67320 CASH AND IN-KIND TRANSFERS FOR AT-RISK L YOUTH, THE DISABLED, AND VULNERABLE ELDERLY A O E E SOCIAL ASSISTANCE PROGRAM AND PUBLIC EXPENDITURE REVIEW 7 THE WORLD BANK, JAKARTA OFFICE Indonesia Stock Exchange Building Tower II/12th Floor Jl. Jend. Sudirman Kav. 52-53 Jakarta 12910 Tel: (6221) 5299-3000 Fax: (6221) 5299-3111 Website: www.worldbank.org/id THE WORLD BANK 1818 H Street N.W. Washington, D.C. 20433, U.S.A. Tel: (202) 458-1876 Fax: (202) 522-1557/1560 Website: www.worldbank.org Printed in February 2012. Designed by Hasbi Akhir (hasbi@aisukenet.com) Cover photograph and photograph on page 9 provided by Anne Cecile Esteve/Matahati Productions/World Bank. Photograph on page 7 provided by Ryca C. Rawung. Copyright protection and all other rights reserved. The Social Assistance Program and Public Expenditure Review policy notes 1 through 8 together comprise Volume 2 of Protecting Poor and Vulnerable Households in Indonesia report. Both the report and the policy notes are products of the World Bank. The �ndings, interpretations, and conclusions expressed herein do not necessarily reflect the views of the Board of Executive Directors of the World Bank or the Governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of the World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. For any questions regarding this report, please contact Vivi Alatas (valatas@worldbank.org) or Jon Jellema (jjellema@worldbank.org). JSLU, JSPACA, PKSA CASH AND IN-KIND TRANSFERS FOR AT-RISK YOUTH, THE DISABLED, AND VULNERABLE ELDERLY SOCIAL ASSISTANCE PROGRAM AND PUBLIC EXPENDITURE REVIEW 7 JSLU, JSPACA, PKSA Cash and In-kind Transfers for at-risk youth, the disabled, and vulnerable elderly Table of Contents List of Abbreviations, Acronyms and Indonesian Terms 4 Executive Summary 6 1. Background 8 2. Objective, Program Size and Bene�t Adequacy 11 3. Targeting 19 4. Potential Impacts 23 5. Cost Effectiveness 25 6. Implementation 28 7. Public Financial Management and Sustainability 34 8. Summary and Recommendations 37 References 40 Annexes 41 List of Figures Figure 1: Intervention logic 13 Figure 2: Spending on the Yanrehsos Program/Directorate General 14 Figure 3: Number of Bene�ciaries: JSLU, JSPACA, and PKSA 14 Figure 4: Spending on the Yanrehsos Program by activity/subdirectorate 15 Figure 5: Evolution of bene�t levels for JSPACA and JSLU over time 16 Figure 6: Spending on the Yanrehsos Program by executing agency 17 Figure 7: Flow of funds within the Yanrehsos Program 18 Figure 8: JSPACA Coverage and Incidence by province, 2009 19 Figure 9: JSLU Coverage and Incidence by province, 2009 20 Figure 10: Funds utilization: JSPACA 23 Figure 11: Funds utilization: PKSA 24 Figure 12: Spending on the Social Services and Rehabilitation Program by economic classi�cation 25 Figure 13: Implementation: PKSA 28 Figure 14: Implementation: JSPACA and JSLU 29 Figure 15: PKSA bene�ciaries assessment process 30 Figure 16: Bene�ciary perception of PKSA features 31 Figure 17: JSPACA funds delivery mechanism 31 Figure 18: Topic discussed during PKSA facilitator visits 32 Figure 19: Budget execution ratios by executing agents 35 Figure 20: Budget execution by month (Monthly and cumulative actual expenditure, percentage of total annual actual expenditure, 2008 through 2010) 35 Figure 21: Financial sustainability of programs for vulnerable groups out to 2014 36 Figure 22: Triage protocol for two types of vulnerable children 38 2 List of Tables Table 1: Yanrehsos cash transfer programs at a glance 12 Table 2: Spending on the Yanrehsos Program by Activity/Directorate 15 Table 3: Summary of Eligibility Criteria of JSPACA, JSLU, PKSA 22 Table 4: Spending Ef�ciency Indicators, 2008-2010 27 Table 5: Financial sustainability of programs for vulnerable groups out to 2014 36 Table 6: Spending on the Yanrehsos Program by activity & Implementing Unit 41 Table 7: Indicators of spending ef�ciency by major activity/Directorate, 2009 and 2010 42 Table 8: Budget execution ratios by Activity 43 List of Boxes Box 1: Baseline estimates for vulnerable subpopulations in Indonesia 21 3 JSLU, JSPACA, PKSA Cash and In-kind Transfers for at-risk youth, the disabled, and vulnerable elderly List of Abbreviations, Acronyms and Indonesian Terms APBD Anggaran Pendapatan dan Belanja Daerah (Regional Budget, Both Provincial and District Budgets) APBN Anggaran Pendapatan dan Belanja Negara (Central Government Budget) Bappenas Badan Perencanaan dan Pembangunan Nasional (National Development Planning Agency) BKSN Badan Kesejahteraan Sosial Nasional (National Social Welfare Agency) BLT Bantuan Langsung Tunai (Unconditional cash transfer) BN/bn Billion BPS Badan Pusat Statistik (Central Statistics Agency - Statistics Indonesia) DG Directorate General Dinsos Dinas Sosial (Kemensos regional-level of�ce) Fakir Miskin Poor people GDP Gross Domestic Product GOI Government of Indonesia INPRES Instruksi Presiden (Presidential Instruction) JSLU Jaminan Sosial Lanjut Usia (Social cash transfer for the elderly) JSPACA Jaminan Social Penyandang Cacat Berat (Social cash transfer for the severely disabled) Kemenkeu Kementerian Keuangan (Ministry of Finance, MOF) KUBE Kelompok Usaha Bersama (Collective Enterprise Group) LAKIP Laporan Akuntabilitas Keuangan Pemerintah (Annual Financial Accountability Reports) LHS Left hand side (of graph) LIPI Lembaga Ilmu Pengetahuan Indonesia (Indonesian Institute of Sciences) LKSA Lembaga Kesejahteraan Sosial Anak (Social Welfare Institution for Children) M&E Monitoring and Evaluation MDG Millennium Development Goal(s) MoU Memorandum of Understanding NIMH National Institute for the Mentally Handicapped 4 OECD Organization for Economic Co-operation and Development PKS-ABT Program Kesejaheraan Sosial - Anak Balita Terlantar (PKSA program for neglected children under �ve) PKS-Antar/Anjal Program Kesejahteraan Sosial - Anak Terlantar/Anak Jalanan (PKSA program for neglected children/street children) PKS-ABH Program Kesejahteraan Sosial - Anak Berhadapan dengan Hukum (PKSA program for children in contact with the law) PKS-AMPK Program Kesejahteraan Sosial - Anak Memerlukan Perlindungan Khusus (PKSA program for children in need of special protection) PKS-ADK Program Kesejahteraan Sosial - Anak Dengan Kecacatan (PKSA program for children with disabilities) PKH Program Keluarga Harapan (Conditional cash transfer) PKSA Program Kesejahteraan Sosial Anak (Social cash transfer for disadvantaged children) PMKS Penyandang Masalah Kesejahteraan Sosial (People with Social Welfare Issues) PMT Proxy-means Testing PPLS Pendataan Program Perlindungan Sosial (Data collection for targeting social protection programs) PT Pos Perseroan Terbatas Pos Indonesia (National post of�ce system) PUSKA PA UI Pusat Kajian Perlindungan Anak Universitas Indonesia (Centre on Child Protection University of Indonesia) RBM Rehabilitasi Berbasis Masyarakat (Community Based Rehabilitation) RHS Right hand side (of graph) Rp Indonesian Rupiah RPJM Rencana Pembangunan Jangka Menengah (Medium-Term Development Plan, MTDP) Susenas Survei Sosio-Ekonomi Nasional (National Socio-Economic Survey) UCT Unconditional Cash Transfer UEP Usaha Ekonomi Produktif (Productive Economic Enterprise) US$ United States Dollars Yanrehsos Pelayanan Rehabilitasi Sosial (Pelayanan dan Rehabilitasi Sosial) 5 Executive Summary Direct cash transfers for vulnerable elderly and disabled populations have been provided by the Ministry of Social Welfare (Kementerian Sosial, Kemensos) since 2006; a similar cash transfer for at-risk youth was inaugurated in 2009. The Government of Indonesia’s (GOI) Pro-Poor development initiatives, international agreements and domestic laws and regulations, and considerable experience delivering more general social assistance programs led to the creation of cash transfers for these historically neglected and dif�cult-to-reach groups. These programs – Jaminan Sosial Lanjut Usia (JSLU), Jaminan Sosial Paca Berat (JSPACA), and Program Kesejahteraan Sosial Anak (PKSA) for the elderly, disabled, and youth respectively – transfer cash directly to bene�ciaries. They account for increasing shares of the Kemensos overall budget, but subsidies directed to care and rehabilitation facilities as well as direct provision of institutional care still account for a noticeable portion of the Kemensos budget for these groups. Estimates of eligible and targeted populations for these programs are not de�nitive, but overall coverage and budgeted recipients in these pilot initiatives remain quite small. In 2010, there were less than 20,000 JSPACA bene�ciaries, less than 15,000 JSLU bene�ciaries, and less than 5,000 PKSA bene�ciaries. Independent estimates of eligible populations imply current coverage levels of 1.4 percent for the elderly, 4 percent for the disabled, and less than 1 percent for at-risk youth. However, nationally representative data do not permit an estimate of the total eligible populations while the estimates of subpopulations that program implementers are using appear to be quite low. These activities were executed under DG Yahrensos which received 17 percent of Kemensos budget allocation. Of this budget, PKSA program receives the largest share of DG’s budget at 42 percent, followed by Disabled program at 35 percent and Elderly program at 12 percent, while the remaining budget was disperse among various activities including DG’s administrative activities. Outreach and facilitation are included with the cash transfer in all programs, but the capacity to effectively deliver these elements varies widely between regions; furthermore, budgets for these activities are limited and depend on local-level support and cooperation. Facilitation includes access to basic social services such as acquiring birth certi�cates, access to, and encouragement to use, local education and health providers; and routine visits to monitor clients’ well-being and cash transfer receipt and utilization. These services can be provided by workers at 6 care facilities or independent advocacy and outreach organizations, implementing agencies, or facilitators recruited by Kemensos or its regional of�ces. The responsibilities a facilitator has are broad, and often the geographic coverage of one facilitator is substantial, but facilitators are generally paid far less than professional facilitators working for Indonesia’s conditional cash transfer (PKH), for example. Average levels of education among the facilitator corps and relatively low quali�cations suggest that the outreach and facilitation provided is quite general, though enthusiastically provided. Program support operations – socialization and outreach; allocation, targeting and prioritization; monitoring and evaluation; and complaints and grievances – have very small budgets and depend crucially on cooperation and enthusiasm from local governments and facilitators. A full range of safeguarding activities is spelled out in program guidelines but these have not been institutionalized at the local implementation level. There is variation in the content, methods, frequency, completion rates, and outcomes in all safeguarding activities, and no easy-to-use reporting process that would ensure information from implementation level reaches the central funding and policy agency, Kemensos. Indicators describing whether safeguarding activities have occurred are monitored, but the indicators do not describe whether activities performed were effective or perfunctory and therefore cannot be used for any “troubleshooting� of the implementation process. To be effective for the target populations – who are neglected, unable to rely on a support network of friends, family, and community members, and confronting serious daily risks to well-being – these cash transfers need reform. The populations targeted by the JSLU, JSPACA, and PKSA cash transfers are not reliably covered by existing social assistance or social insurance initiatives and they are likely excluded from regular community-wide support services as well, so a mixed cash transfer and facilitation package is a valuable bene�t. In order for the JSLU, JSPACA, and PKSA options to be effective: (1) the mix of cash and services should be re-examined while the quality of facilitation, outreach, and triage services provided should be improved; (2) safeguarding operations, including a regular monitoring and evaluation cycle, need major upgrades; and (3) methods for searching and identifying eligible bene�ciaries need serious consideration and improvement. 7 1. Background Cash transfers and facilitated services from the Ministry of Social Affairs are delivered directly to hard-to-reach populations of vulnerable elderly, severely disabled, and at-risk youth. In Indonesia, the Ministry of Social Affairs delivers programs and services to improve welfare and outcomes for vulnerable Indonesian subpopulations, including disadvantaged youth, disabled individuals, and the vulnerable elderly.1 The Social Welfare and Rehabilitation Services of�ce (Pelayanan dan Rehabilitasi Kesejahteraan Sosial, or Yanrehsos), which is responsible for these activities, has traditionally provided services and programs via three different modes: operating grants and for-service subsidies to other non-government service providers (like orphanages, old-age homes, and foundations delivering biomedical devices and life services to disabled groups); community intervention programs; and programs delivering a mix of cash, in-kind services, and training directly to individuals or households to encourage greater participation in a productive and social community life. The last mode – direct cash transfers with facilitated services – is used in the newer household-based social assistance initiatives in Indonesia: �rst payments through the JSLU and JSPACA programs (for the elderly and disabled, respectively) were delivered in 2006 and for PKSA (for neglected children) in 2009. 1 Additional vulnerable subpopulations served are also served by the Yanrehsos of�ce, for example drug addicts, victims of violence and traf�cking, those in trouble with the law, and people without access to social resources ; the homeless and those who have been sexually exploited. 