Adaptive Social Protection Human Capital & Climate Change Identifying Policy Priorities Asha Williams Gracia Hadiwidjaja for Indonesia Rabia Ali Imam Setiawan NOV. 2023 This work is a product of the staff of The World Bank. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Rights & Permissions © 2023 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW Washington DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org Some rights reserved The material in this work is subject to copyright. Because The World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. All queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e-mail: pubrights@ worldbank.org. Please cite the work as follows: “Williams, AT T RI B U T I ON Asha; Hadiwidjaja, Gracia; Ali, Rabia; and Setiawan, Imam. 2023. “Adaptive Social Protection, Human Capital, and Climate Change: Identifying policy priorities for Indonesia.” © World Bank.” P. I Contents 01 09 15 Executive Summary SECTION 1 Introduction: The Interconnectedness of Climate Change, Human Capital, and Poverty SECTION 2 The Importance of Social Protection in Addressing Climate Risk, Contributing to Human Capital Outcomes, & Supporting Indonesia’s National Development Priorities 24 32 60 SECTION 3 The Development of Indonesia’s Social Protection System SECTION 4 Stress Testing Indonesia’s Social Protection System SECTION 5 Conclusions & Recommended Priorities for Government Action P. I I Figures FIGURE 1.1 POVERTY EXPOSURE TO FLOODS5 12 FIGURE 2.1 CHANGE IN FOREST COVER LOSS IN PKH PARTICIPATING VILLAGES 20 FIGURE 2.2 PERCENTAGE OF ECONOMIC INCLUSION PROGRAMS, BY TYPE OF NATURAL RESOURCE MANAGEMENT OR CLIMATE CHANGE ADAPTATION INTERVENTION 21 FIGURE 2.3 PERCENT DISTRIBUTION OF ECONOMIC INCLUSION PROGRAMS AND BENEFICIARIES BY REGION 22 FIGURE 2.4 LINKS BETWEEN SOCIAL PROTECTION INSTRUMENTS AND CLIMATE ADAPTATION AND MITIGATION 23 FIGURE 3.1 EXPANSION OF PKH COVERAGE: NUMBER OF BENEFICIARY HOUSEHOLDS (2007-20) 27 FIGURE 3.2 COVERAGE OF MAIN SOCIAL PROTECTION PROGRAMS BY DECILE (RURAL VS URBAN) 30 FIGURE 3.3 COVERAGE OF TARGET HOUSEHOLDS BY MAIN SOCIAL PROTECTION PROGRAMS BY REGION (2019)29 (%) 31 FIGURE 3.4 PKH BENEFICIARY INCIDENCE IN THE POOREST 20 PERCENT (2014-19) (CONDITIONAL CASH TRANSFERS) 32 FIGURE 4.1 SHARE OF POPULATION (PERCENT) VULNERABLE TO POVERTY (BY RURAL AND URBAN AREAS OF ISLAND REGION) (2011-19) 39 Tables TABLE 4.1 SUMMARY SCORES (PART 2 OF THE STRESS TEST) 40 P. I I I This report was authored by Asha Williams, Gracia Hadiwidjaja, Rabia Ali, and Imam Setiawan. The report is largely based on findings from an Adaptive Social Protection (ASP) Stress Test for Indonesia prepared by Shonali Sen, and the team thanks her for her thorough assessment and diligence. Valuable Acknowledgments comments and contributions to this background paper were received from (in alphabetical order), Department of Foreign Affairs and Trade (DFAT) (Australia) Human Development Team Jakarta, Ugo Gentilini, Sara Giannozzi, David James Kaczan, Ilsa Meidina, Shinsaku Nomura, Utz Johann Pape, Habib Rab, Achim Daniel Schmillen, Ekki Syamsulhakim, and Brian Walsh. The team thanks Rizky Fitriany for her diligent administrative support and Chris Stewart for editing the report. Graphic design by Muhammad Kamal. The team would also like to thank the following persons for their feedback, inputs, and support to the Stress Test, which contributed significantly to this report (in alphabetical order): Atin Parihatin, Benedikt Lukas Signer, Cynthia Clarita R. Kusharto, Dewi Novirianti, Ekki Syamsulhakim, Endro Kristanto, Erita Nurhalim, Francesco Strobbe, Hari Kurniawan, Ilsa Meidina, Imam Setiawan, Jian Vun, Natsuko Kikutake, Nurzanty Khadijah, Rabia Ali, Sara Giannozzi, Silviana Puspita, and Sumati Rajput. The team is especially thankful to staff and officials from the Government of Indonesia for their thorough review, comments, and suggestions on the Stress Test that informed this report and for providing additional supporting evidence and documentation, namely (in alphabetical order by agency): the Ministry of National Planning and Development: Maliki, Ph.D, Deputy Minister for Population and Manpower and Dinar Kharisma, Acting Director for Poverty Alleviation and Community Empowerment, and other staff; the Ministry of Social Affairs: particularly staff from the Directorate General of Social Protection and Security, the Directorate of Family Social Security, and the Directorate for Social Protection of Natural Disaster Victims; and staff from the National Disaster Management Authority. Financial support for this report was generously provided by the Australian Government through the Australia-World Bank Indonesia Partnership (ABIP). Financial support was also provided by the Climate Support facility Whole- of-Economy Program, administered by the World Bank. The team is also grateful to Megha Kapoor for her support. P. I V Acronyms ASP ADAPTIVE SOCIAL PROTECTION BAPPENAS BADAN PERENCANAAN PEMBANGUNAN NASIONAL (NATIONAL DEVELOPMENT PLANNING AGENCY) BPBD BADAN PENANGGULANGAN BENCANA DAERAH (LOCAL DISASTER MANAGEMENT AUTHORITY) BKF BADAN KEBIJAKAN FISKAL (FISCAL POLICY AGENCY) BKKBN BADAN KEPENDUDUKAN DAN KELUARGA BERENCANA NASIONAL (NATIONAL POPULATION AND FAMILY PLANNING AGENCY) BNPB BADAN NASIONAL PENANGGULANGAN BENCANA (NATIONAL DISASTER MANAGEMENT AUTHORITY) BPS BADAN PUSAT STATISTIK (CENTRAL STATISTICS BUREAU) BPJS BADAN PENYELENGGARA JAMINAN SOSIAL (SOCIAL SECURITY ADMINISTRATION AGENCY) BST BANTUAN SOSIAL TUNAI (CASH SOCIAL ASSISTANCE) CCT CONDITIONAL CASH TRANSFER CFW CASH-FOR-WORK CMEA COORDINATING MINISTRY FOR ECONOMIC AFFAIRS DRF DISASTER RISK FINANCE DRM DISASTER RISK MANAGEMENT DRTF DISASTER RESPONSE TASK FORCE DTKS DAFTAR TERPADU KESEJAHTERAAN SOSIAL (INTEGRATED SOCIAL WELFARE DATABASE) DUKCAPIL DIREKTORAT KEPENDUDUKAN DAN CATATAN SIPIL (DIRECTORATE OF POPULATION AND CIVIL REGISTRATION) EAP EAST ASIA AND THE PACIFIC EWS EARLY WARNING SYSTEMS FDS FAMILY DEVELOPMENT SESSION GBV GENDER-BASED VIOLENCE GDP GROSS DOMESTIC PRODUCT P. V GHG GREENHOUSE GAS GOI GOVERNMENT OF INDONESIA ID IDENTIFICATION JADUP JAMINAN HIDUP (LIVING SUPPORT ASSISTANCE) MHEWS MULTI HAZARD EARLY WARNING SYSTEM MOECRT MINISTRY OF EDUCATION, CULTURE, RESEARCH, AND TECHNOLOGY MOF MINISTRY OF FINANCE MOHA MINISTRY OF HOME AFFAIRS MOM MINISTRY OF MANPOWER MOPWPH MINISTRY OF PUBLIC WORKS AND PUBLIC HOUSING MOSA MINISTRY OF SOCIAL AFFAIRS MOV MINISTRY OF VILLAGES NDRF NATIONAL DISASTER RESPONSE FRAMEWORK NGO NONGOVERNMENT ORGANIZATION NIK NOMOR INDUK KEPENDUDUKAN (16-DIGIT UNIQUE IDENTIFICATION NUMBER) P3KE PENSASARAN PERCEPATAN PENGHAPUSAN KEMISKINAN EKSTREM (TARGETING THE ACCELERATION OF EXTREME POVERTY ALLEVIATION) PBI-JKN PENERIMA BANTUAN IURAN - JAMINAN KESEHATAN NASIONAL (NATIONAL HEALTH INSURANCE -RECIPIENT CONTRIBUTION ASSISTANCE) PDHA POST-DISASTER HOUSEHOLD ASSESSMENT PENA PAHLAWAN EKONOMI NUSANTARA (ARCHIPELAGO ECONOMIC HEROES PROGRAM) PFB POOLING FUND UNTUK BENCANA (POOLING FUND FOR DISASTERS) Executive P. 0 1 Summary limate change, and its associated impacts, threatens to reverse decades of global C progress in improving people’s health, human capital accumulation, and poverty reduction. Climate change impacts, such as heatwave intensity and extreme weather patterns, increase the risks of infectious disease, vector-borne infections, undernourishment, morbidity, and mortality (IPCC 2014). Globally, climate change generates hidden environmental, health, and poverty costs estimated at almost US$12 trillion per year (Food and Land Use Coalition 2019). Climate change is also expected to increase the frequency and severity of natural hazards and disasters (Hallegatte et al. 2016). These shocks have disproportionate impacts on the poor (Hallegatte et al. 2018) who routinely engage in negative coping methods such P. 0 2 as selling assets, removing children from school, and reducing consumption. Climate change has also caused adverse impacts on social infrastructure which is critical for delivery of education, health, and social services. These impacts can be direct‒through floods and heat stress‒or indirect‒through increased water and air pollution, and diseases. Climate adaptation and mitigation efforts are, therefore, essential to facilitating continued progress in health, human capital accumulation, and poverty reduction. At the same time, individuals and households with more human capital and are better positioned to withstand climate change impacts. Several studies have established a correlation between higher human capital with faster disaster preparedness and recovery. For instance, evidence from the 2004 Indian ocean tsunami in Indonesia found that those with higher levels of education were better able to minimize dips in spending and were in better psychosocial health than those with less education five years after the event. A recent Welfare Tracking (WelTrAC) survey following the 2018 Central Sulawesi earthquake and tsunami in Indonesia (Purnamasari et al. 2021), found that high school graduates experienced employment recovery faster than lower educated groups, while tertiary-educated household heads demonstrated faster welfare recovery compared to lower educated households. These challenges are particularly pressing for Indonesia, where the poor are disproportionately affected by climate shocks. Indonesia has high exposure to covariate (community-level) shocks that stem from climate change. There were 3,622 disasters caused by natural hazards in 2019 alone (The Jakarta Post 2019), approximately 90 percent of which were hydrometeorological phenomena which are expected to worsen due to climate change. Over 110 million people in approximately 60 Indonesian cities are exposed to negative impacts of climate change (World Bank 2019), with the country’s urban poor being most vulnerable. Of the 76 million flood-exposed people in Indonesia, 40 million live in poverty at less than US$5.50 per day (14.3 percent of the population), and 16 million (5.7 percent) on less than US$3.20 a day. Indonesia is also among ten countries with the highest number of poor people exposed to floods (Hallegatte et al. 2017). Furthermore, studies have found that, in the face of climate-related shocks, Indonesian P. 0 3 households protected their food expenditures at the expense of nonfood expenditures, including on health and education (Skoufias 2012), and reduced food consumption (Purnamasari et al. 2021). A worrying projection that climate change could have direct impacts on global stunting rates has direct implications for Indonesia, where the stunting rate is 21.6 percent (Government of Indonesia 2022). Finally, previous research found that damage to 22,323 hectares of plantation or forest was associated with a 1.69 percent decrease in secondary school enrollment (Rush 2018). The disproportionate impact of climate change on poor households, and those vulnerable to poverty, signals the importance of social protection as a critical interlocutor to help address the pressing threat of climate change and climate shocks. Social protection serves an important connector role in helping to build human capital of the poorest and most vulnerable households so they can adapt to climate risk and contribute to mitigation efforts. A notable example of social protection’s potential in this regard for Indonesia comes from a recent evaluation which found that villages that participated in the country’s flagship conditional cash transfer (CCT) program (Program Keluarga Harapan: PKH) experienced a 30 percent reduction in forest cover loss. This was due to, inter-alia, PKH’s consumption substitution impacts that enabled households to substitute deforestation-sourced consumption goods with market-purchased goods (Ferraro et al. 2020). More recently, provision of government and nongovernment assistance played a significant role in increasing the probability of faster, long-term recovery in employment and restoring household welfare following the 2018 Central Sulawesi earthquake, tsunami, and liquification (Purnamasari et al. 2021). This background paper outlines the important relationship between human capital development and climate change adaptation; and the needs and opportunities for improving the adaptiveness of Indonesia’s social protection system. While the paper signposts the importance of education and health and nutrition interventions for mitigating the impact of climate change on human capital, it focuses on social protection, given its importance for addressing the disproportionate impacts of climate change on the poor and vulnerable. The paper: (i) outlines the importance of social protection for addressing climate risk among households as part of a broader P. 0 4 suite of human capital policies; (ii) provides a brief overview of the development of Indonesia’s social protection system; (iii) stress tests the adaptiveness of Indonesia’s social protection system to respond to climate shocks; and (iv) recommends critical priorities to help ensure that the system is better positioned to help poor and vulnerable households cope with, and adapt to, climate risk. Given these pressing challenges, an Adaptive Social Protection Stress Test of Indonesia’s social protection System (Sen et al. 2022) was carried out to assess its adaptiveness to respond to climate shocks, the results of which are presented in this report. The application of the Stress Test, developed by Bodewig et al. (World Bank 2021), assessed the readiness of Indonesia’s social protection system to build resilience to shocks and to respond to heightened needs through a detailed assessment of four building blocks: (i) programs and delivery systems; (ii) data and information; (iii) finance; and (iv) institutional arrangements and partnerships (Part 2 of the Stress Test). In addition, recent analysis by Ali and Setiawan (2022) which examined levels and sources of vulnerability to poverty and shocks, was leveraged to assess the scale of need for social protection support in an average covariate crisis in Indonesia (Part 1 of the Stress Test). Indonesia has made considerable progress in the development of its social protection system and by leveraging its social protection programs and services to respond to household vulnerabilities and risks. Notable achievements include, inter-alia, provision of a suite of core, poverty-targeted, and post-shock social assistance programs; expansion of the flagship CCT program (PKH) to 10 million households; and establishment of an integrated social welfare database (Data Terpadu Kesejahteran Sosial: DTKS) of the poorest 40 percent of the population (used to identify potentially eligible households for several social protection programs). Importantly, PKH has also been found to have a positive effect on deforestation in villages where households participate in the program (Ferraro et al. 2020). For Part 1 of the Stress Test, the estimations revealed that 12.16 percent of the rural population (11. 4 million households) and 4.26 percent of the urban population (6.6 million households) were likely to need social protection support in an average crisis. Ali and Setiawan (2022) decompose Indonesian households’ P. 0 5 vulnerability to falling into poverty over a defined time period into two components: (i) poverty-induced vulnerability‒when the average expected consumption of a household falls below the poverty line in the absence of shocks; and (ii) risk-induced vulnerability‒when the average expected consumption falls above the poverty line, but during shocks (both idiosyncratic (i.e. household-level) and covariate (i.e. community-level)), consumption is expected to fall below the poverty line. Overall, the analysis found that for Indonesia, the share of risk-induced vulnerability at the national level rose from 80 percent of to­tal vulnerability in 2011 to 90 percent in 2019. The study found that about one in ten Indonesians were vulnerable to falling into poverty due to covariate shocks in 2019, and covariate risk-induced vulnerability was about six times and 8.3 times higher than poverty- induced vulnerability in rural areas and urban areas respectively. These findings have implications for Indonesia’s capacity to scale up social protection. Although the current DTKS covers more than 22 percent of the population, no single social safety net program has covered more than 20 percent of the population. Overall, the ‘Stress Test’ to assess the capacity of Indonesia’s social protection system to respond to climate shocks has found the country to be operating at an Emerging Level of 3.26 (on a scale of 1 to 5 ).* When assessed across four building block areas, the country received a score of 3.47 for Programs and Delivery Systems; 3.08 for Data and Information; and 3.25 for both Financing, and Institutions and Partnerships. Indonesia performed well on: (i) having a range of regular and shock-responsive social protection programs; (ii) national identification and electronic payment delivery; (iii) the ability to quickly issue post-shock payments to existing beneficiaries; and (iv) having Early Warning System (EWS) platforms covering all relevant natural hazards. The country also has a clear disaster risk financing strategy encompassing various shocks‒supported by strategies, policies, and laws governing social protection and Disaster Risk Management (DRM), with clear assignment of roles. *The specific scores and corresponding levels are (1) Latent; Despite this, the Stress Test also found that lingering gaps to (2) Nascent; (3) Emerging; (4) Advanced; and (5) Established. social protection effectiveness hamper the system’s ability to effectively build adaptive capacity of the poorest and most vulnerable to sufficiently prepare for, and respond to, the impacts P. 0 6 of climate shocks and climate change.1 Social registries (under Data and Information) were identified as the area in need of most improvement, with a score of 2.9‒given the need to ensure improved dynamism and recency of data; improve its use for post-shock responses; and ensure more effective links to an EWS to better predict needs and define thresholds for action. Other major gaps included: (i) a lack of deliberate program interventions or linkages to improve climate resilience among social protection beneficiaries; (ii) delays with providing post-shock benefits for non-beneficiary affected households (who could number up to 18 million households during a covariate shock); (iii) limited adequacy of post-shock benefits; (iv) a lack of integrated post-disaster household assessment to inform social protection responses; and (v) a need for improved capacity to model the potential costs of different shocks over time. 1 There have been recent efforts to address some of these Given these findings, the recommended priority areas of focus for challenges since the application of the Stress Test, which are noted government include: later in the report. 1 Close lingering social protection coverage gaps to ensure poor and vulnerable are adequately protected by benefits and services for improved climate resilience and adaptive capacity. Gaps in social assistance and social insurance coverage reveal that the burden on the social protection system to scale-up to non-beneficiaries (i.e. horizontal expansion) during large covariate shocks is likely to be high. Furthermore, these coverage gaps mainly apply to existing poor who currently do not receive core social assistance benefits. In addition, to the extent that those facing risk-induced vulnerability to covariate shocks can be supported by benefits and services for which they are eligible, this will also help address their vulnerabilities and make them more resilient to the impact of future shocks. 2 Improve direct activities in social protection programs and linkages to other sector programs, to build adaptive capacity of beneficiary households. The Stress Test found few systematized approaches, deliberate interventions, or complementary benefits in existing social assistance programs to improve climate resilience among beneficiaries. This can be done by: (i) scaling up education and information on climate change and shock preparedness to beneficiaries; (ii) ensuring that social protection programs (especially cash-for-work and housing-related benefits) have direct P. 0 7 adaptation and resilience-informed design; and (iii) improving access to complementary benefits and assistance offered by other ministries and agencies for which social assistance beneficiaries could be prioritized. 3 Improve the use of social registry systems for shock response and to support climate policy. Specifically: (i) improve the dynamism and quality of regular data updating for social registry systems that support targeting and ensuring such principles are foundational in any future registry systems; (ii) expand social registry coverage of the population in both poor and disaster-prone areas (such as was done through nationwide socioeconomic registration);2 (iii) ensure that the data collected is useful for responding in a shock; (iv) facilitate improved access to social registry data by humanitarian agencies in case of a shock; (v) improve data privacy; and (vi) strengthen disaster protection of social protection information systems. In addition, it would be important for Government to develop a disaster victim’s database to streamline data sharing of post-disaster assessment data of affected households. Optimally, these actions could be linked to the future development of an Integrated Social Protection Information System to facilitate dynamic data updates for eligibility determination; integrated view of benefit and services delivery; and more on-demand access to the population for social protection benefits and services. 