66553 Towards a Social Protection Strategy for the Poor and Vulnerable Outcomes of the consultation process TABLE OF CONTENTS Boxes, Tables and Figures ..............................................................................................4 Acknowledgements........................................................................................................5 Acronyms ......................................................................................................................6 Executive Summary .....................................................................................................9 1. Introduction ........................................................................................................21 Background ..........................................................................................................22 Consultation process and fit with other strategies in the social sectors ..................26 2. Profile of Poverty and Vulnerability .....................................................................31 The dynamic dimension of poverty ......................................................................33 Risks, shocks and vulnerabilities ...........................................................................37 Vulnerable groups.................................................................................................49 3. Social Protection Schemes Targeting the Poor and Vulnerable in Cambodia ......63 Informal arrangements .........................................................................................66 Public support – RGC and development partners ................................................68 Public support – civil society ................................................................................82 Coordination gaps in pro-poor spending of social protection ...............................84 4. Policy Challenges .................................................................................................87 Coverage of main risks .........................................................................................88 Challenges of existing interventions......................................................................90 Institutional challenges .........................................................................................93 5. Conclusions: Towards a Social Protection Strategy for the Poor and Vulnerable.97 Options for the near future ................................................................................102 Elements to consider in program implementation ..............................................116 References .................................................................................................................120 Annex 1. Social Protection and Safety Nets: Concepts and Definitions .....................128 Annex 2. The Current Framework for Social Protection in Cambodia .......................135 Annex 3. Additional Tables........................................................................................138 Boxes, Tables and Figures Box 1: Social protection, safety nets and the Social Protection Floor Initiative .................. 24 Box 2: Poverty and vulnerability ....................................................................................... 37 Box 3: Vulnerability profile of food-insecure households .................................................. 53 Box 4: Selling buffalo to cope with illness ......................................................................... 57 Box 5: Household coping strategies in response to the 2009 global economic slowdown .. 61 Box 6: Poverty in the absence and presence of safety nets ................................................. 65 Box 7: Poor economic status makes it difficult to help one another .................................. 68 Box 8: UNICEF social protection programs ..................................................................... 80 Box 9: The Identification of Poor Households program .................................................... 81 Box 10: Governance and the decentralization process ..................................................... 100 Box 11: Impact of CCT programs .................................................................................. 108 Box 12: Design elements of a cash transfer program for Cambodia ................................ 110 Box 13: Outline of a public works program for Cambodia ............................................. 115 Table 1: Summary of the consultation process .................................................................. 26 Table 2: Government social protection interventions and sectoral policies/strategies......... 28 Table 3: Sources of vulnerability and vulnerable groups (summary) .................................. 32 Table 4: Transition in and out of poverty, 2001-2004 ....................................................... 38 Table 5: Estimated impact of extreme floods and droughts, 2000-2005 ............................ 40 Table 6: Malnutrition rates by wealth quintile .................................................................. 42 Table 7: Poverty levels of selected population groups ........................................................ 49 Table 8: Observed coping and mitigation strategies .......................................................... 57 Table 9: Worst forms of child labor .................................................................................. 59 Table 10: Current social protection against the social risk management framework .......... 64 Table 11: Social intervention chapters by ministries and other RGC agencies (Riel ‘000s) 70 Table 12: Coverage of social health protection for the population..................................... 74 Table 13: WFP food distribution and beneficiaries, 2009 ................................................. 77 Table 14: UNICEF safety nets budget, 2006-2010 (US$ ‘000s) ....................................... 78 Table 15: Coverage of main risks (summary) .................................................................... 88 Table 16: Challenges of existing interventions (summary) ................................................ 90 Table 17: Institutional challenges (summary) ................................................................... 93 Table 18: Options for the near future (summary) ........................................................... 102 Table 19: Existing strategic framework for social health protection ................................. 103 Table 20: Estimated cost of a cash transfer program........................................................ 111 Table 21: Elements of implementation (summary) ......................................................... 116 Figure 1: NSPS development and implementation ........................................................... 27 Figure 2: Poverty headcount, 1994-2007 (%) ................................................................... 34 Figure 3: Geographic variations in poverty rates ............................................................... 35 Figure 4: Youth out-of-school status and education .......................................................... 43 Figure 5: Distribution of workers by category of employment (%) ................................... 44 Figure 6: Food-insecure households (%) ........................................................................... 45 Figure 7: Per capita out-of-pocket expenditure 2004/2007 (%) ........................................ 46 Figure 8: Progress in achieving key CMDG targets ........................................................... 50 Figure 9: Trends in overall social spending (% of GDP).................................................... 69 Figure 10: ODA profile in 2009 – social sectors ............................................................... 82 Figure 11: Per capita ODA (2009) and poverty (2004) by province.................................. 85 4 Acronyms ADB Asian Development Bank ADI Analyzing Development Issues AFSC American Friends Service Committee AIDS Acquired Immune Deficiency Syndrome ART Anti-Retroviral Therapy AusAID Australian Agency for International Development BETT Basic Education and Teacher Training CAMINCO Cambodian National Insurance Company CARD Council for Agricultural and Rural Development CAS Cambodia Anthropometric Survey CBHI Community-Based Health Insurance CCLS Cambodian Child Labor Survey CCT Conditional Cash Transfer CDC Council for the Development of Cambodia CDCF Cambodia Development Cooperation Forum CDRI Cambodia Development Resource Institute CDWS Child Domestic Worker Survey CESSP Cambodia Education Sector Support Project CMDG Cambodia Millennium Development Goal CRC Cambodian Red Cross CRDB Cambodian Rehabilitation and Development Board CSES Cambodia Socio-Economic Survey CWCC Cambodian Women’s Crisis Center DAC Development Assistance Committee DFID UK Department for International Development DHS Demographic and Health Survey ECCD Early Childhood Care and Development EEQP Enhancing Education Quality Project FAO Food and Agriculture Organization FGD Focus Group Discussion FTI Fast Track Initiative GDP Gross Domestic Product GTZ German Technical Cooperation H.E. His/Her Excellency HEF Health Equity Fund HIV Human Immunodeficiency Virus IDPoor Identification of Poor Households ILO International Labor Organization IPEC International Programme on the Elimination of Child Labour (ILO) IT Information Technology IYCF Infant and Young Child Feeding JFPR Japan Fund for Poverty Reduction LBAT Labor-Based Appropriate Technology MAFF Ministry of Agriculture, Forestry and Fisheries MCH Mother and Child Health MEF Ministry of Economy and Finance 6 MIME Ministry of Industry, Mines and Energy MLVT Ministry of Labor and Vocational Training MoEYS Ministry of Education, Youth and Sports MoH Ministry of Health MoI Ministry of Interior MoP Ministry of Planning MoSVY Ministry of Social Affairs, Veterans and Youth Rehabilitation MoWA Ministry of Women’s Affairs MoWRAM Ministry of Water Resource and Meteorology MPWT Ministry of Public Works and Transportation MRD Ministry of Rural Development NAA National AIDS Authority NCDD National Committee for Sub-National Democratic Development NCDM National Committee for Disaster Management NCHADS National Center for HIV/AIDS, Dermatology and STD NCPD National Committee for Population and Development NGO Non-Governmental Organization NIPH National Institute of Public Health NIS National Institute of Statistics NPA-WFCL National Plan of Action on the Elimination of the Worst Forms of Child Labor NP-SNDD National Program on Sub-National Democratic Development NSDP National Strategic Development Plan NSPS National Social Protection Strategy NSSF National Social Security Fund OD Operational District OECD Organisation for Economic Co-operation and Development RACHA Reproductive and Child Health Alliance RGC Royal Government of Cambodia SHI Social Health Insurance SNAP-DDR Strategic National Action Plan for Disaster Risk Reduction SNEC Supreme National Economic Council SWAp Sector-wide Approach TB Tuberculosis TVET Technical and Vocational Education and Training TWG Technical Working Group TWG FSN TWG on Food Security and Nutrition UCW Understanding Children’s Work UK United Kingdom UN United Nations UNDAF UN Development Assistance Framework UNDESA UN Department for Economic and Social Affairs UNESCO UN Educational, Scientific and Cultural Organization UNICEF UN Children’s Fund UPE Universal Primary Education URC University Research Co. USAID United States Agency for International Development UXO Unexploded Ordnance WFP World Food Programme WHO World Health Organization 7 xaxAXa Executive Summary This Note provides background and technical information on social protection in Cambodia to inform the development of a national social protection strategy for the poor and vulnerable by the Council for Agricultural and Rural Development (CARD). This document is the outcome of over a year-long consultation process in which CARD, in collaboration with the Interim Working Group on Social Safety Nets under the Technical Working Group on Food Security and Nutrition (TWG FSN), convened several technical consultations and meetings and produced policy notes discussing selected aspects of a social protection strategy for the poor and vulnerable to build consensus on the meaning of key concepts and the broad direction for policy development. The Note reviews risks and shocks faced by the poor and vulnerable and matches them with coverage of existing programs to identify gaps and challenges in existing interventions. The risks and shocks analyzed can be grouped into: situations of emergency and crises, such as natural disasters and economic crises (with outcomes including reduction in purchasing power, increase in food insecurity, reduced earnings, rise in underemployment and unemployment, increase in insecure and risky employment and destruction/degradation of assets and resources); human development constraints (leading to poor maternal and child health and nutrition, high school dropout rates, low levels of skills attainment and increased incidence of hazardous and unconditional worst forms of child labor); seasonal unemployment and income insecurity (leading to poor health and nutrition and decreased capacity to study or work productively); and health shocks (leading to higher mortality rates and loss of assets and increased debts). In addition, there are special vulnerable groups (for example the elderly, people living with disabilities, people living with chronic illness, ethnic minorities) who are identified as warranting strengthened social protection and/or specific types of intervention. The Note then reviews existing programs that aim to provide social protection for the vulnerable population and identifies gaps with regard to coverage of risks, coordination, monitoring and evaluation, financing and targeting of interventions. In order to address outstanding gaps and challenges, this document provides options for the future that aim at expanding and harmonizing some existing interventions, as well as options suggesting additional programs to reach those most in need. Social health protection (e.g. health equity funds (HEFs), community-based health insurance (CBHI)), promotion of education and child development (e.g. scholarships for the poor, school feeding, take-home rations), fortified food transfer and public works (e.g. food for work and cash for work) are among the programs that can benefit from expansion, harmonization and improvements in their current design. The Note also proposes the development of a cash transfer program to alleviate chronic poverty and promote human development as a new option for the near future, along with programs that promote skills development for out-of-school youth. In addition to program options, proposals on how to better address elements of implementation such as coordination, financing and targeting are also included. Among others, capacity building within CARD in the near future is needed in order to achieve effective coordination, monitoring and evaluation of social protection interventions. How sustainable financing becomes in the future is reliant on the extent to which the Royal Government of Cambodia (RGC) gradually increases ownership and financial commitment to social protection programs. And the main targeting mechanism, which is increasingly being adopted, the Identification of Poor Households program (IDPoor), can benefit from guaranteed medium-term financing and further evaluation to enhance its accuracy. Complementary targeting instruments could also be considered in certain cases, especially to reach special vulnerable groups. 10 Towards a Social Protection Strategy for the Poor and Vulnerable | Executive Summary Poverty and vulnerability Despite substantial poverty reduction in the past decade, almost a third of the population remains in poverty. The poverty headcount for Cambodia fell from 47 percent in 1993/94 to 35 percent in 2004, to 30 percent in 2007. However, a third of the population (just over 4 million people) continued to live in poverty in 2007, with the overwhelming majority of the poor and vulnerable in rural areas. Moreover, the consecutive severe macroeconomic shocks – the food, fuel and financial crises – that have hit Cambodia since 2008 are likely to have further compromised the ability of the poor and vulnerable to cope with shocks. The 2004 regional poverty data for the main agro-climatic zones of the country show striking regional variations. The Tonle Sap and Plateau/Mountain zones experienced the highest poverty headcounts (43 percent and 52 percent, respectively, compared with a national average of 35 percent in 2004), whereas poverty magnitude (the number of poor people) was largest in the more densely populated Tonle Sap and Plains regions. The provinces with the highest poverty rates (proportion of the population in the province living below the poverty line) were Siem Reap, Kampong Thom and Kampong Speu. In addition to the poor, the near poor also remain vulnerable to household, community and nationwide shocks. Many households have a tenuous hold on well-being and are vulnerable to shocks that can push them into poverty. The 2006 Poverty Assessment calculated that Cambodia has a sizeable proportion of households (7 percent in 2004) living within a 10 percent band above the poverty line. The main sources of vulnerability in Cambodia that can push these households into poverty and worsen the living conditions of those already poor can be grouped into five categories: situations of emergency and crisis (i.e. economic crises and natural disasters); human development constraints; seasonal unemployment and income insecurity; health shocks; and other specific vulnerabilities affecting particular groups. With Cambodia’s integration into the global economy, external economic shocks are likely to become more frequent and also more severe. Economic shocks impact on poor and vulnerable households via loss of employment or lower wages/hours of work, increase in food insecurity and lower purchasing power. At times, they also change substantially the profile of the poor. The recent spike in food prices hit consumers while benefitting producers, and the recent global economic crisis hit key sectors such as garments, construction and tourism. The impact of these crises is complex, and comprehensive data on the impact on poverty trends are yet to come. Rapid assessments and surveys have reported constrained circumstances for many of the poor and vulnerable. Cambodia’s unique hydrological regime and low coverage of water control infrastructure makes it also vulnerable to climatic and natural disasters. Most rural households rely almost exclusively on subsistence agriculture for their livelihood, which makes food security heavily dependent on weather conditions. Extreme floods and droughts are among the most damaging shocks for rural households, and climate change will heighten their severity. Livelihoods and sources of income of the rural population can be compromised by natural disasters, leaving them reliant on social protection. The most severe human development constraints are identified as poor maternal and child health and nutrition, and poor access to quality education. Maternal mortality is persistently high (461 deaths per 100,000 live births) and, although under-five mortality rates have 11 decreased (83 deaths per 1,000 children), poor children remain three times more likely to die before the age of five than those who are well-off. Chronic and acute malnutrition also remain high. Both malnutrition and poor quality of education have repeatedly been shown to impact lifelong earnings and life opportunities and to act as a serious bottleneck for economic growth. Evidence from the most recent Cambodia Anthropometric Survey (CAS) shows that, following improvements between 2000 and 2005, chronic malnutrition rates have stagnated since 2005, with acute malnutrition even facing an increase between 2007 and 2008, possibly as a result of the food price crisis. In the case of education, the sector faces important challenges in terms of quality of service delivery. Poor basic school facilities, a shortage of textbooks and an inadequate supply of (trained) teachers have made it difficult to provide quality public education. This situation contributes to high dropout rates and high incidence of child labor, which compromises the skills of the future labor force. Seasonal unemployment and income insecurity are also a source of vulnerability for the poor and particularly for the about 350,000 food-insecure households with poor and borderline food consumption, equaling about 1.7 million individuals. According to the 2009 Global Hunger Index, Cambodia remains within “alarming� levels of hunger. Most food-insecure households live in rural areas, are landless (estimated at 15 percent of the rural population) or land poor (47 percent of the rural population) and have more children and more elderly to be nourished. The period from August to November, representing the “hunger gap,� remains particularly severe for poor households, as demand for agricultural labor is low and households’ rice stocks start running out. Poor rural households are predominantly dependent on their own limited food production and irregular, low-paid casual wage labor. Finally, health shocks were found to be among the major sources of vulnerability faced by the poor. Health shocks affect the poor disproportionately through higher prevalence of injuries and illnesses because of more physical jobs, foregone income by not being able to work when sick and unmet high health care costs. Moreover, access to quality health care is limited for poor individuals owing to factors such as excessive distance from health facilities, lack of qualified health staff and lack of drugs and equipment, especially in remote rural areas. Health shocks and the difficulty of coping with them have serious consequences for the poor, leading in particular to persistently high maternal mortality ratios and under-five mortality rates, as described above. 12 Towards a Social Protection Strategy for the Poor and Vulnerable | Executive Summary Poor infants and children are the most vulnerable group in society, because of their disproportionate representation among the poor and the severity of the vulnerabilities they face. Poverty levels, exposure and capacity to mitigate risks, long-term consequences of shocks for economic and social development and detrimental coping strategies (such as reducing food consumption, eating less nutritious food, selling assets and pulling children out of school and pushing them into child labor) are important criteria to prioritize vulnerable groups for social protection interventions. Under most of these dimensions, poor infants and children (aged 0-14) are the most vulnerable group. They represent 33.5 percent of the population and 37.5 percent of the poor, and the types of vulnerabilities they face (high mortality and malnutrition rates; poor quality of education) impact their lifelong earnings and life opportunities and contribute to the perpetuation of intergenerational cycles of poverty and destitution. Women of reproductive age, food-insecure households and groups such as the elderly or people living with disabilities and chronic illness were also identified as particularly vulnerable in the current context. Sources of vulnerability and vulnerable groups Main risks and shocks Determinants of vulnerability Outcomes Most vulnerable groups A risk is a source of danger; a The vulnerability of an individual or Depending on the While all poor possibility of incurring loss or household depends on their level of vulnerability of the individual and near poor misfortune. When a risk occurs, exposure and ability to cope with a and household, there can are vulnerable it becomes a shock. shock. People living under or near be a range of outcomes that to shocks, the poverty line tend to be more result from experiencing the some groups in vulnerable to negative outcomes of shock. the population shocks. are especially vulnerable to certain shocks. 1. Situations of Economic • Have limited income-generating • Rise in under- or • All poor and emergency and crises opportunities unemployment near poor crisis (price shocks, • Be food insecure • Increase in poorly economic • Be concentrated in insecure, remunerated, insecure and slowdown) unstable employment risky jobs • Reductions in the number of jobs • Lower remittances in the key sectors of the economy • Increase in food insecurity • Reductions in the purchasing power of salaries and earnings Climate, • Rely on crop farming and • Destruction or degradation • All poor and environment, livestock rearing for subsistence of assets and resources near poor natural food production and income • Increase in under or • People living disasters provision unemployment in flood and (floods, • Depend on (often degraded, • Increase in incidence and drought-prone droughts) over-exploited and contested) severity of food insecurity areas common natural resources for • Lower incomes livelihoods • Live in remote, isolated areas and suffer a low level of community infrastructure • Have low base of savings and assets to cover emergency needs 13 Main risks and shocks Determinants of vulnerability Outcomes Most vulnerable groups 2. Human Poor • Have low income and suffer • Maternal mortality • Girls and development maternal and from food insecurity and under- • Infant mortality women of constraints child health nutrition • Increase in incidence and reproductive and nutrition • Have poor access to quality severity of malnutrition, age maternal, new-born and child stunting and poor cognitive • Pregnant health care development women • Small children (0-5 years) Poor access • Come under pull factors to • Higher dropout rates and • School age to quality undertake domestic activities, low level of skills attained (6-14) education help with family business and/ • Increased incidence of or take up external employment, child labor (6-14) given households’ low income • Increase in under- and and food insecurity unemployment • Have poor access to quality • Increase in poorly education services remunerated, insecure and risky jobs Poor access • Come under pull factors to • Increase incidence of • Youth (15-24) to quality remain in paid employment, hazardous or unconditional second however precarious and lowly worst forms of child labor chance paid (15-17) programs • Have poor access to quality training services 3. Seasonal Under- and • Rely on subsistence farming with • Increase in food insecurity • All poor and unemployment poor nutrition low productivity • Maternal mortality near poor and income • Do not have sustained • Increase in incidence and • Families with insecurity employment to supplement severity of malnutrition, greater age incomes from agricultural stunting and poor cognitive dependency activities development ratio • Rely on (often degraded, • Increased likelihood of • Landless or land over-exploited and contested) ill-health poor common natural resources for • Decreased capacity to livelihoods study or work productively • Face a greater age dependency • Are more likely to be landless, or have less access to land and relatively smaller land holdings 4. Health Ill-health, • Have constrained access to clean • Maternal mortality • All poor and shocks injury, illness, water and sanitation • Infant mortality near poor death • Live with poor housing conditions • Increase in incidence and • Pregnant • Have low base of savings and severity of malnutrition, women and assets to cover out-of-pocket stunting and poor cognitive small children expenditures for health care development (0-5 years) • Have poor access to quality • Loss of assets and • Elderly preventive and treatment health increased debt • People with services disabilities • Work in physical jobs with greater • People living risk of accidents and injuries with chronic illness 5. Special Inability • Have limited access to income- • Increased income and • Elderly vulnerable to work, generating activities food insecurity • People living groups marginalazion • Suffer from marginalization in • Increased likelihood of with disability society, constrained access to becoming victims of • People living services and exclusion from violence, labor and sexual with chronic opportunities exploitation and abuse illness • Have extra nutritional and medical • Ethnic minorities needs • Orphans • Child laborers • Victims of violence, exploitation and abuse • Veterans • Families of migrants • Single mothers 14 Towards a Social Protection Strategy for the Poor and Vulnerable | Executive Summary Existing social protection schemes targeting the poor and vulnerable Existing schemes draw from informal arrangements, public support from the RGC and development partners and civil society and non-governmental organizations (CSOs and NGOs). All these play an important role by complementing one another. It remains clear, however, that even together they do not manage to adequately protect the most poor and vulnerable. A strong case remains for expanding social protection coverage for the poor. Mutual help still plays an important role in Cambodia, through kinship, family obligations and informal networks, although it appears to be eroding rapidly. When households face death of a family member, natural disasters such as big storms or lightning strikes, fires or theft and robbery, cultural, traditional and religious values oblige community members to provide assistance. Traditional forms of mutual help are rapidly eroding, however. Rural households report that, while some forms of inter-household assistance remain common (e.g. looking after another family’s children, sharing information on opportunities, pooling resources for funerals), others (providing small loans of cash or food at no interest) are increasingly extended only to close relatives or may be fully or partially commoditized. Those surveyed also report that pagoda associations are increasingly forced to target more narrowly, restricting their support only to the elderly without families. Several reasons underlie such rapid erosion. Rapid population growth in recent decades has increased livelihood competition and depleted natural assets such as fisheries and forests, as well as common lands that used to serve as the main sources of livelihoods. Recent privatization and poor management of natural resources have also had direct and indirect adverse effects on coping mechanisms as well as on community solidarity and collective actions to help the poor. In the case of public support, most of the RGC’s social protection spending targets public employees, with development partners covering most safety net expenditures for the poor and vulnerable. Safety nets are relatively developed – though far from universal – in education and health. In particular, the education sector has vast school feeding and take-home rations programs and is experimenting with quite effective scholarship programs for the poor. The health sector is expanding coverage of HEFs and CBHI. Apart from education and health interventions, safety net programs in Cambodia consist mainly of food distribution, food for work and some public works programs that have limited coverage. Given the limited resources available, there is also growing emphasis on better targeting services to the poor, in particular with the introduction of the IDPoor program. NGOs also play a significant role in assisting households in distress. In 2009, NGOs provided roughly 10 percent of total official development assistance (ODA) in Cambodia, with a total of $95 million spent on social sectors. Within the health sector, much of NGO assistance goes towards primary health care and access to hospitals and clinics. In education, NGOs focus on basic education for the poor and vocational training. NGOs are also very active in providing community and social welfare services through orphanages and general assistance to vulnerable children and youth. Despite covering different areas and sources of vulnerability, many social protection interventions face similar types of challenges. Overall, challenges related to existing social protection provision were identified in terms of inadequate coverage of main risks and vulnerabilities; limited geographic coverage that does not necessarily reach the poorest regions and groups; limited implementation capacity; poor targeting; absence of proper monitoring 15 and evaluation; and institutional and policy challenges such as a preference for immediate responses to crises rather than a longer-term vision for social protection development and the absence of an agency with the mandate and capacity to coordinate and develop social protection interventions. Options for the near future and elements of implementation The key question is how to target better the very limited resources that are available to ensure coverage of main sources of vulnerability and of most vulnerable groups. The RGC and its partners face tight capacity and budget constraints across all sectors. Any spending on social protection interventions has high opportunity costs, as resources are limited and spending for social protection means fewer resources for another sector. Coherence rather than competition between sectors needs to be strengthened, as investments in social protection without adequate investments in basic social services, including health, education, water and housing, would remain ineffective. Better targeting plus coherence between current sector investments in social protection, and a gap analysis-driven identification of priorities for scaling-up across sectors, will ensure maximum coverage and provision for the poor and vulnerable. The matching of main sources of vulnerability with programs that are being implemented suggests the need to scale up and harmonize existing interventions, but also the presence of gaps that could be addressed by the implementation of a few new programs. HEFs, school feeding, scholarships and public works are already addressing most of the major sources of vulnerability faced by the poor, and already have to various extents a strong presence in Cambodia. However, coordination and expansion are needed in the context of sustainable financing of programs. The Table below also highlights potential returns to implementing a few new programs, such as a cash transfer program that could help relieve chronic poverty while at the same time promoting human capital and training programs that would promote skills development for out-of-school youth. Options for the near future Main risks and Progress to date in Gaps and challenges in Options for the near future shocks response response 1. Situations of • Public works have been • Limited coverage and • Harmonize public works approaches emergency and shown to be an effective coordination of existing and guarantee stable financing crisis and rapidly expandable public works programs • Establish unit in RGC in charge of safety net instrument public works for rural development and during crises and natural emergency situations disasters 2. Human • Some maternal and child • Supply of maternal and • Improve and expand nutrition services, development nutrition programs are in child nutrition services including fortified complementary constraints place remains limited and of feeding • Breastfeeding practices poor quality • Develop cash transfer program are improving • Coverage of these targeting poor families with children services is not universal • Design cash transfer programs in • Supply of fortified health and education such that complementary food they can eventually be harmonized/ is low coordinated/merged • Other demand-side factors (eating, feeding and care practices) are not being adequately addressed 16 Towards a Social Protection Strategy for the Poor and Vulnerable | Executive Summary Main risks and Progress to date in Gaps and challenges in Options for the near future shocks response response • Scholarships (both cash • Quality of education • Improve quality and access to and take-home rations) remains poor education and school feeding • Coverage of education • Expand programs addressing demand programs are improving services is variable side (in particular scholarships) both attendance • Coverage of in terms of coverage and covering all scholarships and school years of basic education feeding programs does • Improve coordination of education and not reach all poor areas child labor programs • Establishment of • Quality of training • Boost second chance programs vocational training remains poor • Improve quality of vocational training curricula • Supply of second programs by linking training to • Some programs in place chance program is employers’ needs for second chance minimal • Develop certification/ accreditation education • Poor link between system to regulate quality of training training offered and provided employers’ needs • No certification/ accreditation system for private sector 3. Seasonal • Some targeted food • Limited coverage and • Harmonize public works approaches unemployment distribution coordination of existing and guarantee stable financing and income • School feeding public works programs • Establish unit in RGC in charge of insecurity • Public works programs • Funding and assistance public works for rural development and are providing some remains volatile emergency situations assistance during lean • Facilitate access to public works season or crises program for special vulnerable groups and introduce design features that respond to their needs 4. Health shocks • HEFs are financing health • Quality of health care • Improve and expand social health care for the poor in some poor protection for the poor and vulnerable areas • Coverage/access of (HEFs and CBHI) HEFs is not universal 5. Special • Pensions for civil • No pensions for the poor • Identify and pilot social protection vulnerable groups servants, NSSF for • Very limited assistance programs for the disabled and elderly private sector to the disabled poor and other special vulnerable • Some donor assistance • Limited assistance to groups to people with disabilities other special vulnerable • Extend targeted cash transfer program • Some donor assistance groups to the elderly and the disabled to tuberculosis patients • Benefits and services • Expand food assistance to special and people living with for special vulnerable vulnerable groups, including HIV/AIDS groups are not complementary feed, provision of integrated fortified food • Mainstream social protection programs are not sensitive to the needs of special vulnerable groups The following options were identified during the consultation process as possibilities for priority programs that respond to the needs of the poor and vulnerable: 1. Improving and expanding social health protection for the poor and vulnerable. The RGC has already positioned health care for the poor and vulnerable as a priority for improvement under the National Strategic Development Plan (NSDP) Update 2009- 2013, the Health Strategic Plan 2008-2015 and the Social Health Protection Master Plan (in the process of adoption). Key instruments for expansion are HEFs and CBHI, while a strategic and coordination framework is needed to support the arrangements being developed. Preventing malnutrition needs to be prioritized during the design of instruments related to the health sector. 17 2. Harmonizing and expanding coverage of programs promoting education and child development. High dropout rates and high incidence of child labor are strong outcomes of vulnerability affecting children’s human capital and the likelihood of future generations moving out of poverty. Programs intended to fight child labor and keep children in school (in-kind transfers including school feeding and take-home rations, cash scholarships) should be not only expanded but also harmonized. While addressing demand constraints that prevent parents from sending their children to school, it is equally (if not more) important also to improve the quality of education. 3. Piloting a cash transfer program to alleviate chronic poverty and promote human development. Cash transfers help poor households access services and can provide incentives to change behavioral practices. Given the importance of behavioral practices in determining nutrition outcomes, a cash transfer program could in particular be used to address demand constraints in fighting chronic malnutrition, the main source of vulnerability of infants and small children. Such a program would fulfill the dual objective of alleviating chronic poverty while combating maternal mortality and malnutrition. 4. Improving the effectiveness of public works to respond to food insecurity, crises and natural disasters. Public works are particularly suited to be rapidly expanded in specific areas hit by crises or natural disasters to provide emergency assistance. Moreover, with a third of the population falling below the poverty line, public works can also be used to address chronic poverty and food insecurity, in particular during the lean season. Currently, however, approaches vary significantly across projects. It is important therefore to try harmonizing interventions and unifying approaches, so as to enable transition from a “project� approach that is financed (and often implemented) by donors, to a programmatic approach that is increasingly owned by the RGC. In moving forward, there are also several aspects related to program implementation that should be considered by coordinating and implementing agencies. There are, at the moment, gaps in coordination, monitoring, evaluation and information management that could be addressed by building the capacity of CARD to perform these tasks. Fiscal constraints should also be taken into account when choosing and designing social protection programs. For this reason, targeting and program evaluation should be performed in the best possible way, as to guarantee that financial resources are spent where they are most needed. Specifically, the following elements should be considered: 1. Build CARD’s capacity to effectively coordinate, monitor and evaluate implementation of social protection programs for the poor and vulnerable. Social protection for the poor and vulnerable consists by nature of inter-sectoral programs involving several ministries that need to be coordinated. At the moment, CARD has been designated to perform these tasks but lacks capacity. Therefore, there is a need to build capacity in CARD to perform coordination, monitoring and evaluation of social protection development activities. 2. Build towards sustainable financing through an appropriate approach to investment. A sustainable social protection program for the poor and vulnerable requires predictable and guaranteed financing. While development partner financing will continue to make an important contribution in the coming years, the work plan for operationalizing the 18 Towards a Social Protection Strategy for the Poor and Vulnerable | Executive Summary social protection strategy for the poor and vulnerable should identify the steps towards a program-based approach. In order to achieve this, sustainable financing would also require increasing RGC financial commitment over time. 3. Evaluate and improve the design and financing modality of current targeting systems; update poverty and vulnerability maps to improve coverage of the poorest regions and households. As the IDPoor program is increasingly adopted, there is a need for further evaluation and continuous strengthening of the program. Moreover, there is also a need to guarantee its sustainable financing, for instance by ensuring that programs using it would also contribute to its financing. Post-identification mechanisms and geographic targeting through updated poverty and vulnerability maps are other important targeting efforts that should be considered. 19 xaxAXa 1. Introduction Background Cambodia has emerged from three decades of instability with an impressive record of sustained growth and poverty reduction. Yet Cambodians still face many serious forms of vulnerability. With the majority of the population engaged in single crop subsistence agriculture heavily dependent on the weather, poor harvests and food insecurity (in particular during the lean season) as well as natural disasters are major sources of vulnerability. Illness and injury, high costs of treatment and large out-of-pocket expenditures on health are also critical factors in pushing households into poverty or preventing them from breaking the poverty cycle. For many of Cambodia’s young and generally undereducated population now entering the workforce, unemployment or underemployment is an equally endemic risk. Moreover, the recent crisis has shown that even those who have obtained jobs in export- oriented sectors of the economy (such as garment manufacturing or tourism) remain vulnerable to job loss or drastically reduced earnings. Premature entry into the workforce is also an issue both caused by and contributing to household vulnerability and poverty. These forms of vulnerability, which push people further into poverty or keep them in it, have serious long-term consequences for human and national development. Poverty and periodic shocks result in child malnutrition and high child and maternal mortality rates; sale of land or other household productive assets at low prices to raise cash for food or medical treatment; debt; and taking children out of school so that they can work or help out at home. These detrimental coping strategies adopted by households in distress push them further into vicious cycles of poverty and destitution and prevent them from utilizing their full economic and human potential, with a negative impact on national development. Informal arrangements based on kinship and community practices appear to be gradually eroding and cannot substitute for a well-functioning public social protection system. While community help and transfers from richer to poorer households may play a valuable role in helping households deal with occasional household-level (idiosyncratic) crises, they have limitations. They appear to be gradually eroding over time and to fail in the face of widespread shocks such as extensive floods or rapid economy-wide increases in food prices. A sound and sustainable social protection system that supports the poor and vulnerable in coping with major sources of vulnerability, at the same time promoting human development, should be developed. Coverage of existing social protection programs for the poor and vulnerable is very limited, and important sources of vulnerability (such as health shocks and lack of access to affordable and quality social services such as health and education), which remain poorly addressed, lead poor households further into destitution. In the near future, therefore, priority should be given to the development of effective and affordable social protection programs for the poor and vulnerable which support them in coping with major sources of vulnerability and in reducing exposure to risk while promoting human development. The mutually complementary and reinforcing nature of social protection and basic social service investments needs to be acknowledged and capitalized on. In addition to reducing vulnerabilities, investing in social protection can have a direct impact on economic growth. Direct transfers to the poor allow them to increase their consumption and encourage prudent risk taking. The overall economy can benefit from increasing per capita consumption and more dynamic markets, as poor households can have more available income to spend in their communities. In terms of risk taking, social 22 Towards a Social Protection Strategy for the Poor and Vulnerable | Introduction protection programs that are reliably delivered and transparently operated provide a form of insurance that can encourage households to adopt innovations (Alderman and Hoddinott, 2007). While alleviating immediate needs, in-kind or cash transfers enable households to use their income in acquiring assets, investing in children’s human capital and/or entering profitable activities, all of which have a potential positive impact on economic growth. An effective social protection system can also promote equitable growth and the ability of the Royal Government of Cambodia (RGC) to reform. A well-implemented social protection program ensures that social investment is reaching those who need it, which can decrease social tensions and promote investment at all levels of society. Social protection can also help Government embark on reforms that have long-term benefits in economic efficiency but high short-term social and political costs, by providing effective compensation to those negatively affected by the reform. Objective of the Note This Note provides an overview of vulnerabilities, existing safety nets and gaps. It then identifies policy options and discusses possible next steps for the implementation of a comprehensive, effective and affordable social protection system for the poor and vulnerable. Because of its focus on the poor and vulnerable, this Note will focus mostly on the review and assessment of programs that target the poorest strata of the population (i.e. safety nets – see Box 1), rather than looking at the whole social protection panorama. It is, however, important to contextualize safety nets within the larger social protection framework, comprising contributory programs for formal sector employees and civil servants as well as for the relatively better-off segments of the informal economy, which is in turn part of a broader poverty reduction strategy. As such, safety nets complement policies and programs for social insurance; health, education and financial services, including microfinance, the provision of utilities and local infrastructure and other policies aimed at reducing poverty and managing overall risk and vulnerability. 