22809 Local Level Institutions Working Paper No. 5 Local Level Institutions LOCAL INSTITUTIONS AND SERVICE DELIVERY IN INDONESIA By Christiaan Grootaert The World Bank Social Development Family Environmentally and Socially Sustainable Development Network April 1999 LOCAL LEVEL INSTITUTIONS WORKING PAPER SERIES #1 The Local Level Institutions Study: Overview and Program Description (also available in Spanish and French) #2 The Local Level Institutions Study: Program Description and Prototype Questionnaires #3 Estudio Sobre Las Instituciones Locales En Bolivia: Panorama General Y Descripcion, Muestreo Y Recoleccion De Datos, Y Encuestas (by Godofredo Sand6val, Julio Cordova, Beatriz Ascarrunz, Alifredo Balboa, Griselda Gonzales, and Gloria Velasques) (only available in Spanish) #4 Grassroots Organizations and Local Development in Bolivia: A Study of the Municipalities of Tiahuanacu, Mizque, Villa Serrano, and Charagua (by Godofredo Sand6val, Julio Cordova, Beatriz Ascarrunz, Alfredo Balboa, Griselda Gonzales, and Gloria Velasques) (also available in Spanish) U5 Local Institutions and Service Delivery in Indonesia (by Christiaan Grootaert) #6 Social Capital, Household Welfare and Poverty in Indonesia (by Christiaan Grootaert) REQUESTS FOR DOCUMENTATION These and other documents may be obtained from Ms. Susan Assaf, The World Bank, Social Development Department, 1818 H. Street, Room MC5406, Washington, DC 20433. Tel. 202-473-1763 Fax. 202-522-3247 Email. sassaf@worldbank.org Papers in the Local Level Institutions Working Paper Series are not formal publications of the World Bank. They are published informally and circulated to encourage discussion and comment within the development community. The findings, interpretations, judgements, and conclusions expressed in this paper are those of the author(s) and should not be attributed to the World Bank, to its affiliated organizations, or to members of the Board of Executive Directors or the governments they represent. Local Institutions and Service Delivery in Indonesia 1 Table of Contents Page Acknowledgment ........................................ 2 1. Introduction: Local Institutions and Social Capital ..................................... 3 2. The Data Set ........................................ 8 3. The Map of Local Institutions ........................................ 24 4. Characteristics of the Most Important Associations .................................. 44 5. Social Capital and Poverty ........................................ 51 6. Who Provides Services? ........................................ 58 7. Collective Action and Trust ........................................ 63 8. Summary and Conclusions ........................................ 70 Annex: Tables by Province ........................................ 75 References ........................................ 80 Local Institutions and Service Delivery in Indonesia 2 Acknowledgment The author would like to thank Gloria Davis and Scott Guggenheim for helpful comments and suggestions on an earlier version of this paper. The Local Level Institutions (LLI) study was conducted under the leadership of Gloria Davis, Director, Social Development Department. In the initial phases, task Managers were Anthony Bebbington (July 1995-June 1996) and Christiaan Grootaert (July 1 996-December 1997). The final phase of the study was undertaken as a joint venture between the ESSD and PREM Networks and co-managed by Christiaan Grootaert and Deepa Narayan. The study received financial support from the Government of Norway. The Bolivia country-study was undertaken by a team from the consulting firm Sinergia. Coordinator was Godofredo Sand6val and the research team consisted of Julio Cordova, Beatriz Ascarrunz, Afredo Balboa, Griselda Gonzales, and Gloria Velasquez. The Burkina Faso country-study was coordinated by Paula Donnelly-Roark. The field work was undertaken under the auspices of the Commission Nationale pour la Decentralisation. The Indonesia country-study was coordinated by Scott Guggenheim and the research team consisted of Kamala Chandrakirana, Pieter Evers, Sjari Manaf and Silvia Werner. The processing and management of the data files was the responsibility of Gi- Taik Oh and Kalpana Mehra. General research assistance was provided by Susan Assaf. Staff assistants for the study were Gracie Ochieng and Anju Sachdeva. Local Institutions and Service Delivery in Indonesia 3 1. Introduction: Local Institutions and Social Capital The objective of this paper is to provide a map of local institutions which exist in rural Indonesia and to investigate the extent to which they are involved in the delivery of services (health, education, credit, land registration). We focus on several key dimensions of local institutions: * density of membership * internal homogeneity i attendance of meetings 3 active participation in decision making i type of organization (state or community initiated; formal or informal) 3 membership dues This paper is part of the first-phase (descriptive) analysis of the data collected by the Local Level Institutions Study (discussed in section 2). This phase investigates several aspects of the functioning of local institutions. The role of local institutions in collective action is discussed in Werner (1998) and the role of village government is examined in Evers (1998). The second phase of the analysis undertakes a multivariate analysis of the role of local institutions in determining access to services by households and in affecting household welfare and poverty. In that setting, the impact of local institutions is Local Institutions and Service Delivery in Indonesia 4 compared with the impact of household assets such as education and physical capital (Grootaert, 1998). The literature contains an impressive number of case studies which document that local institutions play a key role in successful project design and in determining project sustainability.' The importance of local associations, whether formal or informal, has been demonstrated in almost all parts of the world and in a wide range of sectoral settings2: 3 Experience with irrigation and water supply projects in Cote d'Ivoire and Pakistan have indicated that effective water user associations improve project performance and reduce maintenance costs (Hino, 1993; Meinzen- Dick et al, 1995; Narayan, 1995; Ostrom, 1995). I We use the term "local institution" interchangeably with "local association" or "local organization". This follows the practice of most social science literature (Uphoff, 1993), but there is a subtle distinction between the two concepts. Uphoff (1993) defines institutions as "complexes of norms and behaviors that persist over time by serving collectively valued purposes" (p. 614), while organizations are "structures of recognized and accepted roles" (p. 614). Examples of institutions are money, the law, marriage. Organizations are PTAs, workers' unions, rotating credit associations. In some cases, the two terms overlap: the army is an institution as well as a group of soldiers, the parliament is a law-making institution as well as an association of law makers. As Uphoff (1993) argues, the distinction is a matter of degree, and organizations can become more or less "institutional" over time. 2 Additional case studies are cited in Uphoff (1993), Narayan (1995), Grootaert (1997), Krishna et al (1997), Uphoff et al (1998) and Woolcock (1998). Local Institutions and Service Delivery in Indonesia 5 * Involvement of community groups in forest management in Gujarat reduced violent conflict over the use of forests and led to cooperation between government and local people (Pathan et al, 1993). * In a remote area of Zimbabwe, a small community obtained joint management with government over wildlife resources, leading to better wildlife protection and increased revenue from safaris and tourism for both the community and government (Scoones and Matose, 1993). * Social capital in Indian villages enhanced the poor's ability to allocate resources efficiently and increased their resilience to hazards (Townsend, 1994). * Improved relations between civil servants and local associations in Northeast Brazil enhanced the effectiveness of municipal programs. When the staff of a Ceara health program focused on building trust with clients in the communities in which they worked, the quality and impact of the program increased-more families were served and infant mortality declined (Freedheim, 1988; Tendler, 1997). * The Grameen Bank in Bangladesh has successfully extended small loans to groups of poor women who are jointly responsible for repayment. Local Institutions and Service Delivery in Indonesia 6 Grameen Bank has two million borrowers and a loan recovery rate of 98% (Yunus, 1997). The debate surrounding the role of local institutions is frequently set in the context of social capital. While there are many definitions and interpretations of the concept of social capital in the literature, usually social capital refers to the institutions, the relationships, the attitudes and values that govern interactions among people and contribute to economic and social development (Coleman, 1988, 1990; Putnam, 1993; Woolcock, 1998). Social capital and institutions are hence not synonyms: social capital is not just the sum of institutions which underpin society, it is also the glue that holds them together (Serageldin, 1996). Social capital includes the shared values and rules for social conduct expressed in personal relationships, trust, and a common sense of "civic" responsibility, without which it would be hard to have a functioning society. At the local level, institutions or associations can be seen as one manifestation of social capital. It must be emphasized though that social capital can and does exist outside the context of local institutions (whether formal or informal). For example, two neighbors who help each other in times of trouble have social capital but may never embody their bond in an association. Vice versa, the mere presence of an association does not prove the existence of social capital. Local branches of political Local Institutions and Service Delivery in Indonesia 7 parties, with mandatory membership, are associations which may display little or no social capital. The way in which social capital, whether through local associations or in other manifestations, performs its role is centered around three mechanisms: the sharing of information among group members, the reduction of opportunistic behavior, and the facilitation of collective decision making (Grootaert, 1997). To assess the extent to which an association can potentially achieve these goals and effectively embodies social capital, it is necessary to look at membership conditions (voluntary or not, payment of fees, etc.) and the degree of effective participation by members. The discussion below will highlight some of these aspects. Before presenting results, we discuss in the next section the data set, and the comparative study of which it is part. Local Institutions and Service Delivery in Indonesia 8 2. The Data Set The data set for this paper comes from the Local Level Institutions (LLI) Study, a comparative study of three countries (Bolivia, Burkina Faso and Indonesia), which aims to investigate the role of local institutions in providing service delivery and in affecting welfare and poverty outcomes.3 Data were collected at the level of the community, the district and the household. At the level of the community, interviews with focus groups of households and with community leaders were held to establish a map of functioning institutions in the community. Three instruments were used: 3 Information on community services was obtained through interviews with key informants such as village chief, teacher, health provider, etc. This was supplemented with information on the local economy (infrastructure and distance to markets), local society (ethnic/religious composition) and local institutions. Recent experience with selected development projects was also discussed. * The community services were also discussed with groups of households, with an objective to learn the community's perspective on the quality of 3 The objectives of the Local Level Institutions Study and the questionnaires are further discussed in World Bank (1998). Local Institutions and Service Delivery in Indonesia 9 service, its experience with collective action, and its views on local institutions and development projects. * For the most important local institutions, interviews were held with leaders and members, as well as with non-members, in order to get a balanced view of the role of the institutions in the village, their development over time, their main activities, relations with other institutions and government, and their main strengths and weaknesses. At the district level (defined as the administrative level above the village or community), data were collected about the extent of service coverage and the institutional arrangements for the provision of services. Information was also obtained about the general functioning of the district administration and its relation with civic organizations, through interviews with general and sectoral managers at the district level. The third and critical part of the data collection was a household survey which aimed to capture households' actual participation in local institutions, their use of services, and information that identifies the welfare level of households and their coping strategies. The questionnaire consisted of six sections: * demographic information on household members * participation in local institutions Local Institutions and Service Delivery in Indonesia 10 * characteristics of the most important groups * service provision profiles * perceptions of community trust and collaboration * household economy and coping strategies. The limited resources available did not make possible a sampling framework such that the studies would be representative of the countries at the national level. Instead, three or four areas were selected in each country (municipios in Bolivia, provinces in Burkina Faso and Indonesia), which represent different economic, social and institutional environments. In the case of Indonesia, the collected data cover the rural areas of three provinces: Jambi, Jawa Tengah, and Nusa Tenggara Timur (NTT). Jambi is located on the island of Sumatra. It is a tropical forest area which was only recently colonized and is still an agricultural frontier zone. It is characterized by low population density and its socio-economic indicators are close to Indonesian averages or slightly below (Table 1). Among the three provinces, Jambi has the lowest level of inequality in the distribution of household expenditure. Jawa Tengah is in the center of the island of Java, about 500 kms away from Jakarta. It has a very high population density (867 people/kM2) and is the most urbanized of the three provinces. It also has the highest income level and the best access to education and health services and to housing amenities. The population of both Jambi and Jawa Tengah is 99% Muslim. Local Institutions and Service Delivery in Indonesia 11 NTT consists of a series of islands in the eastern part of Indonesia (about 2500 kms and two time zones away from Jambi) and is the poorest and least urbanized of the three study areas. It relies heavily on traditional and subsistence-oriented agriculture and fewer than 5% of its economically active population have wage-jobs. The population is almost entirely Christian, evenly divided between Catholics and Protestants. Table 1: Selected Socio-economic Indicators of the Three Study Areas Jambi Jawa Nusa Indonesia l Tengah Tenggara I ~~~~ ~ ~ ~~~~~~~~~Timurl Population ('000) 2,370 29,653 3,577 194,755 Area ('000 km2) 44.8 34.2 47.9 1,919.3 Population Density (people/km2) 53 867 75 101 % Urbanized 27.2 31.9 13.9 35.9 % of Households with Access to Electricity 30.5 71.1 14.5 57.2 Gross Primary Enrollment Ratio 95 97 91 95 Gross Secondary Enrollment Ratio 47 58 44 56 % of Heads of Household who are Farmers 71 67 92 77!' Household Expenditure per Capita ('000 575.3 612.4 453.8 547.1 ' Rupiah/year)2 Gini-coefficient 0.29 0.36 0.37 0.35'| I / Based on the three study areas only. 2/ At the time of data collection (Fall 1996), the exchange rage was in the range $1 = 2,300-2,400 Rupiah. [ources: Statistical Yearbook of Indonesia 1995; Statistik Pendidikan 1994/95; Penduduk Indonesia, Jambi, Jawa Tengah, NTT-Hasil Survei Penduduk Antar Sensus 1995; author's calculations. Within each province two districts (kabupaten) were selected to participate in the study, and within each district two sub-districts (kecamatan) were selected. These units were selected purposively so as to represent a range of social, economic and institutional situations. Within each sub-district, four villages were selected based on location criteria (upland/lowland and near/far to growth center). Within each of the Local Institutions and Service Delivery in Indonesia 12 48 villages, 25 households were selected randomly to participate in the household survey. Table 2 summarizes the LLI Study sample for Indonesia.4 Table 2: The Local Level Institutions Study Sample for Indonesia Province District Number of (Kabupaten) Households Jambi Sarolangon Bangko 200 Batanghari 200 Jawa Tengah Banyumas 200 Wonogiri 200 Nusa Tenggara Timur Timor Tengah Selatan 200 Ngada 200 Total _ 1,200 In order to evaluate whether the characteristics of the respondent, specifically gender and position in the household, would have an effect on responses, the field work was organized to attempt to interview roughly 50% men and 50% women. Table 3 indicates that this objective was fairly well achieved: for the study zone as a whole, 57% of respondents were male and 43% were female. Roughly two-thirds of the respondents were heads of household and one-third were spouses. 4 Data were collected in the fall of 1996, i.e. prior to the start of the recent social and economic crisis in Indonesia. Thorbecke (1998) provides an initial assessment of the social costs of the crisis. Local Institutions and Service Delivery in Indonesia 13 Table 3: Selected Characteristics of Respondents (in %) Jambi Jawa Nusa All Tengah Tenggara Timur Gender of Respondent -Male 62.7 56.2 51.2 56.7 - Female 37.3 43.8 48.8 43.3 Position in Household - Head 72.2 62.2 59.7 64.7 - Spouse 27.8 37.8 40.3 35.3 Language -Bahasa Indonesia 1.0 9.5 36.5 15.7 - Javanese 1.3 50.5 0.2 17.3 - Melayu 41.8 40.0 0.0 27.2 - District 0.8 0.0 23.5 8.1 -Sub-district 28.8 0.0 18.5 15.7 -Other 26.5 0.0 21.3 15.9 Religion - Muslim 99.2 98.7 2.3 66.7 - Buddhist 0.3 0.5 0.2 0.3 - Hindu 0.5 0.0 0.2 0.3 - Catholic 0.0 0.3 50.8 17.0 -Protestant 0.0 0.5 46.5 15.7 Average Years Lived in 32.5 39.6 31.7 34.6 Community Table 3 highlights the regional breakdown of religion and language spoken. While at the country level, Indonesia is a linguistically and ethnically diverse society, this is not the case at the province level (Bachtiar, 1994). In Jambi and Jawa Tengah almost all respondents are Muslim, while in NTT almost all respondents are Christian, evenly divided among the Catholic and Protestant religion. At the village level, respondents have a strong perception of religious and ethnic homogeneity. During Local Institutions and Service Delivery in Indonesia 14 focus group interviews 90% of villages were described as being "of one religion" and more than 75% of villages were seen as "of one ethnic group." Each province has its own pattern of languages spoken. In Jambi, the predominant language is Melayu with sub-district and other languages accounting for the remainder. In Jawa Tengah, the main languages are Javanese and Melayu. NTT shows the greatest diversity of languages: Bahasa Indonesian, district, sub-district and other languages are spoken. It is important to keep the linguistic and religious patterns in mind for the analysis below which for the most part will follow a breakdown by province. This is not just a locational breakdown, but it corresponds also to a religious breakdown, and the interpretation of some of the observed institutional patterns will have to be seen in that light.5 Table 4 provides a broad breakdown of some of the key economic characteristics of the households in the sample.6 Almost one-fifth of the adult sample population is inactive. This figure is lowest in NTT (14.3%) and highest in Jambi 5 The predominance of Islam in Jawa Tengah and Jambi does not imply homogeneity of culture and values in these areas. Historically, different socioeconomic groups in Indonesia have absorbed Islam gradually and in different ways. Traditional structures and values have continued to exist in differing degrees and Islamic teachings have been adopted to these traditions. This has led to a generally moderate form of Islam (Bachtiar, 1994). Geertz (1968) has argued that in Indonesia Islam has been more a force for cultural diversification than for uniformity. 6 It is to be noted that Table 4 is calculated over all adult individuals in the sample whilst Table 3 pertained only to the respondents. Local Institutions and Service Delivery in Indonesia 15 (22.0%). Roughly 80% of economically active household members are farmers. This percentage is higher in NTT and lower in Jambi and Jawa Tengah. The vast majority of economically active people work as self-employed or unpaid family workers, although there are some notable regional differences. Wage earners are virtually absent in NTT but constitute a third of the economically active population in Jawa Tengah. By the same token, unpaid family workers are most prevalent in NTT and least prevalent in Jawa Tengah.7 In terms of education, 14% of the adults in the sample have no education. Two-thirds have undertaken part of primary education or completed it. Only 19% of the sample population have undertaken schooling beyond the primary level, with the incidence of this being highest in Jawa Tengah. Graduates of vocational schools or universities are very rare in the rural areas in the three provinces of the study. 7 Economic differences between provinces and regions in Indonesia are discussed further in World Bank (1996). The latter study also reviews the macroeconomic situation of the 1990s. Thorbecke (1991) reviews the macroeconomic situation of the 1980s. Local Institutions and Service Delivery in Indonesia 16 Table 4: Employment and Education Characteristics of the Adults (aged 15 and above) in the Sample (percentage) Jambi Jawa Nusa All Tengah Tenggara Timur Main Occupation - Farmer/Fisherman 60.8 54.1 78.6 65.4 - Trade 4.2 7.3 0.4 3.8 - Other Private Sector 10.3 15.3 5.0 9.9 -Public Sector 2.7 3.1 1.8 2.5 -Inactive 22.0 20.2 14.3 18.5 All 100.0 100.0 100.0 100.0 Employment Status of Economically Active People -Wage Earner 23.8 31.6 4.6 18.6 - Self-employed 41.9 54.1 51.3 49.3 - Unpaid Family Worker 34.3 14.3 44.1 32.1 All 100.0 100.0 100.0 100.0 Education - None 13.1 16.1 12.9 14.0 - Primary Incomplete 29.4 26.0 24.9 26.6 - Primary Complete 37.3 37.7 45.3 40.4 - Secondary Incomplete 11.3 13.6 9.4 11.3 - Secondary Complete 4.7 4.6 3.8 4.3 -Vocational 3.1 1.6 1.0 1.8 -University 0.6 0.2 0.6 0.5 - Other 0.5 0.3 2.1 1.1 All 100.0 100.0 100.0 100.0 The educational and employment differences across the three provinces translate into differences in household level of living (Table 5). Jawa Tengah has the highest household expenditure per capita level, which is 6% higher than Jambi and 34% higher than NTT. Local Institutions and Service Delivery in Indonesia 17 Table 5: The Level and Composition of Household Expenditure Jambi Jawa Nusa All Tengah Tenggara Timur Composition of Household Expenditure (%/o) Purchased Food 55.1 33.8 26.6 39.3 Home Produced Food 6.3 8.4 32.6 14.4 Education 5.1 3.0 4.9 4.3 Health 3.1 3.4 1.6 2.8 Housing 8.5 25.8 7.0 14.6 Private Transfers 3.7 2.4 6.2 3.9 Other 18.2 23.2 21.1 20.8 All 100.0 100.0 100.0 100.0 Household Expenditure per 575,265 612,366 453,768 547,133 Capita (Rupiah per year) Coefficient of Variation 0.64 0.74 0.76 0.72 Gini-coefficient 0.29 0.36 0.37 0.35 Ratio of Richest Over Poorest 2.1 2.2 2.3 3.2 Village _ - Poverty Incidence -Poverty Head Count* 11.3 18.0 39.0 24.3 - Poverty Gap** 26.5 23.6 34.0 30.6 * Percentage of population which is poor. This is based on a relative poverty line of two-thirds of the median of household expenditure per capita in the three provinces. ** The poverty gap equals the shortfall, relative to the poverty line, of expenditure per capita of the average poor person, expressed as a percentage of the poverty line. These are also some notable differences in the pattern of expenditure between the three provinces. For the country as a whole, households spend slightly more than half of their expenditures on food (including the value of home produced food). This figure is much lower though in Jawa Tengah (42% ) and much higher in the two other provinces. Equally noteworthy is the breakdown between purchased food and home Local Institutions and Service Delivery in Indonesia 18 produced food. In NTT the value of home produced food exceeds that of purchased food whilst in two other provinces, home produced food is only 6-8% of the total.8 The other major difference between the provinces is in the share of housing. In Jambi and NTT the housing share ranges between 7-8% whilst in Jawa Tengah it is 26%. This can probably be explained by the higher degree of urbanization and higher population density of Jawa Tengah. With regard to social services, households spend on average 4.3% of their expenditures on education and 2.8% on health. The share of education is however lower in Jawa Tengah and higher in the two other provinces. The reverse pattern is observed for health. Finally, it is remarkable that the share of outgoing private transfers is highest in NTT, which is the poorest of the three provinces.' Table 5 shows three different measures of the distribution of household expenditure: the coefficient of variation, the Gini-coefficient, and the ratio of household expenditures of the richest over the poorest village. All three indicators suggest that the distribution of expenditure is the most equal in Jambi, followed by 8 The lower food share of Jawa Tengah relative to the two other provinces is consistent with Engel's law. However, the food share of Jambi exceeds that of NTT even though the latter has a lower total expenditure per capita level. One possible reason for this discrepancy could be that the value of home produced food is based on respondents' self-estimation and this figure could be underestimated in the case of NTT. However, if households are ranked by per capita expenditure, regardless of location, Engel's law is met: the poorest quintile spends 69.1% on food, which decreases monotonically to 43.9% for the richest quintile. 9 If households are ranked by per capita expenditure, regardless of location, a different picture emerges. The share of household expenditure going to transfers is almost unchanged over Local Institutions and Service Delivery in Indonesia 19 Jawa Tengah, and the most unequal in NTT. This is surprising because NTT is the poorest of the three provinces. The Gini-coefficient for the three provinces put together is 0.35 which is slightly higher than the Gini-coefficient of 0.32 reported for Indonesia as a whole (World Bank 1997). We have also calculated the incidence of poverty based on a poverty line determined as two-thirds of the median household expenditure per capita for the three provinces. In line with the observed difference in expenditure per capita, NTT records the highest poverty incidence at 39% while the lowest incidence is in Jambi at 11.3%. Jawa Tengah has higher poverty than Jambi, in spite of the former's higher level of expenditure per capita. However, as was noted earlier, the distribution in Jawa Tengah is notably less equal than in Jambi, which explains the higher poverty figure. Still, the poverty gap is lower in Jawa Tengah than in Jambi. The overall poverty gap for the three provinces (30.6%) is fairly high. '° Household welfare is not only measured by the level of expenditures, but also by various other dimensions. Table 6 contains several dimensions which can be calculated from the LLI data set. Specifically, we show indicators relating to how often the household goes hungry, the forced sale of assets in order to buy necessities, the number of children aged 5 to 14 who do not attend school, and access to tap water the bottom three quintiles, at 7-8%, and then rises to 13.7% for the second richest quintile and 22.9% for the richest quintile. Local Institutions and Service Delivery in Indonesia 20 and electricity. The data are shown by quintiles which rank households based on household expenditure per capita. Table 6: Selected Indicators of Household Well-being Quintile | Percent of Households 1 2 3 4 5 All (Poorest) (Richest) Which often go hungry 18.8 11.3 9.6 6.3 0.8 9.3 Which sold land, 32.5 27.9 23.3 18.8 22.9 25.1 livestock or equipment in order to buy necessities Where not all children 33.1 22.4 27.1 16.9 10.9 22.6 5-14 attend school Which pulled children 10.3 6.7 2.6 3.8 1.5 5.2 out of school to help on farm or at home With access to electricity 23.8 35.0 29.2 48.8 50.8 37.5 With access to tap water 29.6 18.3 18.8 17.5 17.9 20.4 Table 6 suggests that only 9% of sample households indicate that they often go hungry in the course of the year. However, the rate is twice as high among the poorest households. One in four households declare that in the past year they had to sell land, livestock or equipment in order to buy necessities. This figure varies less pronouncedly by quintile and is actually slightly higher for the richest households than the next richest quintile (presumably because the richest households have more assets to sell). 