4 Improve mechanisms for faster horizontal expansion and delivery of post-disaster social assistance benefits. This includes: (i) expanding social registry coverage to a larger share of the population; (ii) leveraging technology more effectively to identify non-beneficiary households in affected areas; (iii) establish an integrated post- disaster household assessment (PDHA) process that is deployed rapidly, is interoperable with other relevant social protection information systems, and coordinated across agencies. 5 Improve gender-sensitivity and attention to vulnerable groups in post-shock social protection operations. This includes: (i) improving assessment and information sharing on affected households with vulnerable members; (ii) ensuring direct messaging to women in affected households; (iii) providing post-disaster benefits directly to women, people with disability and the elderly to improve their agency 2 (LINK) and/or ensuring measures to monitor and respond to their needs; P. 0 8 and (iv) ensuring that temporary shelters are accessible and have better measures to prevent gender-based violence. 6 Develop an integrated tool linked to Early Warning Systems (EWS) data to quantify post-shock social protection needs and estimate optimal post-shock benefit levels. A key finding of the Stress Test was that post-disaster financial planning for social protection was largely based on a retrospective view of the previous year’s costs. Better quantification of potential post-shock needs before shocks occur, could help improve this process. This would optimally be complemented by establishing triggers linked to EWS that could facilitate automatic scale-up of social protection depending on established metrics. 7 Continue to build on progress with electronic payment delivery and facilitate broader choice among payment mechanisms, particularly for the post-shock response. Key actions include (i) ensuring more beneficiary-responsive payment modalities; (ii) addressing the gaps that result in payment delays during shock times for both regular social protection benefits and emergency transfers; and (iii) build on ongoing efforts include the plan to develop a Central Mapper for Government-to-Person (G2P) payments, essentially a repository of unique individuals linked to a particular payment information (such as bank account) for the purpose of routing payment transactions, to improve monitoring, accountability and speed of payments. These actions would optimally be supported by, and linked to, an integrated beneficiary database to facilitate onboarding and monitoring. 8 Fully operationalize ongoing reforms to strengthen the social protection system’s adaptability to climate risk. These ongoing reforms include the creation of a Disaster Pooling Fund, which will improve government’s risk layering and financial protection in the event of future shocks with direct linkages to social protection responses. On the policy front, Bappenas is finalizing an ASP Roadmap and its associated regulations that will recommend integration of social protection and climate change action and provide a guide for leveraging social protection to address risks from natural and climate-related hazards, particularly for poor and vulnerable populations. P. 0 9 Although the focus of this paper is social protection, it is equally important to ensure continued investments in education and health to support climate resilience objectives for a comprehensive and integrated human capital approach to addressing climate risk. These could include efforts to: (i) protect social infrastructure in health and education to ensure business continuity in the face of shocks; (ii) continued priority to stunting reduction given the possible climate change impacts on stunting; and (iii) facilitating improved education completion outcomes, particularly length of schooling, given previous study findings that those with higher education attainment are better able to withstand the impacts of shocks. 01 P. 1 0 P.13–18 Introduction: The Interconnectedness of Climate Change, Human Capital, and Poverty limate change and human capital development are mutually impacting. Understanding C the relationship between them is, therefore, critical to ensuring effective climate change mitigation and adaptation and facilitating meaningful human capital development for more sustainable poverty reduction outcomes. Climate change and its associated impacts cause adverse impacts on people’s health, human capital accumulation, and overall wellbeing (World Bank 2017). For health outcomes, climate change impacts, such as heatwave intensity and extreme weather patterns, increase the risks of infectious disease, vector-borne infections, undernourishment, morbidity, and mortality (IPCC 20104). Globally, climate change generates hidden environmental, health, and poverty costs estimated at P. 1 1 almost US$12 trillion per year (Food and Land Use Coalition 2019). Climate change is expected to increase the frequency and severity of natural hazards and disasters (Hallegatte et al. 2016). Past climate shocks have demonstrated that losses disproportionately affect the poor (Hallegatte et al. 2018) who routinely engage in negative coping methods such as selling assets, removing children from school, and reducing consumption to address the impacts of climate shocks. Climate change has also caused adverse impacts on social infrastructure.3 These impacts can be direct‒through floods and heat stress‒or indirect‒through increased water and air pollution and diseases. Climate adaptation and mitigation efforts are, therefore, essential to facilitating continued progress in health, human capital accumulation, and poverty reduction. Indonesia has high exposure to covariate (community-level) shocks4 that stem from climate change. Indonesia experiences frequent natural disasters, with 3,622 total disasters occurring in 2019 alone (The Jakarta Post 2019). Of these, approximately 90 3 The New Zealand Social percent are hydrometeorological phenomena, including tornadoes, Infrastructure Fund defines social Infrastructure as a subset flooding, and landslides‒all of which are expected to worsen due of the infrastructure sector that typically includes assets that to climate change. High population density in hazard-prone areas, accommodate social services. (LINK) coupled with strong dependence on the country’s natural resource base, make Indonesia vulnerable to climate variability. Over 110 4 Covariate shocks affect many households in a region or million people in approximately 60 Indonesian cities are exposed community at the same time. Examples include droughts, to negative impacts of climate change (World Bank 2019)‒with the floods, earthquakes and other natural disasters, spikes in food country’s urban poor being most vulnerable. Of the 76 million flood- prices, and epidemics. These are exposed people in Indonesia, 40 million live in poverty on less than experienced simultaneously by most if not all other households US$5.50 per day (14.3 percent of the population), and 16 million in a community and are, therefore, difficult to insure. (5.7 percent) on less than US$3.20 a day. The country’s reliance on This makes a strong case for provision of insurance by agriculture-based livelihoods heightens the risk of poverty for those government as it can spread risk across communities and create exposed to climate-related shocks such as variation in precipitation risk-sharing mechanisms across and increases in temperature. Finally, the country’s location on the so- geographies. called Ring of Fire exposes its population to high risk of earthquakes, 5 The authors looked at poverty exposure bias for floods volcanic eruptions, tsunamis, and landslides. This finding has been with a 10-year return period (or 10 percent annual probability of reinforced in other analyses that show higher exposure of the poor occurrence) in select countries in people to floods in Indonesia compared to select countries in East Latin America and the Caribbean, Africa, and Asia. Asia (Hallegatte et al. 2020) (Figure 1.1).5 P. 1 2 FIG 1.1 POVERTY EXPOSURE TO FLOODS 6 POOR PEOPLE MORE EXPOSED POOR PEOPLE ARE LESS EXPOSED NOT SIGNIFICANT NO DATA A. LATIN AMERICA AND THE B. AFRICA AND EUROPE, NATIONAL C. ASIA, NATIONAL CARIBBEAN, NATIONAL MOLDOVA DOMINICAN HAITI REPUBLIC ALBANIA ATLANTIC OCEAN HONDURAS MOROCCO JORDAN GUYANA ARAB REPUBLIC COLUMBIA OF EGYPT SENEGAL NEPAL MALI NIGER PERU BENIN NIGERIA ETHIOPIA CENTRAL AFRICAN PACIFIC UGANDA REPUBLIC CAMBODIA KENYA OCEAN GUINEA LIBERIA GABON SIERRA LEONE ANGOLA ZAMBIA INDONESIA MADAGASCAR CAMEROON INDIAN OCEAN ATLANTIC OCEAN NAMBIA MOZAMBIQUE ATLANTIC OCEAN ZIMBABWE Source: Hallegatte et al. 2020. Such covariate shocks make many vulnerable to falling into poverty or deeper into poverty. Indonesia is among the 10 countries with the highest number of poor people exposed to floods (Hallegatte et al. 2017).7 Past research in Indonesia (Pritchett et al. 2000; Chaudhuri et al. 2002; World Bank 2006; Wai-Poi 2014; World Bank 2019; and World Bank 2020) has documented a high aggregate level of poverty risk and `churning’ of households around the poverty line, even as the poverty headcount rate fell through 2019. This vulnerability to poverty matters as even short spells of lowered consumption during large shocks such as those that stem from climate change can reduce productivity in the long run due to adverse impacts on human capital investments at the household level and/or reliance on adverse strategies of coping with income shocks‒for example, sale of productive assets (see Alderman et al. 2006; Gubert and 6 Data for 52 countries: all Robilliard 2007; Rosenzweig and Binswanger 1993; and Klasen and households. Waibel 2014). Furthermore, vulnerable households may anticipate 7 The other countries include India, Bangladesh, Arab Republic shocks and, as a result, adopt conservative or risk-averse production of Egypt, Vietnam, Democratic Republic of Congo, Nigeria, Mexico, and investment strategies that lead to low consumption (Elbers et Iraq, and Sudan. al. 2007). Regardless of whether the adverse coping strategy is adopted ex post or ex ante, productivity is reduced in the long run which, in turn, lowers the chances of securely escaping poverty. P. 1 3 Climate change impacts also threaten Indonesia’s progress on human capital development. For instance, Skoufias et al. (2012) found that rice-farming households residing in areas that experienced low rainfall following the monsoon’s onset in Indonesia experienced a 14 percent reduction in their per capita expenditures and, in the face of weather shocks, these households protected their food expenditures at the expense of nonfood expenditures‒including on health and education‒thereby directly reducing investment in human capital with potential longer-term effects on poverty reduction. More recently, a Welfare Tracking survey found that one of the most common coping strategies for households in the bottom 40 percent affected by the Central Sulawesi earthquake, tsunami, and liquification, was to reduce food consumption (Purnamasari et al. 2021). One worrying projection is that climate change could have direct impact on global stunting rates, with direct implications for Indonesia where the stunting rate is 21.6 percent (Government of Indonesia 2022). Expected agricultural impacts from climate change are expected to increase undernutrition and result in an additional 7.5 million children with severe stunting by 2030 globally (Hales et al. 2014). Education outcomes are also threatened by increased climate change and increased frequency and severity of shocks. For instance, previous research found that damage to 22,323 hectares of plantation or forest was associated with a 1.69 percent decrease in secondary school enrollment (Rush 2018). That study also found that higher poverty incidence exacerbated the negative impact of disasters on school enrollment. These impacts point to the importance of introducing mechanisms to mitigate against the education and health impacts caused by climate change and further 8 For example, studies have found reinforce the disproportionate reductions in human capital faced by that educated individuals were poorer households. more likely to survive and had a lower Stronger human capital is important to improving resilience risk of injuries from the 2004 Indian to climate change and climate shocks.8 Several studies have ocean tsunami and communities established a correlation between higher education outcomes and and countries with higher average disaster preparedness and recovery. For instance, evidence from the levels of education 2004 Indian ocean tsunami in Indonesia found that those with more also experienced much lower losses education were better able to minimize dips in spending following in human lives from climate-related the tsunami, compared to the spending cuts made by those with disasters. P. 1 4 less education (Frankenberg et al. 2013). That study also found that more educated males had a higher probability of surviving the tsunami and, five years after the event, those with more education were in better psychosocial health than those with less education. One study in Thailand and the Philippines found that formal education increased the propensity to prepare for disasters, and that the highly educated exhibited higher levels of disaster preparedness due to their comparatively better abstraction skills in anticipating the consequences of disasters (Hoffmann and Muttarak 2017). These findings have been recently reinforced for Indonesia in the WelTrAC survey that found that high school graduates experienced employment recovery faster than lower educated groups and that tertiary-educated household heads demonstrated a higher probability of, and faster, welfare recovery by October 2019 , compared to lower educated households (Purnamasari et al. 2021).9 As Indonesia embarks on more aggressive climate mitigation strategies, education, health, and social protection policies will be essential complements to ensure that these efforts result in net positive gains for human capital, wellbeing, and climate resilience‒ particularly for those whose livelihoods and incomes depend on carbon-intensive industries and those who are more vulnerable to climate change impacts due to poverty. Ensuring sustainable development outcomes in the context of a changing climate, therefore, requires integrated climate change adaptation and mitigation efforts, combined with climate- sensitive poverty reduction and human capital development policies. Previous analyses have advocated that without measures to facilitate poverty reduction and human capital development, 9 For the WelTrAC study, deliberately coupled with targeted climate resilience measures, welfare was measured using data on households’ assets and other climate change could force more than 100 million people globally socioeconomic and demographic characteristics using a Proxy into extreme poverty by 2030 (Hallegtatte et al. 2016). To be most Means Test (PMT) approach to estimate households’ wealth effective, designing and implementing solutions to end extreme using a consumption regression and categorizing households poverty and to stabilize climate change as an integrated strategy into welfare quintiles. Variables was, therefore, necessary (Hallegtatte et al. 2016). The same holds included demographic structure of households, occupation and true for Indonesia which faces dual challenges of climate change risk education of household members, and the ownership of assets and barriers to effective poverty reduction and shared prosperity. (Purnamasari et al. 2021). This background paper outlines the important relationship between human capital development and climate change P. 1 5 adaptation and the urgent need to improve the adaptiveness of Indonesia’s social protection system to address these challenges. While the paper signposts the importance of education and health and nutrition interventions for mitigating the impact of climate change on human capital; it focuses on social protection, given its importance for addressing the disproportionate impacts of climate change on the poor and vulnerable. The subsequent sections of the paper: (i) outline the importance of social protection for addressing climate risk among households as part of a broader suite of human capital policies; (ii) provide a brief overview of the development of Indonesia’s social protection system; (iii) stress tests adaptiveness of Indonesia’s social protection system to respond to climate shocks; and (iv) draws on this analysis to recommend critical priorities to be addressed if Indonesia’s social protection system is to better help poor and vulnerable households cope with, and adapt to, climate risk. 02 P. 1 6 P.15–23 The Importance of Social Protection in Addressing Climate Risk, Contributing to Human Capital Outcomes, & Supporting Indonesia’s National Development Priorities P. 1 7 ndonesia has set forward an ambitious Vision 2045, which aims to have the country achieve I high income status by its 100th anniversary of the country’s independence. The Vision is supported by National Medium-Term Development Plans (Rencana Pembangunan Jangka Menengah Nasional: RPJMNs) which cover medium-term priorities over subsequent five-year periods. The current RPJMN 2020-2024, outlines the country’s roadmap “to create an independent, advanced, just, and prosperous Indonesian society,” through accelerating development in various sectors supported by qualified and competitive human resources. Human capital development features strongly in the current RPJMN, with improvement in human resources as one of seven development priorities and a recognition that human capital development is critical to inclusive and equitable development. As such, the RPJMN outlines objectives for improving the implementation of social protection, improving health coverage and implementation, and increasing the quality of and equity in the education sector. Human capital targets in the RPJMN include obtaining an average length of school enrollment of 9.18 years for persons 15 years of age and over; a maternal mortality rate of 183 per 100,000 live births; a reduction in stunting prevalence to 14 percent; and enabling at least 66.7 percent of college graduates to obtain work within one year of graduation.10 Social protection is a critical contributor to Indonesia’s ambitious national development priorities. The RPJMN’s 2024 targets include, inter-alia, facilitating access to social protection for 98 percent of the population and ownership of productive assets among 98 percent of the population. More specifically, the RPJMN aims to improve the effectiveness of social assistance programs and delivery systems toward a comprehensive social protection scheme (Major Project Number 18) and expanded coverage and institutional strengthening 10 The RPJMN’s Major Projects of the National Social Security System (Sistem Jaminan Sosial numbers 15 and 17 relate Nasional: SJSN). to targets for the health and education sectors respectively. Social protection also plays an importing cross-cutting role in the RPJMN’s priorities for addressing climate risk. For instance, Major Project Number 13 outlines objectives to improve the quality of life P. 1 8 and speed up recovery and economic conditions for communities affected by disasters, with the Ministry of Social Affairs (MoSA) among the responsible agencies for this major project. Importantly, the RPJMN also includes targets for reducing greenhouse gas emis- sions (GHGs), with a final target of 29 percent by 2030. Finally, the Government of Indonesia (GoI) has announced an ambitious target to eliminate extreme poverty by 2024.11 The plan includes targets to lower the extreme poverty rate to 3-3.5 percent in 2022 through measures in 212 regencies/municipalities. The government plans to expand these measures to 514 regencies/municipalities with an ex- treme poverty rate target of 2.3-3 percent by 2023. Social protection will, therefore, be critically important to the country’s national devel- opment priorities given these ambitious and time-sensitive targets. The disproportionate impacts of climate shocks on poor house- holds, and their threat to push nonpoor households into poverty, requires interventions that include poverty reduction and resil- ience building as central areas of focus in climate policy. These impacts have been documented widely. For instance, Hallegatte et al. in their notable 2016 Shockwaves report, flagged the disproportion- ate impacts that climate change has on the poor, primarily due to: (i) their higher exposure to, and impact from, these climate shocks or trends; (ii) their higher vulnerability relative