October 2010 56931 Articulations of Local Governance in Timor-Leste Lessons for Local Development under Decentralization David Butterworth and Pamela Dale This policy note is a product of the collaboration between the Australian Agency for International Development (AusAID) and the World Bank on the East Asia and Pacific-Justice for the Poor Initiative. Justice for the Poor is a World Bank research and development program aimed at informing, designing and supporting pro-poor approaches to justice reform. It is an approach to justice reform which sees justice from the perspective of the poor and marginalized, is grounded in social and cultural contexts, recognizes the importance of demand in building equitable justice systems, and understands justice as a cross-sectoral issue. Justice for the Poor policy notes provide policy advice to partner governments, NGOs, donors and development organizations. Policy notes are refereed by at least two external referees who are independent of the program. Copyright statement The material in this publication is subject to copyright. However, the World Bank encourages dissemination of its work and this publication may be reproduced in part or in whole with full attribution. Disclaimer The views expressed herein are those of the authors and do not necessarily reflect those of the Executive Directors of the World Bank or the governments they represent, or the Justice for the Poor program. The World Bank does not guarantee the accuracy of the data included in this work. Contact details Justice for the Poor Justice Reform Practice Group Legal Vice Presidency World Bank 1818 H Street, NW Washington, DC 20433 USA Email: j4p@worldbank.org All Justice for the Poor publications are available at http://www.worldbank.org/justiceforthepoor Comments are welcome and should be addressed to Pamela Dale (pdale@worldbank.org) and David Butterworth (david.butterworth@anu.edu.au). Acknowledgements The authors are grateful to Gregory Ellis (Senior Operations Officer, OPCFC), Kay Engelhardt (Consultant, EASHD), Paul Keogh (PFM/Governance Advisor, AusAID Timor-Leste), Susanne Kuehn (UN Capital Development Fund [UNCDF] Timor-Leste), David Mearns (Associate Professor, Charles Darwin University), Douglas Porter (Senior Public Sector Specialist, EASPR, World Bank), Ingo Wiederhofer, Anton Baaré, Samuel Clark, Angela Khaminwa, and Jose Mousaco of the Youth Development Program team, Antonio Franco (former Country Manager, World Bank Timor-Leste), and Lene Ostergaard, Rod Nixon, and Matthew Stephens of the Justice for the Poor Timor-Leste team for their useful feedback. Both the report and the research it grew from benefitted from the cooperation and assistance of the Ministry of State Administration and Territorial Management (particularly Director Miguel de Carvalho), Local Development Programme (LDP) staff, UNCDF, AusAID, Luta Hamutuk, and TIM Works staff and directors 1. Introduction Meaningful inclusion of all relevant actors at the local level is decisive for successful local development, to ensure that different local power structures work with each other (Lutz and Linder 2004, 2). There are no poor people here; we are just not given the means (chefe aldeia, Aileu district). Engaging with different types of In 2012, Timor-Leste is expected to implement a model of governance systems that have decentralization under which some functions of central legitimacy at local and state levels is government would be devolved to 13 municipalities (mapped a key challenge for the government to present-day districts). Municipal government will comprise of Timor-Leste (GoTL) as it pursues an elected assembly and mayor, as well as a municipal decentralization. administration led by an appointed civil servant. Devolved powers will be designated by decree law and are expected Decentralization`s authors should initially to include services such as primary health and water.. recognize the ways in which citizens Municipal funding will be derived from own revenue, and local leaders, whose lives are transfers, block grants, external funding, and cash reserves. grounded in customary, colonial, and While most funds are earmarked for specific expenditures, the resistance heritages, understand and municipal assembly will designate and allocate block grants (based on the existing Local Development Programme (LDP) engage with democratic state block-grant structure), subject to some restrictions. authority. Suco-municipal relations under decentralization are still This policy paper takes an largely undefined. Under the present draft law, which is understanding of this challenge as its pending approval before parliament, suco councils will receive starting point. It supports the a budget from the municipality to execute council functions, conclusion that a clear strategy of and will be charged with creating community-development working with communities to identify plans. However, suco representatives will have no formal role local development priorities and on municipal development planning boards and the design, implement, and monitor these municipality will not be required to adopt sucos' community- programs is essential to establishing development priorities. While municipalities will be required to