8 The new pilot cash transfer programs target dif�cult-to-reach populations currently underserved by other social assistance transfers or pro-poor schemes. The government of Indonesia has made poverty reduction, Pro-Poor development, and a “Justice for All� initiative centerpieces of its medium-term inclusive growth strategy: Presidential Instruction (Inpres) no. 3 (2010)2 stated inter alia that all public of�cials should encourage programs focused on Pro-Poor development, the Justice for All initiative, and the achievement of the Millenium Development Goals (MDGs).3 The GOI has also rati�ed (in 2006) the International Covenant on Economic, Social, and Cultural rights (ICESCR), which states individuals or groups cannot be excluded from either economic rights (for example, fair wages, social security) or social rights (for example, food, clothing, housing, access to basic education and health services ) on the basis of color, ethnicity, gender, language, political af�liation, race, religion or other status. These obligations have resulted in initiatives providing regular social services to groups that have dif�culty accessing such services as they are normally delivered.4 2 Together with Presidential Instruction no. 1/2010 on the Acceleration of the Implementation of National Development Priorities, these commitments are af�rmed into strategic statements and action plans in the RPJM (Rencana Pembangunan Jangka Menengah, or Medium-Term Development Plan) 2010-2014. Before Inpres 3 (2010), Indonesia had already made several national and international commitments to improving welfare outcomes for poor and vulnerable groups, including rati�cation of: Convention on Rights of the Child in 1990, Convention on the Elimination of All Forms of Racial Discrimination in 1999, and Convention on the Rights of Persons with Disabilities in 2007. 3 Each of the Pro-Poor Development, Justice for All, and MDG schemes are policy orientations that explicitly direct attention and resources towards achiev- ing better welfare outcomes for poorer Indonesians. 4 For example, in 2005 it was observed that a new unconditional cash transfer (see “Social Assistance Program and Public Expenditure Review 2: BLT� in this collection) was not reaching the elderly or the disabled, and a cash transfer with delivery methods revised speci�cally to be effective for these subpopulations was quickly put in place. These very same cash transfers – JSLU for vulnerable elderly and JSPACA for the disabled – are discussed in this note. They were initiated by the President of Indonesia on the International Day of the Disabled in December 2005 and inauguration included instruc- tions to the Minister of Social Affairs and all related agencies for taking actions to improve the welfare of the marginal groups. 9 JSLU, JSPACA, PKSA Cash and In-kind Transfers for at-risk youth, the disabled, and vulnerable elderly The following note summarizes quantitative and qualitative evidence in order to build a sound foundation for evaluating the cash transfer programs - JSLU, JSPACA, and PKSA - provided by Kemensos. The evidence on which the evaluation is based here is composed primarily of �rst-hand observation of the programs in operation.5 Where possible information collected from administrative records, including monitoring and evaluation reports, and from Kemensos itself, is summarized. Design features, ef�ciency and effectiveness of program implementation and operation, and impacts (intended or not) the program produces for bene�ciaries are all analyzed in as much detail as possible. Current policy planning within Kemensos assumes expansion of these programs in the coming years6, so an evaluation of the programs’ features is relevant for Indonesian policymakers and stakeholders. 5 World Bank staff observed operations and talked to Kemensos employees over a period of approximately 12 months to understand program logic and delivery processes. In addition, a team of researchers from Pusat Kajian Perlindungan Anak – Universitas Indonesia (PUSKA PA UI) also spent several days in the �eld at four locations across Indonesia interviewing bene�ciaries, providers, facilitators, and associated support staff as well as other stakehold- ers. However, the programs’ small size means they do not leave a “footprint� in the nationally representative datasets that are used in other volumes in this collection; nor have the programs been frequent topics in either private or commissioned analytical research. For this reason, this note provides a preliminary investigation that inevitably will raise more questions than are answered conclusively. 6 In addition, medium-term development policy planning documents (RPJM) also have poverty reduction and the expansion of social safety net programs as explicit goals. 10 2. Objective, Program Size and Bene�t Adequacy Kemensos has developed cash-based and hybrid cash-and-services-based transfers for marginal groups including the severely disabled, vulnerable elderly, and at-risk youth. PKSA operates as a conditional cash transfer program; payments began on a pilot basis in 2009. PKSA combines youth savings accounts with facilitated, in-kind assistance to children and families.7 Facilitators encourage bene�ciaries to use the cash transfer to ful�ll basic needs and access social services (birth certi�cate, education, health, shelter) while strengthening knowledge and good practice among parents and caretakers. The cash transfer amount was determined by consulting daily per-capita subsidies provided by Kemensos to facilities providing care for neglected children: the Rp 6,000 daily rate is equivalent to Rp 1.8 million per year. Program designers consider this to be adequate for consumption of an appropriate food basket and school supplies (including proper clothing). PKSA conditionalities, or behaviors for which bene�ciaries are responsible, include: positive behavior change, increased social function, and increase in utilization of basic social services. PKSA parenting information packages encourage caretakers to adopt preventative strategies to keep children from neglect and risky behavior. PKSA provides rehabilitation and social protection for children outside the institutional care network with the explicit goal of helping well-adjusted children stay with their families, communities, and local networks.8 JSLU is a cash transfer program that targets elderly with diminished potential for supporting themselves or for receiving support through family and social networks; payments began on a pilot basis in 2006. The characteristics (see Table 3 below) of a JSLU bene�ciary include income poverty as estimated by a proxy-means score and indicators summarizing inability to provide for oneself; reports from the subdirectorate in charge of the JSLU program indicate that the “bedridden� characteristic is referred to most often to determine JSLU eligibility status. JSLU’s objectives include ful�lling basic needs and protection and promotion of physical, psychological, and social rights. The program delivers Rp 300,000 per month to its bene�ciaries. In 2010 the program reached approximately 10,000 bene�ciaries in 143 districts across 29 provinces. JSPACA delivers cash transfers of Rp 300,000 per month to severely disabled individuals; payments began in �ve provinces in 2006. Much like JSLU, eligible JSPACA bene�ciaries exhibit characteristics indicating that they have dif�culty raising income and that they depend on others for accomplishing daily activities. Reports from the subdirectorate in charge of the JSPACA program indicate that the “severely disabled� are prioritized for JSPACA as long as they are not already receiving income support; disability severity is judged primarily with reference to physical disability and physical inability to participate in the production of one’s own needs. JSPACA’s objectives include enabling bene�ciaries to ful�ll nutrition and health care needs as well as encouraging social functioning and community integration. JSPACA provides Rp 300,000 per month to its bene�ciaries. In 2010, JSPACA was delivered to approximately 17,000 bene�ciaries in 184 districts in 31 provinces. 7 Within the PKSA program, there are 5 subprograms, each with its own targeted bene�ciary pro�le. There is PKSA for (1) “neglected (under 5 years old)� or PKS-ABT (Program Kesejahteraan Sosial - Anak Balita Terlantar), (2) “street children and neglected children (5- 18 years old)� or PKS-Antar/Anjal (Program Kesejahteraan Sosial-Anak Terlantar/Anak Jalanan), (3) “children in contact with the law� or PKS-ABH (Program Kesejahteraan Sosial-Anak Berhadapan dengan Hukum), (4) �children with disabilities� or PKS-ADK (Program Kesejahteraan Sosial-Anak Dengan Kecacatan); and (5) “children in need of special protection� or PKS-AMPK (Program Kesejahteraan Sosial-Anak Memerlukan Perlindungan Khusus). Table 3 below summarizes eligibility characteristics for each of the programs discussed in the note. 8 Older modes of child protection in Indonesia relied almost exclusively on social services provided by homes for neglected children and victims of violence. 11 JSLU, JSPACA, PKSA Cash and In-kind Transfers for at-risk youth, the disabled, and vulnerable elderly Table 1: JSPACA JSLU PKSA Yanrehsos Of�cial name: Jaminan Sosial Jaminan Sosial Program cash Penyandang Cacat Lanjut Usia Kesejahteraan transfer Berat Sosial Anak programs Program type: Cash transfer (permanent, tax-�nanced) at a glance9 Inaugural year 2006 2006 2009 Coverage: 31 provinces, 184 29 provinces, 143 24 provinces districts districts Number of bene�ciaries (2010): 17,000 10,000 4,187 (estimated)* Of�cial value of bene�t: Rp 300,000 per Rp 300,000 per Rp 1.3 - 1.8 million month for 12 months month for 12 months per year Public expenditure (2009) Rp 4,620 billion (US$ 446 million) Administrative cost per recipient 522,169 576,390 199,882 Percent of poor households 4% poor and 1.4% poor and 1.6% poor and covered (2009)** disabled vulnerable elderly neglected children Key policy and executing agency: Ministry of Social Affairs (Kemensos) Key implementation agencies Kemensos, Regional of�ce (Dinsos), PT.Pos (role): Support operations partners (role) PT Pos (funds distribution to bene�ciaries); Ministry of Finance (Kemenkeu) (silent fund disbursement); PKSA implementing agencies (LKSA) Local Government participation Socialization, card distribution, monitoring and evaluation Source and Notes: Kemensos, Kemenkeu, and World Bank staff calculations. *Estimated number of PKSA bene�ciaries is based on PUSKA PA UI �eld research data. **The coverage number is estimated based on Kemensos data dan Susenas 2009. Together and individually JSPACA, JSLU, and PKSA intend to improve bene�ciary and community welfare by helping bene�ciaries acquire and consume basic necessities and basic social services including health and education10 as well as to encourage and assist in daily functions and quality of life issues that are otherwise at risk. PKSA’s conditionalities are explicitly related to re-integrating bene�ciaries into the regular education system and re-socialization.11 JSLU and JSPACA bene�ciaries are instructed in the appropriate use of funds, but the programs also hope to achieve better psychological functioning, social functioning, and community integration of bene�ciaries by increasing self-con�dence and through increased acceptance of disability and its characteristics in family, community, and social networks. JSPACA, JSLU, and PKSA attempt to intervene before bene�ciaries need more intensive services or care (Figure 1). Potential bene�ciaries – at-risk children or elderly individuals and the severely disabled – are not currently being assisted or can not effectively utilize regular social services like health, education, or social support provided universally to all Indonesians (the “Primary Level� in Figure 1 below). Additionally, potential bene�ciaries lack access to the support networks in family, kin, and local community, including government-delivered programs like PKH or Jamkesmas, (the “Secondary Level� in Figure 1 below) that might otherwise intervene or assist with loans or in-kind assistance, child care, or contacts and advice when dif�culties arise. In fact, potential bene�ciaries are identi�ed partially with reference to a lack of primary and secondary care. The “Tertiary Level�, where JSPACA, JSLU, and PKSA12 operate, attempt to �ll in when universally-provided or privately-available care and assistance are unavailable; they intend to improve bene�ciary and 9 These programs are by and large direct cash transfers to bene�ciaries. However, each program and each directorate varies in its reliance on institu- tions (either state-owned and operated or private) and the human resources available in the institutions to deliver these bene�ts. When programs are delivered through institutions, and when programs rely on institution staff, we make note of it, but we do not attempt a full review of practices or the level and quality of care in the institutions themselves. For reviews of the care provided by institutions, please see “Someone that matters; The Quality of Care in Childcare Institutions in Indonesia (Save the Children, Unicef, & MoSA, 2007). 10 In addition, helping bene�ciaries acquire birth certi�cates, shelter, clean water, skills and other facilitated behaviors are mentioned in program manuals as facilitated activities. 11 Nonful�llment of PKSA’s conditionalities may result in reduction of bene�ts or exit from the program, but PKSA is not a fully-fledged conditional cash transfer with all relevant machinery and processes. 12 Though Figure 1 was developed to summarize the logic of the PKSA program in particular, it also summarizes the logic behind the JSLU and JSPACA interventions as well. At the “Tertiary� level in Figure 1, the reader can substitute “Elderly protection services� or “Disabled protection services� in the red circle that surrounds “Child Protective services� when thinking about the JSLU or JSPACA interventions (respectively). 12 community welfare by helping bene�ciaries acquire and consume basic necessities; basic social services including health and education; and to encourage and assist in daily functions and quality of life issues that are otherwise at risk. Figure 1: Intervention PRIMARY LEVEL logic UNIVERSAL SECONDARY TERTIARY LEVEL TARGETED LEVEL INDIVIDUAL Child Protective services Education Family support- Information, and Early intervention Intensive family sensitisation support Out-of- Home Care CONTINUUM OF CARE Intensity of risk Source: PUSKA PA, UI, 2011 JSPACA, JSLU, and PKSA command a large share of Kemensos resources: currently 33 to 46 percent of each subdirectorate’s budget is accounted for by the cash transfers alone (not counting administrative costs). The total spending of these three subdirectorates (Child, Disabled, and Elderly Social Services) account for nearly 90 percent combined of the Directorate General under which they operate. The programs’ collective share has been consistently increasing and should be expected to increase further given current policy stances. In other words, the Child, Disabled, and Elderly Social Services subdirectorates, and the Directorate General under which they operate - Pelayanan dan Rehabilitasi Kesejahteraan Sosial, or Yanrehsos - are increasingly dedicated to delivering cash transfers. Yanrehsos spending is relatively minor part of Cluster 1 social assistance spending, but represents a sizable share of all Kemensos spending. Spending on all Yanrehsos activities rose substantially in 2006, increasing by around three-quarters in real terms, reflecting the launch of various social cash transfer initiatives including JSLU and JSPACA (Figure 2). Since then, however, nominal spending has increased only marginally each year, with the result that real spending declined by 30 percent between 2006 and 2010.13 In 2010, Rp 559 billion (US$ 56 million) was spent on Yanrehsos activities and staff: this is equivalent to around 17 percent of total Kemensos spending and just 4.3 percent of the Central Government’s total expenditure on Cluster 1 social assistance transfer programs. 