23 Box 1: Social protection, safety nets and the Social Protection Floor Initiative Social protection, safety nets and the Social Protection Floor Initiative are concepts at the heart of this strategy. It is therefore important to understand the similarities and differences between them. Social protection consists of a broad set of arrangements and instruments designed to assist individuals, households and communities to better manage risk, shocks and extreme poverty. This includes social insurance, labor market policies, social funds, social services, social welfare services and safety net (social assistance) programs. Social security is closely related to the concept of social protection and can be defined as the protection that a society provides to individuals and households to ensure access to health care and to guarantee income security, particularly in cases of old age, unemployment, sickness, invalidity, work injury, maternity or loss of a breadwinner. Social insurance is designed to help individuals, especially workers in the formal sector, insure themselves against sudden reductions in income. Social insurance programs include publicly provided or mandated insurance against unemployment, old age, disability, death of the main provider, maternity and sickness. Social insurance programs are contributory: beneficiaries receive benefits or services in recognition of contributions to an insurance scheme. Safety nets are non-contributory transfer programs (in cash, in kind, waivers) targeted in some manner to the poor or those vulnerable to poverty and shocks, including informal workers. These programs are also referred to as social assistance or social welfare programs. The main objectives of safety nets are to: (i) relieve chronic poverty; (ii) help reduce the impact of major shocks; and (iii) promote human capital (e.g. through access to health, education and social welfare services). There is no universal consensus on the types of interventions covered under the safety net label, although they generally include cash transfers (conditional and unconditional), in-kind or resource transfers (i.e. productive inputs), food distribution and nutrition programs, general commodity price subsidies, public works and fee waivers (e.g. health equity funds (HEFs)). Selected programs are reviewed in more detail in Annex 1. Social protection, safety nets and development policies Such as land Includes labor redistribution, policy, enforcement of contributory contracts & social property rights, insurance, and universal social care education, EQUITY SOCIAL PROTECTION services as well safety nets as safety nets SAFETY NETS Foment pro-poor growth while providing For example, for POVERTY services that small farmers, SOCIAL RISK MANAGEMENT REDUCTION facilitate the includes poor’s irrigation, participation in microfinance, growth; safety weather nets are part of insurance, safety providing nets security Source: Grosh et al. (2008). 24 Towards a Social Protection Strategy for the Poor and Vulnerable | Introduction The objective of this Note is also consistent with the objectives of the Social Protection Floor Initiative supported by the Food and Agriculture Organization (FAO), the International Labour Organization (ILO), the United Nations Children’s Fund (UNICEF), the UN Educational, Scientific and Cultural Organization (UNESCO), the UN Population Fund (UNFPA), the World Bank, the World Food Programme (WFP), the World Health Organization (WHO) and many other UN agencies, as well as international non-governmental organizations (NGOS) such as HelpAge, Save the Children, etc. The Initiative consists of promoting a social protection floor as a core element of poverty reduction policies and pursuing wider development policies that enable countries to grow with equity. The Social Protection Floor Initiative corresponds to a set of basic social rights, services and facilities that the global citizen should enjoy. It can be seen as a core obligation in ensuring the realization of minimum essential levels of rights embodied in human right treaties. The Initiative could consist of two main elements that help realize these human rights: 1. Ensuring the availability, continuity and geographical and financial access to essential services, such as water and sanitation, food and adequate nutrition, health, education, housing and other social services, such as life- and asset-saving information. 2. Realizing access by ensuring a basic set of essential social transfers, in cash and in kind, to provide a minimum income and livelihood security for poor and vulnerable populations and to facilitate access to essential services. This includes social transfers (but also information, entitlements and policies) to children, people in active age groups with insufficient income and older persons. These transfers could include: universal access to essential health services; income (or subsistence) security for all children through child benefits; income support combined with employment guarantees through public works programs for the working-age poor who cannot earn sufficient income in the labor market; or income security through basic tax-financed pensions for the old, the disabled and those who have lost the main breadwinner in a family. The Social Protection Floor Initiative promotes a holistic and coherent vision of national social protection systems as a key component of national development strategies. It seeks to support countries in identifying and closing crucial protection gaps through coherent and efficient measures that maximize the effects of scarce resources on the reduction of poverty and insecurity, to ensure “guaranteed access� to essential services and social transfers. Source: Cichon (2008); Grosh et al. (2008); ILO and WHO (2009). This Note reflects the outcome of a year-long consultation process intended to support the establishment of a social protection strategy for the poor and vulnerable. At the Cambodia Development Cooperation Forum (CDCF) in December 2008, the RGC and its development partners discussed the heightened need for a coherent social protection system in the face of the global food, fuel and financial crises and the effects of these shocks on Cambodia. As an outcome of this discussion, the RGC assigned to the Council for Agricultural and Rural Development (CARD) the responsibility for developing the National Social Protection Strategy (NSPS) for the Poor and Vulnerable. Over 2009 and 2010, CARD, in collaboration with the Interim Working Group on Social Safety Nets under the Technical Working Group on Food Security and Nutrition (TWG FSN), convened several workshops and meetings and produced policy notes discussing selected aspects of social protection to build consensus on the meaning of key concepts and the broad direction for policy development. This Note, and in particular the RGC NSPS that is informed by it, should be seen as a living document that will evolve with time and changing socioeconomic conditions. While some fairly general “dos and don’ts� should be adhered to in developing social protection programs, there is not a single best practice social protection system. To be successful, 25 social protection programs should be tailored to local conditions, historical backgrounds and cultural values. Moreover, the type and sophistication of programs should evolve with changing socioeconomic conditions. While nutrition may be a priority in a low-income country, for instance, it is usually less of a problem in wealthier countries. At the same time, other priorities continue to emerge, such as old-age pensions, health financing and keeping children in school up to secondary education. Any social protection strategy should be updated on a regular basis to reflect progress made and the changing socioeconomic environment. The Note is structured as follows. Section 2 presents a profile of the poor and vulnerable in Cambodia, identifying main risks, shocks and sources of vulnerability, vulnerable groups and coping strategies. Section 3 reviews safety nets in Cambodia, distinguishing among informal arrangements and mutual support, public support and support from civil society, and identifying the gaps. Section 4 highlights policy challenges faced by public safety net programs targeting the poor in Cambodia. Section 5 concludes by outlining priorities for the near future in developing social protection programs for the poor and vulnerable. Consultation process and fit with other strategies in the social sectors To prepare the NSPS, in 2009 and 2010 CARD has convened meetings and held technical consultations with a broad set of national stakeholders, giving Government representatives (national and sub-national), development partners, civil society representatives and other development practitioners the opportunity to explore in-depth the priorities and options. This transparent and rigorous consultation process has ensured that the analytical and policy inputs for the NSPS have gone through several rounds of discussion and are the result of a combined effort of stakeholders. Table 1: Summary of the consultation process Date Activity/event Description Papers Dec 4-5 CDCF RGC commitment to develop and implement an integrated 2008 national strategy for social safety nets. Jan-Jun Interim Working Group Shared knowledge and consensus building on the key 2009 on Social Safety Nets concepts and broad direction for policy development and (under the TWG FSN) inventory of ongoing social protection interventions. Jul 6-7 National Forum on Food Circa 400 participants (RGC, development partners, civil Social Safety 2009 Security and Nutrition society), with Samdech Techo Hun Sen, Prime Minister, Nets: Concept under the Theme of providing the closing address. Note and Social Safety Nets in Inventory Cambodia Oct 19-22 Technical Consultation 70+ participants (RGC, development partners and civil Note on Cash 2009 on Cash Transfers with society) consulted during a 1-day workshop in Phnom Transfers a focus on addressing Penh. The core group (20 participants) also visited health malnutrition and educational social protection activities and services, a Commune Council and the Provincial Office in Kampong Speu. The consultation culminated in brainstorming by key stakeholders to produce a Note on Cash Transfers. 26 Towards a Social Protection Strategy for the Poor and Vulnerable | Introduction Date Activity/event Description Papers Jan 12-14 Technical Consultation 80+ participants (RGC, development partners and Note on 2010 on Public Works civil society) consulted during a 1-day workshop in Public Works Phnom Penh. The core group (circa 30 participants) also visited sites of cash for work and food for work projects (Asian Development Bank- (ADB-) and WFP- supported interventions) in Kampong Chhnang, including a consultation with representatives of a Commune Council and beneficiaries of the projects. The consultation culminated in a next steps meeting by CARD and a core group of development partners and the production of a Note on Public Works. Feb 3-4 Technical Consultation 100+ participants (RGC, development partners and civil Note on 2010 on Education and society) consulted during a 2-day workshop in Phnom Tackling Child Child Labor Penh. The consultation built consensus around the Labor and integration of education and child labor issues in a social Increasing protection strategy and the need to explore greater Educational access to safety net schemes for those at risk of child Access labor in particular, and explored the role that (conditional) cash transfer schemes could play in ending child labor. A Note on Tackling Child Labor and Increasing Educational Access was prepared as a contribution to the NSPS. Feb –May Consultations on draft An executive drafting team was set up to prepare Draft NSPS 2010 NSPS and consolidate inputs into the draft NSPS. Several consecutive drafts of the NSPS were shared and discussed in the extended format of the Interim Working Group on Social Safety Nets. This Note and the related NSPS represent important milestones of the process – but are, at the same time, just an intermediate step. The Note summarizes discussions, analyses and achievements of the consultation process that have been at the core of the development of the RGC’s NSPS for the Poor and Vulnerable. It provides a basis for informing key policy decisions as well as inputs into the future NSPS and its subsequent in-depth analyses, scenarios and ways forward (Figure 1). Figure 1: NSPS development and implementation Safety Nets Technical consultations: in Cambodia: Note on Cash Transfers with a focus on addressing nutrition Concept Note Note on Public Works and Inventory Note on Tackling Child Labor and Increasing Educational Access Background Note on a Social Protection Strategy Existing and new support ADB, AusAID, RGC National Social Protection Strategy for the Poor and Vulnerable FAO, GTZ, ILO, UNICEF, WFP, WHO, World Bank More in-depth analyses, Detailed program design preparation and others scenarios, costings, etc Government Program A (supported by Donor X and Y) Government Program B, C etc (supported by U, V, etc) Note: AusAID = Australian Agency for International Development; GTZ = German Technical Cooperation. 27 This Note and the related NSPS complement existing efforts and developments in the social sectors to promote social protection in Cambodia. In its broadest characterization, social protection includes social security – focused mainly on contributory and insurance- based schemes – and safety nets (or social assistance), with the respective instruments of non-contributory transfers and accompanying services, education scholarship schemes and health protection, labor market interventions, social welfare and other such programs for the poor and vulnerable population. By putting attention on safety nets that target the poor, the Note and the related NSPS cover a subset of social protection. They should be seen as complementing other efforts towards social protection development, i.e. as one of the components of the RGC’s current social protection interventions (see Table 2). It will be important to maintain coordination between the various initiatives developing social protection and to discuss financing priorities. One of the priorities highlighted in this Note and in the related NSPS consists of achieving coordination between programs and initiatives, both to avoid gaps and overlaps in coverage and to guarantee that RGC and development partner spending reflects major sources of vulnerability and identified priorities. Traditionally, the RGC has directed most of its spending towards financing social protection programs for the non-poor (in particular veterans and civil servants); development partners implement and finance most safety nets programs. To guarantee ownership and sustainability of safety nets, it will be important for the RGC in future to gradually take an active role in safety net provision, in terms of both financing and implementation. The social protection reform agenda is also being promoted within the context of the Cambodia Millennium Development Goals (CMDGs), achieved through the National Strategic Development Plan (NSDP), where social protection is recognized as a critical policy area. The RGC’s NSDP Update 2009-2013 places significant emphasis on expanding social protection to the poor and vulnerable and proposes a vision for “developing and implementing a comprehensive and sustainable social safety net system aimed at protecting the livelihoods of the poor and most vulnerable segments of the population’’ (RGC, 2009). A number of development partners have aligned their programs of assistance with the priorities of the NSDP. For example, the recently endorsed UN Development Assistance Framework (UNDAF) includes support to strengthening social protection as one of five priorities forming the core of the UN’s support to Cambodia in 2011-2015. Table 2: Current Government social protection interventions and related sectoral policies/strategies Institutions Dimension(s) of social protection and Current sectoral policy/strategy social safety nets Crosscutting national programs Ministry of Interior • Identification of entry points for ensuring • National Program on Sub-National Democratic (MoI) quality and equitable provision of social Development (NP-SNDD) and the 3-Year protection at sub-national levels Implementation Plan (2011-2013) Ministry of Planning • Identification of Poor Households (IDPoor) • MoP Strategic Plan (MoP) program CARD • Food security and nutrition • National Strategic Framework on Food Security and Nutrition 2008-2012 RGC institutions that are mandated to provide emergency response Ministry of Economy • Coordination of emergency assistance and Finance (MEF) 28 Towards a Social Protection Strategy for the Poor and Vulnerable | Introduction Institutions Dimension(s) of social protection and Current sectoral policy/strategy social safety nets National Committee • Disaster response and disaster • Strategic National Action Plan for Disaster Risk for Disaster preparedness Reduction in Cambodia 2008-2013 (SNAP- Management (NCDM) DRR) (launched March 2009) RGC institutions that are mandated to deliver social services to the population and to protect specific vulnerable groups against risks Ministry of Labor and • National Social Security Fund (NSSF) for • The Employment Injury Scheme is established Vocational Training private sector employees and will be extended in 2010; the Social (MLVT) • Vocational training Health Insurance Scheme will be designed in • Creating decent work opportunities for 2010 vulnerable groups and taking proactive • Vocational training program for laid-off steps to reach the Twin Goals on child workers implemented with resources from the labor: to reduce all forms of child labor to 8 Samdech Hun Sen Foundation percent by 2015 and to eliminate the worst • National Plan of Action on the Elimination of forms of child labor by 2016 the Worst Forms of Child Labor (NPA-WFCL) 2008-2012 Ministry of Social • National Social Security Fund for civil • Work Platform 2009-2013 Affairs, Veterans and servants • A sub-decree is under preparation for the Youth Rehabilitation • Services and pensions for veterans Social Health Insurance Scheme (MoSVY) • Services for homeless and destitute, victims of trafficking, children and youths, people living with disabilities • Emergency relief to those affected by natural disasters • Social security law to cover employees of small enterprises (fewer than 8 employees), the self-employed, the informal economy and the poor (under development) Ministry of Women’s • Early childhood care and development • ECCD Policy adopted February 2010 Affairs (MoWA) (ECCD) centers RGC institutions that implement specific safety net interventions Ministry of Health • HEFs • Health Strategic Plan 2008-2015 (MoH) • Community-based health insurance (CBHI) • Strategic Framework for Health Financing for the poor and vulnerable 2008-2015 • Reduce micronutrient malnutrition in • Master Plan on Social Health Protection (in children and women process of approval) • National Nutrition Strategic Plan 2009-2015 Ministry of Education, • Scholarship for the Poor program • Education Sector Strategic Plan 2006-2010 Youth and Sports (MoEYS) RGC institutions with complementary activities Ministry of • Food production, livelihoods Strategy for Agriculture and Water 2006-2010 Agriculture, Forestry and Fisheries (MAFF) Ministry of Public • Implementation of national policy Works and concerning all public works construction Transportation (MPWT) Ministry of Rural • Rural infrastructure works • National Strategy for Rural Water Supply, Development (MRD) Sanitation and Hygiene (forthcoming 2010) • Rural Roads Policy (forthcoming 2010) Ministry of Water • Rural infrastructure works (irrigation) • Strategy for Agriculture and Water 2006-2010 Resource and Meteorology (MoWRAM) 29 2. Profile of Poverty and Vulnerability Table 3: Sources of vulnerability and vulnerable groups (summary) Main risks and shocks Determinants of vulnerability Outcomes Most vulnerable groups A risk is a source of danger; a The vulnerability of an individual Depending on While all poor and possibility of incurring loss or or household depends on their the vulnerability of near poor are misfortune. When a risk occurs, it level of exposure and ability to the individual and vulnerable to shocks, becomes a shock. cope with a shock. People living household, there can be some groups in under or near the poverty line a range of outcomes that the population are tend to be more vulnerable to result from experiencing especially vulnerable negative outcomes of shocks. the shock. to certain shocks. 1. Situations of Economic crises • Have limited income- • Rise in under- or • All poor and near emergency and (price shocks, generating opportunities unemployment poor crisis economic • Food insecure • Increase in poorly slowdown) • Concentrated in insecure, remunerated, insecure unstable employment and risky jobs • Reductions in the number of • Lower remittances jobs in the key sectors of the • Increase in food economy insecurity • Reductions in the purchasing power of salaries and earnings Climate, • Rely on crop farming • Destruction or • All poor and near environment, and livestock rearing for degradation of assets poor natural disasters subsistence food production and resources • People living in (floods, droughts) and income provision • Increase in under or flood and drought- • Depend on (often degraded, unemployment prone areas over-exploited and contested) • Increase in incidence common natural resources for and severity of food livelihoods insecurity • Live in remote, isolated areas • Lower incomes and suffer a low level of community infrastructure • Have low base of savings and assets to cover emergency needs 2. Human Poor maternal • Have low income and suffer • Maternal mortality • Girls and women development and child health from food insecurity and • Infant mortality of reproductive constraints and nutrition under-nutrition • Increase in incidence age • Have poor access to quality and severity of • Pregnant women maternal, new-born and child malnutrition, stunting • Small children (0-5 health care and poor cognitive years) development Poor access to • Come under pull factors • Higher dropout rates • School age (6-14) quality education to undertake domestic and low level of skills activities, help with family attained business and/or take up • Increased incidence of external employment, given child labor (6-14) households’ low income and • Increase in under- and food insecurity unemployment • Have poor access to quality • Increase in poorly education services remunerated, insecure and risky jobs • Increase incidence of hazardous or Poor access to • Come under pull factors to unconditional worst • Youth (15-24) quality second remain in paid employment forms of child labor chance programs however precarious and lowly (15-17) paid • Have poor access to quality training services 32 Towards a Social Protection Strategy for the Poor and Vulnerable | Profile of Poverty and Vulnerability Main risks and shocks Determinants of vulnerability Outcomes Most vulnerable groups 3. Seasonal Under- and poor • Rely on subsistence farming • Increase in food • All poor and near unemployment nutrition with low productivity insecurity poor and income • Do not have sustained • Maternal mortality • Families with insecurity employment to supplement • Increase in incidence greater age incomes from agricultural and severity of dependency ratio activities malnutrition, stunting • Landless or land • Rely on (often degraded, and poor cognitive poor over-exploited and contested) development common natural resources for • Increased likelihood of livelihoods ill-health • Face a greater age • Decreased capacity dependency to study or work • Are more likely to be landless, productively or have less access to land and relatively smaller land holdings 4. Health shocks Ill-health, injury, • Have constrained access to • Maternal mortality • All poor and near illness, death clean water and sanitation • Infant mortality poor • Live with poor housing • Increase in incidence • Pregnant women conditions and severity of and small children • Have low base of savings and malnutrition, stunting (0-5 years) assets to cover out-of-pocket and poor cognitive • Elderly expenditures for health care development • People with • Have poor access to quality • Loss of assets and disabilities preventive and treatment increased debt • People living with health services chronic illness • Work in physical jobs with greater risk of accidents and injuries 5. Special Inability to work, • Have limited access to • Increased income and • Elderly vulnerable marginalization income-generating activities food insecurity • People living with groups • Suffer from marginalization in • Increased likelihood disability society, constrained access to of becoming victims • People living with services and exclusion from of violence, labor and chronic illness opportunities sexual exploitation and • Ethnic minorities • Have extra nutritional and abuse • Orphans medical needs • Child laborers • Victims of violence, exploitation and abuse • Veterans • Families of migrants • Single mothers The dynamic dimension of poverty Cambodia, still dealing with the legacy of conflict, has enjoyed a decade of sustained growth, resulting in a reduction in the proportion of the population living in poverty from 35 percent in 2004 to 30 percent in 2007 – 1.2 percent per year (Figure 2). Cambodia has gone through a period of rehabilitation and reconstruction to rebuild a society that largely collapsed during the Khmer Rouge period (1975-1979), when an estimated 2 million people died and physical and social infrastructure was destroyed. Having established a track record of political and macroeconomic stability, in the past decade there has been a profound transformation in the Cambodian economy, with gross domestic product (GDP) growth above 10 percent for four consecutive years ending 2007. With a substantial rise in real per 33 capita household consumption, the poverty headcount for Cambodia fell from 47 percent in 1993/94 to 35 percent in 2004, to 30 percent in 2007.1 Figure 2: Poverty headcount, 1994-2007 (%) 1994 2004 2007 50 45 40 35 30 25 20 15 10 5 0 Phnom Penh Other urban Rural Cambodia Source: Cambodia Socio-Economic Survey (CSES) 2007 (NIS, 2007), analyzed in World Bank (2009a). A third of the population (just over 4 million people) continued to live in poverty in 2007, however, with the overwhelming majority of poor and vulnerable living in rural areas. The 2007 data show that 92 percent of the poor live in rural areas. Moreover, poverty rates are much higher in rural areas: the average rural poverty rate is 35 percent, compared with an urban rate of 22 percent (excluding Phnom Penh). The percentage of the population under the food poverty line is 18 percent on average across the country and 21 percent in rural areas (CSES 2007 (NIS, 2007), analyzed in World Bank, 2009a). The 2004 regional poverty data for the main agro-climatic zones of the country show striking regional variations. The Tonle Sap and Plateau/Mountain zones had the highest poverty headcounts (43 percent and 52 percent, respectively, compared with a national average of 35 percent). Poverty magnitude (the number of poor people) was largest in the more densely populated Tonle Sap and Plains regions.2 The provinces with the highest poverty rates (proportion of the population in the province living below the poverty line) were Siem Reap, Kampong Thom and Kampong Speu. 1 The overall poverty line includes the food poverty line (based on an estimated cost of a basket of food providing a dietary intake of 2,100 calories per day) plus non-food allowances (derived from non-food consumption of those whose total per capita household consumption is equal to the food poverty line). The poverty line for Cambodia overall in 2007 is 2,473 Riel (3,092 Riel for Phnom Penh, 2,704 Riel other urban areas, 2,367 Riel for rural areas). 2 Up-to-date data on regional and provincial poverty headcounts will be available from the 2009 CSES. The CSES 2007 and 2008 are relatively small in size – 3,593 households compared with almost 12,000 in the 2004 and 2009 CSES – and cannot provide estimates for smaller geographical areas such as regions or provinces. 34 Towards a Social Protection Strategy for the Poor and Vulnerable | Profile of Poverty and Vulnerability Figure 3: Geographic variations in poverty rates Oddar Meanchey Banteay Preah Vihear Meanchey Stung Treng Ratanak Kiri Siem Reap Battambang Krong Kampong Thom Pailin Kracheh Mondul Kiri Pursat Kampong Chnang Kompong Cham Phnom Kampong Koh Kong Penh Less than 25% Speu Prey Kandal Veng 25% to 35% 35% to 45% Svay 45% to 60% Takeo Rieng Kampot Sihanoukville Krong Kep Source: CSES 2004 (NIS, 2004a). Note: The 2004 instead of the 2007 CSES has been used because of its greater representativeness at province level. Poverty estimates are based on the 12-month sample. Some provinces were grouped together to achieve representativity. Groups are as follows: Kampong Chnang/Pursat; Other Coastal (Krong Preah Sihanouk, Krong Kep, Koh Kong); Other Plateau/Mountain (Kracheh, Mondul Kiri, Preah Vihear, Ratanak Kiri, Stung Treng, Otdar Meanchey, Krong Pailin). Since this snapshot of poverty in Cambodia from 2004 and 2007 data, Cambodia has been hit by consecutive severe macroeconomic shocks – the food, fuel and financial crises – which have further compromised the ability of the poor and vulnerable to cope. The impact of these crises is complex, and comprehensive data on the impact on poverty trends will be available only after the forthcoming analysis of the Cambodian household surveys of 2008 and 2009. However, it is clear that these crises have accentuated difficulties and compromised livelihoods and well-being for many Cambodians, in particular those living near or below the poverty line. There is emerging evidence that progress on some key human development indicators (in particular on nutrition) may have stalled or even reversed. While overall poverty has been falling, there has been a rise in inequality. The rate of consumption increase has varied quite widely and the divide between the rich and the poor has increased (with almost half of the country’s total consumption now enjoyed by the richest 20 percent of the population). Average living standards in urban areas are pulling ahead of those in rural areas (CSES 2007 (NIS, 2007), analyzed in World Bank, 2009a). As a result, there has been an overall increase in inequality, from a Gini coefficient of 0.39 to one of 0.43 in only three years (2004-2007) (ibid). Average expenditure per capita in Phnom Penh 35 is now more than 3.5 times that in rural areas, and the population’s top quintile spends on average 8 times that of the lowest quintile (Vu and Glewwe, 2009). Household well-being is increasingly affected by greater resource competition in the context of demographic growth. The poor continue to depend heavily on access to common resources for fishing, collecting firewood, foraging or hunting wild animals (World Bank, 2009a). Fishing and coastal villages tend to be among the most vulnerable, as common property resources in these types of villages suffer a combination of population pressures, over-exploitation and weak governance. In addition, with low education levels, the poor and vulnerable have low levels of knowledge to reap the benefit of available natural resources and face great challenges when there is a change in natural assets. Similarly, villages that are geographically isolated and have poor access to markets to sell produce or labor are also vulnerable (Fitzgerald and So, 2007). With the population increase in recent years, and the degradation of common property resources through over- exploitation or mismanagement, existing resources are becoming increasingly inadequate to meet livelihood demands in the present organization of resources, technology and knowledge (Kim et al., 2002). Lack of basic assets that would ensure a flow of income and act as collateral also impedes poor households from gaining credit and coping with consumption and income shocks. With limited savings and assets (land, animals, other durable assets that can be sold for cash) to draw on, the poor and vulnerable often take out high-interest loans from moneylenders when facing hard times, spiraling into a vicious cycle of increased poverty and destitution. The 2007 CSES confirms that 46 percent of households in the poorest quintile had one or more loans outstanding, with an average value of $131. There is also a concern that social capital – the networks and relationships of trust, reciprocity and exchanges that facilitate cooperation – is increasingly under pressure from demographic changes, including population growth and migration, as well as divorce and family breakdown (see Section 3). There is also a reported increase in anti-social behavior (youth and gang violence, drug and substance abuse, domestic and public violence, a particular threat of rape) to which the poor are especially vulnerable (So, 2009). 36 Towards a Social Protection Strategy for the Poor and Vulnerable | Profile of Poverty and Vulnerability Risks, shocks and vulnerabilities Households face several risks that can push them into poverty. When realized, these risks become shocks that can drastically change households’ socioeconomic situation and well- being. The degree of change depends on how vulnerable the households are to the shocks. Vulnerability can be defined as the ability of households to cope with adverse shocks. The degree of vulnerability comes from (i) the extent of their exposure to the shock and (ii) their susceptibility and sensitivity to adverse consequences (Grosh et al., 2008). Households with few or inadequate coping strategies against adverse shocks are highly vulnerable to negative impacts. Poverty and vulnerability are intrinsically related. Poverty is one cause of vulnerability, as poor households tend to have fewer coping strategies that protect them against shocks, while vulnerable households are more likely to be pushed into or deeper into poverty. However, vulnerability is also a cause of poverty, as well as a perpetuating and defining element of it (Box 2). Box 2: Poverty and vulnerability Vulnerability is best seen as a cause, symptom and constituent part of poverty: Cause: Vulnerability describes the potential for people to enter into poverty or chronic poverty and highlights the experience of the transient poor (people who move in and out of poverty). Symptom: At the same time, vulnerability is an effect of poverty, as poorer households will be more vulnerable to shocks. This focuses attention on the mutually reinforcing nature of poverty and vulnerability. Constituent part: In a multidimensional definition of poverty, vulnerability is also part of the concept of poverty. When poverty is analyzed beyond the scope of income- or consumption-based indicators of welfare, vulnerability, voicelessness and powerlessness, for instance, constitute other dimensions of its definition. Population Vulnerable to poverty Currently Currently non-poor poor Not at risk of At risk of Likely to move out of Chronically or becoming poor becoming poor poverty but may be structurally poor poor again Sources: Grosh et al. (2008); Prowse (2003). 37 Many households have a tenuous hold on well-being and are vulnerable to shocks that can push them into poverty. The 2006 Poverty Assessment calculated that Cambodia has a sizeable proportion of households (7 percent in 2004) living within a 10 percent band above the poverty line: in other words, if the per capita consumption of these households were to decline by just 10 percent, the poverty rate would increase by 7 percent from 35 percent to 42 percent (World Bank, 2006a). Household, community and nationwide shocks continuously push households near the poverty line in and out of poverty. But movements in and out of poverty do not happen only during crises or among near- poor households. A study of nine Cambodian villages found that, between 2001 and 2004, 8 percent of well-off households and 21 percent of moderately poor households became very poor; at the same time, 36 percent of households that were very poor became moderately poor and 13 percent even became well-off (Fitzgerald and So, 2007). Household-level shocks, such as health and labor shocks, are for the most part responsible for these movements (So, 2009). Table 4: Transition in and out of poverty, 2001-2004 Very poor Moderately poor Well-off 2004 Very poor 51% 36% 13% 2001 Moderately poor 21% 43% 36% Well-off 8% 30% 62% Source: Fitzgerald and So (2007). Note: Data are from a panel survey of nine villages and 890 households. They show transition probabilities.3 Main sources of vulnerability in Cambodia can be grouped into four categories: situations of emergency and crisis; human development constraints; seasonal unemployment and income insecurity; and health shocks. Next, we review these sources of vulnerability. Situations of emergency and crisis: macroeconomic shocks Poor households’ vulnerability to crises is reflected by their exposure to the recent food, fuel and financial crises. Cambodia’s poor suffered from the impact of the food and fuel crisis first and, subsequently, from that of the global financial crisis. These economic shocks have impacted on poor and vulnerable households via loss of employment or lower wages/ hours of work and lower purchasing power. With Cambodia’s integration into the global economy, the impacts of external economic shocks are likely to become more frequent (World Bank, 2009b). Households’ vulnerability to macroeconomic shocks such as the food, fuel and financial crises varies according to their economic status, livelihood and location. For instance, the impact of price fluctuations is complex and the aggregate poverty impacts of the 2007/08 price rises (the price of rice, the staple food crop of Cambodia, increased by approximately 3 In this study, Fitzgerald and So constructed poverty lines for each of the nine villages. Households were labeled “very poor� if their consumption was well below the respective poverty line, “well-off� if their consumption was well above the poverty line and “moderately poor� if they were 20 percent below or above the poverty line. 38 Towards a Social Protection Strategy for the Poor and Vulnerable | Profile of Poverty and Vulnerability 100 percent from 2007 to 2008) is yet to be determined (pending analysis of the CSES 2008). Preliminary analyses suggest that there were some winners (for example rice sellers and agricultural day laborers) but there were also serious impacts on food security and nutrition for many of the poor and vulnerable. As net food buyers, the poor were least able to cope with the steep rise in prices. About 50 percent of households reported cutting back on food consumption as a way of coping and, during the lean season prior to the wet season rice harvest in the latter part of 2008, as many as 2.8 million people could have become food insecure (CDRI, 2008). The high prices may have caused a stagnation and possible reversal of the decline in child malnutrition. The 2008 Cambodia Anthropometric Survey also found a higher prevalence of child disease (fever, diarrhea and acute respiratory infection) that appears to be increasing compared to CDHS 2000 and 2005 (Conkle, 2009). On the heels of the food price shock, the impact of the global economic crisis hit key sectors such as garments, construction and tourism. There have been around 70,000 redundancies in the garment sector, 60,000 in construction and other job losses in tourism, and many firms have reported reducing hours or pay for workers (Chandararot et al., 2009). With their earning power reduced, many wage and self-employed workers had to reduce remittances to rural families. With less money around in rural areas, local demand for goods and services also reduced and off-farm activities became less profitable. Poor and vulnerable households have been hit hard by the economic crisis, with a likely significant social and poverty impact. The overall poverty impact will be known when the findings of the 2009 household survey are issued. In the meantime, rapid assessments and surveys have reported constrained circumstances for many of the poor and vulnerable. There are concerns that women may have been disproportionately affected by the crisis, with significant loss of employment in the garment industry (which mainly employs female workers), increased risk of domestic violence and greater vulnerability to trafficking and exploitation in the commercial sex industry. Low-skilled male workers have also been vulnerable, in particular in the male-dominated construction industry. The effects of the current economic crisis for children of poor and vulnerable households are also likely to have been significant, increasing the risk of child labor (as families depend more on children’s income) and causing poor families to switch to less nutritious food and defer health treatment. 39 Situations of emergency and crisis: natural disasters Cambodia’s unique hydrological regime and low coverage of water control infrastructure makes it vulnerable to climatic and natural disasters. Most rural households rely heavily on subsistence agriculture for their livelihood, especially rice, which accounts for 90 percent of total cultivated area and 80 percent of agricultural labor input (World Bank, 2006b). Agricultural production (and thus households’ food security) is heavily dependent on weather conditions and can fluctuate significantly from year to year. Accordingly, the growth rate of the crop sub-sector is highly variable, reflecting high reliance on adequate rainfall and susceptibility to the weather (CDRI, 2008). Livelihoods and sources of income of the rural population may therefore be compromised, leaving them reliant on social protection from the state and development partners – in particular in the case of natural disasters. Extreme floods and droughts are among the most damaging shocks for rural households, and climate change will heighten their severity. In the past decade, unusual floods and droughts have severely affected large parts of the countryside, resulting in three years of negative agricultural growth (Table 5). In 2009, Typhoon Ketsana left 43 people dead and 67 severely injured and destroyed the homes and livelihoods of some 49,000 families or 180,000 people directly or indirectly (equivalent to 1.4 percent of the population). Most of the affected districts were among the poorest in the country. The widespread damage to property and public infrastructure will have a long-term impact on these communities’ livelihoods (CNCDM, 2010). Looking ahead, although many regions in Cambodia are shielded geographically from climate hazards, almost all provinces are considered vulnerable to the impact of climate change, owing to their low adaptive capacity resulting from financial, technological, infrastructural and institutional constraints (UNDP, 2009). Table 5: Estimated impact of extreme floods and droughts, 2000-2005 Year Type Population affected Provinces Estimated Estimated Growth in agricultural (millions) affected deaths damage ($m) GDP (%) 2000/01 Flood 3.4 19 347 157 - 0.4 2001/02 Flood 2.1 62 36 3.6 Drought c. 0.5 6 2002/03 Drought 2 All (8 acutely) 22 - 2.5 Flood 1.5 29 12 2004/05 Drought 2 14 21 -0.9 Source: ADI (2007), compiled from various sources (agricultural GDP from the National Institute of Statistics (NIS)), in CARD et al. (2009). Poor households also rely on use of natural resources such as water resources and forests to generate income. Access to common property provides an important safety net for the rural poor in bad harvest years. The 2006 Poverty Assessment found that one-quarter of the poor depended on only fishery and forest products for over half their income in 2004 and, on average, fishery and forest products accounted for 25 percent of household income among the poor (World Bank, 2006a). However, access to this common property is becoming increasingly limited. As captured in the qualitative Participatory Poverty Assessment by the Cambodia Development Resource Institute (CDRI) (Ballard et al., 2007), many of the extractive activities in the forest are not complying with rules and regulations. Rising 40 Towards a Social Protection Strategy for the Poor and Vulnerable | Profile of Poverty and Vulnerability population numbers have also contributed to overexploitation and a decline in resource availability. In addition, leasing of water bodies to business interests and increasing restrictions on free access to fisheries are already evident in places where the poorest depend on hunting and gathering for their livelihoods. Rural households’ vulnerability to climate and economic shocks is exacerbated by the low productivity and low diversification of their income-generating activities. Most rural households rely heavily on subsistence agriculture for their livelihoods: an estimated 72 percent of Cambodians are dependent on fishing and agriculture (CNCDM, 2010). In addition, household-level agricultural productivity remains low: rice yields, for instance, remain among the lowest in the region, owing to limited and poor use of improved seed, fertilizer, tillage and water management (CARD et al., 2009). Human development constraints: poor maternal and child nutrition Chronic and acute malnutrition remains high in Cambodia, barring young Cambodians from life opportunities, undermining higher levels of economic growth and negatively affecting human development. Malnutrition is caused by inadequate infant and young child feeding (IYCF) practices, high levels of infectious disease and the inability to access and afford nutritious food. Despite efforts by the RGC to address the underlying causes of malnutrition, the percentages of wasted (8.9 percent), short (39.5 percent) and underweight (28.8 percent) children remain among the highest in Asia. Evidence from the most recent CAS (Conkle, 2009) shows that, following improvements between 2000 and 2005, chronic malnutrition rates have stagnated since 2005. Acute malnutrition saw an increase between 2007 and 2008, possibly as a result of the food price crisis. 