encourage the involvement of sucos in planning and ownership and local legitimacy and decision making, municipal accountability will be achieving the full potential of local predominantly upwards and horizontal, rather than development efforts under downwards. decentralization. In an effort to build upon the government`s substantial efforts in this regard, this paper looks specifically at the ways in which communities and local government work together to plan and implement local development initiatives and attempts to answer the question, What steps might the government of Timor-Leste take to amplify the voice of citizens in development planning and improve the state's responsiveness to those voices?1 This paper is premised on the understanding that the GoTL is committed to pursuing a model of decentralized government whereby much of the decision making regarding local development will be devolved to the municipal level. It is also accepted that the Local Development Programme (LDP), which was launched in part to prepare local governments and communities to develop and implement development initiatives, has and will continue to serve as the model 1 This is a key aim of decentralization. For additional information on the purposes of and models for decentralization, see for example World Bank, East Asia Decentralizes; and United Cities and Local Governments, Decentralization and Local Democracy in the World. 1 for the distribution of community-development grants under a decentralized government. (For a broader discussion of LDP, please see Annex 3). The authors recognize that an effective model of decentralization, particularly in an emerging state such as Timor-Leste, must answer a much broader set of questions. These include but are not limited to the ways the state can build a relationship directly with citizens, fiscal and administrative arrangements between branches of government, the impact of decentralization on stability and service delivery, and the devolution and management of traditional state-provided services.2 While these are important considerations, they are beyond the scope of the current study and this policy paper.3 Summary of the report In the remainder of this report, we discuss how the convergence of different governance systems at the local level, and the wider Timorese experience of sociopolitical change, can impact the way local development initiatives are perceived and implemented. We attempt to examine local development through the eyes of local authorities and community members, many of whom have experienced firsthand the various iterations (traditional, Portuguese, Indonesian, UNTAET, CNRT and RDTL4) of statehood and governance in the nation (see for example Fox 2003; Hohe 2002; McWilliam 2009).5 Using this analysis, supported by data from original fieldwork in two districts that examined participation and decision making in local development initiatives, we make four interrelated arguments.6 In each section we address a different theme, but our structure of argumentation is the same. We initially describe the dynamic construction and transformation of authority and decision making at local levels as differing governance paradigms meet, and then, based on this description, examine ways local development models under decentralization can effectively engage with both customary and formal state systems of governance. The four arguments are as follows: Local Political Identities: We discuss how values associated with political legitimacy and decision-making behavior are strongly linked to personal identification with place and community. We highlight the constraints these identities impose on--and also the potential they offer to--local development-planning models. Cooperation and Competition for Development Benefits: We show that patterns of cooperation and competition are underpinned by customary community allegiances and that on top of this, development initiatives can themselves impact patterns of, and incentives to, cooperate or compete for resources. 2 For more discussion of the potential for decentralization efforts to increase, as well as reduce, social and political conflict, see for example Siegle and O`Mahony, Assessing the Merits of Decentralization. 3 This is one of a series of papers describing and analyzing research findings. Accompanying papers, which can be found on the Justice for the Poor Timor-Leste Web site, provide (i) detailed findings and recommendations for the LDP team on operational aspects of the program and (ii) a larger discussion of relationships between citizens, nonstate authorities, and the state in the context of an expanding state. See www.worldbank.org/justiceforthepoor. 4 CNRT is the acronym for Concelho Nacional da Resistência Timorense (National Council of Timorese Resistance), UNTAET refers to the United Nations Transitional Administration in East Timor, and RDTL is the República Democrática de Timor-Leste (Democratic Republic of Timor-Leste). 5 The authors see this question of dynamic relationships between state and non-state authorities as a key point in any discussion of decentralization and development. However, a larger discussion is outside the scope of this paper. 