10 The evolution of poverty and inequality in Indonesia is discussed in detail by Tjiptoherijanto (1996). Local Institutions and Service Delivery in Indonesia 21 About one in five households does not send all of its school-age children to school. The incidence of this is much higher among the poor than among the rich: in the poorest quintile, one in three households does not send all of its children to school while in the richest quintile it is only one out of ten households. On the other hand, it is relatively rare for households to pull attending children out of school in order to work at home or to work on the farm. Only 5.2% of households with children in school report having had to do this in the past year, but this rate is twice as high in the poorest quintile. In terms of access to amenities, electricity is more available to the rich than to the poor although in comparison to many other countries the differential access between the poorest and the richest is not extreme. Lastly, access to tap water is not differentiated by level of income. Table 7 shows the ownership of various assets by the household, such as household durables, land and animals. The assets most frequently owned by the sample households are a radio (present in almost half of all households) a pressure lamp, a bicycle, and an audio system. There is a very distinct and somewhat peculiar regional pattern to this ownership. In Jambi all assets relating to transportation are more frequently owned than elsewhere. This is the case for car/truck, boat and outboard motor, motorcycle and bicycle. In contrast, modem consumer electronics such as television, audio system and radio are all most frequently owned in Jawa Local Institutions and Service Delivery in Indonesia 22 Tengah. In general, the ownership of household durables is lowest in NTT. The average household in NTT owns only 1.3 durables against 2.3 in Jawa Tengah and 3 in Jambi. The only asset which is more frequently owned in NTT than elsewhere is the pressure lamp (because most households in NTT have no access to electricity). The pattern of large animal ownership (i.e. excluding poultry and fish ponds) goes the other way, with the average NTT household owning 7.7 large animals against only 1.3 animals in Jambi. Land ownership is very low in the highly densely populated Jawa Tengah area, where the average household owns or rents less than 1 hectare. In NTT, the average household owns or rents 2.1 hectares and in Jambi this is 3.2 hectares. Local Institutions and Service Delivery in Indonesia 23 Table 7: Ownership of Household Durables, Land and Animals Jambi Jawa Nusa All Tengah Tenggara _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ T im ur Percent of Households Owning Car/Truck 2.5 1.5 0.0 1.3 Boat 24.5 0.3 0.8 8.5 Outboard Motor 2.3 0.3 0.3 0.9 Motorcycle/Moped 16.3 10.3 1.3 9.3 Bicycle 42.3 36.5 13.0 30.6 Color TV 7.0 13.3 1.0 7.1 Black/White TV 20.5 22.8 2.3 15.2 Audio System 36.8 38.5 14.0 29.8 Telephone 0.5 1.5 0.0 0.7 Radio 54.3 62.0 22.8 46.3 Refrigerator/Freezer 1.0 3.0 0.5 1.5 Sewing Machine 18.8 8.8 5.8 11.1 Washing Machine/Dryer 3.3 1.5 0.0 1.6 Pressure Lamp 42.3 23.3 52.0 39.2 Other Major Durable 26.3 5.5 12.0 14.6 Average Asset Score 3.0 2.3 1.3 2.2 Average Number of Large 1.3 2.3 7.7 3.8 Animals Owned Average Hectares of Land 3.2 0.9 2.1 2.1 Owned or Rented In summary, the LLI Study sample shows three very diverse regions. The richest province is Jawa Tengah, with the highest ownership of consumer electronic durables. Households have the highest education level and a larger percentage of them work as wage earners than elsewhere. The distribution of household expenditure is fairly unequal which leads to higher poverty rates than in Jambi which has a lower average expenditure level. The NTT province is distinct from the others in several respects: it is the only province which is almost entirely Christian, it is the Local Institutions and Service Delivery in Indonesia 24 poorest area, and its employment pattern is the most heavily skewed towards agriculture and home production of foods. Asset ownership is low but ownership of animals is high. Against this general background of some of the economic characteristics of the households in the three provinces, we now turn to the discussion of associational membership. Local Institutions and Service Delivery in Indonesia 25 3 The Map of Local Institutions Indonesia has a long tradition of community-based groups (swadaya), which are mostly informally organized (Werner, 1998). The Indonesian government has established many new groups with national scope to promote its objectives locally and bring together people with the same occupations or concerns (e.g., national association of teachers, national organization of housewives, health and family planning groups). In 1979, the government passed the Village Governance Law which put in place a new structure of local government based on neighborhoods (RT/RW) and hamlets (dusun) within villages (Evers, 1998). The key feature of the government-sponsored groups is that they are formally organized and have mandatory membership (Werner, 1998). However, both community-based and government- sponsored associations are found across the functional spectrum of associations (social service groups, production and occupational groups, finance and credit groups, etc.). The three Indonesian provinces in the LLI Study have a very dense and rich associational life at the local level. The LLI Study's questionnaire accounted for 14 different kinds of institutions, and in total 7,187 memberships in these organizations were recorded. This means that on average each household belongs to six associations. Local Institutions and Service Delivery in Indonesia 26 Across the three provinces, 210 different local associations were recorded. They range from broad national level organizations with many local branches, such as the ruling political party, the national organization for Indonesian housewives (PKK) and the farmers' association (Kelompok Tani) to very local groups, such as the local high school drum band, a group of swallow nest farmers, and a local saving group. Of the 7,187 memberships that were recorded in the data, households indicated in 6,210 cases that one or more of its members "participated actively" in the association. Only those cases have been retained for further analysis. Table 8 lists the distribution of those cases over the three provinces. Local Institutions and Service Delivery in Indonesia 27 Table 8: Active Memberships in Local Institutions, by Province (original classification) Jambi Jawa Tengah Nusa Tenggara All % of Timur Total Community-wide 187 14.7% 510 21.6% 443 17.2% 1140 18.4 Organizations l Sub-community 59 4.6% 239 10.1% 297 11.5% 595 9.6 Gender Groups l Sub-community 52 4.1% 79 3.4% 82 3.2% 213 3.4 Generation Groups l Religious Groups 317 24.9% 270 11.5% 395 15.3% 982 15.8 Social Service 266 20.9% 150 6.4% 208 8.1 % 624 10.1 Groups l Environmental 0 0.0% 53 2.3% 318 12.3% 371 6.0 Service Groups Production Groups 65 5.1% 116 4.9% 263 10.2% 444 7.1 Labor Groups 87 6.8% 26 1.1% 258 10.0% 371 6.0 Natural Resource 4 0.3% 36 1.5% 29 1.1% 69 1.1 Management Groups l Credit/Finance 194 15.2% 823 34.9% 258 10.0% 1275 20.5 Groups l Marketing Groups 0 0.0% 7 0.3% 0 0.0% 7 0.1 Supra-community 8 0.6% 30 1.3% 4 0.2% 42 0.7 Level Membership Organizations Recreational Groups 16 1.3% 17 0.7% 22 0.9% 55 0.9 Other 18 1.4% 3 0.1% 1 0.0% 22 0.4 Total 1273 100.0% 2359 100.0% 2578 100.0% 6210 100.0 Credit and finance groups are the most prevalent form of association, accounting for one-fifth of all memberships. This is followed closely by community- wide organizations which represent 18.4% of memberships and religious groups which account for 15.8% of memberships. There is a distinct regional pattern: Jawa Tengah and NTT each account for about 40% of memberships, which means that each household participates actively in approximately six associations. In contrast, the Local Institutions and Service Delivery in Indonesia 28 density of associations is much lower in Jambi: each household participates actively in about three local organizations. The classification in Table 8 was designed as a general reference for the LLI Study, and needed some adaptation in the case of Indonesia. First, a number of classifications proved to be virtually absent in the organizational pattern in Indonesia: natural resource management groups, marketing groups, supra-community level membership organizations and recreational groups as well as the residual category each accounted for 1% or less of the total cases. Second, in the course of collecting the data, several enumerators had reported difficulties with classifying organizations due to the overlap of functional classification with the nature of the group. For example, a church group of teenagers who attend Sunday school could be classified either as a religious group or as a sub-community generation group, depending upon whether one puts most emphasis on the functional aspect or the age aspect. While consistency of classification was achieved within each enumeration team, it proved necessary to make a number of adjustments at the national level in order to ensure consistency of classification. Priority was given to the functional classification, while gender and age dimensions and whether the group was at the community, sub- community or supra-community level would become a secondary cross-classifying variable. Each organization was subsequently reassigned a new code which led to 10 rather than 14 different categories. The new classification is described in detail by Werner (1998) and summarized in Box 1 and Table 9. Local Institutions and Service Delivery in Indonesia 29 Box 1: Types of Local Level Institutions in Indonesia 1. Social Service Groups Social service groups provide assistance and services to community members in the fields of education, health, pest management, security, general neighborhood assistance and development. Among the social service groups are some of the large government organizations which have to be present in every village. Especially worth mentioning are women's group for family welfare (PKK), and its sub-units at the neighborhood level (Dasawisma), the integrated health service post (Posyandu), the organization for birth-control (Kelompok KB) and. the primary school pupils' parents organization (BP3). Within community-based social service groups there is a wide variation concerning their scope of activities and their role for the community. Some groups are concerned with limited, specific purposes, like the assistance of families where somebody has died (burial societies) or the collective purchase of dishes used for large festive occasions. Other groups have a wide scope of activities and a large role in the village community, sometimes even exceeding those of the formal village government. These kind of groups are customary (adat) organizations which can be found in Jambi and NTT. They regulate customs, settle disputes and impose sanctions. 2. Production Groups This category encompasses those associations founded for the purpose of mutual assistance in the production of goods to facilitate an increased production capacity. All groups concerned with agricultural production are classified as production groups. Other members of this category are livestock groups, small industry and handicraft groups, and groups concerned with the harvest of forest products. 3. Occupational Groups Groups mobilizing the labor force of its members to achieve a common purpose or to solve a common problem are classified as occupational groups. Although a major part of these work groups consists of mutual help groups for field work, be it in gardens or rice fields, these groups also contribute to the maintenance of village infrastructure. 4. Credit and Saving Groups The category consists of groups providing credit facilities and informal groups with rotational saving activities (arisan). In Central Java these groups are especially important, because traditional rotational saving groups originate from that island. Besides their saving and credit role, arisan in Central Java probably have an even more important social role in bringing people together. 5. Religious Groups Local level institutions with religious activities like collective praying or related to religion like groups for the construction of mosques or churches are classified as religious groups. They are very important for bringing the community together, and function as a forum for meeting and exchange. Local Institutions and Service Delivery in Indonesia 30 6. Natural Resource Management Groups These groups target the use and management of natural resources. The most prominent are irrigation groups which manage water resources for the purpose of paddy rice irrigation. Some government-initiated groups deal with regreening, social forestry and terracing activities. 7. Environmental Service Groups Environmental service groups represent the smallest category with four different groups present in Jawa Tengah. They are concerned with mutual help for the maintenance of public facilities (cleaning roads, graveyards, religious facilities and the like). 8. Governance Groups This category consists of groups concerned with village governance. It encompasses the village government itself as well as the Village Development Council (LKMD) and Village Deliberation Council (LMD). There are no informal governance groups and the existing institutions and structures are the same all over Indonesia. The lowest level of administration is the RT (Rukun Tetangga = smallest neighborhood unit), followed by RW (Rukun Warga = intermediate neighborhood unit), and the Dusun (hamlet, largest neighborhood unit). These administrative units reflect traditional settlement entities and are the main locus of collective action and mutual help within the village. In Jawa Tengah and NTT, RT and Dusun have a high profile not because of their government task, but because they perform many social tasks. In Jambi, on the other hand, the social functions are taken over by other groups (e.g. religious and customary groups). 