Report No: AUS0000291 . Indonesia Long-Term Generasi Qualitative Study . May 2018 . URS . THE WORLD BANK OFFICE JAKARTA Indonesia Stock Exchange Building Tower II/12th Floor Jl. Jend. Sudirman Kav. 52-53 Jakarta 12910 Tel: (6221) 5299-3000 Fax: (6221) 5299-3111 Website: www.worldbank.org/id THE WORLD BANK 1818 H Street NW Washington, DC 20433, USA Tel: (202) 458-1876 Fax: (202) 522-1557/1560 Website: www.worldbank.org Printed in May 2018 Long-Term Generasi Qualitative Study is a product of the staff of the World Bank. The findings, interpretations, and conclusions expressed herein do not necessarily reflect the views of the Board of Executive Directors of the World Bank or the Government they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of the World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. For any questions regarding this report, please contact Audrey Sacks (asacks@worldbank.org). _____________________________________________________________________________________________ Copyright Statement: The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. The International Bank for Reconstruction and Development/ The World Bank encourages dissemination of its work and will normally grant permission to reproduce portions of the work promptly. For permission to photocopy or reprint any part of this work, please send a request with complete information to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA, telephone 978-750-8400, fax 978-750-4470, http://www.copyright.com/. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA, fax 202-522-2422, e-mail pubrights@worldbank.org. ii Acknowledgements Long-Term Generasi Qualitative Study is a product of Social, Urban, Rural and Resilience Global Practice’s team in the World Bank Office Jakarta. Support for this report has been generously provided by the Department of Foreign Affairs and Trade of the Australian Embassy. This paper was prepared by a core team led by Audrey Sacks (Senior Social Development Specialist, GSU03). Jesse Grayman was the lead researcher for this study. The research team also consisted of Marzi Afriko, Nelti Anggraini, Kelik Endarso, Hestu Prahara, Agung Prabowo, Lina Rozana, Ali Subandoro, and Rob Wrobel. The paper was edited by Kelley Friel (Consultant, GSUID). This report was produced under the overall guidance of Kevin Tomlinson (Practice Manager, GSUID) and Susan Wong (Lead Social Development Specialist, GSUSD). Key comments were provided by Andrew Beath (Senior Economist, GFCSN) and Andrea Woodhouse (Senior Social Development Specialist, GSU03). List of abbreviations AIPMNH Australia Indonesia Partnership for Maternal and Neonatal Health APBdes Village Budget BLM Community Block Grant, Bantuan Langsung Masyarakat BPD Village Council, Badan Perwakilan Desa BPMD Village Community Empowerment Agency, Badan Pemberdayaan Masyarakat Desa BPMPD Community and Village Government Empowerment Agency, Badan Pemberdayaan Masyarakat dan Pemerintahan Desa CDD Community-Driven Development DPRD District Assembly, Dewan Perwakilan Rakyat Daerah FGD Focus Group Discussion FK Subdistrict Facilitator, Fasilitator Kecamatan GoI Government of Indonesia ICT Information and Communication Technology IDM Village Development Index, Indeks Desa Membangun KPMD Village Community Empowerment Cadres (Generasi village facilitators), Kader Pemderdayaan Masyarakat Desa MAD Inter-village Meetings, Musyawarah Antar Desa NTT Nusa Tenggara Timur province P3MD Development and Empowerment of Village Communities Program (village funds) PAUD Early Child Education and Development, Pendidikan Anak Usia Dini PD Village Facilitator, Pendamping Desa PenLok Mentor to Subdistrict Facilitator (same as PL) PJOK Activity Operations Superintendent PK Generasi Activity Implementer, Pelaksana Kegiatan PKK Family Welfare Movement, Pembinaan Kesejahteraan Keluarga PL Mentor to Subdistrict Facilitator PLD Village Law facilitator, Pendamping Lokal Desa PNPM National Community Empowerment Program, Program Nasional Pemberdayaan Masyarakat PONED Basic Emergency Obstetric Care, Pelayanan Obstetri Neonatus Essensial Dasar PMT Supplementary Food PSD Basic Social Services, Pelayanan Sosial Dasar RAB Budget Plan RKPDes Annual Village Development Plan RPJMdes Mid-Term Village Development Plan TA-PSD Technical Assistants for Basic Social Services at the district level, Tenaga Ahli Pelayanan Sosial Dasar TBA Traditional Birth Attendant TPK Village Implementation Team, Tim Pengelola Kegiatan TPMD Village Deliberation Advisory Team, Tim Pertimbangan Musyawarah Desa UPK Subdistrict Implementation Unit, Unit Pengelola Kecamatan 1 List of terms Bantuan Assistance Bupati District Head Camat Subdistrict Head Faskab Generasi District Facilitator (fasilitator kabupaten) Gotong Royong Mutual cooperation Kabupaten District Kecamatan Subdistrict Kepala Desa Village Head Minilok Mini workshops Polindes Village maternity post Poskesdes Village health post Posyandu Health clinic Puskesmas Subdistrict community health center Siskeudes Village finance system 2 Table of Contents Acknowledgements 0 Executive Summary 4 Introduction 7 Methodology 8 Generasi Program 11 Facilitation 13 Facilitation: Three Understandings 14 Role of Facilitators 14 Facilitator Workload Distribution 19 Information Sharing and Learning Among Facilitators 22 Perceptions of (and Interactions with) Village Law Facilitators 22 Facilitators’ Independence from Government 23 Participation 25 Perceptions of Participation 25 Encouraging Participation 26 Community Monitoring of Basic Social Services and Village Governance 27 Community Contributions towards Stimulating Basic Social Service Utilization 29 Are Health and Education Shared Concerns in the Community? 30 Participatory Budgeting Processes (Including for Generasi Block Grants) 31 Beneficiaries’ Motivation to Participate 32 Impact on Village Law Implementation 32 Targets and Performance Bonuses 33 Awareness and Understanding of Generasi’s 12 Target Indicators 34 Do Targets Induce Learning and/or Competition? 35 “Gaming” the Target System 37 Understanding and Effectiveness of Generasi’s Performance Bonus System 38 Generasi’s Impact on Village Governance and Basic Social Services 39 Many Generasi Actors Have Entered Village (or Higher) Government 40 Generasi’s Influence on the Allocation of non-Generasi Village Funds 41 Perceived Barriers to Using Non-Generasi Funds for Health and Education 45 Generasi Impacts on Local Regulations 49 Generasi’s Impact on Basic Social Services 49 Conclusions 50 Recommendations 51 Annex A: Description of Sample Sites 53 Descriptive Statistics 53 Subdistrict and Village Profiles 54 3 Executive Summary In 2007, the Government of Indonesia (GoI) introduced PNPM Generasi (National Community Empowerment Program – Healthy and Smart Generation, Program Nasional Pemberdayaan Masyarakat – Generasi Sehat dan Cerdas) to address key policy priorities and the Millennium Development Goals – reducing poverty, maternal mortality and child mortality, as well as ensuring universal coverage of basic education. Generasi provides over 5,400 villages with an annual block grant, which each village can allocate to any activity that supports one of 12 indicators of health and education service delivery. Each village’s success in meeting these 12 targets helps determine the size of the next year’s grant. Trained facilitators recruited from within the communities help implement the program. To facilitate a rigorous evaluation of the program, GoI (working with the World Bank and the Abdul Latif Jameel Poverty Action Lab) randomly assigned Generasi locations for the pilot phase (2007–09). A randomized evaluation of two different versions of the program (with and without performance bonuses) was conducted in three rounds (Wave I at baseline, Wave II 18 months after implementation, and Wave III 30 months after implementation). In 2016/2017, the Impact Evaluation (IE) team fielded a follow-up survey in the same subdistricts as the first three waves. A separate report analyzes the quantitative findings of this final survey. During the final survey round, the IE team also collected qualitative data in geographically distinct treatment and control communities in order to explore two questions. First, are Generasi’s three components – facilitation, community participation, and the target and performance bonus system – functioning as intended? Second, what is the program’s long-term impact on village governance and service delivery, and how can it influence Village Law implementation? The qualitative findings related to the first question can be summarized as follows. First, Generasi facilitators were found to have technical knowledge about health and education issues, experience in mobilizing communities around basic social service delivery, and creative problem-solving skills. They maintain communication and cooperation with a variety of actors in the community and at different levels of government to assess community needs and address problems. Generasi facilitators were found to be better informed than Village Law facilitators about their roles and responsibilities as well as the technical aspects of their jobs, and were better integrated into the areas in which they work. Second, the fieldwork revealed that many facilitators and beneficiaries interpret community participation as attending meanings and utilizing services, which fails to advance the program’s goal of empowering communities to plan, implement, and monitor the delivery of basic services and influence village governance. Third, the study found that while facilitators at all levels were aware of the 12 health and education related targeted indicators, few understood how they related to the performance bonuses. Public accountability appears to serve as a more important motivation to achieve the targets: village leaders wish to avoid the embarrassment of reporting at inter-village meetings that they failed to meet them. In regards to the second question, the study found that Generasi has had a significant impact on village governance, but not on the delivery of sector-based service providers such as puskesmas (subdistrict community health center). Several program actors have taken on important roles in their communities, which has helped embed Generasi-style consultation and implementation approaches in village planning processes and encouraged villages to allocate funds for health and education in their budgets. Yet there have been fewer contacts with (or advocacy efforts targeted at) service providers, which may explain the program’s uneven impact. Based on the qualitative findings, the following recommendations can be made in three main areas. Facilitation • Village Law facilitators, specifically pendamping lokal desa (PLD), should be recruited from the local communities they serve. Both PLD and pendamping desa (PD) should receive training in basic social services to allow them to effectively advocate for village governments to utilize funds for health and 4 education services once Generasi ends. For training materials, the Ministry of Villages, Disadvantaged Areas and Transmigration (MoV) could draw on curriculum developed under Generasi. • Posyandu (village health clinic) volunteers and other community-based volunteers should receive training in health service delivery. Puskesmas staff could deliver complete and routine training for posyandu cadres and all community-based health volunteers starting with curriculum the Ministry of Health has already developed for this purpose. Village governments could help to pay for these training costs. • While it is important for Village Law facilitators to continue collecting health, education, and other basic indicators, MoV should reduce their data collection burden. MoV and other community-based programs should consider training and paying community volunteers such as former Generasi Village Community Empowerment Cadres (Kader Pemderdayaan Masyarakat Desa, KPMD) to collect routine data, which would free up facilitators’ time for outreach and enlarge the network of community volunteers. • Subdistricts (kecamatans) should continue coordinating inter-village meetings (musyawarah antar desa, MADs) post-Generasi in which village actors discuss community problems and exchange advice. These meetings can also be adapted and used to motivate and incentivize village governments’ performance. Subdistricts should build on the locally developed innovations that contributes to village governance practices, which this report highlights. Participation • Community-driven development projects should encourage equal participation in the full project cycle from both village-level elites and community members, including posyandu volunteers and community facilitators. While ordinary community members may not participate in village-level planning meetings in large numbers, KPMDs are actively in touch with them, and are thus able to present their diverse needs at the planning meetings and follow up with outreach activities. If community-level facilitators are selected from the communities they serve, they can provide ongoing support and help to bring community members’ complaints and needs to higher levels. • Given the important role that posyandu volunteers play in providing maternal and infant health services, village governments should invest more in the posyandu and continue to ensure that posyandu are sufficiently staffed (at least five per village and one per hamlet) and that they are compensated appropriately. Performance targets can help village governments monitor posyandu activities and ensure continued investment in them • The analysis highlights many examples of how local culture continues to influence health and education behaviors and modes of accountability. Under the Village Law, service providers, village governments and facilitators should consider how to tailor health and education activities to local cultures to influence behaviors. Targets and incentives • Generasi’s target system was effective at motivating Generasi facilitators to mobilize communities around the targets. Future programs that consider adopting such a system should ensure that there is a forum like the MAD meetings that puts pressure on village law facilitators, kepala desas (village heads), and village governments to ensure their villages meet their targets. • Programs that adopt a Generasi-style performance bonus may want to consider simplifying the system, and to ensure high levels of awareness of the process among program actors and community 5 beneficiaries throughout the project cycle. The monetary rewards associated with the bonus should also be more substantial. • MoV should consider putting in place a simple set of performance targets for village governments. The above-mentioned MAD or Reporting Day meetings could incentivize village governments to collect data needed to report back, and put pressure on service providers to deliver more and better to make the kepala desas’ performance stand out. • Future programs should consider rewarding individual facilitators with non-monetary bonuses (e.g., a package of household supplies) as a possible alternative or complement to a community-level performance bonus. Evidence from the qualitative study suggests that informal rewards help KPMDs and other village actors feel appreciated and motivated. 6 Introduction In 2007, the Government of Indonesia (GoI) introduced PNPM Generasi (National Community Empowerment Program – Healthy and Smart Generation, Program Nasional Pemberdayaan Masyarakat – Generasi Sehat dan Cerdas) to address key policy priorities and the Millennium Development Goals – reducing poverty, maternal mortality and child mortality, as well as ensuring universal coverage of basic education. In 2014, the Generasi program was renamed Generasi Sehat Cerdas (“Bright Healthy Generation”) when it transferred administration from the Ministry of Home Affairs (MoHA) to the Ministry of Villages, Disadvantaged Areas and Transmigration (MoV). Under Generasi, over 5,400 villages receive an annual block grant. With the assistance of trained facilitators, each village can allocate these grants to any activity that supports one of 12 indicators of health and education service delivery. Generasi employs unique performance incentives: the size of a village’s block grant depends in part on its performance on the 12 targeted indicators during the previous year. While 80% of a subdistrict’ s funds are divided among villages in proportion to the number of target beneficiaries, the remaining 20% forms a bonus pool that is distributed to villages in the subsequent year based on their performance on the 12 indicators. To facilitate a rigorous evaluation of the program, GoI (working with the World Bank and the Abdul Latif Jameel Poverty Action Lab) randomly assigned Generasi locations for the pilot phase (2007–09). A randomized evaluation of two different versions of the program (with and without performance bonuses) was conducted in three rounds (Wave I at baseline, Wave II 18 months after implementation, and Wave III 30 months after implementation). Based on the 2009 Impact Evaluation (IE), which found that performance incentives accelerated improvement in preventative health and malnutrition, the performance bonuses were scaled up and the program was expanded beyond the pilot locations. Since the expansion took place almost entirely in new provinces, the original randomization remained intact. Nine years after the program began, of the 181 subdistricts that were originally randomized to receive the program, 156 continue to receive it; of the 83 subdistricts originally randomized to the control group, 80 remain as controls. This creates a virtually unprecedented opportunity to study the long-term effects of community mobilization on improving health and education, and to understand the long-run impacts of nutritional improvements. Between September 2016 and January 2017, the IE team fielded a survey of households, service providers, program facilitators, and government officials in the same subdistricts as the first three waves. A separate report analyzes the quantitative findings of this final survey. The final quantitative IE has two primary objectives. First, it estimates the impact of community block grants on maternal and child health-seeking behavior and health for new cohorts. Using data from direct observations and interviews with health providers and households, the IE assesses how the community block grants and performance incentives increase the use of pre-natal care, childbirths assisted by trained personnel, post-natal care, immunizations, and participation in growth monitoring for newly born cohorts. Second, it assesses how maternal and infant health improvements translate into better child outcomes later in life. Specifically, the IE examines how cohorts that have been exposed to the program for most of their lives are faring now in terms of health outcomes (such as height and weight), as well as enrollment and performance at the primary and junior/secondary school levels. The quantitative results of the IE can be summarized as follows. Overall, Generasi has continued to be effective at mobilizing community members to attend the posyandu (village health clinic) for infant weighing and to attend maternal health and parenting classes. In the lowest-performing districts, Generasi continues to encourage community members to attend the posyandu and to increase immunization rates and vitamin A distribution. Its initial impact on stunting, which was concentrated in Nusa Tenggara Timur (NTT) province, was not sustained beyond the 2009 IE. There are a few possible reasons for this. First, the overall substantial improvements in stunting in NTT that occurred in both control and treatment areas may have exhausted the 'low-hanging fruit' that Generasi was able to solve in earlier periods. Second, Generasi funding 7 produced crowd-in/crowd-out effects on other program resources that undercut the efficacy of the intervention. Third, implementation issues and delays in the maternal health and parenting classes may have weakened any potentially positive impacts this intervention may have had on behavioral change and malnutrition. Fourth, Generasi’s effects on stunting were limited because the full suite of complementary demand- and supply-side interventions needed to address stunting were not fully implemented. At the same time as the final survey round, the IE team also collected qualitative data in geographically distinct treatment and control communities. The research methodology is described in greater detail in the next section. The main objective of the qualitative study (reported here) is to examine the relationships between Generasi’s three components – facilitation, community participation, and the target and performance bonus system – and the achievement of outcomes. The qualitative findings in these three areas can be summarized as follows. First, Generasi facilitators were found to have technical knowledge about health and education issues, experience in mobilizing communities around basic social service delivery, and creative problem-solving skills. They maintain communication and cooperation with a variety of actors in the community and at different levels of government to assess community needs and address problems. Generasi facilitators were found to be better informed than Village Law1 facilitators about their roles and responsibilities as well as the technical aspects of their jobs, and were better integrated into the areas in which they work. Second, the fieldwork revealed that many facilitators and beneficiaries interpret community participation as attending meanings and utilizing services, which fails to advance the program’s goal of empowering communities to plan, implement, and monitor the delivery of basic services and influence village governance. Third, the study found that while facilitators at all levels were aware of the 12 health and education related targeted indicators, few understood how they related to the performance bonuses. Public accountability appears to serve as a more important motivation to achieve the targets: village leaders wish to avoid the embarrassment of reporting at inter-village meetings that they failed to meet them. The study found that Generasi has had a significant impact on village governance, but not on the delivery of sector-based service providers. Several program actors have taken on important roles in their communities, which has helped embed Generasi-style consultation and implementation approaches in village planning processes and encouraged villages to allocate funds for health and education in their budgets. Yet there have been fewer contacts with (or advocacy efforts targeted at) service providers, which may explain the program’s uneven impact. Methodology Qualitative Research Methods: The Generasi Long-Term Qualitative Study employed the following research methods to identify pathways and processes that determine levels and quality of facilitation, participation, and use of performance bonuses: 1. Focus group discussions (FGDs) 2. Semi-structured interviews 1 The Village Law was passed in 2014 and is a massive decentralization effort that substantially increases direct transfers to villages. Village transfers will be scaled over time. The national government allocated IDR 280 million (US$20,000) in 2015, and district governments are estimated to allocate around IDR 500 million (US$40,000). Each village will receive approximately IDR 1.4 billion (US$122,000) on average each year. 8 3. Observation and description 4. Document collection 5. Mobile information and communication technologies (ICTs) 6. Videography The first three methods listed above were the primary tools for addressing the research questions. Semi- structured interviews and FGDs were based on questions and discussion guides developed during the research staff training, which were refined after the pilot phase. Questions were adapted to the different types of stakeholders (described in more detail below). FGDs solicit normative, agreed-upon master narratives of how facilitation, participation, and incentives work within a small group of similar stakeholders in a community such as village facilitators or mothers of young children. Semi-structured interviews follow the FGDs to solicit specific detailed examples, individual variations in practice, and contested ideas that may veer from the normative FGD responses. Observations validate and add contextual depth to FGD and interview reports. Observations include detailed descriptions of: interview settings; sites visited, including government offices, health facilities, and respondent households; village geography; public resources; and, most importantly, Generasi and Village Law activities such as village consultation and planning meetings, posyandu clinics, and subdistrict inter-village meetings. Site-specific documents collected from government line agencies and programs provide reference and background material for each district, subdistrict, and village, and allow comparisons of the quality of record keeping in each location. ICTs were used to communicate with stakeholder-informants throughout the program cycle even after researchers left the field. ICTs were also useful for clarifying and cross-checking data during the analysis and write-up phases. Generasi stakeholders maintain informal professional networks by using ICTs such as Facebook and WhatsApp; research staff occasionally asked permission to observe or participate in these online communities to learn more about how facilitators talk about their work, which adds significant insight and depth to the research questions about facilitation. Videography provided a visual archive to supplement narrative descriptions, and has been used to report research findings in multi-media formats to partner stakeholders in Jakarta and elsewhere, showing what facilitation meetings and other Generasi stakeholder events look like in the sampled communities. Stakeholder Sampling: Interviews, FGDs, observations, document collection, and digital conversations with and among