6 The analysis in this paper is largely based on study of LDP. To a smaller extent, the TIM Works cash-for-work program and the Youth Development Program, a youth grant mechanism based on the LDP model, were analyzed. For a detailed description of research methodology, please see Annex 2. For information about all programs studied, please see Annex 3. 2 Defining Relationships between Civil Servants and Suco Authorities: We highlight that district- and sub-district-level civil servants are the closest formal state authority to the population, but that aldeia- and suco-level authorities are the most visible and accessible form of governance for the majority of the citizenry. The relationship between chefes suco/aldeia and civil servants is a key intersection between the citizens and the state, and is thus crucial to the design of effective local development models. Creating Equitable Dispute-Management Mechanisms: We show that many problems, misunderstandings, and disputes that occur in the implementation of local development programs are not efficiently resolved because local actors are uncertain of their rights and responsibilities in this regard. While customary, church, and formal (state) dispute- resolution mechanisms are relevant and effective in certain spheres, they can be integrated into a coherent system that would have a better-defined legitimacy. This policy paper is not intended as a static document. Rather, the Justice for the Poor (J4P) Timor-Leste team hopes that it will form part of a continued dialogue with the GoTL and other partners on the potential impacts of local conceptions of justice and governance on local development planning and decentralization more broadly. The purpose of this note, then, is to inform current discussions on community-planning processes, dispute-resolution/complaints mechanisms within local development programs, and other important aspects of the decentralization agenda through a perspective informed by in-depth, local-level policy analysis. Feedback is encouraged, as are suggestions for additional research, policy, and operational activities that could be of use to the GoTL as it moves forward with planned decentralization activities. 3 2. Local political identities Timor`s ideas of governance are expressed in a great variety of myths, legends and genealogical narratives - ancestral parables for social actions. Nor was there ever one system of governance. Rather there were key principles whose expression and application resulted in a variety of historical outcomes (Fox 2009, 121). In aldeias and sucos across Timor-Leste, the authority to make governance decisions is closely tied to community and geographical identity. In simple terms, particular people from certain families in certain areas are customarily given more right than others to make decisions on behalf of the community. In this section we discuss the potential benefits and drawbacks of this system as they are expressed in the implementation of LDP. Customary perceptions of who should and should not hold authority need to be carefully negotiated, to ensure both equal representation (as, for example, in encouraging female political participation) and political legitimacy (so that communities respond to the decisions made). Traditional political systems in Timor-Leste are based on membership to, and alliances between, houses.7 In different areas and among different alliances, a number of houses are linked together and ordered by precedence--a hierarchy in which certain groups are classified as superior to others. For example, the first house to arrive in a domain (area of settlement that generally maps onto aldeia and/or suco) is accorded rights to land and holds sacred and political responsibility over subsequently arriving groups.8 Interestingly, this pattern is sometimes reversed so that political authority is invested in newcomers whose power is held separate from, but must be legitimated by, the spiritual mandate of first-comers.9 In either articulation, politically powerful houses have been heirs to colonial-mandated political positions, such as chefe suco. In sum, in customary governance systems, which are still very much relevant in the region today (see, for example, Mearns 2009; Vischer 2009), political legitimacy is drawn from identification with community (descent group and alliances) and place (domain). Field-research findings support this contention and, moreover, highlight how such local political identities can impact development initiatives. As the GoTL expands its presence across the territory through decentralization, there is an opportunity for the state to open new and positive relationships with the nation`s communities that are responsive to local political identities. Given the importance of these identities to political and cultural life in rural Timor-Leste, the government`s willingness and ability to positively respond to these realities will likely determine its degree of acceptance, legitimacy, and effectiveness in implementing local development activities under a decentralized governance structure.10 Responsive, locally legitimate models of public service delivery and local development will be key to developing these robust state-citizen relationships. The selected 7 Houses are conventionally defined in the anthropological literature on Timor-Leste as lineal descent groups (clans or lineages in which membership is decided through either the father`s or mother`s lines). Houses are also commonly referred to as origin groups because the group collectively recognizes certain ancestors as its founders or origin. 