13 This reflects high rates of inflation in Indonesia over the same period. 13 JSLU, JSPACA, PKSA Cash and In-kind Transfers for at-risk youth, the disabled, and vulnerable elderly Figure 2: Rp billion Percent 1,000 32 Spending on the Yanrehsos Program/ 750 24 Directorate General (Realized 500 16 spending in Rp billion, nominal 250 8 and in constant 2009 prices, 2005-2010) 0 0 2005 2006 2007 2008 2009 2010 Nominal (LHS) Constant 2009 prices (LHS) Share Kemensos spending (RHS) Share of Social Assistance spending (RHS) Sources and Notes: World Bank staff calculations based on Directorate General and BPS data. Nominal spending is deflated using the GDP deflator. Figure 3: 20 Number of Bene�ciaries: JSLU, JSPACA, 15 and PKSA Thousands 10 5 0 2006 2007 2008 2009 2010 PKSA JSLU JSPACA Source: Kemensos. Number of bene�ciaries of PKSA is based on assessment on the implementation of PKSA by PUSKA PA UI. The Yanrehsos budget is divided between �ve activities each of which provides numerous interventions. Each of the �ve activities is managed and administered by a separate subdirectorate (Table 2 and Figure 4). The largest intervention in the Child Social Services subdirectorate is a direct cash transfer to orphanages providing assistance to 145,000 children; it accounts for over half of all Child Social Services spending. The next largest intervention provides cash and in-kind assistance to non-institutionalized children delivered by social workers, drop-in centers and institutions (including two state run orphanages). In 2009 part of this activity was “converted� into the PKSA conditional cash transfers which are payable directly to bene�ciaries and with social workers facilitating the transfers. Disabled and vulnerable elderly subdirectorates have also expanded activities including by piloting and then increasing allocations for the JSPACA and JSLU cash transfers. These subdirectorates also �nance social services delivered by social workers, community-based organizations14 and institutions. Notably, the disabled subdirectorate funds 18 institutions for the disabled throughout Indonesia as well as a Braille printing house. Finally, minor interventions targeting the homeless and drug abuse victims accounted for the remaining 11 percent of Yanrehsos spending. 14 These include mobile rehabilitation units, sheltered workshops and community-based rehabilitation. 14 Table 2: Share in Spending 2005* 2006 2007 2008 2009 2010 2010 (%) on the Child Social Services 301 240 338 296 235 42 Yanrehsos Program by o/w cash transfer for orphanages 0 0 136 159 0 0 Activity/ Directorate Disabled Social Services 186 194 206 217 194 35 (Realized o/w cash transfer for severely spending in 21 31 43 67 0 0 disabled Rp billion, Elderly Social Services 76 73 75 82 69 12 constant o/w cash transfer for abandoned 2009 prices, 14 19 21 38 0 0 elderly 2005-2010) Homeless Social Services 43 40 39 39 30 5 Drug Abuse Social Services 40 42 37 35 26 5 Unclassi�ed 155 175 0 0 0 0 Total Program/Directorate General 453 801 765 694 669 554 100 % change n.a. 77 -5 -9 -4 -17 % of total Kemensos spending 14 25 22 20 21 17 % of total Social Assistance 2.4 1.9 4.2 1.9 2.4 4.3 spending Source: Kemenkeu, Kemensos and World Bank staff calculation. * 2005 data is undetailed up to program level. Figure 4: 1000 Rp billion Spending on the Yanrehsos Program 750 by activity/ subdirectorate* 500 (Share of total program 250 spending, 2005- 2010) 0 2005** 2006 2007 2008 2009 2010 Children Disabled Elderly Homeless Drug Abuse Victims Split unavailable Sources and Notes: World Bank staff calculations based on Directorate General and BPS data. *While original budget data allocates personnel expenses, of�ce maintenance and the DG’s Secretariat as an overhead cost for the DG as a whole, expenses for each individual Activity/Directorate have been estimated based on staff numbers and budget allocations. **Budget allocation data rather than realized spending, and no breakdown available. Bene�t levels for JSPACA and JSLU are generous relative to other Cluster 1 transfers… Transfer levels in JSLU and JSPACA - Rp 300,000 per month - are approximately one and one-half times the March 2010 poverty line. JSLU and JSPACA bene�ts are more than double the average bene�ts paid to households participating in the PKH conditional cash transfer program (see �Social Assistance Program and Public Expenditure Review 6: PKH� in this collection), but it is worth noting that a full year of PKSA bene�ts would not quite be enough to fund a complete year of education at a public junior high school (see �Social Assistance Program and Public Expenditure Review 5: BSM� in this collection) though returning to the education system is a conditionality in the PKSA program.15 15 PUSKA PA UI (2011). 15 JSLU, JSPACA, PKSA Cash and In-kind Transfers for at-risk youth, the disabled, and vulnerable elderly …but elderly and disabled subpopulations face higher costs of living. However, in addition to an earnings handicap, both disabled and elderly populations likely face a higher cost of living due to higher levels of non-optional expenditure in certain categories like transportation, health care, assistive devices, personal assistance, and dwelling adaptation.16 For example, a national survey in India estimated average recurring costs of being disabled were alone approximately equal to the of�cial poverty line. In Bangladesh, 57 percent of families with disabled children reported extra direct costs for specialized child care, medicine, and health care; a child with severe impairments was found to be three times more expensive to raise than a child without disabilities.17 The amount of the increase in living costs also depends on the supply, leading to a perverse result: the disabled and elderly may experience no increases in spending where required goods and services are not available or are too expensive. In such areas the increased cost of living is not a monetary surcharge but is observed in increased effort accomplishing daily activities, increased morbidity, and increased mortality. Kemensos calculates JSLU, JSPACA, and PKSA bene�ts levels based on subsidies paid to institutional care providers rather than on estimates of the increased cost of living (relative to an able-bodied individual) for the non- institutionalized bene�ciary, making it dif�cult to summarize whether bene�ts are adequate. In addition, real transfer values have eroded due to a lack of adjustment for inflation. Like most other cash transfer programs in Indonesia, bene�t levels for JSPACA and JSLU have remained unchanged since the launch of the initiatives in 2006. This has resulted in a 29 percent decline in the real purchasing power of transfers between 2006 and 2010 (using poverty basket inflation). The lack of indexation of bene�ts to account for increases in the cost of living is unsustainable and could undermine the objectives of the programs. Figure 5: Rp 500,000 Evolution of bene�t levels 400,000 for JSPACA and JSLU over time 300,000 200,000 100,000 0 2006 2007 2008 2009 2010 Nominal value of monthly benefit Real value of monthly benefit (Constant 2010 prices) Sources and Notes: Kemensos, BPS and World Bank staff calculations. Bene�ts are deflated using the Poverty Basket Index. Yanrehsos has shifted program implementation away from community-executed towards centrally-executed spending while subsidies to institutions and rehabilitation centers has remained relatively stable. Traditionally Yanrehsos has used three channels for implementation (see Figures 5 and 6): (1) direct central execution by Kemensos, for example cash transfers originating in the central budget, transferred to Indonesia’s postal service agency PT Pos, then delivered directly to recipients; (2) local execution by social workers and community organizations, for example, mobile rehabilitation units, supported by deconcentration funds for provinces and districts; and (3) spending in the form of subsidies executed by institutions. Lately the shift towards centrally-executed spending has picked up pace, reflecting the increasing focus on cash transfers18, and the centrally-executed share of total program spending rose from 46 percent in 2007 to 61 percent in 2010 (Table 6 and Table 7 in Annex). This trend is broadly based: around 75 percent of spending on children and the elderly was centrally executed in 2010 (up from 24 and 45 percent in 2007), while 45 percent of spending on the disabled was centrally executed (up from 32 percent).19 Community-based spending has declined in parallel with this development: real spending fell by more than 80 percent between 2007 and 2010, reducing the share of spending from 31 percent to 8.5 percent over the same period. Meanwhile, spending executed by institutions has remained relatively stable, and accounted for almost 30 percent of total spending in 2010. Changes to implementation 16 See Mitra, Posarac, and Vick (2011). 17 See NIMH (2004) and Chowdhury, J. (2005) . There are no contemporaneous estimates of the additional expenditures associated with caring for either a disabled or vulnerable elderly individual in Indonesia. 18 This shift also reflects, in part, a policy to encourage greater cost sharing by sub-national governments. 19 The lower ratio for the disabled activities reflects a greater predominance of institution-provided care. 16 arrangements in the future are expected, as Kemensos plans to decentralize management and execution of JSPACA to district governments. The resulting increase in deconcentration funds would partially fund the district-implemented program. Figure 6: 100 Spending on the Yanrehsos 80 Program by executing Percent 60 agency 40 20 0 2005** 2006 2007 2008 2009 2010 Centrally executed Institutions & rehab. centers (Pantis) Split unavailable Community executed via deconcentration funds Sources and Notes: World Bank staff calculations based on Directorate General and BPS data. Nominal spending is deflated using the GDP deflator. **Budget allocation data rather than realized spending, and no breakdown available. 17 JSLU, JSPACA, PKSA Cash and In-kind Transfers for at-risk youth, the disabled, and vulnerable elderly Figure 7: Flow Ministry of Social of funds within Affairs (Kemensos) the Yanrehsos Program Directorate General of Social Services and Rehabilitation Directorates (Children, Disabled, Secretariat* Elderly etc) Central Deconcentration Integrated Social funds funds Services Unit (UPT) funds Other functions & Cash tranfers Social workers Social Institutions & interventions (UCTs) (pendamping) rehabilitation centers Community- PT Pos Indonesia based organisations Provincial Pos Inspector District Pos Inspector Sub-district Pos Of�ce (Payer) Targeted bene�ciries and/or their representatives Sources and Notes: World Bank staff based on DG Manuals. *Disburses civil servant salaries at the central level and helps supervise deconcentration funds. 18 3. Targeting Coverage of PKSA, JSPACA and JSLU is small and not proportional to estimated need; bene�ts provided include cash, facilitation, and recruitment into additional social services. It is dif�cult to determine the size of the eligible populations for JSLU, JSPACA, and PKSA at either the national or regional levels (see Box 1 below). External researchers found that PKSA pilot provinces were identi�ed based on the availability of institutions and care organizations (who would become PKSA implementation partners) rather than by attempting to determine where eligible bene�ciaries would be concentrated. There is no nationally representative dataset with the information necessary to enumerate the JSPACA-, JSLU-, or PKSA-eligible populations according to characteristics that Kemensos uses to identify bene�ciaries. As a proxy, the target population for JSPACA is here estimated as those individuals who are disabled and living below the poverty line and the JSLU target population is estimated as those individuals 60 yrs or older who are poor and categorized as vulnerable based on PMKS (Penyandang Masalah Kesejahteraan Sosial or People with social welfare issues) criteria.20 Data limitations preclude calculation of coverage and incidence for the PKSA program. JSPACA, JSLU, and PKSA together cover far less than one-tenth of one percent of Indonesia’s 2010 individual population, and approximately one-half of one percent of Indonesian households. Both JSLU and JSPACA are present almost in all provinces in Indonesia (29 and 31 provinces for JSLU and JSPACA respectively) while PKSA is still essentially a pilot program present in 24 provinces. Kemensos has directed approximately equal absolute shares of the total program resource envelop across all Indonesian provinces, leading to large variation in per-capita coverage and incidence rates by province (see Figure 7 and 8 below)21. Figure 7 and 8 also shows that the expected incidence of either disabled or “poor and elderly� status does not vary widely by province. Figure 8: 25 JSPACA Coverage and 20 Incidence by Percent province, 2009 15 10 5 0 Aceh Riau Jambi North Sumatera West Sumatera South Sumatera Bengkulu Lampung Kep. Bangka Belitung Kep. Riau DKI Jakarta West Java Central Java DI Yogyakarta East Java Banten Bali West Kalimantan Central Kalimantan South Kalimantan East Kalimantan North Maluku West Nusa Tenggara North Sulawesi Central Sulawesi South Sulawesi Gorontalo Maluku West Papua Papua East Nusa Tenggara Southeast Sulawesi West Sulawesi Coverage Incidence Probability Disabled Source: Susenas 2009 & JSPACA administrative data from Kemensos. 20 See Box 1 below for an explanation of the PMKS data module and Kemensos procedures for estimating the size of eligible bene�ciary pools. 