9. Recreational Groups Associations concerned with leisurely pastimes like sports, arts and cultural activities are classified as recreational groups. A large part of these groups are initiated by youths. There is only one government-initiated group, Karang Taruna. Source: Werner (1998). Local Institutions and Service Delivery in Indonesia 31 Table 9: Number of Local Organizations and Number of Memberships (Functional Classification) Type of Group Number of Number of Examples Organizations Active Memberships Social Services Group 44 1695 27.3 Group raising funds for wedding parties or funerals; national organization of Indonesian housewives; night watch group; child care group. Production Group 27 235 3.8 Association of farmers; group of women doing weaving activities; group of swallow nest farmers. Occupational Group 16 360 5.8 Mutual help group for rice cultivation; group for building fences; group of harbor workers. Finance/Credit Group 45 1195 19.2 Rotating savings group; villagers' credit union; group to assist poor villages. Religious Group 35 1147 18.5 Christian prayer group; Koran recital group; communal congregation; Catholic youth group. Natural Resources Group 9 173 2.8 Group to maintain irrigation channels; group for developing community-managed forests. Environmental Services Group 2 10 0.2 Group for cleaning village roads; group for soil rehabilitation. Governrment Affairs Group 6 1179 19.0 Village deliberation council (LMD); neighborhood and hamlet governance groups (RT/RW, Dusun). Recreation Group 18 197 3.2 Village youth group; soccer team; traditional dance'l __________________________ _______ ~~group. Other 8 19 0.3 The ruling political party (TOLKAR); group of children of members of the armed forces. Tl'otal 210 6210 100.0 Local Institutions and Service Delivery in Indonesia 32 Social services groups and finance and credit groups account for the most organizations (44 and 45 associations, respectively). However, in terms of number of memberships, social services groups account for by far the most memberships with 27.3% of the total. This is partly explained by the fact that 20 of the 44 social service groups are government-sponsored or national groups with mandatory memberships. More than half of all memberships in this category are in the national association of Indonesian housewives for improving family welfare (PKK) and its sub-units at the neighborhood level (Darawisma), and the integrated health centers (Posyandu). These formal associations are especially strong in Jawa Tengah, and weakest in Jambi where the community's own initiative for organizing social services is very strong (Werner, 1998). After social service groups, memberships are most concentrated in finance and credit groups, religious groups, and government affair groups, who each have 18% to 19% of total memberships. The high membership of government affairs groups follows from the mandatory character of the membership of these organizations. This explains why only six organizations account for almost one-fifth of all memberships. Those six organizations comprise the civil defense system (Hansip) and five organizations involved in the administration of the village (LKMD, LMD, Perangkat Desa) or neighborhoods and hamlets within a village (RT/RW, Dusun). ' 11 Evers (1998) discusses the village administration structure in detail. Local Institutions and Service Delivery in Indonesia 33 The regional breakdown of the memberships is shown in Table 10. Social service groups are equally important in each of the three provinces. Memberships in finance and credit groups, government affairs groups and religious groups show distinct regional concentrations. More than 60% of all memberships in finance and credit groups occur in Jawa Tengah, where they represent almost one in every three memberships. Religious groups are most strongly concentrated in Jambi, where they constitute 27.3% of all memberships (against only 11.5% in Jawa Tengah, and 20.4% in NTT). In Jambi, religion and related associational life plays a big role in maintaining community cohesion and social values. In NTT, religious groups frequently cooperate with neighborhood groups to undertake collective action. Religion in Jambi and NTT thus has a strong function of bringing the community together and serving as a communication forum which is often the source of development activities. The lower presence of religious groups in Jawa Tengah is related to Javanese traditions and the pre-Islamic Hindu-Buddhist culture which is still a large part of daily life in that province (Werner, 1998). Local Institutions and Service Delivery in Indonesia 34 Table 10: Active Memberships in Local Organizations, by Province Jambi Jawa Tengah Nusa Tenggara All % of Timur Total Social Service 346 27.2% 616 26.1% 733 28.4% 1695 27.3 Group Production Group 45 3.5% 61 2.6% 129 5.0% 235 3.8 Occupational Group 113 8.9% 41 1.7% 206 8.0% 360 5.8 Finance/Credit 148 11.6% 740 31.4% 307 11.9% 1195 19.2 Services Group Religious Group 348 27.3% 272 11.5% 527 20.4% 1147 18.5 Natural Resources 19 1.5% 89 3.8% 65 2.5% 173 2.8 Group Environmental I 0.1% 9 0.4o/ 0 0.0% 1 0 0.2 Services Group I I Governmental 189 14.9% 418 17.7% 572 22.2% 1179 19.0 Affairs Group Recreation Group 63 5.0% 97 4.1% 37 1.4% 197 3.2 Others I 0.1% 16 0.7% 2 0.1% 19 0.3 Total 1273 20.5% 2359 38.0% 2578 41.5% 6210 100.0 Government affair groups are strongest in NTT (22.2%), followed by Jawa Tengah (17.7%) and Jambi (14.9%). Among the smaller categories of associations, production and occupational groups are most prevalent in NTT, while natural resources and environmental services groups have the highest representation in Jawa Tengah. Local Institutions and Service Delivery in Indonesia 35 Table 11: Percentage of Households Participating in Local Associations, by Region Province Jambi Jawa Nusa All Tengah Tenggara Timur Social Service Group 57.5 72.8 79.0 69.8 Production Group 10.5 14.3 29.0 17.9 Occupational Group 25.0 9.8 39.0 24.6 Finance/Credit Services Group 32.3 84.8 52.5 56.5 Religious Group 57.8 58.0 89.3 68.3 Natural Resources Group 4.5 18.5 15.8 12.9 Environmental Services Group 0.3 2.3 0.0 0.8 Government Affairs Group 43.8 67.5 87.5 66.3 Recreation Group 14.8 21.3 7.5 14.5 Others 0.3 3.0 0.5 1.3 So far, we have tabulated the pattern of memberships in local associations, recognizing that a household can have multiple memberships (and counting each membership as one unit of observation). Another way of looking at this information is to ask: What percentage of households participate in what type of local association? This view is portrayed in Table 11 which shows that more than two- thirds of Indonesian households participate in a social service group, a religious group, or a government affair group. This means that although social services groups account for far more memberships than religious or government groups, the same fraction of households participates in each type of group. There is thus a strong concentration of social service group memberships within the same households. Local Institutions and Service Delivery in Indonesia 36 The percentages of participating households vary considerably by province. In line with the earlier observation of higher associational density in Jawa Tengah and NTT, we note that in NTT almost 80% of households belong to a social service group and almost 90% to a religious or government group. In contrast, in Jambi less than 50% of households participate in government affair groups and 57% participate in religious or social service groups. Finance and credit groups are concentrated the strongest in Jawa Tengah with 85% of households participating against only 32% in Jambi and 52% in NTT. About one-third of memberships in local associations relate to groups that have a distinct gender dimension (Table 12). Within those, two-thirds are for female- only and one-third for male only. The strongest presence of gender-oriented groups occurs in Jawa Tengah where 35% of all memberships are in female-only groups and an additional 20% in male-only groups, so that the majority of all memberships have a gender dimension. This is related to the strong presence of savings and credit groups in Jawa Tengah (Table 10) which are traditionally segregated by gender. In contrast, in the two other provinces gender-oriented groups account for only approximately one-fourth of all memberships. Local Institutions and Service Delivery in Indonesia 37 Table 12: Distribution of Associational Memberships by Gender, Age and Organizational Type Jambi Jawa Nusa All Tengah Tenggara Timur Gender - Female Only 18.3 34.8 17.0 24.0 - Male Only 6.9 20.0 8.3 12.5 - Mixed 74.8 45.2 74.7 63.5 All 100.0 100.0 100.0 100.0 Age - Youth 10.4 5.3 4.3 5.9 - Adult/Middle Age 63.5 67.9 70.2 68.0 -Elderly 0.0 0.4 0.1 0.2 - Mixed 26.2 26.4 25.4 25.9 All 100.0 100.0 100.0 100.0 Organization - Village-based Group with 22.6 37.7 56.1 42.3 Supra-Village Network/ Affiliation - Village-based Group without 75.5 58.4 41.3 54.8 Supra-Village Network/ Affiliation - Organization Based Outside 1.9 3.9 2.6 3.0 Village All 100.0 100.0 100.0 100.0 Table 12 also shows that age is not a particularly important membership criterion of associational life in Indonesia. Only 6% of groups have a specific age dimension in being oriented towards young people. Jambi has the strongest concentration of youth-oriented groups accounting for 10% of all memberships. There are virtually no groups that are oriented specifically towards the elderly, so that the bulk of the memberships are either oriented towards adults or are mixed for all ages. Local Institutions and Service Delivery in Indonesia 38 The way local associations are organized and fit in supra-village structures shows a strong regional variation. Three types of organizations are distinguished: (1) those that are village based and which have no affiliation with any organization outside the village-this is the most local of the local organizations; (2) village-based groups which have an affiliation with a supra-village network; and (3) organizations which find their origin outside the village. The latter category is fairly unimportant, accounting for only 3% of all memberships. The truly local groups which are village based and have no affiliation outside the village account for 55% of memberships. However, this percentage is much lower in NTT (only 41%); in part, this is related to the presence of religious groups in that province, associated with the Catholic or Protestant church, which by nature surpasses the area of the village. In contrast, in Jambi, groups without any association outside the village account for three-fourths of all memberships. This is related to the strength of traditional village structures in Jambi, which has meant that government-sponsored groups have a lower penetration than in the other provinces. Not all memberships in groups are equally important. Yet, in general, Indonesians take their associational life very serious: 79% of all memberships are rated as "very important", but this percentage is higher in NTT (88%) and lower in Jambi (67%) (Table 13). Memberships in production groups, occupational groups and religious groups receive the highest rating in each province. The lowest rating goes to environmental groups and recreational groups. Local Institutions and Service Delivery in Indonesia 39 Table 13: Percentage of Memberships in Associations Listed as "Very Important" Jambi Jawa Nusa All Tengah Tenggara Timur Social Service Group 61.8 71.6 88.1 76.8 Production Group 73.3 83.6 95.4 88.1 Occupational Group 70.8 85.4 96.6 87.2 Finance/Credit Services Group 57.4 73.7 82.7 74.0 Religious Group 88.5 84.9 91.1 88.8 Natural Resources Group 63.2 75.3 87.7 78.6 Environmental Services Group - 66.7 - 60.0 Governmental Affairs Group 52.4 75.6 85.3 76.6 Recreation 33.3 65.0 62.2 54.3 Others - 75.0 _ 73.7 Average 67.0 74.9 88.1 78.8 In the LLI Study's questionnaire, households were asked to identify which were the three most important groups to which they belonged. A series of supplementary questions were then asked about these groups, relating to the reasons for joining, membership contributions, active participation in decision making, the nature of the service provided by the group, the internal homogeneity of its membership, and whether the household occupied a leadership position in the group. Table 14 identifies the ten groups most frequently cited among the three most important groups. The government-created village governance structure (Village Governance Law of 1979), in which membership is mandatory, is cited most frequently as the most important local institution. In this structure, villages are divided into hamlets (dusun) with an average size ranging from 300 households in the Sarolangon Banko Local Institutions and Service Delivery in Indonesia 40 sub-district in Jambi to 1,200 households in the Banyumas sub-district in Jawa Tengah. Hamlets are further divided into RT (rukun tetangga = neighborhood solidarity unit) consisting of up to 30 households, and RW (rukun warga = community solidarity unit) which is formed by two or more RTs (Evers, 1998). The RT/RW structure is active and considered very important in Jawa Tengah and NTT, but in Jambi it functions less well and its importance is surpassed by the traditional village structure (of which Lembaga Adat-listed #10 in Table 14-is the prime institution). It is noteworthy that among the ten most important groups, all but two are either religious groups or groups that have been sponsored by the government, i.e., they are local associations with a supra-village structure. The only two genuinely local groups are No. 7 Arisan Dusun, locally based rotating savings groups, and No. 10 Lembaga Adat, the traditional groups which settle disputes and impose sanctions. In total, the ten most important groups in Indonesia (out of 210 groups) account for 52% of all memberships which were cited as among the three most important for Indonesian households. This indicates that local associational life is dominated by a limited number of groups. From an analytic point of view, it will be important to focus on the dimensions brought out by these groups, namely, whether or not they are national in scope or government sponsored, whether they are religious, and whether they are genuinely locally associated. Local Institutions and Service Delivery in Indonesia 41 Table 14: The Ten Most Important Local Associations Group Name Category Government Formal Number of or or Times Cited Community Informal (% of Total Initiated Group Citations) 1. RT/RW (neighborhood Government G F 326 (10.9%) governance group) I 2. Pengajian (Koran recital group) Religious C I 245 (8.2%) 3. IDT (group receiving assistance Finance/ G F 175 (5.9%) from government in poor villages) Credit 4. PKK (national organization for Social G F 166 (5.6%) Indonesian housewives) Services 5. Rayon (organization of Protestant Religious C F 122 (4.1%) church for church maintenance and ministerial support) 6. Kelompok Tani (association of Production G/C F/I 120 (4.0%) farmers) l 7. Arisan Dusun (hamlet or Finance/ C 1 114 (3.8%) neighborhood based rotating Credit savings group) I 8. Kelompok Doa (Catholic prayer Religious C F 95 (3.2%) group) 9. Posyandu (government sponsored Social G F 93 (3.1%) group to promote health of Services children and pregnant/ feeding mothers) _ __ l 10. Lembaga Adat (traditional group Social C F 87 (2.9%) to regulate customs, settle Services disputes and impose sanctions) The list of the ten most important