stakeholder-informants proceeded hierarchically downwards through a snowballing network of officials, facilitators, and (eventually) beneficiaries, moving from districts, to subdistricts, to villages. Data collection proceeded in this fashion because access to village communities required consultations and permissions at every level. It typically required at least two days of fieldwork at the district and subdistrict level before a team of researchers could begin talking to informants at the village level. The stakeholders involved at each level of government include: • Province level: Offices of education, health, finance and community empowerment, and the Generasi Provincial Coordinator.2 • District level: Offices of education, finance, health, and Village Community Empowerment Agency (BPMDes), and the Generasi district facilitator (fasilitator kabupaten, faskab). • Subdistrict level: subdistrict implementation unit (unit pengelola kecamatan, UPK) staff, puskesmas (subdistrict community health center) staff (including the director and the midwife coordinator); teachers or the head of school; health forum members, heads and staff of the Subdistrict Technical 2 Meetings at the provincial level were not a required component of the sampling, but as the research teams passed through the provincial capitals in Gorontalo and NTT provinces, the team opportunistically met with several provincial level officials in these two provinces only. 9 Implementation Unit education offices, the Generasi subdistrict facilitator (fasilitator kecamatan, FK), Generasi support staff, and Village Law staff and facilitators. Village/Desa level: Formal village leadership (including the kepala desa (village head) and village secretary), Generasi Village Community Empowerment Cadres (Kader Pemderdayaan Masyarakat Desa, KPMDs) and Generasi activity implementers (pelaksana kegiatan, PK), Village Law facilitators, posyandu volunteers, health forum members, village clinic staff (especially village midwives), traditional birth attendants, school committee members, early childhood and education development (Pendidikan Anak Usia Dini, PAUD) teachers, and parents of young children. Site Sampling Criteria: The three provinces selected for inclusion in the study were based on the 2009 levels of high, medium, and low levels of stunting – NTT, Gorontalo, and East Java, respectively. One district was randomly selected within each of these provinces for the study. Within the selected district, two subdistricts (one treatment and one control) were randomly chosen. Within the treatment subdistrict, all villages were ranked based on two variables using the 2005 Village Potential Statistics (Potensi Desa in Indonesia): distance to the subdistrict seat and the number of households receiving Jamkesmas (the national health insurance scheme) per capita. Once ranked, two villages were randomly selected: one average village and one poor village (high density of poor households and far away from the subdistrict seat). A similar ranking of the control subdistricts was completed, and then one average-scoring village was randomly selected for inclusion. The results of this sampling procedure are shown in the table below. Sampled subdistricts and villages have been assigned pseudonyms in compliance with qualitative research best ethical practice. Generasi Long-Term Qualitative Impact Evaluation Site Selection Results Province / District / Subdistrict / Kecamatan Village / Desa Propinsi Kabupaten West Java Garut Lebak Siwur (treatment) Pasir Ucing (pilot) (pilot) Cikereteg (pilot) Gorontalo Gorontalo Tarengge (treatment) Mangkawani (weak) Maroangin (average) Telogojoyo (control) Jaton (average) NTT Lembata Nelle (treatment) Rampe (weak) Ilekora (average) Tanabola (control) Mogiye (average) East Java Pamekasan Petis (treatment) Lelaok (weak) Sogiyan (average) Sambingan (control) Tespates (average) Prior to fieldwork in the locations listed above, the research team spent four days piloting the instruments and getting familiar with Generasi program dynamics on the ground in the West Java district of Garut, in two 10 Generasi villages in Lebak Siwur subdistrict: Cikereteg and Pasir Ucing. Some relevant data from these pilot sites appears in this report. The fieldwork schedule proceeded as follows, with the weeks in between fieldwork spent writing up field notes before traveling to the next destination: Garut, West Java: 31 August – 3 September 2016 (pilot phase) Gorontalo, Gorontalo: 7 – 25 September 2016 Lembata, NTT: 16 October – 7 November 2016 Pamekasan, East Java: 6 – 27 December 2016 Documentation of Primary Data: A team of five or six researchers conducted the fieldwork. All interviews, FGDs, descriptive observations, and visual media were documented with comprehensible, legible, and well- organized field notes. Interview and FGD field notes address a detailed research question guide, with a separate narrative summary of interviewer observations (household descriptions, interview dynamics, etc.). Each interview or FGD participant is documented in an informant list mentioning the date of the interview, the interviewer’s name, and the name and position of each informant. Field observations were compiled into district, subdistrict, and village profiles. Field notes were transcribed and expanded from shorthand to narrative within a few days of completing the fieldwork in each province. When permissible and ethically appropriate, researchers documented interview settings and field observations with audio and/or visual recording devices. Analysis and Reporting of Results: The analyses began with the documentation of primary data described above, in the form of individual researchers’ margin notes: themes were noted, patterns were identified, and new questions were posed. Interim comparative analyses proceeded with routine communication among researchers, at scheduled times, via WhatsApp for example. These conversations included the Lead Researcher, regardless of whether he or she was present in the field. This facilitated early and iterative insights that developed over time. After fieldwork in each location, researchers performed a preliminary analysis of the data by excerpting sections from their transcripts that addressed specific research questions. These excerpts were inserted into a database to facilitate an organized thematic analysis for each research question using NVivo.3 Interesting case studies were pursued by referring to complete interview transcripts and triangulating data across local sources, sometimes with follow-up phone calls with informants to cross-check and elaborate on the findings. The sources of many quotes are attributed to a position rather than an individual in order to protect confidentiality. Personal names in the case studies are all pseudonyms. Generasi Program4 Generasi began in mid-2007 in 164 pilot subdistricts spread across five provinces selected by GoI: West Java, East Java, North Sulawesi, Gorontalo, and NTT. By the time of the first IE in 2009, the program was operating in 264 subdistricts across these five provinces. It currently operates across 499 subdistricts in eleven provinces. The current report and analysis focuses on districts considered in the Wave III IE which took place in 2009/10. The Generasi project focuses on 12 indicators of maternal/child health and educational behavior. These indicators are in line with Ministry of Health priorities and protocols and GoI’s constitutional obligation to ensure nine years of basic education for all Indonesian children. These 12 indicators relate to seeking health 3 After identifying patterns in the data, discussions were held with the research team to establish the validity of the analysis by elaborating upon specific examples in the data. 4 Portions of the description of the Generasi program in this section, as well as the experimental and evaluation design sections, draw directly from B.A. Olken, J. Onishi, and S. Wong. 2011. “Indonesia PNPM Generasi Program Final Impact Evaluation Report.” Washington, DC: World Bank. 11 and educational services that are within the direct control of villagers – such as the number of children who receive immunizations, pre- and postnatal care, and the number of children enrolled and attending school – rather than long-term outcomes, such as test scores or infant mortality. As school enrollment rates improved significantly across control and treatment areas over the past decade, in 2014 Generasi revised its education targets to better focus investments on the neediest populations. The new education targets include participation rates for children with disabilities and transition rates from primary to junior secondary school. In addition, Generasi introduced indicators to measure community participation in enhanced nutrition counseling sessions delivered through the posyandu. Under the Generasi program, all participating villages receive a block grant each year to improve education and maternal and child health. For example, these grants can be used for a wide variety of purposes, including hiring extra midwives for the village, subsidizing the costs of pre- and postnatal care, providing supplementary feeding (PMT), hiring extra teachers, opening a branch school in the village, providing scholarships or school supplies, providing transportation funds for health care or school attendance, improving health or school buildings, or rehabilitating a road to improve access to health and education facilities. Trained facilitators help each village elect an 11-member village management team and select local facilitators and volunteers to decide how to allocate the block grants (see Table 1). Through social mapping and in-depth discussion groups, villagers identify problems and bottlenecks in reaching the indicators. Inter-village meetings and consultation workshops with local health and education service providers allow community leaders to obtain information, technical assistance, and support from the local health and education offices and coordinate the use of Generasi funds with other health and education interventions in the area. Following these discussions, the elected management team makes the final Generasi budget allocation. Performance incentives are a central element of the Generasi approach. The size of a village’s block grant depends in part on its performance on the 12 targeted indicators in the previous year. The performance bonus is structured as a relative competition among villages within the same subdistrict. A fixed allocation to each subdistrict also ensures that the bonus system does not result in the unequal geographic distribution of funds. The incentive is designed to encourage a more effective allocation of Generasi funds and to stimulate village outreach efforts to encourage mothers and children to obtain appropriate health care and increase educational enrollment and attendance. The Generasi project design built on GoI’s PNPM Rural program, which, along with its predecessor program (Kecamatan Development Project (KDP)), have funded over US$2 billion in local infrastructure and microcredit programs in some 61,000 Indonesian villages over the period from 1998–2014.5 MoV implements Generasi, which is funded through GoI resources and in part by loans from the World Bank and grants from several bilateral donors. Technical assistance and evaluations have been supported by a multi- donor trust fund with contributions from the World Bank, embassies of the Netherlands, Australia, United 5For more than fifteen years, Indonesia has been pioneering and implementing various community-driven development (CDD) projects and programs. Beginning in 1997, when the Kecamatan Development Project (KDP) was piloted in 25 villages, Indonesia has pioneered the design, management, and expansion of projects that give communities more control over the plans and resources that shape their towns and villages. In 2007, the Government scaled-up the KDP nationwide, combined it with other community-based poverty programs and renamed it the National Program for Community Empowerment (Program Nasional Pemberdayaan Masyarakat Mandiri or PNPM Mandiri). PNPM Rural and PNPM Urban ultimately reached more than 70,000 villages and urban wards across Indonesia. As part of the PNPM umbrella, the Government developed several pilot projects to expand PNPM to vulnerable groups and new activities. These projects included Generasi, PNPM Respek, PNPM Green and PNPM Pedulie. For more information about PNPM and these pilot projects, see J. Friedman. 2012. Expanding and diversifying Indonesia's program for community empowerment, 2007-2012. Washington, DC: World Bank Group. https://hubs.worldbank.org/docs/ImageBank/Pages/DocProfile.aspx?nodeid=23040388 . 12 Kingdom, and Denmark, and the World-Bank-managed Spanish Impact Evaluation Fund. The 2016 IE and qualitative study was supported by the Australian Department of Foreign Affairs and Trade. Facilitation Generasi hires full-time facilitators to work at the district and subdistrict levels and recruits semi-volunteers at the village levels. These facilitators mobilize community members, especially the poor and marginalized, to utilize mother and child health services for the first 1,000 days of life.6 They also ensure that communities access primary education services, including for children with special needs, and encourage out-of-school children to return to school. At the district level, the faskab’s responsibilities include overseeing the work of FKs and intervening at the village-level when an issue came up that the FK couldn’t solve. Other faskab responsibilities include explaining Generasi to the bupati (district government head), the district assembly (Dewan Perwakilan Rakyat Daerah, DPRD), the health and education agencies, and coordinating Generasi activities with the health and education offices. At the subdistrict level FKs support communities with implementing each program stage, including socialization, training, social mapping, planning, implementation, and maintenance. FKs also conduct financial audits of Generasi accounts and activities, and collect monthly performance data on the 12 targets. At the village level, community members select KPMDs to help prepare, plan, and implement activities. The Village Deliberation Advisory Team (Tim Pertimbangan Musyawarah Desa, TPMD) mainly supports communities at the sub-village (hamlet) level with developing and prioritizing proposals for Generasi to fund. The qualitative analysis explores what aspects of facilitation are most effective, and how facilitation helps communities achieve their main targets. The analysis reveals that effective subdistrict – and village-level facilitators have technical knowledge about health and education issues, experience in mobilizing communities around basic social service delivery, and creative problem-solving skills. Effective faskabs proactively address community-level issues by reaching out to (and maintaining frequent communication with) government actors outside the Generasi program – including the bupati, BPMD, Community and Village Government Empowerment Agency (Badan Pemberdayaan Masyarakat dan Pemerintahan Desa, BPMPD), and the health and education agencies. Similarly, effective FKs coordinate closely with the camat’s (subdistrict head) office and service providers; when service delivery problems emerge, they reach out to government partners to help solve them. At all government levels, the coordination between facilitators and service providers was cooperative, especially with respect to data validation and case referral. Recruiting and retaining facilitators remains a problem for Generasi. High-quality KPMDs are often recruited into the village government or to other jobs, and new recruits face steep learning curves; facilitators require at least a year of participation in the program cycle before they are experienced enough to take on leadership and decision-making roles. In addition, the heavy burden placed on facilitators to collect Generasi-related (and, more, recently Village Law-related) data often comes at the expense of advocacy and outreach. The qualitative analysis also compares Generasi and Village Law facilitation, which both emphasize participation, transparency and accountability. While this is a somewhat unequal comparison, since Village Law facilitators were deployed only a few months prior to the qualitative fieldwork and lacked a general understanding of their roles and responsibilities, the differences may have implications for the implementation of the new law, which are discussed in more detail below. The deployment of Village Law facilitators has increased stakeholders’ appreciation of PNPM Rural/Generasi-style facilitation as a tool for community-driven development (CDD) work. 6 Generasi introduced this nutrition-focused, 1,000 days of life paradigm in 2013/14. 13 Facilitation: Three Understandings In every interview, field researchers asked all Generasi stakeholders about their understanding of the word “facilitation.” The responses can be grouped into three overlapping types. I. To push, move, or stimulate communities to make proper use of basic social services: This is a vertical, or top-down, understanding of what it means to facilitate communities in their social development. Descriptions of this style of facilitation suggest that communities need encouragement to access available services. Policy prescriptions focus on incentives or disincentives to entice communities to use government social services by changing the “mindset” that prevails in rural communities. Incentives are often described as assistance (bantuan), to poor households. In many ways, this is a traditional view of rural development that does not accord with core CDD principles. II. To link communities with basic social services: This is a more horizontal understanding of facilitation that seeks to build bridges between communities and basic social services. These responses describe facilitators as engaging in advocacy and empowerment activities that involve figuring out, relaying, and sharing communities’ needs and aspirations (see the case study, “Generasi District Facilitator Advocacy in Pamekasan”, and the contrasting case study, “Facilitators Struggle to Solve a School Dropout in Gorontalo”). Rather than assuming there is a problem with a community’s mindset, this approach to facilitation serves communities by identifying and meeting its needs. III. To help correctly and efficiently administer government programs: Many respondents described facilitation as working to improve the performance of village governance, government programs, service provision, and monitoring activities. These responses describe technical, management, training, and data oriented approaches to facilitation. In all four regions of this study (including the pilot study), respondents most frequently described types I and III (the ratio of response types I:II:III was roughly 3:2:3). There are no significant differences in responses between treatment and control communities, likely because control communities gained an understanding of facilitation from the PNPM Rural program, and are now encountering new Village Law facilitators. Role of Facilitators Facilitators perform a bridging or linking role between multiple levels of government, as well as with efforts to implement the Village Law. At each level of government, facilitators have a variety of coordination, advocacy, and socialization tasks with the line agencies at their level of government and below, which familiarize stakeholders with the Generasi program and support government capacity building. At the provincial level, Generasi coordinators primarily engage in managerial tasks, troubleshooting problems and program gaps with staff across the province to find solutions. At all levels, data collection, administration, and reporting are the most important priorities, and take the most time. Less time is generally devoted to empowerment and outreach activities. Although many development programs have facilitators (including the Village Law), the Generasi model is unique because it employs two facilitators, or KPMDs, for each village who are chosen from within the community. These two factors, facilitators’ responsibility for one village and recruitment from that village, contributes to their high levels of responsiveness to community needs and ability to mobilize community members. By comparison, under the Village Law, one facilitator (pendamping lokal desa, PLD) is responsible for more than one village at a time, and is not necessarily recruited from those villages. Consequently, Village Law facilitators may be less responsive to community needs and less effective at mobilizing community members to participate in planning and oversight of the Village Law. Generasi’s end has triggered a desire to implement Generasi-style facilitation mechanisms to organize and implement village development projects, which presents an opportunity to extend its long-term outreach. 14 Case Study: Facilitators Struggle to Solve a School Dropout in Gorontalo After completing elementary school, Hayat wanted to continue to middle school but was forced to drop out to work odd jobs (such as gathering grass for livestock or picking and drying corn) to provide for his family. His mother is a single parent with four children, the youngest of whom is an infant. The Generasi FK argues that Hayat’s case is an individual household problem that is beyond the program’s authority since it is only designed to help those who attend school, and that Generasi’s role is not to facilitate the education of young people. Village officials have refused to use government funds to cover Hayat’s education costs or household expenses. His right to education continues to be overlooked. Hayat’s aunt is a new KPMD with insufficient skills and experience to advocate his case beyond the village – particularly to the District Education Office – to find an alternative solution such as homeschooling or private classes. The village’s other KPMD resigned in 2017. This case highlights the importance of enhancing advocacy skills. It is shows the FK's misunderstanding of Generasi's role and purpose. Hayat is precisely the type of child the project is supposed to target. Case Study: Generasi District Facilitator Advocacy in Pamekasan Ida Nurbaya served as an FK in the PNPM Rural program before working as a finance facilitator at the district level in the Generasi program. In 2016, during her annual internal audit of the subdistricts where the program operates in Pamekasan – which includes village visits and beneficiary visits with district facilitators – she learned about the case of Nia. Nia had to quit school at 6th grade just before her graduation exam after missing several months of school due to a stomach tumor. The Larangan FK shared her story during their routine coordination meeting with district facilitators. Ida shared the case via Facebook to raise money for the treatment. A donor offered to cover the surgery costs, and the Generasi facilitators paid for the transportation costs to the hospital. After a successful surgery, FKs convinced the Education Office to allow Nia to take the National Exam even though she missed the exam date due to illness. She passed the final exam and graduated from elementary school. Nia’s story was shared at the National Generasi Coordination Meeting as an example of successful facilitation. Such stories are not always shared due to poor documentation and exchange of information between levels and across sectors. District-level facilitators (faskabs) coordinate with district community empowerment agencies7 and the district health and education offices to ensure access for Generasi actors and to raise awareness of health and education issues. Stakeholders at this level reported frequent interactions with the faskab and other Generasi staff in Garut, Gorontalo, and Pamekasan, but not in Lembata. The Education Office secretary in Pamekasan reported that facilitators keep officials at the district level informed of conditions in rural villages. A health 7Some districts have a Community and Village Government Empowerment Agency (Badan Pemberdayaan Masyarakat dan Pemerintahan Desa, BPMPD), and others have a Village Community Empowerment Agency (Badan Pemberdayaan Masyarakat Desa, BPMD). Other district governments have merged these two community empowerment agencies. 