41 While poverty contributes to observed malnutrition, other factors are also at play. Although poor children suffer most from malnutrition, it affects all quintiles of the consumption distribution (Table 6) as, in addition to poverty, malnutrition is influenced by a number of health, hygiene and feeding behaviors. For instance, households may not feed children enough nutritious food. The CAS revealed that mean consumption among children whose ages range between 6 and 35 months is of just 4.7 out of 14 food groups (Conkle, 2009). Particularly low is the consumption of milk products, oils and fats, and legumes and nuts. A recent analysis also showed that washing hands before feeding children (10 percent) and before preparing food (25 percent) is infrequent, as is washing after cleaning a child who has defecated (11 percent) (Johnston, 2006). From these and related practices, many children become ill with diarrhea, skin diseases, measles and malaria, all of which can contribute to malnutrition. Table 6: Malnutrition rates by wealth quintile Quintile 1 Quintile 2 Quintile 3 Quintile 4 Quintile 5 Stunting 2000 52.6% 48.4% 42.6% 41.5% 27.5% Stunting 2005 46.7% 42.5% 36.5% 35.5% 19.4% Underweight 2000 52% 47.9% 42% 43.1% 33.5% Underweight 2005 42.9% 39.8% 33.5% 34.3% 23.1% Source: Cambodia Demographic and Health Survey (DHS) 2000 and 2005 (NIPH et al., 2001; 2006). Human development constraints: poor access to quality education The education sector faces important challenges in terms of quality of service delivery. The limitations of basic school facilities, a shortage of textbooks and an inadequate supply of (trained) teachers have made it difficult to provide quality public education. Cambodia suffers in particular from a severe shortage of teachers for primary and secondary education: in 2009/2010, the number of students per teacher averaged 49 in primary and 26 in secondary school (EMIS Office, 2010). As a result, educational outcomes remain low, with particular deficiencies in writing and mathematics (Marshall et al., 2009). Poor access to quality of education partly influences repetition and high dropout rates. Although enrollment rates are as high as 81.2 percent in primary school, they decrease sharply to 28.6 percent in lower secondary and 13.5 percent in upper secondary school, according to the CSES 2007 (NIS, 2007). Children in rural areas are more than two times less likely to continue to lower secondary than children in Phnom Penh (25 percent of attendance of the former vs. 61 percent of the latter) (ibid). Dropout rates in primary, lower secondary and upper secondary education stand at 8.3 percent, 18.8 percent and 11.2 percent, respectively (EMIS Office, 2010). Dropout rates in rural areas are twice as high as those in urban areas for secondary education: 20.8 percent vs. 11.5 percent in the case of lower secondary and 14.7 percent vs. 6.3 percent in the case of upper secondary. The repetition rate in primary school reaches 8.9 percent (ibid). As a result, Cambodia’s youth tend to have low educational achievements: 63 percent of Cambodia’s youth (15-24) can be defined as out of school: those who have either never attended school or dropped out before completing secondary education. Of these, 12 percent have never been in school, according to the CSES 2007 (NIS, 2007). The percentage of out- 42 Towards a Social Protection Strategy for the Poor and Vulnerable | Profile of Poverty and Vulnerability of-school youth is high to begin with (among those aged 15, more than a quarter), and it increases dramatically with age (more than two-thirds of those of 19 years old (Figure 4)). Moreover, while out-of-school youth do tend to stay in school longer than five years, given high repetition rates (see above paragraph), they de facto barely complete primary school: in rural areas, the average number of completed academic years of education among out-of- school youth is below five years, and urban areas perform only slightly better. Furthermore, the country is undergoing a demographic transition, given the increasingly large youth cohort entering the labor market. For instance, 89 percent of out–of-school youth participate in the labor force. Without adequate social protection and in the face of high job insecurity, concerns for the well-being of Cambodian youth remain high. Figure 4: Youth out-of-school status and education Out of school youth rates (%) Education of out of school youth 100 .12 88.0 90 80 .1 83.0 83.6 79.2 70 72.6 .08 70.4 Density 60 61.0 50 51.0 .06 40 30 36.9 .04 20 26.5 10 .02 15 16 17 18 19 20 21 22 23 24 0 1 2 3 4 5 6 7 8 9 10 11 Age Urban Years of schooling Rural Source: CSES 2007 (NIS, 2007). 43 Most workers begin their professional career in low-paid informal jobs and move at a later age towards self-employment, remaining therefore in low-paid and low-skilled informal sector jobs (Figure 5). In both urban and rural areas, most youth begin their professional careers as unpaid family workers, one of the least-paid activities and one which does not tend to require sophisticated skills. When they grow older, cohorts of unpaid family workers appear to pursue their careers as self-employed persons, remaining therefore in low-paid and low-skilled informal sector jobs. In both urban and rural areas, the share of wage workers among the working population remains in fact quite stable across youth and non- youth, suggesting a somewhat segmented labor market between wage workers and the self- employed. To be sure, wage workers are not necessarily better remunerated than the self- employed.4 But a steady proportion of wage workers across youth and non-youth remains symptomatic of a static labor market that does not offer to the many unskilled the possibility to move toward better remunerated jobs. Figure 5: Distribution of workers by category of employment (%) Urban Rural Non-youth Youth Wage Wage worker worker Unpaid Unpaid family family worker worker Self- Self- employed employed 0 20 40 60 0 20 40 60 70 Source: Calculated from CSES 2008 (NIS, 2008a). Seasonal unemployment and income insecurity In Cambodia, there are about 350,000 food-insecure households with poor and borderline food consumption, equaling about 1.7 million individuals (CDRI, 2008). The highest number of food-insecure households is observed around the Tonle Sap zone, followed by the Plains zone, the Plateau zone and the Coastal zone. The Tonle Sap and Plateau zones are also the ones with the highest incidence of urban food insecurity (Figure 6). Most food-insecure households live in rural areas, are landless (estimated at 15 percent of the rural population) or land poor (47 percent of the rural population)5 and have more children and more elderly to be nourished (ibid).6 4 Unfortunately, the CSES is not a panel and income data at the individual level are available only for wage workers, which makes analyses of labor market dynamics difficult. 5 47 percent of the rural population has less than 0.5 hectares of land (GTZ, 2010). 6 In the average food-insecure household, less than half of members are in the productive age range of 18-59; this proportion is nearly 65 percent in the average household in Phnom Penh (CDRI, 2008). 44 Towards a Social Protection Strategy for the Poor and Vulnerable | Profile of Poverty and Vulnerability Figure 6: Food-insecure households (%) 20 Rural Urban 16 12 8 4 0 Plains Tonle Sap Plateau Coastal Cambodia Source: Calculated from CSES 2008 (NIS, 2008a). According to the 2009 Global Hunger Index, Cambodia remains within “alarming� levels of hunger (von Grebmer et al., 2009). Among the rural poor, the main causes of food insecurity include lack of access to land, livestock, credit, markets and agricultural inputs. Rural households’ food consumption averaged 66 percent of total consumption in 2007 and the share of food in total consumption was 73 percent for the poorest consumption quintile (CSES 2007 (NIS, 2007), analyzed in World Bank, 2009a). The Cambodia Development Research Institute (CDRI) estimates that about 40 percent of rural households and 80 percent of urban households are net rice buyers. Net food buyers are the least able to cope with fluctuations in staple food prices (CDRI, 2010). Qualitative evidence suggests that food insecurity is particularly high during the lean season, where the number of food-insecure households rises significantly. A 2009 baseline survey in the Tonle Sap basin found that poor and very poor families failed to generate sufficient income to meet their needs for most months of the year: during the first 10 months of 2009, for instance, poor and very poor households had an aggregated income gap of 581,000 Riel (circa $145). The period from August to November is particularly severe for poor households, as the demand for agricultural labor is low and households’ rice stocks start running out (Koy and Em, 2009). Poor rural households are predominantly dependent on their own limited food production and irregular, low-paid, casual wage labor. The seasonality of labor requirements in farming means that households, especially those with little or no land, are obliged to find off-farm employment in the slack agricultural season to supplement family income. Given the limited availability of non-farm employment, households increasingly need to rely on income from unskilled wage employment in urban areas or in neighboring countries. Seasonal labor migration is particularly common in provinces near Thailand and Viet Nam. Food-insecure households are also particularly exposed to fluctuations in food prices, as shown by the food and fuel crisis in 2007/08. Cambodia’s food prices rose dramatically, by 37 percent from July 2007 to July 2008 (Chan, 2008), and rice prices even doubled. While the spike in food prices may not have had a strong impact on overall poverty, as it 45 hurt net food consumers but also benefited net food producers, it had a strong impact on food-insecure households, which tend to be landless and had to rely on detrimental coping strategies, such as selling assets and pulling children out of school to face increased prices (CDRI, 2008). Health shocks and poor access to quality health care Health shocks are among the major sources of vulnerability faced by the poor. Health shocks affect the poor disproportionately through three channels. First, the poor may have a higher prevalence of injuries and illnesses, as they are often involved in physical jobs, facing greater risk of accidents and injury; have poor nutrition; have less access to improved water sources; live in poorer housing conditions; and have less access to health and social services (World Bank, 2006b). Second, foregone income by not being able to work has a greater impact on the poor, because of a lack of savings. Lastly, poor households get trapped in a vicious cycle of high health care costs; out-of-pocket expenditures for seeking health care; indebtedness at overwhelmingly high interest rates when household resources are insufficient; selling assets (usually land) when all other funds are depleted; ending in further poverty and destitution. Once the cycle starts, it is very difficult to break out. As such, it is important to prevent it from starting by tackling high health care costs and out-of-pocket expenditures. Between 2004 and 2007, out-of-pocket expenditures for health costs as a proportion of total consumption halved for the richest income quintiles but did not decrease much for the poorest ones. Health expenditures increased for all wealth quintiles from 2004 to 2007, except for the richest quintile. In addition, real per capita consumption increased in all wealth quintiles but the highest increase also occurred in the richest quintile. These two facts explain why health expenditure as a proportion of total per capita consumption showed the most significant decrease in the richest quintile (Figure 7), while the three lowest quintiles showed marginal progress in decreasing this burden despite the favorable economic environment, according to the CSES 2007 (NIS, 2007). Observe also that the numbers presented are averages across households, so that, while out-of-pocket expenditures appear to represent a small proportion of households’ consumption, they can be much higher for households affected by shocks. Figure 7: Per capita out-of-pocket expenditure as a proportion of total consumption, 2004/2007 (%) 7 2004 2007 6 5 4 3 2 1 0 Poorest Next poorest Middle Next richest Richest Source: CSES 2007 (NIS, 2007) analyzed in World Bank (2009a). 46 Towards a Social Protection Strategy for the Poor and Vulnerable | Profile of Poverty and Vulnerability When poor families cannot afford health care, they may resort to detrimental strategies such as taking out debt or going to a range of traditional healers and other unqualified (and often dangerous) private service providers. Analyses of CSES data have found that, while fewer households were in debt for health care in 2007 compared with 2004, and the nationwide household debt for health care dropped to 3.5 percent in 2007 (from 5.1 percent in 2004, possibly because of the introduction of HEFs),7 the situation of poor indebted borrowers worsened between 2004 and 2007: in 2007, 62 percent of borrowers were more severely indebted for health care in 2007 than when they took out the loan, and the overall level of interest that people were required to pay was higher than before (Jordanwood et al., 2009). Studies also find a strong and positive impact of HEFs on out-of-pocket expenditure: in districts covered by a HEF implemented by University Research Co. (URC), 55 percent fewer households have a health-related debt (ibid). However, resorting to unqualified health service providers such as traditional healers or not seeking medical attention altogether remain common practices among poor households, with detrimental consequences. In addition to health care costs, other factors affect poor households’ ability to seek health care. Access to quality health care is limited for poor individuals owing to factors such as distance from health facilities, difficult and expensive transport, lack of qualified health staff in remote facilities, lack of drugs and equipment, limited opening hours of health facilities and negative attitudes of health staff. A serious outcome of poor access to quality health care is persistently high maternal mortality rates. Maternal mortality, at 461 casualties per 100,000 born (Census 2008 (NIS, 2008b)), remains one of the worst outcomes of poor health service provision. Only 63 percent of births were attended by trained birth attendants in 2009, and this actually represented a significant increase from 22 percent in 2003 (MoH, 2010). Less than half (44 percent) of deliveries happen at public health facilities (ibid). The risks to maternal and child survival are very high when a mother delivers outside a health facility, without access to qualified personnel and equipment. Women also frequently skip antenatal and postnatal checkups because of poverty and their precarious employment and income situation. Only 47 percent of women attend the four recommended antenatal checkups (Conkle, 2008) and 64 percent visit a health specialist after delivery (MoH, 2010) 7 Please see Section 3 for a more detailed explanation of Health Equity Funds. 47 Pregnant women are particularly at risk in rural areas, where three delays contributing to maternal mortality often occur: (i) delay in seeking medical care because of a failure to identify danger signs; (ii) delay in reaching medical care because of distance, road conditions and lack of a means of transport; and (iii) delay in receiving adequate medical care as facilities are often poorly equipped and services provided by unskilled personnel. High under-five mortality in the lower wealth quintiles is another indicator of how health shocks disproportionately affect poor children. Apart from the high under-five mortality rate of 83 casualties per 1,000 children (CDHS 2005 (NIPH et al., 2006)), Cambodian children in the lowest quintile have almost three times as much chance of dying before the age of five than those in the highest wealth quintile (ADB et al., 2010). This suggests the importance of income disparities in exacerbating the vulnerability of infants. Poor access to safe water and sanitation also increases the vulnerability of poor households to health shocks and contributes to high maternal and under-five mortality. Of the poorest quintile, half of households do not have access to improved sources of drinking water and 87.2 percent live without sanitation. In contrast, 70 percent of households in the richest quintile have access to the former and 77 percent to the latter (CSES 2007 (NIS, 2007)). According to the 2008 Census, 76.8 percent of all rural households and 18.5 percent of urban households do not have a household latrine (NIS, 2008b). Less than half (46.8 percent) of all Cambodian households had access to an improved source of drinking water in 2008, including 78.8 percent of urban and 40.8 percent of rural households (ibid). 48 Towards a Social Protection Strategy for the Poor and Vulnerable | Profile of Poverty and Vulnerability Vulnerable groups Particular groups in society are especially vulnerable to shocks and to the detrimental effect of inadequate coping strategies, and it is vital to identify them for targeted social protection interventions. Analyzing poverty levels among different groups that are likely to be vulnerable can be a good start in identifying those who deserve special consideration (Table 7). However, poverty is not the only indicator of vulnerability, and additional analysis should be carried out on: the exposure and capacity to mitigate risks that these groups might have; detrimental coping strategies that could exacerbate negative consequences of shocks within groups; and the negative impact of shocks in particular groups that can have long- term consequences for economic and social development. Table 7: Poverty levels of selected population groups Group % Population % Poor Poverty gap Elderly (65 years old or above) 4.3 25.9 5.5 Employed 53.3 28.3 6.6 Members of female-headed households 18.2 27.9 7.2 Members of male-headed households 81.8 31.2 7.5 Members of employed-headed households 91.1 31.0 7.5 People with disabilities 1.3 28.6 7.2 Ethnic minorities 2.2 36.1 10.0 Infants (below 1 year old) 4.0 38.9 10.0 Children (aged 0-14) 33.5 37.4 9.4 Cambodia 100 30.5 7.4 Source: CSES 2007 (NIS, 2007). Progress in achieving the CMDGs reflects the need to prioritize interventions targeting children, women and the food insecure. Figure 8 summarizes progress in achieving the key CMDG targets (UNDP, 2010b, with data for 2009). Although there has been great progress in improving health coverage and effectiveness in relation to child mortality and HIV/AIDS, malaria and other diseases, improving maternal health, basic education and environment CMDGs remains unlikely to be achieved by 2015. In terms of maternal heath, there has been progress of just 37 percent from the baseline, mainly because of an increase 49 in the maternal mortality ratio (from 437 in 1997 to 461 in 2008). In the case of education, persistently low enrollment and survival rates in lower secondary (less than 40 percent in both cases in 2008 (UNDP, 2010a)) are far from the universal nine-year basic education goal. More work needs to be done to achieve the goals on reducing poverty and hunger,8 as well as in achieving gender equality and women’s empowerment. This situation reflects how social protection interventions focused on children, pregnant women and poor and food-insecure people could have a positive impact with regard to short- and medium-term development goals. Figure 8: Progress in achieving key CMDG targets CMDG1: Poverty/Hunger 100 90 80 CMDG9: CMDG2: 70 Demining/UXO Basic education 60 50 45 58 40 30 41 20 10 CMDG7: 0 CMDG3: Enviroment 47 41 Gender equality 37 66 83 CMDG6: CMDG4: HIV/AIDS, other Child Mortality CMDG5: Health Source: UNDP (2010b). Note: Percentage distance from benchmark to latest available data in 2009. Infants and children Strong priority should be given to assistance to infants and children, who represent more than a third of the population and face the highest poverty rates among vulnerable groups (Table 7). Because poor households tend to be of larger size, infants and children are on average the poorest group in Cambodia. They are also one of the most vulnerable, as shown by high rates of malnutrition and child mortality, poor educational attainments and high incidence of child labor. Cambodia is among the 36 countries with the highest burden of child under-nutrition in the world and one of 33 “alarming or extremely alarming� countries in terms of hunger and under-nutrition, with very poor nutritional status of women of childbearing age and very high child and maternal mortality rates (von Grebmer et al., 2009). In recent years, the situation of child mortality has improved substantially, but there are still 8 Worrying indicators for this CMDG are increasing child labor and decreasing share in consumption of the poorest population. 50 Towards a Social Protection Strategy for the Poor and Vulnerable | Profile of Poverty and Vulnerability wide regional variations that need to be addressed: infant and under-five mortality is still almost double the national average in Kampong Speu, Preah Vihear/Stung Treng, Prey Veng, Mondul Kiri and Ratanak Kiri (UNDP, 2010a). Supporting poor infants and children by providing better health care and nutritional support and better quality of education should therefore be a priority. Furthermore, as children represent Cambodia’s next generation of citizens and workers, supporting children is a solid investment into the country’s future. A recent study conducted by UNICEF to measure child poverty and disparities in Cambodia shows that over 1.7 million children younger than 18 were income poor in 2004. It is estimated that child poverty is higher among the younger age groups and reduces as age increases. However, small children are not currently targeted by social protection at all. Children in Cambodia face absolute poverty of 64 percent, when measured by deprivation from at least two areas of well-being (UNICEF, 2009).9 Moreover, child deprivation is higher among the younger age groups. Although disaggregation by age remains limited, poverty assessments by household size and type show that families with children under 15 experience higher poverty levels than other families. Income poverty combined with inadequate access to basic social services seriously affects the well-being of children and can compromise their health, education and protection. Infants and children are at risk of detrimental coping strategies that can have lifetime consequences, including being fed less or lower nutritional quality food, being pulled out of school to take care of younger siblings or to enter into child labor and becoming victims of human trafficking. Infants and children are highly vulnerable to shocks as they lack the ability to mitigate risks and to control adverse circumstances. 9 According to UNICEF, the areas of well-being are food, health, education, information, water, sanitation and shelter. 51 Women of reproductive age (15-49) Girls and women, especially those of reproductive age, also remain highly vulnerable. While there has been progress in advancing gender equality and opportunities for women in Cambodia, young girls and women continue to be particularly vulnerable. Women have particular vulnerabilities arising from their health needs: maternal mortality remains at an unacceptably high rate; women make up a bigger proportion of HIV-infected adults than they did in the past (52 percent in 2009 vs. 38 percent in 1997 (NCHADS, 2007));10 the low nutritional status of women is a growing concern; and the overall number of women reporting constraints in accessing health care remains high.11 Violence against young girls and women is widely prevalent. One-quarter of women in Cambodia aged 15-49 who have been married have experienced some sort of emotional, physical or sexual violence since they were 15 years old (CDHS 2005 (NIPH et al., 2006)). Sexual exploitation appears to be entrenched in Cambodia, and sexual harassment is prevalent in informal, vulnerable occupations (commercial and indirect sex workers, beer promotion girls and karaoke hostesses). The reported incidence of rape and sexual assault is also increasing, particularly that involving young girls and children (MoWA, 2008). Although some progress has been made, low levels of female education compromise the contribution of women and their children to future economic development. There have been some successes in increasing girls’ enrollment rates at primary school level but 10 Of all new infections among women, two-thirds will be among non-sex workers or women at “low risk� (NAA and MoWA, 2009). 11 In 2005, 89 percent of women reported at least one problem in accessing health care. Getting money for treatment remains the main problem, followed by the concern that no provider or drugs are available and not wanting to go to health services alone (CDHS 2005 (NIPH et al., 2006)). 52 Towards a Social Protection Strategy for the Poor and Vulnerable | Profile of Poverty and Vulnerability female share of enrollment declines at each level of education, and low levels of literacy and education constrain women’s employment and life opportunities. Evidence suggests that low levels of education of mothers have a detrimental impact on the education and health of their children (UNDP, 2010). For instance, under-five mortality in Cambodia is 27 percent higher among children of mothers who did not finish primary schooling than among those whose mothers completed primary education (ADB et al., 2010). Food-insecure households Food-insecure households remain highly vulnerable to different types of shocks owing to their poor nutrition and lack of remaining coping strategies. Constant food deprivation increases the chances for food-insecure households of facing health shocks, as their health increasingly deteriorates over time. Also, few alternatives are left for households that have resorted to cutting food consumption to cope with adverse economic shocks (Box 3). Box 3: Vulnerability profile of food-insecure households According to initial findings of WFP Cambodia's recent vulnerability profile, food-insecure households in Cambodia have the following characteristics: • The most vulnerable households are those that are unable to supplement their incomes from agriculture activities. • Food-insecure households, however, tend to rely on agriculture alone, although they have less access to land and relatively smaller landholdings: households with poor food consumption are more likely to be landless than the overall rural population (27 percent compared with 20 percent, respectively). Landless households, including those that lease land, are also more likely to be more food insecure. • Food-insecure households face a greater age dependency burden than other rural households, and they tend to make up a higher proportion of those with only one family member working for a cash income (42 percent). • Food-insecure households have higher primary school absenteeism rates than food-secure households (40 percent compared with 22 percent, respectively). Source: WFP (2009a). The majority of food-insecure households are rural; however, there are also groups living in extreme poverty in urban areas that face food insecurity and low standards of living, with marginalization from access to services and lack of secure housing tenure. It is estimated that informal and other urban poor communities constitute 20 percent (or about 250,000 people) of the population of Phnom Penh.12 With the urban growth rate set to continue increasing (at 2.21 percent from 1998 to 2008) (Census 2008 (NIS, 2008b)), the proportion of vulnerable urban poor will grow. Special vulnerable groups While they represent a minority of the population and may not be income poor, the Rectangular Strategy, the NSDP Update 2009-2013 and sector ministries’ strategies identify special groups that remain particularly vulnerable along one or several dimensions. These groups include: people living with HIV and their families; patients of tuberculosis 12 www.unhabitat.org/content.asp?cid=711&catid=251&typeid=13&subMenuId=0. Last access March 11, 2010. 53 (TB); homeless people; people living with disabilities; orphan children and at-risk children and youth; victims of violence, abuse and exploitation; indigenous and ethnic minorities; families of migrants; veterans; and the elderly. These groups face particular challenges owing to overlapping vulnerabilities that are often experienced on top of income poverty, and can be considered special vulnerable groups because they either warrant priority action in strengthening their social protection or require specific support. Ethnic minorities face higher poverty rates and much higher poverty gaps than the national average (Table 7). Although ethnic minorities represent a small share of the population, their living standards are much lower than the national average. Moreover, ethnic minorities also face non-monetary disadvantages related to language remoteness and discrimination. They therefore deserve particular attention in social assistance, in terms of both tailoring programs to cultural values and finding appropriate targeting mechanisms (e.g. geographical) that allow programs to cover their particular needs. While the number of older persons in Cambodia is increasing, traditional support mechanisms for older people (families, pagodas, etc) are being weakened through changing family profiles and migration. According to the 2004 Cambodia Inter-Censal Population Survey, the proportion of older persons in the population is currently 6.0 percent. Given lower fertility and higher life expectancy, this is expected to increase to 11.7 percent by 2050 (UNDESA, 2002).13 At the same time, traditional support mechanisms for older people (families, pagodas, etc) are being weakened through changing family profiles and migration, which forces the elderly to join the labor force. The proportion of working elders (65+ years old) increased from 46.59 percent in 1998 to 54.52 percent in 2008 (Census 2008 (NIS, 2008b)), with the participation rate of rural elders much higher than that of urban residents (59.4 percent vs. 29.4 percent in 2008), a phenomenon that can be explained by the general involvement of the rural population in agricultural activities. Similarly, Cambodians with chronic illnesses have very little support to obtain independent and sustainable livelihoods. Particular attention needs to be paid to households with people living with HIV and TB. There are 57,900 adults living with HIV in Cambodia and 5,473 children are known to be infected with HIV (NCHADS, 2007). Lack of food security and poor nutrition accelerate progression to AIDS-related illnesses and tend to negatively impact on adherence to treatment and response to anti-retroviral therapy (ART). Vulnerability and poverty are compounded by stigma and discrimination. TB also remains a major public health concern: Cambodia ranks 21st among countries with the highest burden of TB. According to the WHO, the incidence rate for all forms of TB is 500 per 100,000 people per year, with an estimated 64 percent of Cambodians infected with M. tuberculosis. The twin burden of TB and HIV epidemics can have devastating consequences (WHO, 2008). Children in these households are often at greater risk. For instance, when a parent is ill for a protracted period, whether as a result of HIV infection or because of any other chronic illness, a number of changes often take place in the household. Expenditures on health care increase, often with a resulting reduction of funds available for food, education and other household expenses. Moreover, children often assume adult roles, such as caring for the sick adult, running the household or caring for other children in the family. According to 2005 13 Older persons are defined as those above the age 60. 54 Towards a Social Protection Strategy for the Poor and Vulnerable | Profile of Poverty and Vulnerability DHS data (NIPH et al., 2006), children with chronically ill parents14 are significantly less likely than other children to have basic possessions, such as shoes or two or more sets of clothes. Cambodians living with disabilities also have little support to obtain independent and sustainable livelihoods, and often have to rely on limited family support to survive. Very few people living with disability have access to rehabilitation and appropriate basic services. Their vulnerability goes further than mere lack of financial resources at individual and/or family level, however; it also encompasses cultural and social barriers, inadequate availability and access to education, health and rehabilitation services, lack of awareness of their rights and dependency on others. Disability and inability to work affect poor households through both foregone income and the added cost of providing special care. Cambodians with disabilities and the inactive elderly poor have little support to obtain independent and sustainable livelihoods, and often have to rely on limited family support to survive. Most disabled or older Cambodians continue to depend, at least to a certain extent, on their families, given the lack of a welfare system in the country (NCPD, 2007). If these extra costs of support are taken into account (which is rarely the case), households with disabled and inactive members would be significantly poorer and more vulnerable than households without these members (Braithwaite and Mont, 2008). Relative levels of vulnerability Although difficult to quantify, vulnerable groups have different types and “levels� of vulnerability, which ought to influence design and prioritization of safety net interventions. As shown in the next section, the RGC and development partner budget for safety net development remains limited. It is therefore important to suggest priorities with regard to 14 Of Cambodian children (aged 0-17), 6.1 percent had one or both parents who had been very sick for three or more months the previous year. 55 who deserves more attention, so as to be able to address the largest sources of vulnerability first (in terms of both severity and size of the group). Given high levels of poverty and malnutrition, and limited ability to avoid and mitigate adverse circumstances, children are the largest and most vulnerable group in Cambodia. Infants and children (0-14) represent more than one-third of the population, and reducing the vulnerability of this group can have a great impact both today and in generations to come. It is therefore of the utmost importance to give them priority in safety net development. Women of reproductive age are the second-largest vulnerable group, as they are particularly vulnerable to negative coping strategies within the family (like reduced food consumption) and to risks derived from poor access to quality health care. Safety nets targeted to these groups can reduce highly negative outcomes such as child and maternal mortality, child labor and poor cognitive development owing to malnutrition. Food-insecure households have a high level of vulnerability, as the consequences of malnutrition can be persistently severe. Safety net interventions addressing income and food insecurity can help chronically poor households cover basic needs, while preventing households that are not necessarily poor but that are experiencing difficulties in affording food from falling into poverty. Although special groups represent a small proportion of the population, their level of vulnerability remains significant. Limited access to income-generating activities, lack of self-help capacity and welfare benefits, social discrimination and poverty are some of the considerations that should be included when designing safety net interventions for these special vulnerable groups. Additionally, social protection programs should be sensitive to these group’s needs. Coping strategies Households resort to various types of coping strategies when facing adverse shocks. Some of these have detrimental outcomes with regard to household livelihoods and become the driving factors that push households into (deeper) poverty. Vulnerable households are more likely to use detrimental coping strategies, some of which have a negative impact on specific groups, such as women and children. Many of the negative coping strategies have longer- term consequences and often can lead to even greater exposure to risk and diminished ability to manage risks. While these informal strategies tend to become less dominant with growing country income, they remain a cornerstone of risk coping and mitigation strategies, even in the most developed countries. The most frequent coping strategy in times of distress is to sell assets (including land), take out loans or use own savings. Table 8 presents main mitigation and coping strategies cited in a qualitative survey conducted in five provinces. While the survey was not nationally representative, it indicates that households rely overwhelmingly on assets, their own savings (both cash and in kind) and loans in times of distress. Selling livestock is another coping mechanism that has been found to be frequent in time of distress (CDRI, 2008). Given urgent needs to pay medical bills or alleviate food insecurity, households forced to sell assets may quickly resort to selling at unfavorable, below-market prices. Although such sales of assets provide fast access to cash, they may leave households impoverished in the longer term. 56 Towards a Social Protection Strategy for the Poor and Vulnerable | Profile of Poverty and Vulnerability Table 8: Observed coping and mitigation strategies Study site Risk mitigation measures Coping strategies Common Rare/extreme Banteay • Moving to Poipet • Dropping out of school, delaying • Criminal activity Meanchey and school entry • Prostitution/selling virginity Battambang • Remarriage • Alcoholism and drug abuse • Pawning assets to neighbors • Begging • Immigration to Thailand as a day laborer or beggar Phnom Penh • Tontine (savings • Dropping out of school to work • Criminal activity associations) (shining shoes, selling flowers, • Prostitution/selling virginity collecting trash) • Alcoholism and drug abuse • Pawning assets • Returning to home villages • Squatting, living on the streets, seeking refuge in temples or other public places • Begging Prey Veng • Migration to Phnom Penh • Dropping out of school, delaying • Sale of land/livestock or other areas school entry • Debt evasion • Immigration to Thailand or • Visiting private healers/pharmacies • Criminal activity Viet Nam for work • Borrowing from private • Selling virginity moneylenders/credit schemes • Alcoholism and drug abuse • Pawning assets to neighbors • Begging Ratanak Kiri • Not enrolling in/dropping out of • Criminal activity school • Suicide • Sale of land, moving deeper into the • Deforestation of communal forest trees • Visiting traditional healers • Begging Source: Muny et al. (2004), in World Bank (2006a). Cutting back on food consumption is another frequent coping mechanism of households facing difficulties. A CDRI study (2008) revealed that, in times of distress, 71 percent of the households interviewed recognized that they had, at least once in a while, relied on less preferred and less expensive food, 59 percent had purchased food on credit or incurred in debt, 52 percent had reduced their food consumption, 43 percent had restricted adult food consumption to feed children and 41 percent had given less food to mothers and elder sisters to feed others. This last coping mechanism is likely to deteriorate the nutritional status of pregnant women and women of childbearing age. Box 4: Selling buffalo to cope with illness Grandmother Khoeu, an elderly and very poor Lao-Khmer widow living in O Kaan Village in Ratanak Kiri province, explained how she resorted to selling her most valuable asset, a buffalo, to deal with a serious illness in the family. In 2002, Grandmother Khoeu was living with her two adult daughters, both of whom had been abandoned by their husbands. One daughter had four small children; the other fell seriously ill as a result of a miscarriage. To afford care for her sick daughter, Khoeu exchanged her buffalo, the offspring of a cow she had received through the Seila Program, for a female cow. She then pawned the cow for 80,000 Riel (about $20) to her neighbor to pay for care that would help cure her daughter. Afterwards, she was threatened by the local moneylender, who claimed that the buffalo belonged to the Village Development Committee and was not Khoeu’s property to sell. Source: Muny et al. (2004), in World Bank (2006). 57 International and internal migration has also become an increasingly common strategy for poor rural families seeking better opportunities or additional sources of income. Both qualitative and quantitative data indicate that a fair amount of poor households rely on internal migration and migration to other countries as a way to seek better income or to cope with a shock. Evidence suggests that migration to other countries has been rising in East Asia, including in Cambodia. The importance of external labor migration is particularly evident in border provinces such as Battambang and Banteay Meanchey, where jobs across the border in Thailand have become the primary source of income for many households. In addition, because of limited opportunities in rural areas, internal migration to Phnom Penh and other urban areas has become common. Pulling children out of school is another common coping mechanism used in times of financial difficulty. Extensive child involvement in work is a key factor behind Cambodia’s education challenges of late school entry and substantial dropout rates starting in upper primary school. For every three out of four working children, family poverty or the need to supplement family income is given as the primary motive for pulling children out of school to work. Almost 90 percent of economically active children work as unpaid family laborers but, for those children who report cash or in-kind earnings, their earnings are significant to their families. Children earn an average amount of $1 per day, accounting for 28 percent of total household labor income – a major opportunity cost barrier to the schooling of poor children. Empirical analysis shows that child work has a detrimental effect on learning achievements and can also affect health outcomes if children are engaged in hazardous work (UCW, 2006). Child labor is a common coping mechanism, as nearly half of Cambodian children aged 5-14 work. The 2001 Cambodian Child Labor Survey (CCLS) showed that 44.8 percent of all children aged 5-14 (1.52 million) can be considered economically active, and that 8.6 percent of working children do not attend school (NIS, 2001). The RGC in its National Plan of Action on the Elimination of the Worst Forms of Child Labor (NPA-WFCL) (February 2004) identified 16 hazardous forms of child labor in the country, along with three unconditional worst forms of child labor (Table 9). Within working children, worrying cases of child labor are those who are under the age of 12 years (750,000) prescribed for 58 Towards a Social Protection Strategy for the Poor and Vulnerable | Profile of Poverty and Vulnerability “light work,� and those who are working in hazardous jobs or illicit activities as described in Table 9 (313,000) (UCW, 2009). Table 9: Worst forms of child labor Worst forms of child labor identified in Cambodia Child laborers (10-17) Hazardous forms of child Portering 3,234 labor Domestic worker (private home) 18,257 Waste scavenging or rubbish picking 1,664 Fishing 89,080 Processing sea products Work in rubber plantations 7,602 Work in tobacco plantations Work in other semi-industrial agriculture plantations 67,552 Brick making Handicrafts and related enterprises 43,407 Salt production and related enterprises 18,868 Stone and granite breaking Rock/sand quarrying, stone collection Gem and coal mining Restaurant work 53,951 Begging Domestic and restaurant work15 9,649 Total 313,264 Unconditional worst Child commercial sexual exploitation n/a forms of child labor Child trafficking n/a Children used in drug production, sales and trafficking n/a Source: UCW (2009). Child labor is found mainly in agricultural activities. The agricultural sector employs 79.5 percent of all working children. Boys are more likely to work in the agricultural sector than girls (78.9 percent vs. 74 percent), whereas more girls can be found in service-related jobs (20 percent of working girls vs. 15.6 percent of working boys) (ILO-IPEC, 2003). For instance, underage domestic workers in Phnom Penh are mainly girls between 7 and 17 years old (16,380 girls vs. 11,569 boys in 2003), according to the Child Domestic Worker Survey (CDWS) 2003 (NIS, 2004b). Although most children who work combine these activities with their studies, early entrance to the labor force has serious implications for human development. Child labor may result in early dropout from school and negative performance in academic activities, negatively affecting labor productivity in the future. The agricultural nature of child labor in Cambodia poses an additional risk in this respect, as extra hands are needed during the harvesting season and children may skip school to meet this demand. The disruption of the academic cycle, along with physical and psychological hazards, makes child labor a problem with long-term social and economic consequences. 15 These children are working part time as domestic workers and part time in a restaurant. 59 Sexual exploitation is one of the worst forms of child labor existing in Cambodia, but it goes beyond the age threshold. About 30 percent of sex workers in Cambodia are forced. The starting point for 38 percent of female sex workers is the sale of their virginity, with most of them underage at the time (MoWA, 2008). Touristic urban centers like Phnom Penh, Sihanoukville and Siem Reap, along with border towns like Poipet, accommodate the majority of sex workers in Cambodia and serve as bridges for human trafficking in and out of the country (UCW, 2006). Reports also suggest that, when families are poor and have no assets left to sell, a growing number of them sell their daughters (and sometimes their sons) to be trafficked within and outside Cambodia. Thailand and Cambodia are the main centers of human trafficking in the region. Cambodia hosted at least 3,000 Vietnamese victims of trafficking during the 1990s for prostitution (ILO-IPEC, 1998). Victims of trafficking are lured on false pretenses by agents purporting to recruit workers into the garment industry, construction or agriculture. Cambodian women and children are trafficked primarily to Thailand and Malaysia for labor and commercial sexual exploitation, and children are trafficked to Thailand to work as street beggars. In the 2005 Trafficking in Persons Report of the US Department of State, Cambodia remained in the lowest-ranking Tier III group, which represents countries that have “significant problems with regard to human trafficking, have not complied with minimum standards to combat these problems and are not making significant attempts to do so.