associations does differ of course by province (Annex tables Al to A3 show this list for each province). Foremost, the different religious composition of each province causes the Pengajian (Koran recital group) to be ranked first or second in Jambi and Jawa Tengah, while Rayon (Protestant church organization) and Kelompok Doa (Catholic prayer group) are listed second and third in NTT. Among government organizations, the RT/RW governance Local Institutions and Service Delivery in Indonesia 42 groups rank as the most important institution in both Jawa Tengah and NTT, while they rank only ninth in Jambi, reflecting the overall lower penetration of government- supported local institutions in Jambi. As the other side of this coin, the traditional regulatory and adjudication institution of Lembaga Adat ranks as the second most important institution in Jambi. Finance and credit groups are much more frequently listed among the important associations in Jawa Tengah, while mutual-help occupational groups are considered more important in Jambi and NTT. One direct measure of actual importance to the household is the number of times a member attends a group meeting. On average, a household attends group meetings 10 times a month (Table 15). The figure is very much influenced by the extent of attendance to religious groups: in Jambi, religious group meetings are attended 9 times per month, in Jawa Tengah slightly more than 6 times per month, and in Christian NTT about 5 times per month. Attendance to meetings of other types of associations is much lower. It is lowest for meetings of production groups, natural resources groups, and governmental affairs groups. While on average there is not much regional variation, within each category there is quite a bit of regional variation. For example, attendance to meetings of social service groups is much higher in Jawa Tengah than in other provinces, while meetings of production and occupational groups are much better attended in NTT than in the other provinces. Local Institutions and Service Delivery in Indonesia 43 An interesting difference occurs for attending government affair groups: households attend about 1.5 meetings per month in Jawa Tengah and NTT, but in Jambi government affair groups are virtually not attended at all. This reflects again the strength of traditional structures in Jambi. We observed earlier that the presence of finance and credit groups is most pronounced in Jawa Tengah and attendance of these group meetings is also much stronger in that province than elsewhere. Table 15: Average Number of Attendances to Groups Meetings (Per 3-Month Period) LLI Categories Jambi Jawa Nusa All Tengah Tenggara Timur Social Service Group 3.3 8.2 5.1 5.7 Production Group 1.9 2.3 5.1 3.7 Occupational Group 7.4 5.6 10.3 8.7 Finance/Credit Services Group 4.2 11.1 4.2 7.7 Religious Group 27.2 19.6 14.2 19.4 Natural Resources Group 6.2 2.2 5.1 3.8 Environmental Services Group 10.0 9.3 - 9.4 Government Affairs Group 0.4 4.4 4.7 3.6 Recreation Group 3.2 5.2 7.5 4.9 Others 2.0 13.3 0.0 10.7 All 25.2 33.2 30.3 29.7 The final aspect of group membership to be noted is the perception of whether the group is initiated by the state or the community and whether the group is formally or informally organized. Table 16 is based on household responses to questions about the origin and the type of organization of the group and as such it tabulates household perceptions about group origin and structure (which may not always coincide with the factual situation). Local Institutions and Service Delivery in Indonesia 44 As Table 16 indicates, almost 100% of households perceive religious groups to be community initiated, while quite clearly the majority of them belong to churches which are national or provincial in scope. Muslim groups are more frequently perceived as informally organized while Catholic or Protestant groups are seen as more formally organized. Except for religious groups, there is a fairly close parallel between the perception of state initiation and that of formal organization on the one hand, and community initiation and informal organization on the other hand. Social service groups, natural resource groups, and environmental groups are most frequently initiated by the state while production and occupational groups have the strongest community initiation and also the most informal activity-based organization. Table 16: Type of Initiation and Organization of Groups State Community Formally Informally Initiated Initiated Organized Organized Social Services Group 62.8% 37.2% 73.3% 26.7% Production Group 42.6% 57.4% 36.2% 63.8% Occupational Group 1.4% 98.6% 1.7% 98.3% Finance/Credit Services Group 32.6% 67.4% 35.1% 64.9% Religious Group 0.2% 99.8% 47.8% 52.2% Natural Resources Group 63.6% 36.4% 76.3% 23.7% Environmental Services Group 60.0% 40.0% 0.0% 100.0% Government Affairs Group 100.0% 0.0% 99.9% 0.1% Recreation Group 50.3% 49.7% 66.5% 33.5% Others 78.9% 21.1% 78.9% 21.1% Local Institutions and Service Delivery in Indonesia 45 4. 'Characteristics of the Most Important Associations As we mentioned in the previous section, supplementary information was collected about the three most important groups to which each household belongs. Table 17 tabulates some of the characteristics of these groups. There are some distinct regional patterns in the reasons for joining groups and for considering them important . In Jambi, the most important reason is protection in the case of emergency or to derive future benefits. In Jawa Tengah in contrast, the most important reason for joining is the impact on the community and on the livelihood of the household. In NTT the predominant reason is the impact on the livelihood of the household: 27% of memberships are motivated by this (against only 11% in Jambi). It is also noteworthy that mandatory membership applies to about one out of every five cases in Jawa Tengah and in NTT, but only about one in ten cases in Jambi. By and large no membership fee is required to join and to remain a member of local associations in Indonesia. Only about 30% of memberships require the regular payment of a fee. This too shows a distinct regional pattern. Membership fee requirements are much higher in Jawa Tengah where almost 50% of memberships require a monthly fee. For the most part this is less than 1,000 Rp per month. In Jambi, membership fees are required in fewer cases, only about 1 in 4, but they tend to be higher and concentrated in the 1,000-10,000 Rp per month range. In NTT, less than 20% of associations require a membership fee, but in NTT it is much more Local Institutions and Service Delivery in Indonesia 46 customary to provide regular labor contributions, which occurs in 73% of the memberships. The amount of labor required in NTT as part of membership contributions is not insignificant: in 44% of the cases the contributed amount of labor exceeds 25 days per year. Table 17: Characteristics of the Memberships in the Most Important Groups Jambi Jawa Nusa All Tengah Tenggara Timur Reasonsfor Joining (%) Impact on Household Livelihood 11.3 21.0 26.9 20.3 Impact on Community 14.5 27.4 15.0 18.8 Everybody Belongs/Mandatory 11.0 18.5 19.6 16.6 Emergency/Future Benefit 35.5 13.6 17.1 21.6 Other 27.7 19.5 21.4 22.7 Total 100.0 100.0 100.0 100.0 Membership Fees (%) No Membership Fee 74.3 51.6 82.6 70.2 1-500 Rp/Month 2.3 20.3 6.0 9.4 501-1,000 Rp/Month 4.8 13.4 2.7 6.7 1,001-10,000 Rp/Month 16.3 13.1 7.1 11.8 10,001 and up Rp/Month 2.3 1.7 1.6 1.8 Total 100.0 100.0 100.0 100.0 % Also Providing Labor Contribution 2.1 10.9 73.3 31.7 Extent of Active Participation in Decision Making (°lo) Very Active 42.4 33.9 61.4 46.8 Somewhat Active 46.4 46.6 23.6 37.9 Not Active 11.2 19.5 15.0 15.3 Total 100.0 100.0 100.0 100.0 Households are most actively involved in the decision making of groups in NTT. In 61.4% of membership cases the household considers itself to be very Local Institutions and Service Delivery in Indonesia 47 actively involved in decision making. In contrast, the equivalent percentage in Jawa Tengah is only 34% and in Jambi it is 42.4%. Local associations play a wide variety of roles in community life in Indonesia. This is already clear from the previous tables which showed the distribution of these groups over widely different categories. Table 18 focuses on the provision of services and suggests that improving access to services is not a prime role of local associational life in Indonesia. Local institutions only play a significant role in the provision of education and credit and savings services. In the case of education, this is primarily concentrated in Jambi and Jawa Tengah, where respectively 62% and 43% of the memberships considered most important by households was in groups playing an active role in the provision of education. As we have noted earlier, credit and savings groups are most active in Jawa Tengah where slightly more than half of the memberships in the most important groups relate to access to credit and savings. Local groups play a secondary role in the provision of health services and water supply and sanitation and this function is primarily concentrated in Jawa Tengah. They play virtually no role in the provision of electricity. Local associations also play only a small role in the provision of services that directly relate to agriculture. This is surprising in view of the rural nature of the sample and the fact that the majority of respondents are farmers. Nevertheless, only 7.2% of memberships is in local associations playing a role in the provision of agricultural Local Institutions and Service Delivery in Indonesia 48 inputs or technology (primarily in Jawa Tengah), 4.9% in issues relating to land, forestry, and water rights (primarily in Jambi), and 2.4% in irrigation issues (primarily in NTT). Table 18: Involvement in Service Provision of the Most Important Groups % of Memberships in Jambi Jawa Nusa All Groups Active In Tengah Tenggara Timur Education 62.2 43.1 19.6 40.2 Health 6.9 12.2 5.1 7.9 Water Supply and Sanitation 0.0 6.5 4.5 3.8 Electricity 0.0 0.8 0.1 0.3 Credit 13.6 56.2 15.7 27.9 Savings 21.3 53.4 9.6 27.1 Agricultural Inputs/Technology 4.8 9.2 7.4 7.2 Irrigation 2.0 2.0 3.2 2.4 Land, Forestry, Water Rights 7.6 2.7 4.6 4.9 The conclusion one can draw from this is that Indonesian households do not primarily join local associations in order to improve their access to particular services. The only exception to this would be education and, in the case of Jawa Tengah, credit and savings. Instead, as was clear from previous tabulations, the prime function of local groups seems to be in the area of practicing religion and the dissemination at the local level of certain government functions. Nevertheless, a word of caution is in order here. Many associations may well have multiple functions but the LLI Study's questionnaires recorded only the primary objective of each association. For example, many religious groups are active in the provision of credit or education (Werner, 1998). The possibility exists, therefore, that respondents' answers about associations' Local Institutions and Service Delivery in Indonesia 49 primary role in the provision of services may underestimate the role of associations in this area. This possibility is illustrated if one tries to match the functional classification of associations with that based on the provision of services. In the case of education, for example, more than half of the organizations involved in education delivery are classified as religious organizations. In fact, less than 20% of associations delivering education services fall specifically under the social service headings. Even some organizations whose main function is in the area of natural resources or recreation are reported as having a function in the provision of schooling or training. The same is true for finance and credit groups. Less than half of the groups actually involved in the provision of credit or savings are classified primarily as a finance/credit services association. Social services groups, religious groups, government groups all play a role in the provision of credit. These observations underline the difficulty of classifying a specific association against a functional typology, and are testimony to the diversity and broad range of functions which most associations in Indonesia display. This suggests also that the role of social capital in improving access to services and/or improving welfare outcomes of households may stem from the overall relationships with others in the Local Institutions and Service Delivery in Indonesia 50 community that are being established in local associations, and which may well be more important than the specific function of the associations in question.'2 Local associations in Indonesia show a high degree of internal homogeneity along many important criteria (Table 19). The most striking of these is religion, which reflects of course the religious homogeneity of the three provinces in the study. Internal homogeneity by gender is very high in Jambi and Jawa Tengah (65%) and low in NTT (25.7%) which might also be a reflection of the different religious make- up of the provinces. Homogeneity by occupation and economic status is very high in Jambi (exceeding 80%) but quite significantly lower in the other two provinces. A remarkable distinction also occurs along the dimension of education, where 60% of memberships in important associations in Jambi display homogeneity of educational achievement of the members against only 14.6% in Jawa Tengah and 6.3% in NTT. This is a surprising result because as we saw earlier, the educational composition of the population in the three provinces does not differ very much. 12 This is in line of course with the findings of Putnam's (1993) seminal study on the role of civil society and social capital in Italy. Local Institutions and Service Delivery in Indonesia 51 Table 19: Internal Homogeneity of the Most Important Local Associations Jambi Jawa Nusa All Tengah Tenggara Timur l % of Memberships in Associations whose Members are "Mostly from the Same" Neighborhood 51.5 79.3 58.7 63.0 Kin Group 12.5 6.3 32.6 18.1 Occupation 85.6 46.6 61.4 64.2 Economic Status 84.0 32.3 37.7 50.2 Religion 98.6 83.7 77.9 86.1 Gender 65.9 65.4 25.7 50.6 Age 26.5 6.0 2.0 10.8 Education 60.1 14.6 6.3 25.5 From the perspective of social capital, it is not a priori obvious whether internal homogeneity is an advantage or a disadvantage. On the one hand, members that are similar in important social or economic characteristics may well find it easier to interact and establish trust, thus making the association more effective. On the other hand, they may also have similar information so that less is gained from exchanging information. Furthermore, the coexistence of a series of associations which are each internally homogeneous but along different dimensions could render the