15 officer in Pamekasan appreciated the faskab’s efforts to find out what activities the District Health Office covers in its budgets so that Generasi does not overlap in its allocations. Ideally, the faskab should spend at least 15 days each month in the subdistrict providing support to the FKs and their teams, including overseeing phases of the program cycle, ensuring targets are met, and reviewing planning documents. Pamekasan and Gorontalo faskabs engaged in productive coordination and advocacy efforts. For example, the Pamekasan faskab developed good relations and coordination with the head of the district Family Welfare Movement (Pembinaan Kesejahteraan Keluarga, PKK) (who is the bupati’s, or head of district’s, wife) to raise her awareness about nutrition in Pamekasan (see the case study, “Cross-Sectoral Coordination to Solve Malnutrition Issues in Pamekasan”). The faskab also found strategic opportunities to lobby the bupati to issue a regulation on the allocation of village funds for health and education. In Gorontalo, because of the faskab’s routine communication and advocacy work, the secretary of the District Health Office promoted the allocation of Village Law budgets for health at subdistrict puskesmas and stakeholder workshops, which kepala desas typically attend. The faskab also coordinates closely with Gorontalo’s district- level technical consultant for basic social services in Village Law implementation (known as Tenaga Ahli Pelayanan Sosial Dasar, TA-PSD), who also strongly advocates the inclusion of health and education in village budgets. However, the Lembata faskab engaged in little outreach, which he attributed to a heavy workload in his district office and to difficulties accessing the subdistricts in his district. Case Study: Cross-Sectoral Coordination to Solve Malnutrition Issues in Pamekasan Two faskabs in the Pamekasan district of Jawa Timur province conducted an internal audit by visiting the UPK and the homes of selected beneficiaries. In April 2016, they visited Larangan Luar village to see Sahlan, a five-month-old baby suffering from malnutrition who lives in an unhealthy and abusive environment. Sahlan is the youngest of three children; his father has an intellectual disability and abuses his wife, who suffers from depression as a result. He was brought to the attention of the program after a neighbor witnessed him being thrown forcefully into a bucket by his elder sister and brought him to the puskesmas. The Generasi program provided Sahlan with PMT, which was administered by his neighbor, a posyandu volunteer; there were concerns that the mother was not sterilizing bottles and that the sisters might drink the milk. The KPMD and village midwife performed the initial intervention, but cross-sectoral advocacy at the district level was required to deal with the poor housing conditions and the risk of neglect and abuse. The two faskabs maintained excellent communication with staff in the community empowerment agency, a section head of which had a very close relationship with the head of the district PKK (Ibu Ani), who was also the bupati’s wife. Through the formal coordination process, Ibu Ani learned of Sahlan’s case and personally donated funds for his family’s short-term treatment. Shortly afterward, in August 2016, Ibu Umi from the district’s Women Protection and Family Planning Body visited the family with the facilitators. Ibu Umi suggested placing the children in a shelter or foster home, which would require the parents’ consent. A proposal was also submitted to the Social Office to renovate Sahlan’s house; this proposal was under assessment at the time of writing. While coordination with the head of the district PKK is not regular, the Generasi facilitator usually tries to encourage the bupati’s wife to attend program activities in order to promote them within the community. A good relationship with the PKK chairwoman has indirectly helped speed up the handling of nutrition issues in Pamekasan, which may help extend Generasi’s influence after it ends. Attempts to involve the PKK have begun to show some positive results. The Pamekasan district PKK has encouraged the subdistrict PKK to pay more attention to malnutrition issues. In turn, the 16 head of the subdistrict PKK in the Sambingan puskesmas persuaded the village PKK to provide PMT at the posyandu. Ibu Dian, the wife of the kepala desa and the head of Tespates (Sambingan subdistrict) PKK, coordinates with the posyandu volunteers and midwives regarding the types of food supplied. She also runs a monthly session on the importance of children’s nutrition with kindergarten students and their parents. While district-level coordination in Pamekasan has not fully resolved the malnutrition problem, facilitation by Generasi actors has at least encouraged the district to promote multi-sectoral engagement on malnutrition, particularly for related issues. Through coordinated facilitation, in this case with the PKK, there are hopes that the PKK will reprise its former role in family welfare development. The PKK can encourage governments at all levels to pay more attention to malnutrition. Thus far, the emphasis has been limited to allocating funds for PMT, which is considered an important motivator for communities to participate in and access health services at the posyandu. Subdistrict-level facilitators (FKs) report that they focus on five main activities: administration, supervision, coordination, advocacy, and mediation. At this level, respondents reported frequent and productive coordination with Generasi staff. The heads of puskesmas, for example in Petis subdistrict (Pamekasan district) and Tarengge subdistrict (Gorontalo), coordinate extensively with Generasi on health services, primarily through village midwives, including efforts to prevent overlap and cross-check and validate health data. Generasi also frequently hires puskesmas staff to administer training sessions in the villages, and helps ensure that the posyandu are adequately staffed. The wide range of stakeholders interviewed for the study knew the names of at least one, if not all, of the Generasi actors in their village, but not all knew the name of the program. One of their most important administrative tasks is validating reports from KPMDs because these relate to Generasi’s scoring system and achievement of targets. FK supervision requires routine visits to the villages to ensure that the program cycle is on schedule. The FK also prepares follow-up planning documents with the KPMD, and monitors the extent to which these plans are carried out. As part of the rollout of the new Village Law, MoV has assigned FKs work that was previously done by Village Law facilitators, who experience a high turnover rate. This work involves reporting the village development index (indeks desa membangun, IDM) and village profiles. The FK in the subdistrict of Nelle (Lembata) reported that they now spend more time filling in IDM forms than supervising and facilitating Generasi activities in the villages, and are sometimes unable to attend important village planning meetings. The FK in Pamekasan reported managing this extra workload with data collection help from KPMDs. Subdistrict-level coordination succeeds when Generasi actors coordinate closely with health and education agencies and the Activity Operations Superintendent (PJOK), who is responsible for running general activities in the subdistrict. The FK attends coordination meetings held by other agencies, such as tri-monthly stakeholder workshops (minilok) held at the puskesmas. The FK also convenes monthly coordination meetings with all KPMDs (called rakor), as well as three inter-village meetings (musyawarah antar desa, or MAD). MAD meetings are considered more prestigious, and are attended by village leaders (who send delegates to minilok).8 8For more information about these meetings, see the case study, “A Generasi Inter-Village Meeting and Improving Access to Health Services”. 17 In addition to routine supervision and coordination work, the FKs also assist with outreach and empowerment in the villages. In Pamekasan they report on efforts to make village posyandu more self-reliant, without depending on external sources of funding such as Generasi or village budgets. For example, the FKs have worked on demonstrations of banana gardens and catfish ponds, which are managed by posyandu volunteers. The FK hope these agricultural activities might provide a small income to support posyandu activities in each village. The FKs in the treatment subdistricts of Gorontalo and Garut have also provided support in the villages such as assisting with the writing of village budget plans and monitoring posyandu services. Facilitators at this level may play a mediating role if there is a conflict. For example, in Pamekasan, the FK intervened with the kepala desa on a specific nutrition issue on behalf of a village midwife. KPMDs focus on four main activities: data collection, coordination, outreach, and advocacy. Data collection includes maintaining tallies of the total number of pregnant women, infants, and elementary and middle school children to conduct outreach to these groups so they are included in planning and participation in Generasi activities, with the goal of raising awareness of health and education issues and services. KPMDs reported that their success relies on their ability to collect data on Generasi’s indicators and meet targets, and that these and other administrative tasks take up most of their time, leaving insufficient time for outreach. A focus on maintaining accurate data can stimulate direct outreach to beneficiaries through home visits as they attempt to collect or verify data. For example, in Gorontalo, the KPMD visits beneficiaries to let them know about upcoming activities. During these visits, the KPMD gently applies pressure to participate in program activities. The 12 KPMDs interviewed for this study all had high school diplomas and were of at least middle-class status relative to their community, but only three or four of them showed a good understanding of the health and education situations in their villages. This subset generally had prior active experience in village affairs such as volunteering for posyandu, working as a teacher in the local schools, or serving in some capacity in village governance. This variation in knowledge of health and education greatly affects KPMDs’ performance. For example, in a treatment village in Gorontalo, the KPMD had only a cursory understanding of the village’s health and education issues, and their funded activities were standard items such as school uniforms and food supplements (for a specific example that points to limited KPMD knowledge resulting in limited solutions from this village, see the case study, “Facilitators Struggle to Solve a School Dropout in Motoduto Village”). Yet the KPMD in a Pamekasan treatment village found innovative solutions to problems solicited from target beneficiaries, such as a sturdy tripod for the posyandu scale because mothers feared that their infants would fall out of the normal scale, and a megaphone to announce posyandu schedules so that more distant households would hear. Generasi training of KPMDs appears to be more focused on completing reports and validating data than improving their understanding of health and education issues. Importance of understanding targets: The KPMD in Lelaok (Petis subdistrict, Pamekasan) understands that achieving complete immunization is one of Generasi’s targets, and that a full battery of immunizations must occur on a schedule according to the infant’s age. She helps young mothers find alternate immunization clinics if they are unable to attend the local clinic, since they are reluctant to communicate directly with the midwife. The KPMD’s knowledge and understanding of how immunizations work also reassures and educates mothers. In a treatment village in Lembata, the KPMD looks out for newly pregnant women to encourage them to visit the midwife so they can meet the Generasi pre-natal exam target. Facilitator outreach increases basic social service utilization: As integrated members of the community, KPMD facilitators work on the front line of program implementation and serve as a bridge between target beneficiaries and service providers such as midwives who are often considered “outsiders” even if they live in the community. At the village level, stakeholders such as village midwives, teachers, and leaders or administrators report frequent interactions with Generasi KPMD, Generasi activity implementers (PK) and often the FK. 18 All KPMDs encourage and remind families to bring their young children to posyandu, sometimes leveraging collaborative support from formal village leaders. Likewise, in a treatment village in Lembata, the KPMD conducts outreach to children at risk of dropping out of school. The KPMD usually receives information from community members if particular students are frequently absent from school, and often visits the parents to find out why in order to try to prevent the student from dropping out. In a few cases, the KPMD discovered that students were dropping out of school because they were ashamed that they could not pay school fees. The KPMD will also speak with school officials to find a way for the student to continue their education. School officials might also try to contact the households of students who drop out, but usually after several months have passed, when the chances of finding a solution have diminished. By comparison, in the control village in Lembata, researchers found that many students drop out of school. The village government readily admitted that it was a problem, but it does not know how exactly many children have dropped out. While village residents recognize the signs of children who are at risk of dropping out, the prevailing opinion in this community is that it is the parents’ responsibility to ensure their kids stay in school, rather than a shared public responsibility. Coordination is an important aspect of a KPMD’s work. They must organize consultation meetings with target beneficiaries in each village hamlet; convene activities with posyandu volunteers and local schools; collect and share data from posyandu, village midwives, and teachers; secure permission from and report results to village leaders on all Generasi activities; and attend coordination meetings at the subdistrict level. Researchers found that KPMD coordination with service providers is cooperative in most sampled treatment villages. There are a few exceptions. A teacher from a treatment village in Lembata pointedly remarked that only the school has the authority and the right to distribute assistance to communities because school officials know better than Generasi facilitators who should be the beneficiaries of educational assistance. The head of the Education Office in Petis subdistrict (Pamekasan district) reported that he knows little about Generasi: “I’ve only heard that PNPM [Generasi] provides assistance to students. PNPM coordinated with me when they had an activity for special needs children. They wanted to coordinate with the School Committee, so I came.” He then complained that “Generasi only asks for data, total number of students, but there is never any follow-up … we never find out what they do with it.” He could not describe Generasi’s targets or aims, and could only recall that students received school uniforms from the program. Such critiques may be the result of poor communication and coordination with Generasi, but they also suggest there are misunderstandings about how CDD programs work (from the bottom up), which involves holding service providers accountable. Facilitator Workload Distribution Case Study: Exceptional Administration Burden in Lembata district In Lembata, Generasi actors at all levels complained about the new MoV administrative requirements, which involve submitting at least 50 IDM reports each month. The former faskab for Lembata said the majority of his work involved making sure that every subdistrict completed its administrative and planning documents. The current faskab agreed, and claimed that all his time is spent completing administrative duties, leaving no time to coordinate with line agencies or the bupati’s office. The former FK for the subdistrict of Nelle believes that many of the reporting requirements are unnecessary; the current FK spends time in the villages, but mainly to ensure the KPMDs are completing their administrative tasks. While they appreciate the hiring of an assistant FK and a mentor to the FK (PenLok) to provide support, the assistant FK for Nelle asked: “how can we do our coordination and facilitation work if we are too busy delivering a variety of data requests?” The KPMDs in both treatment villages said they focus on collecting and reporting data during and 19 immediately after the posyandu and hamlet consultations, and conduct few home visits with beneficiaries. The administrative burden appears to be heavier in Lembata due in part to weaker administrative capacities and the remote island geography, which prevents the provincial team from completing site visits as frequently as is done in other provinces. They also tend to be responsible for more villages than in other subdistricts. The assignment of PenLok and assistant FK has been helpful, but reportedly not enough to meet this burden. Local politics resulted in the turnover of the faskab and FK in Lembata in 2016. Research staff also detected poor working relationships between Generasi program actors at different administrative levels, particularly between the district and province. In addition, administrative capacities may be weaker in NTT compared to other provinces. At the village level, the KPMDs from the two treatment villages in the subdistrict of Petis subdistrict (Pamekasan district) gave contrasting views of their workload distribution. In one village, the KPMD reported that few people want to work as a KPMD because the work is too difficult and time consuming (see also the case study, “Exceptional Administration Burden in Lembata”). Filling in forms and taking notes takes up most of her time, leaving little time for outreach and advocacy. Meanwhile, the KPMD in the other treatment village, who served as a posyandu volunteer for more than 20 years, said it only takes her about two days each month to complete her Generasi reports. Her prior data recording and reporting experience, and knowledge of posyandu and how the data matters, were of tremendous benefit to her current job. She also said that a good learning and collaborative relationship with the village midwife makes her job easier. Some proactive KPMDs manage to combine their data collection with advocacy, primarily through home visits with Generasi beneficiaries. Such visits allow them to complete their data collection or meet their targets, for example when parents skip posyandu or their children are missing school. Subdistrict-level informants also report a higher percentage of administrative work compared to outreach and empowerment activities. Researchers observed monthly coordination meetings, led by the FK and attended by nearly all village KPMDs, that involved teaching the KPMDs how to fill in the forms that were due soon in the annual program cycle. The former FK for the Nelle subdistrict (Lembata) reported that he visits villages 15 days each month, but that the work usually involves checking the administration and bookkeeping of the KPMDs and PKs rather than planning and monitoring village activities (see the case study, “Exceptional Administration Burden in Lembata district”). The FK in Tarengge subdistrict described a lot of administrative work as well, but did not complain because she said her team members (including the Generasi PenLok and Kelompok Kerja (PokJa, Working Group) staff, the UPK staff, and the KPMDs) each completed and delivered their tasks, which makes her job much easier. Each treatment subdistrict had a PenLok on the Generasi team, and only the subdistrict of Nelle (Lembata) had an assistant FK. Each Generasi subdistrict team described different divisions of labor, but every FK expressed gratitude for the support. For example, in Nelle, the former FK described dividing up the villages between the FK, PenLok, and an assistant FK (but with final supervisory authority with the FK), whereas in Tarengge subdistrict the FK tracked the finances and the PenLok tracked the program cycle. In each subdistrict, researchers observed good working relationships between the FK, PenLok, assistant FK, and UPK staff. At the district level, the faskab for Lembata reported that he had no time for coordination or empowerment work, and spent all his time on administrative matters (see the case study, “Exceptional Administration Burden in Lembata district”). In Pamekasan and Gorontalo, respondents were less absolute but also unwilling to estimate how much time was spent on various tasks. The faskabs play an important role performing outreach and coordination with district-level leaders and officials, which can result in significant political buy- in for the program (see case study, “Cross-Sectoral Coordination to Solve Malnutrition Issues in 20 Pamekasan”). The Pamekasan faskab and finance facilitator described how they incorporate empowerment and advocacy work into their administrative work routines by conducting site visits as part of their internal audit procedures, which allow them to observe program activities and meet beneficiaries. When Village Law facilitators were asked how their workload is allocated between administrative and empowerment tasks, it was clear that the district-level TA-PSDs are far more concerned with advocacy and outreach than the village-level facilitators (see the case study, “A Village Law Facilitator in Rampe Village, Lembata District”). The TA-PSDs described various socialization activities with village leaders and district officials regarding the importance of including health and education activities in village budgets. Pamekasan’s Technical Assistants for Basic Social Services at the district level (Tenaga Ahli Pelayanan Sosial Dasar, TA-PSD) described how difficult it is to persuade village leaders to allocate their budgets to basic social services when they prefer physical infrastructure that serves as a tangible signal of their contribution to the village. He also lobbies the bupati and other district officials to implement a mandate to include basic social services in village budgets. A Village Law facilitator9 (PLD) in Tarengge subdistrict admitted that “compared with the Generasi KPMD, the work of the PLD is much lighter, not complicated… we only have to ask for data.” All village facilitators (pendamping desa, PDs) and PLDs in all three districts conceded that they work with the village governments, but not the beneficiaries of government services; they deliver documents and remind village governments to complete their paperwork, but do not get actively involved in the planning process. Case Study: A Village Law Facilitator in Rampe Village, Lembata District Almost all kepala desas in Nelle subdistrict complained about the ill-defined, minimum role of PLDs in the Development and Empowerment of Village Communities Program (P3MD), also known as Village Funds. The kepala desas expect PLDs to be familiar with the technical aspects of planning as well as the legal and administrative requirements of utilizing the village funds. However, P3MD facilitators do not yet have technical regulations to guide their work in the villages. Over the past year, they have been left to define their own tasks by interpreting Village Law No. 6/2014. Many cannot name the program they are working on. In multiple villages in Nelle, Generasi KPMDs are covering the PLD roles in supporting their respective village governments. Generasi facilitators are generally acknowledged in the villages as experts on education and health issues. The village governments allow them to influence the dynamics of village development policy, which could help promote the integration of Generasi basic social services (Pelayanan Sosial Dasar, PSD) issues into regular village planning, which now falls under the P3MD program. Case Study: Generasi Facilitator and Kepala Desa Hold Health Service Provider Accountable The treatment village of Sogiyan (Petis subdistrict, Pamekasan district) held its midwife accountable after she charged a national health insurance cardholder for delivery fees. Bidan (midwife) Intan charged Ibu Lalan for her treatment at the polindes (village maternity post), which was supposed to be free. Her family complained to the KPMD, who privately conveyed these concerns to the kepala desa, a former posyandu volunteer, who followed up with the head of the puskesmas. Intan was asked to return the money to Ibu Lalan’s family. The incident eventually led to her replacement by Bidan Wiwid, who previously served as a midwife in the puskemas. 9 PD and PLD are facilitators recruited and funded by MoV. They are not Generasi program actors. The PD works at the subdistrict level, while the PLD works at the village level. 