� Some groups of workers who have income to pool participate in informal savings associations. Individuals pool their resources and lend to members when needed, for example to cover medical care or costs associated with getting married. Informal savings associations operate on the basis of common practice and without any collateral, relying solely on the trust and goodwill of the participants. Although members have in some cases absconded with the pool of money, informal savings associations continue to be used because they provide a flexible means of accumulating savings and accessing credit that is not available through formal channels. A qualitative study documented the use of tontine as a common risk mitigation instrument among garment workers, beer and cigarette sellers and day laborers working in the fishery industry in Thailand. Informal savings associations are also used by Cambodian immigrant communities abroad as a risk management instrument and to finance personal or business needs. 60 Towards a Social Protection Strategy for the Poor and Vulnerable | Profile of Poverty and Vulnerability Box 5: Household coping strategies in response to the 2009 global economic slowdown A preliminary assessment of the impact of the global financial crisis in Cambodia provides some insights into households’ working decisions and coping strategies. In Cambodia, the crisis has initially manifested itself through job losses in the formal sector, reduction of hours worked in the main job and a decline in hourly wages. In response to the crisis, households increased their supply of labor: male workers shifted from wage jobs back to self-employment and women in unpaid family jobs had to move into self-employment activities. As more workers, especially women, joined the labor market, overall labor force participation rates increased by 4.6 percent with respect to 2008 (see Table below). As was observed in many countries during the 1997/98 financial crisis, the agricultural sector absorbed laid-off workers and new entrants in the labor market. Therefore, in 2009, the share of the sector in total employment climbed back to 60.1 percent, from 52 percent in 2007. Many workers coped with the crisis by finding a second job and/or working longer hours to offset loss of earnings resulting from a reduction in hours worked in their primary job. The share of multiple job- holding workers more than doubled (from 18 percent in 2007 to 40 percent in 2009), and total number of hours worked in all jobs rose from 42 hours in 2007 to 48 in 2009. Households also withdrew their children from school, so that between 2008 and 2009 school attendance rates dropped by 7 percentage points for children aged 6-11 years and by 4.6 percentage points for teenagers aged 12-17 years. Correspondingly to a decline in school attendance, child labor rates increased by 3.1 percentage points. There are also indications that households sold some durable goods. Some effects of the crisis on labor market and education 2007 2008 2009 2007-2009 2008-2009 Labor force participation rate 79.1% 77.0% 81.6% 2.5% 4.6% Average number of hours worked per week in the first 39.2 38.8 36.8 -2.4 -2.0 job Average number of hours worked per week in all jobs 42.2 42.3 48.2 6.0 5.9 Average hourly wage, first job (in 2009 Riel) 1,731 1,698 1,000 -42.2% -41.1% Share of waged workers 26.1% 29.9% 26.2% 0.1% -3.7% Share of self-employed workers 39.8% 38.8% 50.7% 10.9% 11.9% Share of unpaid family workers 34.1% 31.2% 23.0% -11.1% -8.2% Share of multiple job-holding workers 16.8% 18.2% 39.7% 22.9% 21.5% Share of workers in agricultural sector 55.8% 51.6% 60.1% 4.3% 8.5% Share of workers in non-agricultural sectors 44.2% 48.4% 39.9% -4.3% -8.5% School attendance rate (ages 6-11) 83.9% 88.2% 81.2% -2.7% -7.0% Child labor rate (ages 5-14) 27.0% 27.3% 30.4% 3.4% 3.1% School attendance rate (ages 12-17) 77.2% 73.7% 69.1% -8.1% -4.6% Source: CSES 2007, 2008 and 2009 (NIS, 2007; 2008a; 2009). Note: The analysis was conducted based on the first six-month sample of the CSESs. One important caveat to keep in mind is that the full impact of the global economic crisis will be more accurately gauged with the annual sample, once the 2009 survey is fully released. While coping strategies such as building up assets and savings are important and should be encouraged, they also face limitations. For instance, households can draw only on the amount of money they have saved or borrowed. These mechanisms therefore do not offer any protection against major risks that exceed an individual’s ability to pay (e.g. hospitalization). When shocks are repeated over time, the household may also deplete its assets as a response to the first shock and have no assets left when the second shock occurs. For many poor households, which possess few or no assets and live “from hand to mouth,� setting aside part of their earnings in order to build up assets is very difficult. For better-off households, the absence of a safety net may discourage “prudent risk taking� that may stimulate investment and economic growth. 61 3. Social Protection Schemes Targeting the Poor and Vulnerable in Cambodia The social risk management framework, which links sources of vulnerability to interventions, is a useful instrument to identify gaps in coverage. The social risk management framework (Holzmann and Jorgensen, 2000) analyzes social protection issues by type of risk and the three strategies available to deal with risk: prevention strategies taken in advance of a shock, which have the objective of reducing the likelihood of the shock occurring; mitigation strategies taken in advance of a shock, which have the objective of reducing the impact of the shock; and coping strategies (described in Section 2), which include both common and more extreme ways in which individuals and households deal with minor shocks and more devastating crises (Table 10). Across these three strategies, the framework reviews three levels of risk management: informal household or community-based arrangements; public sector and civil society support; and market-based arrangements such as private insurance. This expanded view of social protection emphasizes the double role of risk management instruments in protecting basic livelihoods as well as promoting risk taking. Because of its focus on the poor, this section will mainly review mutual help and public support. Table 10: Current social protection in Cambodia against the social risk management framework Prevention Mitigation Coping Informal arrangements • Diversification of • Agricultural techniques • Pulling children out of school/ income adapted to local weather into child labor • Mutual and self-help groups • Migration • Building up assets • Borrowing money Public State • Vaccination/ • Weather-proof agricultural • Emergency food distribution Support nutrition programs techniques • Public works • Social safety nets • Social security, social safety • HEFs, CBHI Civil society nets, universal benefits Market-based • Savings arrangements • Insurance In Cambodia, as in all societies (especially those without an extensive governmental social protection program), social practices provide an element of informal social protection. Individuals and households may be able to draw on help from relatives, neighbors or other social contacts to manage risk; smooth out highs and lows in household income and consumption; and obtain help in times of crisis (Kim, 2001; McAndrew, 1998). A national social protection strategy should reflect an understanding of these informal social protection practices of households and communities, and policies and programs should seek to complement the positive effects of these practices, or at least avoid undermining them (Morduch and Sharma, 2002; Norton et al., 2001). However, informal social protection practices are eroding and can give only limited support to households in distress. As this section shows, traditional practices, rooted in trust that existed within small, stable rural communities, have been eroded over time by the effects of the civil war, revolution and displacement in the 1970s and 1980s. They have also been transformed by the spread of market economics since the early 1990s (Krishnamurthy, 1999), which has improved living standards but has also increased population mobility, intra-community inequalities and commoditization of inter-household exchanges previously conducted on the basis of reciprocity (e.g. allowing others to cultivate a portion of one’s land but now for a fee). 64 Towards a Social Protection Strategy for the Poor and Vulnerable | Social Protection Schemes Targeting the Poor and Vulnerable in Cambodia Comparative research also suggests that informal safety net practices start to fail in the face of an economy-wide crisis. When everyone’s incomes and living standards fall at the same time, it becomes harder for households to help each other out. The price and financial crises are compounding a rise in vulnerability driven by more long-term processes of increasing land concentration and decreasing access to common property resources. In this context, households are finding it more difficult to extend assistance to one another and are rationing this assistance (CDRI, 2010). Box 6: Poverty in the absence and presence of safety nets Ms Meas Sombou, a 43-year-old rice field worker and head of a household with four children in Peil Heil village, has been divorced for almost six years. She used to be better-off but became poor after her husband left her to marry a new wife in another village. Owing to a lack of labor and draft animals, and with four dependent children (three girls and one boy) to feed and bring up, she decided to sell her 1 hectare of rice land for about $3,000. She wanted to use this to repay her loans, which resulted from the long conflict with her husband before the divorce and from the limited income from farming, and to buy food for her children. She became landless then, and dependent on selling her labor to transplant and harvest rice, to earn around 8,000-10,000 Riel per day. Her income in the past six months has decreased by about half owing to the decline in local demand for farm labor, which has happened because many farmers’ sons and daughters have returned home from Phnom Penh to help their parents in farming after losing their jobs. In the past seven months, she has received some assistance from neighbors and the Cambodian Red Cross (CRC). A rich neighbor is temporarily allowing her to use a small plot of land free of charge to cultivate vegetables for some extra income and food for consumption. She also often receives food and money when she helps her neighbors organize weddings or religious ceremonies. When her house was destroyed by a big storm in January 2009, the village chief allowed her and her children to stay under his house for two months while her house was rebuilt. Several houses were destroyed at that time and the village chief reported the cases to TV11 and the CRC. Two of her daughters (8 and 10 years old) and one son (7) were not able to go to school for a few weeks but returned to school after they received a CRC assistance package (25 kg of milled rice, 50,000 Riel, two cotton scarves, two sarongs, one plastic tent, one bottle of dishwashing soap and one packet of powdered soap) in February 2009. Before receiving CRC assistance, she also received 50 kg of milled rice through a village collection. This assistance has helped her and her children at least to survive for a few months. However, her livelihood remains in a bad condition as there is no more work for her to do in Peil Heil and neighboring villages. Source: So (2009). 65 Similarly, there are limits to the safety net functions that can be played by gifts from wealthier individuals in urban business or governmental positions to rural communities. These may provide useful short-term assistance, but the size of these flows and their effectiveness in helping households to manage risks and crises is likely to vary greatly over time and from one community to the next. This lack of predictability limits their value in helping households make long-term decisions (such as whether to keep children in school or sell assets to deal with a protracted crisis). Thus, there is a clear need for an expanded and coherent governmental safety net system to complement informal arrangements. Being able to depend on availability of predictable benefits when needed is (or should be) a key advantage of a formal, government-provided safety net system. As discussed below, a review of existing programs reveals that, in relation to the critical risks faced by Cambodian households, the ongoing set of social protection interventions tends to fall short of needs, in terms of typology of programs, their coverage and overall spending. Informal arrangements Mutual help plays an important role in Cambodia through kinship, family obligation and informal networks (Kim, 2001; McAndrew, 1998; Phlong, 2009). This might entail: assistance from family; exchange of labor (provas dei in Khmer); exchange of animals; sharecropping; sharing of household equipment; renting; informal credit arrangements; information exchange; provision of food; lending of money at no interest; and self-help initiatives (i.e. funeral associations). Such practices are shot through by Buddhist beliefs in karma (doing good things in order to achieve a better situation in the next life) and by generations of family and community ties. These systems are made up of informal networks, working on the basis of trust, cooperation, solidarity and community spirit (Kim, 2001). Recent studies have also shown an economic rationale for mutual help, in that households that help expect to be helped in times of distress. The same studies also tend to find that, on average, informal networks of mutual help do not manage to fully insure households against risks (Alderman and Paxson, 1994; Dercon and de Weerdt, 2002; Townsend, 1994). These practices were particularly strong in the pre-war period and were revitalized after the collapse of the Khmer Rouge regime to play a crucial role in helping poor and vulnerable people (Kim, 2001). People shared food, draft animals and farming equipment free of charge, based on the principles of reciprocity, community norms and obligation. When people had no food to eat they could go to better-off neighbors and/or the pagoda for help. Exchange of labor for farming and exchange of animals were also common. Overall, community spirit, solidarity and cooperation served as a strong informal safety net for the poor and vulnerable in Cambodian society for generations. Villagers and extended community members often contribute money, food and labor to help households in distress. When households face death of a family member, natural disasters such as big storms or lightning strikes, houses destroyed by fire, theft and robbery, cultural, traditional and religious values oblige community members to provide assistance (So, 2009). When a community member falls seriously sick or dies, for instance, other community members contribute money, food and labor to organize sang kahak tien, that is, inviting a monk to pray for better health or safe passage from the world, and to arrange the funeral. When community members hold engagement or wedding ceremonies, the poor 66 Towards a Social Protection Strategy for the Poor and Vulnerable | Social Protection Schemes Targeting the Poor and Vulnerable in Cambodia provide labor and receive food in return. And when the house of a poor or rich family is destroyed, people provide cash or in-kind assistance. Accordingly, So (2009) reports that between 12 and 36 percent of households report relying on support from friends and relatives in times of distress. Although mutual help constitutes an important safety net for poor households in distress, there are several limitations of these informal arrangements. First, they are not efficient in dealing with major or repetitive shocks, or with covariate shocks that affect the whole community. Second, they are not adapted to situations of emergency, as the security provided is not known in advance and often is not sufficient, leading to delays in finding urgent support. Third, informal arrangements usually solve immediate needs but often do not include risk prevention strategies that can provide a longer sense of security. Last, informal arrangements are mostly mechanisms based on selective membership and guided by a principle of balanced reciprocity, such as solidarity networks, which might function well for “insiders� but not for vulnerable households (e.g. in-migrants to communities, very poor households, ethnic minorities, the elderly and infirm, people living with chronic illness) that are excluded from membership, particularly since no counter-gift can be expected. Traditional forms of mutual help are also rapidly eroding in Cambodian society. Rural households report that, while some forms of inter-household assistance remain common (e.g. looking after another family’s children, sharing information on opportunities, pooling resources for funerals), others (providing small loans of cash or food at no interest) are increasingly extended only to close relatives or may be fully or partially commoditized. Those surveyed also report that pagoda associations are increasingly forced to target more narrowly, restricting their support only to the elderly without families (CDRI, 2010). Several reasons underlie such rapid erosion. Rapid population growth in recent decades has 67 increased livelihood competition and depleted natural assets such as fisheries and forests, as well as common lands that used to serve as the main source of livelihoods and as a coping mechanism for poor and vulnerable people in rural areas (Ballard et al., 2007; Fitzgerald and So, 2007). Recent uncontrolled implementation of privatization and mismanagement of natural resources have also had direct and indirect adverse effects on coping mechanisms as well as on community solidarity and collective actions to help the poor (Ballard et al., 2007; Fitzgerald and So, 2007; World Bank, 2006b). Such erosion is particularly marked among the poorest households, and has been heightened by the recent food and fuel price and financial crises. Focus group discussions (FGDs) held in the context of various recent studies all point out to a rapid erosion of mutual help among the poorest families (Box 7). The economic rationale behind such a trend appears to be the limited income available, which constrains the ability to help other families. FGDs also suggest that informal safety nets have been affected by the recent food and fuel price and financial crises, at a time when they are needed most. The limits of mutual help have been observed in the garment sector, for instance, where workers have been more at risk of being laid off since mid-2008 (So, 2009). Vulnerable workers confirmed that assistance to the poor and very poor from better-off neighbors, except in cases of ill- health and funerals, has changed dramatically since the mid-1990s. More and more better- off households prefer to hire labor to help them organize wedding ceremonies, rather than relying on help from others, especially in villages closed to urban centers. Box 7: Poor economic status makes it difficult to help one another FGDs held in the context of various CDRI studies tend to confirm an erosion of mutual help practices. “I will not let my relatives die of hunger … but I cannot afford to help other poor people who are not my relatives in this village.� Better-off woman in FGD group in Trapieng Prei. “Besides my son and daughter, who are living with me, who else can help me? I have to work even though I feel sick … Otherwise, we will have no food to eat … It is difficult for us now since there is not much work to do in this rainy season and, for the past few months, there is no remittance from my son who is working as a construction worker in Phnom Penh. He is still there.� 58-year-old mobile laborer. “Before, people always had some money to donate to help people in the village during any difficulties or events but now things are different. We still want to help each other but we do not have money like before … and we need to save for our family also. How can we be so kind if we are also poor?� FGD with the poor in Donn Vong. Source: So (2009). Public support – RGC and development partners Despite improving trends, the RGC budget for the social sector remains low by East Asian standards. The RGC budget for social ministries increased by one-half of a percentage point between 2004 and 2009, to 3.1 percent of GDP. The detailed budget allocation for these two chapters across all Government agencies is presented in Table 11.16 This proportion remains lower than the regional level: in East Asia, average public spending on education alone was 4.3 percent in 2004 (Figure 9). 68 Towards a Social Protection Strategy for the Poor and Vulnerable | Social Protection Schemes Targeting the Poor and Vulnerable in Cambodia Figure 9: Trends in overall social spending (% of GDP) 3.5 Health 3.0 Education 2.5 Social Affairs Labor 2.0 Women’s Affairs 1.5 TOTAL 1.0 0.5 0.0 2004 2005 2006 2007 2008 (est.) 2009 (est.) Source: Data from the MEF Budget Department (2009). GDP data from World Development Indicators online: http://data. worldbank.org/data-catalog. Given the current budget structure, it is hard to determine the level of Government spending on social protection and safety nets. The new budget structure, in use since 2007, has two chapters (64 and 65) in which there are items labeled as social interventions.17 Government expenditure on these items amounted to a total of $181 million across all Government agencies in 2008, showing a 55 percent increase since 2007. Nevertheless, given the level of aggregation in the budget, it is not possible to determine how much of this amount goes to social protection activities and how much to other types of social interventions. Furthermore, nearly all existing Government agencies report spending on these chapters, though not all are actually mandated with a social agenda.18 Most of the RGC’s explicit social protection spending currently targets public employees and formal sector workers. MoSVY, responsible for social security programs for public sector employees, and MLVT, responsible for private sector workers, spent roughly $35.7 million on social interventions in 2008, 98 percent of which was spent by MoSVY ($35.0 million). These ministries were created in 2004 from the previous Ministry of Labor, Vocational Training and Youth and the Ministry of Women’s and Veterans’ Affairs. The bulk of MoSVY spending goes towards civil service pensions and veterans’ benefits: all retired and disabled civil servants and veterans, as well as their widows and orphans, are entitled to benefits in the form of pension and compensation. MLVT, on the other hand, is responsible for the NSSF, established in 2008 and set to provide, by 2012, all private sector employees of firms with more than eight employees with: (i) employment injury coverage (employment injury 16 The RGC includes in the classification of social ministries the Ministries for Information, Culture and Fine Arts, Environment and Public Worship and Religion. For this review, a more conventional definition of social sectors is adopted, limited to MoEYS, MoH, MoSVY, MLVT and MoWA. 17 Chapter 64.5 is called “Allowances and Social Transfers� and includes: 1. Family 2. Health and birth giving 3. Death 4. Retirement 5. Demission 6. Work accidents and invalidity 7. Orphans of personnel 8. Other 9. Return attenuation. Chapter 65.7 refers to “Social Assistance,� which includes 1. Assistance for hospitalization 2. Medicaments 3. Food and supplies 4. Natural disaster 5. Research stipends 6. Local scholarships and research 7. Scholarships and research for abroad 8. Others. Chapter 65.8 refers to “Subsidies to cultural and social entities,� including subsidies 1. To communities 2. To orphan centers 3. For travel scholarships 4. For the CRC 5. For sport and culture community 6. To rehabilitation centers 7. For King’s affairs (King’s charity programs). 18 In fact, to a large extent, the observed increase in social spending between 2007 and 2008 owes to increases in spending by non-social ministries, such as MoWRAM, the Constitutional Council, the Ministry of Industry, Mines and Energy (MIME) and MAFF. As a consequence, it is hard to assess whether the observed increase corresponds to an actual increase in spending on social programs (in particular safety nets). 69 insurance was launched in November 2008 and was already (in December 2009) covering 350,000 workers from roughly 900 enterprises); (ii) health insurance; and (iii) pension coverage. With the introduction of the NSSF, social spending by MLVT increased by over 300 percent between 2007 and 2008, to roughly $668,000. As of February 2009, the NSSF had enrolled roughly 400 firms in the Employment Injury Scheme, about 250 of which are garment factories, covering a total of roughly 300,000 workers.19 Beyond these priority areas, budget constraints have made it difficult for the RGC to support even minimal direct interventions for the poor and vulnerable, and the RGC relies on development partners and NGOs to fill the gaps in reaching these groups. Table 11: Social intervention chapters by ministries and other RGC agencies (Riel ‘000s) 2007 2008 Total expenditure (central + Total Total expenditure (central + Total provincial) provincial) Chapter 64.5 65.7 65.8 64.5 65.7 65.8 (1) (2) (3) (1)+(2)+(3) (1) (2) (3) (1)+(2)+(3) I. General Administration 1.1 Royal Palace 17 19,505.60 19,522.60 157.78 49.92 19,477.20 19,684.90 2.1 National Assembly 492.6 1,824.00 300 2,616.60 3,713.90 1,824.00 300 5,837.90 3.1 Senate 1,839.80 1,485.00 3,324.80 2,229.71 1,800.00 4,029.71 4.1 Constitutional Council 62.16 10 10 82.16 433.50 12.98 10 456.48 5.1 Council of Ministers 109.91 66,383.18 66,493.09 127.24 207,317.20 207,444.44 5.2 Public Function (Civil 20.17 20.17 38.87 38.87 Service Secretariat) 7.1 Interior (Public Admin) 7,139.14 20,131.86 27,271.00 16,249.14 8,539.55 24,788.69 7.2 Provinces and 150.27 14,819.80 840 15,810.07 245.23 16,729.16 920 17,894.39 Municipalities 8.1 Assembly Relations 74.98 74.98 128.02 0 128.02 and Inspection 9.1 Foreign Affairs 40.66 92.44 133.1 44.68 82.95 127.63 and International Cooperation 10.1 Economy and Finance 16,805.59 2,501.97 2,000.00 21,307.56 30,438.87 10,624.82 41,063.69 14.1 Planning 243.14 11.97 255.12 376.30 12.01 388.31 26.1 Justice 247.64 247.64 217.77 217.77 31.1 National Audit 12.31 12.31 20.85 20.85 Authority 26.2 Supreme Court 8.87 2.8 11.67 18.01 4.19 22.20 26.3 Appeals Court 15.4 15.4 6.65 6.65 26.4 Appeals Court 0.56 0.56 0.85 0.85 II. Defense and Security 5.4 National Defense 2.4 2.4 0.00 6.1 Interior (Security) 14,433.67 8,039.96 22,473.62 22,147.27 8,089.85 30,237.12 19 Data and the description of NSSF coverage and workings were collected through personal interviews with NSSF representatives. 70 Towards a Social Protection Strategy for the Poor and Vulnerable | Social Protection Schemes Targeting the Poor and Vulnerable in Cambodia 2007 2008 Total expenditure (central + Total Total expenditure (central + Total provincial) provincial) Chapter 64.5 65.7 65.8 64.5 65.7 65.8 (1) (2) (3) (1)+(2)+(3) (1) (2) (3) (1)+(2)+(3) III. Social Administration 11.1 Information 317.6 1.2 318.8 317.84 317.84 12.1 Health 1,335.69 146,381.53 43.25 147,760.47 2,026.26 186,104.13 49.82 188,180.21 16.1 Education, Youth and 10,733.04 4,970.15 4,163.19 19,866.38 15,257.13 6,806.32 4,405.69 26,469.14 Sports (Education) 16.2 Education, Youth 100.91 250.87 57.49 409.27 122.42 229.97 49.08 401.47 and Sports (Higher Education) 16.3 Education, Youth 66.15 25.7 2,630.07 2,721.92 64.10 395.01 4,602.08 5,061.19 and Sport (Youth and Sports) 18.1 Culture and Fine Arts 393.32 35.1 428.42 462.37 28.87 491.24 19.1 Environment 213.03 80 293.03 321.54 119.44 440.98 21.1 Social Affairs 95,814.54 9,358.66 9,584.32 114,757.52 120,212.86 12,321.47 11,382.37 143,916.70 Veteran and Youth Rehabilitation 23.1 Public Worship and 176.34 1 0 177.34 157.52 0 0 157.52 Religion 24.1 Women’s Affairs 278.58 596.33 3,599.37 4,474.29 379.46 568.77 4,288.48 5,236.71 32.1 Labor and Vocational 257.12 146.65 278.72 682.5 309.96 1045.24 1387.92 2,743.12 Training IV. Economic Administration 5.3 Secretariat of Civil 31.11 6 37.11 38.61 9 47.61 Aviation 13.1 Industry, Mine and 319.61 319.61 323.92 1,140.16 1,464.08 energy 15.1 Commerce 207.09 2,800.00 3,007.09 294.77 2,800.00 3,094.77 17.1 Agriculture, Forestry 843.13 665 66.9 1,575.03 1,191.55 5,721.71 36 6,949.26 and Fishery 20.1 Rural Development 348.73 348.73 471.32 471.32 22.1 Post and 159.93 159.93 234.98 234.98 Telecommunication 25.1 Public Works and 644.97 644.97 827.49 827.49 Transportation 27.1 Tourism 120.34 120.34 167.63 167.63 28.1 Land Management, 340.27 340.27 440.35 440.35 Urban Planning and Construction 29.1 Water Resources and 179.27 179.27 256.35 4,850.01 5,106.36 Meteorology Total 154,597.04 280,621.16 43,078.92 478,297.11 220,473.07 477,226.73 46,908.64 744,608.44 Source: Data from the National Treasury Accounting Department (2009). In 2010, the NSSF will strengthen the Employment Injury Scheme by addressing several issues. Registration procedures need to be improved to avoid problems when claiming for benefits, as some workers use the same ID cards or do not possess marriage certificates. The 71 database of registered workers and enterprises needs to be updated regularly, as many workers change companies. The information technology (IT) system needs to be improved to be ready for extension to the provinces, and appropriate management information tools and procedures need to be introduced. Relationships with health care providers also need to be improved and, more generally, a great deal of effort has to be made to increase understanding of the scheme’s rules and procedures among workers, employers and those in health care structures. It is also planned to expand the scheme to new enterprises, as for the moment only 10 percent of enterprises are covered. In 2010, design of the Social Health Insurance Scheme is also planned, along with the corresponding sub-decree, although implementation will not occur immediately. Civil servants are currently entitled to old-age, invalidity, maternity and death of breadwinner benefits. These are financed through the state budget and channeled by MoSVY (old age and invalidity) and other ministries. It is planned to transform these non- contributory schemes into contributory schemes in the near future. Civil servants are not covered for employment injury and non-work-related health care. It is planned to develop two contributory and compulsory schemes to cover these contingencies. MoSVY is currently developing a social health insurance (SHI) sub-decree with the support of development partners and the strong collaboration of MoH and MEF. This will provide the general principles of the new Social Health Insurance Scheme for civil servants. An actuarial study will also be conducted in 2010 to provide various options for the concrete design of the scheme, one of which will be endorsed. The Social Health Insurance Scheme is planned to be implemented by 2012. Education There have been significant improvements in the education sector in the past five years. This is seen particularly in terms of primary net enrollment gains, the introduction of program-based budgeting and the development of a sound pro-poor policy framework. Over the past decade, the RGC has steadily pursued its policy of reallocating public finances in favor of priority sectors. As a consequence, significant progress has been made in securing increasing and predictable resource for education. In 2009, the RGC increased the locally financed education budget from 1.5 percent to 1.6 percent20 of GDP (MEF, 2010). The total budget provided by donors to education in Cambodia during the school year 2006/07 was in excess of $56 million. Since 2006, MoEYS has implemented an education-based pro-poor intervention in the form of the Scholarship for the Poor program (lower secondary scholarships and pilot of primary scholarships). With support from various development partners, in 2009 the program covered 18,684 children of lower secondary age across all provinces except Phnom Penh. Eligible children are selected through a poverty assessment mechanism. School management committees collect information on students’ living standards through a survey: this information is consolidated in a database, on the basis of which a poverty index is calculated for each child. Scholarships are awarded to the 50 percent of all students with the highest poverty index. The poorest half of that 50 percent is labeled “very poor� and the other half “medium poor.� The level of scholarship varies depending on the poverty level and the school grade. At lower secondary level, very poor children receive $60 per year and 20 2009 estimated figure. 72 Towards a Social Protection Strategy for the Poor and Vulnerable | Social Protection Schemes Targeting the Poor and Vulnerable in Cambodia medium poor children get $45. Payments are made in three installments throughout the year. An evaluation of the Scholarship for the Poor program shows that it has contributed to increasing enrollment by 20 percent (Filmer and Schady, 2009): building on this success, a pilot is currently under way to extend the program to primary schools. However, concerns are raised regarding the effectiveness of the targeting mechanism and whether additional verification criteria, including through IDPoor, would not be required in order to ensure the pro-poor orientation of scholarship-related expenditures and therefore a greater contribution towards reducing inequalities and disparities in education outcomes. Despite efforts in increasing school enrollment, dropout rates remain high and quality of education presents serious challenges, as detailed in the vulnerability analysis. Moreover, the Scholarship for the Poor program does not have national coverage, and at the moment targets only secondary school children – while dropout rates continue to be a challenge even in primary schools. The pilot to extend to primary schools is in operation, but sources of financing for a large scale-up remain to be secured. Health In 2009, the RGC health budget represented 6.5 percent ($128 million) of national locally financed expenditure (MEF, 2010). In recognition of the shortage of adequate public health care provision, the RGC has substantially increased the budget allocation for health in the past few years. From 2007 to 2008, for instance, the budget increased by 25 percent. Total health expenditure per capita is nowadays $40, comprising expenditures from Government ($7.95) and development partners ($8.27) and out-of-pocket expenditures ($24). Currently, household out-of-pocket expenditures still outweigh Government and donor expenditures. Health spending is a significant source of indebtedness for the poor. However, progress has been made to address health expenditure issues through social protection 73 mechanisms (contributory and non-contributory). There are currently several schemes being developed or fully operational (Table 12). The policy framework for social health protection has been developed through the 2005 Master Plan for Social Health Insurance, the 2009 Health Financing Strategic Framework and the 2010 Master Plan on Social Health Protection (pending endorsement), and aims to develop and implement a sustainable national system for social health protection. An inter-ministerial SHI committee has been established between MoH, MLVT, MoSVY, MoP, MEF and the Council of Ministers. Table 12: Coverage of social health protection for the population % of population Coverage of social health protection by wealth and mechanisms 5 Wealthy Public health care with user fee schemes SHI coverage Complementary private coverage 10 Formal sector Public health care with user fee schemes SHI coverage 50 Informal sector Public health care with user fee schemes CBHI coverage Gradually moving to social health insurance coverage 35 Poor Public health care with user fee schemes HEFs and other subsidies Source: Annear (2008), in Sok (nd). User fee exemption policies backed by HEFs are therefore used to improve access to health care for the poor (UNICEF, 2008). The HEF is a mechanism to reimburse health facilities for treating patients who are classified as too poor to pay. The aim is to provide poor 74 Towards a Social Protection Strategy for the Poor and Vulnerable | Social Protection Schemes Targeting the Poor and Vulnerable in Cambodia people with access to appropriate health services and to protect them against health-related impoverishment. HEFs were brought into the national framework developed in 2003 and introduced into the Health Sector Support Project from 2004 to 2008, following through into the second project for 2009-2013. HEFs work closely with MoP’s IDPoor program (see Box 9 below). Poor individuals, whether identified through the IDPoor process (pre- ID) or through an in situ assessment of poverty level by medical staff or CSOs (post-ID), are covered for all costs of hospitalization and other medical services and, in some cases, expenses for transportation and food during their stay in the hospital. Income from the HEF is retained at the health facility and used to support service improvements and staff incentives. After original piloting, HEFs were introduced using four models, involving different implementers and operators, benefits packages and financing mechanisms. In 2008, HEFs were functional in 50 out of the 77 Operational Districts (ODs) in Cambodia, covering (in principle) about 3,168,883 people or about 68 percent of the poor population in Cambodia (although actual use by the poor is lower) (MoH, 2009). HEFs are managed through cooperative arrangements between the RGC, development partners and local NGOs and financed by a combination of a regular Government budget and additional support from development partners ($6.6 million in 2009 (HEF Strategic Meeting, 2010)). HEFs serve as an important complement to existing supply-side subsidies, which aim at lowering the cost of health care but generally do not reach the poor because of barriers to access such as unofficial demands for payment. Moreover, the poor face the costs of travel to the health facility and opportunity costs associated with time spent away from work and care, which also need to be factored into a HEF arrangement. Plans to scale up HEFs are currently being discussed. Health insurance schemes are also being developed for the near and non-poor. One drawback of HEFs is that they do not protect people who are above the poverty line but at risk of falling into poverty as a result of health care costs. One possible way to address this is the implementation of SHI schemes. In 2005, an inter-ministerial SHI committee was established between MoH, MLVT, MoSVY, MoP, MEF and the Council of Ministers. This oversees the development of both compulsory and voluntary SHI. As described above, MoSVY is currently developing a scheme for civil servants. Another scheme will be designed in 2010 for private sector employees (by MLVT through the NSSF). Coverage of all civil servants and all private sector employees through SHI is unlikely to be achieved in the near term, however. In addition to compulsory SHI, voluntary CBHI schemes have been mainstreamed and operational under common guidelines since 2006. CBHI targets the near poor who can afford to pay a minimal amount for premium contribution in exchange for a defined health care benefit package. In 2008, there were 12 schemes operated by local and international NGOs in the country, with about 79,873 members. Total premiums came to $183,238, with expenditures for direct medical benefits $169,729 and non-medical benefits $132,733 (totaling $302,462) (MoH, 2009). In 2009, there were 13 schemes with 122,829 members. MoH is considering scaling up CBHI but there are no costed plans for this expansion yet. Although contributory and non-contributory social health protection schemes have been an increasing presence in the Cambodian health sector, several key gaps remain. The geographic coverage of HEFs is relatively good but CBHI coverage is limited. Both of these mechanisms utilize numerous operators and implementers and are funded from various 75 sources, so coordination of this process is vital to their success. Consumer engagement and service accountability require strengthening if these schemes are to reach their full potential. Beyond these implementation gaps, there are a range of significant challenges within the health sector that require addressing to enable social health protection and social safety nets to achieve their targets. These include supply bottlenecks, weaknesses in key institutional areas and constrained community-level engagement. Finally, harmonization between RGC and development partners’ interventions remains a challenge. Other safety nets and programs Apart from education and health interventions, safety net programs in Cambodia consist mainly of food distribution, school feeding, take-home rations, food for work and, more recently, cash for work public works programs. In October 2008, MEF launched the $40 million Emergency Food Assistance Project to mitigate the effects of the increase in food prices on poor households. The emergency package consisted of a $12.5 million grant and a $17.5 million concessional loan from the ADB. Approximately $19 million is intended for social protection measures (including rice distribution, school feeding and public works), with the remaining allocated to measures aimed at increasing agricultural productivity (including subsidized seed and fertilizer), designing a food security system and strengthening RGC capacity in implementing food security operations. In October to November 2008, roughly 342,000 beneficiaries in 200 communes received in-kind assistance (35 kg of rice rations) in seven provinces surrounding the Tonle Sap Lake. MEF is now coordinating with MRD, MoWRAM and the WFP to implement food for work and cash for work programs to help poor households in food-insecure areas cope with the lean season. The WFP remains among the largest implementers of safety net programs. In 2009, the WFP distributed 27,887 metric tons of food to 988,659 beneficiaries, at a total value of $23.36 million for food and other support programs (Table 13). Through its food for work program in 2009 the WFP, working with MRD, provided roughly 3,726 metric tons of rice to 106,923 vulnerable beneficiaries in identified food-insecure communes from the most 76 Towards a Social Protection Strategy for the Poor and Vulnerable | Social Protection Schemes Targeting the Poor and Vulnerable in Cambodia food-insecure provinces, in exchange for work on public infrastructure projects, helping to mitigate critical food shortages among vulnerable households. The WFP also implements a nutritional program – the Mother and Child Health (MCH) program. Targeted at food- insecure areas, the MCH program seeks to reduce under-nutrition among pregnant and lactating women and children 0-24 months of age by integrating micronutrient-fortified food, nutrition education and other health interventions provided through local health clinics. In 2008, the program distributed a total of 4,129 metric tons of food in cooperation with the National Maternal Child Health Center and NGO cooperating partners, such as the Reproductive and Child Health Alliance (RACHA) and World Vision, which handle distribution, as well as health centers, which provide basic health education during distribution. The school feeding program, implemented by local Government authorities and school committees, provides a daily breakfast for students throughout the year. This is an incentive for children from poor families to enroll in school, attend class regularly and complete their education. It also helps mitigate short-term hunger and improve concentration. Other WFP programs include food distribution to TB and HIV/AIDS patients. TB patients receive food as an incentive to attend treatment and adhere to the national TB program. Food assistance to people living with HIV is part of home-based care, which stabilizes household food access, diminishes the proportion of income spent on food, helps improve health and reduces negative coping strategies (Table 13). An ongoing evaluation is attempting to improve the design of school feeding and scholarships, based on international best practices. Table 13: WFP food distribution and beneficiaries, 2009 Activity No. beneficiaries Food distribution (mt) Value ($) Total Female Rice Fish Veg., Salt Bean Corn Sugar Total oil soya blend Basic Education School feeding 532,186 255,669 5,330.74 866.66 431.61 122.36 963.37 - - 7,714.75 6,467,734.46 Take-home 86,925 43,464 2,443.69 159.68 297.81 - - 2,901.18 2,432,235.78 Rations Vocational - - - - - - - - - - - training Sub-total 619,111 299,133 7,774.44 866.66 591.30 122.36 1,261.18 - - 10,615.93 8,899,970.24 Health & Nutrition TB 38,268 19,500 3,334.74 - 83.21 50.29 - - - 3,468.24 2,907,635.36 HIV/AIDS 75,879 41,211 4,877.72 - 164.52 71.45 - - - 5,113.70 4,287,118.18 MCH 64,273 48,379 1,646.15 - 108.95 - - 2,110.20 263.77 4,129.07 3,443,188.51 Sub-total 178,420 109,090 9,858.61 - 356.68 121.74 - 2,110.20 263.77 12,711.01 10,637,942.06 Disaster Risk Reduction Food for work 106,923 53,461 3,726.80 - - - - - - 3,726.80 3,124,400.05 Food for training - - - - - - - - - - - Relief 84,205 42,103 833.00 - - - - - - 833.00 698,353.88 Sub-total 191,128 95,564 4,559.80 - - - - - - 4,559.80 3,822,753.93 Total 988,659 503,787 22,192.85 866.66 947.98 244.10 1,261.18 2,110.20 263.77 27,886.74 23,360,666.22 Source: WFP (2009b). 