decision making process at the village level more difficult. The evidence suggests that both effects are present. When internal homogeneity is included together with other aspects of associations in a multivariate model which assesses the impact of local associations on access to services by household and on household welfare, the results suggest that on balance heterogeneity is a positive factor. The exception is collective action which is facilitated by internal homogeneity (Grootaert, 1998). Local Institutions and Service Delivery in Indonesia 52 5. Social Capital and Poverty The analysis of the links between social capital and poverty is one of the main objectives of the Local Level Institutions Study (World Bank, 1998). There are two aspects to this issue. First is the question whether the poor "own" more social capital than the rich, and whether this compensates for their lack of other assets. The second question is the extent to which social capital helps to escape from poverty and improves access to services (especially credit) which the poor would otherwise not be able to procure. These questions can only be addressed fully in the context of a multivariate analysis of the role of social capital and this is done elsewhere (Grootaert, 1998). In this paper, we would however like to make a start by tabulating the membership in local organizations according to the income level of the household (proxied here by the level of household expenditure per capita). Households can be classified in three different ways: first, according to the expenditure per capita quintile in which each individual household falls; second, according to the average level of income of the village; and third, by combining the province and whether the households' village is in the poorest or richest group of villages. For each of these three classifications, Tables 20-22 show the percentage of households who are Local Institutions and Service Delivery in Indonesia 53 members of an association in each particular category.'3 (Table A4 and A5 show provincial breakdowns of tables 20 and 21). Table 20: Percentage of Households who are Members of Local Associations, by Category of Association and Household Quintile Quintile of PCHHEXP, Household Level LLI Categories 1 2 3 4 5 All (Poorest) (Richest) Social Service Group 71.7 66.7 70.0 68.3 72.1 69.8 Production Group 20.0 20.0 18.8 14.6 16.3 17.9 Occupational Group 29.2 27.5 25.8 17.5 22.9 24.6 Finance/Credit Services Group 54.2 59.2 57.1 55.0 57.1 56.5 Religious Group 71.3 73.3 67.5 70.8 58.8 68.3 Natural Resources Group 10.0 13.8 12.5 12.5 15.8 12.9 Environmental Services Group 0.0 2.1 0.4 1.7 0.0 0.8 Governmental Affairs Group 70.0 62.5 61.7 68.8 68.3 66.3 Recreation Group 9.6 15.0 15.4 11.7 20.8 14.5 Others 0.4 0.8 0.8 1 .3 2.9 1 .3 13 Tables 20-22 are not membership tables as we used earlier. In a membership table, each membership in an organization is counted as one unit. In Tables 20-22 the unit of analysis is the household. For example, if a household is a member of two social service groups, it is counted only once for the calculation of the percentage of households who are members in the social service group. Local Institutions and Service Delivery in Indonesia 54 Table 21: Percentage of Households who are Members of Local Associations, by Category of Association and Village Quintile Quintile of PCHHEXP, Village Level LLI Categories 1 2 3 4 5 All (Poorest) (Richest) Social Service Group 84.9 56.8 68.9 63.2 76.4 69.8 Production Group 20.4 16.4 35.1 12.0 7.6 17.9 Occupational Group 43.1 26.8 18.7 17.6 18.0 24.6 Finance/Credit Services Group 63.6 54.8 61.8 42.0 61.6 56.5 Religious Group 86.7 66.0 71.1 57.2 62.8 68.3 Natural Resources Group 11.6 13.6 11.1 6.4 21.6 12.9 Environmental Services Group 0.0 0.8 0.4 0.4 2.4 0.8 Governmental Affairs Group 74.2 63.6 63.1 62.8 68.0 66.3 Recreation Group 12.0 9.6 15.1 18.4 17.2 14.5 Others 0.4 0.4 3.1 2.0 0.4 1.3 Local Institutions and Service Delivery in Indonesia 55 Table 22: Percentage of Households who are Members of Local Associations, by Category of Association and Region/Village Group Jambi Jawa Tengah NTT LLI Categories Richest Poorest Richest Poorest Richest Poorest All Villages Villages Villages Villages Villages Villages Social Service Group 65.5 49.5 72.0 73.5 77.5 80.5 69.8 Production Group 7.5 13.5 8.5 20.0 37.0 21.0 17.9 Occupational Group 28.0 22.0 6.0 13.5 36.0 42.0 24.6 Finance/Credit Services 22.5 42.0 83.0 86.5 54.0 51.0 56.5 Group Religious Group 52.0 63.5 67.0 49.0 86.0 92.5 68.3 Natural Resources 4.0 5.0 19.5 17.5 21.5 10.0 12.9 Group Environmental Services 0.0 0.5 3.5 1.0 0.0 0.0 0.8 Group Governmental Affairs 55.0 32.5 67.5 67.5 86.5 88.5 66.3 Group Recreation Group 13.0 16.5 26.0 16.5 10.5 4.5 14.5 Others 0.5 0.0 2.5 3.5 0.5 0.5 1.3 Local Institutions and Service Delivery in Indonesia 56 The overriding observation which emerges from Tables 20-22 is how uniform the membership in local organizations is in Indonesia. There is generally speaking no conclusion to be drawn that membership is distinctly higher among poor or among rich household groups. This is true whether households are classified according to their own income level, according to the village income level or by taking the regional dimension into account. This being said, there are some patterns by specific category of organization that need to be mentioned. Social Service Groups. Membership in social services groups bears no relation to the household's own income level. It is however more concentrated in the poorest villages. There is a regional dimension to this since the membership in social service groups is highest in NTT and lowest in Jambi. Within Jambi, it is more concentrated among the richer villages. Production Groups. Membership in production groups is primarily concentrated among the lowest two quintiles. It is strongly concentrated in villages in the middle quintile and falls of very sharply in richer villages. This also has a regional dimension. In Jambi, membership in production groups is quite low, particularly among high income villages. The same is true in Jawa Tengah, but in NTT the pattern is reversed and production group membership is strongest in the richest villages (mainly in the villages in the third and fourth quintiles). Local Institutions and Service Delivery in Indonesia 57 Occupational Groups. Membership in occupational groups is quite strongly related to income level and much more pronounced among the poorest income groups. This also holds true in the village classification, where 43% of households in the poorest quintile of villages are members of such groups against only 18% in the two richest quintiles. The regional pattern confirms this, except in Jambi, where membership of occupational groups is more pronounced in higher income villages. Finance/Credit Services Groups. Membership in finance and credit groups is not related to the household's income level. It is also unrelated to the income status of the village, but as we pointed out in earlier pages, there is a strong regional focus with a very high membership in Jawa Tengah. Only in Jambi is there some relation between membership of finance and credit groups and income level in the direction that the poorest villages have a higher degree of membership. Religious Groups. Membership in religious groups is for the most part uncorrelated with the household's income level except at the upper range of the distribution where membership declines. Membership in religious groups is, however, significantly more pronounced in the poorest villages(except in Jawa Tengah). Local Institutions and Service Delivery in Indonesia 58 Natural Resources and Environmental Services Groups. Membership in these groups is fairly limited and no specific pattern can be discerned according to income level. Governmental Affairs Groups. Membership in governmental affair groups is very slightly related to income level, especially at the lower end of the distribution where membership is slightly higher. This is true both at the household and the village level. A more pronounced pattern is present in Jambi where membership is significantly higher in the richer villages. In the other two provinces, no correlation with income exists. Recreation Groups. This category displays the most pronounced correlation with income whereby membership steadily rises with income level. This is not surprising since recreational activities are generally considered a luxury good more accessible to the rich. Local Institutions and Service Delivery in Indonesia 59 6. Who Provides Services? In the previous sections we have described the role of local associations according to various functions and according to a number of services in which they are active. In this section, we focus four specific services-education, health, access to credit, and assistance with titling of land or houses-to see whether governments, NGOs and/or local associations are active in the provision of these services. The majority of children in Indonesia go to schools which are owned and maintained by either the district government or the central government (Table 23). Next in line are religious organizations, which own and manage a significant number of schools in NTT. Only in rare cases (fewer than 5%), does the community own and operate the school. NGOs likewise play a very small role in the provision of education services. 14Parents are not particularly active participants in school affairs, at least not as measured by attendance to parent-teacher association meetings. The average parent in the sample attends only 1.7 PTA meetings per year. The figure is slightly higher in NTT. Attendance to PTA meetings is also not related to the level of income of the household. 14 Werner (1997) discusses in detail the role of the community in the provision of education services in Jambi. Local Institutions and Service Delivery in Indonesia 60 Table 23: Who Owns and Manages Schools? Ownership Management Jambi Jawa Nusa All Jambi Jawa Nusa All Tengah Tenggara Tengah Tenggara Timur Timur Community 6.0 0.6 9.9 6.0 5.7 0.6 1.7 2.7 District 42.0 72.7 45.6 52.0 69.4 73.3 66.7 69.4 Government Central 46.5 12.0 9.3 22.2 19.7 11.7 9.1 13.3 Government Religious 3.1 9.2 26.6 14.1 3.1 8.9 16.0 9.8 Organizations NGOs 0.0 2.5 6.5 3.2 0.0 2.4 5.0 2.6 Other 2.3 3.1 2.2 2.5 2.1 3.1 1.5 2.1 All 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Note: Figures show the distribution of children aged 5-14 in school, according to entity who owns/manages the school. The provision of health care is likewise largely a government affair. Central or district governments provide anywhere from approximately 50% of health care in Jambi and Jawa Tengah to well over two-thirds in NTT. The community and religious organizations play a small part only in NTT, and the role of NGOs is essentially nonexistent. However, private businesses and "other" sources of health care account for over 40% of consultations. This includes traditional medicine which is still actively practiced in many rural areas (Werner, 1997).' 15 Consultations with traditional providers are likely to be classified under "private business" or "other" in Table 24, because the LLI Study's questionnaire did not have an explicit response code for traditional health care. Local Institutions and Service Delivery in Indonesia 61 Table 24: Who Organizes Health Care? Jambi Jawa Nusa All Tengah Tenggara Timur Community 0.2 0.5 3.8 1.5 Central Government 11.6 2.2 4.7 5.5 District Government 36.7 45.0 64.9 49.1 Religious Organizations 0.1 0.1 4.2 1.4 NGO 0.1 0.3 0.7 0.4 Employer/Private Business 51.0 11.7 5.1 20.2 Other 0.3 40.2 16.7 21.9 All 100.0 100.0 100.0 100.0 Note: Figures show the distribution of health consultations, according to ownership of the health care facility. Indonesian households have good access to credit. Overall, more than 50% of households report having obtained credit in the past year (Table 25). The lowest figure is in NTT at 41.5%, and the highest figure is in Jawa Tengah at 68.8%. The main source of credit in Jawa Tengah is a community credit group, in line of course with the high presence of such groups in that province (see Table 8). In Jambi, the main single source of credit are traders or stores, while in NTT it is relatives. However, in Jambi and NTT, people's personal networks (i.e., household members, relatives and friends) account for most of the credit, respectively 46.4% and 56.2%. In all three provinces, the role of banks and NGOs is unimportant. Local Institutions and Service Delivery in Indonesia 62 Table 25: Who Provides Credit? Jambi Jawa Nusa All Tengah Tenggara Timur % of Households who obtained credit 54.0% 68.8% 41.5% 54.8% in the past year Distribution of sources Household member 9.3 3.0 4.9 5.2 Friends 9.3 15.6 13.2 13.3 Relatives 27.8 13.0 38.1 22.6 Money lender 0.4 1.7 4.9 2.1 Trader/store 41.3 17.5 2.4 20.8 Community credit group 5.0 32.8 22.9 22.9 Government bank 5.4 6.5 6.3 6.2 Commercial bank 0.4 0.4 0.5 0.4 NGO 0.0 0.0 1.0 0.2 Other 1.2 9.5 5.9 6.4 All 100.0 100.0 100.0 100.0 Lastly, we look at the organizations who are helpful in arranging land titling. This is a potentially important issue since 90% of households in the sample own land. Nevertheless, only 38.4% of those who own land have title documents (Table 26). This figure is quite low in Jambi and in NTT (respectively 17.7% and 24.2%), but much higher in Jawa Tengah at 72.2%. For the most part assistance with title registration is received from village officials. For the remaining part, other government officials assist in the matter. Table 26 indicates that neither village organizations nor NGOs nor religious organizations play any active role in the provision of this service. Presumably one can conclude that the government organization of title registration is adequate at least for those people who obtain documents. The low ownership of title documents, however, suggests that action may be needed in the future to ensure that more land owners register their claim. Local Institutions and Service Delivery in Indonesia 63 Table 26: Who Arranges Land Titling? Jambi Jawa Nusa All Tengah Tenggara Timur % of Households who own land 84.8% 91.8% 94.0% 90.2% % of Owners who have title 17.7% 72.2% 24.2% 38.4% documents Registration assistance received from - village officials 62.5% 69.3% 52.4% 63.9% - village organizations 0.0% 0.5% 1.2% 0.7% -government officials 25.0% 7.3% 42.7% 18.9% -NGO 0.0% 0.0% 0.0% 0.0% - Religious organizations 0.0% 0.6% 0.0% 0.4% - Other 12.5% 22.3% 3.7% 16.1% Total 100.0% 100.0% 100.0% 100.0% Local Institutions and Service Delivery in Indonesia 64 7. Collective Action and Trust The final section in this paper will address the issues of collective action and trust in the community. Collective action is an important aspect of village life in Indonesia. It is known as "Gotong Royong" and consists of community-organized activities to build and maintain infrastructure (roads and bridges, community buildings, water supply systems) and to provide related public services (Werner, 1998). Collective action is organized both at the village level and at the neighborhood level. In the LLI Study's questionnaire, the extent of collective action was recorded and people were asked to give their perception about the fairness of the arrangements and possible criticism or fines that would be incurred in the case of non-participation. As a direct inquiry into the amount of trust in the immediate community, households were asked whether in case of shortness of money or food or in case of illness, they would be able to obtain assistance from people beyond their immediate household and relatives, and, if so, which organizations would be most likely to provide assistance. The tables in this section present the results from these questions. Eighty percent of households participated in collective action at the neighborhood level at least once in the past year. Two-thirds of them also participated in such action at the village level. This is quite an impressive Local Institutions and Service Delivery in Indonesia 65 demonstration of people's collective spirit and their ability to put it into action. 