to their income or wealth; (iii) their comparatively lower levels of support from family and 24.6% of households community networks; and (iv) less access to financial tools or social safety nets to help prevent, prepare for, and manage these impacts (Hallegatte et al. 2016). In Indonesia and other East Asia and the Pacific (EAP) countries, wealthier households were found to be more in the poorest 40 percent in likely to take proactive ex ante climate change adaptation measures, Indonesia saved while poorer households were most likely to react to shocks ex post at a financial (Francisco et al. 2011). As with other countries, Indonesia’s poor institution in 2021 have fewer assets and lack access to savings to help them mitigate the impact of climate shocks on their wellbeing. For instance, only 24.64 percent of households in the poorest 40 percent in Indonesia saved at a financial institution in 2021, compared to 43.83 percent in the top 60 percent of the income distribution.12 Social protection is important to addressing climate risk as 11 (Link) part of a broader suite of human capital policies. It serves as an 12 Susenas 2021. important connector role in helping to build human capital of the P. 1 9 poorest and most vulnerable households. In addition, numerous studies have demonstrated that social protection programs have contributed to mitigation efforts; helped households adapt to climate risk; and have been critical to preparedness, coping, and recovery from climate shocks among households. A notable example of social protection’s potential in this regard for Indonesia comes from a recent evaluation which found that villages that participated in the country’s flagship conditional cash transfer (CCT) PKH program experienced a 30 percent reduc- 13 Results at 95 percent tion in forest cover loss (Ferraro and Simorangkir 2020) (Figure confidence interval, 10 to 50 2.1).13 Among the factors attributed to the correlation were longer percent. The study included 7,468 rural forested villages exposed exposure to the program and higher PKH participation density per to PKH between 2008 and 2012 across 15 provinces‒with 266,533 hectare of forest, which was associated with larger reductions in households in these villages receiving transfers by 2012. The forest cover loss. Notably, the study found no discernible impacts in study found that approximately villages before they were exposed to the PKH program. The factors at- one-half of the avoided losses were in primary forests and tributed to the relationship in the study included PKH’s consumption that the economic value of the avoided carbon emissions alone smoothing impacts which helped beneficiary households substitute compared favorably to program implementation costs. deforestation for cash; and consumption substitution, where house- 14 Results include a decline in holds were able to substitute deforestation-sourced consumption stunting among PKH beneficiaries goods with market-purchased goods. More recently, the WelTrAC by 9 to 11 percentage points; an increase in primary school survey following the 2018 Central Sulawesi earthquake, tsunami, enrollment by 4 percentage points; an enrollment rate of around 93 and liquification in Indonesia found that provision of government percent; and an increase in junior secondary school enrollment by and nongovernment assistance played a significant role in increas- about 8 percentage points. ing the probability of faster, long-term recovery in employment and 15 The study found that 98.4 restoration of household welfare (Purnamasari et al. 2021). percent of prosperous-independent graduate household members 7-12 years of age continued attending Social protection has been proven to facilitate improved human school after they left PKH. This did not differ much from active capital among the poorest in Indonesia and, therefore, serves beneficiary household members. as an important contributor to facilitating improved resilience to In addition, graduate household members 13-18 years of age were climate change impacts. Social protection programs, particularly attending school at even higher rates (87.1 percent) compared CCTs, have been critical to linking poor households to education and to active beneficiary household members (83.5 percent). For health services and facilitating human capital outcomes among those health, 92 percent of former PKH beneficiaries who gave birth in most vulnerable. For instance, results from PKH impact evaluations 2020 reported checking their have shown that the program has had positive impacts on improved pregnancy at least four times during the pregnancy‒as one of consumption patterns, education enrollment, and positive health the PKH conditions‒and 94.2 percent who gave birth in 2020 behaviors such as maternal and neo-natal practices (World Bank delivered their babies in healthcare facilities. 2020).14 In addition, recent analysis showed that these behaviors are sustained by PKH graduates after their exit from the program (Syamsulhakim and Khadijah 2021).15 These human capital impacts, P. 2 0 PKH SEMBAKO FIG 2.1 PIP CHANGE IN FOREST COVER LOSS IN PKH PARTICIPATING VILLAGES BPJS HEALTH 15 10 PKH IMPACT ON DEFORESTATION (HA/YEAR/VILLAGE) 5 7468 VILLAGES 0 –5 –10 –15 –20 IMPLEMENTATION –25 YEAR OF 4 3 2 1 1 2 3 4 YEAR(S) BEFORE YEAR(S) AFTER Source: Ferraro and Simorangkir, 2020. supported by social protection initiatives, also help to contribute to Indonesia’s broader development objectives for an improved, more skilled, and healthy human resource base that contributes to a growing economy through participation in more productive sectors. Social protection also has strong potential to support climate adaptation and mitigation through complementary benefits; climate-sensitive economic inclusion; and by leveraging social protection interventions and delivery systems more deliberately for shock preparedness, response, and recovery. Safety net programs in particular are increasingly providing non-social protection linkages to beneficiaries to provide a more comprehensive suite of benefits and services to facilitate poverty reduction. These can help further climate 16 Examples of climate-smart adaptation objectives‒for instance, in the housing sector, where housing projects in complementary benefits can support climate-smart and climate- EAP include those implemented in resilient housing for the poor16 and in the energy sector to facilitate Cambodia and Vietnam. diversification away from fossil fuels and harmful energy sources. P. 2 1 Economic inclusion17 interventions have also become a more salient feature of social protection provision across the globe, including in Indonesia. These have strong potential to facilitate preparedness, 17 Defined by Andrews et al. behavior change, and diversification away from harmful livelihood (2021) as “a bundle of coordinated practices and towards more climate-sensitive sources of livelihoods. multidimensional interventions that support individuals, This is playing an increasingly important role among the objectives of households, and communities to increase their incomes and assets. economic inclusion interventions, with the 2021 State of Economic Economic inclusion programs therefore aim to facilitate the dual Inclusion report (Andrews et al. 2021) noting that 55 percent of all goal of strengthening resilience programs surveyed had a focus on climate change mitigation. The and opportunities for individuals and households who are poor.” types of these interventions are illustrated in Figure 2.2. FIG 2.2 PERCENTAGE OF ECONOMIC INCLUSION PROGRAMS, BY TYPE OF NATURAL RESOURCE MANAGEMENT OR CLIMATE CHANGE ADAPTATION INTERVENTION 18 A. Overall B. by entry point PKH SEMBAKO PIP BPJS HEALTH SOCIAL SAFETY NETS (SSNs) LIVELIHOODS AND JOBS (L&J) LAND TENURE SYSTEMS 24.2% 18.6% 27.2% WATER MANAGEMENT 51.6% 55.8% 49.4% ENERGY ACCESS 19.4% 20.9% 18.5% CLIMATE-SMART AGRICULTURE 68.5% 53.5% 76.5% CLIMATE-RESPONSIVE SP SYSTEMS 29.8% 41.9% 23.5% FOREST PROTECTION/MANAGEMENT 46.8% 37.2% 51.9% OTHER 17.7% 16.3% 18.5% Source: Andrews et 18 Note: Panel a shows the percentages of all programs supporting natural resource management or climate change adaptation al. 2021. or both (N = 124). Panel b shows the percentages of these programs by entry point (N = 43 Social Safety Net programs and 81 Livelihoods and Jobs programs). Financial inclusion programs are excluded from this analysis because they are few in number. Programs may include more than one type of intervention. P. 2 2 Where implemented, these interventions could have positive results on beneficiary households’ preparatory and adaptive capacity. For example, an evaluation for Nicaragua demonstrated the utility of economic inclusion in improving resilience to climate shocks, where provision of vocational training or a productive investment grant in addition to a cash transfer to beneficiaries who were vulnerable to drought, provided full protection against drought "Economic shocks two years after the end of the intervention, relative to the inclusion control group who only received a cash transfer (Macours et al. interventions 2012). Despite this potential, economic inclusion appears to be have also become operating at limited scale in the region (Figure 2.3). a more salient feature of social Beyond economic inclusion and complementary benefits, social protection protection interventions such as public works can play an provision across important role in supporting mitigation projects at the community the globe, level and contribute to disaster recovery efforts after climate including in shocks. Previous analysis for Indonesia found that access to credit Indonesia." and public works projects in communities can help households cope with weather shocks and, therefore, help play a strong protective 19 Useful guidance on how these delivery systems can be role during times of crisis (Skoufias 2012). Social protection delivery effectively leveraged for shock response can be found in Bowen systems such as social registries and payment mechanisms can et al. 2020 and Williams and help ensure faster, more effective response to poor and vulnerable Moreira 2020. households affected by climate shocks.19 Finally, active labor 20 While not the focus of this note, a separate background market programs and jobs interventions are critical social protection note will assess the labor market demands for green measures for climate resilience.20 A summary view of how social skills in Indonesia as an input protection can contribute to both adaptation and mitigation efforts to the Country Climate and Development Report. is summarized in Figure 2.4 (Rigolini 2021). FIG 2.3 PERCENT DISTRIBUTION OF ECONOMIC INCLUSION PROGRAMS AND BENEFICIARIES BY REGION PROGRAMS EAP ECA LAC MENA SA SSA 18.8% 14.2% 51.4% 6.0% 7.3% BENEFICIARIES 4.6% 31.3% 61.2% Note: ECA: Europe and Central Asia; LAC: Latin America and Caribbean; MENA: Middle Source: Andrews et al. 2021. East and North Africa; SA: South Asia; and SSA: Sub-Saharan Africa. P. 2 3 FIG 2.4 LINKS BETWEEN SOCIAL PROTECTION INSTRUMENTS AND CLIMATE ADAPTATION AND MITIGATION Social protection and labor protects & prepares people for­: SOCIAL PROTECTION ADAPTATION DECARBONIZATION INSTRUMENT & MITIGATION Cash transfers Adaptive Social Protection (ASP) Payments for Environmental helps households build resilience Services (PES) help manage critical and cope with shocks by facilitating ecosystems, and cash transfers savings, food security and livelihood can reduce deforestation. Social adaptation and diversification. ASP protection measures cash transfers also provides post-shock support are also a central element of Just to mitigate impacts and avoid Transition policies and post-carbon/ damaging coping strategies. energy subsidy reforms. Public works Public works promote food security, Public Works contribute to carbon shock coping and livelihood capture through reforestation and diversification at the household restoration of ecosystems. level, and the creation of adaptive assets at the community level through better management of land and natural resources. They also support post-disaster reconstruction efforts. Livelihoods / Economic Multidimensional programs with Livelihoods and Economic Inclusion Inclusion livelihoods and economic inclusion programs support Just Transition components support medium and policies and enhance resilience to long-term resilience building through labor market disruptions caused by food security, higher productivity, Green transitions in both rural and savings and diversification of urban settings. livelihoods. Training and Active Labor Training and ALMPS support Training and ALMPs support post- Market Programs reskilling and job transitions in carbon "Just Transitions" in energy, urban and rural areas for household coal, agriculture, transport and whose livelihoods are affected by other sectors. They also prepare climate shocks. the workforce for new post-carbon jobs and enhance resilience to labor market disruptions caused by Green transitions. Source: Rigolini 2021. P. 2 4 The design of social protection programs is also critical to ensuring their potential for addressing these challenges. The mere presence of varied social protection benefits and services is not sufficient to ensure that program objectives and spillover effects for climate resilience can be achieved. Program design, benefit and service delivery processes, and monitoring and evaluation are all essential elements to ensuring that: (i) programs identify and include who they are intended for; (ii) beneficiaries receive the benefits and services intended for them; (iii) these benefits and services are adequate to address the risks faced and appropriate to need; and (iv) outcomes are evaluated and delivery is monitored and adjusted as needed. Furthermore, as delivery often requires coordination within and across sectors, institutional arrangements and coordination mechanisms 21 For more detailed guidance also need to be clearly defined and adhered to.21 This is even more and country experiences on social crucial when leveraging social protection benefits and services to help protection delivery systems, please refer to Lindert et al. facilitate improved climate resilience and adaptive capacity. 03 P. 2 5 P.24–29 The Development of Indonesia’s Social Protection System he decade to 2021 has been characterized by significant progress in the reform of T Indonesia’s social protection system to address risks to chronic and transient poverty. What exists today is now a suite of contributory and non-contributory programs offered across the lifecycle, supported by social protection delivery systems. Central to Indonesia’s social protection system is social assistance which today includes a set of core, mostly poverty-targeted, permanent programs, including the flagship PKH CCT program which saw a significant expansion of coverage and in 2020 it covered approximately 10 million households (Figure 3.1). PKH is the second largest CCT in the world after Brazil’s Bolsa Familia program. Other social assistance programs include a cash transfer for poor and vulnerable students (Program Indonesia Pintar: PIP); a food assistance voucher program (Program Sembako/ Bantuan Pangan Non-tunai: BPNT); the subsidized health insurance premium (Penerima Bantuan Iuran - Jaminan Kesehatan Nasional); cash-for-work (Padat Karya Tunai: PKT) programs managed by the P. 2 6 Ministry of Villages (MoV) and Ministry of Manpower (MoM), and a recently piloted program for Social Rehabilitation of Uninhabitable Houses (Rehabilitasi Sosial Rumah Tidak Layak Huni: RS-Rutilahu) offered by MoSA.22 Indonesia has also recently scaled up efforts to improve linkages between poor households and economic empowerment programs. The most notable is the Archipelago Economic Heroes (Pahlawan Ekonomi Nusantara- PENA)23 economic inclusion intervention provided by MoSA. The program provides grants, business mentoring, and facilitation for beneficiaries of MoSA social assistance programs and other vulnerable groups and individuals who experience social risks. An integrated social welfare database (Data Terpadu Kesejahteraan Sosial: DTKS) of nearly 29 million poor and vulnerable households24 has been put in place as a common 22 The program is managed by MoSA and aims to improve mechanism to identify potential beneficiaries for these centrally the quality of housing for the financed poverty-targeted social assistance programs. Overall social poor through repair and/or rehabilitation of uninhabitable assistance spending has seen increases in recent years largely due housing conditions with a priority on roofs, floors, walls, and toilet to increased spending on COVID-19 social assistance responses. facilities. It has a community- based approach, with work done Social assistance spending was 1.5 percent of GDP in September by groups of between five to 15 2021 (1.6 percent in 2020) (World Bank 2021), which is close to heads of households. It has been piloted in several areas including the global average of 1.54 percent.25 Indramayu Regency, West Java Province. The program had 958 beneficiaries in West Java Province in 2021, with nominal assistance provided of Rp 20 million per beneficiary. (LINK) 23 PENA was launched in December 2022. Previous iterations of this program include Prokus and Kube. PKH-CCT and Sembako food voucher program beneficiaries sign an agreement that they would withdraw from these programs once they receive PENA. https://kemensos.go.id/ diluncurkan-desember-2022-pena- telah-graduasi-1800-an-penerima- bantuan-sosial 24 DTKS coverage for 2020, hence some indicators may not reflect the latest situation of DTKS. 25 Global average from the Atlas of Social Protection Indicators of Resilience and Equity (ASPIRE) prepared by the World Bank. P. 2 7 FIG 3.1 EXPANSION OF PKH COVERAGE: NUMBER OF BENEFICIARY HOUSEHOLDS (2007-20) 10,000,2 32 10,000,0 00 9,841,270 6,228,810 5,981,528 3,511,088 2,871,827 2,326,533 1,454,65 5 774,293 1,052,20 1 387,947 620,848 726,376 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 Source: Administrative data Beyond core/regularly provided social assistance programs, from MoSA. Indonesia also provides a suite of emergency non-contributory transfers which are important social protection responses to disasters‒often those caused by climate shocks. These include various in-kind transfers provided by MoSA including emergency food assistance, evacuation equipment, household items and necessities (Bantuan Isi Huntara dan Huntap); a Living Support Assistance cash transfer (Jaminan Hidup: Jadup); a cash transfer to compensate heirs of disaster victims; and cash assistance for house renovation. The National Disaster Management Authority (Badan Nasional Penanggulangan Bencana: BNPB) also provides in- kind transfers for disaster recovery, including assistance for housing renovation, temporary shelter (Huntara), permanent housing (Huntap), and household items for persons in in temporary shelter. The Ministry of Public Works and Public Housing (MoPWPH) also provides a Housing Stimulus Assistance Program (Bantuan Stimulan Perumahan Swadaya) and assistance for house renovation which is jointly implemented with the respective local disaster management authority (Badan Penanggulangan Bencana Daerah: BPBD). These benefits have been routinely provided in past emergencies, including in response to the 2018 Central Sulawesi disaster and to numerous 26 Recent examples include the provision of compensation localized shocks.26 by MoSA to heirs of flood victims in Jakarta in 2020; and MoSA provision of in-kind supplies to The COVID-19 crisis has taken a severe toll on lives, livelihoods, flood victims in Sumatra in January 2022. and the Indonesian economy; and the government responded quickly to protect the poor, vulnerable, and newly poor through a 27 Data as of January 10, 2022. (LINK) suite of emergency social protection measures. By January 2022, over 4.26 million confirmed COVID-19 cases were recorded; 144,136 persons had died;27 1.8 million Indonesians became unemployed P. 2 8 between February 2020 and 2021; a further 3.2 million people exited the labor force; and 2.8 million people fell into poverty as of September 2020 (World Bank 2021a). To protect purchasing power amidst widespread income losses, the GoI launched an array of social assistance, jobs/skills, and social insurance measures. The benefit level of Program Sembako28 food assistance program was increased by 33 percent for nine months and its coverage expanded from 15.2 million to 20 million families while PKH families received double benefits for three months which was also extended to an additional 800,000 families. A new unconditional cash transfer program was introduced to cover approximately nine million additional households registered in the DTKS but who were not receiving PKH or Program Sembako benefits. Spending on social assistance expanded to 1.6 percent of GDP in 2020 and 1.5 percent of GDP in 2021, essentially doubling the pre-COVID level of spending on core social assistance programs. World Bank simulations estimate that these packages of social support have helped minimize the scale of the increase in poverty caused by the pandemic’s economic