20.55 3 140,780 17.94 4 136,703 17.42 5 107,192 13.66 Total 784,836 100 Source: SW/PMT data, World Bank staff estimates. REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA Figure 33: Percentage of Individuals in Each Household PMT Score Quintile and the Proportion of Child Money Program Beneficiaries in Each Household PMT Score Quintile Source: SW/PMT data, World Bank staff estimates. 76 ANNEX II: INFLATION-ADJUSTED EXPENDITURES FOR TRANSFERS 2010 – 2013 2010 2011 2012 2013 Expenditure types (Thousands MNT MNT MNT MNT MNT) Number MNT MNT MNT (base total Number total Number total of total total total year) (Real) of (Real) of (Real) Social Welfare Pension 56,708 57,925 60,658 63,423 Pregnant and Lactating Mothers 87,171 87,171 88,971 94,882 Social Welfare Allowance 33,074 31,910 31,971 42,260 Community-Based Services 17,588 42,845 37,366 24,104 4,063,080 Social Welfare Support for 33,910 35,252 25,272 36,411 Caregivers Allowance for the Disabled 35,888 36,451 32,852 38,743 4,763,805 Allowance for the Elderly 118,672 114,103 125,941 Food Stamp Program 97,919 8,011,278 Benefits for Mother Heroes 204,313 204,378 202,474 Benefits for the Elderly with State 5,228 5,025 4,042 3,622,362 Merits Specialized Care Services 48,207 44,227 57,606 45,719 100,000 72,993 Child Money Program (Approved 970,000 Budget) Inflation Factor Year 2010 2011 2012 2013 Inflation Factor 100 109 126 137 Source: World Development Indicators. REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA 77 78 ANNEX III: PROPORTION OF BENEFICIARIES IN EACH SOCIAL WELFARE QUINTILE ACCORDING TO PMT SCORE First three Number of Proportion of Beneficiaries by Estimated Wealth Quintiles digits of Beneficiaries in the (PMT score) the Service SW Admin/PMT Q1 Q5 Codes Program database (Poorest) Q2 Q3 Q4 (Better off) Social Pension service111 Social pensions for elderly (single seniors) 724 41.6% 31.2% 15.2% 7.6% 4.4% Social pensions for elderly (guardians are elderly or service112 disabled) 169 51.5% 30.8% 11.8% 5.3% 0.6% service121 Disabled with lost labor capacity by 50 to 60 percent 17894 49.5% 22.8% 14.4% 9.2% 4.1% Disabled with lost labor capacity by more than 70 service122 percent 16479 51.4% 23.4% 13.3% 8.3% 3.6% service130 Vertically challenged (dwarfs) over 16 121 38.0% 24.8% 19.0% 10.7% 7.4% service140 Welfare benefits for single mother or father 22 59.1% 31.8% 0.0% 4.5% 4.5% service150 Children under 18 with no breadwinner 10927 39.1% 23.5% 17.3% 12.3% 7.7% service160 Elderly who do not qualify for Social Insurance 742 46.2% 27.4% 14.7% 8.4% 3.4% Social Welfare Allowance service211 Cash transfers for caregivers 29678 38.0% 23.3% 17.7% 13.6% 7.5% Benefits for those who adopted or nursing or caring for service212 an orphan 1574 38.2% 21.5% 16.7% 14.1% 9.4% service213 Benefits for those nursing children who need protection 300 41.7% 21.3% 18.0% 11.3% 7.7% service214 Benefits for those taking care of single, disabled, elderly 274 37.6% 26.3% 18.2% 12.4% 5.5% Cash/pensions for children under 16 needing service215 permanent care 5144 49.1% 20.3% 14.2% 11.4% 5.0% service224 Cash for those with twins or triplets 3578 37.8% 19.6% 16.2% 15.8% 10.6% Assistance for under-aged orphans or lost shelter due service225 to a disaster/accident 3115 45.6% 24.3% 15.2% 10.4% 4.6% REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA service227 Benefits for citizens requiring permanent nursing 25835 36.0% 26.2% 18.4% 12.8% 6.6% service228 Single parent with more than 3 children under age 14 1813 76.6% 13.1% 6.5% 2.8% 1.1% Community-based care First three Number of Proportion of Beneficiaries by Estimated Wealth Quintiles digits of Beneficiaries in the (PMT score) the Service SW Admin/PMT Q1 Q5 Codes Program database (Poorest) Q2 Q3 Q4 (Better off) Advice and training for elderly, disabled, orphans, service310 victims of violence 950 39.7% 24.4% 16.1% 13.6% 6.2% service311 Migrant family 0 Assistance in income generation for elderly, disabled, service320 orphans, victims of violence 1052 33.1% 24.4% 18.5% 15.2% 8.7% service330 Rehabilitation services 4206 22.9% 21.0% 21.8% 22.0% 12.3% service340 Temporary shelter 165 21.2% 16.4% 17.6% 23.0% 