21 The KPMD, traditional birth attendant, kepala desa, and Generasi beneficiaries all reported that Wiwid’s work is better than that of the previous midwife. The community members prefer Wiwid because she is more patient, friendlier, politer, and charges less for medicines. Unlike Intan, Wiwid is available to her patients: she always attends every posyandu and otherwise stands by at the polindes (where she lives with her husband) except for her weekly visits home. The kepala desa monitors the new midwife’s performance to ensure the community receives quality service, and will request a replacement if it does not. Information Sharing and Learning Among Facilitators Formal Generasi program mechanisms include a variety of opportunities for facilitators at all levels to share information and learn from each other, including monthly coordination meetings and routine training activities. Many Generasi stakeholders referred to the program’s annual training activities as an example of the mutual learning among facilitators, especially KPMDs. Generasi actors in the villages, especially KPMDs, agree that their administration and data collection provide the best learning opportunities. For example, the two KPMDs in Mangkawani village (Tarengge subdistrict, Gorontalo) described how the longer-serving, more experienced KPMD teaches the newer KPMD about data collection and administration on the job. The more often a facilitator convenes meetings and works with the community to access health and education services, the more they improve their public speaking and meeting management skills, as well as their capacity to deliver the program. For example, when a KPMD makes a home visit, they hear stories about whether and how Generasi beneficiaries access basic social services, including their barriers to access as well as their suggestions. Learning also occurs on both sides through coordination and collaboration between Generasi actors and service providers. For example, Generasi facilitators in Pamekasan learned about the nutritional values of various foods when they asked the nutrition officer at the puskesmas for advice about food supplements. Generasi facilitators also organize and support health classes and counseling sessions (typically at a posyandu, or special pregnancy classes), working with service providers to deliver the training, such as the nutrition officer from the puskesmas. Generasi facilitators at the subdistrict and district levels use social media tools such as WhatsApp and Facebook. There is little social media use in Lembata, due to limited data access and user capacity, while Generasi actors in Gorontalo and Pamekasan frequently use social media to access a variety of information in different ways. In Gorontalo, Generasi actors use a closed Facebook group to coordinate activities, share invitations to meetings, and inform facilitators about the payment of their salaries. In Pamekasan, Generasi facilitators run a Facebook account (not a group) that links Generasi actors with the community to share news about program activities and to build a network with target beneficiaries and provide widespread support. Generasi actors in Pamekasan use WhatsApp to internally coordinate activities and meeting invitations. Village-level actors mainly share information through the FK’s routine coordination work as well as SMS and telephone calls. The UPK office is a frequent meeting point between KPMD sand FKs, or even among KPMDs from different villages, to coordinate and share information. Perceptions of (and Interactions with) Village Law Facilitators While most residents knew their Generasi facilitators, at the time of this data collection which occurred in 2016, very few respondents knew who the Village Law PDs were in their communities. This is likely due to two reasons. First, the Village Law was only recently passed, and the terms and scale of facilitators’ work are still being finalized. Second, Village Law facilitators must cover at least three villages, and they are not 22 typically members of the communities in which they work. Generasi has facilitators at the village, subdistrict, and district levels, so respondents are far more likely to interact with Generasi actors than Village Law actors, and to know them personally. Village Law facilitators may struggle to understand a community’s challenges and to gain the trust of its members, which is essential for effective facilitation. They are more focused than their Generasi counterparts on collecting village-level data for MoV. There is also significant coordination between Village Law and Generasi facilitators at the district level. TA- PSDs often coordinate with Generasi faskabs and are sometimes former Generasi facilitators. For example, the TA-PSD for Gorontalo attends Generasi coordination meetings, and the Gorontalo faskab attends Village Law coordination meetings at the district level. Likewise, in Pamekasan the TA-PSD and faskab coordinate their efforts to introduce health and education into village budgets; such overlap extends Generasi’s reach into non-participating villages. At the subdistrict and village levels, there was significantly less coordination and uneven interactions with Village Law facilitators. Residents often complained about the absence of PDs and their lack of qualifications and skills. Most stakeholders, including some village officials, do not have a clear sense of PDs’ role and responsibilities. Many local residents believe PD recruitment is a tool of political patronage. Kepala desas in Lembata’s treatment and control villages found KPMDs to be better trained and more informative than their counterparts in Village Law implementation. In Gorontalo, although the Generasi FK shares office space with Village Law facilitators and communicates with them a few times per month (usually by telephone, SMS, or Facebook), she rarely meets them in person. Facilitators’ Independence from Government The level of facilitators’ independence from government varied by location. In Lembata, informants reported that facilitator independence is maintained at all levels of government. From subdistrict up to provincial levels, facilitators characterize their relationships with the government as involving limited coordination. At coordination meetings with government agencies, government officials provide recommendations about what activities they think need support, but they do not intervene directly. For example, the Generasi provincial coordinator for NTT has refused requests from various line agencies to disburse Generasi funds through the BPMPD, explaining that fulfilling their requests would violate Generasi’s standard operating procedures and the program’s independence from the government. In Gorontalo and especially Pamekasan, facilitator dynamics depend on local village politics. Rather than selecting KPMDs through the formal consultation mechanisms prescribed by Generasi procedures, in Pamekasan they are in practice chosen by the kepala desa. These KPMDs represent the head rather than Generasi beneficiaries, and view their facilitation work for Generasi as an extension of the head’s governing mandate, which reinforces the perception that Generasi is a top-down bantuan program associated with the kepala desa’s personal prestige: “The kepala desa also requested that [Generasi’s] assistance (bantuan) does not cause a disturbance, nor any bad words directed towards him. He asked that if possible, everyone should receive assistance because it will have an impact on his electoral success.” – PK in a Generasi treatment village, Pamekasan district However, Generasi facilitators have more bargaining power than TA-PSDs, who are subordinate to the kepala desa in Pamekasan. KPMDs can still advocate in order to convene consultation sessions in keeping with Generasi’s operating procedures. The kepala desa cannot intervene too much in the selection of Generasi activities. By contrast, Village Law facilitators there see themselves as information providers and collectors, who are constrained to administering activities. According to a district TA-PSD, “We have to be proactive, 23 but when it comes to implementation in the villages we can’t fully deliver the programs that the agencies asked us to bring because the kepala desa has the authority to decide whether or not to include it.” Case Study: Replacing a Generasi Facilitator in Lelaok Village, Pamekasan In the treatment village of Lelaok (Petis subdistrict), the new kepala desa elected in 2015 replaced the long-serving KPMD with her own daughter without following proper community consultation and deliberation procedures. Political dynamics in the village were the primary factor behind the replacement: the new head wanted to appoint someone loyal to her. The Generasi program was thus forced to accept the village government’s decision to replace the KPMD with someone much less qualified for the role. The previous KPMD, Hajjah Danisa, has been a posyandu volunteer since 2006, and she continues to in this role. The community members trust her, and the village midwife acknowledged her initiative in organizing health-related activities in the village (such as immunizations, weighing, and pregnant mother visits to poskesdes (village health post)) and supported keeping her staffed as a KPMD. The new kepala desa has demonstrated a narrow interpretation of the Generasi program in general, and the role of KPMDs in particular. The new KPMD does not perform the tasks that Hajjah Danisa usually performed, such as hamlet-level consultations to identify new ideas. Subairi, the other new KPMD (also a relative of the kepala desa) is rarely present since he works in Java. Therefore, the kepala desa is running KPMD tasks. The main KPMD, who also serves as the village secretary, has been tasked with preparing PMT for the posyandu. Her husband is the new Generasi PK. Although PMT is a vital element to encourage villagers’ participation in posyandu, its provision represents only one part of the KPMD’s tasks in the village. Other tasks include socialization and data collection. The program seems to have failed in informing the village government, especially the kepala desa, of its scope. According to the kepala desa, the process of selecting activities for 2016 was quite simple: the villagers told the posyandu volunteers what kinds of PMT to provide, such as types of juices and snacks, and she will prepare and deliver them. At the time of the fieldwork, Hajjah Danisa’s absence had not directly affected village health services. The midwife and four nurses have been able to cover the implementation of health services, and the four posyandu were still running. This case demonstrates how village government capacity and local politics may influence facilitators’ performance. The presence of the kepala desa at the posyandu may indeed encourage villagers’ participation, but rather than perform the tasks of the volunteers, she should be supporting the overall progress of the program in the village. Village governments seem to value social prestige over development outcomes: a kepala desa is deemed successful when he or she helps an individual villager, not necessarily when they aim to develop the entire community. They also prefer to perform such tasks themselves, as a matter of pride. The village government under this new leadership was not able to articulate its vision for strengthening health services, particularly by allocating village funds to this area. Although she vaguely praised Generasi’s provision of PMT, the new head has made no effort to use KPMD instruments, such as the recording forms, working steps, or even considering how a role like the KPMD can be developed in the village to support access to basic social services. Instead, she is waiting for instructions from the district government. 24 For facilitation in villages like Lelaok, Generasi must pay attention to local political dynamics, such as kepala desa elections, which have proven to heavily impact the KPMD role. Although Generasi actors at the subdistrict level have urged the Lelaok village government to maintain Hajjah Danisa as KPMD, the program must prepare alternative KPMDs in case such events happen in the future. Participation Generasi promotes community participation, especially of the poor and women, in the planning and implementation of Generasi activities, monitoring of service providers, and the utilization of health and education services. This section addresses three questions related to participation. First, to what extent do communities meaningfully participate in the planning and prioritization of Generasi activities and the monitoring of service providers? Second, in what ways do the 12 targets stimulate or constrain community participation? Third, has participation helped achieve the program’s outcomes? The qualitative data point to a gap between the policy and practice of participation. For example, the PK in Desa Maroangin (Tarengge subdistrict, Gorontalo) described participation as datang, duduk, dengar (come, sit, listen). By understanding participation as simply attendance, local elites, program actors, and beneficiaries perceive Generasi’s work as a form of bantuan for the poor, which is contrary to its stated goal of empowerment. All villages surveyed in this research have formal and informal participatory mechanisms for a variety of village activities. Control villages rely more on consultations with formal and informal leaders in community planning meetings, whereas treatment villages are required to include program beneficiaries and stakeholders in these meetings. Outside the formal participation mechanisms, all communities including those in the control group have different ways of providing oversight of service delivery, although these are not always effective. Generasi beneficiaries are more likely to attend program activities when village leaders invite and encourage them to attend. They attend as a sign of respect for their leaders, and/or to ensure their leaders’ support when they require assistance. In most treatment villages, community participation was higher in hamlet-level meetings than in village ones, partly due to the time and costs involved in travelling to villages’ headquarters. Further, participants felt that they had a more direct impact on decision making at the hamlet than the village level, which is discussed in more detail below. Generasi actors and beneficiaries consistently describe participation as attendance at activities, especially planning meetings, and utilization of services including the posyandu. In some cases, village governments used sanctions to ensure that community members utilized Generasi-funded services. While facilitators’ encouragement of community members to participate in Generasi activities contributed to achieving the 12 targets, the program did little to empower community members to influence village governance and service delivery beyond prioritizing Generasi block grants.10 Perceptions of Participation Participatory processes are used in all four phases of Generasi’s annual program cycle: socialization, planning, implementation of activities (e.g., supplementary feeding at posyandu, nutrition classes) and monitoring the implementation of the activities. Project socialization is conducted in subdistrict, village, and hamlet-level meetings and small discussion groups. Throughout this stage, facilitators introduce the project objectives, the 12 indicators, the welfare benefits of achieving these indicators, and project rules and principles. More 10 The quantitative IE analysis did not find any spillover effects of Generasi on participation in community meetings. 25 ordinary community members participate in meetings at the hamlet and village levels compared to the subdistrict level. In the village planning stage, KPMDs invite all target beneficiaries to a hamlet-level meeting for a brainstorming session, or visit target beneficiaries’ houses if attendance is low. Respondents cited this phase as the most effective in eliciting community participation in the program cycle. These ideas form the basis of the village-level planning phase: the KPMD opens the consultation meeting with the results of these brainstorming sessions. Yet higher-level officials such as the FK or the FK’s mentor, PL, often usurp the communities’ power to define the program’s concrete activities. For example, the nutrition officer in Petis subdistrict decides Generasi’s PMT menu for the entire subdistrict. Likewise, the KPMD for Maroangin village (Tarengge subdistrict, Gorontalo) reported that Generasi actors cross-check proposed program activities against health and education service providers’ plans before finalizing program plans to avoid overlap. Generasi actors and beneficiaries appear to reinforce the notion that participation is a formality that entails mere attendance. For instance, a program beneficiary from the treatment village of Ilekora (Nelle subdistrict, Lembata) interprets his obligation to participate as attending all Generasi meetings and using the program funds as instructed (to pay for his children’s school transportation costs), and feels it would violate a social norm to speak out at these meetings. In some places, speaking up directly would be interpreted as a lack of confidence in the village leadership, because Generasi assistance is understood to come from them. A PL in the subdistrict of Sambingan explains: “People here believe that to criticize village government is an act of rebellion, disturbance, treachery, and other negative connotations ... Instead, community members share their concerns or requests with the hamlet leader, and they do not directly object or criticize the village government.” In Sogiyan (Petis subdistrict, Pamekasan), beneficiaries could only recall participating in meetings at which they collectively decide who will be added to the list of Generasi beneficiaries for that year’s program cycle; they did not know they had an opportunity to participate more generally. It is possible that Generasi facilitators do not actively try to elicit participation through project meetings as they are aware that social norms constrain that type of interaction. Community participation and engagement takes place in other ways. Village leaders, program volunteers, and frontline service providers (such as teachers or village midwives) usually share ordinary citizens’ ideas, suggestions, or questions on their behalf at program meetings. Program actors in some areas have developed workaround solutions to gather community input, for example by interviewing target beneficiaries individually at the posyandu, or exchanging information through social media. The program may thus be inadvertently contributing to the perception that its activities represent bantuan rather than attempts to empower communities to take charge of their own planning and monitoring of service provision. Encouraging Participation Communities play a central role in deciding how to allocate their block grants. Trained facilitators help each village select an 11-member TPMD and KPMDs. The KPMDs and TPMD then organize participatory decision-making meetings at the hamlet (sub-village) and village levels. At the hamlet level, the facilitators hold focus group discussions with target groups to identify: the main problems in the community, the types of activities that can help to address the problems, and community members who are not benefiting from the activities. Following the announcement of block grant allocations for each village at the MAD meeting, a village-level public consultation is held to establish and decide on a list of activities and activity implementers (PKs) to be funded by Generasi. Public village-level accountability meetings occur at least twice a year to discuss the outcomes and financial reports of all Generasi activities. Similarly, Village Law facilitators check whether the village governments are organizing their planning activities in line with participatory principles. For example, the PLD in Mangkawani village (Tarengge 26 subdistrict, Pamekasan district) provides guidance to the village government, and translates the Village Law’s planning guidelines into technical requirements, such as including all elements of the community in all planning phases. Community leaders such as religious leaders and kepala desas help stimulate community participation in Generasi activities, particularly if they understand the community’s health and education issues, and how Generasi and other CDD programs work. The two treatment villages in Pamekasan provide contrasting examples. Sogiyan’s kepala desa actively encourages the community to participate in Generasi activities because he understands health issues; he previously served as a posyandu volunteer. His counterpart in Lelaok (Petis subdistrict, Pamekasan) does far less to encourage participation, as she knows little about health issues; she replaced the long-serving KPMD with her daughter, as described in the case study, “Replacing a Generasi Facilitator in Lelaok Village, Pamekasan.” KPMDs invite beneficiaries to meetings or activities either through personal interactions or by making announcements via text message, loudspeaker, or letters. In Gorontalo, villagers prefer to be directly invited by their leaders because it is considered a sign of respect. In Lembata and Pamekasan, it is usually enough to publicize activities to the community by loudspeaker. Rampe village (Nelle subdistrict, Lembata) fines residents who do not come (or come late) to posyandu based on an agreement between posyandu volunteers and the village government. In Gorontalo, fines are also used to compel pregnant women to use health facilities for pre-natal exams and deliveries. The midwife in the control village of Jaton (Telogojoyo subdistrict, Gorontalo) (reported that the threat of fines is effective at persuading pregnant women to use accredited health facilities. She also threatens to report mothers who refuse to deliver their infants at the puskesmas to the kepala desa. Leaders sometimes discourage participation. For example, religious leaders in Petis subdistrict, in Pamekasan, prohibited their followers from immunizing infants after they heard an untrue rumor that vaccines contain pig products. As a result, many members of the community refused to immunize their children. The head of the puskesmas in Petis subdistrict now recommends conducting outreach to religious leaders before introducing a health program into the community. Community Monitoring of Basic Social Services and Village Governance The program guidelines hold community members responsible for monitoring program activities. In the village consultation forum, the community may choose to form special groups to voluntarily monitor activities in the interest of the village community. Facilitators are also tasked with encouraging community members to actively participate in and benefit from the activities, which should help ensure that the village meets the 12 targets. This research yielded several interesting examples of ways that communities monitor basic social services in their villages. In a Generasi treatment village in Lembata, village and community authorities hold annual evaluation meetings for posyandu volunteers that beneficiaries also attend. The volunteers’ performance is assessed, and inactive volunteers are usually replaced. The Lembata control village of Mogiye (Tanabola subdistrict), by contrast, has never evaluated its posyandu volunteers’ performance. Only six of its 21 volunteers were active; inactive volunteers still receive incentives from the village and have not been replaced. Outside the posyandu, there is little community monitoring and evaluation of health services (see the case study, “Generasi Facilitator and Kepala Desa Hold Health Service Provider Accountable” for a positive example of community monitoring of a health service provider). Even if they receive poor service, citizens often choose not to report it, sometimes due to fears that their family will suffer: 27 “I am afraid that reporting will come back to haunt us. We might face difficulties when we complain, everything will become more complicated for us. That is why we remain silent. If they only give me complications, that’s fine, but what about my kids and family? I am afraid they would be ‘marked’” — Generasi Beneficiary, Cikereteg village, Lebak Siwur subdistrict, Garut district One young mother in Maroangin Village (Tarengge subdistrict, Gorontalo) uses the more expensive private clinic instead of the puskesmas rather than complain about slow service. Community members often feel they cannot communicate their health concerns to the puskesmas staff because of administrative and technical language barriers. “I don’t want to waste my time on administrative forms. Registering this and that, I have a lot of work at home, apart from that I feel clumsy to communicate with them… that’s what makes me unable to complain, what I have been through with those people (health employee), they are so stiff.” — Generasi Beneficiary, Maroangin Village, Tarengge subdistrict, Gorontalo district In the Sambingan subdistrict in Pamekasan, residents of control villages heard about Generasi activities in treatment areas and demanded similar activities in their villages, which were implemented on a cost-sharing basis between village governments and BPMPD at the district level. There is some evidence of communities monitoring education services in treatment villages. In Ilekora village, (Nelle subdistrict, Lembata), the community complained about increased school committee fees to the KPMD. The KPMD conveyed this concern to the school and Generasi subdistrict coordinator, but there was no response from the education subdistrict office. The community then suggested covering the cost of these fees from the Generasi block grants. Communities also monitor education in their village by reporting truant students to schoolteachers, as researchers found in Mangkawani village (Gorontalo). One legacy of the PNPM Rural program is the use of public information signboards that document agreed- upon projects, their progress, and budgets. In Village Law implementation, this remains an important tool that helps communities monitor activities and ensure that the implementation matches the original plans as illustrated in the case study, “Community-Based “Pressure Groups” Monitor Village-Level Development.” In the Lembata treatment village of Rampe (Nelle subdistrict), Village Law funds were used to build a concrete road, but the result was not consistent with the specifications in the planning documents. The community protested by refusing to participate in village labor activities to build it. “If the community is not involved in a meeting, it would be difficult to mobilize them to work; they will say do it yourself, because you chose this activity, not us. It is very hard to invite them to work.” — Head of the village council (Badan Perwakilan Desa, BPD), Rampe village, Nelle subdistrict, Lembata district In Lembata, Generasi facilitators posted lists of all the pregnant women in the village along with their progress and health status. Likewise, when health and Generasi officials conduct immunizations or pregnancy exams, they ask whether there is anyone else in the neighborhood who might not have immunized their infant or had a pre-natal exam. They report these results to the Generasi facilitator, posyandu volunteer, or village midwife. Case Study: Community-Based “Pressure Groups” Monitor Village-Level Development The control village of Mogiye (Tanabola subdistrict, Lembata) has multiple community “pressure groups” (kelompok penekan) that have formed a local civil society organization to monitor whether the village government is fulfilling its development agreements with the community. The members of this group are young, unemployed adults who have completed their higher education and 28 returned home to their village to search for employment. They direct their protests directly to the kepala desa. For example, when the village government supported the construction of a concrete road, the members of this group requested a picture of the project design to monitor whether it was proceeding according to the specifications, and to ensure no stakeholders were siphoning off personal profits. The kepala desa was pressured to agree that all infrastructure projects in the village must be accompanied by publicly posted project design images. “So now I request designs for all our development projects. That seems fair. So that if there is ever a suspicion from the community, we have a shared reference point. It's actually not that hard, does not require advanced technical skills. We just need to know, for example, the width of the road, how high, etc., to include in the design.” — Kepala desa, Mogiye Village, Tanabola subdistrict, Lembata district The kepala desa does not automatically see these groups as a threat to his government; he has tried to get to know them better and include them in village planning. Now these groups no longer spontaneously protest and there are spaces for dialogues organized by the village government. Sogiyan, a treatment village in Petis subdistrict, Pamekasan, has also experienced the formation of pressure groups. An organized group that explicitly referred to itself as an LSM (NGO) has monitored the village’s development activities. But unlike the kepala desa in Mogiye (Tanabola subdistrict, Lembata), Sogiyan’s kepala desa sees these groups as a nuisance. In his opinion, when the LSM asks questions about village development, it is looking for opportunities to threaten village programs. He worries this information will create suspicion and cause problems in the community. “Some people are frightened of these LSM, but I’m not, as long as we continue to follow the regulations. What matters is that we continue to share information with the public. If we don’t, then it means we are not following the regulations!” — Kepala desa, Sogiyan Village, Petis subdistrict, Pamekasan district Community Contributions towards Stimulating Basic Social Service Utilization Posyandu volunteers play an important role in encouraging communities to use health services. They interact with and link other actors such as the village midwife, Generasi facilitators, government officials, community leaders, and families with young children. The head of the puskesmas in Petis subdistrict, Pamekasan, explained that even though the village midwife is the knowledgeable health official in a village, she is still often seen as an outsider without a strong social position in the community. Since posyandu volunteers are from (and live in) the village where they work, their collaboration with the midwife is a crucial link for conducting outreach to and monitoring beneficiary households. The volunteers also help the village midwife mobilize participation in the monthly posyandu activities. The volunteers publicize the posyandu schedule and encourage families with children under five to attend. In turn, Generasi supports the posyandu volunteers, for example, by providing transportation support to volunteers who meet the first trimester exam target. Treatment villages have many more active volunteers involved in posyandu activities than control villages. For example, in the treatment village of Ilekora in Lembata, 16 of 20 registered volunteers were active at the time of the fieldwork, compared to five out of 20 in the control village of Mogiye (Tanabola subdistrict, Lembata) in the same district. Similar differences between treatment and control villages were also found in Pamekasan and Gorontalo, which had only two to five active volunteers in control villages. 