impact. The poverty rate could have risen to 11.4 percent but peaked at 10.6 percent under the new package and revised growth assumptions (World Bank 2021). Despite this strong effort, the poverty headcount was 10.1 percent in February 2021, compared to 9 percent in September 2019. While not a core focus of this report, Indonesia provides a range of livelihoods and employment programs to help improve jobs and income outcomes. These are implemented by various government agencies, including the Ministry of Finance (MoF), Ministry of Cooperatives, Small and Medium Enterprises, MoM, Ministry of Education, Culture, Research, and Technology (MoECRT), and MoSA. Beyond training programs, there are also a host of credit programs, 28 Program Sembako is Indonesia’s primary food including those that link training with credit provision, largely assistance safety net program, (previously BPNT). The program managed by MoF and the Ministry of Cooperatives, Small and Medium is managed by MoSA and aims to Enterprises. In addition, the government has introduced multiple boost food security and improve nutrition. Program Sembako’s emergency livelihood measures following the start of the COVID-19 benefits are provided via vouchers redeemable at distribution points pandemic for people suffering from expected income losses. Notably, called e-Warongs. The program covered 15.6 million households this includes a newly launched training program (Kartu PraKerja) in 2020. under the Coordinating Ministry for Economic Affairs (CMEA) which provides online training and stipends to recently unemployed workers. P. 2 9 Indonesia has also made significant progress towards rationalizing SJSN and expanding its coverage, particularly health insurance coverage through JKN. The increase of JKN coverage–from 130 million to over 220 million people in the five years to 2022–is a major achievement. The expansion of social insurance coverage for employment-related risks, however, has been much slower, partly due to the absence of contribution subsidies for informal sector workers. In mid-2020, the GoI introduced the much-needed job loss guarantee (Jaminan Kehilangan Pekerjaan) as a new scheme under SJSN. The scheme provides a cash transfer, access to labor market information, and access to training opportunities for formal sector employees during unemployment shocks. P. 3 0 Despite progress in reducing poverty, inequality, and vulnerability, there remain lingering gaps to inclusion and effectiveness of social protection in Indonesia. The development of the country’s social protection system has contributed to concrete achievements PKH SEMBAKO PIP in poverty PBI reduction in recent decades, with the country recording a sustained decline in poverty, from 19.1 FIG 3.2 COVERAGE OF MAIN SOCIAL NATIONAL PROTECTION PROGRAMS BY 50 percent of the population in 2000 to 9.2 DECILE (RURAL VS URBAN) 29 40 percent in 2019 (World Bank 2020). As B40 30 URBAN noted in the previous section of this report, 20 evaluations of the PKH program have found 10 positive impacts on welfare, consumption 0 of protein-rich food, and improvements in B30 RURAL healthy behaviors among beneficiaries (World Bank 2020). Although, Indonesia was able to reduce extreme poverty to the single digits prior B20 B10 to the COVID-19 crisis, the pace of poverty reduction has stalled since 2018. An estimated 25 million Indonesians lived below the poverty line prior to the crisis, and vulnerability was also high, with 20 percent of the population living above the poverty Source: World Bank calculations using Susenas (March 2021). line but below 1.5 times the poverty line. "Despite progress Inequality, as measured by the Gini coefficient, rose from its lowest value of 30 points in 2000 to 38 in September 2019. For social in reducing assistance, programs such as PKH and PIP have understandably poverty, prioritized targeting of households with children, but other vulnerable inequality, and groups such as the elderly and people with disability have not been vulnerability, adequately covered by the social assistance system. there remain lingering gaps to inclusion and effectiveness of social protection in Indonesia." 29 PBI here refers to the subsidized health insurance premium waiver under BPJS Health. P. 3 1 FIG 3.3 COVERAGE OF TARGET HOUSEHOLDS BY MAIN SOCIAL PROTECTION PROGRAMS BY REGION (2019) 30 (%) Rural Urban PKH SEMBAKO PIP BPJS HEALTH PKH SEMBAKO PIP BPJS HEALTH MALUKU-PAPUA MALUKU-PAPUA 80 80 60 60 NUSA TENGGARA KALIMANTAN NUSA TENGGARA KALIMANTAN 40 40 20 20 0 0 SULAWESI JAVA-BALI SULAWESI JAVA-BALI SUMATERA SUMATERA Note: Coverage is reported for the following groups: (i) PIP in the bottom 20 percent of households; (ii) PKH in the Source: Authors’ calculations using bottom 10 percent; (iii) Program Sembako in the bottom 30 percent; and (iv) the Social Security Implementation Susenas (2019). Agency (Badan Penyelenggara Jaminan Sosial: BPJS) Health in the bottom 40 percent. Furthermore, despite their “de-jure” design, the “de-facto” coverage of several social assistance programs differ in reality, with many eligible households with children not receiving all the 30 BPJS Health benefits they are eligible for. Overall, the country’s main social here refers to PBI - the subsidized health assistance programs reach about 61.39 percent of the poorest two insurance premium quintiles.31 Recent analyses have found that exclusion errors are high, waiver under BPJS Health. with none of the main social assistance programs covering more than 31 Susenas. 60 percent of the poorest 10 percent of the eligible population (See 2020. Includes PKH, Program Sembako, Figure 3.2), with regional variations (Figure 3.3) (Susenas 2020). A PIP, PBI, and local troubling development has been that coverage expansion has been social assistance programs. When accompanied by reduced beneficiary incidence among the poorest using international poverty lines, these 20 percent, reaching 39 percent in 2019 (Figure 3.4). programs cover 70.49 percent of the population Reforms are needed to improve the coverage, adequacy, and below US$1.90 and 65.66 percent below sustainability of social insurance schemes. The social insurance US$3.20 per day. system continues to face challenges with expanding coverage, 32 This is below particularly among informal sector workers. For protection from short- the developing country average term risks, only 28 percent of the working-age population is actively of ~ 30 percent of the working-age contributing, while only one in ten among the working-age population population covered contributes to the pension scheme.32 The total replacement rate for for retirement income protection. retirement savings is still below the International Labour Organization’s P. 3 2 international standard of 40 percent. Finally, as Indonesia starts experiencing demographic ageing, pension coverage will have to be addressed swiftly and will require a move towards retirement schemes that respond to the persistent challenges caused by high informality. The design of Indonesia’s social protection system has direct implications for the country’s ability to effectively contribute to climate change adaptation and mitigation efforts. The subsequent section will, therefore, summarize the findings of an Adaptive Social Protection (ASP) Stress Test of Indonesia’s social protection system to assess its adaptiveness to respond to climate shocks and to support climate resilience objectives. These findings will help inform strategic recommendations to improve the social protection system’s performance to better help poor and vulnerable households cope with, and adapt to, climate risk. " The design of Indonesia’s social protection system has direct implications for the country’s ability to effectively contribute to climate change adaptation and mitigation efforts." FIG 3.4 PKH BENEFICIARY INCIDENCE IN THE POOREST 20 PERCENT (2014-19) (CONDITIONAL CASH TRANSFERS) 53.3% 45.2% 40.6%39.0% 2014 2017 2018 2019 Source: World Bank, 2021 based on analysis of household survey data. 04 P. 3 3 P.32–40 Stress Testing Indonesia’s Social Protection System iven the issues identified in the previous G section, a Stress Test of Indonesia’s social protection system (World Bank 2021b) has been carried out to assess its adaptiveness to respond to climate shocks. The application of the Stress Test focused on an assessment of the system’s readiness to build resilience to shocks and to respond to heightened needs through a detailed "Indonesia assessment of four building blocks: Programs and Delivery Systems, has made Data and Information, Finance, and Institutional Arrangements and considerable Partnerships (Part 2 of the Stress Test). In addition, recent analysis progress in the (Ali and Setiawan 2022) which examined levels and sources of development vulnerability was leveraged to inform possible sources of risk and of its social required social protection needs to respond to households affected protection by covariate shocks in Indonesia (Part 1 of the Stress Test). system" Overall, the ASP Stress Test has found that Indonesia has made considerable progress in the development of its social protection system and leveraging programs and services to respond to the vulnerabilities caused by climate change and climate shocks. P. 3 4 Despite this, the Stress Test also found that lingering gaps to social protection effectiveness hamper the system’s ability to effectively build adaptive capacity of the poorest and most vulnerable to sufficiently prepare for, and respond to, the impacts of climate shocks and climate change. The results of the Stress Test are summarized in this report but detailed separately in a summary presentation and annotated Stress Test Report (World Bank 2022). The Adaptive Social Protection (ASP) Stress Test The ASP Stress Test is a new tool developed by the World Bank which aims to, inter-alia: (i) outline a risk profile of a country and connect it more deliberately to its social protection system; (ii) assess how existing national capacities could be scaled up before and after a shock; and (iii) identify gaps and guide investment priorities to build capacity for crisis management. Part 1 of the Stress Test examines the main sources of risk that are likely to require social protection scale-up, and provides an estimate of the number of people in need of support in the aftermath of different types and intensities of shocks; the degree to which they are covered by existing programs; and the extent to which the social protection system needs to increase support to existing beneficiaries (vertical expansion) or new beneficiaries (horizontal expansion). Part 2 of the Stress Test assesses the social protection system’s readiness to build resilience, and respond to, shocks through a detailed assessment of four building blocks: Programs and Delivery Systems, Data and Information, Finance, and Institutional Arrangements and Partnerships. Part 2 assigns a score from 1 to 5 for each building block subcomponent and produces an overall average score to identify strengths and priority areas for attention. The Stress Test combines both qualitative and quantitative assessment. While the tool, in-principle, examines social protection as a whole, there is a focus on noncontributory social assistance, namely cash transfers. P. 3 5 3.1 Approach & Caveats he application of the ASP Stress T Test for Indonesia was an intensively consultative process. First, the initial proposals for country scores were drafted through consultation with various World Bank Global Practices,33 given the inter- sectoral nature of the topic at hand. Secondly, the scores proposed for Part 2 of the Stress Test went through detailed consultation with various government agencies working on the related sectors. The proposed scores and associated references were shared via an annotated Stress Test with Bappenas (National Development Planning Agency, Directorate of Poverty Reduction and Social Welfare and Directorate of Spatial Planning and Disaster Management); MoSA (Family Social Security; Directorate of Social Protection for Natural Disaster Victims: PSKBA; and the Center for Social Welfare Data and Information: Pusdatin); BNPB; and the Fiscal Policy Agency: (BKF) at MoF.34 In addition, detailed consultative meetings to discuss the scores were held with the Directorate of Poverty Reduction and Social Welfare, Bappenas; and Directorate of Social Protection and Security (Perlindungan Jaminan Sosial: Linjamsos); and PSKBA. Written feedback was also provided by BNPB. These consultations formed the basis to finalize the Stress Test scores. The Stress Test leveraged a range of information sources including household survey and administrative data; regulations and laws; program manuals; and analytic reports. In addition to the 33 These included consultations, data sources for the Stress Test included household Urban, Disaster Risk Management survey data (namely Susenas); program-level administrative data; (DRM), Resilience and Land; Finance, regulations and legislation covering social protection, DRM, and Competitiveness and Innovation; Poverty Finance, particularly Disaster Risk Finance (DRF); program manuals and Equity; and and operational guidelines; and recent analytic studies and reports Social Sustainability and Inclusion. on ASP, social protection, DRM, and DRF in Indonesia. Given the 34 PSKBA: nature and scale of recent events, the social protection responses to Perlindungan Sosial Korban Bencana the 2018 Central Sulawesi earthquake and tsunami and COVID-19 Alam; Pusdatin: Pusat Data dan pandemic served as the main references for assessing ASP shock Informasi; BKF: response capacity, together with the documentation of social protection Badan Kebijakan Fiskal. responses to these shocks and assessment of their effectiveness. P. 3 6 The consultative nature of the Stress Test was particularly important for Part 2 of the Stress Test, as application of the tool is partially qualitative in nature. Part 2 of the Stress Test included a questionnaire with over 35 questions on each of the four ASP building blocks. In addition, two questions were added to Part 2 of the Stress Test for Indonesia to gain further insight on the linkages between social protection programs and building climate resilience among beneficiaries and to assess the shock-preparedness of social protection information systems.35 The Programs and Delivery Systems building block focused on Indonesia’s regular and disaster responsive social protection programs–primarily social assistance– which have been outlined previously in this report. The building block on Data and Information looked at the country’s various EWS and registries with a focus on the DTKS social registry. There are some important caveats to the findings for Part 2 of the Stress Test. A key caveat is that the overall aggregate score in Part 2 of the Stress Test is primarily indicative and is recommended to be used to identify areas of focus for strengthening the social protection system, rather than being used as a benchmark across countries. An important caveat to the scoring process is that scores are based on existing regulations, policies, programs, and systems. While the narrative acknowledges where there are plans and reforms underway, these do not impact scores unless they are already operational or enacted. A final caveat relates to the Building Block on Data and Information. The DTKS registry is currently undergoing reforms to its design which are expected to change its coverage, intake process, and a shift from household to individual registration. As such the findings for these related questions apply primarily to the DTKS’ design as of end 2020 and should be reapplied once the reform to the system is completed. Finally, several important policy 35 The added and program reforms were initiated following the application of the questions were as Stress Test. These are flagged where applicable in the subsequent follows: (i) Do social protection programs sections of this report. support improved resilience to climate risks among beneficiaries and their communities? and (ii) What mechanisms are in place for the protection of data and information in times of disasters? P. 3 7 3.2 Part 1 of the Stress Test Results: Possible Social Protection Needs Generated by Climate Risk in Indonesia36 or Part 1 of the Stress Test, the team drew on analysis done by Ali and F Setiawan (2022) which decomposed households’ vulnerability to poverty over a defined time period into two components. The first is `structural’ or poverty-induced vulnerability which occurs when the average expected welfare of a household falls below the poverty line in the absence of shocks. This situation could arise when, for example, physical assets and human capital endowments are too low to allow consumption above the poverty line. The second component is risk-induced vulnerability, which occurs when the average expected welfare level of a household falls above the poverty line, but during shocks consumption is expected to fall below the poverty line. Finally, the share of risk-induced vulnerability that is due to idiosyncratic (household-level) and covariate (community-level) shocks is estimated. It is important to note that, although risk-induced vulnerability due to covariate shock in the analysis includes climate shocks, the estimations do not distinguish between this and other types of covariate shocks such as food price shocks and epidemics. This analysis was then applied to assess the need for social protection scale-up to address climate risk in Indonesia, using the Multilevel Approach detailed by Bodewig et al. 2021. Specifically, for this approach, the Stress Test estimates the number of people that the social safety net needs to cover by calculating the total of poverty-induced vulnerability and covariate-risk*risk-induced vulnerability. These results are detailed at the end of this section of the report. 36 For more details on this analysis, see Shocks have been the dominant source of vulnerability to poverty Ali and Setiawan (2022). across Indonesia. Between 2011 and 2019 in urban areas, and P. 3 8 in rural areas to a lesser extent, vulnerability to poverty was mainly driven by high consumption volatility, while low expected mean of consumption played a small role (Ali and Setiawan 2022). More precisely, 6 percent of rural Indonesians had an expected per capita consumption in 2019 that was below the poverty line (that is, structural or poverty-induced vulnerable) and 38 percent were vulnerable because of high consumption volatility (that is, “risk- induced” vulnerable). In contrast, only 2 percent in urban areas faced “structural” or poverty-induced vulnerability and 20 percent faced risk-induced vulnerability. Risk-induced vulnerability was about six times higher than structural-induced vulnerability in rural areas, and the ratio was even higher in urban areas. The contribution of risk-induced vulnerability grew over time: at the national level, the share of risk-induced vulnerability rose from 80 percent of total vulnerability in 2011 to 90 percent in 2019. Nevertheless, in both urban and rural areas, on aggregate, both risk- and poverty-induced vulnerability fell between 2011 and 2019. Nationally, about one in ten Indonesians were vulnerable to falling into poverty due to covariate shocks in 2019. Some 15 percent of rural and 9 percent of the population in urban areas were vulnerable due to covariate shocks, while 42 percent of rural and 22 percent of urban Indonesians were vulnerable to idiosyncratic shocks in 2019. As expected, given the nature of covariate of shocks, vulnerability due to covariate shocks in most regions exhibited surges and reversals from year to year, especially in urban areas (Figure 4.1). Urban Sulawesi was a case in point–covariate vulnerability declined sharply between 2011 and 2012, increased until 2018 and then declined sharply to one of the lowest levels in the country in 2019. Urban and rural Nusa Tenggara stood out as they experienced an overall increase in covariate vulnerability during the period of observation (despite some fluctuation from year to year), while rural Kalimantan–a much richer part of the country–showed no progress at all. Using the data from this analysis and applying the methodology in Part 1 of the Stress Test, the estimations revealed that 12.16 percent of the rural population (11.4 million households) and 4.26 percent of the urban population (6.6 million households) were likely to need social protection support in the event of a shock. The findings have implications for the country’s capacity for P. 3 9 scale-up. Although the current social registry has covered more than 22 percent of the population, not a single social safety net program has covered more than 20 percent of the population. Some of these gaps are detailed later on in this report. FIG 4.1 SHARE OF POPULATION (PERCENT) VULNERABLE TO POVERTY (BY RURAL AND URBAN AREAS OF ISLAND REGION) (2011-19) 2011 2012 2014 2016 2018 2019 COVARIATE VULNERABILITY 0% 20% 40% 60% 80% RURAL TOTAL VULNERA- MALUKU-PAPUA BILITY: HIGH & DROPPING SLOWLY/STAG- NANT NUSA TENGGARA ...HIGH & JAVA-BALI DROPPING FAST SUMATERA SULAWESI ...LOW & KALIMANTAN DROPPING