77 UNICEF is also active in supporting the RGC in implementing safety net programs across various lines of social protection, covering education, health (including HEFs), nutrition and child protection. Since 2006, UNICEF has invested roughly $16 million per year on strengthening basic social service provision and planned to invest an additional $13 million in each of 2009 and 2010. UNICEF supports the RGC by providing funds and developing public capacity to define and implement its agenda. It works closely with the RGC at both national and sub-national levels, including support to provincial, district and commune structures in six priority provinces (Kampong Speu, Kampong Thom, Otdar Meanchey, Prey Veng, Stung Treng and Svay Rieng). Social protection assessment through the decentralization structures, especially Commune Councils and Commune Committees for Women and Children, is a feature of UNICEF’s engagement with strengthening social protection provision at sub-national level. Moreover, UNICEF experiments with targeting mechanisms, towards improving identification and response to social protection needs of families with children, cross-referral and also linkages between service providers, in order to ensure a continuous stream of services to children and families with children. Table 14: UNICEF safety nets budget, 2006-2010 (US$ ‘000s) 2006 2007 2008 2009 2010 Total Child Survival 3,624 3,624 3,624 3,624 3,624 18,120 Seth Koma (Child Rights) 3,260 3,260 3,260 3,260 3,260 16,300 Expanded Basic Education 6,100 6,700 5,450 4,000 3,450 25,700 Child Protection 3,380 3,230 3,030 3,030 2,930 15,600 Total 16,364 16,814 15,364 13,914 13,264 75,720 Source: UNICEF and RGC (2006). The RGC also relies on development partners, mainly FAO, UNICEF, WHO and WFP, for support to nutrition. The WHO and UNICEF provide significant technical assistance to MoH, and more specifically the National Maternal and Child Health Center and the National Nutrition Program, in the area of nutrition-specific policies and guidelines, integrated nutrition monitoring and capacity building of health staff at various levels in the area of infant and child feeding (i.e. early and exclusive breastfeeding, appropriate complementary feeding), micronutrient supplementation and management of malnutrition. Additionally, UNICEF works to support the establishment of national community-based systems for the delivery of nutrition-specific messages, services and food commodities and for the development of improved practices in relation to nutrition practices at community, family and individual level. Despite these efforts and the progress already achieved, malnutrition levels in Cambodia remain high and much work remains to be done in this area. Over the past eight years, the ILO has channeled an amount of nearly $15 million in development partner funds in concerted efforts to eliminate child labor and trafficking. These efforts in many cases take the form of social safety nets, such as direct support interventions to prevent children from joining the worst forms of child labor and to remove child laborers from work through education services, such as provision of counseling and tutorials, referral to non-formal education centers and in-kind transfers (e.g. uniforms and stationary) for target children. The poorest of the poor families of these children have also been given training and support to enhance household incomes, which in turn reduces household dependence on children’s income and keeps children in school more sustainably. 78 Towards a Social Protection Strategy for the Poor and Vulnerable | Social Protection Schemes Targeting the Poor and Vulnerable in Cambodia Cambodia has a tradition of public works to assist the poor with under- and unemployment and improve local infrastructure, but knowledge remains low since programs close when funding ends. In addition to the WFP’s food for work program and the ADB’s cash for work under the Emergency Food Assistance Project, the RGC has endorsed labor-based appropriate technologies (LBAT) to generate employment through improvement and maintenance of essential rural infrastructure. The ILO and the ADB have been the main development partners providing assistance to the RGC in implementing the Mainstreaming Labor-Based Road Maintenance to the National Road Network Program. Between 2006 and 2008, the ILO channeled nearly $4.8 million to this, roughly half from the RGC and half from the Japanese Fund for Poverty Reduction (JFPR), to this project. Between 2007 and 2008, the ADB provided $690,398. Despite a tradition of public works, however, program knowledge remains low, since programs end with available funding, impacting the pro-poor effectiveness of public works. Ways to build this know-how (for instance by supporting an agency in charge of coordinating public works across the country) are currently being explored. Public works programs need to be made women and child sensitive, in order to avoid the pitfalls of child labor and disadvantaged female presence and earnings. Mechanisms for making special vulnerable groups benefit from public works programs as participants or beneficiaries of built infrastructure could also be considered. 79 Box 8: UNICEF social protection programs The Child Survival Program helps national counterparts achieve universal coverage of low-cost and high-impact life-saving interventions to enhance child survival. The program consists of three projects: (i) the Promoting Child Survival Project, which contributes to the nationwide expansion of immunization, control of diarrheal diseases and pneumonia and malaria prevention; (ii) the Improving Maternal Health and Newborn Care Project, which promotes access to antenatal care by skilled health personnel, iron supplementation, immunization with tetanus toxoid, promotion of clean delivery and referral for emergency obstetric care, knowledge and skills for exclusive breastfeeding and vitamin A supplementation for lactating women; and (iii) the Health Behavior Change Communication Project, which promotes key health and nutrition practices for child survival and development by strengthening the capacity of national, provincial and district health staff and other relevant actors to induce positive behavior changes through a range of communication efforts. The Expanded Basic Education Program supports the RGC to independently manage a sector-wide approach (SWAp) for education and achieve the goals of Universal Primary Education (UPE). The program consists of three projects: (i) the Capacity Building for Sector-wide Education Reform and Decentralization Project, which assists national and provincial counterparts to effectively manage the SWAp with decreased reliance on external technical assistance; (ii) the Improving Equitable Access and Quality of Basic Education Project, which provides support to primary teacher training colleges in promoting child-friendly teaching and learning methods and supporting skills-based health and hygiene promotion; and (iii) the Expanded Learning Opportunities for Disadvantaged Children and Youth Project, which addresses the specific needs of disadvantaged children without access to formal preschools or basic education by supporting community or home-based preschools, a school readiness program at the beginning of Grade 1, life skills education for in- and out-of-school youth, multi-grade teaching and bilingual education for hard-to-reach and ethnic minority areas, accelerated learning for overage children and development and implementation of an inclusive education policy for children with disabilities. The Child Protection Program helps develop laws, policies and standards and raise national awareness, understanding and capacity to protect children at particular risk. The program consists of three projects: (i) the Social Protection Project, which supports the strengthening of systems to protect, care and reintegrate children without primary caregivers and children subjected to violence, abuse, trafficking and sexual exploitation; (ii) the Legal Protection Project, which strengthens legislation to protect the rights of children in need of protection and enhances government capacity and systems to develop and enforce legislation, particularly with regard to trafficking, sexual exploitation and abuse, alternative care including inter-country adoption as a last resort, children in conflict with the law and birth registration; and (iii) the Accidents, Injuries and Disabilities Project, which supports actions to prevent, care for and rehabilitate children affected by landmines and unexploded ordnance (UXO) as well as by other accidents, injuries and disabilities. The Local Governance for Child Rights Project (part of the Seth Koma Program) facilitates collaboration among local government institutions, Commune Council members and local social service providers to promote positive behavior changes in communities and improve the delivery and utilization of basic social services for children and women. Protection, care and support to children and families affected by HIV/AIDS aims to mitigate the impact of HIV on poor households. It includes a focus on provision of psycho-social support, care and access to basic services for families affected by HIV. Source: UNICEF and RGC (2006). Targeting The RGC and development partners are also making efforts to develop targeting mechanisms to channel safety net assistance in a more cost-effective manner. The most important targeting system is the IDPoor program of MoP, supported by GTZ, which creates lists of poor households in all villages covered, compiles a Database of Poor Households and issues Equity Cards to identified poor households (see Box 9). Currently, HEF operators are the largest users of IDPoor data, for provision of free or subsidized health care assistance to 80 Towards a Social Protection Strategy for the Poor and Vulnerable | Social Protection Schemes Targeting the Poor and Vulnerable in Cambodia the poor. However, there are many other current and potential users of the data. To further consolidate the IDPoor program as the national targeting system, a sub-decree has been drafted and submitted to the Council of Ministers, mandating the use of IDPoor procedures for identification of poor households and the data generated by these procedures for appropriate targeted assistance programs. Box 9: The Identification of Poor Households program The RGC and development partners are developing interventions to better target assistance to the poor. The RGC’s officially mandated targeting program in rural areas is the IDPoor program of MoP, which has received support from the German Government through GTZ. The IDPoor identification procedures are consultative and participatory, with village representatives themselves conducting interviews using a standard questionnaire with objective criteria based mainly on assets, combined with community consultation and validation. MoP coordinates and monitors the entire process, distributes data on poor households to decision makers and service providers and issues Equity Cards to poor households. During 2007 and 2008, IDPoor conducted identification of poor households in five provinces and a total of 2,109 villages, and in 2009 extended its coverage to 2,971 more villages in seven provinces. All areas covered by the IDPoor system during that period were areas where HEFs operate, and poor households identified are entitled to receive free or subsidized medical assistance at public health centers. As well as partial or full coverage by MoP of a total of 11 provinces between 2007 and 2009, in the same period HEF operators used IDPoor procedures to identify poor households in many other provinces and districts around the country, covering around 2,200 villages. Other organisations have carried out pre-identification procedures in other areas of the country, but their methodologies differ among each other and from official MoP procedures and their results are not comparable. While the IDPoor system is currently mostly used by HEFs for provision of services, many other programs and projects targeting the poor already use it or have the potential to use it (e.g. Scholarships for the Poor). Although IDPoor data are not available for all provinces, they have proven to be an effective targeting tool for emergency assistance. For instance, they have been partly used to target emergency rice distribution by the ADB-financed Emergency Food Assistance Project, whereby in October and November 2008 about 68,000 households around the Tonle Sap Lake (20 percent of poor families in 200 communes) received 35 kg of rice rations. A preliminary evaluation of IDPoor in 2008 showed a good degree of satisfaction among communities involved in the process, in particular regarding its accuracy (i.e. limited inclusion and exclusion errors), fairness and level of community participation. The IDPoor program therefore has the potential to become a national targeting system, one which will improve the effectiveness, objectivity and transparency of targeting of assistance to poor households in such fields as education, health and social protection, as well as in emergency operations. To further consolidate IDPoor as the national targeting system, a sub-decree has been drafted and submitted to the Council of Ministers mandating the use of IDPoor procedures for identification of poor households and the data generated for appropriate targeted assistance programs. At the moment, IDPoor covers only households living in rural areas and no alternative targeting system has been developed to identify poor individuals living in urban areas. Although initial analyses have shown a good degree of satisfaction with IDPoor’s accuracy, transparency, fairness and level of participation, a question mark remains as to whether, under current implementation procedures, IDPoor would not be subject to capture if more benefits were linked to it. A thorough study on implementation procedure performance, system robustness in terms of governance risks and accuracy of results is therefore being implemented to understand where further improvements could be made according to international best practices. The expansion of the IDPoor methodology would not, however, preclude other complementary and context-specific identification and targeting mechanisms from being applied where they have added value in terms of facilitating the identification of specific vulnerabilities related to women, children, etc. In fact, the combined application of the methodologies could be expected to generate better and more focused targeting outcomes. 81 This would require assessment of IDPoor towards identifying areas where complementarities would make a difference and facilitating the choice of such complementary methodologies. Public support – civil society NGOs play a significant role in assisting households in distress. Data on NGOs show that their disbursement of core funds, i.e. excluding the funds that are delegated to them by other development partners, amounted to $103 million in 2009, representing 10 percent of total aid (CRDB, 2010).21 When development partner funding to NGOs is also considered, NGOs provide or manage approximately 20 percent of all aid to Cambodia. NGO activities are focused primarily on the social sectors: a total of $95 million (93 percent) of NGO core funds in 2009 was spent on the social sectors. Figure 10 shows the ratio of development partner and NGO assistance to these sectors. Within the health sector, much of the NGO assistance goes towards primary health care and access to hospitals and clinics. In education, NGOs focus on basic education for the poor and vocational training. NGOs are also very active in providing community and social welfare services through orphanages and general assistance to vulnerable children and youth. World Education, Winrock International, Friends International and World Vision are some NGOs that have been actively implementing social safety nets that prevent child labor, withdraw and rehabilitate child laborers and prevent trafficking. Figure 10: ODA profile in 2009 – social sectors 100 Donors NGOs 80 60 40 20 0 Health Education HIV/AIDS Education Community and Emergency social welfare and food aid Source: CRDB (2010). 21 This section draws on analysis in the Cambodia Aid Effectiveness Report 2010 of data from the RGC’s ODA database. The NGO component of the database was established in 2009 and to date 293 international and national NGOs record their data. This is an improvement with regard to data gathering on NGO activities but it by no means covers all NGOs or all NGO activities and therefore still represents only a partial picture of NGO support. For example, data for the CRC, a major emergency and food aid provider, are not included. 82 Towards a Social Protection Strategy for the Poor and Vulnerable | Social Protection Schemes Targeting the Poor and Vulnerable in Cambodia The nature and focus of support for social protection have shifted over the past decade. Some development partners and NGOs work directly at district, village or commune level and not through central government. They cite lack of transparency and governance issues, in particular, as major concerns in working with central Government. In addition, the geographic focus of some agencies has shifted from nationwide initiatives to more targeted interventions for particular regions of the country. Pagodas, churches and political parties are also providing safety net assistance. Religious institutions have historically been important providers of social protection in Cambodia. Even since the pre-war period, pagodas and temples have been places not just of worship but also of care giving. The role of Buddhist temples in providing housing for orphans and the elderly is recognized but largely unstudied. Monks frequently assist those individuals who are desperate for help and lack other channels of support, often because of social exclusion. HIV-infected and AIDS-affected populations are among the target groups of the Buddhist Leadership Initiative, which has been proved successful at awareness raising and also at maternal support to affected families. The question lies in identifying strategies for mainstreaming this kind of support and learning lessons from it to inform social protection focused on targeting the HIV/AIDS population and the population at risk. Meanwhile, FGDs suggest that assistance from pagodas and churches remains limited and focuses mostly on the elderly poor (So, 2009). In contrast, participants in a FGD reviewed by So (2009) felt that local politicians tend to be more active and responsive than NGOs in terms of providing assistance. Poor and vulnerable workers in some villages received a gift kit (50 kg of milled rice, a sarong, MSG, a cotton scarf and/or 50,000-100,000 Riel per family) before the national elections. 83 Assistance from religious groups and political parties may complement formal safety nets but should not be a substitute for development of an effective formal safety net system. Assistance from the church and religious groups has historically been the only help offered to the poor, and remains crucial in many countries. However, such assistance is often delivered to members of the religious community and at the discretion of the religious leader. Therefore, although it may complement formal safety net efforts, it usually does not substitute for them. Similarly, aid from political parties tends to be one-off and highly volatile over time, and often remains related to elections. Coordination gaps in pro-poor spending of social protection A study of selected Government, development partners and NGOs shows common constraints and difficulties in building a viable and effective social protection system in Cambodia (Chan et al., 2004, in World Bank, 2006b). The study examined six local and international NGO- and donor-supported programs covering people living with disabilities, women and children and integrated community development. Successful programs share a number of features, including an emphasis on local ownership and participation – with an eye towards communities eventually assuming responsibility for activities – and decentralization of staff and resources to provincial and village levels. The study also found common challenges, from the need for funding to the lack of human and financial resource capacity among partner RGC agencies to administer or take ownership of programs. A more in-depth study of selected programs would be needed to identify successful elements and perhaps even opportunities for scaling up or replicating interventions elsewhere. Government, development partners and NGOs have had limited success in reaching the poorest and most vulnerable. Although social protection programs are targeted towards the needs of very poor, disadvantaged and vulnerable groups, many acknowledge that reaching these groups has been difficult in practice. Geographic disparities in support are reflected in the overall disbursement of external assistance. Development partner-supported programs and NGO activity (measured in dollars spent per capita) vary greatly across the country. The average amount of per capita of development partner and NGO assistance across the country is $73, but it varies from $10 to $15 per person in Kampong Cham and Svay Rieng to $161 in Krong Preah Sihanouk and $410 in Mondul Kiri. Phnom Penh receives $53 per capita (CRDB, 2010). Of this, per capita NGO assistance across the provinces ranges from around $5 (Kampot, Prey Veng, Kampong Cham, Svay Rieng) to over $20 per capita in Phnom Penh, Siem Reap and Mondul Kiri. The geographical targeting of the funding does not appear always to be well correlated with poverty rates. Some very poor provinces, such as Mondul Kiri, Pailin, Preah Vihear, Ratanak Kiri and Stung Treng receive considerable support, whereas other provinces that are among the poorest, such as Kampong Speu, Kampong Thom and Siem Reap, do not receive commensurately higher levels of support. However, support does seem to have reached the northeastern region, neglected in the past because of difficult access and the dispersed population, with increased per capita contribution in provinces such as Mondul Kiri and Stung Treng. Drawing conclusions at this aggregate level on the targeting of ODA can only go so far, however; further analysis would require better spending data; an up-to-date poverty rate by province; and investigation of provincial vulnerabilities. 84 Towards a Social Protection Strategy for the Poor and Vulnerable | Social Protection Schemes Targeting the Poor and Vulnerable in Cambodia Figure 11: Per capita ODA (2009) and poverty (2004) by province 180 Development partner per capita ($) 160 NGO per capita ($) $410 140 Poverty rate (%) 120 100 80 60 40 20 0 Kampong Speu Kampong Thom Siem Reap Mondul Kiri Pailin Preah Vihear Stung Treng Ratanak Kiri Otdar Meanchey Kracheh Kampong Chhnang Pursat Prey Veng Banteay Meanchey Kampong Cham Svay Rieng Battambang Kampot Takeo Preah Sihanouk Kep Koh Kong Kandal Phnom Penh Source: ODA data from CRDB (2010) referring to spending in 2009; population figures are from the NIS and refer to 2008; the poverty rate is calculated using the CSES 2004 (data from the later CSES are not representative at provicial level and therefore cannot be used for this analysis). 85 4. Policy Challenges Safety net interventions in Cambodia face several challenges related to implementation, institutional arrangements and financing. Given the many sources of vulnerability faced by the poor in the country, safety nets ought to be a key component of growth and poverty reduction strategies. Many challenges remain, however, in terms of ability to cover the most important risks; effectiveness and coverage of existing interventions; programs used to address vulnerability; safety net financing; and the institutional organization in place to implement programs and facilitate coordination among government agencies and donors. Coverage of main risks Table 15: Coverage of main risks (summary) Main risks and shocks Most vulnerable Progress to date in Gaps and challenges in groups response response 1. Situations of Macroeconomic, • All poor and • Public works have been • Limited coverage and emergency and financial near poor shown to be an effective coordination of existing crisis and rapidly expandable public works programs Climate, safety net instrument environmental, during crises and natural natural disasters disasters 2. Human Poor maternal and • Pregnant • Some maternal and child • Supply of maternal and development child health and women nutrition programs are in child nutrition services constraints nutrition • Small children place remains limited and of (0-5 years) • Breastfeeding practices poor quality are improving • Coverage of these services is not universal • Supply of fortified complementary food is low • Other demand-side factors (eating, feeding and care practices) are not being adequately addressed Poor access to • School age • Scholarships and school • Quality of education quality education children (6-14) feeding programs are remains poor improving attendance • Coverage of education services is variable • Coverage of scholarships and school feeding programs does not reach all poor areas Poor access to • Youth (15-24 • Establishment of • Quality of vocational quality training years) vocational training training remains poor curricula • Supply of second chance • Some programs in place program is minimal for second chance • Poor link between training education offered and employers’ needs • No certification/ accreditation system for private sector 3. Seasonal Under-and poor • All poor and • Some targeted food • Limited coverage and unemployment nutrition near poor distribution coordination of existing and income • Pregnant • School feeding public works programs insecurity women and • Public works programs are • Funding and assistance small children providing some assistance remains volatile (0-5 years) during lean season or • Families with crises greater age dependency ratio • Landless or land poor 88 Towards a Social Protection Strategy for the Poor and Vulnerable | Policy Challenges Main risks and shocks Most vulnerable Progress to date in Gaps and challenges in groups response response 4. Health shocks Ill-health, injury, • All poor and • HEFs are financing health • Quality of health care poor illness, death, near poor care for the poor in some • Coverage/access of HEFs pandemics • Pregnant areas is not universal women and small children (0-5 years) 5. Special Inability to work, • Elderly • Pensions for civil servants, • No pensions for the poor vulnerable illness • Households NSSF for private sector • Very limited assistance to groups living with • Some donor assistance to the disabled disability and people with disabilities • Limited assistance to other chronic illness special vulnerable groups A review of programs against main risks and shocks suggests that much more can be done to support children and youth development. A juxtaposition of risks and vulnerable groups with current program coverage highlights some key areas in which future support is needed (Table 15). The analysis indicates that greater support is needed in particular for children and youth, in terms of addressing malnutrition, access to early childhood care, improving education quality and attendance, preventing entry into and removal from child labor and providing second chance education for out-of-school youth. Particular attention should be given to children and youth for several reasons. First, given current demographics, children and youth represent the largest vulnerable group. At the same time, given their young age, they have little voice and capability to express their needs and vulnerabilities. Moreover, children and youth represent the next generation of citizens participating in socioeconomic life. Failing to address their vulnerabilities puts at risk future productivity growth and may also endanger social cohesion through growing inequality and social disparities. Poor households also continue to remain vulnerable to unexpected health and natural shocks and to food shortages during the lean season. Health shocks have one of the worst impacts on households, but subsidized health care coverage is not yet universal and services provided often remain of poor quality. Natural disasters also have a strong impact on poor households. Public works can be an effective safety net to help households cope with disasters, but current coverage remains very limited; often, more emphasis is put on building infrastructure than on the social protection role of public works programs. Finally, poor households remain particularly vulnerable during the lean season – a source of vulnerability that public works could also help address. If left unaddressed, these sources of vulnerability could lead poor households to adopt coping strategies with long-lasting consequences (such as selling land or pulling children out of school), sucking them into vicious cycles of increased vulnerability and destitution. Little support is also given to the disabled and the elderly poor, who rely on family support for survival. Whereas efforts are ongoing to develop a social security system for the formal sector, there is currently no allowance for the elderly poor, who must rely on family support. A similar picture holds for the disabled poor. Family support remains one of the major sources of assistance to the elderly and disabled in most low- and middle-income countries but it represents a significant burden on the poorest families. Assistance to the elderly poor could be envisaged in the context of a cash transfer program, for instance. 89 Challenges of existing interventions Table 16: Challenges of existing interventions (summary) Areas Best practice Current state Geographic • Geographic coverage is universal or • Coverage depends strongly on factors beyond coverage based on poverty status poverty, such as accessibility • Use of poverty and vulnerability • Program location is sometimes based in cities, maps and consultations to making access difficult determine priority areas • Little use of poverty and vulnerability maps to determine coverage Type of • The instrument used is appropriate • Most assistance is in kind, even in areas with vibrant assistance to address the need/vulnerability food markets identified • The potential benefits of cash transfers are yet to be • The instrument used will work in the explored and evaluated particular context Governance/ • Financial (and in-kind) flows are • Some programs do not have strong financial financial monitored closely from the source management monitoring, leading to leakages of management to final delivery to avoid leakages funds so that only part of the support reaches beneficiaries • Disbursement modalities need to be informed by beneficiary preference and should be allowed to include women empowerment mechanisms Targeting • Programs use transparent targeting • Targeting has not yet been mainstreamed into safety methodologies to ensure delivering net implementation assistance to those in need and • Even among safety net programs that make use manage the distribution of transfers of targeting, many still rely on ad hoc targeting procedures whose accuracy has not been investigated • Various (often complementary) targeting mechanisms are used depending on local needs and context Monitoring and • Monitoring and evaluation are • Formal monitoring and evaluation are lacking in most evaluation integrated into program design and programs; basic program data are often not available implementation in a user-friendly format • Monitoring and evaluation • No agency in the RGC currently has the capacity to continuously influence program perform program evaluations that could influence design policy directions • Programs have well-developed • Feedback and complaint resolution systems are feedback and complaint resolution often poorly developed systems • Disaggregation of data by age, gender, etc, is not • Participatory monitoring and common and could ensure that future programs are evaluation are preferred where local well informed by evidence elite capture is an issue • Databases are not cross-referenced and a common ID system is not yet used, which might enable convergence of programs and databases in future social protection programming Safety net • RGC co-finances safety net • Public safety net spending is among the lowest in financing development for increased East Asia ownership and sustainability • RGC contributes minimally to safety net spending • Most programs are on budget and • Most safety net programs are still treated as institutionalized “projects� • Development partner financing • Development partners and NGOs tend to finance makes increasing use of SWAps own programs with little coordination 90 Towards a Social Protection Strategy for the Poor and Vulnerable | Policy Challenges Geographic coverage of existing programs, even the largest ones, is far from being universal, and programs do not necessarily prioritize poor areas. Support is not necessarily flowing to the areas where the need is greatest. A more detailed analysis will be needed when the necessary data on vulnerability and social protection outlays become available. Such an analysis should explore the reasons for such large discrepancies and the factors that may justify such differences. However, the need for prioritization does not obviate the need to support vulnerable groups in less poor provinces. Given the extent of poverty and vulnerability in Cambodia, even in the areas where resources are concentrated, much more could be done to make better use of existing resources and help reduce vulnerability. There is a need to set priorities in resource allocation so as to be able to allocate scarce resources to some priority social protection components in line with a consistent national social protection strategy for the poor and the vulnerable. Existing interventions are largely food based. These arrangements have the advantage of directly addressing shortages and providing households with a stable flow of food. However, in building a more comprehensive safety net system, there may be the potential to complement this in-kind assistance with cash-based interventions, which would help address the whole range of vulnerabilities faced by households (such as health shocks and inability to satisfy non-food needs). In particular, it is worth exploring the potential for cash transfers, with or without conditions, to address the challenges of food insecurity and resulting child malnutrition, of poor access to education and high incidence of child labor, especially its worst forms, and of access to health care and prevention of diseases (e.g. immunization). The fact that few cash-based interventions have been piloted to date may reflect the strong 91 fiduciary challenges that Cambodia faces. Nonetheless, as the success of HEFs shows, solutions to these challenges exist. The Scholarship for the Poor program, which has raised enrollment by providing poor families with cash transfers to keep their children in school, suggests that there is unmet potential for cash-based solutions. The quality of governance and financial management arrangements varies significantly across programs. Given the limited amount of resources available from the RGC and other partners for social protection-related programs, it is important to ensure that such support reaches beneficiaries. Factual evidence suggests, however, that the quality of governance and financial management arrangements varies greatly across programs. The challenges of tackling corruption in donor-funded projects and of promoting good governance more generally have also been widely acknowledged. Targeting has not yet been mainstreamed into safety net implementation and even many safety net programs that make use of targeting still rely on ad hoc targeting procedures whose accuracy has not been investigated. Given the relatively high cost of targeting, it is important to limit the number of targeting programs on which several safety net programs could draw. A strong candidate for further development is the RGC’s IDPoor targeting system. However, although initial analyses have shown a good degree of satisfaction with IDPoor’s accuracy, transparency, fairness and level of participation, a question mark remains as to whether, under current implementation procedures, IDPoor would not be subject to capture if more benefits were linked to it. A thorough study on implementation procedure performance, system robustness in terms of governance risks and accuracy of results is therefore necessary to understand where further improvements could be made according to international best practices. Within the area of social protection, mechanisms such as greater community participation in planning and implementation of projects could help in this regard. Complementarities of targeting mechanisms should be allowed where these make sense in terms of facilitating reach to target populations. Alternative targeting mechanisms should be considered for vulnerable groups that might not be well captured by a poverty targeting mechanism such as IDPoor, for groups that are difficult to qualify through this tool given that they face multiple vulnerabilities or for at-risk groups for which social protection would play the role of a prevention mechanism rather than just a mitigation or coping one (i.e. potential food-insecure households relying heavily on agriculture). There is also a need to improve monitoring systems and to perform more rigorous evaluations of existing interventions. Few programs or institutions are actually collecting critical monitoring information beyond inputs, outputs and mere lists of beneficiaries, which makes it difficult to assess the effectiveness of programs and improve them on an ongoing basis. Even fewer are using monitoring data to improve their procedures on a continuous basis. Moreover, there are few rigorous and thorough evaluations of existing safety net interventions, making it difficult to assess how well they perform by international standards and what areas there are for improvement. Finally, feedback and complaint resolution systems – a central pillar in guaranteeing good governance, transparency and effectiveness in safety net interventions – tend to remain underdeveloped. Very few programs have evaluated the effectiveness of their feedback systems in the Cambodian context. As an underlying challenge, the budget for safety net implementation remains low, with the majority of funding provided by development partners and earmarked for interventions 92 Towards a Social Protection Strategy for the Poor and Vulnerable | Policy Challenges that are often implemented in parallel with the RGC system. The RGC budget for safety nets remains low. Moreover, most programs are implemented by development partners, prompting concerns regarding their long-term sustainability. Ways to incorporate programs led by development partners into the RGC’s own agenda should be explored as part of the process of developing a comprehensive safety net strategy. Institutional challenges Table 17: Institutional challenges (summary) Issue Current status Areas for potential improvement Vision • Safety net implementation often reflects • Safety net implementation reflects a immediate priorities shared long-term vision for addressing vulnerability and promoting equitable growth Capacity • Implementation and financial management • Sustained capacity building efforts capacity remains weak at the local level of local institutions to progressively • Program management often remains incorporate programs into local implemented by external agents (NGO) Government Coordination • In many areas, each agency/development • Increased use of SWAp to coordinate partner/NGO implements its own safety net development program with little coordination (although • Single (or few) programs addressing each coordination is increasing in HEFs) vulnerability with large coverage • Programs are of small size and often • Nomination/creation of an agency in overlap in coverage charge of coordinating, monitoring and • Lack of an agency in charge of coordinating evaluating safety net development safety net development and interventions Safety net implementation often reflects immediate priorities (such as the need to respond to the food and fuel price and financial crises) rather than a shared longer-term vision for safety net development. Frequent changes in priorities dictated by unexpected events such as the food and fuel price and financial crises hinder the establishment of a longer- term strategy for safety net development, which would eventually allow not only long-term protection of the poor and vulnerable but also more effective rapid responses. In the long run, systems need to be put into a legal framework, based on a sustainable financing/fiscal strategy, and adequately monitored to serve as a basis for the build-up or strengthening of more comprehensive and self-sustained national social protection systems. Moreover, safety net spending sometimes reflects priorities dictated by development partner interests and earmarked funding sources. This leads to relatively large sums for safety net interventions in some sectors, with other areas remaining largely uncovered. While implementation of effective safety nets requires some degree of decentralization, capacity of local Government to effectively manage large scale-programs remains limited. Local Government faces significant staffing and capacity constraints, which generates additional challenges for the effective implementation of safety nets. As a consequence, programs are often implemented in parallel with the RGC structure, failing to build capacity in local Government to gradually take over safety net management and therefore generating a vicious cycle of low local capacity and sustained parallel implementation of programs. However, with the recent emphasis on decentralization and de-concentration, and with the role of the Commune Councils as well as the District Councils growing, there is a visible effort to build capacities within local Government. 93 The RGC and development partners should also seek better coordination across interventions. The limited scale and weak coordination of most initiatives have resulted in uneven coverage, duplication of efforts and lack of sustainability and overall impact. But coordination is essential, in particular because few NGOs or development partners have the resources necessary to scale up or replicate successful projects. The decision by NGOs and development partners to work at local level (for example through village committees formed for particular projects), combined with the concentration of efforts in particular areas, has also placed a strain on the resources of some target communities. Lack of coordination also pulls the RGC in different directions and taxes its already very limited capacity. With over 500 and by some accounts 1,000 NGO and donor programs, coordination of support for social protection as well as other sectors poses a formidable challenge (World Bank, 2006b). Even in areas where there have been attempts to achieve coordination, such as with HEFs, challenges remain: these mechanisms utilize numerous operators and implementers and are funded from various sources. A major constraint for addressing these institutional challenges is a lack of the required capacity within CARD to coordinate interventions across ministries and development partners and to facilitate and monitor implementation of cross-sectoral interventions. As in many countries, safety net interventions are scattered across several ministries. MoEYS, MoH, MoSVY, MLVT and MoWA are all mandated to manage state social services for the wider population and to help protect specific vulnerable groups against risks. However, coordination of social protection interventions across ministries and development partners requires capacity building within CARD and any other body designated by the RGC to carry out this task, for instance in areas such as monitoring and evaluation. 94 5. Conclusions Given the extent of poverty and vulnerability in Cambodia, the need for social protection support is immediate and widespread. As described in Section 2, a large proportion of Cambodians live below the poverty line, particularly in rural areas. Subgroups within this population, such as children and youth, have specific needs arising from their different sources of vulnerability. In addition, the overall population is vulnerable to idiosyncratic and covariant risks that could push the non-poor into poverty and the poor even deeper into poverty. The continued use of extreme coping strategies indicates that informal safety nets remain weak. Therefore, formal social protection support is needed to help the poor and disadvantaged and to help the vulnerable better manage risks and cope with adversity and shocks. The key question is how to target better the very limited resources that are available to ensure maximum coverage and provision for the poor and vulnerable. A compelling case can be made for a large social protection sector and support in many different areas. However, the RGC and its partners face tight capacity and budget constraints across all sectors. Any spending on social protection interventions has high opportunity costs, as resources are limited and spending on social protection means fewer resources for other sectors. Even within the area of social protection, tradeoffs will be involved in determining how to channel resources. Moreover, social protection and safety net spending will have to be traded off with other spending that may not relate directly to social protection but may promote equitable growth and thus alleviate poverty in the long run, such as improving people’s skills, asset ownership and off-farm opportunities. Coherence rather than competition between sectors needs to be strengthened, as investments in social protection without adequate investment in basic social services, including health, education, water and housing, would remain ineffective. At the same time, not all social protection investments will require additional financing, since line ministries currently fund sectoral social protection interventions. Better targeting plus coherence between current sector investments in social protection, and a gap analysis-driven identification of priorities for scaling-up across the sectors, will ensure maximum coverage and provision for the poor and vulnerable. An effective social protection strategy for the poor and vulnerable requires a balance among the three objectives of relieving chronic poverty, helping the poor cope with the most severe shocks and promoting human capital development. In a low-income country such as Cambodia, which has specific subgroups that experience severe deprivation, it may be tempting to focus support only on the most destitute and vulnerable and address only the worst forms of coping strategies. However, providing pure transfers alone will have a very limited impact in the longer term and will not address the underlying causes of poverty and vulnerability. Nor will it provide the poor and vulnerable with a platform for escaping poverty. Transfers to the poor and vulnerable in low-income countries should be used not only to help the poor who are living at unacceptably low consumption levels but also to simultaneously finance investments for longer-term poverty reduction. Given the landscape of poverty and risks in Cambodia, the development of a social protection system for the poor and vulnerable should be directed towards three different but complementary purposes: (i) support the poorest and most disadvantaged who do not have the means to ensure their own well-being, for example through transfer programs; (ii) help reduce the potential impact of impoverishing risks that could trigger the use of negative coping strategies among the larger population; and (ii) support the poor who are capable of helping themselves move out of poverty, in part by building human capital and expanding opportunities. These objectives for social protection development are in line with the RGC’s strategic objectives for social protection and provide a framework within which current support can be assessed. 