29 This difference can be explained by two main factors. First, the incentive payments that Generasi offers to posyandu volunteers are not seen as a wage, but rather as a sign of appreciation from the government. Second, there are more training opportunities available for posyandu cadres in treatment villages. Control villages have more limited funds for training, so usually only one or two volunteers can attend each year, with support from the puskesmas clinic. The same volunteers generally attend every year and hold a higher status than those who have not been invited. In the treatment villages, Generasi provides support for a variety of trainings for all posyandu volunteers in addition to the routine annual trainings provided by the puskesmas. Many posyandu volunteers say there has been increased awareness among target beneficiaries about using health services, indicated by increased participation in posyandu activities and a reduction in home visits. The KPMD in Sogiyan village (Petis subdistrict, Pamekasan district) observed, “There are a lot who come to posyandu on their own, and I no longer have to invite them.” Many informants credit Generasi with increasing participation in health services, particularly posyandu, due to the provision of food supplements. “Food supplements are really important because without them community participation in posyandu would decrease drastically. The participants always ask about the snacks. That’s why I feel that Generasi has greatly assisted me through its provision of food supplements at posyandu.” — Village Midwife, Sogiyan Village, Petis subdistrict, Pamekasan district Informants did not offer criteria to measure “increased awareness” other than their increased attendance at program meetings, activities, and school. Since the number of children who drop out of school in a village reflects Generasi’s failure to reach its school attendance targets, KPMDs require parents to “participate” in the program by ensuring their children attend school. Markers of increased awareness may also correlate with decreased participation in other activities such as using traditional birth attendants (TBAs). In Lembata and Pamekasan, in addition to using public health facilities, some members of the community also use the services of TBAs, typically for massage during and immediately after pregnancy. TBAs also often occupy a respected social position in the community, so midwives, who may be young or newly placed in the village, rely on them to encourage community members to use health facilities for pregnancy exams and deliveries. In Pamekasan, a TBA in Lelaok village (Petis subdistrict) reported that she is frequently called to the homes of pregnant women to give massages, and she advises pregnant women to attend pregnancy exams if they have any complaints that the TBA is unable to handle. The TBA will often accompany laboring mothers to the midwife and stay with them during the delivery; the midwife often gives the TBA a small cash incentive for these referrals and support. Are Health and Education Shared Concerns in the Community? Most respondents reported that health and education are shared concerns in the community that require shared attention, for example via district and subdistrict programs to increase community utilization of basic services. In Gorontalo, for example, the bupati formed a cross-sectoral “rapid response” team coordinated via a WhatsApp group to ensure that pregnant women receive health care at accredited health facilities. There were conflicting accounts of the extent to which village governments promote basic services – both informally and in their village development plans, which compete with infrastructure projects for resources. In Sogiyan village (Petis subdistrict, Pamekasan), the village government has not included health and education activities in its planning budgets, but the kepala desa (a former posyandu volunteer) encourages residents to access health services. Health appears to be more of a shared concern in the villages than education. This can be seen in the high levels of concern shown among neighbors when health concerns arise. In Pamekasan, the kepala desa of Sogiyan (Petis subdistrict) proudly explained how residents visit the hospital in large groups when a neighbor falls sick or gives birth. In the control village of Tespates (Sambingan subdistrict, Pamekasan), the kepala desa’s 30 wife reported that neighbors often remind one another to attend upcoming posyandu activities; the village midwife agreed that it is not too difficult to reach her target beneficiaries. In Gorontalo, in Mangkawani village, many respondents described a case in which a neighbor used his motorized pedicab to bring a laboring woman to the city hospital. When the woman miscarried during the trip to the hospital, it focused the community’s attention on Mangkawani’s difficult access to adequate health services, triggering demands to reopen the village clinic and bring a midwife to live in the community. In Lembata, in the control village of Mogiye (Tanabola subdistrict), a mother who recently gave birth reported that the shared savings for childbirth organized by the village worked well for her. Having borrowed from the fund, she is slowly paying back the small loan by taking on additional farm work. She feels that it is her obligation to return the funding because she knows that another mother will use the money for childbirth expenses. In Lembata, education has become a shared concern. The head of the Ilekora (Lembata) BPD reported that children rarely skip school in the village; residents immediately report truant children to their parents. Yet in Gorontalo, residents in the sampled villages tended to see education as an individual household responsibility. When children skip school in Mangkawani village (Gorontalo) it is not considered a serious problem (for two examples, see the case studies, “The Limits of Generasi’s Facilitation and Participation Mechanisms in Preventing School Dropouts” and “Facilitators Struggle to Solve a School Dropout in Mangkawani Village, Tarengge subdistrict, Gorontalo district”). Many neighbors might not even realize if children in their community are skipping school, much less understand the reasons why. Case study: The Limits of Generasi’s Facilitation and Participation Mechanisms in Preventing School Dropouts Ihsan, a 16-year-old Generasi beneficiary in Maroangin village (Tarengge subdistrict, Gorontalo) was still registered as a student in grade 6 but has dropped out. He appears to be malnourished and his family lives in poverty. He is the oldest of four children and often works to gather coconuts to sell and takes care of his siblings while his parents are away working as farm laborers. The village PK said Ihsan wants to work, and that attempts to help him return to school have been unsuccessful. Ihsan said he does not like going to school because his peers often insult, bully, and beat him. His family is not supportive: his father would prefer him to quit school. The research team asked him several times if he wanted to go back to school, and he consistently answered that he wants to continue his education. Although more comprehensive outreach may be required, Generasi educational assistance in this area is limited to providing uniforms and transport assistance. Maroangin’s only active KPMD also works as the village secretary, so he engages in limited Generasi outreach. Generasi actors in Maroangin reported that the community is motivated to participate based on a desire to receive tangible benefits such as uniforms rather than to identify and analyze deeper social and educational problems. Participatory Budgeting Processes (Including for Generasi Block Grants) The TPMD analyzes the list of proposals at the village level, taking into account input from service providers, and at the MAD meetings, where the budget is allocated to the top-ranked activity proposals. The number of beneficiaries is used to determine how much to allocate for each activity: “When we recap our figures in the village, we cross-check against the actual number of beneficiaries. If they [KPMD and kepala desa] realize – ‘Hey, wait! We have another 31 pregnancy here. It’s not three, but four!’ – we can fix it right then and there. – Generasi FK, Tarengge subdistrict, Gorontalo district Purchases that relate to Generasi’s targets and can be directly used by the community are prioritized. “Once the village midwife asked for a stethoscope, however, Generasi rejected that one because budget allocations for instruments must be for use by the posyandu volunteers and not just the midwife… she also requested a tool for doing hemoglobin tests, but this was also rejected because the volunteers can’t use it.” – Generasi KPMD, Cikereteg village, Lebak Siwur subdistrict, Garut district Beneficiaries’ Motivation to Participate Generasi beneficiaries’ motivation to participate in the planning, supervision, and evaluation of village development activities varied. Seasonal sugar cane workers and factory workers in Tarengge subdistrict (Gorontalo) are unable to attend meetings due to their work schedules. Respondents in Pamekasan complained about the time commitment involved in attending Generasi meetings, and leaders acknowledged the difficulties of securing attendance. In Lembata, a PAUD teacher in one of the treatment villages in Nelle subdistrict who is a Generasi beneficiary reported that she is motivated to participate in Generasi meetings because she is invited to them and receives incentive support from Generasi for her work. She said she does not attend village government planning meetings because she is never invited, which suggests the importance of personal recognition from village leaders for some participants. In the control village of Mogiye (Tanabola subdistrict), in Lembata, despite the kepala desa’s efforts to maintain the PNPM program’s legacy of participation and inclusion after its completion in 2015, posyandu volunteers no longer felt the need to participate in the planning, monitoring, and evaluation of village development because these roles had been taken over by the village government. Apart from the monitoring of posyandu and distribution of Generasi assistance, there is little community oversight of service delivery, especially outside the villages. Direct contact with service providers seems to depend on the level and quality of interaction between service providers and Generasi actors. Problems with service providers are more likely to occur with schools. This may be because 10 out of 12 target indicators are health related, and KPMDs spend more of their time coordinating with the health sector (village midwives, mostly) than with the schools. In addition, midwives are frequently based in the village posyandu, and clinics are sometimes based in the village, whereas teachers and schools (apart from PAUD) have few opportunities to coordinate directly with village actors or residents. Impact on Village Law Implementation The Village Law increased villages’ authority in both village governance and in the allocation of greatly increased village funds. This study found that, as with Generasi, participatory processes in village fund allocation are most effective in the planning stages – i.e., soliciting proposals from the hamlets, prioritizing them at the village level, and ranking the proposals for inclusion in mid-term village development plans and annual village development plans (RPJMDes and RKPDes, respectively). According to the PL for Sogiyan (Petis subdistrict, Pamekasan district), community participation in Village Law-funded projects is most visible in infrastructure development activities such as roads and irrigation canals. The head of the control village of Mogiye (Tanabola subdistrict, Lembata) said he relies on village labor (gotong royong) to implement Village Law projects because it guarantees faster results, but acknowledged that “If the project does not meet their needs, the community will not participate.” 32 Politics influence the village planning meetings in every region (see, for example, the case study, “Replacing a Generasi Facilitator in Lelaok Village, Pamekasan”). For example, two PDs from the control subdistrict of Sambingan (Pamekasan) reported that everyone in such meetings supports the kepala desa, who invited them; supporters of the head’s political opponents are not included, which precludes dynamic debates and exchanges of ideas. Targets and Performance Bonuses Generasi focuses on 12 indicators of maternal and child health behavior and educational behavior. These indicators apply to those seeking health and educational services that are within the direct control of villagers – such as the number of children who receive immunizations, pre- and post-natal care, and the number of children enrolled and attending school – rather than long-term outcomes, such as test scores or infant mortality. As school enrollment rates improved significantly in Indonesia over the past decade, in 2014 Generasi revised its education targets to focus investments on the neediest populations. The new education targets include participation rates for children with disabilities and transition rates from primary to junior secondary school. In addition, Generasi introduced indicators to measure community participation in enhanced nutrition counseling sessions delivered through the posyandu. The size of a village’s block grant depends in part on its performance on the 12 target indicators. If villages exceed the targets, they are eligible for a performance bonus, which is structured as a relative competition among villages within the same subdistrict. 33 This section addresses three main questions related to how the target and performance bonus system works in practice. First, to what extent are Generasi stakeholders knowledgeable about the bonus system? Second, what are the main strategies to achieve the targets, and do the performance bonuses stimulate competition? Third, how does the bonus system contribute to overall outcomes? This study’s findings point to a widespread understanding and awareness of the 12 targets among facilitators at all levels, but little awareness of how the performance bonuses work among village facilitators. The target system – specifically, the public announcement of villages’ progress in achieving the indicators and the desire to avoid embarrassment at meetings – motivates KPMDs to mobilize communities around achieving the targets. At the same time, the target system has led facilitators to spend too much time and focus on collecting data at the expense of outreach with communities. The amount of time facilitators spends collecting program-related data has been compounded by requests from MoV to collect additional data. Few actors believe that performance bonuses motivate facilitators to exceed the health and education targets. Respondents offered two possible explanations for why this is the case. First, since the bonuses are added to the subsequent year’s grant allocation, by the time communities receive the allocations, most facilitators have forgotten that they received a bonus. Second, performance bonuses do not “feel” like a bonus because almost all villages earn them. There is some evidence to suggest that individual rewards (even small rewards such as a meal, or a small package of household supplies) for successful, high-performing KPMDs are more effective. Monthly coordination MAD meetings, convened by FKs at the subdistrict level, are most frequently reported as the best way for KPMDs to learn about the targets and how to achieve them, through the FK’s instructions and through sharing information among facilitators. Village leaders attend these meetings to hear about the villages’ achievements, the ranking of villages, and the awarding of bonuses. The competitive nature of these meetings, and facilitators’ embarrassment if their village fails to meet its targets, stimulates facilitator effort. In the villages, Generasi actors and stakeholders report two main strategies to achieve their targets: ensuring community members attend posyandu and KMPD “sweeping” (going door to door to ensure that pregnant women and mothers attend posyandu). As a community-based health institution, posyandu represents the best opportunity for beneficiaries to convene, service providers to achieve various health goals, and for the KPMD to collect data on a range of indicators. Through experience, KPMDs learn that not all users attend posyandu; they receive their health services through a variety of channels. Only through home visits and meeting with users directly can KPMDs encourage beneficiaries to attend posyandu and motivate children to attend school, as well as keep their records accurate and complete. Awareness and Understanding of Generasi’s 12 Target Indicators “Yes, I’ve heard about the indicators, but they are confusing because of so many terms and abbreviations.” – Kepala desa, Maroangin, Tarengge subdistrict, Gorontalo district Knowledge of Generasi’s 12 target indicators (see Table 1) decreases at lower levels of government: Generasi stakeholders at the district and subdistrict levels have all the annual indicators memorized; village-level actors such as KPMDs and PKs know about the targets, but do not always have them all memorized; Generasi’s immediate partners in the UPK offices, puskesmas clinics, education offices, village midwives, and village government leaders understand that the targets exist, but may not know what they are or how they work; and program beneficiaries generally do not know about them. While the Generasi target system is designed to inform communities about how best to allocate their block grant resources, there is no evidence that this is taking place in the treatment communities. Only the KPMDs monitor progress through targets, and many view them as administrative requirements; other village-based actors do not know about the target system or how it works. 34 KPMDs’ knowledge of the 12 targets, which have changed several times over the years, varies considerably depending on their length of service in the program. Some KPMDs memorize them all, and can recite which targets were achieved (or not), while others cannot list even one, and have only a vague understanding that Generasi is a health and education program. Those who know about the 12 targets generally report that they are “a tool [or benchmark] for measuring Generasi’s success.” Only a few knowledgeable stakeholders, usually at the subdistrict level or above, mention that the targets help villages identify areas for improvement and allocate their resources more strategically. A few mentioned that the purpose of the targets is to meet Generasi’s overall goals of improving maternal and child health and education outcomes, improving community participation in village governance and performance monitoring, and enhancing the quality of health and education services in their communities. Slightly more informants in Garut and Pamekasan reported an understanding of the targets and their purpose compared to Gorontalo and Lembata. When Generasi stakeholders were asked whether the target indicators achieved their intended purposes, most respondents reported a summary of their annual results, such as which targets their village achieved or failed to achieve. At higher levels of government, respondents mentioned which targets the subdistrict or district typically achieve easily, and which remain a challenge. The head of the Ilekora BPD (Nelle subdistrict, Lembata) was dissatisfied with the KPMD’s explanation of how the target system works (they stated that the procurement of food supplements had already been decided by the program and could not be changed), and questioned whether Generasi’s support for providing food supplements achieves those targets. The KPMD’s explanation foreclosed further discussion, and the BPD leader wondered whether Generasi’s reported achievements each year were only “success on paper” that was used to guarantee additional funding support. Do Targets Induce Learning and/or Competition? Targets lead directly to outreach, which indirectly contributes to beneficiary learning. The head of the UPK office in Tarengge subdistrict (Gorontalo) reported that the annual system “makes it easier for [Generasi] actors to measure their performance.” Some Generasi actors explained that targets encourage more direct interaction with individual beneficiaries to collect more accurate and complete data, which gives beneficiaries more opportunities to access government services. The PK in Desa Rampe (Nelle subdistrict, Lembata) mentioned that the data collection process (which they described as “direct monitoring of individuals in the field”) enabled Generasi actors in the village to identify individual cases that health and education service providers might not be able to otherwise find. Likewise, the PK in Cikereteg village (Lebak Siwur subdistrict, Garut) described how data collection served as an impetus for home visits. When their village did not reach its immunization target, the KPMD discovered some beneficiaries had taken their infants for shots at a private clinic. The KPMD visited to check the health cards of beneficiaries and inform them about program services. Other actors described making home visits to ensure pregnant mothers take their iron pills, or to encourage parents of children at risk of dropping out of school to improve their attendance. A KPMD in Pasir Ucing village (Lebak Siwur district, Garut) indirectly connected the performance targets with behavior changes in service utilization. She stated that the targets have prevented maternal and child mortality cases during her tenure as a KPMD by stimulating monitoring and advocacy, and focusing the community’s attention on maternal health (i.e., routine pre-natal exams with midwives, consumption of iron pills). Sharing progress on meeting targets appears to serve as a more important motivation than the bonus system. Generasi actors at all levels (particularly KPMDs) seek to avoid the embarrassment of having to report incomplete or negative results. The FK for the subdistrict of Tarengge (Gorontalo) reports that four months into the annual program cycle, she presents the interim results for each village’s targets at their monthly coordination meeting. This allows the KPMD from each village to measure their progress and share ideas about how to improve their outcomes. For example, the KPMD from one village was not meeting their posyandu attendance targets, and the KPMD from another village advised using food supplements to attract 35 higher attendance. Generasi officials from the subdistricts of Lebak Siwur (Garut) and Petis subdistrict (Pamekasan district) also noted the utility of such information exchanges at monthly coordination meetings. The annual Generasi program cycle features three MAD meetings, held in the subdistrict seat, where all facilitators gather to agree upon the allocation of funds, discuss activity plans, and compare results (see the case study, “A Generasi Inter-Village Meeting and Improving Access to Health Services”). Kepala desas and village midwives attend, as do the subdistrict head, UPK