SLOWLY URBAN ...HIGH & DROPPING FAST NUSA TENGGARA JAVA-BALI SUMATERA ...LOW & MALUKU-PAPUA DROPPING SLOWLY ...LOW & KALIMANTAN DROPPING FAST SULAWESI Source: Authors’ calculations using Susenas 2011, 2012, 2014, 2018, and 2019; and PODES 2011, 2014, and 2018. P. 4 0 3.2 Part 2 of the Stress Test Results: Assessment of Indonesia’s Social Protection System’s Adaptability he Stress Test of Indonesia’s social protection system has found that T the country is operating at an emerging level for most metrics. On a scale ranging from 1 (Latent) to 5 (Advanced),37 the country scored 3.26 overall which corresponds to an Emerging level. Each building block 37 The specific scores and also averaged scores in the Emerging level, ranging from 3.08 and corresponding levels are (1) Latent; 3.47 (Table 4.1). When disaggregated further, a similar picture (2) Nascent; (3) Emerging; (4) emerged, with little variation across variables. The building block Advanced; and (5) Established. areas that performed comparatively better included Programs and Payment Systems, while Social Registries was the only category with 38 The assessment and score for Social a Nascent score (of 2.9).38 The results confirm previous findings Registries primarily focused on the that, while the country has made considerable advancements in DTKS social registry managed by the the development of its social protection system and its adaptability Ministry of Social to climate and other shocks, there remain several areas for Affairs. There have been subsequent enhancement of effectiveness, preparedness, and responsiveness. efforts to improve the coverage and The specific findings demonstrate areas of focus that the GoI may timeliness of data for social registry wish to prioritize to operate on a more established and advanced purposes, which are acknowledged later level. These findings are detailed in the subsequent sections of this in this report. report and in the recommendations presented in the conclusions. P. 4 1 TAB 4.1 SUMMARY SCORES (PART 2 OF THE STRESS TEST) Building Block Dimension Latent 1 Nascent 2 Emerging 3 Established 4 Advanced 5 Programs Programs 3.60 and delivery systems Delivery 3.14 (3.47) systems Payments 3.67 Data and Early warning 3.25 information systems (3.08) Social 2.9 registries Financing 3.25 Institutions Government 3.00 and leadership partnerships (3.25) Institutional 3.50 arrangements Overall: 3.26 BUILDING BLOCK Programs OVERALL SCORE & Delivery 1 Systems 3.47 1 Programs SCORE 3.6 ndonesia’s mix of regular social I assistance programs, disaster 39 Overall, responsive social assistance programs, Indonesia received an Advanced Score and livelihood programs is one of (4) on both social assistance and the strengths of the country’s social livelihood programs and an established protection system.39 These programs score (5) on overall were outlined earlier in this report and coverage of social protection programs have been successfully leveraged to (73.21 percent) (including social help poor households smooth consumption, provide support to assistance, labor markets, and those affected by climate shocks, and help build resilience of the social insurance). Nevertheless, close most vulnerable. In particular, the suite of disaster-responsive social to one-quarter of the assistance programs provides a useful complement to regular social population remains uncovered. assistance, which can help facilitate scale-up in post-shock periods. P. 4 2 These programs are supported by regulations that outline the roles and responsibilities of ministries such as MoSA in the provision of both regular and post-shock social assistance. For instance, PKH and Program Sembako fall under MoSA Regulation No. 1/2018, MoSA’s disaster-responsive social assistance programs are regulated under Regulation No. 4/2015, while PIP falls under MoEC Regulation No. 9/2018 (Lubis et al. 2021). In terms of overall program coverage, Indonesia performs well when assessed on the Stress Test‒with 73.21 percent covered by social protection programs.38 Despite this performance, there remain important gaps in beneficiary incidence and coverage gaps among vulnerable groups‒as detailed previously in this report. Furthermore, fragmentation and duplication are still challenges, with some programs having similar goals or target populations, but implemented by different ministries, as is the case for different cash-for-work programs and some disaster-response programs (Lubis et al. 2021). One area that could benefit from attention under the Programs "In terms metric relates to improving post-shock benefit adequacy. of overall Specifically, the Stress Test found that while the amount of the program social protection benefit provided during shocks covers a significant coverage, proportion of the consumption impact, there is room for improvement. Indonesia One important measure of this was to review the adequacy of performs regular program benefits to existing beneficiaries during shock well" times. For poor families with children who receive regular assistance through PKH and Program Sembako, for example, the temporary enhancements due to the pandemic, if received in full, are likely to cover a reasonable proportion of the consumption impact. This is because the combined value of these two main programs going to the poorest 15 percent was considered “adequate” in normal pre- COVID times, as it provides transfers sufficient to bring the median recipient household above the poverty line. The poorest 15 percent of households entitled to receive PKH 40 Susenas. should receive an average 21 percent of median consumption in 2020. This includes social assistance direct cash transfer (World Bank 2020). Adding on PIP, Program programs, BPJS Kesehatan, BPJS Sembako, and PBI-JKN to PKH would render a very adequate package Ketenagakerjaan, and severance pay. of protection for the poorest 15 percent of households with children, It is 73.55 percent at an average 49 percent of median consumption. For the same with company health insurance included. group, however, PIP and Program Sembako benefits each constitute P. 4 3 an average of 7 and 10 percent of median consumption respectively. Neither PIP nor Program Sembako or a joint payment from both programs would provide an adequate package of assistance (World Bank 2020). The population beyond the poorest 20 percent, therefore, receives a minimal package of protection and households without children are less likely to receive benefits due to eligibility rules of the main social assistance programs. Consideration could be given to improving the adequacy of post- shock social assistance benefits. For the 2018 Central Sulawesi disaster, findings from the WelTrAC household survey revealed that overall household welfare had not fully returned to its pre-disaster level more than one year after the disaster (Purnamasari et al. 2021). For COVID-19 social protection responses, a World Bank High Frequency (HiFy) survey found that 63 percent of PKH and Program Sembako beneficiaries reported that their needs were only partially met one year after the pandemic in March 2021, while 5 percent reported that their needs had not been met at all (World Bank 2021c). Furthermore, while 55 percent of pre-pandemic program beneficiaries felt that social assistance benefits and their own income were covering basic needs prior to the pandemic, only 40 percent of them reported the same level of adequacy in March 2021 (World Bank 2021c). Finally, households in the bottom 40 percent were more likely to report that their monthly basic needs were not covered. The question related to providing more direct linkages to help beneficiaries of social assistance programs improve their resilience to climate change impacts41 that was added for Indonesia revealed that there was limited documentation or systematic implementation of interventions to help social assistance beneficiaries build their adaptive capacity. The Stress Test score reflected that, to some extent, social protection programs are risk informed and credited programs that have direct activities that seek to improve resilience among beneficiaries or their communities. Notably, PKH Family Development Sessions (FDS) complement cash transfers by providing training and mentoring recipients on key topics and have helped foster improved resilience on a range of human capital metrics. 41 Indonesia received a score of Another social protection program with significant potential 2 (Nascent) on this metric. for climate resilience is cash-for-work (CfW). In 2018, the GoI P. 4 4 introduced the Village CfW (Padat Karya Tunai: PKT) policy to address unemployment and under-employment in rural areas. The policy mandates that villages allocate 30 percent of their Village Fund toward CfW interventions–50 percent of which must be used for wages (World Bank 2021d). Common agendas for CfW activities from Village Funds include clean water source development and drainage development (Effendi et al. 2020). The government is working to better integrate climate resilience into village development by establishing an integrated information management system (InfoDesa) that will draw on data collected by public agencies and integrate them into a single platform. Data to be integrated in this platform will include village poverty status, income, health, nutrition, education, infrastructure, exposure to disasters and climate-related hazards, community assets at risk, land and forest fire hotspots and related GHG emissions. Linkages to existing real-time databases (for example, for disaster early warning, weather forecasts) and indexes (climate vulnerability and food security) which are relevant for village-level planning will also be included (World Bank 2019a). Despite these developments, there are few systematized approaches evident in existing social assistance programs (including CfW) that 42 Indonesia include deliberate programs linkages to improve climate resilience received a score of 5 (Advanced) on among beneficiaries. This has, therefore, emerged as a priority area the percent of the poorest 40 percent of focus for improved ASP implementation in Indonesia. with a government- authorized/ recognized ID. The score for grievance redress and inclusion 2 Delivery Systems SCORE 3.14 of other vulnerable (elderly and people with disability) was 3 (Emerging); while the n delivery systems, Indonesia performs well on identification (ID) coverage O score for inclusion of women was 2 (Nascent). among the poorest and modalities to 43 Calculated enroll beneficiaries in times of shocks.42 from Susenas March 2020 based on According to Susenas 2020 data, 94.97 National ID number percent of the poorest 40 percent have a (Nomor Induk Kependudukan: NIK) government authorized/recognized ID.43 ownership. Those who already have a This should help limit challenges with NIK may not have a valid national ID, but identity verification and payment delivery in post-shock response. they are recorded in MoHA's Population There remains potential for improving ID among the poor and, among Administration its priorities, the Ministry of Home Affairs (MoHA) is focusing on Information System (SIAK). ensuring adequate ID coverage among poor and vulnerable families, P. 4 5 and in the more remote and isolated areas in Indonesia (World Bank 2020). Prior to COVID-19, enrollment in times of shocks was not automatic, with no reliance on DTKS and significant emphasis on self-enrollment which often includes lengthy bureaucratic processes before approval, enrollment and payment. In contrast, pandemic- related social protection support has relied primarily on expansions of regular programs and the use of other (multiple) mechanisms including self-enrollment (for example, support for small businesses).44 The use of automatic enrollment and other multiple mechanisms leveraged to enroll beneficiaries during the pandemic have informed an Established score (4) on this metric, but there remains room for improvement to ensure these mechanisms are appropriately leveraged for disasters‒ particularly those caused by natural hazards. For example, DTKS has not been used for enrollment for post-disaster benefits. The 2018 National Disaster Response Framework (NDRF) "only around recognizes the importance of both internal and public 25 percent communication to disaster response and includes public of the adult communications to reach the affected population. There is also a disaster-affected Guideline on a Disaster Response Media Center (BNPB Regulation population No. 8/2013) which explains the organization and procedures for the had any prior “Media Center” applicable to BNPB and BPBD when establishing an knowledge of emergency response command center at national and local levels how to evacuate to provide information to all stakeholders during emergencies. during a A Disaster Response Command System (BNPB Regulation No. disaster" 03/2016) includes provision of information on the distribution of non-food items. In practice, however, there is scope for enhancement in the implementation of these regulations and the effectiveness of these communications mechanisms is uncertain. For instance, the WelTrAC survey found that only around 25 percent of the adult disaster-affected population had any prior knowledge of how to evacuate during a disaster and this knowledge came mainly from 44 A notable self-learning using e-media. Only 15 percent reported receiving example is Kartu PraKerja which set information from local government (Purnamasari et al. 2021). The up an on-demand application process NDRF is a very high-level document and, as it is not operational in that worked well and quickly to enroll nature, BNPB plans to formulate Standard Operating Procedures applicants although, (SOPs) under the framework. given its online set- up, there were issues with access for those At the program level, PKH’s implementation guidelines outline with limited online literacy and access. that direct socialization should be carried out by human P. 4 6 resources, mass media, and online media, however, beneficiaries rely extensively on facilitators and, to some extent group leaders, for information (Microsave Consulting 2021). While some modifications to communication to address the COVID-19 social distancing requirements have been implemented, their effectiveness remains unclear. A recent survey found that, although 84 percent of PKH beneficiary respondents were aware about the new (monthly) PKH disbursement schedule briefly introduced during the pandemic, awareness about other PKH modifications was low‒with only 5 percent aware of the increased benefit amounts for certain categories of beneficiaries (Microsave Consulting 2021). On whether the delivery of post-disaster assistance is informed by a post-disaster household assessment (PDHA),45 Indonesia received a Nascent score (2). The process is fragmented, largely uncoordinated, links between the various agency needs assessment instruments and delivery of assistance is uncertain, and there is need for an integrated tool for needs assessment which is currently carried out by different government agencies. The PSKBA (MoSA) uses a Social Assistance Data Collection Form for Victims of Natural Disasters that collects data on household members, their NIK, and living conditions. MoSA Regulation No. 28/2012 regulates matters related to disaster preparedness cadets (Taruna Siaga Disaster/ Tagana). These are social volunteers or social welfare workers active in disaster management and who are also tasked with identifying disaster victims and recording material losses. On the DRM side, BNPB/ BPBD collect data during emergencies using a ‘Delivery of Disability 45 Post-Disaster Household Compensation Aid for Disaster Victims’ needs assessment form Assessments (PDHAs) are instruments which are deployed intended to provide complete and reliable data on the identification to collect information from and verification of cash and non-cash needs of disaster victims who households affected by shocks to assess their needs and experience disability (BNPB Regulation No. 15/2010). There are also determine what assistance will be provided. These instruments needs assessments carried out by nongovernment organizations differ from Post Disaster Needs Assessments which are high-level (NGOs). The fragmentation was found to be evident in the aftermath and not granular at the household of the 2018 Central Sulawesi disaster with needs assessments level. The management of PDHA processes in countries vary carried out by several agencies, including BPBD and NGOs, who from DRM and social protection agencies. Humanitarian agencies collected data to provide shelter, living support, and compensation to also have their own PDHA instruments in times of crisis. For heirs. The planned disaster victims’ database could help streamline more lessons on this instrument, these processes. It would be useful to ensure that the database’s see Williams 2020. development is accompanied by improving integration and data sharing of post-disaster assessment data to affected households. P. 4 7 While there are multiple ways to register complaints which can be used by beneficiaries and non-beneficiaries, implementation during emergencies is unclear, in part due to shock-induced disruptions.46 Mechanisms include MoSA’s Contact Center which includes multiple access methods and Ombudsman complaints which was leveraged during the COVID-19 pandemic. Nevertheless, during the pandemic, as few as one percent of PKH beneficiaries reported that they were aware of the PKH helpline number (Microsave Consulting 2021). BNPB also deploys complaints handling mechanisms in times of shock, however, following the Central Sulawesi disaster, the complaint handling system was found to be inadequate and was marked by slow responses, a lack of clear procedures, and complicated tiered complaint resolution processes (Lubis et al. 2021). Social protection shock responses could be reviewed to ensure "PKH and Program there are adequate and appropriate design features to ensure Sembako inclusion of women and other vulnerable populations such as elderly transfers are and people with disability.47 On design features to ensure inclusion prioritized of women, the country received a Nascent score (2). Although the GoI to women in has strengthened gender inclusion in social protection and DRM, more families which so for some local governments facing high disaster risk, there is need may foster for more explicit efforts to improve access and outreach to women in empowerment shock response programs. Gender mainstreaming is obligatory for all of women in government ministries and agencies through Presidential Instruction beneficiary No. 9/2000 on Gender Mainstreaming in National Development. BNPB households" has also enacted a Gender Mainstreaming in Disaster Management Regulation No. 13/2014 to ensure gender mainstreaming in disaster management. Several other regulations and policies on gender equality and social inclusion as part of disaster management have also been developed since 2000. An evaluation from the 2018 Central Sulawesi disaster, however, flagged challenges experienced by women, including limited access to information on facilities and 46 Indonesia’s score on this metric is 3. services available, and acts of gender-based violence (GBV) in and around the temporary housing (Global Facility for Disaster Reduction 47 The score for design features to ensure inclusion of women is 2 and Recovery – GFDRR 2019). At the program level, PKH and Program and the score for design features to ensure inclusion of other Sembako transfers are prioritized to women in families which may vulnerable is 3. foster empowerment of women in beneficiary households. Leveraging these programs for response to climate shocks, therefore, has an added benefit of improving gender outcomes in ASP responses. P. 4 8 In Indonesia there are limited platforms, regulations, policies, awareness materials, and implementation on DRM that accommodate people with disability. Government Regulation No. 42/2020 on Accessibility to Settlements, Public Services, and Disaster Protection for People with Disability (a recent implementing rule of Law No. 8/2016) is an umbrella regulation covering accessibility in DRM. BNPB Regulation No. 14/2014 also includes considerations for people with disability in DRM, but its framework is general. Furthermore, while the expansion of social assistance has been concentrated mostly on poor households with children, other groups‒such as the elderly and people with disability‒are not adequately covered (World Bank 2020). When program top-ups have been provided to beneficiary households with a member with disability, it is unclear if there is any agency given to the beneficiary with disability. Finally, a MoSA survey of 251 beneficiaries across five provinces found that about 30.5 percent of them reported that they do not have sufficient help nor ability to provide for the elderly and members with disability in their households (Ministry of Social Affairs 2020). 