98 Towards a Social Protection Strategy for the Poor and Vulnerable | Conclusions The matching of main sources of vulnerability and programs being implemented (Table 18) suggests the need to scale up and harmonize existing interventions. HEFs, school feeding, scholarships and public works are already addressing the major vulnerabilities faced by the poor and all already have a strong presence in Cambodia. Some of these programs, however, such as public works, tend to be implemented by more than one development partner on an ad hoc basis without much coordination and, since they depend on specific financing sources, their medium-term sustainability is often in question. In scaling these interventions up, it will be of the utmost importance to harmonize processes and ensure more regular financing, so as to guarantee medium-term sustainability. It also suggests the presence of gaps that could be addressed by the implementation of a few new programs. While the safety net panorama already contains most of the main types of interventions suited to its income level and capacity, there seem to be a few gaps in program implementation. Table 18, for instance, highlights the need for a program that could help both relieve chronic poverty and promote human capital. Given the high proportion of out- of-school youth and the marked skills gaps, which could put a ceiling on future productivity growth, second chance programs promoting skills development for out-of-school youth should also be included in safety net development. Also, many special vulnerable groups currently benefit from no or only limited and inadequate social protection, so there is a need to identify, pilot and expand social protection mechanisms for them and to take their specific needs and concerns into consideration when designing social protection programs. Coverage of existing programs should also be reassessed and better aligned with poverty levels. Currently, the geographic implementation of safety nets follows only partially the ranking of provinces and districts along poverty and vulnerability dimensions, in part (although not only) because existing data can provide only imperfect information about the poorest and most vulnerable districts. Such a gap could be addressed through the development of updated poverty and vulnerability maps. To be precise, these maps require the use of up-to-date information from both the Census and household surveys. Given that a Census was collected in 2008, and that a large-scale, nationally representative household survey should be available in 2010, such maps could be developed soon. Developing poverty and vulnerability maps is only the first step, however – and most likely the easiest one. Once the maps have been produced, there will be a need to coordinate and redirect safety net support towards the poorest (in terms of both poverty incidence and the number of poor) and most vulnerable areas. Such a process can take a long time – in particular in an environment where similar programs are implemented in a poorly coordinated manner by various agencies, development partners and NGOs. Essential implementation features to ensure coherent and integrated social protection interventions would include cross-checking of beneficiary data, cross-referral and consideration of common IDs for beneficiaries. Addressing governance and transparency issues will also be critical to improve the effectiveness of social protection programs and to maintain development partners’ support. If the RGC wants to increase its ownership of and role in programs, particularly over the long run, it will have to address governance and transparency issues to build trust with donors and other partners. Some of these challenges (although not all) stem from limited implementation and program management capacity of local Government, which makes it difficult to effectively implement safety net programs while at the same time guaranteeing sound fiduciary and good governance procedures. It is hoped that the ongoing decentralization process will address some of these constraints (see Box 10). On the other 99 hand, much more should be done during program design to ensure that good governance and sound monitoring procedures, including beneficiary complaint mechanisms and consultation on the design of future programs, are included from the beginning. Box 10: Governance and the decentralization process The National Program on Sub-National Democratic Development (NP-SNDD) is the RGC’s 10-year strategy and implementation program through which it aims to achieve substantial progress toward its strategic vision of decentralization and de-concentration. The NP- SNDD Design Team includes key officials from the National Committee for Sub-National Democratic Development (NCDD) from member ministries and is supported by a group of national and international consultants. The purpose of the 10-year NP-SNDD is encapsulated by the RGC’s objective to develop a management system of sub-national government that will operate with transparency and accountability in order to promote local development and delivery of public services to meet the needs of citizens and contribute to poverty reduction within the respective territories. In particular, the reform of the sub-national governance system will aim to: (i) consolidate and deepen the process of democratization at the grassroots; and (ii) promote local development and poverty reduction. The reform will be guided by the following principles: • Democratic representation: Strengthen local councils which are democratically elected (either directly or indirectly) and expand their powers, responsibilities and resources. • Popular participation: Introduce systems and procedures for people’s participation in decision making at all levels of the sub-national governance system. • Public sector accountability: Strengthen the accountability of public administration at all levels and facilitate people’s oversight of administrative and financial performance. • Effectiveness: Bring providers of services closer to users and allow users to participate in the planning and monitoring of public service delivery in order to make availability of public services responsive to local needs and priorities. • Efficiency: Improve the administrative system, coordination and management capacity of the sub-national governance system to improve quality and access to public services at all levels. • Poverty focus: Enhance the capacity of integrated territorial authorities at all levels to better target public expenditures to eradicate poverty, by focusing on vulnerable groups, and to achieve the CDMGs. In formulating programs, one factor to be considered is the extent to which existing RGC mechanisms and processes can be used. Given governance concerns, NGOs and development partners have often tried to use non-government channels to deliver support. However, bypassing the RGC does not help develop longer-term capacity or ensure the sustainability of programs once NGO and donor funding has ended. To the extent possible, working within the RGC system and fostering partnerships between the RGC and NGOs/communities for implementation should be considered. In particular, over time implementation of effective and coordinated social protection for the poor and vulnerable should lead to some delegation of tasks from current implementers (mostly local NGOs) to the RGC, and development partners’ financing should increasingly be directed towards RGC programs as opposed to the implementation of parallel programs. 100 Towards a Social Protection Strategy for the Poor and Vulnerable | Conclusions Ways to move towards cash assistance should also be explored. Currently, food distribution and food for work remain the most common forms of assistance during crises and natural disasters and even during normal times, when surplus rice production has allowed distribution of locally procured food for in-kind transfers to the poor through various programs such as food for work and school feeding. In-kind assistance presents some advantages but can also carry some disadvantages. When food markets do not function well and are highly volatile, such as in cases of natural disasters or market fluctuations, in-kind assistance is the best option to ensure the basic needs of households are covered. In-kind assistance may also tend to oblige households to give stronger priority to food consumption and may be more “self-targeting.� On the other hand, it may at times (though not necessarily, in particular if food is procured at lower prices than those offered by local markets) lead to higher administrative costs and erect stronger barriers to households using assistance to address basic needs other than food, such as out-of-pocket expenditures on health. If distributed to the right people (such as mothers), there is strong international evidence that cash is used mainly to cover households’ and children’s basic needs and is rarely spent on less important goods such as alcohol and tobacco. Cash- based assistance, including vouchers, should therefore be piloted in either existing or new programs. The types of risks in cash-based assistance are different, and there must be understanding on how to design sound and transparent fiduciary procedures for cash management that are suited to the Cambodian context. Coordination and monitoring of social protection programs should be improved. Limited coordination among partners has resulted in a multitude of social protection- related activities that have not been reviewed for consistency or duplication. Improved coordination among partners, focused on country needs and social protection objectives, is a critical task. Increased coordination could significantly enhance the impact of current interventions, both in terms of reinforcing and scaling up efforts as well as with regard to eliminating duplication in order to use resources more effectively. Monitoring and evaluation of safety net programs should also be improved, at both program and aggregate level. The quality of program monitoring varies significantly across programs. Most programs track numbers of beneficiaries and quantity of in-kind assistance but few of them extend the monitoring system to track more detailed information, such as basic outcomes and service utilization. Moreover, formal program evaluations remain poor. Existing evaluations rarely go beyond surveying overall beneficiary satisfaction with the program, and questions on targeting accuracy, cost effectiveness and actual program impact (as opposed to outputs) are rarely analyzed. At the aggregate level, the picture looks even dimmer. Information reported by the RGC, development partners and NGOs in the Cambodian Rehabilitation and Development Board (CRDB) database currently does not distinguish between assistance categories of safety net spending and overall social spending. Limited fiscal space calls for prioritization of proposed options for social protection development in the short term. The proposed options in the next section could be weighted in discussing medium-term implementation of a national social protection strategy. However, short-term reforms should be prioritized by first addressing the major (uncovered) sources of vulnerability and at the same time establishing the foundations for an effective safety net system on which further developments can be built. It is particularly important that short-term priorities match available fiscal space, which consists of the RGC’s commitment to social protection spending and resources from development partners. 101 Options for the near future Table 18: Options for the near future (summary) Main risks and Progress to date in Gaps and challenges in Options for the near future shocks response response 1. Situations of • Public works have been • Limited coverage and • Harmonize public works approaches emergency and shown to be an effective coordination of existing and guarantee stable financing crisis and rapidly expandable public works programs • Establish unit in RGC in charge of safety net instrument public works for rural development during crises and natural and emergency situations disasters 2. Human • Some maternal and child • Supply of maternal and • Improve and expand nutrition development nutrition programs are in child nutrition services services, including fortified constraints place remains limited and of poor complementary feeding • Breastfeeding practices quality • Develop cash transfer program are improving • Coverage of these services targeting poor families with children is not universal • Design cash transfer programs in • Supply of fortified health and education such that complementary food is low they can eventually be harmonized/ • Other demand-side factors coordinated/merged (eating, feeding and care practices) are not being adequately addressed • Scholarships (both cash • Quality of education • Improve quality and access to and take-home rations) remains poor education and school feeding • Coverage of education • Expand programs addressing programs are improving services is variable demand side (in particular attendance • Coverage of scholarships scholarships) both in terms of • Establishment of and school feeding coverage and covering all years of vocational training programs does not reach basic education curricula all poor areas • Improve coordination of education • Some programs in place • Quality of training remains and child labor programs for second chance poor • Boost second chance programs education • Supply of second chance • Improve quality of vocational training program is minimal programs by linking training to • Poor link between training employers’ needs offered and employers’ • Develop certification/ accreditation needs system to regulate quality of training • No certification/ provided accreditation system for private sector 3. Seasonal • Some targeted food • Limited coverage and • Harmonize public works approaches unemployment distribution coordination of existing and guarantee stable financing and income • School feeding public works programs • Establish unit in RGC in charge of insecurity • Public works programs • Funding and assistance public works for rural development are providing some remains volatile and emergency situations assistance during lean • Facilitate access to public works season or crises program for special vulnerable groups and introduce design features that respond to their needs 4. Health shocks • HEFs are financing health • Quality of health care poor • Improve and expand social care for the poor in some • Coverage/access of HEFs health protection for the poor and areas is not universal vulnerable (HEFs and CBHI) 5. Special • Pensions for civil • No pensions for the poor • Identify and pilot social protection vulnerable groups servants, NSSF for • Very limited assistance to programs for the disabled and elderly private sector the disabled poor and other special vulnerable • Some donor assistance • Limited assistance to other groups to people with disabilities special vulnerable groups • Extend targeted cash transfer • Some donor assistance • Benefits and services for program to the elderly and the to tuberculosis patients special vulnerable groups disabled and people living with are not integrated • Expand food assistance to special HIV/AIDS • Mainstream social vulnerable groups, including protection programs are complementary feed, provision of not sensitive to the needs fortified food of special vulnerable groups 102 Towards a Social Protection Strategy for the Poor and Vulnerable | Conclusions Accordingly to the strategy objectives, a mix of programs should be expanded and developed to cover both chronic and transient poverty and also help promote human capital. While there is some overlap between programs addressing transient and chronic poverty, they tend to be of different natures. Safety nets for the chronically poor place a strong emphasis on solving households’ main sources of vulnerability, such as poor health, malnutrition, low education levels, children in hazardous work and food insecurity, and on building human capital to lift households out of poverty permanently. They therefore consist mostly of programs that enter into a medium- to long-term engagement with beneficiaries, to tackle the sources of vulnerability in-depth. These programs, such as (conditional) cash transfers, scholarships and training and employment programs, also tend to use targeting techniques (such as proxy means testing) that are best adapted to capture chronic poverty but less so people who have recently fallen into poverty. On the other hand, programs such as public works, which often rely on self-targeting (i.e. everybody is free to enroll albeit at some non-monetary cost, so only the neediest will apply), are much better suited to address temporary episodes of poverty deriving from global shocks, such as natural disasters or economic crises, but also from shocks at the individual level, such as unexpected unemployment or the lean season for agricultural workers. These programs are often run at a reduced scale during normal times and are rapidly expanded during crises. They tend therefore to be less effective in tackling the roots of poverty and vulnerability. Option I: Improving and expanding social health protection for the poor and vulnerable Support should be given to achieving the RGC’s existing health strategy and social health protection plan outcomes to address the main vulnerabilities of the general population related to poor health, and in particular maternal and child health and malnutrition (as resulting from inadequate provision and lack of affordability of health and nutrition services and poor health- and nutrition-related practices). The RGC has already positioned health care for the poor and vulnerable as a priority for improvement under the NSDP Update 2009-2013, the Health Strategic Plan 2008-2015 and the Social Health Protection Master Plan (in the process of adoption). The Health Strategic Plan prioritizes improving the health outcomes of the poor and vulnerable and reflects the need to help the poor cope with health shocks. The Social Health Protection Master Plan sets out the RGC’s strategic approach to providing effective and equitable access to quality health services to all Cambodians. Table 19: Existing strategic framework for social health protection The NSDP Update 2009-2013 prioritized policies and planned actions for the health sector guided by the strategic framework outlined in the Health Strategic Plan 2008-2015 Vision: Goals: “To enhance sustainable development 1. Reduce maternal, newborn and child morbidity and mortality and improve of the health sector for better health and reproductive health well-being of all Cambodian, especially 2. Reduce morbidity and mortality of HIV/AIDS, malaria, TB and other of the poor, women and children, thereby communicable diseases contributing to poverty alleviation and 3. Reduce the burden of non-communicable diseases and other health socio-economic development� problems Master Plan on Social Health Protection Vision: Goals: “To provide effective and equitable access To develop and implement a sustainable national system for social health to quality health services for all Cambodians protection that will ensure: by [2015-tbc]� 1. Social health coverage to all, priority for the poor 2. Basic package of responsive health care benefits at an affordable price 3. Accessible client-oriented quality health care services 103 Key instruments for expanding social health protection are HEFs and CBHI. CBHI schemes refer to private, non-profit pre-payment schemes that target people who can afford to pay some insurance premium. HEFs, the demand-side financing mechanism that covers user fees, are a central mechanism for protecting the poor and vulnerable. HEFs aim to prevent health shocks driving families into deeper poverty (leading to loss of livelihoods and a deterioration of nutritional status) and to promote access to essential public health services. The RGC is already prioritizing work in this area, with plans to scale up HEFs as a standard package to provide access for all poor patients nationwide. Up to now, HEFs have been run largely by development partners and NGOs, which is an unsustainable resource arrangement and has resulted in different services in different places and gaps in coverage. An ongoing dialogue is attempting to identify key issues, opportunities and risks relating to national coverage of HEFs and to agree on the main elements of a strategic approach for scale-up. A social protection strategy for the poor and vulnerable could design a strategic and coordination framework complementing and supporting arrangements for the coordination of social health protection that are being developed. The Master Plan on Social Health Protection sets out an institutional framework for social health protection, including the roles and responsibilities of a new inter-ministerial body (the Social Health Protection Committee) to provide strategic policy development and coordinated implementation of social health protection priorities. This moves HEFs and other forms of insurance under the supervision of the national health insurance agency, to increase efficiency in the system and to allow for risk pooling and cost sharing. Designing effective coordination on social protection for the poor and vulnerable will need to take into account and complement the health sector’s ongoing work in developing an institutional framework for the coordination of social protection. There is an opportunity here to identify synergies in policy priorities and strategic approaches between social protection in health and in other sectors, to ensure that social protection programs are coherent and integrated across the population and also geographically. Synergies could be found in reducing duplication, ensuring learning across sectors, building on economies of scale and capacity in targeting and monitoring and evaluation. The key actors developing the NSPS should work closely with the key actors engaged in developing social health protection to ensure an integrated approach. 104 Towards a Social Protection Strategy for the Poor and Vulnerable | Conclusions Addressing the challenges of maternal and child health and nutrition requires strengthened investments in respective services and awareness of the population to ensure adequate utilization levels. Social protection can play a role in covering the costs of access to these services; however, in the absence of such services in the first place, social protection investments might not produce equally successful outcomes. Addressing supply-side constraints is essential in ensuring that social health protection meets its desired objectives in reference to maternal and child health and nutrition. Investing in nutrition can lead to considerable social but also economic returns. In Guatemala, a study following individuals who received nutritional supplements found that, at 25-42 years old, those men who had received the supplements prior to the age of three earned on average 44 percent higher wages than those who did not (Hoddinott et al., 2008).22 Accordingly, at the Copenhagen Consensus, an initiative by means of which leading international economists and policymakers attempted to prioritize economic investment based on the highest returns, addressing micronutrient deficiency was determined as the most profitable investment among more than 30 possible fields of investment. Potential benefits in reducing malnutrition include reducing the costs of neonatal and child care, reducing the costs of adult care through reduced obesity and chronic diseases and improving the health of the next generation. However, the strongest evidence of the economic benefits of reducing malnutrition rests in improved productivity, and is thus usually linked to schooling: improved nutritional status positively impacts a child’s cognitive development and educational performance. These, in turn, are widely accepted as main determinants of wages and productivity. Preventing malnutrition needs to be prioritized during the design of instruments related to the health sector, ensuring that enough human resources are allocated to support interventions already identified. The National Nutrition Strategy (2009-2015) has identified effective interventions for reducing under-five and maternal mortality, improving children’s nutritional status and reducing vitamin and micronutrient deficiency. However, the National Nutrition Program has few and under-skilled human resources, none of whom operates at the provincial or district level. Rural communities are not always aware of outreach activities and thus may miss the distribution of key nutrition supplements. Awareness raising on good nutrition practices and outreach are typically conducted by volunteers who lack incentives and compensation for their time and efforts. A social health protection intervention could benefit from skilled nutritionists and more government staff involved in prevention and outreach for the poor and vulnerable in rural areas. Further efforts will, however, need to be made to raise the profile of nutrition within the health sector and to build on a positive coordination experience in the National Nutrition Program. Although the National Nutrition Strategy focuses on activities recommended by MoH (supplementation, feeding practices) and its operational period matches that of the Health Strategic Plan 2008-2015, there is no specific goal in the Health Strategic Plan related to nutrition. Nutrition projects are typically not national in scale, with each project tending to cover only a small proportion of the country, although RGC/development partner/NGO coordination of activities and messages may be credited with recent improvements in the breastfeeding rate. More and better coordination and harmonization will be necessary to help health facilities manage effectively at community level the provision of critical services, such as management of malnutrition, treatment of anemia and others. 22 While this result held only for men, and not for women (possibly because of labor force participation differentials), the study also showed that nutritional supplements resulted in increased schooling by 1.2 grades for women and improved reading comprehension for both sexes. 105 Option II: Harmonizing and expanding coverage of programs promoting education and child development High dropout rates and child labor are strong outcomes of vulnerability, affecting children’s human capital and the likelihood of future generations moving out of poverty. In addition to high malnutrition rates, children from poor families are very likely to drop out from school at early stages and join the workforce as child laborers. Such behavior may be rational from a poor household’s perspective, in weighing the uncertain long-term gains from education against immediate income needs. From a global perspective, however, a poorly educated labor force not only perpetuates a vicious cycle of poverty and destitution but also affects productivity and a country’s potential for growth. It is important therefore to support efforts to keep children in school and to withdraw and prevent children from entering child labor, especially its worst forms, while at the same time improving the quality of education. Programs intended to fight child labor and keep children in school should be expanded but also harmonized. Currently, several interventions attempt to fight child labor and keep children in school, the most important ones being school feeding and scholarships. ILO and MLVT implement various time- bound measures to eliminate the worst forms of child labor, such as in-kind transfers of school materials to encourage enrollment, income generation training for the poorest of the poor families of child laborers and partnerships with NGOs t o i m p l e m e n t g r a s s ro o t s interventions for children (e.g. non-formal education and community monitoring of child labor). Other stakeholders work closely with MoEYS to enhance access to schools, improve instruction quality and review the curriculum. These initiatives share similar goals but are currently implemented separately, with little coordination. In the short term, it is important to create linkages between education and child labor programs and reduce overlaps in coverage. In a second step, the effectiveness of the various programs should be rigorously evaluated, so as to understand the advantages and disadvantages of all approaches 106 Towards a Social Protection Strategy for the Poor and Vulnerable | Conclusions and to agree on a unified program design (which, ex post, may well consist in keeping some programs separated). Finally, once a harmonized approach has been agreed on and possibly piloted, financing should be sought and the program gradually expanded at national level, or at least to cover the neediest areas based on poverty map analyses. Underlying effective harmonization is the need for rigorous evaluations of the different programs, so as to be able to come up with a unified approach that best fits the national context. Evaluation of the Scholarship for the Poor program, for instance, has highlighted strong success in increasing enrollment rates but few improvements in beneficiaries’ cognitive abilities (relative to non-beneficiaries), which may be a consequence partly of the poor quality of education. Before discussing how to harmonize different approaches, similar analyses, possibly going beyond qualitative reviews, should be performed for other programs (including an analysis of the relative benefits of food vs. cash support), so as to have a solid basis of each program’s best practices to draw on. It is also important to find a balance between addressing demand constraints and improving education quality. Several factors are impacting the quality of education, and returns to education remain among the lowest in the region. Poor returns to education are usually found to affect dropout rates and child labor, as poor parents may find it more profitable to send children to work than to keep them in school. Therefore, while addressing demand constraints that prevent parents from sending their children to school, it is equally (if not more) important to improve quality of education. Option III: Piloting a cash transfer program to alleviate chronic poverty and promote human development In addition to alleviating chronic poverty, cash transfers (both conditional and unconditional) can help promote human development by addressing demand for services. A major (though far from unique) cause of low utilization of education and health services by poor households is often the monetary (opportunity) costs that this entails. Poor households may not have the necessary resources to pay for transportation to health facilities (even if care is offered for free), or for textbooks and uniforms; often, in particular if the immediate returns to education are low, child labor seems a rational decision and families may send their children to work to supplement family incomes. Some of these children may end up in hazardous working conditions, which can permanently damage their physical and moral development. Accordingly, international best practices show that providing cash support to poor household helps reduce financial barriers and improve utilization of education and health services. Cash transfers (in particular conditional ones) can also be used to provide incentives to households to change behavioral practices. The way payments are made, or conditions attached to receiving payments, can be used to provide incentives to households to change their behavioral practices. International evidence shows that attaching some form of conditions to payments can improve utilization of health care services, attendance and enrollment of children in schools, household behavior with regard to nutrition and educational attainments (see Box 11). This has been demonstrated in the Cambodian context as well: the Scholarship for the Poor program, which, by paying a scholarship conditional on attending school, is a form of CCT, was found to have significantly increased enrollment (although impact on attainments was of lower magnitude; Filmer and Schady, 2006). 107 Box 11: Impact of CCT programs CCT programs have been subject to rigorous evaluations, using experimental or quasi-experimental methodologies to identify causal impacts. On average, programs have been found to have a strong impact on human development indicators (although in health the impact varies quite significantly across programs, as it depends on additional factors such as quality of service delivery and how severe children’s health outcomes are). The Table below summarizes selected evidence presented in Fiszbein et al. (2009). All reviewed CCT programs have had an impact on school enrollment, while impact on health outcomes is more heterogeneous. Mexico’s Oportunidades has had an estimated impact on child height of approximately 1 cm, but only for children between 12 and 36 months of age. In Colombia’s Familias en Acción, Attanasio et al. (2005) found that the Z scores of covered children younger than two years of age improved by 0.16 points with the program, but there was no statistically significant impact on older children. In contrast, Brazil’s Bolsa Alimentação appears to have had a negative impact on weight for age and a borderline significant negative effect on height for age (Morris et al., 2004). Impact of CCT programs on human development indicators Program School enrollment Growth monitoring Age range Impact (%) Outcome Age range (months) Impact Chile Solidario (Chile) Ages 6-15 7.5 n/a Familias en Acción Ages 8-13 2.1 Height-for- <24 0.161 (Colombia) age Z score Ages 14-17 5.6 24-48 0.011 >48 0.012 Oportunidades (Mexico) Grades 0-5 1.9 Height (cm) 12-36 0.959 Grade 6 8.7 Change in 4-12 0.503 height (cm) Grades 7-9 0.6 12-36 1.016 36-48 –0.349 Red de Protección Ages 7-13 12.8 Height-for- <60 0.17 Social (Nicaragua) age Z score Bolsa Escola/Bolsa Ages 10-15 3 Height for- <24 –0.110 Alimentação (Brazil) age Z score 24-48 –0.190 49-83 –0.040 Source: Fiszbein et al. (2009). Note: Significant estimates are in bold. Results must be interpreted with caution owing to the methodological limits of the study. Some studies have also estimated the impact of CCTs on children’s hemoglobin levels and anemia. Gertler (2004) estimates that, in Mexico’s Oportunidades, covered children were 26 percent less likely to be anemic after the first year than children not exposed to the program were. Other studies have also found impacts on infant mortality (Barham, 2005, Mexico), infant and maternal mortality (Hernández et al., 2005, Mexico), children’s self-reported illnesses (Gertler, 2004, Mexico) and diarrhea incidence among children aged 48 months or less in rural areas (Attanasio et al., 2005, Colombia). Most evaluations suggest positive program impacts on health center visits by children. Maluccio and Flores (2005), for instance, estimate that the Red de Protección Social in Nicaragua increased by 13 percentage points the probability that a child aged 0-3 had been taken to a health center and weighed in the past six months. Attanasio et al. (2005) report even larger effects for the Familias en Acción program in Colombia: the study finds an increase of 22 percentage points in the probability that a child aged 0-1 had been taken to growth and developing monitoring, 33 percentage points for children aged 2-4 and 1.5 percentage points for children aged 4 years old and above. Only in a few countries (Chile and Mexico) does there appear not to be significant impacts of CCTs on preventive health care usage, partly because of already relatively high usage of services. 108 Towards a Social Protection Strategy for the Poor and Vulnerable | Conclusions Given the importance of behavioral practices in determining nutrition outcomes, a cash transfer program could be used to address demand constraints in fighting chronic malnutrition, the main source of vulnerability of infants and small children. CCTs have been used widely to increase the frequency of preventive visits and address malnutrition. Although effectiveness varies across programs, some have achieved good impacts. In Mexico, preventive visits increased by 16 to 18 percent in communities covered by Oportunidades, the national CCT. Similarly, growth monitoring in communities covered by CCTs increased by 20 percentage points in Honduras and by 23 to 33 percentage points in Colombia, where the program also reduced stunting among children younger than two years of age by 6.9 percent. Questions remain open on the extent to which strict enforcement of conditions is necessary to achieve such an impact, as well as on the relative importance of investments to boost demand vs. addressing supply-side constraints (i.e. quality of service delivery). However, there is little doubt that, if well implemented, cash transfers can be an effective tool to influence the behavior of poor households in key areas of human development. Poor households have higher rates of malnutrition and poorer health outcomes compared with other groups, despite the fact that levels of malnutrition remain high for all income groups. The consultation process identified the potential for piloting a cash transfer program targeted at poor pregnant mothers and poor families with young children. Such a program would fulfill the dual objective of alleviating chronic poverty and combating maternal mortality rates and malnutrition. Consultations held in October 2009 suggest that the program could be integrated into the existing national policy framework on maternal and child health and would be part of an early childhood development approach that aims to improve child health, development outcomes and readiness for school, among others. The cash transfer could be associated with soft conditions for compliance with a set of actions related to maternal and childhood health. A food transfer component, notably as a delivery mechanism for micronutrients, could also be integrated. If covering all children of five or younger proves to put excessive strain on fiscal resources, then children of two or younger should be covered as a priority. Given the prevalence of malnutrition across all the income distribution quintiles, such a program may need complementary interventions to promote behavioral changes among the rest of the population who do not qualify for the cash transfer. In particular, public information and outreach activities may be needed in addition to the CCT program targeted at the poor. Linkages to other complementary welfare services, to address other, overlapping vulnerabilities facing families with children, should also be considered in an attempt to ensure a comprehensive and integrated approach. If well implemented, such a cash transfer program could be expanded to become one of the core pillars of support to the poor and vulnerable. Once an effective cash transfer program is in place and the poor have been identified, it would be easy to expand it to cover additional poor beneficiaries, such as the disabled or the elderly poor. Such an expanded cash transfer program, covering various categories of beneficiaries, could become one of the core pillars of support to the poor and vulnerable and could substitute in the long run for some smaller interventions, thereby improving harmonization and monitoring. If proven effective, targeted cash transfer programs linked to education and health should be integrated, or at least harmonized. In the short term, cash transfers linked to education and health need to be piloted and implemented independently, so as to limit capacity requirements, enable better measurement of effectiveness and identify gaps and challenges. 109 In the medium term, however, keeping such programs separated may lead to inefficient outcomes linked to households’ double dipping, potential tensions over budget allocations and separate administrations with similar purposes. Following international best practices, it is advisable that, if successful, these separated programs be integrated into a harmonized cash transfer program that supports poor children all along their development path. Box 12: Design elements of a cash transfer program for Cambodia During a week-long consultation held by CARD and development partners in October 2009, a short Note was written on how a cash transfer program covering poor pregnant women and families with young children could be designed in the Cambodian context. A summary of the implementation features discussed in the Note follows. The program would aim to increase utilization of a pre and postnatal care package that conforms to existing national policies and international best practices. First, women of childbearing age could receive iron and folic acid supplementation, ideally prior to conception. Second, mothers could be encouraged to make up to four prenatal visits to the health facility prior to delivery to monitor pregnancy risks and to receive iron and folic acid supplementation, tetanus immunization, de-worming (Mebendazole) and treatment of malaria (if needed). Such visits would also be an opportunity for health workers to communicate the importance of placing the baby on the breast immediately after delivery and to promote exclusive breastfeeding. Third, mothers could be encouraged to make use of birth delivery and postnatal services, such as delivering at a health facility; receiving a postnatal visit in a health facility (or at home), including vitamin A, de-worming, iron and folic acid supplementation, as well as additional counseling on breastfeeding; children between 6 and 59 months receiving vitamin A twice a year; children between 12 and 59 months receiving de-worming twice a year; and children receiving full immunization, which will also provide an opportunity for health service providers to communicate best practices for weaning and complementary feeding. On registration in the program, each mother and child could be issued a card (or booklet) that is evidence of enrollment and on which a record of compliance with receiving the services listed above can be kept. For each visit or other instance of receiving a service, the card would be updated by the service provider with a form of proof of attendance (stamp, signature, hole punch). The beneficiary would then present the card during the distribution of the benefit to prove eligibility. In order to avoid moral hazard issues, the cash benefit should not be distributed in conjunction with receiving the service, or at least the service provider should not be the same agent as the payment provider. While payment of the benefit would be subject to fulfillment of the conditionality, it is advisable that the latter not be applied too strictly. The purpose of the conditions is not to move people off the register but to ensure meaningful behavioral change. For example, mothers showing sufficient evidence of behavioral compliance with the program, even if not perfect, should remain eligible: a mother who attends two or three out of four recommended prenatal visits should be kept in the program. Similarly, an inability to deliver in a clinic should not exclude a mother from participation; it may be possible to have a minimum of visits before and immediately after birth which fulfills the recommendation. However, monitoring of conditions should be subject to well-defined criteria that leave little room for interpretation to avoid differential treatment across different locations. The size of the benefit should be sufficiently large to make an impact on beneficiaries’ welfare, but not so large as to induce unwanted behavioral changes, such as leaving work in light of the cash transfer or falling pregnant to be eligible for the program. The payment should be frequent enough to allow women some liquidity to pay for transport and other expenses in order to be able to meet the conditionalities: a bi-monthly or quarterly payment would allow for this, while keeping administrative costs under control. Estimates should also take into account the monthly costs of nutritious food and food preparation per woman and child. A benefit of between $10 and $15 quarterly emerged as a reasonable amount from technical discussion, although more analysis needs to be performed based on household surveys before determining the optimal benefit level. The program could start as a pilot in to-be-defined priority areas, with the plan of scaling up coverage once implementing mechanisms are fine-tuned and once there is evidence that the program is beneficial. A gradual rollout would also provide the option of conducting an impact evaluation with a credible counterfactual; this would allow for a demonstration of the performance of the program and impacts on nutrition and maternal mortality outcomes. A decision on whether and where to pilot will require an analysis and discussion of all constraints as well as priority needs. The consultation also highlighted the need to take supply constraints and quality of service delivery into account, since the impact of the cash transfer program in areas with poor quality of service delivery is likely to be minimal. More analysis will therefore be needed to assess supply-side capacity at the moment of choosing the pilot areas and, once pilot areas are selected, on the specific constraints that are present there. Source: World Bank (2010). 110 Towards a Social Protection Strategy for the Poor and Vulnerable | Conclusions Preliminary estimations suggest that such a cash transfer program would cost around $20 million per year, or 0.21 percent of GDP (Table 20).23 At such a level, the cost of the program is in line with international comparators if assessed as a share of GDP. The flagship CCT programs in Mexico and Brazil, for instance, have an annual cost of 0.3 to 0.5 percent of GDP. However, it is worth noting that the estimation below does not reflect the full cost of administering the program, including targeting. It is worth noting, for example, that the annual cost of implementing IDPoor alone is about $2 million, which should be at least partially borne by the program. The estimation also does not include the supply-side cost of providing nutrition services. Table 20: Estimated cost of a cash transfer program Item Value Total population (2008 Census) 13,395,682 Share of population living in rural areas (2008 Census) 80.5 Population living in rural areas (2008) 10,783,524 Extreme poverty rate in rural areas (2007, %) 20.78 Number of extremely poor people (2008) 2,240,816 Composition of household – children under 5 years of age in lowest quintile (%) 11.5 Total number of extremely poor children under 5 years of age 257,694 Number of extremely poor pregnant mothers 43,302 Total number of beneficiaries 300,996 Size of benefit per child/mother ($) 15 Yearly frequency 4 Total yearly transfer per child ($) 60 Total cost of benefits only – per year ($) 18,059,752 Administration costs (%) 10 Total cost of program ($) 19,865,728 GDP (2008) ($) 9,573,000,000 Total cost of program as % of GDP 0.21 Source: World Bank (2010). Option IV: Improving the effectiveness of public works to respond to food insecurity, crises and natural disasters An important social protection instrument that has the potential to address chronic poverty and food insecurity during normal times and to respond to natural and economic crises is a nationwide public works program. Cambodia already has significant experience here, as various donor partners have been working with the RGC in financing and implementing labor-intensive public works programs. However, none of these programs has been implemented nationwide. Moreover, different public works projects at different points in time and in different locations have used diverse design features, such as those relating to the level of the wage rate and procedures for selection of projects and beneficiaries. As a result, which particular beneficiary gets how much employment and at what remuneration 23 Different scenarios of benefit size and coverage result in different cost estimates (see Note). 