team, and head of puskesmas, particularly when targets are announced and ranked. At their best, MAD meetings are festive and good humored in the public presentation of achievements from each village, with lively discussions and explanations. The FK in the Petis subdistrict (Pamekasan district) described a MAD meeting in 2014, at which one village was teased for failing to achieve six of its annual targets; the following year, the village reached its targets. Case Study: A Generasi Inter-Village Meeting and Improving Access to Health Services MAD meetings attract cross-sectoral stakeholders, including kepala desas, as well as BPMD representatives, because villages receive their block grants at these meetings. Interviews with MAD attendees in Petis demonstrated that these meetings encourage KPMDs, village governments, and service providers to play a more active role in achieving their targets and increasing access to services. The MAD can also encourage service providers to identify service gaps and address problems. For example, the 2013 MAD in Petis subdistrict (Pamekasan district) revealed a discrepancy in data collection between the KPMD and midwife from Sumber Petis village due to access problems. Although the midwife claimed that all targets had been covered regularly, according to the KPMD the indicator for weighing children under five was not achieved. The midwife had only collected the data of those who attended the posyandu; beneficiaries who never attended were not included. Soon after the MAD, the village midwife established a new posyandu in the other hamlet. The head of one of the two puskesmas in Petis admitted that he feels ashamed if the villages under his administration fail to reach their targets. Usually, after the MAD he organizes follow-up meetings at the puskesmas to discuss the underperforming targets in the villages under his administration. Villages can be removed from the program if they “cannot manage the program well, for instance: violating principles and provisions of the program, misuse of funds or authority. Funding for the subdistrict or village in question will either be postponed or removed altogether the following year.” Researchers only found one example of this – the control village of Kolipadan in the treatment subdistrict of Nelle (Lembata), where disincentives are a more accepted method of compelling participation. All the villages in Nelle had agreed that only those that successfully completed all phases of the annual program cycle would be eligible to receive block grants the following year. Kolipadan was removed from the list of beneficiaries after it had insufficient attendees at a village-level socialization meeting. While interviewers used the word “competition” to ask about targets, Generasi stakeholders typically described the target system as a motivational tool. According to the PK from Sogiyan village (Petis subdistrict, Pamekasan district), “there is a desire among the facilitators from each village to improve the indicators they have not achieved yet. When their data is announced at the coordination meetings, they will be embarrassed if they haven’t reached their targets.” This emphasis on motivation rather than competition appears to be due to two factors. First, a competition suggests a prize, but since the potential bonuses earned for the best-performing villages are awarded to next 36 year’s village budgets, rather than to individual KPMDs, there is little incentive to “win.” Second, the motivation to avoid shame and embarrassment at the public forums was repeatedly mentioned. The head of the UPK office in Petis subdistrict (Pamekasan district) reported that in 2015 and earlier, KPMDs more actively checked on their target performance. He assumes this is due to the withdrawal of Generasi community block grants (Bantuan Langsung Masyarakat, BLM) in 2017 as Generasi ends. “Gaming” the Target System When asked if they used any tricks to reach their targets, most Generasi facilitators answered by sharing their tips for success, such as their “sweeping” activities to collect complete and accurate data. KPMDs in Pamekasan and Garut described going door to door to weigh children who failed to show up at posyandu or to deliver vitamins and ensure that infants receive their immunizations. The second-most common response from village-level Generasi actors was to describe how they coordinated and copied data from posyandu volunteers. Two respondents mentioned providing transportation money (to young mothers or midwives) for post-natal care visits. Another two respondents reported routine liaising with the village midwife (such as letting her know when women show signs of early pregnancy) to achieve targets such as pre-natal exams. The only reported suggestion of cheating was from a KPMD who mentioned keeping in touch (via phone or text messages) with mothers who came home to their village to deliver their babies and then returned to the city where they live. This KPMD would continue collecting data from them on their weight, immunizations, and exams to boost the total number of beneficiaries in her village. Given the disincentives to report such practices during the research interviews, it is difficult to determine how widespread they are. Yet due to the potential benefits of inflated numbers of beneficiaries to increase their village’s budget for Generasi activities, this is unlikely to be the only case of cheating. In the subdistrict of Nelle, (Lembata), the head of the BPD in one of the treatment villages had extensive knowledge of the Generasi program even though he has never been directly involved in it. He strongly criticized the purchase of food supplements, questioning whether they addressed actual needs in the village, and accused the Generasi actors of gaming the program with the same activities every year because it guaranteed their targets would be met even if the supplements did not improve maternal and child health. He also critiqued the allocation of annual block grants, which he argued funded the same activities every year, without community consultation to find out what the people need. Generasi actors at the subdistrict and district levels reported other ways for the villages to meet their targets. The FK in Petis subdistrict (Pamekasan district) reported prioritizing training and other outreach opportunities for underachieving villages, while the faskab in Lembata detailed the benefits of combining the PAUD and posyandu in one building, as it made coordination easier and more efficient. Lembata’s former faskab said advocacy for village-level regulations had been an effective way to guarantee that certain targets were achieved. Generasi beneficiaries generally did not have any tips for maximizing their benefits from, or the targets for, the Generasi program, except two mothers who mentioned that if they are unable to attend posyandu, they usually send their children with a neighbor: one wanted to avoid a fine, while the other wanted to stay up to date with her child’s growth, and for her child to receive the PMT. Generasi actors are thus motivated to meet their targets, and have devised a variety of ways to achieve them. There is still no consensus on whether targets help communities meet their needs: opinions range from the skeptical BPD head in Ilekora (cited above) to the former puskesmas head in Petis subdistrict (Pamekasan district) who praised the Generasi program in Lelaok for using their block grant to buy a loudspeaker, which has been used to publicize posyandu, boost attendance, and more easily meet targets. 37 Understanding and Effectiveness of Generasi’s Performance Bonus System Only Generasi actors at the subdistrict level and above knew about the performance bonus system. Even at the subdistrict level, Generasi staff refer to the FK to give a detailed explanation. Indirect Generasi stakeholders such as UPK officials or village leaders may be aware that such a system exists, particularly if they had direct experience in the PNPM Rural program, but they believe it is an internal Generasi matter and do not understand how it works. All FKs in sampled treatment areas understand the performance bonus system; they all correctly described how it is supposed to work and how it is calculated. All Generasi actors at the district level also understand the system, including one former Generasi facilitator in Gorontalo who is now the TA-PSD for Village Law implementation. In the villages, there is no guarantee that even the KPMDs know about performance bonuses (see the case study, “Generasi Village Facilitators and the Performance Bonus System in Rampe”). Some KPMDs have heard of it, but know very little about how it works. One former KPMD in Petis subdistrict (Pamekasan district) had limited, but correct, knowledge about the system, however she described it disparagingly as merely a supplement to the BLM budget. A former KPMD in Lebak Siwur subdistrict (Garut) made a similar remark. An important exception is Lebak Siwur subdistrict (Garut), the pilot research area in West Java, where most of the KPMDs and other Generasi staff in the villages were knowledgeable about performance bonuses and how they work. High village-level staff turnover may partially explain why Generasi staff in the villages are so unfamiliar with the bonus system. The entire Generasi team in Desa Lelaok (Petis subdistrict, Pamekasan district) and the PK in Desa Rampe (Nelle subdistrict, Lembata) had never heard of the bonus system, but all of them had started within the past year. Bonuses are not a consideration in 2016, as Generasi removes the BLM budgets in anticipation of the transition. Higher-level Generasi staff (mainly at the subdistrict and district levels) explain the purpose of the performance bonuses in terms that reflect Generasi’s reason for offering them: to motivate KPMDs (and the communities they work in) to achieve their targets. However, most (but not all) stated that the bonuses do not achieve this stated goal for various reasons. Some reported that the bonuses are too small to make a difference once they are absorbed into the following year’s general operational budget. Their impact is also decreased by the fact that almost all villages earn them. Case Study: Generasi Village Facilitators and the Performance Bonus System in Lembata Generasi’s bonus system is designed to stimulate competition between treatment villages over achieving health and education indicators. Yet many communities do not fully understand (or even know about) the system, which is mainly discussed at the subdistrict-level MAD meetings attended by the KPMD and TPMD and key figures in the village. If these attendees do not actively disseminate the information to all community members, they will not know about it. Generasi actors in Rampe village do not attribute their achievements to a competition for the bonus, which the village receives every year despite lingering data collection problems. One of the KPMDs in Rampe reported that she is aware of Generasi’s bonus system, and that it encourages her to meet the program’s targets. The village’s senior KPMD is Fransiska Kewa, known as Mama Kewa, who has served as a volunteer in the health sector for many years. Even she demonstrated an incomplete understanding of the bonus system and made no connection to achieving Generasi’s targets. Mama Kewa recited Generasi’s efforts to reduce the number of children under five suffering from undernutrition and malnutrition. For her, Generasi’s success in identifying and treating the target group was due to the provision of PMT rather than the bonus. Although the number of malnourished in Tanjung Baru decreased from nine to six cases from 2014 to 2015, the total bonus 38 did not increase. Likewise, the health personnel and posyandu volunteers attributed their success at increasing the number of attendees every month to the fact that villagers must pay a fine if they do not bring their children to posyandu, rather than to the bonus. Village-based Generasi actors generally have a less comprehensive understanding of the program than higher- level actors. For example, those below the subdistrict level are unaware of coordination meetings at the district level where performance results are discussed. In Gorontalo, for example, Generasi activities are reported at the district level every three months, and shared with health, education, PKK, religion, social affairs, and regional development board officials. Provincial-level meetings in Gorontalo are also convened once every three months. District-level actors in Lembata, Gorontalo, and Garut also mentioned these coordination meetings; some mentioned that this data reaches Jakarta. The thoughtful and knowledgeable FK for Tarengge subdistrict felt strongly that posyandu are the key to improving the target and performance bonus system. She recommended increasing funding for this community-based institution and introducing a process indicator to incentivize participation. In keeping with Generasi’s community-driven approach, she also suggested making the posyandu a purely village-based institution, without relying on resources from higher levels of government. This study has shown that targets are motivating, especially when publicized in a well-attended public forum (such as MAD) that provides a level of public accountability to motivate target achievement. Performance bonuses, as currently conceived, are confusing and do not work as intended. The evidence suggests that individual rewards for village facilitators are more effective than community-based rewards such as the performance bonus system. Generasi’s Impact on Village Governance and Basic Social Services After nine years, Generasi is ending and handing over its goals and mechanisms to village governments, which are implementing the Village Law, which also aims to increase village communities’ access to basic infrastructure and social services. This section analyzes the program’s impacts on village governance and the delivery of basic social services. The findings suggest that Generasi (and the larger PNPM family of programs it is associated with) has had an impact on village governance, but not on basic social service delivery. There are a few possible explanations for this finding. Generasi and other PNPM programs are village-based CDD programs that encourage community members to frequently interact with village governments and influence village government practices. By comparison, within Generasi and other PNPM programs, there is less day-to-day contact between community members and service providers. Several Generasi facilitators have joined village governments, bringing with them their value of participatory decision-making processes and knowledge of health and education issues. By comparison, there is no evidence of Generasi facilitators joining the health and education departments. Based on instructions issued by the Generasi department within MoV, Generasi facilitators are actively advocating for village governments to allocate portions of their budgets to health and education activities that Generasi has historically funded. There are no such advocacy efforts targeted at service providers. The qualitative study uncovered a few different examples of the program’s spillover effects on village governance. In some treatment subdistricts, village governments have adopted participatory planning mechanisms based on the Generasi experience (and PNPM Rural experience in the control areas). Some treatment villages plan to retain Generasi KPMDs to facilitate village activities related to health and education. Several village governments in treatment areas, and some in control areas, spend part of their village budgets on routine activities that were formerly covered by Generasi grant allocations, such as food supplements at posyandu and incentives for posyandu volunteers. Given the usefulness of Generasi data for 39 planning activities, there are also cases of Village Law facilitators utilizing Generasi data; some control areas use Generasi forms to collect data on target beneficiaries and the condition of basic services. Nonetheless, there are a few different challenges related to integrating Generasi activities into the Village Law implementation process. Several village administrations expressed concerns about allocating their own village budgets for health and education activities because they say no regulations have been issued on this. Also, while KPMDs have held socialization activities with village government leaders on how to use the village budget for community service delivery, that knowledge gets lost with turnovers in government administration. Many Generasi Actors Have Entered Village (or Higher) Government The program’s most significant and obvious impact on local governance is the recruitment of Generasi facilitators and other actors into formal structures of village (and higher) governance. In all research locations, there were examples of former Generasi and PNPM Rural actors who have become village actors. The head of BPMD Gorontalo reports, for example, that in Tibawa subdistrict (one of the Generasi intervention subdistricts in Gorontalo, but not in this research sample), five out of six kepala desas were previously involved in either the Generasi or PNPM Rural programs. In the sampled research sites, the research team also found similar examples. The current village secretary of Maroangin village (Tarengge subdistrict, Gorontalo) is also a KPMD, while the kepala desa was previously a member of the village implementation team (Tim Pengelola Kegiatan, TPK), the village implementation team that implements approved activities, in the PNPM program. Likewise, in Ilekora village (Nelle subdistrict, Lembata), the kepala desa is a former PK for Generasi and a former TPK member for PNPM Rural. This transfer of experience to village governments encourages the carry-over and implementation of Generasi mechanisms into village governance in two ways. First, these actors bring first-hand knowledge and understanding of the Generasi program to their new position. Second, it ensures the new leader is supportive of Generasi goals and principles. For example, a former KPMD in a pilot study village in Lebak Siwur subdistrict (Garut) was appointed head of tax administration and plans to implement an incentive mechanism to increase tax revenues from each neighborhood in her village. Researchers also observed the KPMD/village secretary in Maroangin Tarengge subdistrict, Gorontalo using a Generasi-style ranking system to define village planning priorities at a consultative village planning event (see the case study, “Generasi Process Mechanisms Used in a Village Planning Meeting”). In Tarengge subdistrict, Gorontalo, the camat appointed the FK to the monitoring team related to the RKPDes that should be discussed in a village-wide meeting before deciding on the annual village budget.11 He also issued a decree to make all KPMDs members of the RPJMdes drafting team. This allows Generasi actors to incorporate health and education services into the village planning process. Meanwhile, the head of Bapermas in Pamekasan district encourages kepala desas through information dissemination meetings to utilize PNPM or Generasi actors in village governance as they are considered to be well trained. However, this is difficult to achieve because of the district’s political dynamics, since members of the village governments usually have personal connections to the kepala desa. By contrast, kepala desas in Lembata prioritize competence over politics: one of the KPMDs is even a political opponent. In the control villages in this study, village government actors also have a history of involvement in the PNPM program, such as the head of Jaton village (Telogojoyo subdistrict, Gorontalo). The Mogiye (Tanabola subdistrict, Lembata) kepala desa was not directly involved in PNPM, but has held his position since 2008 (when PNPM first arrived) and became familiar with the program mechanisms. He said the village still uses 11 RKPDes should refer to, and be in line with, the Rencana Jangka Menengah Desa (RPJMDes), which describes the village development plan for the next 5–6 years. Any changes to the RPJMDes should be discussed at a village-level meeting. 40 participatory planning tools introduced by the PNPM program such as calendars and organizational diagrams to help ensure that development plans reflect the community’s needs. Case Study: Generasi Process Mechanisms Used in a Village Planning Meeting Generasi concepts and mechanisms have become embedded in village government development activities in Gorontalo. During a visit to Maroangin village (Tarengge subdistrict, Gorontalo), researchers observed that Generasi actors directly influenced the village development planning meeting. The kepala desa assigned Pak Sulaiman, the KPMD/village secretary, to facilitate the meeting with the cooperation of the BPD chairman, Pak Arifin. Pak Sulaiman prepared the RKPDes and other documents to distribute to the participants. He also talked with the head of the BPD and the PK, Pak Abjan, about how to effectively discuss the community’s proposals. When a hamlet representative wanted to submit a new proposal that was not listed in the RKPDes as the basis for further discussion, Pak Sulaiman responded that any new proposal could still be included, but the main priorities were those already listed in the RKPDes. To rank the proposals, Pak Sulaiman divided the meeting participants into three groups based on their hamlet and each group prioritized three proposals. After a discussion, it was decided that the proposals would be ranked according to three criteria: (1) the proposed activity can be useful to many people, (2) it targets the poor, and (3) the community agrees it is urgent. Posyandu cadres also had active roles in the planning meeting. When ranking the proposed activities, a group of posyandu volunteers demanded the prioritization of PMT for children under five and pregnant women, and provided information on the target number of children under five and pregnant women in budget discussions. Generasi’s Influence on the Allocation of non-Generasi Village Funds There is significant coordination at the district level between the faskabs and TA-PSDs (who both have prior experience in PNPM/Generasi) working on Village Law implementation in Pamekasan and Gorontalo. This regular coordination, which includes data sharing, briefings, and progress reports, has influenced non- Generasi villages in both districts to allocate village funds for health and education in their village budgets. The two officials described their relationship as mutually beneficial; they both advocate for including health and education in village plans and budgets. Gorontalo’s faskab perceives this coordination as a strategic way to integrate Generasi’s planning mechanisms into regular village planning processes, which the TA-PSDs facilitate. The Pamekasan district TA-PSD noted that it is crucial to coordinate with Generasi facilitators, as those from his own program do not understand education and health issues. In both districts, routine meetings helped them identify problems and discuss possible solutions. One of the problems identified in the meetings was a lack of health and education data in non-Generasi villages. For example, the control village of Tespates (Sambingan subdistrict) in Pamekasan district lacks reliable data on the total number of infants and pregnant women in the village, since the posyandu volunteers only record the number of visitors who come to posyandu or access health services. This was also the case in the control village of Mogiye (Tanabola subdistrict, Lembata); the village secretary reported that some children had dropped out of school, but he did not know the exact number. The Pamekasan TA-PSD thus collaborated with the faskab to collect data using Generasi Forms 6A and 6B (targeted data and mapping service) and Form 8 (proposed activities form) in non-Generasi villages. TA-PSDs in Pamekasan and Gorontalo have gathered data using those forms to help non-Generasi villages conduct RPJMdes reviews, so that health and education issues can be accommodated in the village planning and budget processes. In 2016, the Gorontalo TA-PSD encouraged seven non-Generasi villages to serve as a model for the allocation of village funds for 41 education and health activities. In 2017, 32 villages (50% of the district’s non-Generasi villages) allocated their village funds for basic social services in Gorontalo. In Lembata, there was no such coordination, since no district-level facilitators had been deployed when the fieldwork was carried out. The importance of BPMD/Bapermas support: There is a correlation between BPMD/Bapermas support, in terms of bridging coordination between Generasi and P3MD facilitators, and village governments’ interest (particularly in non-Generasi villages) in funding education and health activities. In Pamekasan district, Bapermas has encouraged the underqualified TA-PSD to learn from the more experienced Generasi faskab. In Gorontalo, the TA-PSD is a former Generasi FK, who recognizes the importance of coordinating with Generasi actors to make sure that PSD, including health and education, have been covered in village planning and budget processes. Especially in Pamekasan and Gorontalo, the BPMD, Generasi, and P3MD facilitators held coordination meetings at least once a month to discuss their progress and provide updates on their activities. The Pamekasan faskab acknowledged that Bapermas was helping them solve problems in the field, for instance by sending bupatis “advice letters” to ensure villages prioritize health and education in their allocation of village funds after the end of the Generasi program. Similarly, the head of Gorontalo’s BPMD issued appeal letters to kepala desas emphasizing that villages should involve Generasi actors in the planning process, and allocate funds for health and education activities in their budgets. In Lembata district government agencies, the high staff turnover, including in the BPMD, inhibits the performance of the Generasi working group, which is based in BPMD. During 2016 and 2017, the head of the BPMD and its working group were replaced four times. For example, at the beginning of 2016, the BPMD planned to publish a decree that Generasi KPMD should be absorbed into village planning staff to perform a similar role. However, there was no follow-up, because Lembata’s new bupati replaced the BPMD head, which led to another staff turnover (including a new Generasi faskab). Differences between treatment and control villages: All villages have allocated some of their budgets to health and education activities, including the control villages in this study. In Pamekasan district, the activities that are supported by village funds in non-Generasi villages generally duplicate activities funded by Generasi, such as nutrition posts, pregnancy and nutrition classes, and PMT. Although the budget allocations and activities are nearly the same, they are implemented differently because Generasi villages have had complete, detailed, and targeted planning with data collected by trained posyandu volunteers, and established coordination with service providers. In control villages in Gorontalo and Lembata, health issues are included in the village budgets partly because both benefited from previous facilitation from PNPM Rural and other health-related programs. The kepala desa in Mogiye (Tanabola subdistrict), the control village in Lembata, revealed that facilitation mechanisms from PNPM Rural helped them identify village problems including health issues. The kepala desa still uses PNPM-style participatory planning tools that involve community members in village planning to ensure all priority issues in the village are accommodated. Facilitation processes also help connect village communities with service providers, particularly the local puskesmas. Although there was coordination, researchers observed that the control villages have limited data, which makes it hard to prioritize activities and allocate funds. Village officials in control areas sometimes merely consider what the service providers told them to do. The treatment villages have more active and trained posyandu volunteers than the control villages, in part due to the program’s training and incentives for volunteers. In control villages, not all volunteers have access to training due to puskesmas budget limitations: in Tespates village (Sambingan, Pamekasan district), only one out of 10 posyandu volunteers attended a training session, and the same volunteer (usually the most senior) attends each session. 