3 Payment Systems SCORE 3.67 ndonesia performed best on the I Payment Delivery metric for the Stress Test with an Established score (4) with electronic payment mechanisms 48 Following the application of the Stress Test, MoSA initiated largely used to transfer benefits to changes to the delivery mechanism beneficiaries–for both regular social for the BPNT/Sembako food assistance in 2022 switching protection benefits and post-disaster from a fully e-voucher distribution mechanism to a current mixed benefits. In 2017, the GoI initiated modality with partial cash distribution at post offices. The reforms to transform all social assistance payment modalities to a reform has not been confirmed as bank account-based electronic payment model to support a financial a permanent model and was later expanded to include PKH. The PKH inclusion agenda (Lubis et al. 2021). PKH, PIP, and Program Sembako benefit was also paid in cash at the end of 2022 and for the first now utilize “cashless” mechanisms.48 The Family Welfare Card (Kartu payment period covering January to March, 2023. The PKH March- Keluarga Sejahtera) payment instrument has electronic money and/ May 2023 payment was paid in a mixed approach at both Banks or savings features and is used by various programs, including PKH. and cash at post offices in some MoSA’s direct cash assistance for disaster victims (governed by MoSA locations. (LINK) Regulations No. 10/2020 and No. 4/2015) is also almost always provided through cash transfer to bank accounts as well. P. 4 9 Some emergency social protection programs use digital payment methods‒payments for the COVID-19 cash transfer program (Bantuan Sosial Tunai: BST) were largely made through the post office, while the BLT-Dana Desa cash transfer under the Village Fund is distributed in cash by village governments. Indonesia also received an Advanced score (4) for the speed with which the payment system can scale, largely due to progress with delivering rapid emergency social protection benefits in response to the COVID-19 pandemic and ability to deliver vertical expansion benefits to existing beneficiaries rapidly. One survey, for example, found that about 95 percent of respondents reported receiving PKH funds on time during October-December 2020 (Microsave Consulting 2021). Consideration could be given to facilitating broader choice among payment mechanisms and by simplifying administrative processes that delay post-disaster payments, particularly to non- beneficiaries. Although there is very good process on electronic payments, the absence of options in many instances limits beneficiary choice and may complicate post-shock payment delivery. Being able to select among payment delivery options helps implementers to adjust payment delivery in the event that the regular payment delivery systems are interrupted during a shock‒particularly during large-scale disasters that adversely affect payment infrastructure. P. 5 0 The WelTrAC survey for the 2018 Central Sulawesi disaster revealed that, while regular social assistance programs continued to work relatively well post-disaster albeit with some delays in the normal payment schedules, disaster-response assistance experienced protracted delays. Compensation for heirs and Jadup assistance were received on average between eight and 12 months after the disaster; while households that were eligible for house repair compensation waited 20 months on average (Purnamasari et al. 2021). This varies drastically with the time frames set out in program regulations‒for example, Jadup should be distributed a maximum of 90 days after the disaster (Lubis et al. 2021). Finally, capacity of the payment system to handle a horizontal expansion of the main program received a score of 3, given limitations for ensuring rapid expansion to non-beneficiary households. This is largely due to enrollment challenges (as bank account information and verification are also required), but also due to flow of funds required for horizontal expansion–some of which are discussed in the Financing and Institutional sections of this report. BUILDING BLOCK Data & OVERALL SCORE Information 2 3.08 4 Early Warning Systems SCORE 3.25 ndonesia performs well on the I availability of Early Warning Systems (EWS), which collectively are capable of monitoring one or more hazard. On availability of EWS and their multi- hazard capabilities, the country received an Established score (4). Notably, there are approximately 33 EWS platforms and systems, covering all relevant natural hazards managed mostly by government, while some are managed by universities. RAs recent analysis found that no single EWS covers all elements required for an effective EWS (Deltares, The Netherlands and Indonesia 2020), P. 5 1 development of a Multi Hazard Early Warning System (MHEWS) could be given priority. BNPB developed a Master Plan for MHEWS in 2019, which aims to support the government in realizing a people- centered, inclusive, end-to-end EWS that reaches all community groups. The Grand Design remains to be endorsed in the form of a Law/ President Regulation. When completed, it will serve as a guide for creating an integrated and efficient platform for early warning with relevant agencies for increased coordination and dissemination of disaster risk information (Deltares, The Netherlands and Indonesia 2020). The MHEWS’ design covers a broad range of climate and natural hazards but could be expanded to include early warning for biological hazards or pandemics.49 On whether government undertakes vulnerability and risk assessment to assess the impact of shocks based on EWS data, Indonesia is operating at an Emerging level. While government capacity to carry out vulnerability assessment is high, whether these are informed systematically by EWS remains uncertain. BNPB’s InAware EWS helps monitor natural hazards and assesses their potential threats to people and critical assets, while an InaRisk50 risk and vulnerability assessment assesses the scope of natural hazards, affected populations, infrastructure, and potential losses. Establishment and maintenance of the system received a score of Nascent (2) on whether there was an agreed trigger to initiate shock response or to scale up social protection systems in shock response based on EWS data. Shock response or scale-up primarily relies on real time disaster data, usually captured in rapid assessments. Finally, while there are protocols in place for scale-up, there is no agreed trigger to initiate a social protection shock response. 5 Social Registries SCORE 2.9 49 Indonesia’s ndonesia has several registry systems and I InaRisk disaster risk information platform databases with information on households does, however, and individuals. These include the coun- include an additional layer of COVID-19 try’s civil registry Population Administration information and advisories. Information System (Sistem Informasi 50 Supported by Administrasi Kependudukan: SIAK) – man- UNDP. aged by the Directorate of Population and Civil P. 5 2 Registration (Dukcapil), the Social Security Administration Agency (BPJS) registry, and the DTKS social registry, among others. These databases primarily fulfil sector- and program-specific functions but have also been recently used to provide social protection-related COVID-19 assistance. The scores in this section of the Stress Test are informed by the design of all these registries, but focus on DTKS, given that it is a social registry with the broadest coverage at the time of the assessment, and is used for targeting multiple social protection benefits and services. Despite this, it is important to note that the assessment of DTKS in this section is primarily based on the design and coverage of the system up to 2020, as MoSA is cur- rently undertaking reform of DTKS, the full details of which were not available at the time of carrying out this assessment. There are also additional developments in this space following completion of the Stress Test, with two databases currently being developed and/or adopted to target poverty alleviation programs. In 2022, the Coordinating Ministry for Human Development and Cultural Affairs (Kementerian Koordinator Bidang Pembangunan Manusia dan Kebudayaan: Kemenko PMK) adopted the database of the National Population and Family Planning Agency (Badan Kependudukan dan Keluarga Berencana Nasional: BKKBN), which was updated in 2021, and leveraged it as the data for the Acceleration of Extreme Poverty Alleviation (Pensasaran Percepatan Penghapusan Kemiskinan Ekstrem: P3KE) to target extreme poverty programs, in conjunction with DTKS.51 Additionally, Bappenas, funded by different resources, piloted and later scaled-up, a national socioeconomic registration (Registrasi Sosial Ekonomi: Regsosek) to support the government’s plan to build a population social registry including data on household social conditions and welfare.52 Nation- wide data collection was conducted in November 2022 by the Central Bureau of Statistics (Badan Pusat Statistik: BPS) to cover the total population. Indonesia performs strongly on coverage of the social registry 51 Based on relative to need, and registry coverage extends to most disaster- Presidential Instruction No. 4 year prone areas. The DTKS comprises 40 percent of households with 2020 (Inpres Nomor 4 Tahun 2022). the lowest socioeconomic conditions nationally, including in disaster- 52 (LINK) prone areas. The 2015 update was implemented in 34 provinces, 511 regencies/cities, 7,074 subdistricts and 82,190 villages across P. 5 3 Indonesia. When using the results for Part 1 of the Stress Test to estimate possible need, it appears that DTKS coverage, in principle, may fully cover the needs for rural and urban households likely to be affected by shocks.53 The estimations revealed that 12.16 percent of the rural population and 4.26 percent of the urban population were likely to be in need. When assessing the gap between these estimates and the rural and urban populations, DTKS was found to have no gap and fully covered those in need based on these estimates. Indonesia, therefore, received a score of 5 (Established) for rural and urban coverage relative to need. On coverage in disaster- prone areas, the country received a score of 4. While disaster prone areas may be broadly covered in the DTKS system, a large share of the population in these areas may not be included. At the household level, high exclusion errors may signal that the right households do not receive assistance despite being in DTKS and are, therefore, not in receipt of social protection benefits. Despite expanded coverage of PKH, the findings on reduced beneficiary incidence signal that this is a possible challenge. In fact, although the DTKS was, in principle, assessed to fully cover those in need based on the results from Part 1 of the Stress Test; when this need is compared to safety net coverage, gaps emerge. For instance, based on the estimations, Program Sembako falls short of covering everyone in need by 4 million households, while PKH appears to fall short by 8.2 million households. It is, however, important to note that while this population may not be the primary target group for these two programs, it is an indicator that the potential need for scaling-out through horizontal expansion in severe post-shock contexts is likely to be large. Finally, identification and targeting for 53 Based on estimates of the post-disaster social assistance benefits in the past have not relied on share of population and number of DTKS, and the Stress Test has found some limitations with the data people that the safety net needs to collected on their sufficiency for scaling-out during shock times. These be able to expand would need to be addressed if DTKS or another social registry system to as the total of the poverty-induced is utilized in the future for supporting post-shock horizontal expansion vulnerability and the covariate-risk*risk- and identification of beneficiaries for post-disaster social assistance induced vulnerability from the analysis benefits. On the DRM side, most of the disaster-prone areas are presented in Part 1. included in InaRisk according to BNPB, however, the completeness of (Stress Test Tool and Guide 2021). data for indicators in each disaster-prone area varies. P. 5 4 Areas for improved performance on social registry, include the recency of the data in the registry and whether it is sufficient for post-shock targeting. Due to DTKS’ previous update mechanisms, just over one-half (57.53 percent) of the records in the system are over three years old,54 resulting in a score of 2 (Nascent) on this metric. Beyond the impact this has on the registry being able to effectively perform its targeting functions for regular social protection programs, this has implications for the system’s ability to effectively identify and expand to affected households in post-shock contexts. A dynamic data updating process for local government to update information on existing registrants exists, but performance is mixed in terms of the completeness, timeliness, and quality of data provided. There are efforts to improve these standards and to incentivize local governments. Some of these update processes may be revised under ongoing DTKS reforms. Finally, DTKS also received a Nascent score (2) on whether its data allows targeting, identifying, contacting, and paying beneficiaries during shock response. While DTKS is used by several major social protection programs for beneficiary selection, it has not been systematically used to select beneficiaries for post-disaster benefits. For disaster-affected households already in DTKS, data collected includes name, address, ID number, family card number, and socioeconomic characteristics (Lubis et al. 2021). While the system includes only the bottom 40 percent population, it still has not been leveraged to facilitate rapid expansion to these households in affected areas. Furthermore, DTKS is not yet interoperable with other systems, such as InaRisk (Purnamasari et al. 2021). There does not appear to be data sharing with humanitarian partners. BNPB notes that while humanitarian agencies have access to government registry systems, they use their own beneficiary lists both during the emergency response and early recovery. This was also confirmed by the qualitative study on the social protection responses to the Central Sulawesi disaster which noted that several NGOs collected their own data on disaster victims (Lubis et al. 2021). In relation to data privacy, it is uncertain if DTKS has all protocols in place to ensure adequate disaster protection for the system and its data in times of shock. Indonesia received a Nascent score (2) on 54 Per Pusdatin 2021. data privacy, but there are efforts underway to improve performance P. 5 5 on this metric. Notably, data protection is regulated in different laws and regulations,55 and a draft Personal Data Protection Bill was subsequently ratified in October 2022. The Stress Test Assessment was only able to confirm that DTKS had one of the recommended five mechanisms56 in place to protect data and information in times of disasters, namely an off-site location in Surabaya. 55 For example: the Law No. 11/2008 on Electronic Information and Transactions, Law No. 23/2006 on Population Administration and the Ministry of Health Affairs Regulation No. 269/2008 and Law No. 29/ 2004 on Medical Practice which regulate health and medical records. Law No. 27/2022 on Personal Data Protection (PDP Law) establishes responsibilities for processing personal data and rights for individuals, including defining data and entities covered by the law as well as accountability measures. 56 These mechanisms include: (i) ensuring offline functionality for key business processes for the system such as intake and verification; (ii) establishing a Disaster Recovery and Business Continuity Plan; (iii) identifying a disaster recovery site independent of the principal system site which can be used for post-disaster operations; (iv) establishing back- up registry information off-site and virtually (for example, cloud storage) to mitigate against data losses in post-disaster environments; and (v) ensuring effective communication and training on post-disaster protocols ex ante to system users and staff. This question was added by the Indonesia team for the Stress Test assessment. P. 5 6 BUILDING BLOCK Financing OVERALL SCORE 3 3.25 he Stress Test rated Indonesia as Established (5) when it comes to T having a clear disaster risk financing strategy for a wide range of shocks with supporting legal and financial instruments in place, and that also mentions Adaptive Social Protection. A National Disaster Risk Financing and Insurance Strategy, launched in October 2018, aims to protect state finances and the population through sustainable and efficient risk financing mechanisms that meet disaster-related expenditures in a planned and timely manner, and that deliver well-targeted and transparent assistance following shocks (World Bank 2020b). The Strategy also includes the piloting of a public asset insurance scheme, developing a dedicated Pooling Fund for Disasters (Pooling Fund untuk Bencana: PFB) to manage budgetary allocations for disasters, and strengthen fiscal coordination and transfer mechanisms. Specifically, the Pooling Fund is expected to help manage disaster-related contingent liabilities. The PFB will collect, accumulate, and distribute special disaster funds and protect against budgetary pressures arising from disasters through proactive efforts which include fund accumulation and risk transfer through insurance (Ministry of Finance 2022). Indonesia also counts on several established mechanisms to finance disaster responses, including: (i) an On-Call Fund (Dana Siap Pakai) which provides rapid funding during a declared state of emergency; (ii) the Contingent Fund (Dana Darurat); and (iii) the Special Allocation P. 5 7 Fund (Dana Alokasi Khusus) which is often used in post-disaster situations for state budget allocation to finance recovery. Several laws and regulations also provide policy oversight and guidance on DRF and budget allocation.57 Indonesia has made progress in ensuring financing and disbursement mechanisms for social protection responses to climate shocks, with room for improved timeliness. Social assistance for disaster victims is subject to budgetary allocations for several technical ministries. This process has some challenges. For instance, while PSKBA at MoSA is tasked with preparing an annual budget projection for social assistance for disaster victims, these projected figures are often not enough to respond to actual needs and have been experiencing increases each year (Lubis et al. 2021). There are currently five different and overlapping budget procedures for disaster-related expenditures. The budget process is also denoted by slow disbursement and limited flexibility. There are notable developments that should help address outstanding issues‒importantly, the government has established the PFB to help ensure sufficient and well-planned funding for disaster-related expenditures, including health-related and natural disasters. The PFB will help protect the state budget against pressures caused by disaster impacts, through fund accumulation and risk transfer mechanisms. It is expected to improve the government’s risk layering and financial protection during future shocks. A noteworthy aspect of the pooling fund is that it will be linked to the social protection system to facilitate faster social assistance payments for disaster victims, through possible pre-arranged 57 These include: (i) Law No. 24/2007 that requires disbursement channels and linking them to financial solutions to the government to allocate an adequate fund for disaster help funds reach the right beneficiaries. The ASP Roadmap is also management in the State Budget, as well as allocate the On-Call expected to address ASP financing challenges, particularly related Fund in the budget; (ii) Government to improving mechanisms for rapid financing for social protection Regulation No. 21/2008 that regulates budget allocations shock response scale-up. On these metrics, the Stress Test found and reallocations to finance disaster related expenditures; that Indonesia performs at an Emerging level. (iii) Regulation No. 22/2008 on Disaster Financing and Aid Management; and (iv) Law No. The government’s ability to analyze and model the potential 33/2004 on Fiscal Balance Between the State and Subnational cost implications of different shocks over time was found to be Governments. operating at a Nascent level (2). Prior assessments have noted that there has been progress, however, there is currently no centralized planning for disaster-related spending and government primarily P. 5 8 relies on the national budget and international assistance to cover disaster losses (World Bank 2020b). In addition, consolidated information on disaster spending, and information on spending on disasters, is spread across the 542 separate budgetary systems with split oversight by MoF and MoHA. BUILDING BLOCK Institutions OVERALL SCORE & Partnerships 4 3.25 6 Government Leadership SCORE 3.00 ndonesia performs at an Emerging level on ensuring government strategy, I contingency planning, and effective leadership of social protection responses. A strong point of the system is that updated strategies and policies exist, with some recognition of the role of ASP in DRM (and vice versa). Notably, laws governing social protection and DRM in Indonesia clearly articulate the right to ensuring basic needs are met and to guaranteeing social protection to the population and disaster victims. 