21 From administrative records, it appears that there are roughly three levels of province quota. Provinces with larger populations get the largest-sized quota while provinces with smaller populations get the smallest-sized quota, so there is some proportionality in quota assignment. Bene�t amounts do not vary, so roughly equal shares of overall bene�ts means roughly equal absolute numbers of bene�ciaries across provinces. Flat bene�ts also means that the number of bene�ciaries covered is typically determined by overall program budget (for transfers) divided by the per-bene�ciary amount trans- ferred. 19 JSLU, JSPACA, PKSA Cash and In-kind Transfers for at-risk youth, the disabled, and vulnerable elderly Figure 9: JSLU 25 Coverage and Incidence by 20 province, 2009 15 Percent 10 5 0 Maluku North Maluku West Papua Papua Aceh Riau Bengkulu Lampung Kep. Bangka Belitung Kep. Riau DKI Jakarta West Java Central Java DI Yogyakarta East Java Banten Bali West Kalimantan Central Kalimantan South Kalimantan East Kalimantan Gorontalo Jambi West Sumatera West Sulawesi North Sumatera South Sumatera South Sulawesi West Nusa Tenggara East Nusa Tenggara North Sulawesi Central Sulawesi Southeast Sulawesi Coverage Incidence Probability Poor Elderly Source: Susenas 2009 & JSLU administrative data from Kemensos. Average coverage in the JSPACA initiative is 4 percent but the poor disabled living in Maluku and Aceh are covered at rates of less than 0.5 percent while those in Jambi and DKI Jakarta are covered at rates exceeding 20 percent.22 For JSLU, the average coverage rate is 1.4 percent but South Kalimantan and North Maluku are covered at rates exceeding 15 percent.23 Provinces in Eastern Indonesia24 received nearly 25 percent of all JSLU bene�ts but represent approximately 10 percent of the overall Indonesian population. 22 The provinces appointed to implement JSPACA in the its �rst year (2006) are West Java, Central Java, Yogyakarta, West Sumatra, and South Sumatera. These pilot areas were de�ned based on the number of disabled population from BPS using PMKS data from 2003. However, discrepancies between �eld-based observation and BPS data are likely as Susenas records only contain information regarding “regular� disabilities. Susenas does not observe the additional criteria that JSPACA uses to identify bene�ciaries. In 2007, becoming a pilot JSPACA area was a possibility for all districts, but only districts providing data on potential bene�ciaries, as well as an agreement allocating budget and staff time for data collection and additional facilitators, were chosen. The number of bene�ciaries added is determined by the quality of incoming applications and the willingness of local government partners to support implementation. 23 In 2006, �ve provinces in Java were appointed as the �rst JSLU pilot areas. Java was chosen as support facilities and infrastructure, as well as PT Pos readiness, was thought to be higher than in other regions. 24 South Sulawesi. Southeast Sulawesi, Maluku, Papua, and East Nusa Tenggara. 20 Box 1: Baseline Currently there are no de�nitive, authoritative, or consistently-estimated data resources, at either local estimates for or national levels, which reliably capture the frequency of the subpopulations targeted by cash transfers vulnerable (see Table 3 for eligibility criteria). In 1998 (with updates in 2000 and every 3rd year following) the na- subpopulations tional statistics agency (BPS) surveyed households with social welfare problems (Penyandang Masalah Kesejahteraan Sosial, PMKS) as part of its household welfare survey (Susenas). Kemensos referenced in Indonesia these PMKS populations to estimate 2003 populations of neglected children (4.6 million, or 6 percent of all Indonesians below age 18), elderly (2.4 million, or 15 percent of all Indonesians aged 60 yrs and above), and disabled (3 million or 1.4 percent of all Indonesians).25 By 2009, the PMKS module in Suse- nas estimated an absolute decrease in the disabled population to 2 million or 0.9 percent of the total population. Kemensos’s in-house data center reports an even lower number of disabled persons (1.5 million) for 2009, while non-government sources report much higher overall rates of disability at nearly 10 percent or 23 million individuals.26 Internationally, large countries at all income levels estimate much higher disability rates than does In- donesia. In a recent update, using only high-quality data, of disability prevalence in 15 developing countries in Sub-Saharan Africa, Asia, and Latin America, rates among working-aged individuals ranged from lows of 7.4, 7.6, and 8.5 percent (Mexico, Pakistan, and Kenya, respectively) to highs of 16.8, 19.6, and 21.5 percent (Malawi, Bangladesh, and Brazil, respectively).27 In the United States in 1991, disability prevalence ranged from approximately 6 percent for under-18 year olds to nearly 30 percent for 45-64 year olds. Disability prevalence tends to rise with income: from under 5 percent (generally in low-income African countries) to 20 percent or more for some OECD countries.28 Internationally, disability prevalence is frequently, though not always: (1) higher in rural areas than urban, (2) higher among females than among males, and (3) higher among poorer individuals than non-poor individuals (across several differ- ent poverty measures).29 Given different de�nitions (of both disability and the severity of disability) and the methods and in- struments used to record such characteristics, estimated rates can be expected to vary over different sources. At-risk youth and vulnerable elderly subpopulations are similarly dif�cult to estimate as they are often dif�cult to locate and have much higher “miss� rates in national household surveys. Even so, GOI estimates (of disability prevalence at least) for Indonesia appear to be an entire order of magnitude lower than average rates estimated for many different types of countries in many different regions and many different income levels. Compounding this dif�culty, JSPACA, JSLU, and PKSA are prioritized to individuals with speci�c characteristics that do not precisely correspond to any records in, for example, the Susenas household survey. This makes a reasonable guess about the size of the potential bene�ciary pools doubly dif�cult. Eligibility criteria are numerous, complex and not well understood by the local-level staff responsible for �nding and nominating potential bene�ciaries. Eligibility criteria (see Table 3 below) were designed so that transfers would reach the neediest households and so that the negative publicity and misunderstanding surrounding targeting that greeted the BLT program in 2005 (for example) would be avoided. For JSPACA and JSLU, the more criteria that apply to a potential recipient, the higher the priority of that recipient on the �nal recipient list. For PKSA, each group of characteristics represents a sub-program (PKSA for abuse victims, for example) and any one, but not necessarily all, of the listed characteristics may apply to a bene�ciary of that program. 25 The PMKS variables which were used to estimate the incidence of vulnerable elderly, disabled, neglected, and Fakir Miskin were mutually agreed by BPS, BKSN (Badan Kesejahteraan Sosial Nasional), Bappenas, and (Lembaga Ilmu Pengetahuan Indonesia) LIPI. PMKS criteria are associated with socio- economic conditions, for example criteria for neglected children and elderly are: never enrolled in school/have not �nished 9-year basic education, do not have proper meals (eat less than 14 times/week, consume meat/�sh/egg no more than twice/week), do not have enough clothes (less than 4 pairs), do not have proper place to sleep, are not able to get health treatment when needed, have to work to earn family income, and orphaned (only for neglected children). The children/elderly are categorized into not-neglected, nearly neglected, and neglected for satisfying less than 2 criteria, 2 criteria, and more than 2 criteria respectively. See BPS/KemenSos publication on: Analisa deskriptif Penyandang Masalah Kesejahteraan Sosial 2006. 26 United Cerebral Palsy and Wheels for Humanity pamphlet; www.ucpwfh.org. 27 LMitra, Posarac, and Vick (2011) 28 Loeb, Eide, and Mont (2008) 29 Loeb, Eide, and Mont (2008) 21 JSLU, JSPACA, PKSA Cash and In-kind Transfers for at-risk youth, the disabled, and vulnerable elderly Table 3: JSPACA JSLU PKSA Summary 1. Severely disabled who depend 1. 70 years or older 1. Abuse victims, abandoned by of on others to support their the parents/family, vulnerable Eligibility daily activities. 2. 60 years and older if chronically street children. Criteria of ill & depending on others for JSPACA, 2. Without �xed income support in daily activities 2. In violation of law, being JSLU, PKSA support. processed judicially, or serving a 3. Bedridden. criminal sentence. 3. Prioritized for disabled who are poor and/or not members 4. Do not have �xed income support. 3. With physical and/or mental of care institutions. disabilities. 5. Not a disabled or person who 4. Are not currently receiving received a permanent assistance 4. Victims of traf�cking/ other assistance from from government. exploitation, from minority/ government. isolated groups, or HIV/AIDS 6. Other PMT criteria.25 positive, or drugs abusers. In JSLU and JSPACA, local Kemensos employees (from the regional Kemensos of�ces called Dinas Sosial) together with enumerators collect data on potential bene�ciaries. These pro�les are reviewed by Kemensos at both district and central level (see �gure 9). Kemensos in Jakarta applies a scoring system to the data in the pro�les and potential bene�ciaries are ranked according to that score. In the early years of the JSPACA and JSLU programs, �nal bene�ciary selection was made by Kemensos on the basis of this ranking; however JSLU has delegated �nal bene�ciary selection back down to regional of�ces. Reports from JSLU and JSPACA staff in Kemensos indicated that the scoring system and associated criteria are not widely known or well understood by the local program of�cials responsible for making initial nominations. 30 The criteria included in the Proxy Mean Test (PMT) used are: do not have proper meals (eat less than 2 times/day, consume meat/�sh/egg no more than twice/week), do not have enough clothes (less than 4 pairs), do not have proper place to sleep. 22 4. Potential Impacts Impacts are not known, but bene�ciaries appreciate the combination of cash transfer and facilitated services. Bene�ciaries report that cash transfers are appreciated and are spent almost immediately to purchase basic necessities like food, milk, health care, transportation, and clothes while JSPACA bene�ciaries purchased medical equipment.31 When late disbursements and subsequent bunching of fund transfers made it possible (see below), bene�ciaries reported making house renovations or rent payments, purchasing televisions and other assets, and paying down debt. Internal monitoring and evaluation reports from the 2009 JSPACA program indicated that 92 percent of bene�ciaries bought food, 11 percent paid for health services or assistive devices, 17 percent bought clothing, and 18 percent bought assets (usually precious metals or livestock). Only 6 percent of JSPACA facilitators interviewed stated that bene�ciaries utilized funds improperly. PKSA bene�ciaries indicated that common uses of funds were to buy food (nearly 60 percent of bene�ciaries), pay for festive activities and religious holiday observances (14 percent), obtaining birth certi�cates (16 percent), and for business capital or debt payments (12 percent). Over 70 percent of PKSA bene�ciaries mentioned positive bene�ts from PKSA cash support. Some PKSA families indicated that they depend on PKSA to educate children who are otherwise not involved in school. Direct observation of JSPACA and JSLU bene�ciaries revealed similar sentiments: cash transfers help individuals ful�ll daily needs and without such transfers, they would be forced to ask for assistance from family and relatives who are often also poor. Figure Other 17.9 10: Funds utilization: Handicap tools (e.g.wheelchair) (0.2) JSPACA Rent (0.2) Clothes 17.3 Health care 10.8 Nutritional food 92.4 Source: Internal M&E JSPACA - Kemensos 2009 31 There is no trace of these programs in the secondary data sources (like the Susenas household surveys) used to measure impacts in other programs covered in the Indonesia Public Expenditure and Program Reviews (Reviews 2 through 5 in this collection) and no consolidated administrative data that is ready or appropriate for analysis. The available �rst-hand, directly-observed qualitative information (from bene�ciaries, program implanters, and other stakeholders) is summarized below. 23 JSLU, JSPACA, PKSA Cash and In-kind Transfers for at-risk youth, the disabled, and vulnerable elderly Figure11: Funds Religious festive utilization: Business capital PKSA Family debt Transportation School related costs Birth certificate Toys and education Health care Nutritional needs Source: PUSKA PA UI �eld research data PKSA cash transfers are often controlled by pendamping (facilitators) or LKSA staff. PKSA bene�ciaries usually receive cash payments – as a lump sum, as several staged payments, or as payouts based on an assessment of needs – only after consultation with and implicit approval from a social worker.32 Field research also uncovered a few instances of local neighborhood leaders or others in positions of authority (legitimate or otherwise) co-opting the control of PKSA funds. There is no guarantee that these (legitimate or illegitimate) fund managers are aware of individual or household needs; direct observation revealed a lack of thoughtful assessment of needs and instead a reliance on previous assumptions and experiences.33 Nonetheless, some parents ultimately preferred to have fund managers: they considered LKSAs to be better informed regarding both the types of goods, services or activities’ that PKSA funds could be used for as well as decisions regarding how much of a PKSA disbursement to save.34 Very little is known about the quality and effectiveness of the facilitator-provided services that accompany these cash transfers. The facilitator is responsible for making regular visits intended to provide constructive assistance such as access to basic social services like free health care, birth certi�cate or identity registry, and education (for PKSA bene�ciaries) and monitor the client’s condition as well as funds utilization. The facilitator is also expected to be able to give motivational support to increase bene�ciaries’ self esteem. While �eld research and interviews with Yanrehsos of�cials indicate that the number of well-trained facilitators is growing, and that facilitators themselves acknowledge their lack of adequate training, the quality and frequency of facilitated services still varies widely. PKSA (and also JSLU and JSPACA) often relies on the efforts of local-level volunteers to deliver both cash and facilitation. Facilitators (volunteer or otherwise) do not always have the Yanrehsos-desired minimum education level. As a result facilitators are not always equipped to address the complex physical, mental, and social dif�culties that bene�ciaries present; this is compounded by the lack of a triage system within PKSA (and also JSLU and JSPACA) that could direct the worst-off or most complex cases to agencies and personnel with the relevant expertise. An additional complication is that several bene�ciaries under one facilitator’s care may be spread across great geographic distance, which means a facilitator may spend most of his or her time traveling to bene�ciary households and performing only a perfunctory check before having to begin travel to the next site.35 32 The PKSA program for neglected children under 5 years of age (PKS-ABT) provides part of its transfer as supplementary nutrition packages. It uses day care centers to distribute this portion of the assistance. 