42 Case Study: Management and Utilization of the Siaga Program in Mogiye Village, Lembata The control village of Mogiye (Tanabola subdistrict) is running Desa Siaga, a community health empowerment program, more effectively than most of the neighboring villages. The village government reports that the Siaga program targets maternal and child health issues because the mortality rates in Buyasuri remain high, partly due to the persistence of home deliveries, which carry a high risk of infection and death in both mothers and babies. The program was established in 2008 under the instruction of the puskesmas, yet it did not start actively functioning until 2012, after the strengthening intervention from the Australia Indonesia Partnership for Maternal and Neonatal Health (AIPMNH): “The Desa Siaga was merely a government instruction, where each village must establish their own committee. But there was no follow up after it was being formed. In addition, there was no supporting fund for the committee. Only after 2012 that the program was actually starting and keep running until today.” – Village Secretary and Chairman of Desa Siaga program in Mogiye In addition to AIPMNH funding, this program relies on village governments, health volunteers, and village budget allocations, as well as reinforcement from the subdistrict government. These factors also contribute to Generasi’s success outside Lembata; their absence in the Lembata treatment villages may explain why the program is less effective there. The AIPMNH funded many interventions throughout NTT province. The sampled treatment villages in Lembata did not receive AIPMNH intervention, but in the control village of Mogiye, APIMNH provided active support until 2015. AIPMNH successfully promoted village government involvement in managing the Desa Siaga program. The villagers also remain actively involved. Each posyandu area coordinator plays a role in the Siaga program. Funding networks help villagers pay for transportation and delivery costs. The village government and BPD raised awareness among community members to encourage them to contribute to the funding network. It was agreed that all villagers would contribute Rp1.000/household/month. The heads of neighborhood associations were assigned to collect contributions to the birth solidarity fund (dasolin) from all villagers; pregnant mothers collected birth savings (tabulin) to support their delivery costs. The Panama government manages the dasolin fund: the villagers’ annual fees (about Rp 4 million) are collected like village taxes and are considered village income and expended as part of the village’s expenses. According to the village secretary, the fund is currently valued at about Rp 25 million; villagers can request an interest-free loan from this fund and repay it in installments within three months. Those who fail to repay their loan on time will have their Family Hope Program (PKH) fund payment or their Hari Orang Kerja (HOK) (“working days”)12 incentives cut. Tabulin is managed separately, because this fund depends on how many pregnant women contribute to it; they receive the funds when they deliver. 12 This term refers to daily wage labor on village infrastructure projects. 43 This lending mechanism fosters collective responsibility within the community and helps the village government ensure that villagers can access short-term funding when they need it. The mothers generally welcome this lending system. One said, “Paying the Dasolin is for our good, as we know we might be the one in need of a loan in the future.” One villager remarked that she was requesting a Dasolin loan to cover her delivery costs in the Lembata district capital. The village midwives also benefit from the program, since their performance assessments depend on improved health services. For example, the Siaga program’s transportation network helps pregnant women get to their pre-natal exams and deliveries, which helps meet the target for pre- natal exams and early detection of at-risk births. Gradually midwives can increase the number of facilities-based births in Mogiye. As of August 2016, there were no home births; 100% of deliveries took place in health facilities. In 2015 there was one birth at home, while in 2014 there were two. The success of the Siaga program in Mogiye is also due to the role of the subdistrict. According to the subdistrict secretary, at the time of the fieldwork only five out of 20 villages in Tanabola had active Siaga programs, but they were committed to promoting it to other villages: “We will not tolerate excuses like difficulty in transportation, I constantly emphasize this to kepala desas. Please take more active roles, and avoid making the village being commanded, I am asking for their support, not as an instruction. Besides, the money will eventually be paid back, because the Siaga program has some saving, that the villagers will pay back, the village government just needs to cover it in advance. That’s what the Tabulin is for.” – Secretary, Tanabola subdistrict office, Lembata district The secretary also mentioned the importance of strong village government involvement in the program’s success. The subdistrict government and the puskesmas acknowledged the limitations of health services in Buyasuri villages; for example, the availability of health workers in the village is still low. Although the number of births at the puskemas is increasing, Buyasuri is still classified as a “red area,” with high maternal and infant deaths. Mogiye’s experience with the Desa Siaga program is an important reference point for Generasi, especially with respect to how the village government followed up earlier processes of routine facilitation by AIPMNH. Generasi has not strengthened the Desa Siaga program, as Generasi has its own mechanism to work through the KPMD facilitators. The treatment village of Ilekora (Nelle subdistrict, Lembata) held a meeting during the field visit to re-activate the Siaga program. In that meeting, the BPMD and puskesmas encouraged the village government to re-establish the Siaga committee. The Ilekora kepala desa responded that they had issued a decree on new management of Desa Siaga and encouraged the posyandu volunteers to play a more active role by allocating monthly incentives from the village budget. The decree did not specifically involve Generasi volunteers or a new working strategy as lessons learned for Generasi, yet the head of Generasi’s TPK was appointed the chairperson of the Desa Siaga program due to their active role in village affairs. The two villages had similar levels of health performance. Ilekora, despite having an inactive Siaga program and a rather ineffective Generasi program, benefited from its proximity to health services in the district capital. Even though Mogiye is remote and isolated, its health outcomes appear relatively similar thanks to its commitment to sustain the Siaga program. Other non-Generasi villages with poor health facilities and difficult access will be more disadvantaged. Activating the Siaga program can be useful for villages such as Mogiye, although it requires active management by the village government. At least the kepala desa of Ilekora admitted that he already knew from the BPMD that a portion of the village budget could be allocated to support Siaga. The 44 attempt to re-activate Siaga in Ilekora also demonstrated the role of district and subdistrict governments in encouraging village governments to improve village-level services. In such cases, it would be easier for the program and the government to encourage the sustainability of the basic services in the villages. Village-level dynamics: Activities funded by Generasi such as PMT, posyandu, and volunteer honorarium have caused jealousy in non-Generasi villages in Pamekasan district. One informant from the control subdistrict of Sambingan said: “Looking at the neighboring subdistrict get support, such as posyandu from Generasi, communities here also demanded the same thing from their kepala desas, who then pass it to head of subdistrict and Bapermas.” – Head Section of Government Empowerment of Sambingan subdistrict In response to demands from kepala desas, since 2011 Bapermas Pamekasan has proposed budget allocation for PMT and pays honoraria to posyandu volunteers in five non-Generasi subdistricts from the Pamekasan district assembly (DPRD). The proposal was approved by DPRD, but only for six months each year; village governments must cover the remaining six months. The head of the control village of Tespates (Sambingan subdistrict, Pamekasan) said he must sometimes spend personal funds to cover the shortage or get assistance from the midwife in the village/puskesmas. He does not know if the activity could be paid for out of the village budget (APBdes). In 2017, the honoraria and PMT were included in Tespates’s (Sambingan subdistrict, Pamekasan) RKPdes, as the kepala desa was informed by village funding staff and Bapermas that such activities can be budgeted from village funds. Based on interviews with Bapermas in Pamekasan, almost all non-Generasi villages in Pamekasan did this. Researchers also heard that almost all control villages included posyandu activities in their village budgets. Perceived Barriers to Using Non-Generasi Funds for Health and Education Perceptions about using village funds for health and education: District- and subdistrict-level governments generally perceive the main obstacle to using village funds for health and education to be village governments’ preferences for infrastructure development activities, which are easier to implement, show tangible signs of “development,” and offer opportunities for kickbacks. Dr. Ina, Head of the Health Promotion and Public Health sections in the provincial health office, noted: “Since village funds have been made, the kepala desas are required to understand everything, while their capacity is limited. The kepala desa needs to understand health affairs, education, and administration, so all the ministries are in the village. The kepala desa acts as if he is the small president of the village … They then make a simple activity, build embung (water storage) and road for instance, while they have limited knowledge of health.” From the village-level perspective, kepala desas choose to take on infrastructure projects not because they are “easy,” but rather because they are “safe” to implement administratively. Village officers in Pamekasan district use the siskeudes (village finance system) application developed by the Finance and Development Supervisory Agency to manage village-funded activities. Since village financial reports are required to correspond to the menu of approved activities in the siskeudes application, the kepala desas worry about allocating village funds for health and education. “If we allocate village fund budgets for something like the transportation costs of the posyandu volunteers, but there has not been a regulation statement issued yet, then we will have audit problems from [the Finance and Development Supervisory Agency], say the village chiefs.” — Bapermas Secretary, Pamekasan District 45 In addition, almost all kepala desas reported that the rules from Jakarta change frequently and are still unclear. According to the Gorontalo faskab and BPMD, central regulations prioritize infrastructure development for economic livelihoods during the first year of the program. The district officer also has the authority to cut budget items that are not in line with the technical guidelines, which deters some villages in Gorontalo from including health and education proposals in their budgets. The head of Rampe village (Nelle subdistrict, Lembata) is willing to distribute KPMD incentives from village funds so that KPMDs can still help villages with their education and health services, but he is concerned that central regulations may not allow this. The kepala desas from Pamekasan district asked their bupati to produce guidance on allocating village funds for health and education activities. The health and education proposals included in the RPJMDes are still so general that they cannot be included in RKPDes. For example, one of the health service proposals in a Petis subdistrict village’s (Pamekasan district) RPJMdes refers to “improving maternal health” without defining specific activities. The Generasi facilitator encouraged the villages to review RPJMDes to accommodate more detailed health and education proposals (e.g., pregnancy classes or transport incentives) so they could be put into RKPDes and APBdes. Although almost all the villages visited in this study have allocated funds for health and education in their village budgets, some of them duplicate the activities undertaken by Generasi from their prior RAB (Budget Plan) design. Villages in the control subdistricts in Gorontalo and Pamekasan districts generally duplicate activities funded by Generasi, while the treatment villages have slight variations in their proposed education and health activities. Generasi villages in Pamekasan proposed more varied activities than those in Lembata, mainly due to the ability and intensity of Generasi facilitation from the FK and faskab in helping villages design their plans. The Generasi FK and his assistant FK in Nelle subdistrict (Lembata) said they did not have time to facilitate the planning process in 2016 because of complicated administration such as filling out reports requested by MoV. In contrast, Generasi facilitators in Petis subdistrict (Pamekasan) are actively involved in facilitating village planning and discussion; the faskab team occasionally attends the village planning activities. Generasi facilitation is crucial in designing village plans, as many village governments lack detailed information about RAB. According to a Petis subdistrict government report, few kepala desas had dealt with RAB and RPJMdes before village funds were available. Village planning documents are created by the subdistrict and tend to be similar among the constituent villages. Village fund facilitators are not trained to help the villages in this process. According to TA-PSDs in the control subdistrict of Sambingan, they usually use health and education service models from Generasi villages as a template. “I have joined the RKPDes consultation meetings several times; most kepala desas are afraid and confused, firstly about how to design activities to eradicate malnutrition, and secondly about how to allocate for that in the budget. In the end, they focus only on routine activities, or items for which they previously allocated.” – FK of Larangan subdistrict, during a focus group discussion with all FK in Pamekasan district Case Study: Efforts to Integrate Generasi and Increase Demand for Basic Social Services in Pamekasan district In Pamekasan, kepala desas play a significant role in determining the direction of village development; they rely on an elite group of community and religious elders to sustain their political power. Elite interests are often disguised as a representation of community member interests. The village government staff usually consists of people with family ties or other close relationships with the kepala desa. A new kepala desa may replace all village government personnel, including Generasi actors. District and subdistrict facilitators in Pamekasan acknowledge that the sustainability of Generasi activities after the program ends depends on individual kepala desas. If they believe Generasi 46 activities need to be continued, they will integrate them into the village development plan. The program appears to have achieved its goal of cultivating community demand for basic social services: there has been an increased use of health services in Pamekasan. The extent to which this increasing demand can encourage village governments to prioritize basic social services in their village development plans, however, remains unclear. In the treatment village of Sogiyan (Petis subdistrict), the kepala desa has demonstrated a strong commitment to basic social services and intends to ensure that the programs Generasi implemented can be financed through the village fund. The village midwife noted that awareness of health services has been increasing, and that Generasi (particularly PMT and milk distribution) played a vital role in increasing villagers’ enthusiasm to attend posyandu. Generasi FKs shared a similar assessment, noting that community involvement decreased drastically when the PMT was temporarily suspended due to the reduction of the BLM fund in early 2015. The sudden drop in participation levels convinced the village government to support the resumption of PMT to restore posyandu attendance. Aside from whether community awareness of the importance of health services has grown, posyandu participants now consider PMT a norm, and removing it raises questions. Since Generasi was seen as responsible for providing (or not) the PMT, it is perceived as an aid program. The kepala desa seemed to be aware of this perception, as he regularly received information about Generasi activities from the KPMD and village midwife, especially if there were complaints from the community. Although the kepala desa exercises considerable control over Sogiyan’s development, he seemed quite sensitive to the villagers’ complaints, especially those related to what people perceive as government assistance programs. This sensitivity is closely related to protecting his electoral advantage: “If the community keeps complaining, they will not vote for me any longer.” A hamlet leader in Tengangser Laok told researchers that villagers assessed the performance of the kepala desa based on (1) his or her ability to easily provide civil administration documents, especially regarding statements of households living in poverty for education and health care purposes; (2) village infrastructure development, especially roads; and (3) ensuring the distribution of government assistance to the poor. Thus, the supply of PMT is associated with the kepala desa’s efforts to maintain his reputation. In this case, Generasi created a condition in the community, in which PMT became the expected norm; this ultimately encouraged village governments to continue implementing Generasi using the new village funds. The facilitators repeatedly mentioned that PMT was intended to serve as a stimulant to generate public interest in health services in the village, and raise awareness of the importance of basic social services; however, beneficiaries perceive it as the main activity. Because the number of participants at posyandu is highly dependent on the supply of PMT, it is very hard to say that the community has gained an understanding of the importance of basic social services. The public in all areas visited for this study equates PMT with other government assistance such as raskin (rice for the poor). Before the Generasi program intervention, the kepala desa had no idea what basic social services could be included in the village development plan. After several years of Generasi, PMT has proven to be necessary and widely appreciated by the community. 47 Case Study: “Reporting Day” Forum and Allocation of Village Funds for Basic Social Services in Tarengge Subdistrict, Gorontalo district Villages can continue PSD provision activities like those supported by Generasi in two ways – either village governments are compelled to do so by their experiences receiving several support programs, or they receive instructions from the subdistrict level (or higher). The ‘Reporting Day’ forum, which is now regularly conducted in all subdistricts in Gorontalo, encourages villages to continue supporting PSD activities. Kepala desas gather each week at the subdistrict office to discuss progress and problems in their villages. According to the Tarengge camat, they frequently discuss efforts to continue improving the health and education sectors, because in 2016 some villages in Tarengge had cases of malnutrition, deliveries without health personnel attending, chronic energy deficiency pregnancies, and school dropouts. According to the camat, the subdistrict government was considering the sustainability of Generasi- supported activities in the villages at the end of the program. He believes the availability of increased village funds has given villages more opportunities to pursue their own development plans. While there have been some general technical guidelines on the utilization of village funds, more specific guidance on PSD is still needed. Reporting Day meetings always discuss the use of village funds, including the administrative complexities of fund management. “We are worried that after Generasi has ended, the quality of health services will decrease. It is indeed possible to use village funds to budget for these activities, but it depends on the village government whether they budget for it or not. In addition, it also depends on the technical guidance.” – Tarengge Camat, Gorontalo district. Other subdistrict-level forums are less effective in Gorontalo. The puskesmas convenes a coordination forum called minilok (mini workshops), held once every three months, which includes many stakeholders, yet many health issues remain unsolved at these meetings. Although final health-related policy decisions in the village rest primarily with health service providers, as the camat has no formal authority over health and education services at the village level, the subdistrict and village governments can play a monitoring role in village-level service delivery. The experiences of PNPM Rural and Generasi were different: officials in the subdistrict were mandated to oversee the operational activities and maintain the programs’ village development processes. Under the Village Law, the villages are more responsible to BPMPD (at the district level), which evaluates them. Therefore, BPMPD should play a larger role, but this is generally not the case yet, perhaps due to the large areas that district-level offices must monitor. Through routine, and more local, forums such as Reporting Day, the camat can learn more about village issues and encourage the mutual exchange of ideas. The camat can also directly monitor the mechanisms and progress of village development, even though his or her invested authority is still relatively weak. At the very least, Reporting Day meetings represent a useful way for bureaucratic structures to facilitate empowerment initiatives. If implemented in other places, such a forum may help over the longer term, particularly to give the subdistrict government a strategic role in the context of Village Law implementation. Pamekasan district, for instance, uses a different model – the Kepala desa Association, a subdistrict forum attended by kepala desas and the camat, led by one of the kepala desas. In Gorontalo, Reporting Day meetings are useful to discuss development issues in the villages. The Reporting Day structure can be improved and scaled up to other places to encourage villages to address basic social services in their planning and budgeting processes. When the head of the Tarengge puskesmas realized that village funds could be used to support health development in the 48 villages, she began attending the Reporting Day meetings. They may have also influenced the villages of Mangkawani to and Maroangin to include the purchase of health and education items in their budgets. Reporting Day can become a consultative forum, not unlike Generasi’s MAD meetings, which involves more parties to improve stakeholder coordination, especially to overcome village government