111 has remained random, since a common, nationwide approach to the design and targeting of the program and predictable funding has been lacking. Equally important, because of a lack of continuity no domestic capacity to run the program has been created. Public works are particularly suited to be expanded rapidly in specific areas hit by crises or natural disasters to provide emergency assistance, especially if there is a need for reconstruction. One of the main advantages of public works is that there is no need to preselect beneficiaries, in particular if the wage rate has been set slightly below market rates so that only the poorest workers will “self-select.� Such features make public works a good complement to targeted cash transfers during crises and natural disasters, since poor beneficiaries (who may have fallen into poverty as a consequence of the crisis and therefore may not correspond to the chronically poor) do not need to be identified in advance. This role of public works as an emergency safety net has not yet been fully exploited in Cambodia, owing to coordination and capacity challenges that prevent rapid expansion of existing programs. With a third of the population falling below the poverty line, public works can also be used to address chronic poverty and food insecurity, in particular during the lean season. The seasonality of the labor requirement in farming means that households are obliged to find off- farm employment in the slack agricultural season to supplement livelihoods. It is estimated that, in the wet season, 76 percent of households are involved in agricultural activities; in the dry season, this falls drastically to only 29 percent of households.24 In the dry season, therefore, rural households, especially those with little or no land, have to rely on income from unskilled wage employment in urban areas and across the border. Recognizing the relevance and extent of food insecurity in the country, the WFP’s public works program operates in food-insecure areas that are identified through vulnerability maps. Cambodia has a strong history of public works, although these are implemented by various partners and agencies that have adopted differing design features, as we have seen. For instance, village infrastructure works carried out in Battambang in 2009 with the support of ILO and ADB-supported civil work under the Emergency Food Assistance Project use contractors to build infrastructure and spend a larger share on non-labor inputs than the WFP’s food for work program and the Emergency Food Assistance Project’s cash for work program. Counterparts in the RGC also vary: the Emergency Food Assistance Project is supervised by a Central Project Management Unit under MEF and implemented by Technical Support Units under MRD and MoWRAM; the WFP partners only with MRD. These differences in design stem in part from programs with slightly different objectives all having been classified as public works. While all public works programs share the goal of generating employment and supporting the poor and the unemployed, differences in design influence the extent to which they reach the poorest of the poor and their ability to produce quality infrastructure. Public works can be classified into two broad categories: (i) infrastructure and rural development programs that use LBAT to maximize the employment impact of construction projects; and (ii) self-targeting social protection programs that have as their first priority assisting the poorest households and therefore place a lower (although non-negligible) emphasis on infrastructure. While the distinction between these two types of programs is not always clear cut, it is important to understand conceptual differences and These percentages are estimated with data from the CSES 2007 (NIS, 2007), taking as the dry season the 24 months from January to April. 112 Towards a Social Protection Strategy for the Poor and Vulnerable | Conclusions differences in design, since international evidence has shown that mixing design features can, at times, lower their effectiveness. In particular: • Rural development programs using LBAT share the dual objective of constructing much-needed rural infrastructure (such as roads, bridges, schools and irrigation canals) while at the same time maximizing income and employment opportunities for unskilled workers – though not necessarily the poorest. Because of the emphasis on building quality infrastructure, these types of programs tend to spend a high share (usually 50 to 60 percent) of project costs on non-labor inputs such as materials and machinery and technically skilled personnel. They also sometimes use contractors, although they instruct them on how to maximize employment and on the wages they have to pay. In these programs, the wage rate tends to be set competitively at levels equal to the local market rate for workers with the required qualifications. If the market wage is deemed excessively low, however, these programs sometimes offer a higher rate, so as to try to force the market wage rate up to less exploitative levels. While the wage rates that are offered tend to attract poor workers, they do not necessarily attract only the poorest. Also, the need to build large-scale quality infrastructure (such as irrigation dams) may impose some requirements in terms of the skills of the employed workforce. Therefore, if implemented at national level, these types of programs tend to have substantial fiscal implications, because of the high numbers of workers they attract. In situations where too many workers demand work and the program is heavily oversubscribed, job rationing is used to select participants, using communities to identify their poorest members or relying on other targeting systems such as IDPoor. The need to resort to job rationing poses some challenges, in particular during emergencies; although these are not insurmountable, international evidence suggests that they can affect programs’ ability to target the poorest members, given the discretion involved in selecting program beneficiaries. The ILO has extensive expertise in providing technical assistance supporting the adoption of LBAT. • Self-targeting social protection programs aiming to assist the poorest households put stronger emphasis on reaching the poorest, with infrastructure development as a subsidiary but complementary objective. Also, projects are typically small-scale community infrastructure-type projects which are of good quality but require a lower skills level to execute (e.g. dugouts rather than dams). These programs are usually designed to provide employment to anybody who is willing to work at a given wage rate, usually set below the market wage level. (In addition, there is sometimes a cap on the maximum number of days a beneficiary is allowed to work, to render the program fiscally sustainable.) The wage rate therefore becomes the targeting criterion, as only poorest households without alternative employment opportunities are willing to work at a low wage rate. These programs are usually implemented in the absence of other targeting mechanisms, or when the profile of the poor has changed substantially and there is a need to act fast, such as in the case of crises or natural disasters. They can also be implemented to complement other safety nets, for instance to reach poor households that have “fallen into the cracks.� While these programs tend to be well targeted and effective in providing emergency assistance, they also have some limitations. The main one is the need to keep wages below market rates, which limits the “generosity of benefits� and therefore the ability to fully support poor households to levels that enable them to cross the poverty line. Unfortunately, international evidence shows that raising wage rates to attractive levels at or above the market rate attracts individuals of less poor households to apply, leading either to job rationing (raising related governance issues in beneficiary selection) or to unsustainable fiscal costs. 113 A choice has to be made by the RGC as to which type of public works to implement, taking into account the development impact and the fiscal implications of each program. A priori, no alternative can be deemed “better� or “worse,� as the type of program to implement depends on program objectives (i.e. focus on infrastructure vs. social protection) and available fiscal resources. It is important that the implications of design choices be fully understood, however, and that adequate fiscal resources be provided. Independently from the chosen approach, there is a need to unify approaches and harmonize donor interventions. Currently, approaches vary significantly across projects – and sometimes even within the same project. While it is important to keep some degree of flexibility in piloting alternative approaches, it is also important to try harmonizing interventions and unifying approaches, so as to allow for a transition from a “project� approach that is financed (and often implemented) by donors to a programmatic approach that is increasingly owned by the RGC. Furthermore, harmonization and unification of approaches would also facilitate the building-up of capacity in the RGC to supervise and implement public works and, in the medium term, its ability to respond rapidly to crises. A permanent unit should therefore be set up in the RGC to be progressively involved in the implementation of public works. Currently, public works are implemented by various agencies or donors and follow the availability of funds. This prevents the RGC from building its capacity to implement public works and to react effectively to crises and natural disasters, as each time there is a need to rely on donor know-how or to set up a new implementation unit. To be effective, there should be a unit in the RGC with strong and continuous expertise in public works implementation. Potentially, the unit could be in charge of LBAT-based rural development, with the additional mandate of implementing self-targeted public works programs during emergencies. While different in spirit, both types of public works share many commonalities – for instance the need to identify community needs for infrastructure; the need for an engineering unit; and the need to set up a decentralized system for project supervision and implementation. Synergies could therefore be achieved if the same unit implementing LBAT-based rural development implements – with slight changes in design, such as the offered wage rate – self-targeted public works programs during emergencies. In particular, during “normal� times, the program would deal with the plight of the chronically poor and vulnerable, complementing other safety nets (such as targeted cash transfers) as a means for the RGC to deal with low demand for labor and acting as a significant vehicle for infrastructure development at local level. With a continuous public works program already in place, the RGC would then have an ongoing program that can be utilized to channel additional support into rural areas to assist vulnerable households facing additional challenges when a new crisis occurs. A Note has been prepared by CARD and the ILO on harmonization and optimal design of public works in Cambodia. In collaboration with relevant stakeholders in the RGC and among development partners, CARD and the ILO are preparing a Note on issues and approaches to be considered in implementing public works in Cambodia (ILO, 2010a). The Note aims to take stock of all the experience gathered in implementing public works in Cambodia and to harmonize approaches used to improve their effectiveness (see Box 13). 114 Towards a Social Protection Strategy for the Poor and Vulnerable | Conclusions Box 13: Outline of a public works program for Cambodia A harmonized public works program in Cambodia could have the following features: • Objective: To provide employment to poor and vulnerable members of the population, creating sustainable infrastructure assets that will support their economic and social development. • Coverage: Nationwide, with geographical intensity varying according to seasonality, poverty and need for infrastructure. • Overall size of the program: Depends on capacity for implementation at national and sub- national levels, extent of infrastructural needs that can be addressed by a public works program, available funding and demand for work, particularly during periods of income and food insecurity. • Impact on social protection: The public works program would contribute to the overall objective of a national social protection strategy directly by providing cash and food at times of food and income insecurity and by providing employment in areas of significant year-round underemployment. In addition, it would indirectly contribute by creating physical assets that provide economic and social benefits for rural people. Moreover, the program can be linked to other components such as interventions in health protection and education support in terms of infrastructure provision/maintenance. • Monitoring and evaluation: It is essential to justify expenditure in a public works program and to clearly assess the impact of such programs on beneficiaries. Source: ILO (2010a). 115 Elements to consider in program implementation In moving forward, several aspects related to program implementation should also be considered by the RGC’s coordinating and implementing agencies. Overall coordination, monitoring, evaluation and information management are areas in which capacity building is needed within the coordinating agency. Fiscal constraints should also be taken into account when choosing and designing social protection programs. For this reason, targeting and program evaluation should be performed in the best possible way, so as to guarantee that financial resources are spent where most needed. Table 21: Elements of implementation (summary) Elements of Progress to date in response Gaps and challenges in Options for next steps Implementation response Coordination and • CARD has been designated • Poor financial and • Design program of technical development of social to coordinate and develop human resources assistance to improve CARD’s protection social protection programs capacity in CARD to capacity to coordinate, monitor for the poor and vulnerable coordinate, monitor and and evaluate implementation • The draft National Social evaluate implementation and development of social Protection Strategy for and development protection programs the Poor and Vulnerable of social protection discussed at the June CDCF programs Financing of social • Development partners • Financing remains • Develop a single protection programs have secured financing uncoordinated and comprehensive program and of main social protection project based budget framework interventions for the poor • Explore ways to increase use and vulnerable of local systems for program design and implementation, financial management, monitoring and evaluation • Increasing RGC ownership and financial commitment over time Targeting of social • IDPoor is increasingly • Good governance • Evaluate and improve design protection programs adopted and is expected to procedures and of current targeting systems (in reach national coverage accuracy of IDPoor particular IDPoor) • Existing evaluations of have not yet been • Ensure medium-term financing IDPoor have shown good rigorously evaluated of IDPoor (for instance by degree of satisfaction, but • Financing of IDPoor is having programs contributing also some accuracy and guaranteed for the short to it) governance challenges term only • Update poverty and vulnerability maps to improve coverage of the poorest regions and households Element I: Build CARD’s capacity to effectively coordinate, monitor and evaluate implementation of social protection programs for the poor and vulnerable Social protection for the poor and vulnerable consists by nature of inter-sectoral programs involving several ministries that need to be coordinated. The very nature of interventions – ranging from food distribution to public works, targeting and cash transfers – makes them inter-sectoral. But, while each ministry should be responsible and held accountable for its own program, there is a need to coordinate interventions across ministries to avoid overlaps and improve harmonization. Issues to be coordinated range from geographic coverage (ensuring that coverage priority is given to the poorest regions or those with the highest number of poor households); to coordination of funds towards programs addressing main 116 Towards a Social Protection Strategy for the Poor and Vulnerable | Conclusions vulnerabilities, as opposed to development partners’ own priorities; to cross-checks across programs to avoid double-dipping issues; to continuing the development of social protection programs for the poor and vulnerable in collaboration with line ministries and development partners. To this end, there is a need to build capacity in CARD to perform coordination, monitoring and evaluation of social protection development activities. In the NSDP, CARD has been designated as the agency in charge of monitoring and coordinating the development of social safety net programs for the poor and vulnerable. While a few of CARD’s staff have been leading the dialogue with ministries and development partners on social protection development, CARD’s newly assigned role to coordinate, monitor and promote social protection programs for the poor and vulnerable will require financial and staffing resources that exceed those that are currently available. Potential tasks for CARD include: • Monitoring and evaluation: Improving monitoring of social protection interventions will be key in supporting any dialogue on social protection development. The minimum of monitoring activities includes keeping updated inventories of existing programs, containing geographical coverage, types and number of beneficiaries and budget. Such information should be made public so as to provide the many actors involved in social protection (RGC, development partners, civil society) with the opportunity to better evaluate, plan and coordinate interventions. In a second stage, CARD could also become involved in program evaluations to support a more informed dialogue on program coordination and harmonization following local best practices. CARD’s involvement would also provide an independent review of evaluations, currently performed mostly by programs’ implementation agencies. Evaluations should range from simple qualitative studies to understand beneficiary satisfaction and implementation challenges, to more rigorous quantitative (impact) evaluation to assess program impact on a range of development outcomes, along the lines of what has been done for the Scholarship for the Poor program. A social protection module could also be added to the CSES (although not necessarily every year) for broader evaluations of overall social protection coverage and performance. While CARD itself does not need to be involved in performing the evaluations, it is important to build capacity to understand the pros and cons of various approaches to develop a critical and independent view of program effectiveness. • Coordination: Good monitoring and evaluation data represent the basis for sound program coordination, but they are not enough. CARD should have the mandate (and financial resources) to hold regular meetings on program coordination and regularly review and promote coordination efforts across programs of a similar type (i.e. public works, cash transfers, etc), as well as across various interventions, to avoid excessive coverage of certain beneficiary categories at the expense of other groups. While CARD’s recommendations should not necessarily be binding and, to avoid any conflict of interest, CARD should not be involved in program implementation, it is important that CARD be given the moral authority for its recommendations to be heard. • Social protection development: In addition to coordination of existing interventions, CARD should have the capacity (and financial resources) to promote social protection development for the poor and vulnerable. CARD’s role should include updating strategy papers to reflect a changing environment and sources of vulnerability; promoting, coordinating and disseminating social protection analyses; and leading the inter- 117 ministerial dialogue and the dialogue with development partners on social protection development by bringing forward new ideas, holding regular meetings and workshops and supporting requests for information. Element II: Build towards sustainable financing through an appropriate investment approach A sustainable social protection program for the poor and vulnerable requires predictable and guaranteed financing. For ensured sustainability, a move away from donor-financed projects to a RGC-owned program is needed. While donor financing will continue to make an important contribution in the coming years, the work plan for operationalizing the NSPS should identify the steps to a program-based approach, defined as: (i) leadership by the host country or organization; (ii) a single comprehensive program and budget framework; (iii) a formalized process for donor coordination and harmonization of donor procedures for reporting, budgeting, financial management and procurement; and (iv) efforts to increase the use of local systems for program design and implementation, financial management, monitoring and evaluation.25 In addition, sustainable financing would also require increasing the RGC financial commitment over time: modest increases in social protection expenditure would start to bring Cambodia up to the levels of average expenditure in the region. Any expansion of existing programs or piloting of new schemes should be designed as a stepping stone towards the end goal of a full program-based approach with harmonized and aligned financing. Element III: Evaluate and improve design of current targeting systems; update poverty and vulnerability maps to improve coverage of the poorest regions and households As the IDPoor program is increasingly adopted, there is a need to further evaluate it so as to guarantee accuracy and sound governance procedures that can be put under strain as more benefits are attached to Equity Cards. Implementing good targeting is costly – hence the need to share targeting instruments across programs. For the sake of effectiveness, the Council of Ministers will therefore issue a sub-decree mandating the adoption of the IDPoor program in relevant programs targeting the poor. But the more benefits attached to Equity Cards, the bigger the risk that the process of allocating Equity Cards will be captured and biased. There is therefore a need to evaluate further the accuracy and implementation procedures of the IDPoor program and to continuously update them following international best practices, so as to guarantee a fair and transparent targeting process. There is also a need to guarantee the medium-term financing of IDPoor. Currently, MoP has secured short-term financing of IDPoor to expand coverage nationwide, but medium- term financial sustainability has not yet been achieved. One of the major challenges is that IDPoor is perceived as an “independent� program run by MoP, so interventions that make use of it (such as HEFs) do not contribute directly to it. In the future, it would be desirable to develop systems that ensure that programs using IDPoor contribute to its implementation. Since some poor households “fall into targeting cracks,� there is also a need to evaluate the effectiveness and accuracy of “post-identification� mechanisms, as well as to develop a few 25 Definition by the Development Assistance Committee of the Organisation for Economic Co-operation and Development (OECD DAC). See www.aideffectiveness.org/Tools-Definitions-AE-portal-lexicon.html under program-based approach. 118 Towards a Social Protection Strategy for the Poor and Vulnerable | Conclusions programs using alternative targeting methodologies. While post-identification mechanisms are essential, there exists at the moment no rigorous evaluation of their accuracy and good governance procedures. Moreover, while, for the sake of effectiveness, the number of targeting systems should be minimized, it is good to think about developing a few programs that select beneficiaries in ways other than the IDPoor program, such as public works through self-targeting or programs targeting specific, easily identifiable groups. In doing so, it will be important to consider the fine balance that will arise between missing some households while at the same time excessively covering others through multiple programs (e.g. providing one household with scholarships, cash transfers and school feeding). CARD’s role in coordinating interventions will therefore be crucial in achieving effective program coverage and composition. At the same time, updating of poverty and vulnerability maps should allow for improved geographical coverage by reaching the most vulnerable regions. Existing poverty and vulnerability maps that are used by some program implementers (such as the WFP) to determine program coverage are now outdated. The recent collection of Census and household survey data will allow for an update and fine-tuning of poverty and vulnerability maps that estimate incidence of poverty and vulnerability at high levels of disaggregation – i.e. commune level. With these updated maps in hand, it will be easier to prioritize coverage areas according not only to poverty but also to other vulnerabilities (i.e. food insecurity, malnutrition, health), so as to respond to specific household needs. 119 References A ADB (Asian Development Bank), ESCAP (Economic and Social Commission for Asia and the Pacific) and UNDP (United Nations Development Program (2010). Achieving the Millennium Development Goals in an Era of Global Uncertainty. Asia-Pacific Regional Report 2009/10. Bangkok: ADB, ESCAP and UNDP. ADI (Analyzing Development Issues) (2007). Understanding Social Capital in Response to Floods and Droughts: A Study of Five Villages in Two Ecological Zones of Kampong Thom Province. Phnom Penh: CCC. AFSC (American Friends Service Committee) (2000). Socio-Cultural Vulnerability and Coping Strategies Research Project. Phnom Penh: AFSC. Alderman H. and Hoddinott, J. (2007). Growth-Promoting Social Safety Nets. Washington, DC: IFPRI. Alderman, H. and Paxson, C. (1994). Do the Poor Insure? A Synthesis of the Literature on Risk and Consumption in Developing Countries. International Economics Association Moscow Meeting Proceedings Volume 4. Alderman, H., Hoddinott, J. and Kinsey, B. (2006). Long Term Consequences on Early Childhood Malnutrition. Oxford Economic Papers 58(3): 450-474. Aker, J. (2009). Cash (or Food?) for Thought: The Debate on Cash Versus Food Isn’t Over (Yet). Blog. http://blogs.cgdev.org/globaldevelopment/2009/03/cash-or-food-for-thought- the-d.php. Annear, P.L. (2008). Developing a Strategy for Social Health Protection. Presented for discussion at the National Workshop on Promoting Sustainable Strategies to Improve Access to Health Care. Phnom Penh, January. Attanasio, O., Gómez, L.C., Heredia, P. and Vera-Hernández, M. (2005). The Short-Term Impact of a Conditional Cash Subsidy on Child Health and Nutrition in Colombia. Report Summary: Familias 03. London: Institute for Fiscal Studies. B Ballard, B., Sloth, C., Wharton, D., Fitzgerald, I., Murshid, K.A.S, Hansen, K., Phim, R. and Lim, S. (2007). We Are Living with Worry All the Time: A Participatory Poverty Assessment of Tonle Sap. Phnom Penh: CDRI. Barham, T. (2005). Providing a Healthier Start to Life: The Impact of Conditional Cash Transfers on Infant Mortality. Berkeley, CA: Department of Agriculture and Resource Economics, University of California at Berkeley. Mimeo. Behrman, J.R. and Hoddinott, J. (2000). An Evaluation of the Impact of PROGRESA on Pre-school Child Height. Washington, DC: IFPRI. Mimeo. 120 Braithwaite, J. and Mont, D. (2008). Disability and Poverty: A Survey of World Bank Poverty Assessments and Implications. SP Discussion Paper 0805. Washington, DC: World Bank. C CARD (Council for Agricultural and Rural Development), WFP (World Food Programme) and World Bank (2009). Safety Nets in Cambodia. Concept Note and Inventory. June 30. Carter, M.R., Little, P.D., Morgues, T. and Negatu, W. (2004). Shocks, Sensitivity and Resilience: Tracking the Economic Impacts of Environmental Disasters on Assets in Ethiopia and Honduras. BASIS Research Program on Poverty, Inequality, and Development. Washington, DC: USAID. CDRI (Cambodia Development Resource Institute) (2008). Impact of High Food Prices in Cambodia, Phnom Penh, CDRI. CDRI (2010). Annual Development Review 2009-2010. Phnom Penh: CDRI. Chan, S. (2008). Impact of High Food Prices in Cambodia. Policy brief. Phnom Penh: CDRI. Chandararot, K., Sok, S. and Liv, D. (2009). Rapid Assessment of the Impact of the Financial Crisis in Cambodia. Bangkok: ILO Subregional Office for East Asia. Cichon, M. (2008). Building the Case for a Global Social Floor. ILO PowerPoint Presentation. Conkle, J. 2009. English Supplement to the National Institute of Statistics 2008 Cambodia Anthropometrics Survey. Phnom Penh: UNICEF. CNCDM (Cambodia National Committee for Disaster Management) (2010). Cambodia Post-Ketsana Disaster Needs Assessment. Phnom Penh: CNCDM. CRDB (Cambodian Rehabilitation and Development Board) (2010). The Cambodia Aid Effectiveness Report 2010. Phnom Penh: CRDB, CDC. D del Ninno, C., Subbarao, K. and Milazzo, A. (2009). How to Make Public Works Work: A Review of the Country Experiences. HDNSP Working Paper. Washington, DC: World Bank. Dercon, S. and de Weerdt, J. (2002). Risk Sharing Networks and Insurance Against Illness. Working Paper. Oxford: CSAE. Devereaux, S. and Gorman, C. (2006). Unconditional Cash Transfers in Africa. IDS in Focus 1: Social Protection. 121 E Education Management Information System (EMIS) Office (2010). EMIS 2009/2010. Phnom Penh: EMIS Office, Department of Planning, MoEYS. F Fafchamps, M., Udry, C. and Czukas, K. (1998). Drought and Saving in West Africa: Are Livestock a Buffer Stock? Journal of Development Economics 55(2): 273-305. Farrington, J. and Slater, R. (2006). Cash Transfers: Panacea for Poverty Reduction or Money Down the Drain? Editors’ introduction to Special Cash Transfers Theme Issue of Development Policy Review 24(5). Ferreira, F. and Schady, N. (2008). Aggregate Economic Shocks, Child Schooling, and Child Health. Policy Research Working Paper 4701. Washington, DC: World Bank. Filmer, D. and Schady, N. (2006). Getting Girls into School: Evidence from a Scholarship Program in Cambodia. Policy Research Working Paper 3910. Washington, DC: World Bank. Filmer, D. and Schady, N. (2009). Promoting Schooling through Scholarships in Cambodia: Evidence from the Education Sector Support Project. Impact Evaluation Brief. Phnom Penh: World Bank. Fiszbein, A., Schady, N., Ferreira, F.H.G., Grosh, M., Kelleher, N., Olinto, P. and Skoufias, E. (2009). Conditional Cash Transfers: Reducing Present and Future Poverty. Policy Research Report: Washington, DC: World Bank. Fitzgerald, I. and So, S. (2007). Moving Out of Poverty: Trends in Community Well-Being and Household Mobility in Nine Cambodian Villages. Phnom Penh: CDRI. G Gentilini, U. (2008). Cash and Food Transfers: A Primer. Occasional Paper 18. Rome: WFP. Gertler, P. (2004). Do Conditional Cash Transfers Improve Child Health? Evidence from PROGRESA’s Control Randomized Experiment. American Economic Review 94(2): 336- 341. Grosh, M., del Ninno, C., Tesliuc, E. and Ouerghi, A. (2008). For Protection and Promotion: The Design and Implementation of Effective Safety Nets. Washington, DC: World Bank. GTZ. (2010). Documentation on Landlessness in Cambodia. Phnom Penh. Draft H HEF (Health Equity Fund) Strategic Meeting (2010). Intercontinental Hotel, Phnom Penh, March. Hernández, B., Ramírez, D., Moreno, H. and Laird, N. (2005) Evaluación del Impacto de Oportunidades en la Mortalidad Materna e Infantil. In Hernández, B. and Hernández, M. 122 External Evaluation of the Impact of the Human Development Program Oportunidades. Cuernavaca: National Institute of Public Health. Hoddinott, J. Maluccio, A., Behrman, J.R., Flores, R. and Martorell, R. (2008). Effect of a Nutrition Intervention during Early Childhood on Economic Productivity in Guatemalan Adults. Lancet 371(9610): 411-416. Holzmann, R. and Jorgensen, S. (2000). Social Risk Management: A New Conceptual Framework for Social Protection and Beyond. Social Protection Discussion Paper 0006. Washington, DC: World Bank. I International Labour Organization (ILO) (2004). Child Domestic Labour in Cambodia: Why It Has to Stop and How We Can Stop It. Phnom Penh: ILO. ILO (2010). A Strategy for a Public Works Program. Phnom Penh: ILO. ILO (2010b). The National Social Protection Strategy: Input on Tackling Child Labor and Increasing Educational Access. Contributions from UNICEF and UNESCO. Draft. ILO-IPEC (International Programme on the Elimination of Child Labour) (1998). Combating Trafficking in Children for Labour Exploitation in the Mekong Sub-Region: A Proposed Framework for ILO-IPEC Actions and Proceedings of a Mekong Subregional Consultation. Bangkok: ILO-IPEC. ILO-IPEC. (2003). Cambodia Child Labour Country Brief. Based on the Child Domestic Worker Survey 2003. Phnom Penh: ILO-IPEC. ILO (International Labour Organization) and WHO (World Health Organization) with collaborating agencies and development partners (2009). Social Protection Floor Initiative: Manual and Strategic Framework for Joint UN Country Operations. Geneva: ILO/WHO. J Jacobs, B., Price, N.L. and Sok, S.O. (2007). Do Exemptions from User Fees Mean Free Access to Health Services? A Case Study from a Rural Cambodian Hospital. Tropical Medicine and International Health 12(11): 1–11. Johnston, R. (2006). Twelve Key Family Practices for Health. Baseline Survey Report. Phnom Penh: National Centre for Health Promotion. Jordanwood, T., van Pelt, M. and Grundmann, C. (2009). Health Equity Funds Implemented by URC and Supported by USAID. University Research Co. and USAID Evaluation Report. K Kim, S. (2001). Reciprocity: Informal Patterns of Social Interactions in a Cambodia Village. Masters’ thesis. 123 Kim, S., Chan, S. and Sarthi, A. (2002). Land, Rural Livelihoods and Food Security in Cambodia. A Perspective from Field Reconnaissance. Working Paper 24. Phnom Penh: CDRI. Koy, R. and Em, S. (2009). Livelihood Knowledge in the Tonle Sap Basin: Baseline Survey for Building Community Capacity for Poverty Reduction Initiatives. Working Paper 42. Cambodia: CDRI. Krishnamurthy, V. (1999). The Impact of Armed Conflict on Social Capital: A Study of Two Villages in Cambodia. Study conducted for the World Bank by Social Services of Cambodia. M Maluccio, J.A. and Flores, R. (2005) Impact Evaluation of a Conditional Cash Transfer Program: The Nicaraguan Red de Protección Social. Research Report 141. Washington, DC: IFPRI. Marshall, J., Ung C., Puth Nessay, Ung Ngo Hok, Va Savoeun, Soeur Tinon and Meung Veasna (2009). Student Achievement and Education Policy in a Period of Rapid Expansion: Assessment Data Evidence from Cambodia. International Review of Education 55(4): 393- 414. McAndrew, J. (1998). Independence in Household Livelihood Strategies in Two Cambodian Villages. Working Paper 7. Phnom Penh: CDRI. MEF (Ministry of Economy and Finance) (2010). 2010 Budget Book. Phnom Penh: MEF. MoH (Ministry of Health) (2009). Annual Health Financing Report 2008. Bureau of Health Economics and Financing Department of Planning and Health Information. March 2009. Phnom Penh: MoH MoH (2010). 2009 Annual Performance Monitoring Report, Second Health Sector Support Program 2009-2013. Phnom Penh: MoH. Morduch, J. and Sharma, M. (2002). Redes de Seguridad, Seguro Informal y Microfinanzas. World Bank Social Protection Unit Human Development Network. Morris, E. (2007). Promoting Employment in Cambodia: Analysis and Options. Phnom Penh: ILO. Morris, S., Olinto, P., Flores, R., Nilson, E. and Figueiró, A. (2004). Conditional Cash Transfers Are Associated with a Small Reduction in the Weight Gain of Preschool Children in Northeast Brazil. Journal of Nutrition 134(9): 2336-2341. MoWA (Ministry of Women Affairs) (2008). A Fair Share for Women, Gender Assessment 2008. Phnom Penh: MoWA. N NAA (National AIDS Authority) and MoWA (Ministry of Women’s Affairs) (2009). Preventing Spousal Transmission of HIV in Cambodia: A Rapid Assessment and Recommendations for Action, supported by UNAIDS and UNIFEM. 124 NCHADS (National Center for HIV/AIDS, Dermatology and STD) (2007). Report of a Consensus Workshop: HIV Estimates and Projections for Cambodia 2006-2012. Surveillance Unit, Phnom Penh: June 25-29, MoH. NCPD (National Committee for Population and Development) (2007). Elderly in Cambodia. Presentation for the Seminar on the Social, Health and Economics Consequences of Population Ageing in the Context of Changing Families. Bangkok, July 25-27. NIPH (National Institute of Public Health), NIS (National Institute of Statistics) and ORC Macro (2001). Cambodia Demographic and Health Survey 2000. Phnom Penh and Calverton, MD: NIPH, NIS and ORC Macro. NIPH, NIS and ORC Macro (2006). Cambodia Demographic and Health Survey 2005. Phnom Penh and Calverton, MD: NIPH, NIS and ORC Macro. NIS (National Institute of Statistics) (2002). Cambodia Child Labor Survey 2001. Phnom Penh: NIS, MoP. NIS (2004a). Cambodia Socio-Economic Survey 2004. Phnom Penh: NIS, MoP. NIS (2004b). Child Domestic Worker Survey 2003. Phnom Penh: NIS, MoP. NIS (2007). Cambodia Socio-Economic Survey 2007. Phnom Penh: NIS, MoP. NIS (2008a). Cambodia Socio-Economic Survey 2008. Phnom Penh: NIS, MoP. NIS (2008b). General Population Census of Cambodia 2008. Phnom Penh: NIS, MoP. NIS (2009). Cambodia Socio-Economic Survey 2009. Phnom Penh: NIS, MoP. Norton, A., Conway, T. and Foster, M. (2001). Social Protection Concepts and Approaches: Implications for Policy and Practice in International Development. London: ODI. NSSF (National Social Security Fund) (2009). Achievements in 2009 and Action Plan in 2010. Phnom Penh: NSSF. P Phlong, P. (2009). Informal Credit System in Cambodia. Masters thesis. Prowse, M. (2003). Towards a Clearer Understanding of “Vulnerability� in Relation to Chronic Poverty. Working Paper 24. Manchester: CPRC. R Ravallion, M. (2003). Targeted Transfers in Poor Countries: Revisiting the Trade-Offs and Policy Options. Social Protection Discussion Paper 0314. Washington, DC: World Bank. Ravallion, M. (2008) Bailing out the World’s Poorest. Policy Research Working Paper 4763. Washington, DC: World Bank. 125 RGC (Royal Government of Cambodia) (2009). National Social Development Plan Update 2009-2013. Phnom Penh: RGC. RGC (2010). Resume of Discussion and Agreed Actions. Cambodia Development Cooperation Forum (CDCF), Phnom Penh, June 2-3. Rivera, J.A., Sotres-�lvarez, D., Habicht, J.-P., Shamah, T. and Villalpando, S. (2004) Impact of the Mexican Program for Education, Health, and Nutrition (PROGRESA) on Rates of Growth and Anemia in Infants and Young Children: A Randomized Effectiveness Study. Journal of the American Medical Association 291(21): 2563-2570. S So, S. (2009). Informal Risk Management/Safety Net Practices: Experiences of Poor and Vulnerable Workers and Households. CDRI, Mimeo. Sok, K. (nd) Social Health Protection Mechanisms in Cambodia. Presentation. Department of Planning and Health Information, MoH. T Townsend, R. (1994). Risk and Insurance in Village India. Econometrica 62: 539-592. U UCW (Understanding Children’s Work) (2006). Children’s Work in Cambodia: A Challenge for Growth and Poverty Reduction. Phnom Penh: UCW. UCW (2009). Towards Eliminating the Worst Forms of Child Labour in Cambodia by 2016: An Assessment of Resource Requirement. Phnom Penh: UCW. UNDESA (United Nations Department for Economic and Social Affairs) (2002). Population Ageing in Cambodia: Planning for Social Protection. Phnom Penh: UNDESA. UNDP (United Nations Development Program) (2009). Cambodia's First National Forum on Climate Change. Press Release. Phnom Penh: UNDP. UNDP (2010a). Cambodia’s Millennium Development Goals. Preliminary Summary of Progress. Working Paper. January 2010. Phnom Penh: UNDP. UNDP (2010b). Cambodia’s Millennium Development Goals. Preliminary Summary of Progress. Executive Summary. Phnom Penh: UNDP. UNICEF (United Nations Children’s Fund) (2008). Progress for Children 7. September UNICEF (2009). Study on Child Poverty and Disparities in Cambodia. Phnom Penh: UNICEF. Draft, April. UNICEF (United Nations Children’s Fund) and RGC (Royal Government of Cambodia) (2006) Country Programme Action Plan 2006-2010. Phnom Penh: UNICEF and RGC. US Department of State (2005) Trafficking in Persons Report. Washington, DC: Office of the Under Secretary for Global Affairs, US Department of State. 126 V von Grebmer, K., Nestorova, B., Quisumbing, A., Fertziger, R., Fritschel, H., Pandya Lorch, R., Yohannes, Y. (2009). 2009 Global Hunger Index: The Challenge of Hunger: Focus on Financial Crisis and Gender Inequality. Washington, DC, Bonn and Dublin: IFPRI, Deutsche Welthungerhilfe and Concern Worldwide. Vu, L. and Glewwe, P. (2008). Impact of the Food Crisis on Poverty in Cambodia. University of Minnesota. Mimeo. W WFP (World Food Programme) (2009a). Comprehensive Food Security and Vulnerability Analysis (CFSVA) 2009. Phnom Penh: WFP. WFP (2009b). Standard Project Report 2009. Phnom Penh: WFP. WHO (World Health Organization) (2008). Global TB Control: Surveillance, Planning, Financing. Geneva: WHO. World Bank (2006a). Halving Poverty by 2015? Poverty Assessment 2006. Phnom Penh: World Bank. World Bank (2006b). Managing Risk and Vulnerability in Cambodia: An Assessment and Strategy for Social Protection. Phnom Penh: World Bank. World Bank (2009a). Poverty Profile and Trend in Cambodia. Findings from the CSES. Poverty Reduction and Economic Management Sector Unit, East Asia and the Pacific Region. World Bank (2009b). Sustaining Rapid Growth in a Challenging Environment: Cambodia Country Economic Memorandum. Poverty Reduction and Economic Management Sector Unit, East Asia and the Pacific Region. World Bank (2010). Cambodia: A Cash Transfer Program to Support the Poor While Addressing Maternal and Child Malnutrition – A Discussion Note. Contributions from UNICEF, WFP, GTZ, WHO and CARD. 127 Annex 1. Social Protection and Safety Nets: Concepts and Definitions Social protection consists of a broad set of instruments designed to assist individuals, households and communities to better manage risk, shocks and extreme poverty. Social protection comprises interventions targeted at both the poor and the non-poor. The main objective of social protection for the non-poor is to help individuals manage risks along their life cycle, such as health and unemployment shocks and retirement. In most cases, benefits are linked to people’s income level, which entails substantial fiscal implications. Hence, social protection schemes for the non-poor have often a contributory element, with some public subsidies for the poorest beneficiaries. In contrast, social protection for the poor (or safety nets) aims at guaranteeing a minimum socially accepted living standard (through income support and access to basic services); encouraging human capital investment; and, for groups facing special challenges (such as unskilled youth), improving labor productivity. Social protection should be one of the two core pillars of a country’s growth and poverty reduction strategy. Poverty reduction and equitable growth should be achieved through a two-legged strategy. On the one hand, there is a need to promote overall growth, in particular the economic potential of the poor. Therefore, it is important to support investments and reforms that promote overall growth and to invest in pro-poor programs such as those dealing with rural development, microfinance, education and human capital development. However, the effectiveness of these programs in alleviating poverty could be severely dampened if the chronic poor are not given a basis (or “springboard�) from which to escape from poverty and the near poor are not protected from severe shocks that could lead them to adopt negative coping strategies such as pulling children out of school or selling productive assets. It is important to complement growth-enhancing programs with effective safety net that help relieve chronic poverty and provide a springboard to escape from poverty. This Annex focuses on safety nets, which are a subset of social protection interventions targeted at the poor. Safety nets tend to be non-contributory interventions financed out of general revenue (taxation or ODA). To manage costs, they also tend to be targeted at the poor and those most at risk of poverty. Safety nets usually include workfare (food for work or cash for work), transfers (of cash or food, unconditional or conditional) and, under some definitions, subsidies for basic goods and services for the poor. Objectives of safety nets Objective Design features of interventions Examples Help the poorest and most Targeted, medium- to long-term Targeted cash and in-kind transfers vulnerable assistance to address chronic poverty Lower the impact of impoverishing risks Link assistance to the source of HEFs; public works vulnerability Provide the poor with a springboard to Link assistance with ways to improve CCTs; school feeding programs escape poverty investment in human capital 128 Towards a Social Protection Strategy for the Poor and Vulnerable | Annex 1. Social Protection and Safety Nets: Concepts and Definitions Safety nets play multiple roles in development policy. Most societies hold strong convictions that adequate provision for the poor is required (although they may differ in how this should be achieved). By transferring income to the poorest and most vulnerable, safety nets have an immediate impact on poverty and inequality. In addition to this direct and immediate function, safety nets make other key contributions to long-term national development policy: • Safety nets help households manage risk and vulnerability. At a minimum, safety net programs help households facing hard times to avoid irreversible losses, allowing them to maintain the household and business assets on which their livelihoods are based and to adequately nourish and school their children. Throughout the world, falling income among poor households leads to increased child malnutrition (with long-lasting consequences for child growth, cognitive and learning ability and schooling attainments: Alderman et al., 2006; Ferreira and Schady, 2008); higher school dropout rates with long- term consequences on children’s labor market performance (Ferreira and Schady, 2008); and the selling of productive assets such as land and livestock (which impedes households’ recovery from crisis: Carter et al., 2004; Fafchamps et al., 1998). At best, safety nets can provide an insurance element that lets households make choices about livelihoods that yield higher earnings. Thus, safety nets both protect households and promote their independence. • Safety nets can enable households to make better investments in their future by reducing poverty and improving human capital and labor productivity, promoting an environment favorable to equitable and sustainable economic growth. In this role, safety nets basically act to remedy market failures, allowing households to take up opportunities they would otherwise miss – in terms of investment in both the human capital of their children and the livelihoods of the earners. • Safety nets can directly assist governments in making reforms. Short-term safety net programs can compensate those who are negatively affected by the reforms that are necessary for economic development or who may oppose and stall these reforms (Grosh et al., 2008). For example, with safety nets in place, energy sectors can price for efficiency and trade policy can focus on growth rather than job protection. 