33 PUSKA PA UI (2011). 34 PUSKA PA UI (2011). 35 Indonesia’s conditional cash transfer (PKH) also relies heavily on facilitators for motivation, to encourage compliance with conditions, and to ease access to social services. PKH Facilitators indicate that the quality and intensity of services they have time to provide is limited when large distances separate bene�ciaries while all stakeholders note the quality and frequency of facilitators efforts are the best predictors of household success under the PKH program. See “Social Assistance Program and Public Expenditure Review 6: PKH� in this collection. 24 5. Cost Effectiveness Assessing spending ef�ciency in the Yanrehsos program as a whole is dif�cult due to data limitations and the complexity of comparing vastly different interventions. Administrative costs in the numerous Yanrehsos interventions can be very different and dif�cult to compare, making high-level data and budget classi�cations uninformative. For example, just over half of Yanrehsos program spending is classi�ed as social assistance with the rest split between goods and services, salaries and capital expenditures (Figure 12). However, this breakdown disguises large variations: centrally-executed spending – which is increasingly dominated by cash transfers – is now largely classi�ed as social assistance, whereas institution-executed spending – which delivers services rather than cash – is almost entirely classi�ed as salaries or goods and services. Second, while centrally-executed spending is documented in annual �nancial accountability reports (Laporan Akuntabilitas Keuangan Pemerintah or LAKIPs), reporting on supporting activities or stand alone interventions executed by community organizations, institutions and local governments is sparse and not readily accessible.36 This makes it dif�cult to build a comprehensive picture of total program administrative costs. Figure 12: 100% Spending on the Social Services and 75% Rehabilitation Program by economic 50% classi�cation (Share of total 25% spending, 2006- 2010) 0% 2006 2007 2008 2009 2010 Social Assistance Goods & Services Salaries Capital Sources and Notes: World Bank staff calculations based on Directorate General data. Staff and systems are shared across multiple functions and interventions making the ef�ciency of spending dif�cult to determine. Even for centrally-executed interventions, it is dif�cult to isolate staff and administrative costs associated with each intervention. First, civil servant salary expenses are not recorded under the budget – nor reported in the LAKIP – of each individual Directorate, but grouped together and recorded under the budget of the Directorate General’s Secretariat, necessitating estimates based on staff numbers. Second, staff and systems within each Directorate are shared across both policy and implementation functions and across multiple interventions, all of which are not strictly comparable. Thus, ef�ciency indicators for an entire Directorate are simply an average of numerous activities and interventions and may not be meaningful (Table 6 in Annex). While there may be some synergies from this arrangement, the spreading of staff across different functions and interventions could have ef�ciency costs if a lack of specialization hinders the development of streamlined processes and specialized skills. There are also potential conflicts of interest from having the same staff designing, implementing and evaluating activities, which is not generally considered best practice. Assessing the ef�ciency of JSLU, JSPACA, and PKSA is more straightforward; such assessment indicates moderate and declining administrative overheads despite their small scale and pilot status. JSLU and JSPACA spending is unique because it is isolated within the LAKIPs of their implementing Directorates, allowing for a more straightforward examination of their administrative costs. Estimates indicate that administrative costs are moderate and have declined as the scale of the interventions has increased. For example, average annual administrative costs per bene�ciary were US$ 37 and US$ 31 for the disabled and elderly cash transfers respectively in 2010 (down from US$ 49 and US$ 67 in 2008), while the overall administrative overhead was 9 and 8 percent respectively (down from 12 and 36 While reports by community organizations and institutions putatively exist, these are not systematically submitted to Kemensos, collated or analyzed. Detailed local government budgets (with information down to the program level) are also not readily accessible. 25 JSLU, JSPACA, PKSA Cash and In-kind Transfers for at-risk youth, the disabled, and vulnerable elderly 16 percent).37 In comparison, an international survey of 16 cash and near cash programs found that the administrative costs of well-executed interventions cluster in range of 8 to 15 percent of total costs, with a mean of around 8 percent (Grosh et al 2008, Pg.391). Thus, the ef�ciency indicators for the cash transfers appear to be reasonable, especially considering the small scale and pilot status of the transfers. However, this analysis excludes local government contributions and thus may underestimate the full extent of administrative costs. According to Kemensos of�cials, local governments are expected to allocate funds from their budgets (APBD) to provide socialization and to top up the salaries of local social workers. Local governments which do so are rewarded with increased bene�ciary quotas when budgets allow. However, these cost-sharing arrangements are not explicitly speci�ed in any regulation, which means compliance is variable.38 Field visits to a limited set of districts and an examination of a small selection of district budgets obtained from Kemenkeu both indicate variation in district contributions, though it is dif�cult to identify subnational budget allocations for individual programs.39 However, because limited district budget data makes it dif�cult to quantify the extent of local government contributions, they are not included in the ef�ciency indicators reported above, which thus may underestimate the full extent of administrative costs (and thus overstate the intervention’s spending ef�ciency). More resources may be necessary for developing the management and implementation systems of the interventions in order to enhance their impact. A breakdown of current administrative costs indicates that the small non-bene�t budget is largely split between civil servant salaries, general administration and targeting. Very little is spent on socialization, monitoring and evaluation, follow up (such as complaints handling), or training – see Section 6 below for the on-the-ground impacts on service quality of these low allocations. In the near term, maximizing the impact of the program and improving its implementation may require more resources for administration. Budget fragmentation and duplication of program implementation functions could also be imposing ef�ciency costs. The overall size and budget of the Yanrehsos program is relatively small; that small pot of money is then divided between �ve major activities/subdirectorates, each of which has multiple interventions, each of which are executed by multiple actors at various levels of government. This is likely to produce signi�cant ef�ciency costs arising both from the small scale of the interventions resulting in high administration overheads and the challenges of managing, coordinating and ensuring accountability across different levels of government.40 There is also some duplication of implementation systems: for example, JSLU and JSPACA are implemented separately by their respective subdirectorates, and have parallel structures for fund flow, targeting, socializing and monitoring and evaluation as well as separate local-level social workers. Yanrehsos could reap ef�ciency savings by realizing greater economies of scale and consolidating common processes across subdirectorates. Ef�ciency gains could also be realized by consolidating smaller interventions and merging some implementation functions into a single implementation unit. 37 These �gures include estimates of local social worker honorarium, which is not recorded in the LAKIP. In 2009, 640 social workers were employed to assist in JSPACA implementation and 1,000 to assist in JSLU implementation. The primary function of the social workers is to help identify potential ben- e�ciaries and facilitate the delivery of payments. Social workers are paid by local governments using deconcentration funds transferred from the Central Government. 38 In 2010, Presidential Regulation No. 3/2010 on ‘Justice for All’ requested that all levels of local government (Province, District, City/Village) support a nominated group of social assistance programs (including most of the interventions within Yanrehsos) but did not explicitly clarify spending responsibili- ties. 39 During a JSPACA monitoring and evaluation mission in Maluku Province in 2010, Ambon district of�cials reported providing additional salary of Rp 350,000 per year to each social worker, while districts of�cials from Maluku Tengah and Pulau Buru reported no additional salaries. During an JSLU monitoring and evaluation mission in West Java, of�cials reported that seven districts were allocating resources to the elderly cash transfer program, although documents detailing these allocations were not available upon request. Some of�cials reported that allocations for these speci�c programs were not always uniquely identi�ed, but rather grouped under broader categories (for example, socialization activities for these interventions may simply be recorded under the goods and services expenditure). 40 It may, however, be an understandable product of (a) the desire to target a wide array of vulnerable groups, (b) Indonesia’s decentralized system of government and (c) the need for local social workers and government of�cials to support implementation. 26 Table 4: Severely disabled Abandoned Elderly Spending cash transfer** (JSPACA) cash transfer** (JSLU) Ef�ciency 2008 2009 2010 2008 2009 2010 Indicators, Unit cost 2008-2010 (Total spending/No. 4,091,930 4,050,100 4,004,475 4,268,472 4,123,563 3,913,598 bene�ciaries, Rp) Administrative costs per bene�ciary 491,930 450,100 366,098 668,472 523,563 313,598 (Non-bene�ts/No. bene�ciaries, Rp) in US$*** 49 45 37 67 52 31 Administrative overhead ratio 12% 11% 9% 16% 13% 8% (Non-bene�ts/Total spending) Cost of delivering bene�ts ratio 14% 13% 10% 19% 15% 9% (Non-bene�ts/ Bene�ts) Civil servants per 9 9 14 14 10,000 bene�ciaries Memo items: No. of bene�ciaries 10,000 17,000 17,000 5,000 10,000 10,000 Number of civil 16 16 14 14 servants Value of annual UCT 3,600,000 3,600,000 3,600,000 3,600,000 3,600,000 3,600,000 (Rp) Total spending (Rp 41 69 68 21 41 39 bn)* o/w Bene�ts 36 61 61 18 36 36 o/w Non-bene�ts 4.9 7.7 6.2 3.3 5.2 3.1 o/w Civil servant 0.6 0.7 0.5 0.6 0.6 0.4 salaries o/w Admin/other 1.6 3.3 2.7 0.5 0.2 0.1 o/w Targeting**** 1.1 1.8 1.5 1.4 2.9 1.7 o/w Follow-up 0.0 0.0 0.0 0.0 0.1 0.0 o/w Socialization 1.0 0.9 0.7 0.5 0.3 0.2 o/w Training 0.0 0.0 0.0 0.0 0.0 0.0 o/w Evaluation (M&E) 0.5 1.0 0.8 0.4 1.2 0.7 Sources and Notes: World Bank staff calculations based on Financial Accountability Reports (LAKIPs) and Directorate General data. *While original budget data allocates personnel expenses to the DG as a whole, expenses for each individual Activity/ Directorate have been estimated based on staff numbers, and salary expenses associate with the cash transfers are assumed to be 50% of total Directorate personnel costs (while only around 5 full-time staff work on each UCT year round, the majority of staff in each Directorate are involved in processing UCT applications between January and May of each year). **Includes some minor non-UCT activities. ***Rp/US$ rate assumed to be 10,000. ****Includes estimates of social worker honorarium under deconcentration budget. 