concerns about how to use village funds effectively. The kepala desa of Maroangin, for instance, worried about budget overlap with activities implemented directly by health and education services. An important generalized finding of this research is that inter-village forums at the subdistrict level that require and incentivize the participation of village leaders and other stakeholders – such as Reporting Day in Gorontalo, Kepala desa Association meetings in Pamekasan, and Generasi’s own MAD meetings – ensure that village-level concerns are shared and addressed jointly. Generasi Impacts on Local Regulations The program’s impact on local rules and regulations varies by region, often depending on cultural factors. In Lembata, Generasi facilitators encourage the formulation of village regulations (often fines) to help achieve program targets. Lembata’s former faskab explains: “When we see the indicators do not work, such as pregnant women do not want to attend pre-natal exams, how do we make her and her family aware? Then we make the perdes (village regulation). In perdes, there are sanctions… and these perdes are effective, they serve as a deterrent, because life should have rules. The development ‘cakes’ are plenty, but if they are not bound by the rules, everything becomes more difficult.” – Former Faskab in Lembata Pamekasan district, by contrast, has a tradition of authoritative decrees from a single leader such as the bupati or kepala desa, hence many informants there said they hoped for guidance through executive decrees. Generasi facilitators in that area encouraged the bupati to write to all kepala desas to encourage them to allocate funds for health and education activities in their budgets. Generasi’s Impact on Basic Social Services This field study did not reveal strong evidence that Generasi processes affected the provision of basic health and education services, perhaps because village-level Generasi actors consider service providers to be more knowledgeable about service provision. For example, Generasi actors often consult with puskesmas staff about nutrition or what activities the program should support in the villages. The head of the puskesmas in the subdistrict of Boliyhuto (Gorontalo) explained: “I advised [the Generasi FK] Ms. Rara to use the posyandu volunteers trained by our puskesmas… Generasi has their own volunteers, but I told them to use our volunteers, who have already received training, because they already understand what food is appropriate for infants, toddlers, and pregnant women. They already have the health knowledge.” – Head of Puskesmas Tarengge subdistrict, Gorontalo district Service providers are also not deeply involved in Generasi processes, which are more intensely implemented in the villages than in the subdistrict; they usually attend Generasi meetings at the subdistrict level such as the MAD. One exception may be Generasi data collection processes: the program encouraged health service providers in the villages and puskesmas in Gorontalo and Pamekasan districts to improve the quality of their own data collection. According to the midwife coordinator at the Tarengge puskesmas, when Generasi data presented at a MAD meeting revealed a high-risk pregnancy that the village midwife was unaware of, this 49 pushed the coordinator to increase the number of village midwives throughout Tarengge subdistrict to ensure more careful monitoring of potential service users. In Lebak Siwur subdistrict, the puskemas nutrition officer coordinates with FKs to gather data on children and pregnant mothers with nutritional risk, and determine which cases should receive food supplements. Generasi data appears to have had little impact on education service providers, perhaps because the education department collects data at the school (rather than village) level. However, Generasi has encouraged valuable connections between service providers and village communities, which will be useful during the implementation of the Village Law. For example, the secretary of Pamekasan’s District Education Office reported that Generasi meetings raised awareness of the importance of education for disabled children, which resulted in a proposed budget allocation in 2016 for teacher training in inclusive schools (those that educate special needs children). Conclusions The qualitative fieldwork yielded findings in four main areas: facilitation, participation, targets and incentives, and village governance. Facilitation: Village-level Generasi facilitators were effective at mobilizing communities to achieve program targets for three main reasons. First, the participatory processes through which community members selected the facilitators helped to ensure that they were representative of the community and responsive to its needs. Second, the facilitators invested substantial time into maintaining relationships with their communities, facilitators in other villages, and service providers. Third, their knowledge of health and education issues enabled them to fulfill their core tasks of mobilizing mothers to attend the monthly posyandu and encouraging students to attend school. Ensuring that data on beneficiaries’ progress towards achieving the 12 health and education targets is complete is a core responsibility of KPMDs and FKs. Yet the time required to collect this data gives facilitators less time for outreach and advocacy. This burden has been compounded during the Village Law implementation process by additional requests from MoV for monthly village reports. Participation: Ensuring widespread participation in planning meetings requires significant investment from facilitators. Beneficiary participation in Generasi planning activities was found to decline after the hamlet-level meetings in part due to the widespread understanding of participation as attending (but not necessarily speaking up at) meetings. Some respondents reported that they did not feel they had the knowledge or the right to speak at such meetings, or that it might be interpreted as a sign of disrespect against village leaders. Therefore, the program’s goal of empowering communities to plan, implement, and monitor the delivery of basic services and influence village governance was not entirely met. Targets and Incentives: Generasi’s 12 health and education targets serve an important role in motivating facilitators to mobilize communities around meeting the targets. There is a widespread understanding and awareness of the 12 targets among facilitators at all levels. Pressure on facilitators to report at inter-village meetings that their village met its targets serves as an important motivation. By comparison, few frontline program actors demonstrated an in-depth understanding of program’s performance bonuses. Because the bonuses are added on to subsequent years’ budgets, are quite small, and are awarded to almost all villages, they do not stimulate competition and motivate greater facilitator effort. Village Governance: Generasi was found to have had a significant impact on village governance. Generasi-style consultation and implementation mechanisms are becoming important elements of village planning processes, and Generasi actors are serving as important resources during the Village Law implementation phase. KPMDs are beginning to advocate the inclusion of health and education activities in village budgets. 50 Recommendations Facilitation • Village Law facilitators, specifically PLD, should be recruited from the local communities they serve. Both PLD and PD should receive training in basic social services to allow them to effectively advocate for village governments to utilize funds for health and education services once Generasi ends. For training materials, MoV could draw on curriculum developed under Generasi. • Posyandu volunteers and other community-based volunteers should receive training in health service delivery. Puskesmas staff could deliver complete and routine training for posyandu cadres and all community-based health volunteers starting with curriculum MoH has already developed for this purpose. Village governments could help to pay for these training costs. • While it is important for Village Law facilitators to continue collecting health, education, and other basic indicators, MoV should reduce their data collection burden. MoV and other community-based programs should consider training and paying community volunteers such as former Generasi Village KPMD to collect routine data, which would free up facilitators’ time for outreach and enlarge the network of community volunteers. • Subdistricts should continue coordinating inter-village meetings post-Generasi in which village actors discuss community problems and exchange advice. These meetings can also be adapted and used to motivate and incentivize village governments’ performance. Subdistricts should build on the locally developed innovations that contributes to village governance practices, which this report highlights. Participation • Community-driven development projects should encourage equal participation in the full project cycle from both village-level elites and community members, including posyandu volunteers and community facilitators. While ordinary community members may not participate in village-level planning meetings in large numbers, KPMDs are actively in touch with them, and are thus able to present their diverse needs at the planning meetings and follow up with outreach activities. If community-level facilitators are selected from the communities they serve, they can provide ongoing support and help to bring community members’ complaints and needs to higher levels. • Given the important role that posyandu volunteers play in providing maternal and infant health services, village governments should invest more in the posyandu and continue to ensure that posyandu are sufficiently staffed (at least five per village and one per hamlet) and that they are compensated appropriately. Performance targets can help village governments monitor the posyandu activities and ensure continued investment in them • The analysis highlights many examples of how local culture continues to influence health and education behaviors and modes of accountability. Under the Village Law, service providers, village governments and facilitators should consider how to tailor health and education activities to local cultures to influence behaviors. Targets and incentives • Generasi’s target system was effective at motivating Generasi village-level and subdistrict-level facilitators to mobilize communities around the targets. Future programs that consider adopting such a system should ensure that there is a forum like the MAD meetings that puts pressure on village law facilitators, kepala desas, and village governments to ensure their villages meet their targets. 51 • Programs that adopt a Generasi-style performance bonus may want to consider simplifying the system, and to ensure high levels of awareness of the process among program actors and community beneficiaries throughout the project cycle. The monetary rewards associated with the bonus should also be more substantial. • MoV should consider putting in place a simple set of performance targets for village governments. The above-mentioned MAD or Reporting Day meeting (see the case study, “Reporting Day” Forum and Allocation of Village Funds for Basic Social Services in Tarengge Subdistrict, Gorontalo district”) could incentivize village governments to collect data needed to report back, and put pressure on service providers to deliver more and better to make the kepala desas’ performance stand out. • Future programs should consider rewarding individual facilitators with non-monetary bonuses (e.g., a package of household supplies) as a possible alternative or complement to a community-level performance bonus. Evidence from the qualitative study suggests that informal rewards help KPMDs and other village actors feel appreciated and motivated. 52 Annex A: Description of Sample Sites Descriptive Statistics Village Characteristics (Source: PODES) Receiving Distance to Government Location Population (in Person) # of Education Facilities # of Health Facilities Jamkesmas Offices (in km) # of Bidan District Sub-district Village Male Female All Households Kecamatan Kabupaten Elementary Junior High Praktek Polindes Posyandu Year=2005 Pamekasan Sambingan Tespates 815 887 1702 320 0.5 9 3 0 0 0 2 Pamekasan Petis Sogiyan 1466 1561 3027 253 18 33 4 1 0 1 2 Pamekasan Petis Lelaok 1760 2223 3983 620 3.5 37.5 7 2 1 0 3 Lembata Nelle Ilekora 250 269 519 125 4 13 1 0 0 0 4 Lembata Nelle Rampe 128 295 423 103 8 20 0 0 0 1 0 Lembata Tanabola Mogiye 557 676 1233 179 9 65 2 0 0 1 3 Gorontalo Tarengge Mangkawani 520 526 1046 121 5 49 1 0 0 0 3 Gorontalo Telogojoyo Jaton 1017 1032 2049 134 2 3.5 2 1 0 0 1 Year=2014 Pamekasan Sambingan Tespates 876 962 1838 306 1 9 3 1 1 1 3 Pamekasan Petis Sogiyan 1258 1415 2673 288 12 25 4 1 0 1 3 Pamekasan Petis Lelaok 1916 2030 3946 793 2 37 4 1 1 1 5 Lembata Nelle Ilekora 292 305 597 1 15 24 1 1 0 1 1 Lembata Nelle Rampe 197 238 435 2 9 15 0 0 0 1 1 Lembata Tanabola Mogiye 728 811 1539 0 7 65 2 0 0 1 3 Gorontalo Tarengge Mangkawani 575 615 1190 219 7 51 1 0 1 0 3 Gorontalo Tarengge Maroangin 527 526 1053 0 5 64 1 1 0 0 3 Gorontalo Telogojoyo Jaton 1184 1174 2358 283 3 7 2 1 0 0 0 53 Subdistrict and Village Profiles District Subdistrict Village Description (Kabupaten) (Kecamatan) Tarengge The total population of Tarengge is 16,764, in 4,866 households. The main livelihoods are agriculture, particularly rice and corn, and work in factories. Tarengge subdistrict has one puskesmas clinic with basic emergency obstetric care (Pelayanan Obstetri Neonatus Essensial Dasar, PONED) status, which means it can provide emergency services for women with pregnancy complications and newborns with other health complications. Yet currently only the midwife coordinator has PONED training; the other 13 midwives are mainly temporary staff or volunteers. Most of the midwives do not live in the villages where they are assigned, both because they are needed at the puskesmas and due to a lack of housing. Maroangin Most of the village’s 1,096 residents work in agriculture. More than half of the 303 households (165) are categorized as poor based on 2015 rice subsidy data. Maroangin has good road access to and from the village, and it has an elementary school and a middle school; two nearby high schools are easily reached. Maroangin is assigned a volunteer midwife who works at the puskesmas clinic in Tarengge (two hours away) and visits the village twice a month for posyandu activities (there is no clinic with housing in the village). Generasi began in Tarengge in 2007, before Maroangin was split from the neighboring village of Diloniyohu. Since then the program has supported activities such as food supplements and volunteer incentives for posyandu, school uniforms, food supplements for undernourished infants, transportation for high-risk pregnant women, and transportation to school for poor students. Gorontalo Maroangin has two posyandu, one for each hamlet. Each has five volunteers: one receives incentive payments from the puskesmas, while the other four receive payments from 54 Generasi and the village government. Before Generasi, the puskesmas could only provide for two volunteers. Posyandu activities usually take place in the Early Childhood Education and Development (PAUD) building or the kepala desa’s office. Each posyandu has a full set of supplies; village funds recently purchased new tools and supplies (height measuring tool and a blood pressure tool). The volunteers frequently fail to fill in the register books for the posyandu. Mangkawani Mangkawani’s five hamlets are home to 1,101 residents and 283 households, 105 of which are categorized as poor. Most residents work in agriculture or in the village’s coconut flour processing factory. It has good road access to the district (48 km) and subdistrict (5 km) capitals. It has a PAUD and an elementary school. The nearest middle and high schools are 3 km away. Generasi has supported the following activities in Mangkawani since 2007: incentives for posyandu volunteers, provision of posyandu supplies, food supplements, school uniforms, and transportation to school. Mangkawani has a satellite puskesmas clinic with one midwife and one nurse, which had reportedly been inactive for the past year. The two staff work at the subdistrict puskesmas and only come to Mangkawani for posyandu activities. The midwife does not want to stay at the satellite clinic in Mangkawani because she is single and has concerns about living there alone. Village officials have conveyed their concerns to the puskesmas but without response. The three posyandu are usually held in the volunteer’s house and at the PAUD building. Telogojoyo Telogojoyo (also called Limbar) has ten villages with 25,005 residents in 7,306 households. The poverty rate is 39.9%. Most of the population works in agriculture, and the farthest village is only 3km from the subdistrict seat, with good road access. It has one puskesmas clinic (A-class status, with 24-hour service), in charge of seven satellite clinics. In the villages, there are 20 posyandu and three poskesdes (village health posts that should be staffed by a village midwife and two village cadres) clinics. There are six general practitioners and one dentist at the puskesmas. All the villages in Limbar have a midwife (though not all of them live there), seven of whom are permanent civil servants. The puskesmas and the subdistrict government have an “Integrity Pact” with 55 the Gorontalo bupati to reduce the number of maternal and child mortalities. Limbar has sufficient education facilities, and enrollments are very high. Almost all the PAUD buildings in the villages were built by PNPM Rural. Although it has not participated in Generasi, it has taken part in other programs such as the 2015 Pekan Sayang Ibu Anak program and the Gegas Program that encourages pregnant mothers to deliver their infants at health facilities with a certified health service provider. It also received maternal and child health service improvement programs from 2012 until 2015 from the international NGO Adventist International Relief Agency. Jaton Jaton’s population of 2,500 is spread among four hamlets and 738 households, 102 of which are considered poor. Although approximately 100 residents are civil servants in the nearby subdistrict capital, most work in agriculture. The village hosts the subdistrict puskesmas. There is one village midwife for Jaton, as well as two posyandu but only five volunteers to staff them. Jaton has two PAUD schools, two elementary schools, and one middle school. The subdistrict seat (2 km away) has a high school and a school for special needs children. Ilekora Ilekora’s four hamlets contain 635 people living in 166 households. Most residents work in agriculture or fishing. It has good road access to the subdistrict (15 minutes) and district (30 minutes) capitals. The village has one elementary school, and middle and high schools are relatively close and easily reached. Ilekora has four posyandu and 14 volunteers to staff them, as well as one polindes clinic with one village midwife, which provides medication and pregnancy exams; all mothers in the province are required to deliver in puskesmas or hospitals. Rampe Rampe has three hamlets. Its 467 residents mainly work in agriculture or fishing, although some work abroad as laborers. Of the 134 households, 50 are categorized as poor. It is 30 minutes from the subdistrict seat. Rampe has a pre-school; elementary, middle, and high schools are located in neighboring villages. It has one 56 village midwife, one nutrition nurse, and one posyandu with ten volunteers. Villagers also use a satellite health clinic in a neighboring village, which is only 500 m from the kepala desa’s office. Generasi has been active in Rampe since 2007, and the budgets are generally used to support posyandu and to pay the school committee fees for poor families. Tanabola Tanabola has 20,135 residents, living in roughly 5,387 households. The main source of income is fishing and agriculture. Access to its 20 villages is difficult, with hilly terrain and poor roads. Nearly all villages have elementary schools and PAUDs, however the availability of teachers is uneven; some schools have only two or three. Tanabola has an isolated, basic puskesmas, five satellite puskesmas, two poskesdes, and one polindes. The puskesmas has one general practitioner and 36 midwives, 26 of whom are village midwives who stay in village-based clinics, most of which were built with funds from PNPM Rural. Every village has one or two midwives, most of whom are not permanent civil servants. Although it has not participated in the Generasi program, it has received support from PNPM Rural, the Australia- Indonesia Partnership for Maternal and Neonatal Health (AIPMNH), the NGO Plan International, Desa Siaga, and the province-wide Revolusi KIA (maternal and child health revolution) program. Mogiye Mogiye has 1,625 people in 417 households, 315 of which are categorized as poor. Most residents within the four hamlets work in agriculture. It takes three hours on poor roads to reach the district capital, and limited transportation options impede access to the district hospital. Mogiye has a pre-school and an elementary school, but the middle and high schools are an hour’s walk away. It has one polindes clinic, and one village midwife, as well as 20 traditional healers used by many residents. There are four posyandu staffed by 21 volunteers, half of whom are inactive (but receive incentives from the kepala desa, who is a relative). A dedicated posyandu building was built in 2009 with funds from the PNPM Rural program. Each hamlet’s posyandu takes turns using it on different days. 57 Mogiye has received support from recently completed programs such as: PNPM Rural (phased out in 2014), AIPMNH (2012–15), and Plan International. Petis Petis’s population of 63,989 is located in 16,273 households across 12 villages. Most residents work in the agricultural sector or overseas in Malaysia. While most roads provide good access, three villages have difficult access due to the hilly terrain. The subdistrict has a new Type-D hospital as well as two puskesmas, one in Petis (with PONED status), and another in Tampujong (with overnight inpatient status). Most villages in Petis have a midwife, however due to large populations or difficult access, this is not always enough. One of the puskesmas has a “waiting house” near the clinic paid for by the district government for patients who come from distant villages. Education facilities from elementary through high school are generally easy to access, although there is insufficient access for students with special needs. Lelaok Lelaok has a population of 3,896 in 1,616 households, spread out across five hamlets; 784 of these households are registered as poor. Most work in agriculture, but some work abroad. The village is 3 km from the subdistrict seat and 2.5 hours from the district capital district. There are education facilities in the village from PAUD through tertiary education. It also has one polindes clinic with a village midwife and a nurse. Lelaok also has five posyandu with 25 volunteers. All babies in the village are delivered with assistance from accredited health service providers. Generasi has been active in Lelaok since 2007, supporting posyandu-related activities such as supplies, incentives for volunteers, food supplements, as well as providing school uniforms. Sogiyan Sogiyan has 3,005 inhabitants in 879 households, 470 of which are classified as poor. Most residents are farmers, but some work in Surabaya as laborers. It is 45 minutes from the district capital but 75 minutes Pamekasan from the subdistrict seat. The village has private religious schools from PAUD through high school; not all teachers have proper 58 qualifications. The secular government elementary school closed in 2013 because of low enrollment. Sogiyan has one polindes staffed by one village midwife, as well as three posyandu with six volunteers. Each posyandu is held in a different location: the polindes, the kepala desa’s house, and at a volunteer’s home. The nearest puskesmas clinic is 4 km away. Sambingan Sambingan subdistrict has 27 villages, all with flat and low topography and good roads. Its population is 80,388, in 18,646 households. Most residents work in agriculture and forestry, while many others work outside the district. Sambingan subdistrict has education facilities under the authority of the Education or Religion ministries, from the PAUD and elementary level up through high school as well as several religious boarding schools in Sambingan. Sambingan subdistrict has two puskesmas clinics and five satellite clinics. It has three general practitioners, 13 nurses, and 41 midwives who are based in clinics and villages. The main puskesmas is in the process of upgrading to PONED status by building a delivery room and nutrition section. At the village level there are polindes or poskesdes clinics, and monthly posyandu. The polindes and poskesdes clinics were built using health department funds supplemented with village budgets and PNPM Rural support. PNPM Rural built 17 polindes built in Sambingan subdistrict, some of which are difficult to access. The puskesmas assists with funding food supplements for posyandu. Tespates Tespates’s 1,932 residents live in 450 households within five hamlets. It is the subdistrict seat. Although the village has elementary, middle, and high schools, families prefer to send their children to religious boarding schools; the village thus has one of the lowest school enrollments for elementary and middle schools in the district. In addition to the subdistrict puskesmas, Tespates also has a village midwife and three posyandu with 15 volunteers, though only nine are currently active. Posyandu are typically held in the kepala desa’s or a volunteer’s house. Tespates has a polindes, which was built with PNPM Rural funds, but it remains unused because it is considered too far away. 59