16 The distribution of the frequency with which people participate in collective action (Table 27) indicates that the vast majority of households participate between once every month or once every two months, but there are important regional differences which is the traditional unit of organization in that province (Werner, 1998). This explains why collective action is least well established in Jambi where only 20% of the households participate more than once every two months. In contrast, in Jawa Tengah, almost 40% of households participate more than twice a month at the neighborhood level, which is the traditional unit of organization in that province. This explains why collective action at the village level is quite rare in Jawa Tengah. In NTT 25% of households participate more than twice a month in collective action at the neighborhood level and about 15% participate with that frequency at the village level. 16 As Werner (1998) points out, high incidence of collective action can be an indicator of both strong or weak local organizing capacity depending upon whether the collective action complements or substitutes for the activities of local groups. Werner argues that in NTT, which has a densely-structured landscape of local institutions, much collective action is carried out by existing groups with related objectives (e.g., parent organizations may do school repair). In contrast, in Jawa Tengah, neighborhood-based collective action is most often short-term and replaces long-term institutional activities. Local Institutions and Service Delivery in Indonesia 66 Table 27: Participation in Collective Action (Gotong Royong) l__________ Neighborhood Frequency of Jambi Jawa Nusa All Participation (Per Year) Tengah Tenggara l________ Timur 0 30.5 9.8 17.5 19.3 1-6times 51.3 19.8 31.3 34.1 7-12 times 9.0 15.8 13.8 12.8 13-24 times 3.5 15.5 12.5 10.5 25-48 times 4.8 22.3 15.8 14.3 49+ times 1.0 17.0 9.3 9.1 Total 100.0 100.0 100.0 100.0 l ________________________ ____________ V illag e Frequency of Jambi Jawa Nusa All Participation (Per Year) Tengah Tenggara l________ _________ T im ur 0 27.8 55.5 17.3 33.5 1-6 times 51.8 29.5 40.3 40.5 7-12 times 12.3 11.0 19.3 14.2 13-24 times 2.0 1.8 7.8 3.8 25-48 times 5.5 1.3 10.8 5.8 49+ times 0.8 1.0 4.8 2.2 Total 100.0 100.0 100.0 100.0 Participation in collective action is not systematically related to income level (Table 28). The richest households participate in collective action at the neighborhood level slightly above average. Collective action at the neighborhood level appears to be more prevalent in middle-income villages, but this is offset by a lower level of collective action at the village level. Within each province, there is little difference in participation frequency between the richest and poorest groups of villagers, except in Jawa Tengah where collective action at the neighborhood level is more frequent in poorer villagers. Local Institutions and Service Delivery in Indonesia 67 Table 28: Average Frequency (times per year) of Participation in Collective Action, by Quintile Neighborhood Village Household Quintiles Poorest 16.6 7.6 2 17.9 6.5 3 17.8 6.6 4 12.6 7.2 5 20.2 7.4 Richest 17.0 7.1 Village Quintiles Poorest 16.9 8.4 2 14.7 9.0 3 22.5 5.6 4 14.5 7.7 5 17.1 4.5 Richest 17.0 7.0 Province Jambi ........ Poorest 6.2 6.8 ........... Richest 6.0 5.6 Jawa Tengah.. Poorest 32.3 3.0 ........... Richest 22.6 3.8 NTT ........ Poorest 18.1 10.1 ........... Richest 17.1 13.1 Social pressure can be a powerful motivator for people to participate in collective action or otherwise behave according to social norms. Table 29 suggests that this factor plays a role in the Indonesian context. Almost two-thirds of households are fearful of being criticized or being levied a fine for not participating in collective action at a neighborhood level. The figure is slightly lower (57.4%) at the village level. The highest social pressure occurs in Jawa Tengah and NTT (at the neighborhood level), which are also the two provinces where participation in collective action at a neighborhood level is the strongest. Respondents in Jawa Local Institutions and Service Delivery in Indonesia 68 Tengah indicate in large numbers (86.7%) that they feel "encouraged" to participate because of the possibility of a criticism or fine. However, the other side of the coin is that the system of social pressure is seen to yield a fair result, since 97% of respondents in Jawa Tengah indicate that they think others are making a fair contribution to collective action. For the three provinces together, 87% of respondents indicate that they perceive contributions to be fair at the neighborhood level and 76% indicate that contributions are fair at the village level. Table 29: Motivation for Collective Action Neighborhood Jambi Jawa Nusa All Tengah Tenggara Timur l % of Households fearing criticism or fine 46.9 67.2 84.1 65.9 % of Households encouraged to participate 34.5 86.7 63.3 61.5 because of criticism or fine % of Households perceiving others to make 75.5 97.3 88.5 87.1 fair contribution I I Village Jambi Jawa Nusa All Tengah Tenggara Timur % of Households fearing criticism or fine 50.3 37.3 84.6 57.4 % of Households encouraged to participate - - - - because of criticism or fine I % of Households perceiving others to make 81.5 63.5 83.5 76.2 fair contribution There are also distributional aspects to the fear of fines or criticism: 71% of households in the poorest quintile express such fear whilst only 62% of the richest two quintiles express this fear. Poor people also have a slightly lower perception that others make a fair contribution than do the rich. Local Institutions and Service Delivery in Indonesia 69 An alternative way to assess social cohesion and people's faith and trust in one another is to ask hypothetical questions about the perceived ability of people to obtain help from their fellow villagers. An impressive 63% of households think that they will "definitely" be able to receive assistance from people beyond their immediate household and relatives, should the need arise (Table 30). Another 19% think that they will "probably" receive such assistance. Only 14% of households are of the opinion that they would definitely not be able to obtain such assistance. Among the sources most likely to provide such assistance, community organizations rank at the top with 36% of respondents listing them first.'7 The poor are more likely than others to turn to community organizations for help, as are residents of NTT. 17 There are however a multitude of organizations to which people would turn beyond those listed in the table since 50% of the answers fall into the category "other." Local Institutions and Service Delivery in Indonesia 70 Table 30: Community Assistance Jambi Jawa Nusa All Tengah Tenggara Timur % of Households who think they can obtain assistance beyond immediate household and relatives - Definitely 61.3 69.3 60.0 63.5 - Probably 30.3 11.5 15.5 19.1 - Probably Not 0.3 4.0 2.5 2.3 - Definitely Not 7.5 14.5 19.8 13.9 - Difficult to Answer 0.8 0.8 2.3 1.3 Total 100.0 100.0 100.0 100.0 Organization most likely to help - Community Organizations 9.3 40.0 58.5 35.9 - District Government 1.3 0.5 1.0 0.9 -Religious Organizations 9.5 3.3 9.0 7.3 -NGO 0.0 0.0 0.3 0.1 - Businessman/Trader 15.5 3.0 0.5 6.3 - Other 64.5 53.0 30.8 49.5 Total 100.0 100.0 100.0 100.0 Local Institutions and Service Delivery in Indonesia 71 8. Summary and Conclusions This paper provides a map of local institutions in rural Indonesia and investigates the extent to which they are involved in the delivery of services. The study covers three provinces: Jambi on the island of Sumatra, central Java, and Nusa Tenggara Timur (NTT) in the eastern part of Indonesia. These three provinces represent very different social, cultural, and economic settings. Jawa Tengah is the richest province, with the highest level of per capita household expenditure and ownership of consumer durables. Households have the highest education level and a larger percentage work as wage earners than elsewhere. Jambi has a slightly lower average expenditure level, but because it has a fairly equal distribution of expenditure poverty rates are lower than in Jawa Tengah. The population of Jambi and Jawa Tengah are almost entirely Muslim. NTT is the poorest of the three provinces. Its employment pattern is the most heavily skewed towards agriculture and home production of foods. Asset ownership is low but ownership of animals is high. NTT's population is for the most part Christian, evenly divided between Protestants and Catholics. Across the three provinces, 210 different local associations were recorded. The average household in Jawa Tengah and NTT is an active member of six associations, but in Jambi households belong on average to only three associations. Social service groups account for the largest number of memberships (27% of the Local Institutions and Service Delivery in Indonesia 72 total) and are equally important in each of the three provinces. Finance and credit associations, religious groups and government affairs groups each account for just under 20% of memberships. However, more than 60% of all memberships in finance and credit groups are in Jawa Tengah, where they represent almost one in every three memberships. Religious groups are most strongly concentrated in Jambi, while production and occupational groups are most prevalent in NTT. Households in Jambi also have a low participation rate in government affairs groups, reflecting the strengths of traditional village structures in that province. About one-third of all memberships are in local associations which are segregated by gender (mostly female-only). The majority of these occur in Jawa Tengah. In contrast, age is not a particularly important membership criterion of associational life in Indonesia (6% of memberships are in youth-oriented groups). The way local associations are organized and fit in supra-village structures shows a strong regional variation. In Jambi, 75% of all memberships are in genuinely local groups, which are not part of networks outside the village. This reflects in part the strength of the traditional village structure in Jambi, which has meant that government-sponsored or nationwide associations have had lower penetration than in other provinces. In NTT on the other hand, the majority of associations have links outside the village, in part because of the strength of Catholic and Protestant organizations, which by nature surpass the area of the village. Local Institutions and Service Delivery in Indonesia 73 Respondents in the study were asked to identify the three memberships which they considered to be the most important. Among the ten most important associations, all but two were either religious groups or government-sponsored groups, i.e. associations with a supra-village structure. The government-initiated local RT/RW structure was at the top of the list. The most important genuinely local groups were Arisan Dusun, locally based rotating savings groups popular in Jawa Tengah, and Lembaga Adat, the traditional structure for regulating customs, setting disputes and imposing sanctions, which is very strong in Jambi. Households join associations and consider them important because of perceived benefits to their household or community and because of the protection (insurance) membership provides in the case of an emergency. In NTT, the prime reason is the immediately impact on the household's livelihood, which is consistent with the strength of production and occupational associations in NTT. In Jawa Tengah, the prime reason given is the beneficial impact on the community, while the insurance consideration ranks more highly in Jambi. Local associations play a wide variety of roles in community life in Indonesia. The pattern of memberships described earlier highlights this role in religious life and the management of local government affairs. The study results, however, indicate that improving access to services is not a prime role of local associational life. The exceptions are education in Jambi and Jawa Tengah and access to credit in Jawa Local Institutions and Service Delivery in Indonesia 74 Tengah. In those cases, a majority of associations considered by households as most important are active in the provision of those services. When it comes to other services-health, water supply and sanitation, electricity, irrigation, agricultural technology or land/forestry/water rights-fewer than 8% of important memberships are in associations dealing with these issues. However, this finding needs to be treated with some caution since the data recorded only the prime function of each association. Some associations may well have secondary functions in the provision of services (e.g., some religious groups also provide credit or education). This paper also made a start in investigating the links between social capital and poverty, although it is recognized that a full analysis requires estimating a multivariate model of the effect of social capital on household welfare and poverty. The overriding initial finding from the data is the uniformity of memberships in local associations in Indonesia. Membership is not distinctly higher among the poor or the rich, although there are some patterns by specific category of associations. For example, membership in production and occupational associations is most concentrated among the poor, while membership in recreation groups rises with income. Lastly, this paper noted the high incidence of collective action in Indonesia: 80% of households participated at least