27 JSLU, JSPACA, PKSA Cash and In-kind Transfers for at-risk youth, the disabled, and vulnerable elderly 6. Implementation Program support operations – socialization, facilitator training, monitoring and evaluation, and a complaints feedback system – are weakly implemented. Program guidelines in published manuals are comprehensive but not practical. For example, eligibility criteria are enumerated, but there is no guidance for identifying and prioritizing bene�ciaries when not all criteria are met or when two potential bene�ciaries equally meet all criteria; in practice, these judgements are left to local-level implementers and facilitators. For PKSA especially, there are many overlapping criteria between subprograms: abandoned children who are victims of violence could equally belong to the “PKS-Antar/Anjal� subcategory (street and neglected children) or the PKS- AMPK subcategory (children in need of special protection). For socialization and information dissemination, manuals state that socialization should be delivered to all levels of government and to bene�ciaries, but direction for the content, length, frequency, and quality monitoring of socialization activities are not in the guidelines – see below for more detail. PKSA relies heavily on partner institutions to identify bene�ciaries and deliver cash plus facilitated services, but the partner selection process is not open or competitive. PKSA often partners with implementing agencies called LKSA (Social Welfare Institution for Children). LKSA need to be appointed (but not certi�ed) by Kemensos before they can begin delivering PKSA funds. In order to be appointed, LKSA must submit a proposal to the PKSA managing unit (see Figure 13 below); the proposal contains supporting data and a list of potential PKSA bene�ciaries that are either residents or users of the LKSA or are known to LKSA staff. PKSA then reviews proposals, sometimes with an accompanying veri�cation at the LKSA facility. According to Kemensos staff, criteria that determine LKSA appointment include adequacy of facilities, resources, and networks to support PKSA bene�ciaries and extent of experience with children with family, community, and social dif�culties, but standard selection procedures are not detailed in PKSA manuals (or elsewhere). This LKSA recruitment process is not widely publicized; this is partly a self-imposed weakness, as Kemensos instructed LKSA to be able to begin implementing PKSA (including disbursing the savings) only one month after fund disbursement (from PT. POS to LKSA account), leading to an abbreviated LKSA search and vetting period. Many LKSA were invited to participate (based on the recommendations National Commission for Child Protection) and many others were already working in partnership with Kemensos at either the central or local levels.41 See below for the effect of LKSA partners on other program processes like bene�ciary selection. Figure 13: Implementation: Submission of LKSA Selection Issuance of SK PKSA LKSA proposal Kemensos PT. Pos Disbursement of funds PKSA saving account LKSA Bene�ciaries’ Signing contract/ PKSA savings Account MoU LKSA operational account Source: PUSKA PA UI �eld research report. 41 PUSKA PA UI (2011). 28 JSLU and JSPACA perform bene�ciary selection in-house; their processes are thorough but costly and time- consuming. After Kemensos receives local nominations, the subdirectorates complete an individual-by-individual review of all nominees. Both programs compile limited demographic and socio-economic information (for bene�ciaries and their households) while JSPACA records type and severity of an individual’s disability as well (see Table 3 above). A full body photograph is included to capture nominee characteristics not evident in the data collected from nominee materials.42 Once data is compiled, a scoring system is applied to rank nominees according to the severity of vulnerability.43 Often assessments based on photographs do not match assessments or scores from data alone. Many applications with a high score (low score) based on characteristics were ultimately rejected (accepted) when the accompanying photographs showed mild or no vulnerability (high vulnerability). Figure 14: Kemensos announced the Dinsos Province announced The selected Dinsos Implementation: plan to implement the JSPACA the plan to implement prepares the . JSPACA and program to the pilot provinces. JSPACA program to the pilot enumerators and districts. facilitators. JSLU Kemenos conducts The enumerators socialization & training, conduct the data invites relevant agents who collection together with are involved in the program Dinsos Districts implementation. Kemensos receives the The assessment result proposed list of bene�ciaries submitted to Dinsos Province from Dinsos. and Dinsos District. Kemensos conduct assessment of all incoming applications and random checking veri�cation. Kemensos make a �nal assessment based on the random checking veri�cation. Kemensos issued a ministerial Direct funds disbursement to the decree on the list of bene�ciaries follows the MoU bene�ciaries. with Pos andKemenkeu. Source: World Bank staff based on interview with Kemensos staff. The bene�ciary selection process consumes large amounts of staff time: approximately four months to complete �nal bene�ciary lists. JSLU found nominee assessment and �nal bene�ciary selection too burdensome and has delegated this function to regional of�ces. JSLU has also dropped the scoring protocols, so there are effectively no national guidelines for districts to use for prioritizing nominees.44 See Section 3 above for more detail on the outcomes of targeting and bene�ciary selection procedures in PKSA. PKSA has delegated bene�ciary selection to implementation partners; these partners have wide discretion. LKSA are asked to nominate potential bene�ciaries as part of their Kemensos partnership application, but LKSA selection and nomination practices vary widely. In some areas, LKSA do bene�ciary assessment outside their current client lists, but 42 Due to applications that seemed dubious, Yanrehsos did a random review of nominee eligibility Many candidates were deemed ineligible upon review; full body photographs were then added (in 2007) as an application requirement. 43 The scoring system consists of two high-level criteria: type and condition of disability, including assistive devices necessary for performing daily activities; and level of dependency. Economic status is then the last �lter applied: those with major disabilities from poor households are prioritized. 44 Prioritization and selection is left to the determination, by any method they choose, of the regional- and local-level of�ces, program administrators, and partner organizations. 29 JSLU, JSPACA, PKSA Cash and In-kind Transfers for at-risk youth, the disabled, and vulnerable elderly these assessments are typically done under great time pressure (which is itself a result of Kemensos’ desire to distribute funds quickly). In most areas, however, LKSA nominate their own clients because of insuf�cient time and money for additional data collection and assessment.45 As a result, over 75 percent of current PKSA bene�ciaries were LKSA clients; only 15 percent applied themselves, were nominated by their parents, or were nominated by other participants (see Figure 15). See Section 3 above for more detail on the outcomes of targeting and bene�ciary selection procedures in PKSA. Figure 15: PKSA 4% bene�ciaries 5% assessment 8% process 2% Assessment done by respective institution (LKSA) 5% Parents come to apply for assistance School/anyone other than parents apply for assistance Children who have been assisted by an institution Other Do not know 76% Source: PUSKA PA UI �eld research data. Activities and budgets for socialization of JSLU, JSPACA, and PKSA to implementers are extremely limited. When the programs were �rst piloted, socialization took place at the province level for JSLU and district level for JSPACA with village, subdistrict, and district of�cials, facilitators, regional Kemensos staff, and PT Pos staff as the main audience. Kemensos delivered all information regarding program goals and processes in one-day events.46 To date there have not been any additional socialization events for these groups, nor monitoring of the effectiveness of the previous activities, nor attempts to measure staff knowledge. Socialization budgets for new program areas and new program staff are small. For example, JSPACA socialization for 7 of 18 new provinces (in 2009) took place during a yearly staff meeting where program guidelines were distributed together with data collection instruments. Written guidelines were expected to suf�ce for �rst-time implementers. However, program manuals do not have enough operational detail, which leaves regional implementers confused and with much discretion, which itself leads eventually to high rates of potential nominee rejection by Kemensos (see above) and ineffective socialization and facilitation to bene�ciaries (see below). Socialization activities for bene�ciaries and communities occur unevenly among districts; effectiveness is limited. Target groups are by de�nition vulnerable and hard to reach47 and substantial effort must be made to draw these populations into programs and other social services to which they are otherwise not exposed. In practice, the programs rely heavily on local government or partner institutions for hosting and funding socialization activities; these partnerships do not typically include extra funds or support staff from Kemensos. For example, the local government in Maluku spent its own funds on a media campaign supporting JSPACA socialization. However, program manuals do not contain enough detailed direction for the content, length, frequency, and monitoring of socialization activities, leaving the quality and effectiveness of socialization highly variable and leading to too little knowledge among bene�ciaries and communities. For example, most bene�ciaries recall their �rst exposure to a program being a visit by a program facilitator. Only a very few recall having received socialization materials from Kemensos of�cials.48 45 PUSKA PA UI (2011). 46 Socialization to a broad group of stakeholders is usually followed by training for enumerators and facilitators to ensure accurate understanding of eligi- bility criteria and their responsibilities to bene�ciaries. See this section for more detail on facilitator training. 47 As well as being physically remote or unable to travel to regular socialization points or public spaces, they are often not registered and sometimes kept out of sight by their families who fear social shame and stigmatization. 48 Interview results with 174 PKSA bene�ciaries in 4 provinces showed that nearly 75 percent recalled that their �rst source of information was an LKSA. Only 8 percent claimed that they received socialization from Yanrehsos of�cials. 30 Crucial information on program objectives and co-responsibilities is not fully delivered. For example, PKSA bene�ciary interviews indicated that bene�ciary rights and responsibilities, LKSA rights and responsibilities, penalties for noncompliance, program tenure, and where to submit complaints are not well understood (Figure 16). While program objectives seem to be well-communicated, fewer bene�ciaries knew how much PKSA would provide nor when and where to go to receive the PKSA cash transfer. Shortcomings may constrain the effectiveness of other good cash transfer design principles. For example, JSLU and JSPACA provide membership cards, which record the date and amount of each payment made, that are to be signed by PT Pos and countersigned by bene�ciaries each time payment is made. The card is meant to reduce the possibility of corruption but will be less effective when bene�ciaries are not aware of rights, responsibilities, and technical details. Figure 16: Complaint mechanism Bene�ciary Aid time period perception of Penalty of the program PKSA features Program evaluation Rights and responsibilities of LKSA Rights and responsibilities of beneficiaries Disbursement location Amount of assistance Disbursement time Purpose of PKSA 0% 20% 40% 60% 80% 100% Percent with information regarding feature Percent unaware of feature Percent without information regarding feature Source: PUSKAPA UI �eld research data JSLU and JSPACA bene�ciaries have not been receiving funds on time. PKSA reports more timely disbursement, perhaps because of unintentional incentives. In JSLU and JSPACA, PT Pos is responsible for delivering disbursements directly to bene�ciaries by no later than the 15th of every calendar month49, but insuf�cient staff at post of�ces and dif�cult geographic locations slow down delivery. Consequently, disbursements often arrive as large sums that have accumulated over several months (see also Section 7 below on public �nancial management and Section 4 above on fund usage). Figure 17: 2% 2% 3% JSPACA funds delivery 6% mechanism POS officer delivered the funds to beneficiaries address Caretaker drawn the funds at the post office Regional office staff delivered the funds to the caretaker address 24% Regional office staff together with POS officer delivered the funds Withdrawn at the village office 63% Through regional office & facilitator Source: Internal M&E JSPACA-Kemensos 2009 49 If the bene�ciary or guardian is twice not present at his/her address during attempted delivery, then bene�ts are held at and can be withdrawn from the nearest designated post of�ce. Monitoring activities in 2009 showed that 24 percent of bene�ciaries went to the post of�ce to receive disbursement while 65 percent stated that post of�cers delivered funds to their address. 31 JSLU, JSPACA, PKSA Cash and In-kind Transfers for at-risk youth, the disabled, and vulnerable elderly PKSA, which uses PT Pos to transfer funds directly to bank accounts established by LKSA on behalf of bene�ciaries50, reports more timely fund disbursement. This is likely because some LKSA rely almost exclusively on PKSA transfers to run regular activities for all clients under their care and therefore are motivated to advocate for better disbursement outcomes and apply pressure on PT Pos. There are no such incentives in the JSLU and JSPACA programs for applying pressure when funds are delivered late. Facilitation is crucial for program success…. Facilitators are responsible for weekly visits to clients to advocate for and encourage healthy living practices and social integration, assist with cash transfer spending, and provide updates on disbursement schedules and co-responsibilities (see Figure 18 below for most commonly-discussed topics). Facilitators are the primary interface between a bene�ciary and the program, so any obstacles, shortcomings, or malfeasance that occurs is usually reported to facilitators �rst. Facilitators are also one of Kemensos’s primary sources of monitoring and evaluation information and contribute reports for such activities. They also assist enumeration and data collection teams in locating potential bene�ciaries. …but few resources are devoted to quality upgrading in facilitation or facilitators. JSLU and JSPACA facilitators earn approximately Rp 167,000 per month, which is below the average wage for domestic help or childcare (for example) in urban areas in Indonesia. PKSA facilitators employed by Kemensos can earn as much as Rp 1.4 million per month, which higher salaries are meant to attract facilitators with higher education levels.51 Facilitators must be mobile and cover large areas, but neither the transportation allowance (on average Rp 300,000 per month - budgeted for 5 months) nor overall salary varies by ground covered or with the number and distance to bene�ciaries. Internal JSPACA monitoring (in 2009) showed that of 151 facilitators operating in 84 districts, 11 percent served more than 40 clients while the desired bene�ciary-to-facilitator ratio is 10 to 15.52 The effectiveness of facilitator training content and frequency is unknown. As designed, facilitator training occurs as a vertical cascade, with senior facilitators delivering training material and any program updates to cadres in their regions. Senior facilitator training was completed during the inaugural program year and JSPACA and JSLU have each conducted one regionally-staggered training refreshment per year. These refreshers are intended primarily for newly- hired facilitators but any facilitator may come to receive updates. They are staggered regionally so newly-hired facilitators receive training only if it is their region’s turn. PKSA facilitators must often coordinate activities with staff from partner LKSA, but these same LKSA staff have noted that PKSA facilitators are often not well-prepared and thus ineffective. LKSA staff also observed that PKSA facilitators were in danger of becoming gatekeepers of cash transfers (delivered through PKSA) rather than providing more comprehensive support and advocacy.53 Figure 18: Topic Complaints on PKSA discussed during PKSA facilitator PKSA funds utilization plan Percent with discussion visits regarding topic Family health condition Children or parents' job Percent without discussion Children daily life regarding topic School-related problems 0% 20% 40% 60% 80% 100% Source: PUSKA-PA �eld research data 50 In addition to establishing the account, LKSA retains access to that account. 