129 What role for safety net transfers in low-income countries? Low-income countries are the ones most in need of safety net transfers and at the same time are the ones that can least afford them. Lack of adequate information on beneficiaries, deficient administrative capacity and, particularly, limited fiscal resources pose constraints to the establishment of safety net transfer systems in low-income countries. However, there are compelling reasons to consider some form of transfers in these countries, mainly the inability of growth alone to help the large numbers of the very poor living in unacceptable conditions and the potential contribution of redistributive transfers to growth. In very poor countries, safety nets transfers can help fill in the deepest part of the poverty gap, bring all or many of the poor up to an acceptable consumption level, smooth consumption, protect the poor against major shocks, insure against individual risks and serve as an investment in human capital. The challenge is to find ways in which the choice of programs can both limit costs and reduce the opportunity costs of other investments foregone. The amount of expenditure justified, the types of programs and how these programs are to be designed will depend on: (i) the degree and types of insured risks faced by the poor; (ii) the feasibility of identifying groups that are subjected to high levels of risk; and (iii) the depth and severity of poverty. The design and selection of programs require a country-by-country approach, but a number of general principles apply to most very poor countries. On general design of the system: • Safety net transfer systems need to be placed in the context of a wider development strategy. • A few simple program designs should be selected and adhered to for a sustained period. As long as there is adequate pre-evaluation and program continuity, the source of funds is irrelevant. • Potential displacement of private transfers by public ones should be assessed carefully. • Periodic evaluation of programs is absolutely essential. On choice of programs: • Safety nets expenditures should be used to fund investments that lift longer-run impediments to growth (e.g. public works programs that construct roads and irrigation works). • Food distribution programs should be restricted to emergency situations and otherwise limited to “food for education� or nutritional programs. • For “agricultural inputs subsidies� programs, free distribution of small amounts of fertilizer and seed may be preferable to subsidies, as they are less distorting of agricultural input markets and may not be attractive to larger farms. • Safety net expenditures that simultaneously contribute to human capital development should be used (e.g. food for education and nutritional programs for children). • Transfers that have a multiplier effect should be chosen. On coverage: • For pure transfers, it is important to be as selective as possible (i.e. try to identify a subgroup of ultra-poor or select very distinct groups that everyone agrees are deserving of support). • A policy to fill the entire food poverty gap may be problematic if adopted without careful evaluation of affordability, tradeoffs and adverse incentives. • It is best to opt for self-targeting interventions, such as public works. Geographical targeting criteria seem to be more suitable for very low-income countries, particularly rural and community targeting. The latter would help especially in addressing administrative and informational constraints. • Timing should be considered to optimize the impact of transfers. Countercyclical measures that are targeted seasonally can potentially have a major welfare benefit for the poor. Source: Smith and Subbarao (2003), in World Bank (2006b). 130 Towards a Social Protection Strategy for the Poor and Vulnerable | Annex 1. Social Protection and Safety Nets: Concepts and Definitions Safety net instruments Different international organizations adopt somewhat different definitions of safety nets, but a core set of interventions are common to all (Grosh et al., 2008). Kinds of programs typically included as common elements in a safety net are unconditional transfers in cash and in kind, price subsidies, public works schemes, conditional transfers and fee waivers. • Unconditional in-kind transfers can help ensure access to critical basic goods. They are particularly useful where private markets do not work well enough to ensure that supply will respond to increased demand. Transfers of food (the most common form of in-kind transfer) can be used to improve nutritional status as well as to help maintain consumption levels, by addressing food security directly. Examples of in-kind transfers include mother/child food supplement programs, take-home food rations, school supplies and uniforms. The rice distribution undertaken in November 2008 under the Emergency Food Assistance Project supervised by MEF and the MCH program supported by the WFP are examples of in-kind transfers. • Unconditional cash and near-cash transfers help the poor to maintain basic consumption levels. These may be in the form of cash or vouchers/stamps for purchasing goods and services. Vouchers or stamps limit the choice beneficiaries have for the use of funds; cash transfers allow beneficiaries to use the funds on whatever they see fit to smooth household income and consumption but may be less helpful in times of national crises, when transfers do not have the same purchasing power as a result of higher prices. Programs include needs-based social assistance, family allowances, non-contributory pensions and disability transfers and food stamps. Cambodia currently does not have large programs for cash and near-cash transfers. • Conditional transfers (in cash or in kind) link (i) the provision of resources to poor households to maintain consumption levels over the short term with (ii) incentives for these households to invest in human capital (health and education), with long term, intergenerational benefits. Poor households are given transfers in exchange for compliance with specific conditions (or “co-responsibilities�), encouraging them to use education and health services and lowering the cost of accessing such services. School feeding is an example of a conditional in-kind transfer with wide reach in Cambodia. CCTs have proved very effective in improving lagging human development indicators in a number of countries (Farrington and Slater, 2006; Fiszbein et al., 2009; UNDP, 2006). In Cambodia, the Scholarship for the Poor program under the Education Sector Support Project has increased enrollment rates in the program area by 20 percent (Filmer and Schady, 2006; 2009). As with any instrument, it is important to remember that conditional transfers cannot serve as a complete safety net, as households without access to health or education facilities or without children of school age are not eligible. • Workfare schemes (food for work and cash for work) offer low-skilled poor individuals jobs in improving, repairing and building local infrastructure. Programs provide those who are willing to work with some protection from under- and unemployment through the generation of income. If well implemented, workfare is an effective safety net intervention (del Ninno et al., 2008), as it provides employment while at the same time contributing to the development of local and national infrastructure. A key advantage of workfare is that it “self-targets� by offering a low wage rate: it could therefore be a viable 131 safety net intervention in urban areas, where other targeting methods have proven difficult to implement, but also in rural areas, to help households cope with natural disasters and the lean season. However, international experience shows that political considerations often threaten the long-term sustainability of workfare programs: often, Governments have sought to win favor by setting the offered wage rate excessively high, attracting the near and non-poor in addition to the poor and leading to unsustainable program costs or to the severe rationing of assistance, excluding many households in need. As with all safety nets, there will be some people who need assistance for whom workfare is not the most suitable intervention (i.e. those who are not able to work, such as the elderly; those who are ill or disabled). Among others, MEF, MRD, MOWRAM, ADB, ILO and WFP have implemented workfare programs in Cambodia. • Fee waivers assist in providing the poor with access to essential services, including health care, schooling, utilities and transport. Where other users would be charged a fee, the poor receive these services for free or at a reduced rate. However, if the institutions providing these services are not fully compensated for the loss of revenue, they may seek to avoid serving the poor, provide a lower quality of service or face a financing gap that may ultimately reduce the quality of services more generally. In Cambodia, HEFs provide funds to compensate health facilities for operating exemption policies which allow the poor to obtain treatment without paying user fees. Of the safety net provisions currently existing in Cambodia, health fee exemptions backed by HEFs are probably the program with the most extensive coverage. More than most, this safety net is also reasonably thoroughly mainstreamed in Government sector policy (even if financing is still largely dependent on development partners). • Price subsidies for basic commodities also act as a transfer, ensuring that the poor maintain minimum consumption levels of basic goods and services such as food, energy, housing and utilities. Subsidies are generally seen as a second-best safety net intervention, however: when they are available equally to the poor and non-poor, they dramatically increase the cost of assistance. Differentiated tariffs can achieve greater targeting efficiency but entail costs of their own. Food- and cash-based transfers each have different advantages and disadvantages (Devereaux and Gorman, 2006; Gentilini, 2008; Grosh et al., 2008). The most important issue in deciding between food and cash is the functioning of local food markets. Food transfers may be preferable when food markets do not work well (e.g. for remote communities): under these circumstances, cash transfers that are not managed well may push up the local price of food, resulting in limited net welfare gains for recipients (and welfare losses for poor households not receiving transfers (Aker, 2009)). By contrast, cash transfers may be preferable when food markets function reasonably well and can meet increased household demand: in these circumstances, the added flexibility of cash may help poor families meet needs that will include food but maybe also other essential non-food expenditures.26 26 From the recipient’s point of view, food may also be preferable to cash during periods of rapid inflation. However, when food prices are rising rapidly and fiscal space is limited, the agency supplying the transfer will find it difficult to continue supplying the same quantity of food, as the purchasing power of their procurement budget declines. In the absence of additional financial resources and/or large stockpiles, rising food prices may result in rationing food transfers (reducing the size of the transfer and/or the number of recipients) or stop-start program implementation. In these circumstances, the difference between food and cash may ultimately be minor. 132 Towards a Social Protection Strategy for the Poor and Vulnerable | Annex 1. Social Protection and Safety Nets: Concepts and Definitions The choice of form of transfer needs to reflect how the transfer will affect not only consumers in the short term but also food production in the medium term. An increase in local food prices as a result of cash transfers will have a negative effect on households that do not receive benefits and that need to buy food, but may have a positive impact on poor farmers who can sell a surplus: after some time, higher prices will result in higher incomes for these households and improved local food availability. Conversely, if not implemented carefully, food transfers in unconnected markets may undermine local agricultural production (by driving down farm-gate prices for food), even as they ensure immediate food security (and can help drive prices inflated by hoarding and speculating back down). Whether food- or cash-based transfers work better in any particular situation will also depend on intra-household behavior, the specifics of poverty and food security challenges, and logistical and institutional issues. Food-based arrangements address food insecurity directly: the content of food transfers can be tailored to address specific nutritional problems (e.g. using fortified foods to tackle micronutrient deficiencies), with some evidence that food transfers have a longer-term impact on child growth than do unconditional cash transfers (Gentilini, 2008). Small transfers of good quality food may be more effective than cash when there is a concern that a cash transfer might be diverted to non-food consumption (e.g. alcohol or cigarettes). However, cash is logistically easier to manage, with considerably lower administrative costs (food entails costs and systems for procurement, storage and distribution) and capacity for rapid response. In practice, the relative benefits of cash or food transfers depend heavily on context and design. The value of transfers to beneficiaries and their effects on markets (multipliers/ distortions) in a given situation may reflect not so much the intrinsic advantages and disadvantages of food vs. cash but specific program design choices to do with benefit levels, targeting and delivery mechanisms (when and where benefits are distributed). Under some circumstances, there may also be a case for combining food and cash transfers (Gentilini, 2008). There are often fears that, in a weak fiduciary environment, transfer schemes may be subject to large-scale leakage, with a significant proportion of benefits going to the non- poor. This may be why transfer schemes in Cambodia remain relatively small. This issue requires serious attention: however, successful schemes both in Cambodia and in other countries suggest that there are ways of preventing diversion, which can and should be built into systems for targeting, enrollment, monitoring and auditing. Globally, there is a wide variation in both the manner in which each of these types of program is implemented and the relative success that they enjoy. In other words, there are few universal lessons: the details of design and implementation matter a great deal, with the result that the same broad type of program (e.g. workfare) might be very successful in one country and unsuccessful in another. Program objectives should match the needs of the target population; design and implementation choices need to reflect country context (including socio-cultural issues) and a realistic assessment of institutional capacity. The case for targeting Good targeting is essential to the success of safety nets. Targeted assistance makes it possible to provide a more generous package of benefits to a more restricted set of vulnerable people in need, often at lower overall costs. Targeting aims to include all who are intended 133 to benefit (i.e. the poor) and exclude those who are not intended to benefit (the non-poor). Targeting is particularly important with regard to safety nets, as non-contributory transfers are primarily a private good and households are unlikely to possess an upper limit of what they would wish to receive (Grosh et al., 2008). The choice of targeting mechanism will reflect administrative, social and political considerations. Achieving good targeting requires both good design at the outset and fine-tuning during operation in order to balance different considerations (e.g. minimizing exclusion of the poor versus minimizing inclusion of the non-poor). An effort to target narrowly may not be helpful when there are only minor differences between poorer groups, particularly when there is a high degree of “churning,� with people moving in and out of poverty over time (in which case the scheme would need to conduct frequent reassessment of households’ circumstances to ensure accurate targeting). Those designing and implementing targeted schemes should also pay attention to the costs that the poor may have to incur to prove eligibility; the potential risk that targeted benefits may create perverse incentives (resulting in people changing their behavior to qualify); and the risk that targeting criteria may be subject to political manipulation. There are a number of ways to address each of these issues and so ensure that targeting does deliver its potential. These need to be considered carefully at the design stage and their effectiveness monitored during implementation. A comprehensive safety net system can make use of various methods to balance responsiveness in case of emergencies with efficient medium- to long-term targeting. Targeting can be geographic, if pockets of poverty exist, or can be household based if poverty is widespread, with poor households mixed up with non-poor households. Often, the most effective targeting systems combine two (or more) different targeting mechanisms. The most advanced systems use means-tested or proxy means-tested methods that ask detailed questions to households about their wealth, income and assets to determine who is eligible. MoP’s IDPoor is an example of proxy means-tested targeting tailored to the rural Cambodian context, combined with community validation (see Box 9 in the main text). As proxy means-testing is relatively costly to maintain, it should be used to target various social protection programs at the same time, since economies of scale can be reached by splitting costs across interventions. However, some programs should also adopt other targeting methods, both because proxy means-testing tends to be less good at responding in the short term to changes in the need for assistance and because, despite its sophistication, it would still miss some vulnerable households (see Ravallion, 2003; 2008). Other targeting methods include: • Simpler categorical targeting (i.e. children, elderly, disabled, pregnant women, etc); • Geographical targeting (universal coverage in areas with very high poverty incidence); • Self-targeting, where limited assistance is offered to everybody but at some cost to households, so that only the neediest participate (the most common intervention of this type is public works, where work is offered to all people willing to work, but at low wages). 134 Towards a Social Protection Strategy for the Poor and Vulnerable | Annex 2. The Current Framework for Social Protection in Cambodia Annex 2. The Current Framework for Social Protection in Cambodia The Government reiterated its commitment to alleviating poverty through social protection. In the Second Phase of the Rectangular Strategy for Growth, Employment, Equity and Efficiency, announced in September 2008 by H.E. the Prime Minister of Cambodia, the RGC sets out to: 1. Give priority to improving working conditions for workers and employees; 2. Enforce Social Security Law; 3. Implement benefit and pension schemes for people with disabilities and their dependents; 4. Ensure protection of those covered by the Labor Law; and 5. Make available to all employees insurance coverage against workplace accidents. The RGC also intends to continue to strengthen support to people with disabilities and families of veterans, as well as retired civil servants and veterans. This will be achieved by implementing a comprehensive pension system under the NSSF (see Section 2). In this context, the Government intends to promulgate a Law on the Comprehensive National Social Security Fund and a Law on the Establishment of National Pension for Veterans. The Second Phase of the Rectangular Strategy will continue to be operationalized in the NSDP, which has been extended to 2013. In the NSDP 2008-2010 Midterm Review, the RGC explicitly recognizes as a challenge for the coming years the provision of safety nets to the poor during times of inflationary stress, particularly shortages of staple rice, through targeted subsidies and tax relief on essential commodities. The RGC also identifies an urgent need to implement measures to ensure safety nets for the most vulnerable through subsidies and targeted labor-intensive work like food for work programs, as a means to reduce poverty. The NSDP Midterm Review explicitly addressed the issue of increasing inequality. It recognizes that this trend, evident in increasing disparities in income, consumption and economic opportunities between urban and rural areas and high- and low-income groups, can be addressed by more and focused attention on rural areas and agricultural production and by ensuring that adequate safety nets exist to stop low-income groups sliding back into poverty. The NSDP Midterm Review also reaffirms that proper nutritional status of all Cambodians, particularly the vulnerable, is an important element of national development. Food security and nutrition are recognized in the NSDP as a crosscutting issue: achieving progress here depends on the synergy of growth and output from many sectoral activities as outlined in the Strategic Framework on Food Security and Nutrition. The NSDP envisages that targeted efforts in various sectors will ensure that “poor and food-insecure Cambodians, by 2010, have substantially improved physical and economic access to sufficient, safe and nutritious food at all times to meet their dietary needs and food preferences to an active and healthy life.� The need for a Government safety net strategy was discussed at the Second CDCF in December 2008. H.E. The Deputy Prime Minister Keat Chhon assigned to the TWG FSN, co-chaired by CARD and the WFP, the responsibility to undertake a review of safety net interventions and elaborate options for a safety net strategy. At the Third Cambodian Economic Forum (organized by the Supreme National Economic Council (SNEC) in February 2009), development partners welcomed and supported the RGC’s recent commitment to implement a coordinated social safety net system as an important initiative and an essential element in protecting the poorest and most vulnerable from possible economic shocks. 135 The NSDP Update (2009-2013) further specifies the need to strengthen and streamline social protection. In guiding the development of the NSPS and the priorities over the short to medium term, it highlights the need to: • Give preference to social protection measures that not only provide immediate relief but also contribute to building the beneficiary population’s ability/capacity to contribute to the social and economic development of their community; • Ensure greater transparency and better targeting in the delivery of social protection for the poor through the use of the IDPoor program and through another appropriately adapted targeting mechanism for urban poor while IDPoor is being adjusted to urban areas; • Minimize the planning and delivery costs (overheads) of social safety net programs to achieve a maximum net transfer of resources to beneficiary populations; and • Ensure cross-sectoral coordination and integration of social protection measures with decentralized development planning. At the Third CDCF in June 2010, the Government’s commitment to ensuring effective social protection was reaffirmed and the draft NSPS was discussed. The joint RGC– development partner summary of the CDCF discussion and agreed actions (RGC, 2010) outlines the next steps for strengthening social protection provision in Cambodia: “In the session on Social Protection, an outline of the new National Social Protection Strategy was provided, showing how the Strategy will contribute to the reduction of chronic poverty, help the poor to cope with shocks, promote human capital, improve productivity and promote sustainable economic growth. Implementation of the Strategy will require scaling up of existing interventions, design of institutional arrangements and the introduction of new programmes to cover existing gaps. CARD is mandated to establish an interim coordination unit which will examine ways to establish well-resourced institutional capacities and systems. The NSDP’s resourcing profile has identified the Strategy as a priority programme of the Royal Government. Over the next six months CARD will produce a costed action plan toward strengthening institutions and capacities, and will also prepare a costing of the National Social Protection Strategy.� Legal framework of social protection in Cambodia The fundamental commitments to provide social protection for Cambodians lie in the Constitution, which contains articles on the rights of citizens and responsibilities of Government related to social protection. For example: • Article 36: “Every Khmer citizen shall have the right to obtain social security and other social benefits as determined by law. Khmer citizens of either sex shall have the right to form and to be member of trade unions. The organization and conduct of trade unions shall be determined by law.� • Article 46: “The commerce of human beings, exploitation by prostitution and obscenity which affect the reputation of women shall be prohibited […] The state and society shall provide opportunities to women, especially to those living in rural areas without adequate social support, so they can get employment, medical care, and send their children to school, and to have decent living conditions.� • Article 72: “The health of the people shall be guaranteed […] Poor citizens shall receive free medical consultation in public hospitals, infirmaries and maternities. The State shall 136 Towards a Social Protection Strategy for the Poor and Vulnerable | Annex 2. The Current Framework for Social Protection in Cambodia establish infirmaries and maternities in rural areas.� • Article 75: “The State shall establish a social security system for workers and employees.� • Article 73: “The State shall give full consideration to children and mothers. The State shall establish nurseries, and help support women and children who have inadequate support.� Article 74: “The State shall assist the disabled and the families of combatants who sacrificed their lives for the nation.� Several organic laws have been passed to codify some of the social protection-related tenets of the Constitution. The most relevant laws that have been adopted or are currently under consideration include: • The Labor Law passed in October 1998. This provides for a standard legal workweek of 48 hours, not to exceed eight hours per day. It also stipulates pay regulations for overtime, Sundays and holidays. The minimum allowable age for a salaried position is set at 15 years, or 18 for anyone engaged in work that may be hazardous, unhealthy or unsafe. The Labor Law embodies most of the ILO Conventions on core labor standards, all of which Cambodia has ratified. • The Insurance Law passed in June 2000. This provides a legal framework for better regulation of insurance market activities. To help develop the insurance sector, the RGC envisages an expansion of the activities of the state-owned Cambodian National Insurance Company (CAMINCO) to include life insurance, pensions, credit and natural disaster insurance. However, these insurance schemes will benefit only a small proportion of the population in the initial stages and most likely will not be accessible by the poor, who lack the surplus funds necessary to insure their risks. • The Law on Social Security Schemes for Persons defined by the Provisions of the Labor Law passed in September 2002. This entitles workers and employees in the private sector to old-age, invalid and survivors’ benefits as well as workmen’s compensation. The law has been promulgated but not implemented since it requires a sub-decree on the NSSF, which will cover employment injury insurance, the pension scheme and a short-term benefit system. Notably, the Employment Injury Scheme would cover the formal sector as well as the self-employed based on voluntary participation. However, as in the case of the Insurance Law, only a very small segment of the population is expected to benefit from the Law on Social Security Schemes, given the small size of Cambodia’s formal sector. • A National Action Plan to Combat Violence against Women has been developed and is being implemented in accordance with the Law on the Prevention of Domestic Violence and Protection of Victims adopted in 2005. • The Law on Suppression of Trafficking in Humans and Sexual Exploitation was adopted in late 2007, consistent with the UN Palermo Protocol. This provides for heavier punishment if victims are below 15 years of age and gives police more power to investigate and arrest suspected traffickers. • The Law on Protecting and Improving Rights of People with Disabilities was adopted in May 2009. 137 138 Risks and shocks Program type Programs Lead Supporting DP Annual Year of Bene�ciary pro�le Coverage ministry expenditure expenditure (est. US$) data No. of target No. of No. of female Mostly Mostly Whole bene�ciaries bene�ciaries bene�ciaries poor non-poor population 1. Situations of Food distribution Emergency Food MEF ADB 6.5m 2008 342,853 342,853 x 200 targeted emergency and Assistance Project (free communes crisis distribution of rice) General food distribution NCDM WFP 698,353 2009 25,000 84,205 (2009) 42,103 x Rural (Ketsana) Package of emergency MoSVY UNICEF 206 new 87 x Rural relief to vulnerable and landmine victims of emergency victims (2009) (including victims of mines) Annex 3. Additional Tables Various projects NCDM CRC, NGOs x Budget support Agriculture smallholder MEF World Bank 13m 2009-2010 n/a n/a n/a x n/a and social protection (AusAID: development policy supporting Inventory of existing interventions for Cambodia operation technical assistance for agriculture smallholder and social protection development) Commune Emergency assistance MoI UNICEF 27,600 (2009 158 158 x Rural transfers for – cash and in- – UNICEF communes communes emergency kind assistance to contribution) assistance communes to support achievement of CMDGs 2. Human development constraints Poor maternal and Nutrition Child survival: MoH UNICEF 3.6m 2010 budget x child health and programs components on nutrition improving maternal health and newborn care, promotion of key health and nutrition practices MCH and Nutrition MoH WFP 3.4m 2009 46,620 64,273 (2009) 48,379 x program Other interventions MoH WHO, others x Various projects NGOs x Social security Maternity bene�ts for MLVT n/a x National all workers except domestic workers, civil servants, armed forces and police; 90 days maternity leave; pay at half salary covered by employer (Labor Law development) Commune Emergency assistance MoI UNICEF 27,600 (2009 158 158 x Rural transfers for – cash and in- – UNICEF communes communes emergency kind assistance to contribution) assistance communes to support achievement of CMDGs 2. Human development constraints Poor maternal and Nutrition Child survival: MoH UNICEF 3.6m 2010 budget x child health and programs components on nutrition improving maternal Risks and shocks Program type Programs newborn health and Lead Supporting DP Annual Year of Bene�ciary pro�le Coverage care, promotion of key ministry expenditure expenditure health and nutrition (est. US$) data No. of target No. of No. of female Mostly Mostly Whole practices bene�ciaries bene�ciaries bene�ciaries poor non-poor population 1. Situations of Food distribution Emergency Food MCH and Nutrition MEF MoH ADB WFP 6.5m 3.4m 2008 2009 342,853 46,620 342,853 64,273 (2009) 48,379 x 200 targeted emergency and Assistance Project (free program communes crisis distribution of rice) Other interventions MoH WHO, others x General food distribution NCDM WFP 698,353 2009 25,000 84,205 (2009) 42,103 x Rural (Ketsana) Various projects NGOs x Social security Package of emergency Maternity bene�ts for MoSVY MLVT UNICEF n/a 206 new 87 x x Rural National relief to vulnerable and all workers except landmine victims of workers, civil domestic emergency victims (2009) servants, victims of (including armed forces mines) and police; 90 days maternity leave; pay at Various projects by half salary covered NCDM CRC, NGOs x employer (Labor Law Budget support Agriculture smallholder Article 183) MEF World Bank 13m 2009-2010 n/a n/a n/a x n/a and social protection (AusAID: Poor access to Scholarships in development policy FTI (Grades 4-6) MoEYS supporting DP group 217,112 2009 5,174 3,459+ 1,713 x Rural quality education cash operation technical CESSP (Grades 7-9) MoEYS assistance WB 819,847 2009 No target set 18,684 12,036 x Rural for agriculture (2009) smallholder JFPR (Grades 7-9) MoEYS and social ADB 15,087 x protection scholarships development) (2003-2006) Commune Emergency assistance BETT (Grades 7-9) MoI MoEYS UNICEF BTC 27,600 (2009 158 158 6,427 x Rural transfers for – cash and in- – UNICEF communes communes scholarships emergency kind assistance to contribution) (2004-2007) assistance communes to support EEQP (Grades CMDGs achievement of10-12) MoEYS ADB $1.22m Total 4,000 by 2014 500 (2009- 0.6 x Rural 2010) 2. Human development constraints Dormitory (Grades MoEYS ADB 216 (2006- x Poor maternal and Nutrition Child survival: 10-11) MoH UNICEF 3.6m 2010 budget 2008) x child health and programs components on nutrition improving maternal Various projects (Grades NGOs 4,876 (2003- x health and newborn 7-9) 2008) care, promotion of key health and nutrition Emergency Food MEF ADB 305,350 2010 budget 12,645 x Targeted Assistance practices Project communes (Grades 5-6 & 8-9) MCH and Nutrition MoH WFP 3.4m 2009 46,620 64,273 (2009) 48,379 x Child labor, Direct Project of program Support to the MLVT ILO 665,000 Until 2012 12,000 18,000+ Minimum of Only 15 provinces Towards a Social Protection Strategy for the Poor and Vulnerable | Annex 3. Additional Tables especially its intervention NPA-WFCL 2008-2012 children 2,000 children 50% poor by 2012 worst forms (prevention and Other interventions MoH WHO, others families (2010- 4,5000+ x withdrawal) 2012) families (2006- and livelihood Various projects NGOs 2008) x improvement Social security Maternity bene�ts for MLVT n/a x National Poor access to Second chance TVET pilot except all workers skills bridging MLVT ADB 360,000 Total 700 by 2012 x National quality training education domestic workers, civil program program servants, armed forces 139 and police; 90 days TVET post-harvest MLVT ADB 1.15m Total 3,000 by 2012 x Rural maternity leave; pay at processing half salary covered by employer (Labor Law TVET voucher skills MLVT ADB 6.13m Total 210,000 by x Rural BETT (Grades 7-9) MoEYS BTC 6,427 x scholarships (2004-2007) 140 EEQP (Grades 10-12) MoEYS ADB $1.22m Total 4,000 by 2014 500 (2009- 0.6 x Rural 2010) Dormitory (Grades MoEYS ADB 216 (2006- x 10-11) 2008) Various projects (Grades NGOs 4,876 (2003- x 7-9) 2008) Risks and shocks Program type Programs Lead Supporting DP Annual Year of Bene�ciary pro�le Coverage Emergency Food MEF ministry ADB 305,350 expenditure 2010 budget expenditure 12,645 x Targeted Assistance Project (est. US$) data No. of target No. of No. of female Mostly Mostly Whole communes (Grades 5-6 & 8-9) bene�ciaries bene�ciaries bene�ciaries poor non-poor population Child labor, 1. Situations of Direct Food distribution Project of Support Emergency Food to the MEFMLVT ILO ADB 665,000 6.5m Until 20082012 12,000 342,853 18,000+ 342,853 Minimum of Only x 15 provinces 200 targeted especially its emergency and intervention NPA-WFCL 2008-2012 Assistance Project (free children 2,000 children 50% poor by 2012 communes worst crisis forms (prevention and distribution of rice) families (2010- 4,5000+ withdrawal) 2012) families (2006- and livelihood General food distribution NCDM WFP 698,353 2009 25,000 2008) 84,205 (2009) 42,103 x Rural improvement (Ketsana) Poor access to Second chance TVET pilot emergency Package ofskills bridging MLVT MoSVY ADB UNICEF 360,000 Total 700 by 2012 206 new 87 x National Rural quality training education program relief to vulnerable and landmine program victims of emergency victims (2009) TVET post-harvest (including victims of MLVT ADB 1.15m Total 3,000 by 2012 x Rural processing mines) TVET voucher skills Various projects MLVT NCDM ADB CRC, NGOs 6.13m Total 210,000 by x Rural training program (non- 2015 Budget support formal) Agriculture smallholder MEF World Bank 13m 2009-2010 n/a n/a n/a x n/a and social protection (AusAID: Various projects development policy NGOs supporting x operation technical 3. Seasonal Public works Food for work MRD WFP assistance 3.1m 2009 156,249 106,923 56,461 x Rural unemployment programs for agriculture (2009) and livelihood smallholder opportunities Food for work MEF ADB (to WFP) and social 0.6m 2009 5,429 x Targeted (Emergency Food protection 1.4m 2010 households communes Assistance Project) development) Commune Cash for work Emergency assistance MEF MoI ADB UNICEF 0.1m 27,600 (2009 2009 158 2,824 158 x Targeted Rural transfers for (Emergency Food – cash and in- 6.6m – UNICEF 2010 communes households communes communes emergency Assistance Project) kind assistance to contribution) assistance communes to support School feeding achievement of CMDGs School feeding MoEYS WFP 8.1m 2009 564,660 532,186 255,669 x Rural (20% (2009) of schools) 2. Human development constraints Emergency Food MEF ADB (to WFP) 1.2m 2010 60,500 60,500 28,984 x Targeted Poor maternal and Nutrition Child survival: Assistance Project MoH UNICEF 3.6m 2010 budget x communes child health and programs components on nutrition Take-home improving maternal Take-home rations MoEYS WFP 92,400 86,925 (2009) 43,464 x Rural rations health and newborn care, promotion of key Various projects NGOs x health and nutrition 4. Health shocks Insurance practices NSSF health insurance MLVT n/a Private sector x National (planned to be employees of MCH and Nutrition implemented 2011) MoH WFP 3.4m 2009 46,620 8< �rms of 64,273 (2009) 48,379 x program NSSF employment injury MLVT n/a Private sector 350,000 x National Other interventions coverage MoH WHO, others employees of workers (Dec. x �rms of 8< 2009) Various projects NGOs x Health insurance for MSVY n/a x National Social security Maternity bene�ts for retired civil servants MLVT n/a x National all workers except (planned) domestic workers, civil Fee waiver servants, armed forces Exemptions at rural MoH x Rural and police; 90 days facilities for poor maternity leave; pay at patients half salary covered by HEFs employer (Labor Law HEFs in 50 ODs MoH USAID, BTC, 6.6m 2009 Est. 4 million 68% of target x National Cash for work MEF ADB 0.1m 2009 2,824 x Targeted (Emergency Food 6.6m 2010 households communes Assistance Project) School feeding School feeding MoEYS WFP 8.1m 2009 564,660 532,186 255,669 x Rural (20% (2009) of schools) Emergency Food MEF ADB (to WFP) 1.2m 2010 60,500 60,500 28,984 x Targeted Assistance Project communes Take-home Take-home rations MoEYS WFP 92,400 86,925 (2009) 43,464 x Rural rations Various projects NGOs x Risks and shocks Program type Programs Lead Supporting DP Annual Year of Bene�ciary pro�le Coverage 4. Health shocks Insurance NSSF health insurance MLVT ministry n/a expenditure expenditure Private sector x National (planned to be (est. US$) data employees of No. of target No. of No. of female Mostly Mostly Whole implemented 2011) �rms of 8< bene�ciaries bene�ciaries bene�ciaries poor non-poor population 1. Situations of NSSF employment Food distribution Emergency Food injury MEFMLVT n/a ADB 6.5m 2008 Private sector 342,853 350,000 342,853 x x National 200 targeted emergency and coverage Assistance Project (free employees of workers (Dec. communes crisis distribution of rice) �rms of 8< 2009) Health insurance for General food distribution MSVY NCDM n/a WFP 698,353 2009 25,000 84,205 (2009) 42,103 x x National Rural retired civil (Ketsana) servants (planned) Package of emergency MoSVY UNICEF 206 new 87 x Rural Fee waiver Exemptions at rural relief to vulnerable and MoH landmine x Rural facilities for poor victims of emergency victims (2009) patients (including victims of mines) HEFs HEFs in 50 ODs MoH USAID, BTC, 6.6m 2009 Est. 4 million 68% of target x National Various projects NCDM World Bank, CRC, NGOs living under the population: x ADB, DFID, poverty line 3,168,883 Budget support Agriculture smallholder MEF AusAID World Bank 13m 2009-2010 n/a people (2008) n/a n/a x n/a and social protection (AusAID: CBHI 13 CBHI schemes development policy MoH GTZ and others supporting Target 122,829 2009 Near poor National operation technical population: the (coverage less assistance near poor than 1% of for agriculture target) smallholder 5. Special Social welfare Elderly persons' MoSVY and social x National vulnerable groups for elderly association support and protection services development) Commune Various projects Emergency assistance NGOs MoI UNICEF 27,600 (2009 158 158 x Rural transfers for – cash and in- – UNICEF communes communes emergency Pensions kind assistance to for Invalidity pensions MoSVY n/a contribution) 31,121 people x National assistance parents or to support communesguardians with disabilities of deceased of CMDGs achievement soldiers, spouses of people living 2. Human development constraints with disabilities, retirees and people who have Poor maternal and Nutrition lost their ability Child survival: to work MoH UNICEF 3.6m 2010 budget x child health and programs components on nutrition Social welfare improving maternal Physical rehabilitation MoSVY UNICEF x National for families living health and newborn centers/community- with disabilities care, promotion of key based rehabilitation services for people health and nutrition with practices disabilities MCH and Nutrition Various projects MoH WFP NGOs 3.4m 2009 46,620 64,273 (2009) 48,379 x program Social welfare Orphans: allowance, MoSVY UNICEF x National Towards a Social Protection Strategy for the Poor and Vulnerable | Annex 3. Additional Tables for children and Other interventions alternative/residential MoH WHO, others x orphans care; Child victims of traf�cking, sexual Various projects NGOs x exploitation and abuse; Social security Children bene�ts with Maternityin conflictfor MLVT n/a x National the law and drug- all workers except addicted workers, domestic children civil servants, armed forces 141 Various projects and police; 90 days NGOs x maternity leave; pay at Social welfare half salary covered by Child protection: helps MoSVY UNICEF 2.9m 2010 budget x National and policy employer (Labor Law develop laws, policies, Various projects NGOs x Pensions Invalidity pensions for MoSVY n/a 31,121 people x National parents or guardians with disabilities of deceased soldiers, 142 spouses of people living with disabilities, retirees and people who have lost their ability to work Social welfare Physical rehabilitation MoSVY UNICEF x National for families living centers/community- with disabilities based rehabilitation Risks and shocks Program type services for Programs people with Lead Supporting DP Annual Year of Bene�ciary pro�le Coverage disabilities ministry expenditure expenditure (est. US$) data No. of target No. of No. of female Mostly Mostly Whole Various projects NGOs bene�ciaries bene�ciaries bene�ciaries x poor non-poor population 1. Situations of Social welfare Food distribution Orphans: allowance, Emergency Food MoSVY MEF UNICEF ADB 6.5m 2008 342,853 342,853 x National 200 targeted emergency and for children and alternative/residential Assistance Project (free communes crisis orphans care; Child of rice) distribution victims of traf�cking, sexual exploitation and abuse; General food distribution NCDM WFP 698,353 2009 25,000 84,205 (2009) 42,103 x Rural Children (Ketsana)in conflict with the law and drug- addicted children Package of emergency MoSVY UNICEF 206 new 87 x Rural relief to vulnerable and landmine Various projects victims of emergency NGOs victims (2009) x (including victims of Social welfare Child protection: helps mines) MoSVY UNICEF 2.9m 2010 budget x National and policy develop laws, policies, development standards and raise Various projects NCDM CRC, NGOs x awareness to protect Budget support children at smallholder Agriculture particular risk MEF World Bank 13m 2009-2010 n/a n/a n/a x n/a and social protection (AusAID: Social welfare Social services and development policy care MoSVY Global Fund, supporting x National for families living to children operation and families UNAIDS, technical with HIV/AIDS of victims and people UNICEF assistance affected by HIV/AIDS; for agriculture children in conflict with smallholder the law and; drug- and social addicted children protection development) HIV/AIDS workplace MLVT UNICEF 100,000 2009 x Urban Commune program for garment Emergency assistance MoI UNICEF 27,600 (2009 158 158 x Rural transfers for factory workers – cash and in- – UNICEF communes communes emergency kind assistance to contribution) assistance Food Assistance to communes to support MoH, WFP 4.2m 2009 77,355 75,879 (2009) 41,211 x National People Living with HIV achievement of CMDGs MoSVY and AIDS 2. Human development constraints Various projects NGOs x Poor maternal and Nutrition Child survival: MoH UNICEF 3.6m 2010 budget x child health and For TB patients programs Food Assistance components on to TB MoH, WFP 2.9m 2009 36,000 38,268 (2009) 19,500 x National nutrition Patients improving maternal MoSVY health and newborn 6. Other Pensions care,servants/ veterans Civil promotion of key MoSVY n/a 16.4m tbc x National retirement nutrition health and pensions practices NSSF employer-based MLVT n/a x National pension schemes MCH and Nutrition MoH WFP 3.4m 2009 46,620 64,273 (2009) 48,379 x (planned) program Other interventions MoH WHO, others Fast Track Initiative; TVET = Technical and Note: BETT = Basic Education and Teacher Training; CESSP = Cambodia Education Sector Support Project; EEQP = Enhancing Education Quality Project; FTI = x Vocational Education and Training; DFID = UK Department for International Development; USAID = United States Agency for International Development. Various projects NGOs x Social security Maternity bene�ts for MLVT n/a x National all workers except domestic workers, civil servants, armed forces and police; 90 days maternity leave; pay at half salary covered by employer (Labor Law Towards a Social Protection Strategy for the Poor and Vulnerable | Annex 3. Additional Tables Summary disbursement from development partners and NGOs, by province, 2009 ($ ‘000s) No Province Development NGO Total partners 1 Banteay Meanchey 19,845 2,690 22,535 2 Battambang 31,848 5,611 37,459 3 Kampong Cham 17,549 3,880 21,429 4 Kampong Chhnang 18,643 3,988 22,631 5 Kampong Speu 9,377 2,648 12,025 6 Kampong Thom 32,795 3,600 36,395 7 Kampot 12,841 904 13,745 8 Kandal 30,573 6,390 36,963 9 Koh Kong 2,171 1,912 4,083 10 Kracheh 15,268 1,344 16,612 11 Krong Kep 1,160 162 1,322 12 Krong Pailin 9,878 200 10,078 13 Preah Sihanouk 32,199 4,052 36,251 14 Mondulkiri 24,081 1,352 25,433 15 Otdar Meanchey 11,309 1,542 12,851 16 Phnom Penh 43,539 27,556 71,095 17 Preah Vihear 22,942 2,201 25,143 18 Prey Veng 15,332 1,580 16,912 19 Pursat 12,328 1,424 13,752 20 Ratanakkiri 13,969 659 14,628 21 Siem Reap 34,726 21,386 56,112 22 Stung Treng 13,015 863 13,878 23 Svay Rieng 5,487 1,123 6,610 24 Takeo 10,644 4,463 15,107 25 Unknown 41,185 295 41,480 26 Nationwide 403,473 1,456 404,929 Total disbursements 886,177 103,281 989,458 Source: CRDB (2010).A 143 xaxAXa