once in the past year. This participation is not systematically related to income level. Social pressure is clearly part of the Local Institutions and Service Delivery in Indonesia 75 motivating factor since almost two-thirds of households expressed being fearful of being criticized by their neighbors or being levied a fine for not participating in collective action. However, the other side of the coin is that the system of social pressure is seen to yield a fair result since four out of five respondents indicate that contributions are fair. Levels of trust are at a high level in the villages in the sample: an impressive 63% of households think that they will "definitely" be able to receive assistance from people beyond their immediate household and relatives, should the need arise. Among the sources most likely to provide such assistance, community organizations rank at the top, especially among the poor and in NTT. In summary, this paper has documented a dense associational life in rural Indonesia. It shows strong geographic variation, reflecting social, cultural and economic differences. Jambi is characterized by a strong traditional village structure, Jawa Tengah has a long history of local associations in rotating credit groups, and NTT has a focus on agriculture-related production and occupational groups. The study results indicate that concerns over the immediate livelihood of the household or community, or protection against future emergencies motivate people to join local associations. Village life is characterized by high participation in collective action. While not all of this participation is voluntary, and social pressure is high, the outcome is seen as fair by most. We hope that this paper has set the stage for a Local Institutions and Service Delivery in Indonesia 76 further analysis and quantitative measurement of the contributions made by memberships in associations to household welfare and to the alleviation of poverty. Local Institutions and Service Delivery in Indonesia 77 Annex: Tables by Province Table Al: The Ten Most Important Local Associations in Jambi Group Name Category Government Formal or Number of or Informal Times Cited Community Group (% of Total Initiated Citations) 1. Pengajian (Koran recital Religious Community Informal 118 (12.9%) group) 2. Lembaga Adat Social Community Fornal 74 (8.1%) (traditional group to Services regulate customs, settle disputes and impose sanctions) 3. Yasinan (Koran recital Religious Community Informal 74 (8.1%) group in the event of death) 4. IDT (group receiving Finance/ Government Formal 61 (6.6%) assistance from Credit government in poor villages) 11 5. Gotong royong Sawah Occupational Community Informal 58 (6.3%) (mutual help group for rice cultivation) l 6. Pengajian Anak Religious Community Informal 51 (5.6%) (children's Koran recital group) l 7. Posyandu (government Social Govemment Formal 47 (5.1%) sponsored group to Services promote health of children and pregnant/ feeding mothers) _ 8. Kelompok Kematian Social Community Informal 40 (4.4%) (group for funeral Services purposes) l 9. RT/RW (neighborhood Government Government Formal 36 (3.9%) governance group) 10. PKK (national Social Government Formal 32 (3.5%) organization for Services Indonesian housewives) l Local Institutions and Service Delivery in Indonesia 78 Table A2: The Ten Most Important Local Associations in Jawa Tengah Group Name Category Government Formal or Number of or Informal Times Cited Community Group (% of Total l ______________________________ _______________ j Initiated _ C itations) 1 I . RT/RW (neighborhood Government Government Formal 150 (15.9%) governance group) 2. Pengajian (Koran recital Religious Community Informal 127 (13.5%) group) 3. Arisan Dusun (hamlet Finance/ Community Informal 112 (11.9%) or neighborhood based Credit rotating savings group) 4. PKK (national Social Government Formal 94 (10.0%) organization for Services Indonesian housewives) 5. Dusun (hamlet Government Government Formal 39 (4.1%) governance group) 6. IDT (group receiving Finance/ Government Formal 37 (3.9%) assistance from Credit government in poor villages) 7. Arisan umum (general Finance/ Community Informal 34 (3.6%) rotating saving group) Credit 8. Kelompok Tani Production Government/ Formal/ 33 (3.5%) (association of farmers) Community Informal 9. Kelompok KB (family Social Government Formal 29 (3.1%) planning group) Services 10. Siskamling (night watch Social Government Formal 27 (2.9%) group) Services Local Institutions and Service Delivery in Indonesia 79 Table A3: The Ten Most Important Local Associations in NTT Group Name Category Government Formal or Number of or Informal Times Cited Community Group (% of Total Initiated Citations) 1. RT/RW (neighborhood Government Government Formal 140 (12.4%) governance group) 2. Rayon (organization of Religious Community Formal 122 (10.8%) Protestant Church for church maintenance and ministerial support) 3. Kelompok Doa Religious Community Formal 95 (8.4%) (Catholic prayer group) 4. IDT (group receiving Finance/ Government Formal 77 (6.8%) assistance from Credit government in poor villages) 5. Kelompok Kebun Occupational Community Informal 75 (6.7%) (agricultural group for garden cultivation and housing construction) 6. Kelompok Tani Production Government/ Formal/ 69 (6.1%) (association of farmers) Community Informal 7. Dasawisma Social Government Formal 54 (4.8%) (neighborhood group of Services housewives) . l 8. Lingkungan (Catholic Religious Community Formal 51 (4.5%) church communal congregation) l 9. PKK (national Social Government Formal 40 (3.6%) organization for Services Indonesian housewives) l 10. Posyandu (government Social Government Formal 33 (2.9%) sponsored group to Services promote health of children and pregnant/ feeding mothers) Local Institutions and Service Delivery in Indonesia 80 Table A4: Percentage of Households Who Are Members of Local Associations, by Category of Association, Household Quintile and Province (a) Jambi Quintile of PCHHEXP, Household Level Local Level Institutions Categories 1 2 3 4 5 All Social Services Group 47.4 57.0 55.4 59.8 62.2 57.5 Production Group 13.2 15.1 8.7 9.8 7.3 10.5 Occupational Group 31.6 31.4 26.1 16.7 24.4 25.0 Finance/Credit Services Group 36.8 41.9 33.7 33.3 17.1 32.3 Religious Group 44.7 60.5 57.6 66.7 50.0 57.8 Natural Resources Group 2.6 5.8 4.4 4.9 3.7 4.5 Environmental Services Group - - - - - - Government Affairs Group 15.8 39.5 46.7 51.0 48.8 43.8 Recreation Group 13.2 14.0 17.4 11.8 17.1 14.8 Other _ _ _ _ _ _ (b) Jawa Tengah Quintile of PCHHEXP, Household Level Local Level Institutions Categories 1 2 3 4 5 All Social Services Group 70.6 62.7 78.2 75.0 75.7 72.8 Production Group 16.2 14.7 14.1 10.5 15.5 14.3 Occupational Group 8.8 10.7 11.5 5.3 11.7 9.8 Finance/Credit Services Group 76.5 82.7 88.5 88.2 86.4 84.8 Religious Group 48.5 70.7 57.7 65.8 49.5 58.0 Natural Resources Group 10.3 17.3 18.0 21.1 23.3 18.5 Environmental Services Group 0.0 6.7 0.0 5.3 0.0 2.3 Government Affairs Group 64.7 61.3 59.0 75.0 74.8 67.5 Recreation Group 11.8 26.7 21.8 13.2 29.1 21.3 Other 1.5 2.7 1.3 2.6 5.8 3.0 (c) Nusa Tenggara Timur Quintile of PCHIHEXP, Household Level Local Level Institutions Categories 1 2 3 4 5 All Social Services Group 79.1 81.0 80.0 74.2 80.0 79.0 Production Group 23.9 30.4 37.1 27.4 30.9 29.0 Occupational Group 38.8 39.2 41.6 33.9 41.8 39.0 Finance/Credit Services Group 47.8 55.7 52.9 50.0 61.8 52.5 Religious Group 90.3 89.9 91.4 83.9 89.1 89.3 Natural Resources Group 11.9 19.0 17.1 14.5 20.0 15.8 Environmental Services Group - - - - - - Government Affairs Group 88.1 88.6 84.3 90.3 85.5 87.5 Recreation Group 7.5 5.1 5.7 9.7 10.9 7.5 Other _ _I _ _I _ _I _ _I _I I Note: Household quintile boundaries were defined at the national level (see Table 20), which means that provincial quintiles do not necessarily contain 20% of households. Local Institutions and Service Delivery in Indonesia 81 Table A5: Percentage of Households Who Are Members of Local Associations, by Category of Association, Village Quintile and Province (a) Jambi Quintile of PCHHEXP, Village Level Local Level Institutions Categories 1 2 3 4 5 All Social Services Group 100.0 52.0 28.0 52.8 73.0 57.5 Production Group 0.0 12.0 30.0 9.6 3.0 10.5 Occupational Group 48.0 27.0 10.0 29.6 19.0 25.0 Finance/Credit Services Group 92.0 51.0 20.0 14.4 27.0 32.3 Religious Group 92.0 62.0 66.0 49.6 51.0 57.8 Natural Resources Group 0.0 10.0 0.0 0.8 7.0 4.5 Environmental Services Group - - - - - Government Affairs Group 0.0 37.0 8.0 61.6 57.0 43.8 Recreation Group 56.0 6.0 26.0 15.2 7.0 14.8 Other _ _ _ _ _ (b) Jawa Tengah Quintile of PCHHEXP, Village Level Local Level Institutions Categories 1 2 3 4 5 All Social Services Group 68.0 46.0 83.0 74.0 75.2 72.8 Production Group 36.0 8.0 23.0 7.0 11.2 14.3 Occupational Group 32.0 10.0 14.0 0.0 9.6 9.8 Finance/Credit Services Group 100.0 74.0 88.0 84.0 84.0 84.8 Religious Group 40.0 44.0 63.0 56.0 64.8 58.0 Natural Resources Group 24.0 0.0 16.0 15.0 29.6 18.5 Environmental Services Group 0.0 4.0 0.0 1.0 4.8 2.3 Government Affairs Group 56.0 68.0 70.0 63.0 71.2 67.5 Recreation Group 16.0 22.0 10.0 26.0 27.2 21.3 Other 0.0 2.0 6.0 4.0 0.8 3.0 (c) Nusa Tenggara Timur Quintile of PCHHEXP; Village Level Local Level Institutions Categories 1 2 3 4 5 All Social Services Group 85.1 67.0 77.3 72.0 96.0 79.0 Production Group 21.1 25.0 54.7 44.0 8.0 29.0 Occupational Group 44.0 35.0 30.7 28.0 56.0 39.0 Finance/Credit Services Group 54.3 49.0 54.7 12.0 88.0 52.5 Religious Group 92.6 81.0 85.3 100.0 100.0 89.3 Natural Resources Group 11.4 24.0 12.0 0.0 40.0 15.8 Environmental Services Group - - - - - - Government Affairs Group 87.4 88.0 90.7 68.0 96.0 87.5 Recreation Group 5.1 7.0 14.7 4.0 8.0 7.5 Other __ _I __ _I _ II_ _l Note: Village quintile boundaries were defined at the national level (see Table 21), which means that provincial quintiles do not necessarily contain 20% of villages. Local Institutions and Service Delivery in Indonesia 82 References Bachtiar, H. 1994. "Indonesia's Society and Culture in 1994." Indonesian Quarterly, 23(2): 175-81. Coleman, J. 1988. "Social Capital in the Creation of Human Capital." American Journal of Sociology. 94 (Supplement) S95-S 120. 1990. Foundations of Social Theory. Cambridge, Mass.: Harvard University Press. Evers, P. 1998. "Village Government-Does the Current System Enhance Good Government?" Mimeo, Social Development Department. Washington, DC: World Bank. Freedheim, S. 1988. "Why Fewer Bells Toll in Ceara: Success of a Community Health Worker Program in Ceara, Brazil." Master's thesis, Massachusetts Institute of Technology, Cambridge, Mass. Geertz, C. 1968. Islam Observed-Religious Development in Morocco and Indonesia. Chicago: University of Chicago Press. Grootaert, C. 1997. "Social Capital: The Missing Link?" Chapter 6 in Expanding the Measure of Wealth-Indicators of Environmentally Sustainable Development. Washington, DC: World Bank. 1998. "Social Capital, Household Welfare and Poverty in Indonesia," mimeo, Social Development Department. Washington, DC: World Bank. Hino, T. 1993. "Community Participation in 'Programme de Restructuration de l'Hydraulique Villageoise' in Cote d'Ivoire." Africa Country Department II, Washington, DC: World Bank. Krishna, A., N. Uphoff, M. Esman (eds.). 1997. Reasons for Hope-Instructive Experiences in Rural Development. West Hartford, Conn.: Kumarian Press. Meinzen-Dick, R., R. Reidinger, and A. Manzardo. 1995. "Participation in Irrigation." Environment Department Paper3. Washington, DC: World Bank. Narayan, D. 1995. "Designing Community-Based Development," Environment Department Paper No. 7, Washington, DC: World Bank. Local Institutions and Service Delivery in Indonesia 83 Narayan, D. and L. Pritchett. 1997. "Cents and Sociability-Household Income and Social Capital in Rural Tanzania." Policy Research Working Paper No. 1796, Washington, DC: World Bank. Ostrom, E. 1995. "Incentives, Rules of the Game, and Development." In M. Bruno and B. Pleskovic, eds., Annual Bank Conference on Development Economics 1995. Washington, DC: World Bank. Pathan, R., N. Arul, and M. Poffenberger. 1993. "Forest Protection Committees in Gujarat-Joint Management Initiatives." Reference Paper 8. Prepared for Sustainable Forest Management Conference, sponsored by the Ford Foundation, Delhi. Putnam, R. with R. Leonardi and R. Nanetti. 1993. Making Democracy Work: Civic Tradition in Modern Italy. Princeton, NJ: Princeton University Press. Scoones, 1. and F. Matose. 1993. "Local Woodland Management: Constraints and Difficulties for Sustainable Resource Use", in: Bradley, P. and K. McNamara (eds.) Living with Trees: Policies for Forestry Management in Zimbabwe. World Bank Technical Paper No. 210. Washington, DC. Serageldin, I. 1996. "Sustainability as Opportunity and the Problem of Social Capital." Brown Journal of World Affairs 3(2): 187-203. Tendler, J. 1997. Good Government in the Tropics. Baltimore: Johns Hopkins University Press. Thorbecke, E. 1991. "Adjustment, Growth and Income Distribution in Indonesia." World Development 19(11): 1595-1614. Tjiptoherijanto, P. 1996. "Poverty and Inequality in Indonesia at the end of the 20'h Century." Indonesian Quarterly 25(3): 251-75. Townsend, R. 1994. "Risk and Insurance in Village India." Econometrica 62: 539-91. Uphoff, N. 1992. Learning from Gal Oya-Possibilities for Participatory Development and Post-Newtonian Social Science. Ithaca, NY: Cornell University Press. 1993. "Grassroots Organizations and NGOs in Rural Development: Opportunities with Diminishing States and Expanding Markets." World Development 21(4): 607-22. Local Institutions and Service Delivery in Indonesia 84 Uphoff, N., M. Esman and A. Krishna. 1998. Reasons for Success-Learning from Instructive Experiences in Rural Development. West Hartford, Conn.: Kumarian Press. Werner, S. 1997. "Public Development Services-Experiences from Jambi, Sumatra." Mimeo, Social Development Department, World Bank, Washington, DC. 1998. "Local Level Institutions and Collective Action." Mimeo, Social Development Department, Washington, DC: World Bank. Woolcock, M. 1998. "Social Capital and Economic Development: Toward a Theoretical Synthesis and Policy Framework." Theory and Society 27(2): 15 1-208 World Bank. 1996. "Indonesia-Dimensions of Growth." Country Department III, East Asia and Pacific Region, Report No. 1 5383-IND, Washington, DC. 1997. World Development Report 1997: The State in a Changing World. Washington, DC. 1998. "The Local Level Institutions Study: Program Description and Prototype Questionnaires." Local Level Institutions Working Paper No. 2, Social Development Department, Washington, DC: World Bank. Yunus, M. 1997. "The Grameen Bank Story: Rural Credit in Bangladesh." Chapter 1 in Krishna, Uphoff and Esman (eds.) Reasons for Hope-Instructive Experiences in Rural Development, West Hartford, Conn.: Kumarian Press.