51 The minimum level of education for a PKSA facilitator employed by Kemensos is a college degree in social welfare science; JSLU and JSPACA facilitators are required to have graduated from senior high school. 52 Internal M&E of JSPACA program, (2010) 53 However, there is no clear division of authority and mandate between LKSA staff and PKSA facilitators which has lead to mistrust, jealousy, and frustra- tion (for both parties) in some cases, so LKSA evaluations of PKSA facilitators should be taken with a grain of salt. See PUSKA PA UI (2011) for further detail. 32 Monitoring and evaluation activities are provided for program manuals…. Program manuals indicate that activities should include questionnaires, focus group discussions, interviews, and direct observation of bene�ciaries, bene�ciary families, PT Pos staff, facilitators, and local government of�cials. In addition, regional of�ces are responsible for delivering quarterly implementation reports to Kemensos while JSPACA requests a monthly funds delivery report from PT Pos. Facilitators also contribute through monthly reports covering fund disbursement, bene�ciary data updates, and complaints, to regional Kemensos of�ces. Monitoring results should be disseminated at yearly staff meetings (with staff representatives from every province). Feedback for improving program implementation is expected to be generated through discussion of results and regional success stories and dif�culties.54 …but regular monitoring and evaluation is not successfully carried out everywhere. In 2010, Kemensos at the central level, in coordination with regional of�ces, completed JSPACA and JSLU monitoring activities in 20 and 28 provinces (respectively).55 However, the regular regional-to-central reporting process is not monitored nor are there any penalties for noncompliance. Facilitator reports have become a condition for salary payment, which has encouraged hasty reporting and “ticking boxes�. An internal JSPACA monitoring report in 2009 showed that 12 percent of PT Pos staff thought their reports were the responsibility of the Kemensos regional of�ce; approximately 9 percent of post of�ces did not submit the report. The monitoring instruments used and information reported are inadequate. Each directorate has a standard form to be used by facilitators and local of�ces. However, the form is general and does not capture critical issues or explore the root causes of implementation dif�culties. For example, information like bene�ciary socioeconomic characteristics or bene�ciary program knowledge of selection process, disbursement timing and amounts, the complaint mechanism, or bene�ciary perceptions are not fully covered. A functioning complaints and grievance system is mostly not provided. Technically, complaints and suggestions (from any community members) for the three programs can be communicated to Kemensos or its regional of�ces directly. Communities are largely unaware of this, however, and bene�ciaries usually lodge complaints through facilitators. Facilitators are not required to, and have no incentives to, pursue the complaint with superiors or program administrators. When complaints are �led, they do not enter any flow of information from the ground up to the central level; nor is there usually any remedy directed back down to the local level.56 54 Dissemination, evaluation, and feedback procedures are not explicitly mentioned in program manuals. 55 PKSA has not yet completed any monitoring activities.but plans to do so in 2011. 56 JSLU is the only one of the three cash transfers that clearly elaborates a complaint handling system – including a dedicated telephone, fax, and email connection to Kemensos and PT Pos – in the program guidelines. In addition, JSLU (and JSPACA) membership stickers placed on each bene�ciary’s dwell- ing are meant to encourage community reporting of mistargeting. 33 JSLU, JSPACA, PKSA Cash and In-kind Transfers for at-risk youth, the disabled, and vulnerable elderly 7. Public Financial Management and Sustainability Public �nancial management – covering issues such as budget formulation and reporting, indexation of bene�ts, and budget execution and disbursement – could be strengthened. Budget formulation and reporting is relatively comprehensive…. The overall budget ceiling for the Yanrehsos program as a whole is determined annually through consultations between the Directorate General, Bappenas and Kemenkeu, and then allocated by the DG to each of the activities and interventions, subject to approval by the Minister. Annual �nancial accountability reports (LAKIPs) are produced for each of the major activities and provide detailed documentation on centrally-executed spending (down to the sub-activity level). Budget planning is also informed by Indonesia’s Medium Term Development Plan (RPJM) 2010-14, which outlines indicative bene�ciary number targets and budgets for the Yanrehsos budget and for each of the three major activities (Children, Disabled and Elderly) for each year out to 2014. …but the usefulness of the reports is limited by incomplete coverage and lack of detail. For example, the above-mentioned documents do not include spending on civil servant salaries (which are grouped together under the DG’s Secretariat budget) nor spending on local social workers or community-based organizations and social institutions (for which separate reports putatively exist but which are not easily accessible). A single audit document which provides a comprehensive overview of all spending on each activity including salaries and spending by at the local level would allow for greater tracking and analysis of performance. Furthermore, the LAKIPs merely list expenditure items and do not present any analytical categorization of spending nor information on outcomes. The streamlining of expenditure reporting by categories such as: (1) function (e.g. policy or implementation); (2) sub-function (e.g. targeting, M&E, socialization); (3) different interventions; or (4) bene�t versus non-bene�t costs, would allow for faster and easier analysis of program expenditure which could better support future improvements. Spending on JSPACA and JSLU is unique in being isolated within their Directorate’s LAKIPs, allowing for easier analysis, and may serve as an example to replicate. Many, perhaps all, of these issues are not unique to Yanrehsos, but reflect broader public �nancial management practices in Indonesia. Budget execution rates for the Yanrehsos program are high…. The budget execution ratio for the Yanrehsos program, relative to the �nal revised budget, declined to below 90 percent in 2007 and 2008, largely reflecting weak disbursement of community-executed spending which was being wound down at the time (Figure 19 and Annex Table 8). Yet in 2010, the execution ratio remains under 90 percent of the budget, reflecting poor implementation, particularly on non bene�t disbursement. …although disbursement is sluggish early in the year due to implementation delays. Disbursement tends to be slow early in the �scal year and lumpy (Figure 20). In 2008 and 2009, for example, only 6 percent of the budget was disbursed in the �rst quarter, while 2010 shows even worse execution ratio at 3.5 percent. A major cause of delay is JSLU, JSPACA, and PKSA implementation, for which disbursement of funds from Yanrehsos to PT Pos typically only occurs in June of each year after the bene�ciary list is �nalized and a Memorandum of Understanding (MoU) is signed between the two agencies (typically in April or May of each year). These delays are caused by the need for districts to re-con�rm the eligibility of existing bene�ciaries each year and propose new bene�ciaries to Kemensos, which holds �nal authority to approve the bene�ciary list. The Elderly Directorate initiates this process in October or November of the preceding year to try to avoid delays, but then must wait for the MoU anyway, which covers all interventions within Yanrehsos. As noted above, this Directorate is planning to decentralize its cash transfer intervention in 2012, which could help speed up disbursement since bene�ciary lists will not need to be sent to the Center for veri�cation. The delay in disbursement of funds to PT Pos means that recipients often receive their �rst transfer in the second half of the year in a lump sum which covers the �rst six months of the year. Thereafter, the stipend is either paid monthly or, more often, in tranches every two to three months (especially in remoter districts). 34 Figure 19: Budget execution ratios by Figure 20: Budget execution by month executing agents (Monthly and cumulative actual expenditure, (Realized spending as a share of �nal revised percentage of total annual actual expenditure, budget allocation, percent, 2006-2010) 2008 through 2010) Percent 100 Percent 100 2008 monthly 2009 monthly 80 2010 monthly 92 2008 cumulative 2009 cumulative 60 2010 cumulative 84 40 76 20 68 2006 2007 2008 2009 2010 0 Center Social institutions Jan Feb Mar Apr MayJ un Jul AugSep Oct Nov Dec Community Total Program Sources and notes: World Bank staff calculations based on Directorate General data Current expansion plans are minor from a Cluster 1 perspective and have negligible consequences for the central government’s overall budget. The RPJM for 2010 to 2014 outlines modest expansion plans for each of JSLU, JSPACA, and PKSA (as well as other initiatives within the Yanrehsos portfolio). The target number of bene�ciaries is planned to rise by around 80 percent to almost 400,000 by 2014, but coverage of the potential target population (based on household survey estimates) remains below 20 percent. Given the small scale of the programs, the budget consequences of this expansion are negligible: the overall budget doubles to around Rp 1.1 trillion by 2014, roughly equivalent to what is spent on PKH today, and remains a tiny share of total central government spending at 0.1 percent. A big push to expand coverage to all potential bene�ciaries would have larger, although still relatively modest, budget implications. A massive scale up of the program starting in 2012 with the aim of reaching 100 percent coverage (over 2,000,000 bene�ciaries) by 2014 could result in the annual budget rising to around Rp 7,000 billion by 2014, equivalent to around 0.7 percent of central government spending. 35 JSLU, JSPACA, PKSA Cash and In-kind Transfers for at-risk youth, the disabled, and vulnerable elderly Table 5: 2009 2010 2011 2012 2013 2014 Financial Baseline (RPJM) sustainability of programs Total budget allocation 595 538 504 646 899 1,122 for vulnerable (Nominal, Rp billion) groups out to Share of central government 0.1 0.1 0.1 0.1 0.1 0.1 2014 spending (%) Target number of bene�ciaries 228,466 228,466 216,650 260,650 336,480 390,060 - Children 169,306 169,306 157,800 185,600 234,100 263,000 - Disabled 34,195 34,195 35,750 45,000 60,180 75,640 - Elderly 24,965 24,965 23,100 30,050 42,200 51,420 Share of potential target 8 9 9 11 14 17 population covered (%) Big Push Total budget allocation 1,658 5,133 7,043 (Nominal, Rp billion) Share of central government 0.2 0.5 0.6 spending (%) Target number of bene�ciaries 603,464 1,745,643 2,238,551 - Children 352,047 1,018,368 1,305,920 - Disabled 94,041 272,032 348,844 - Elderly 157,376 455,242 583,787 Share of potential target 25 75 100 population covered (%) Figure 21: 8,000 Rp billion (nominal) Number 2,400,000 Financial sustainability 6,000 1,800,000 of programs for vulnerable 4,000 1,200,000 groups out to 2014 2,000 600,000 0 0 2009 2010 2011 2012 2013 2014 Baseline budget (LHS) Big push budget (LHS) Baseline beneficiaries (RHS) Big push beneficiaries (RHS) Sources and Notes: RPJM 2010-14 and World Bank staff estimates and projections. 36 8. Summary and Recommendations Indonesia is committed to protecting welfare for all citizens and the Yanrehsos facilitated cash transfer programs have the potential to help by delivering income and facilitated services to vulnerable and at-risk populations of elderly, youth, and the disabled. These populations face dif�culties bene�tting from universal services provided by local and national governments and privately- or socially-provided services and care. Additionally, individual members of these populations are often identi�ed by a lack of support and care coming from family, kin, and community- based networks. At the same time, and especially for the elderly and disabled populations, daily costs of living are much higher while they may be completely or partially unable to generate income. Welfare protection for these groups must be specialized; the Yanrehsos facilitated cash transfers – JSLU, JSPACA, and PKSA – recognize the increased complexity of protecting welfare in targeted subpopulations and provide generous cash bene�ts for daily costs of living as well as facilitated services that aim to increase participation in social services and community life. However, the Yanrehsos programs are not ideal. Four main areas limit effectiveness: (1) bene�t package design… The cash transfers portions of the JSLU, JSPACA, and PKSA programs are relatively generous by Indonesia standards, but the facilitated services may be equally valuable for bene�ciaries in terms of overall health, mental and social well-being, and inclusion. Though facilitators are instructed to provide bene�ciaries help, advice, and encouragement to access the entire array of services that are locally available (health services, legal services, advocacy services, community and social services), a standard package is not detailed, and there is no guarantee that motivated and enthusiastic facilitators will have the training or experience necessary (see below) to identify and provide remedies for the most relevant physical, mental, and social dif�culties that bene�ciaries face. (2) intervention design and capacity constraints… Relatedly, the cash transfers as designed have no triage system to deliver the worst-off bene�ciaries to agencies and care institutions that can provide specialized assistance when needed. Figure 22 provides a summary of a triage system that distinguishes between “at risk� children and children who are “at risk� and in need of immediate or rapid response; the design can equally apply to “at risk� elderly and in some cases, the disabled. In this scenario, the children in need of immediate response are �rst referred to the institutes or care providers for front-line emergency care or immediate intervention. Only after those services have begun would those children be considered eligible for (and referred back to) the cash transfers and other facilitated services available in PKSA (for example). In PKSA currently (and in both JSLU and JSPACA), any bene�ciaries identi�ed are given the same basic bene�t package regardless of special circumstances.57 This is less effective for those who would bene�t from a front-line emergency response. It also leads to increased burdens on Yanrehsos staff and local af�liates, who are not equipped as front-line emergency providers. 