58 There is: (i) direct conditional cash assistance for the poor For example, Article 26, Section 1, Point A of Disaster Management affected by disaster; (ii) disaster aid distributed for schools affected Law No. 24 /2007 states that “everyone has the right to social by natural/social disasters; (iii) support for social recovery of protection and security, especially for disaster-prone community disaster-affected economies; (iv) groups.” Social protection laws also mention the rights of those an increase in mitigation capacity and community preparedness affected by shocks, with Article 14, Section 1 of Law No. 11/2009 in the disaster-prone area and for victims/communities of stating that "social protection is intended to prevent and reduce the disasters; (v) social protection for communities in disaster-prone risk of shocks and social vulnerability of a person, family, group, areas and disaster-affected and/or community so that their basic needs can be fulfilled." ASP is communities; and (vi) Increased economic capacity and livelihoods stated as a priority focus and action area under the National Disaster for affected communities (Indonesia National Disaster Management Plan 2020-2024, which also includes indicators to Management Plan 2020-2024, pp 170-171). track progress.58 The National Disaster Risk Finance and Insurance Strategy 2018 also recognises the role of ASP in DRM, particularly related to the pooling fund. P. 5 9 One area for further consideration is improving the alignment of Social Protection and DRM laws. On contingency planning and links to risk assessment, contingency arrangements are to some extent covered in policy and frameworks, however, they are not adequate. There is no single contingency plan or response plan, but several including the 2018 NDRF. All cities and regencies are expected to have updated risk assessment and contingency planning for each relevant hazard at the district level and hazard-specific contingency plans in high-priority districts, however, according to BNPB, more than 50 percent of districts do not have a contingency plan. Finally, improved contingency arrangements for PKH could be achieved with the completion and operationalization of emergency response guidelines that are planned to be developed. While there is clear guidance on inter-agency post-shock coordination, previous assessments have noted challenges with on-the-ground coordination arrangements (Lubis et al. 2021). "MoSA has a 7 Institutional Arrangements SCORE 3.5 clear role in the National Disaster nstitutional arrangements received Management the second highest score on the I Plan 2020-2024 Stress Test for Indonesia. This is a to plan for credit to the country’s clear assignment the needs of responsibilities and roles for the of displaced main shock types and fair performance persons and on coordination mechanisms linking volunteers" DRM and social protection responses to climate and other shocks. Notably, BNPB has a clear leadership role to oversee and coordinate disaster management activities at all stages of the DRM cycle and this is supported in law. BNPB provides guidance on disaster management, ensures communications with stakeholders, and coordinates disaster management activities. In addition, BPBDs have been established in all 34 provinces and most of Indonesia’s 514 districts and cities. MoSA has a clear role in the National Disaster Management Plan 2020-2024 to plan for the needs of displaced persons and volunteers, including preparing community social protection instruments for those affected by disasters that also facilitate P. 6 0 community social resilience. A Disaster Management Steering Committee chaired by BNPB comprises 19 members including ten Echelon I government officials or equivalent, including MoSA. Indonesia also has seven National Disaster Response Task Forces (DRTFs), one of which is a Displacement and Protection DRTF, coordinated by MoSA and whose key function is to coordinate the delivery of emergency food, temporary shelter, and emergency assistance to families.59 In the area of coordination arrangements, further consideration could be given to the overlap of roles within the local Social Affairs Offices (Dinas Sosial: Dinsos) and BPBD‒including in the provision of basic assistance and shelter. For instance, program rules specify that house renovation assistance should be provided by two institutions: MoSA and BPBD (in cooperation with MoPWPH). A further area of misalignment is that the established benefit amounts by MoSA and BPBD differ according to their program regulations. Various agencies also implement their own assessment of disaster 59 National Disaster Risk victims with limited coordination, as was the case for the Central Framework. 2018. Sulawesi disaster. 05 P. 6 1 P.60–40 Conclusions & Recommended Priorities for Government Action Concluding Observations: his paper has provided an outline of how climate T change, human development, and poverty interact and are mutually reinforcing. Evidence from numerous studies has demonstrated the debilitating impacts that climate risk has on poverty and human capital accumulation. These include asset losses, loss of livelihoods and income sources, consumption impacts, undernourishment, stunting, strain on non-food expenditures including on health and education, risks of infectious disease, and vector-borne infections, to name a few. Conversely, households less vulnerable to poverty and with higher human capital can better withstand the impacts of climate shocks and climate risks than poor households and those with less human capital. The evidence reinforces the importance of placing people at the center of climate policy and ensuring that integrated climate change adaptation and mitigation efforts are deliberately combined with climate-sensitive poverty reduction and human capital development policies. P. 6 2 Social protection is an important interlocutor for addressing climate risk, poverty and improving human capital and, in recent years, Indonesia has made considerable progress in the development of its social protection system and leveraging its social protection programs and services to reduce poverty, improve human capital, and respond to the vulnerabilities caused by climate risk. Notable achievements include: (i) expansion of PKH to cover 10 million households; (ii) establishment of a social registry (DTKS) for the bottom 40 percent by welfare that covers 29 million households; (iii) provision of a suite of core, poverty- targeted social assistance programs; (iv) a suite of post-shock social assistance benefits; and (v) introduction of interventions to improve socioeconomic empowerment and economic inclusion among social assistance beneficiaries. Evaluations of Indonesia’s CCT have revealed important human capital outcomes among beneficiaries in health and education, and recent analysis has shown that these positive behaviors are sustained by those who graduate from the program after their exit (Syamsulhakim and Khadijah 2021). The program has also been found to have a positive effect on deforestation in villages where households participate in the program (Ferraro and Simorangkir 2020). This signals strong potential for the country’s social protection system to better support climate change adaptation and mitigation efforts‒particularly among the poorest and most vulnerable households. Addressing lingering gaps to effectiveness will enhance the social "To build on its protection system’s ability to effectively build adaptive capacity achievements, of the poorest and most vulnerable, and to sufficiently prepare Indonesia’s for, and respond to, the impacts of climate shocks. To build on its social protection achievements, Indonesia’s social protection system could expand system could to cover a larger share of those in need. Recent analyses have expand to cover found that exclusion errors are high, with none of the main social a larger share of assistance programs covering more than 60 percent of the poorest those in need" 10 percent of the eligible population and, while PKH coverage has expanded significantly in recent years, it has been accompanied by reduced beneficiary incidence among the poorest 20 percent. Many poor households without children and other vulnerable categories of beneficiaries are not eligible for several of the core poverty-targeted social assistance benefits. Results from P. 6 3 Part 1 of the Stress Test illustrate that, based on current coverage of PKH and Program Sembako, there will be significant need for horizontal expansion to include everyone in need due to risk-induced vulnerability caused by covariate shocks. There are also issues with coverage of near-poor and informal sector workers who are at risk of falling into poverty due to climate shocks and longer-term climate change impacts. This implies that the need for scaling-out through horizontal expansion in the post-shock response is large, as many in need do not receive the main poverty-targeted social assistance household benefits and are also not be covered by contributory social protection programs. Beyond coverage, gaps with linkages to, and scale of, economic inclusion shows there is room to improve efforts to strengthen resilience of social assistance beneficiaries. The application of the ASP Stress Test for Indonesia has "There are also recognized that progress and highlighted the gaps to ensuring issues with performance at an established level. It has found that Indonesia coverage of is operating at an EMERGING level for most metrics. On a scale near-poor and ranging from 1 (Latent) to 5 (Advanced), the country scored 3.26 informal sector overall which corresponds to an EMERGING level. Each building block workers who are also averaged scores in the Emerging level, ranging from 3.08 to at risk of falling 3.47 (see Table 4.1). When disaggregated further, a similar picture into poverty due emerged, with little variation across variables. The building block to climate shocks areas that performed comparatively better included PROGRAMS and and longer-term PAYMENT SYSTEMS , while SOCIAL REGISTRIES was the only category climate change with a Nascent score of 2.9. The results confirm previous findings impacts" that the country has made considerable advancements in the development of its social protection system, and its adaptability to climate and other shocks, with several areas for ensuring improved effectiveness, preparedness, and responsiveness. It is important to acknowledge that while the scores present a static view of the country’s performance on social protection system adaptability, there are considerable reforms underway, particularly at the policy level, which are not yet fully operational, but will likely facilitate improved performance on many of these metrics. A major development in this regard is the creation of a PFB that will improve performance under the Finance building block through direct linkages to support ASP objectives, thereby helping to protect the state budget against pressure due to disasters and P. 6 4 improving the government’s risk layering and financial protection in the event of future shocks. Additionally, following the application of the Stress Test there have been efforts to improve the coverage and timeliness of data for social registry purposes, including national socioeconomic registration (Regsosek) to support the government’s plan to build a population-wide social registry; and leveraging BKKBN and DTKS data to target extreme poverty programs for the Acceleration of Extreme Poverty Alleviation (P3KE). On the social protection policy front, Bappenas is finalizing an ASP Roadmap that and its associated regulations which will recommend integration of social protection and climate change action and provide a guide for leveraging social protection to address risks from natural and climate-related hazards, particularly for poor and vulnerable populations. 3.1 Recommendations to Support Improved Climate Resilience his report and the specific findings of T the ASP Stress Test have identified areas of focus that the GoI may wish to prioritize to enhance the performance of its social protection system and, by extension, human capital for improved climate resilience outcomes. This section outlines possible areas of focus that have been prioritized based on the scores emerging from the ASP Stress Test and major gaps to social protection effectiveness. The report also proposes recommendations for education and health, recognizing the importance of deliberately pairing targeted climate resilience measures with interventions to build human capital. 1 Close remaining social protection coverage gaps to ensure poor and vulnerable people are adequately protected by social protection benefits and services for improved climate resilience and adaptive capacity, including expansion of social insurance coverage to mitigate the impact of idiosyncratic shocks such as job loss or health shocks, and closing social assistance coverage P. 6 5 gaps where they persist. Rationale: Gaps in social assistance and social insurance program coverage illustrate that the burden on the social protection system for horizontal expansion during large covariate shocks is likely to be high. These coverage gaps mainly apply to existing poor who currently do not receive core social assistance benefits. Furthermore, to the extent that those facing risk-induced vulnerability to covariate shocks can be supported by benefits and services for which they are eligible, this will help build their human capital and resilience to future shocks. Part 2 of the Stress Test also found that while post-shock payments can be made with little delay for those already enrolled in social protection programs; enrollment and payment of new beneficiaries in the post-shock response face significant delays and the process is complicated by fractured mechanisms to assess and enroll these beneficiaries. The GoI efforts to increase coverage of SJSN employment programs are a critical element of broader risk management, and thus also can play an important role in the context of climate change. 2 Improve direct activities in social protection programs and linkages to other sector programs to build adaptive capacity of beneficiary households. Rationale: A major finding of this report is that there are few systematized approaches evident in existing social assistance programs that include deliberate interventions or complementary program linkages to improve climate resilience among beneficiaries. In addition, although there are resilience-building interventions offered by some core programs–particularly PKH through FDS sessions, and economic inclusion for social assistance beneficiaries through PENA and other programs–their focus has not yet included climate adaptation and resilience as priorities. A Scale up education and information on climate change and shock preparedness to social protection beneficiaries. Completing and rolling out an FDS session on disaster preparedness for social assistance beneficiaries would be a ‘quick win’ in this regard, particularly if it is complemented with additional messaging on climate adaptation geared towards the poor. This should also be complemented with broader communications efforts to other social assistance beneficiaries and poor households. In this regard, a family- based training module on integrating disaster-risk reduction, P. 6 6 child protection, and family-based social protection programs, developed by PSKBA, BNPB, and Save the Children will also have an important role to play. B Ensure programs (especially CfW and housing-related social assistance benefits) have direct adaptation and resilience- informed design. To the extent that the government is already investing in housing-related social assistance, it would be important to assess whether these interventions (particularly for permanent housing) include appropriate disaster-resilient design, clean energy, and sustainable utility usage. This would, of course, require multi-sector engagement in program design and supervision. CfW is perhaps the social protection program type most used to support climate resilience and disaster response in several countries. Leveraging CfW more deliberately in support of these objectives in Indonesia has immense potential for improving both household and community resilience. This could be facilitated through deliberately prioritizing mitigation subprojects at the community level (for example, mangrove restoration and forest replanting), complementing housing benefits with clean energy solutions; and deploying CfW more readily to disaster-affected households to support recovery. C Improve links and access to complementary benefits and assistance offered by other ministries and agencies for which social assistance beneficiaries could be prioritized. Although climate policy is not a primary focus of social protection, the sector can play an important intermediary role in linking poor households to benefits, services, information, and training aimed at improving climate resilience‒similar to how CCTs have successfully helped the education and health sectors achieve improved outcomes among the poorest. Some potential linkages include improving those to training, credit, and grants that encourage livelihood diversification, sustainable livelihoods, and green jobs; inputs for climate-smart agriculture; and climate-resilient housing support offered by the housing sector. In addition, as government seeks to reduce GHG emissions, social assistance beneficiaries whose livelihoods rely on these sectors could be specifically targeted for compensatory benefits and support to use alternative energy sources. 3 P. 6 7 Improve and pre-position social registry systems for shock response and to support climate policy. It would be important for Indonesia to: (i) improve the dynamism and quality of regular data updating of systems to support targeting (including DTKS), and ensuring such principles are foundational in any future registry systems; (ii) expand social registry coverage of the population in both poor and disaster- prone areas (such as was done through nationwide socioeconomic registration (i.e. Regsosek)); (iii) ensure that the data collected is useful for responding in case of a shock; (iv) facilitate improved access to social registry data by humanitarian agencies in shock times; and (v) improve data privacy of social protection systems. In addition, it would be important for Government to complete a planned disaster victim’s database and streamline processes for integration and data sharing of post-disaster assessment data to affected households. Optimally, all these actions could be linked to the future development of an Integrated Social Protection Information System, which could leverage shared digital public infrastructure to create end-to-end, digitalized processes supported by data exchange across existing government administrative databases. Such a model could facilitate dynamic data updates for eligibility determination; integrated view of benefit and services delivery for monitoring purposes; and more on-demand access to the population for social protection benefits and services. While it is unclear if adequate disaster protection mechanisms are in place, these should be ensured for all registry systems. The measures include: (i) ensuring offline functionality for the system’s business processes; (ii) establishing a disaster recovery and business continuity plan; (iii) identifying a disaster recovery site independent of the principal system site that can be used for post- disaster operations; (iv) establishing a back-up registry information off-site and virtually (for example, cloud storage) to mitigate against data losses in post-disaster environments; and (v) ensuring effective communication and training on post-disaster protocols ex ante to system users and staff. Finally, as Indonesia pursues climate policies to facilitate reduced carbon emissions, which could include subsidy reforms, social registries could have a critical role to play in helping to identify potentially affected households whose livelihoods and incomes will need to be protected during the transition. Rationale: Indonesia’s Stress Test scores on social registry require a dedicated focus to improve the use of these systems for climate resilience P. 6 8 goals. Despite covering the bottom 40 percent of the population, DTKS has not systematically been used to identify and target households in past climate shocks. Shortcomings to data updates may be addressed by ongoing DTKS reform efforts but require deeper assessment to know if they include adequate information for targeting in shock times. 