21.8% service350 Home nursing and welfare services 475 26.5% 25.3% 23.4% 16.2% 8.6% service360 (program name unknown) 30 33.3% 43.3% 6.7% 10.0% 6.7% service370 Training on nursing and caring service skills 7 71.4% 14.3% 14.3% 0.0% 0.0% Specialized care service410 Specialized nursing services for the elderly 100 5.0% 8.0% 16.0% 38.0% 33.0% service420 Specialized nursing services for the disabled 24 0.0% 4.2% 4.2% 58.3% 33.3% Specialized nursing services for (disabled) children service430 under 16 18 5.6% 0.0% 5.6% 55.6% 33.3% Specialized nursing services for those with triplets or service440 more 0 Cash Assistance for the Elderly service510 Orthopedic facilities for the elderly 49107 13.5% 21.3% 24.2% 23.2% 17.9% service520 Sanatoriums for the elderly 13534 7.2% 14.8% 20.7% 27.1% 30.2% service530 Cash assistance for house rent/firewood for the elderly 12391 30.9% 31.0% 22.9% 11.3% 3.9% 50% of transportation to the sanatorium for senior service540 honored donors 18002 7.0% 13.6% 19.2% 27.1% 33.2% service550 Burial expenses for the elderly 34 23.5% 23.5% 14.7% 32.4% 5.9% Transport for treatment for elderly living >1000 km service560 from the capital 3360 9.5% 22.6% 30.7% 23.4% 13.8% service570 (program name unknown) 10066 18.3% 24.8% 24.7% 21.6% 10.6% service573 Elderly without affiliation 2401 19.7% 26.0% 24.1% 19.5% 10.6% Cash Assistance to those with Merit REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA 79 80 First three Number of Proportion of Beneficiaries by Estimated Wealth Quintiles digits of Beneficiaries in the (PMT score) the Service SW Admin/PMT Q1 Q5 Codes Program database (Poorest) Q2 Q3 Q4 (Better off) Cash assistance to State and Labor Heroes and war service610 veterans and spouses 460 12.0% 18.7% 22.2% 24.8% 22.4% Cash assistance to elders with state honors and prize service620 winners 39 5.1% 5.1% 15.4% 15.4% 59.0% service630 Transport for elders with honors and heroes 185 1.1% 6.5% 16.8% 29.2% 46.5% Transport to local sanatorium for elders with honors service640 and heroes 309 0.6% 2.6% 11.0% 26.9% 58.9% service650 Housing or firewood for elders with honor and heroes 1735 4.6% 8.3% 14.0% 24.1% 49.0% Fixed benefits for State Hero, Labor Hero, People's service670 honors seniors 244 3.3% 8.2% 12.3% 24.2% 52.0% service671 Fixed benefits for State Honors Seniors from State 1071 1.4% 3.8% 10.0% 23.8% 61.0% Cash Assistance to those with Disability service710 Housing or firewood for the disabled 17243 46.8% 24.0% 15.0% 10.0% 4.2% service730 Orthopedic facilities for the disabled under 18 2633 30.4% 23.4% 20.5% 15.3% 10.3% service740 Orthopedic facilities for the disabled 6954 31.5% 25.1% 20.2% 15.4% 7.9% Orthopedic facilities and other facilities for children service750 under 18 and adults 6771 31.0% 23.4% 21.6% 16.1% 7.8% service760 Transport to school for disabled children 1373 42.2% 18.3% 17.1% 14.4% 7.9% service770 Transport to sanatorium for the disabled 2376 22.2% 19.6% 20.2% 21.5% 16.5% service791 Kindergarten food for the disabled 6 50.0% 33.3% 0.0% 16.7% 0.0% service792 Telephone for the disabled above 18 years of age 1437 42.7% 24.1% 17.7% 11.4% 4.0% Travel for treatment for those living >1000 km from the service793 capital 1579 29.8% 25.1% 21.8% 15.5% 7.8% Transportation for blind and disabled children and service794 adults for treatment 108 29.6% 26.9% 15.7% 16.7% 11.1% REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA service795 Summer camp for disabled children 20 15.0% 45.0% 15.0% 15.0% 10.0% service796 Burial for the disabled 79 53.2% 22.8% 15.2% 5.1% 3.8% Allowance for Mothers and Children service810 Pregnant and lactating mothers 62942 23.6% 20.0% 18.6% 19.3% 18.5% First three Number of Proportion of Beneficiaries by Estimated Wealth Quintiles digits of Beneficiaries in the (PMT score) the Service SW Admin/PMT Q1 Q5 Codes Program database (Poorest) Q2 Q3 Q4 (Better off) service820 Mother heroes I 64735 29.5% 26.3% 22.4% 15.1% 6.7% service821 Mother heroes II 111793 30.4% 22.3% 19.7% 16.7% 10.8% REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA 81 REVIEW OF PROGRAM DESIGN AND BENEFICIARY PROFILES OF SOCIAL WELFARE PROGRAMS IN MONGOLIA 82