57 PKSA has developed protocols for a “case conference� system, which primarily means a consultation between PKSA facilitators, bene�ciary families (if available) as well as other social service providers and representatives of the legal and security services (if necessary). Such case conferences are not yet guaranteed for every PKSA bene�ciary (See PUSKA PA UI, 2011) and in most cases depend on the willingness and capability of the facilitators from the PKSA program and/or those from the PKSA-implementing institutions (the LKSA). 37 JSLU, JSPACA, PKSA Cash and In-kind Transfers for at-risk youth, the disabled, and vulnerable elderly Figure 22: Children in need Triage protocol At Risk Children of immediate for two types Repsonse of vulnerable children SOCIAL PROTECTION FOR CHILDREN Child Welfare and Protection Assistance Plan Child Saving Cash Assistance Immediate Needs Account Registration, Health, Access to Basic Medical, Psychosocial Education Services & Legal Aid Counseling, Counseling, Mentoring, Helping Family Services Mentoring, Helping relationship relationship Specialized Services (if Registration, Health, Protective Services Education and when needed) MONITORING AND Child Saving EVALUATION Account Source: PUSKA PA UI, 2011 (3) implementation weaknesses…. Socialization, targeting and prioritization, facilitator capacity and services delivered, monitoring and evaluation, and a complaints and grievances system are not effectively and consistently provided. The large amounts of time and staff resources that JSLU and JSPACA in particular spend verifying eligible bene�ciaries could be reduced if all programs rely on the upcoming registry of poor households (PPLS11) for identifying their potential bene�ciary pools and for determining quotas to regions. The Yanrehsos programs should be able to rely on PPLS11 and the uni�ed targeting database to extract list of poor households that additionally meet some or all of the eligibility criteria for their program. JSLU and JSPACA should follow the PKSA lead in setting higher salaries for facilitators and beginning a recruitment process that will result in a facilitator corps with upgraded skills and capabilities while The facilitator corps should be given every opportunity to upgrade the services that they themselves provide, for example with cross-region forums for facilitators where experts and others can give advice and answer practical concerns. The Yanrehsos cash transfers should economize on already scarce resources by pooling their targeting, socialization, and monitoring and evaluation activities by exploring the option of introducing its facilitator corps to the PKH MIS systems and monitoring and evaluation apparatus. and (4) extremely narrow coverage. Cumulatively these programs cover less than one-tenth of one percent of Indonesia’s population. The targeted populations are dif�cult to reliably and consistently estimate, but estimates of overall need used by Yanrehsos appear to be an order of magnitude lower than what would be expected from applying international incidence rates to the Indonesian population. While program design and implementation needs signi�cant reform, current low coverage levels mean that even the most effective program will not reach enough of Indonesia to affect nationwide rates of well-being or social function in these populations. 38 While Yanrehsos may not be taking full advantage of the opportunity to intervene positively, achieving a full slate of feasible reforms could make the programs effective and worthwhile pro-poor expenditures. The attempt to provide bene�ts that are in many ways the most dif�cult to reach, with the highest costs of daily living, and most likely the highest rates of social exclusion should be commended. The current narrow coverage of the pilot initiatives has a hidden bene�t: reform can be achieved without major disruptions in services. With concerted attention to program design and implementation, Yanrehsos will be ready to bring a tested program to all of Indonesia. 39 JSLU, JSPACA, PKSA Cash and In-kind Transfers for at-risk youth, the disabled, and vulnerable elderly References Badan Pusat Statistik, Kementerian Sosial Republik Indonesia. 2006. Analisa deskriptif Penyandang Masalah Kesejahteraan Sosial. Jakarta. Chowdhury, J. 2005 Disability and Chronic Poverty: An Empirical Study on Bangladesh. MPhil Thesis, Oxford University. Grosh, M., del Ninno, C., Tesliuc, E., & Ouerghi, A. 2008. For Protection and Promotion: The Design and Implementation of Effective Safety Nets. Washington, DC: World Bank. Loeb, Mitchell E., A. H. Eide, D. Mont. 2008.“Approaching the measurement of disability prevalence: The case of Zambia�, Alter: European Journal of Disability Research. Mitra, Posarac, and Vick. 2011. “Disability and Poverty in Developing Countries: A Snapshot from the World Health Survey� Social Protection Discussion paper No. 1109. Washington, DC: World Bank. National Institute for the Mentally Handicapped NIMH. 2004. Disability Management in India: Challenges and Commitments. New Delhi. Pusat Kajian Perlindungan Anak. Universitas Indonesia. 2011. “Building Social Protection System for Children in Indonesia; An Assessment on the Implementation of the Ministry of Social Affairs’ Social Assistance Program PKSA and Its contribution to the Child Protection System�. Jakarta: Pusat Kajian Perlindungan Anak. Universitas Indonesia. Save the Children, Unicef, Kemensos. 2007. Someone that matters: The Quality of Care in Childcare Institutions in Indonesia. Jakarta, Indonesia. SMERU. 2006. “A Rapid Appraisal of the Implementation of the 2005 Direct Cash Transfer Program in Indonesia: A Case Study in Five Kabupaten/Kota.� Research Report. Jakarta, Indonesia: The Smeru Institute. SMERU. 2009. “A Rapid Appraisal of the Implementation of the 2008 Direct Cash Transfer Program and Bene�ciary Assessment of the 2005 Direct Cash Transfer Program in Indonesia.� Research Report. Jakarta, Indonesia: The Smeru Institute. Son, H., & Sparrow, R. 2009. “Impact Evaluation of Social Health Insurance for the Poor in Indonesia: A Panel Data Analysis.� Asian Development Bank Working Paper. Manila: Asian Development Bank. Sparrow, R., Suryhadi, A., & Widyanti, W. 2008. “Public Health Insurance for the Poor: Targeting and Impact of Indonesia’s Askeskin Program.� SMERU Working Paper. Jakarta, Indonesia: The SMERU Institute. World Bank. 2006. Making the New Indonesia Work for the Poor. Jakarta, Indonesia: World Bank. World Bank. 2010. Indonesia Jobs Report: Towards Better Jobs and Security for All. Jakarta, Indonesia: World Bank. World Bank. 2012a. Targeting Poor and Vulnerable Households in Indonesia. Jakarta, Indonesia: World Bank. World Bank. 2012b. Protecting Poor and Vulnerable Households in Indonesia. Jakarta, Indonesia: World Bank. World Bank. 2012c. “Social Assistance Program and Public Expenditure Review 1: Public Expenditure Review Summary.� Jakarta, Indonesia: World Bank. World Bank. 2012d. “Social Assistance Program and Public Expenditure Review 2: BLT.� Jakarta, Indonesia: World Bank. World Bank. 2012e. “Social Assistance Program and Public Expenditure Review 3: Raskin.� Jakarta, Indonesia: World Bank. World Bank. 2012f. “Social Assistance Program and Public Expenditure Review 4: Jamkesmas.� Jakarta, Indonesia: World Bank. World Bank. 2012g. “Social Assistance Program and Public Expenditure Review 5: BSM.� Jakarta, Indonesia: World Bank World Bank. 2012h. “Social Assistance Program and Public Expenditure Review 6: PKH.� Jakarta, Indonesia: World Bank. World Bank. 2012i. “Social Assistance Program and Public Expenditure Review 7: JSLU, JSPACA, and PKSA.� Jakarta, Indonesia: World Bank. World Bank. 2012j. “Social Assistance Program and Public Expenditure Review 8: History of Social Assistance in Indonesia.� Jakarta, Indonesia: World Bank. 40 Annexes Table 6: Share in Spending on 2006 2007 2008 2009 2010 2010 (%) the Yanrehsos Child Social Services 300 239 337 296 235 100 Program by activity* & Centrally executed 52 58 194 203 177 75 Implementing Institutions & rehab. centers (Pantis) 28 44 46 44 41 17 Unit Community based (Deconcentration (Realized 220 138 97 50 17 7 funds) spending in Rp Disabled Social Services 186 194 206 217 194 100 billion, constant 2009 prices, 2006- Centrally executed 51 61 81 96 86 45 2010) Institutions & rehab. centers (Pantis) 90 96 106 106 96 49 Community based (Deconcentration 45 36 19 15 12 6 funds) Elderly Social Services 76 73 75 82 69 100 Centrally executed 27 33 41 60 54 77 Institutions & rehab. centers (Pantis) 5 7 8 8 7 10 Community based (Deconcentration 44 33 26 14 8 12 funds) Total Program/Directorate General 800 763 693 669 551 100 Centrally executed 306 349 341 388 338 61 Institutions & rehab. centers (Pantis) 139 174 189 184 167 30 Community based (Deconcentration 355 240 163 97 47 8 funds) Sources and Notes: World Bank staff calculations based on Directorate General and BPS data. Nominal spending is deflated using the GDP deflator. *While original budget data allocates personnel expenses, of�ce maintenance and the DG’s Secretariat as an overhead cost for the DG as a whole, expenses for each individual Activity/Directorate have been estimated based on staff numbers and budget allocations. 41 JSLU, JSPACA, PKSA Cash and In-kind Transfers for at-risk youth, the disabled, and vulnerable elderly Table 7: Child Services Disabled Services Elderly Services Indicators 2009 2010 2009 2010 2009 2010 of spending ef�ciency by Unit cost major activity/ (Total spending/No. 1,198,643 1,459,786 2,815,170 2,518,987 2,410,424 2,243,767 Directorate, bene�ciaries, Rp) 2009 and 2010 Administrative costs per bene�ciary 199,882 528,959 522,169 729,252 576,390 801,749 (Non-bene�ts/No. bene�ciaries, Rp) in US$ 20 53 52 73 58 80 Administrative overhead ratio 17% 36% 19% 29% 24% 36% (Non-bene�ts/Total spending) Cost of delivering bene�ts ratio 20% 57% 23% 41% 31% 56% (Non-bene�ts/Bene�ts) Civil servants per 4 4 9 9 11 11 10,000 bene�ciaries Memo items: No. of bene�ciaries 169,306 169,306 34,195 34,195 24,965 24,965 Number of civil servants 69 69 31 31 28 28 Total spending (Rp bn)* 203 247 96 86 60 56 o/w Bene�ts 169 158 78 61 46 36 o/w Non-bene�ts 34 90 18 25 14 20 o/w Civil servant 2.9 2.9 1.3 1.3 1.2 1.2 salaries Sources and Notes: World Bank staff calculations based on Financial Accountability Reports (Laporan Akuntabilitas Keuangan Pemerintah or LAKIPs) and Directorate General data. *While original budget data allocates personnel expenses, of�ce maintenance and the DG’s Secretariat as an overhead cost for the DG as a whole, expenses for each individual Activity/Directorate have been estimated based on staff numbers and budget allocations. 42 Table 8: Budget 2006 2007 2008 2009 2010 execution ratios Child Social Services 92 78 87 97 89 by Activity (Realized Centrally executed 79 75 92 98 97 spending as a Institutions & rehab. centers (Pantis) 65 86 90 94 66 share of �nal Community based (Deconcentration funds) 99 77 78 97 94 revised budget allocation, Disabled Social Services 90 94 91 95 82 percent, 2006- Centrally executed 95 95 95 98 97 2010) Institutions & rehab. centers (Pantis) 86 97 91 93 71 Community based (Deconcentration funds) 95 85 79 93 98 Elderly Social Services 90 86 85 98 91 Centrally executed 88 90 88 98 95 Institutions & rehab. centers (Pantis) 65 79 92 95 68 Community based (Deconcentration funds) 96 86 80 97 95 Total Program/Directorate General 92 89 86 96 86 Centrally executed 96 96 96 96 94 Institutions & rehab. centers (Pantis) 76 100 91 93 87 Community based (Deconcentration funds) 97 79 79 93 96 Sources and Notes: World Bank staff calculations based on Directorate General data. 43 JSLU, JSPACA, PKSA Cash and In-kind Transfers for at-risk youth, the disabled, and vulnerable elderly 44 Indonesia Social Assistance Program and Public Expenditure Reviews are policy notes summarizing current issues in the Government of Indonesia’s major household-targeted social assistance programs. Design, implementation, coverage, impacts, budgetary demands, cost effectiveness, stakeholder perceptions, and program history will be examined through qualitative and quantitative analysis for each program: conditional and unconditional cash transfers, subsidized rice distribution, a health fee waiver, cash scholarships, and cash transfers for the neglected elderly, disabled, and at-risk youth. Evidence-based appraisal can assist policymakers in evaluating programs and deciding whether and how they are achieving Indonesia’s social protection goals. Reviews 1 through 8 together comprise a companion volume to the “Protecting Poor and Vulnerable Households in Indonesia� report. “Protecting Poor and Vulnerable Households in Indonesia� provides a public expenditure review of the wider social assistance sector, summarizes results from Reviews 1 through 8, explores gaps in both population coverage and risk coverage of current social assistance programming, and offers recommendations and suggested reforms for achieving an integrated and coordinated household-based social assistance system. THE WORLD BANK, JAKARTA OFFICE Indonesia Stock Exchange Building Tower II/12th Floor Jl. Jend. Sudirman Kav. 52-53 Jakarta 12910 Tel: (6221) 5299-3000 Fax: (6221) 5299-3111 Website: www.worldbank.org/id