4 Improve mechanisms for faster horizontal expansion and delivery of post-disaster social assistance benefits. Rationale: A key finding of the Stress Test was the delays experienced in identifying, enrolling, and delivering benefits to previously unenrolled households in post-shock times, particularly those caused by large-scale natural hazards. Even delivery of post-disaster social assistance benefits took much longer than the maximum timeframes set out in their regulations. In addition, the scale of horizontal expansion need is likely to be high given coverage gaps and data shortcomings in existing registry systems. It would, therefore, be imperative for the government to consider possible options for improving timeliness in this regard. Some options include: A Expand social registry coverage to a larger share of the population (see related recommendation above); B Leverage technology more effectively for identification of non-beneficiary households in affected areas; C Establish an integrated PDHA process that is deployed rapidly, linked to social protection and other relevant information systems, and whose data is shared across agencies to identify eligible households for post-shock benefits; D Simplify business processes related to determining eligibility for support; and E Include horizontal expansion in ex ante estimations of need and link the initiation of business processes for horizontal expansion to established triggers (see related recommendation below). P. 6 9 5 Improve gender-sensitivity and attention to vulnerable groups in post-shock social protection operations. Possible measures include: (i) improve assessment and information sharing on affected households with vulnerable categories; (ii) ensure direct messaging to women in affected households; (iii) explore the feasibility of providing post-disaster benefits directly to women, people with disability and the elderly to improve their agency and/or ensuring measures to monitor and respond to their needs; and (iv) ensure that temporary shelter support includes more effective measures to prevent GBV and ensure accessibility. Rationale: A main finding of the Stress Test was that social protection shock responses were not deliberately designed to ensure gender sensitivity and inclusion of particular vulnerable groups, such as people with disability and the elderly. It would, therefore, be important to address this, particularly in post-disaster social assistance benefits. 6 P. 7 0 Develop an integrated tool to quantify post-shock needs and estimate optimal post-shock benefit levels and explore opportunities to link EWS data more systematically with social protection scale-up planning. This would optimally be complemented with establishing triggers for social protection scale- up linked to EWS that could facilitate automatic scale-up depending on established metrics. These could be importantly informative to the PFB. Such systems will help make Indonesia’s social protection system more prepared and better able to rapidly respond to the poorest in times of shocks. Rationale: A key finding of the Stress Test was that post-disaster financial planning for social protection was largely based on a retrospective view of previous year’s costs. Better ex ante quantification of potential post-shock social protection needs could help improve this process. 7 Complete ongoing and planned ASP-related reforms for improved performance. Rationale: This report has flagged numerous important reforms currently planned or underway that will likely improve adaptability. A ‘quick win’ on this agenda would be for these reforms to be completed and operationalized as soon as possible, with effective measures in place for routine monitoring and adjustments as needed. These include the PFB and ASP Roadmap. 8 Strengthen operational processes to improve communications on disaster preparedness to poor and vulnerable households. These efforts should also include messaging to help ensure thorough understanding of post-shock assessment and enrollment processes, and adaptations to program design in emergencies. These communications efforts should also extend to informing beneficiaries and the public on grievance redress in post-shock response. 9 Continue to build on progress with electronic payment delivery and facilitate broader choice among payment mechanisms, particularly for the post-shock response. Expanded choice would ensure more beneficiary-responsive payment modalities and provide beneficiaries with options that are more relevant to their needs, particular in post-shock contexts. Additionally, it would be important for Government to address the gaps that result in payment delays during shock times for both regular social protection benefits and emergency transfers. Related ongoing efforts include the plan to develop a Central Mapper for G2P payments, essentially a repository P. 7 1 of unique individuals linked to a particular payment information (such as bank account) for the purpose of routing payment transactions, would improve monitoring, accountability, and speed of payments. This would optimally be supported by and linked to an integrated beneficiary database that facilitates onboarding and monitoring for social programs. Rationale: While Indonesia has done well to rely on electronic and banking payment mechanisms for most regular and post-shock social protection benefits, these often feature reliance on a single payment mechanism with limited flexibility for adjustment. This has implications for post-disaster environments, particularly since payment delivery systems are often interrupted during shocks. Enabling broader choice and back-up payment delivery arrangements, with appropriate fiduciary controls, will help government be better able to avoid payment delays in large-scale climate shocks. 10 P. 7 2 Reduce lingering misalignment and overlap of roles between Social Protection and DRM agencies, both in program regulation, and operationally on-the-ground. Rationale: While Indonesia performs well on several institutional metrics in the Stress Test, this is complicated by lingering misalignment and coordination challenges between Social Protection and DRM agencies. Notably, the Stress Test found an overlap of roles between local Dinsos and BPBD‒including in the provision of basic assistance and shelter‒ and misalignment in the established benefit amounts provided by MoSA and BPBD according to their program regulations. 11 Although the focus of this paper is social protection, it is equally important to ensure continued investments in education and health to support climate resilience objectives for a comprehensive and integrated human capital approach to addressing climate risk. As noted previously, climate change, human development, and poverty interact and are mutually reinforcing and require deliberate and integrated strategies. Although the recommendations in this paper have focused on social protection, it would be equally important for the GoI to invest in deliberate education and health investments to mitigate potential sectoral impacts and to build human capital for improved resilience to climate change impacts. These could include: A Protect social infrastructure in health and education to ensure business continuity in the face of shocks; B Continue to give priority to stunting reduction given the possible climate change impacts on stunting; and C Facilitate improved education completion outcomes, particularly length of schooling, given previous study findings that those with higher education attainment are better able to withstand the impacts of shocks. P. 7 3 Appendix Programs Included in the Adaptive Social Protection Stress Test59 Program Name Program Type/Brief Description Responsible Ministry/ Agency Regular / Foundational SA Programs Asistensi Sosial Lanjut Usia A S LU T : Cash transfer to poor elderly, 60 MoSA Terlantar (Social Assistance for the years of age and above, with a Elderly) physical condition that makes them reliant on other people, have no other source of income, and are not PKH recipients. A SPDB : Asistensi Sosial Penyandang Cash transfer to people 2 to 59 MoSA Disabilitas Berat (Assistance for Persons years of age who are unable to with Severe Disability) fulfil personal needs, have no other source of income, cannot be fully rehabilitated, and require assistance to perform activities. Bantuan BPNT/ PROGR A M SEMBA KO : Food voucher social assistance for MoSA Pangan Non Tunai/Sembako (Non-Cash poor families in the poorest 25% of Food Assistance) households included in the DTKS social registry. PBI-JKN :Penerima Bantuan Iuran Subsidized health insurance Ministry of - Jaminan Kesehatan Nasional premiums for poor and vulnerable Health, BPJS (Contribution Assistance for National households, and free use of all Kesehatan Health Insurance) available health care services and facilities. Program Indonesia Pintar (Smart PI P : Cash transfer to enrolled students MoECRT Indonesia Card) or school-age children from the Ministry of poorest 25% of households. Religious Affairs Program Keluarga Harapan (Family PKH : Flagship conditional cash transfer to MoSA Hope Program) poor households in the poorest 20% of households included in the DTKS social registry. PK T : Padat Karya Tunai (Cash-for-Work) Cash-for-work social assistance – MoV operated by various ministries. MoM MoPWPH PROKU S : Program Kewirausahaan Sosial Grants and mentoring to PKH MoSA (Social Entrepreneurship Program)60 beneficiaries and other poor and vulnerable individuals. Rehabilitasi Sosal Rumah RS -RU T IL A HU : Cash transfer to poor households MoSA Tidak Layak Huni (Social Rehabilitation of that aims to improve the quality of Uninhabitable Houses) their housing through repair and/or rehabilitation. 60 The Stress Test primarily focused on social assistance programs. Social Insurance and labor market programs were also considered, although not assessed in detail and are, therefore, not included in this summary list. P. 74 Program Name Program Type/Brief Description Responsible Ministry/ Agency Emergency Social Assistance Programs Emergency Assistance (Bantuan Isi Emergency in-kind assistance, MoSA Huntara dan Huntap) including evacuation equipment, household items, and necessities. Compensation for heirs of disaster Emergency cash transfer. MoSA victims Housing Stimulus Assistance Program Cash assistance, often MoPWPH (Bantuan Stimulan Perumahan Swadaya) complemented with cash-for-work, to help improve the quality of uninhabitable houses Huntara: Assistance for temporary In-kind assistance for housing BNPB shelter renovation and temporary shelter. Huntap: Assistance for permanent In-kind assistance for permanent BNPB housing housing. Jadup: Jaminan Hidup (Living Support Emergency cash transfer. MoSA Assistance) COVID-19 Social Protection Response Programs BST: Bantuan Sosial Tunai (Cash Social Unconditional cash transfer for MoSA Assistance) households listed in DTKS, but who were not receiving any existing benefits. Bantuan Langsung BLT-DA NA DE SA : Unconditional cash transfer MoV (Village Tunai - Dana Desa (Village Fund Direct targeting rural households affected Fund/Dana Cash Assistance) by COVID-19 who were not covered Desa) by Program Sembako, PKH, BST, and PraKerja programs at the time. K A R T U PR A KER JA : Pre-Employment Card Cash transfer plus training targeting CMEA jobseekers, 18 years of age or older, and not in formal education nor receiving PKH, Program Sembako, or BST at the time. P. 7 5 References Alderman, H., J. Hoddinott, and B.H. Kinsey. 2006. “Long Term Consequences of Early Childhood Malnutrition.” Oxford Economic Papers 58(3): 450-474.. Ali, R., and I. Setiawan. 2022. “Looking Beyond Aggregate Poverty and Vulnerability Rates in Indonesia: Aligning Social Protection with Sources of Vulnerability to Poverty.” The World Bank. Andrews, C., A. de Montesquiou, I.A. Sánchez, P.V. Dutta, B.V. Paul, S. Samaranayake, J. Heisey, T. Clay, and S. Chaudhary. 2021. “The State of Economic Inclusion Report 2021: The Potential to Scale.” The World Bank. Bodewig, C., E. Calcutt, G. Conner, S. Cook, U. Gentilini, R. Hill, Z. Majoka, A. Narayan, M. Saidi, and C. Tassot. 2021. “Stress Testing Social Protection: A rapid appraisal of the adaptability of social protection systems and their readiness to scale-up ‒ A Guide for Practitioners. Version 1.” The World Bank. Bowen, T., C. del Ninno, C. Andrews, S. Coll-Black, U. Gentilini, K. Johnson, Y. Kawasoe, A. Kryeziu, B. Maher, and A. Williams. 2020. “Adaptive Social Protection: Building Resilience to Shocks.” International Development in Focus. Washington, DC: World Bank. Chaudhuri, S., J. Jalan, and A. Suryahadi. 2002. “Assessing household vulnerability to poverty from cross-sectional data: A methodology and estimates from Indonesia.” Columbia University Department of Economics Discussion Paper Series. Deltares, The Netherlands and Indonesia. 2020. “Baseline Study and Analytics on Systems Readiness and Stakeholder Capacity for an End-to-End Multi-Hazard Early Warning System Platform in Indonesia.” Global Facility for Disaster Reduction and Recovery (GFDRR) and The World Bank. Effendi, G., E.P. Purnomo, and A.D. Malawani. 2020. “Cash for Work? Extreme Poverty Solutions Based on Sustainable Development.” Journal of Economics and Policy 13 (2): 381-394. Elbers, C., J. Willem Gunning, and B.H. Kinsey. 2007. “Growth and Risk: Methodology and Micro Evidence.” World Bank Economic Review 21 (1): 1-20. Ferraro, P.J., and R. Simorangkir. 2020. “Conditional cash transfers to alleviate poverty also reduced deforestation in Indonesia.” Science Advances 6 (24). Food and Land Use Coalition. 2019. “Growing Better: Ten Critical Transitions to Transform Food and Land Use.” The Global Consultation Report. https://www.foodandlandusecoalition.org/global-report/. Francisco, H.A., C.D. Predo, A. Manasboonphempool, P. Tran, R. Jarungrattanapong, B.D. The, L. Peiialba, N.P. Tuyen, T.H. Tuan, D. Elazegui, Y. Shen, and Z. Zhu. 2011. “Determinants of Household Decisions on Adaptation to Extreme Climate Events in Southeast Asia.” Economy and Environment Program for Southeast Asia (EEPSEA). Frankenberg, E., B. Sikoki, C. Sumantri, W. Suriastini, and D. Thomas. 2013. “Education, Vulnerability, and Resilience after a Natural Disaster.” Ecology & Society 18 (2): 16. Global Facility for Disaster Reduction and Recovery (GFDRR). 2019. “Gender Inclusion in the Post-Disaster Rehabilitation and Reconstruction of Central Sulawesi: Workshop Summary: December 5th to 6th, 2019.” Government of Indonesia. 2021. Integrated Monitoring Dashboard for the Acceleration of Stunting Prevention. https:// dashboard.stunting.go.id/ Gubert, F., and A.S. Robilliard. 2008. “Risk and Schooling Decisions in Rural Madagascar: A Panel Data Analysis.” Journal of African Economies 17 (2): 207-238. Hales, S., S. Kovats, S. Lloyd, and D. Campbell-Lendrum (eds). 2014. “Quantitative risk assessment of the effects of climate change on selected causes of death, 2030s and 2050s.” Geneva: World Health Organization. P. 7 6 Hallegatte, S., M. Bangalore, L. Bonzanigo, M. Fay, T. Kane, U. Narloch, J. Rozenberg, D. Treguer, and A. Vogt-Schilb. 2016. “Shock Waves: Managing the Impacts of Climate Change on Poverty.” Washington, DC: World Bank. Hallegatte, S., A. Vogt-Schilb, M. Bangalore, and J. Rozenberg. 2017. “Unbreakable: Building the Resilience of the Poor in the Face of Natural Disasters.” Washington, DC: World Bank. Hallegatte, S., J. Rentschler, and B. Walsh. 2018. “Building Back Better. Achieving Resilience through Stronger, Faster, and More Inclusive Post-Disaster Reconstruction.” Global Facility for Disaster Reduction and Recovery. Washington, DC: World Bank. Hallegate, S., A. Vogt-Schilb, J. Rozenberg, M. Bangalore, C. Beaudet. 2020. “From Poverty to Disaster and Back: a Review of the Literature.” Economics of Disasters and Climate Change 4: 223-247. Hoffmann, R., and R. Muttarak. 2017. “Learn from the past, prepare for the future: Impacts of education and experience on disaster preparedness in the Philippines and Thailand.” World Development 96, Issue C. IPCC 2014. “Climate Change 2014: Synthesis Report. Contribution of Working Groups I, ll and lll to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change.” Geneva: Intergovernmental Panel on Climate Change. Jakarta Post. 2019. “3,622 natural disasters occurred in 2019: BNPB.” The Jakarta Post. https://www.thejakartapost. com/news/2019/12/18/3622-natural-disasters-occurred-in-2019-bnpb.html Klasen, S., and H. Waibel. 2014. “Vulnerability to Poverty: Theory, Measurement and Determinants, With Case Studies from Thailand and Vietnam.” New York: Palgrave and Macmillan. Lindert, K., T.G. Karippacheril, I. Rodriguez Caillava, and K. Nishikawa Chavez (eds). 2020. “Sourcebook on the Foundations of Social Protection Delivery Systems.” Washington, DC: World Bank. Lubis, R., B. Satriatna, and A. Nugroho (unpublished manuscript, 2021). “Policy and Delivery System Review of Indonesia’s Social Protection Response to the 2018 Central Sulawesi Earthquake and Tsunami: Qualitative Study Report.” The World Bank. Macours, K., P. Premand, and R. Vakis. 2012. “Transfers, Diversification and Household Risk Strategies: Experimental Evidence with Lessons for Climate Change Adaptation.” Policy Research Working Paper 6053. The World Bank. Microsave Consulting. 2021. “Impact of the COVID-19 pandemic on Program Keluarga Harapan (PKH) beneficiaries and program implementation.” Ministry of Finance, Fiscal Policy Agency (Government of Indonesia). 2022. “Pooling Fund Bencana.” https://fiskal. kemenkeu.go.id/strategi-drfi/pooling Ministry of Social Affairs, Indonesia. 2020. “Results of Integrated Evaluation of Social Security and Protection Programs.” Presentation. Pritchett, L., A. Suryahadi, and S. Sumarto. 2000. “Quantifying vulnerability to poverty: a proposed measure with application to Indonesia.” SMERU Working Paper. Purnamasari, R.S., A. Febriady, B.A. Wirapati, M.N. Farid, P. Milne, Y. Kawasoe, J. Vun, R. Engstrom, and M. Nasiir. 2021. “Welfare Tracking in the Aftermath of Crisis: The Central Sulawesi Disaster Response.” Jakarta: World Bank. Rigolini, J. 2021. “Social Protection and Labor: A Key Enabler for Climate Change Adaptation and Mitigation.” Washington, DC: World Bank. Rosenzweig, M.R., and H.P. Binswanger. 1993. “Wealth, Weather Risk and the Composition and Profitability of Agricultural Investments.” The Economic Journal 103 (416): 56-78. P. 7 7 Rush, J.V. 2018. “The Impact of Natural Disasters on Education in Indonesia.” Economics of Disasters and Climate Change 2: 137-158. Sen et al. (unpublished manuscript, 2022). “Adaptive Social Protection Stress Test for Indonesia. Part 2 of the Stress Test: Annotated Assessment of Scalability and Adaptivity of Social Protection.” The World Bank. Skoufias, E. 2012. “The Poverty and Welfare Impacts of Climate Change: Quantifying the Effects, Identifying the Adaptation Strategies.” Washington, DC: World Bank. Syamsulhakim, E., and N. Khadijah. 2021. “Graduating from a Conditional Cash Transfer Program in Indonesia: Results of a Household Survey of Prosperous-Independent Graduates of the Family Hope Program (PKH) in 2020.” Washington, DC: World Bank. Wai-Poi, M. 2014. “Mengukur Kerentanan Indonesia” (Measuring Indonesian Vulnerability). Forum Kebijakan Publik Indonesia. Jakarta. Williams, A. 2020. “Post-Disaster Household Assessments and Eligibility Determination for Post-Disaster Social Protection Benefits.” Washington, DC: World Bank. Williams, A., and V. Moreira. 2020. “Making Social Protection Information Systems Adaptive.” Washington, DC: World Bank. World Bank. 2006. “Making the New Indonesia Work for the Poor.” Washington, DC: World Bank. World Bank. 2017. “Climate-Smart Healthcare. Low-Carbon and Resilience Strategies for the Health Sector.” Washington, DC: World Bank. World Bank. 2019. “Strengthening the Disaster Resilience of Indonesian Cities: A Policy Note.” Washington, DC: World Bank. World Bank. 2019a. “Project Appraisal Document for a Proposed Loan for an Institutional Strengthening for Improved Village Service Delivery Project.” Jakarta: World Bank. World Bank. 2020. “Investing in People: Social Protection for Indonesia’s 2045 Vision.” Jakarta: World Bank. World Bank. 2020a. “Indonesia Social Assistance Reform Program for Results Additional Financing - Project Appraisal Document.” World Bank. 2020b. “Indonesia Disaster Risk Finance and Insurance Investment Project Financing - Project Appraisal Document.” World Bank. 2021. “A Green Horizon: Toward a High Growth and Low Carbon Economy.” Indonesia Economic Prospects December 2021. Jakarta: World Bank. World Bank. 2021a. “Boosting the Recovery.” Indonesia Economic Prospects June 2021. Jakarta: World Bank. World Bank. 2021b. “Stress Testing Social Protection: A Rapid Appraisal of the Adaptability of Social Protection Systems and their Readiness to Scale Up.” World Bank. 2021c. “Indonesia High Frequency (HiFy) Monitoring of COVID-19 Impacts: Round 5.” World Bank. 2021d. “Indonesia Inclusive Livelihoods for Poor Rural Communities in Eastern Indonesia Project: Project Information Document.”https://documents1.worldbank.org/curated/en/921881625726691146/pdf/ Project-Information-Document-PID-Indonesia-Inclusive-Livelihoods-for-Poor-Rural-Communities-in-Eastern-Indonesia- Project-P174902.pdf World Bank. 2022. “Adaptive Social Protection Stress Test for Indonesia Part 2 of the Stress Test: Annotated Assessment of Scalability and Adaptivity of Social Protection.” Jakarta: World Bank.