8 A cosmological foundation of autochthonous Timor-Leste governance is the separation, but complementarity, of sacred authority and political power. See Fox, Inside Austronesian Houses. Both can be held by the same house, but are then embodied by different lineages or individuals. 9 Such reversal is a formal property of precedence and common in many spheres of social life. See Fox, Category and Complement. 10 Note that while suco representatives are generally well respected within their communities and surveys show both strong confidence in these leaders and a preference for traditional leaders in decision making and dispute resolution, there are certainly variations within and between communities. 4 local development model and processes will also need to maintain their legitimacy when measured against both customary systems and the wide range of planning and implementation arrangements used by local and international nongovernmental organizations (NGOs), whose work often overlaps with (or substitutes for) the state`s service delivery and local development- planning programs. LDP has made strong headway in gaining the trust and support of local communities. The participatory planning and prioritization processes and the consistent cycle of funding have created a significant feeling of local ownership over the infrastructure projects developed through the program. This is no small point, as communities largely measure the value of a government-sponsored project in terms of the degree to which it is owned by their community, consistent with the overall localization of identities.11 Ownership of projects through direct community participation is also proven by global experience to typically deliver better quality community infrastructure at a lower price (Guggenheim et al. 2004; World Bank 2004).12 Further, the engagement of communities in decision making over their own development requirements can build trust in the state as an entity that respects community views and needs, thus contributing to the goals of stability and democratic participation.13 While the program`s participatory processes contributed to community satisfaction with LDP and an accompanying recognition of the benefits that can be provided by the state, some small changes in program implementation could make it more responsive to community expectations. For example, although LDP prioritizes information sharing and community involvement, customary power relations and authority structures can limit consultation and access to information. Much of the program knowledge is held by powerbrokers within the suco councils, and citizen participation in project-prioritization meetings is often limited. Even when access to information exists, it is often confined to basic details about a subproject, making it challenging for community members to participate actively in project design, monitoring, and evaluation. This lack of information has contributed to a sense that LDP, though in some ways a participatory program, is outside the community. The importance of local political identities, and the way information and participation converge to either enhance or detract from the localness of a project, are explored in more detail in the case study below. 11 This is not to say that communities object to or do not see the value in projects such as secondary schools, hospitals, and large roads that are selected and built by actors outside the community. Doubtlessly, such projects also increase the legitimacy of government in the eyes of underserved citizens. However, we are speaking of ownership in the sense of an informed citizenry that both receives from and feels responsibility towards the community and its resources, and that will (for example) willingly give time and labor to ensure the maintenance and repair of those community resources. With LDP, where there is a clear expectation that communities will maintain a facility built using project funding, there is equally an expectation on behalf of communities that the project be their own. 12 For example, evidence from three community-driven development projects in India showed a savings of 11­56 percent on community-contracted projects as compared to similar projects implemented through the State Public Works Department. Approval, contracting, and construction also took place in approximately half the time. See A. K. K. Kumor et al., Community Based Procurement: Value for Money Analysis (New Dehli: World Bank, 2009), http://go.worldbank.org/JF75ZX62T0. Accessed July 2010. A review of the KALAHI-CIDSS community development program in the Philippines found that the unit cost of project infrastructure was substantially less than the cost of similar projects funded by government agencies, ranging from 8 percent for schools to 6 percent for water and sanitation projects. See E. Araral and C. Holmemo, Measuring the Costs and Benefits of CDD: The KALAHI-CIDSS Project, World Bank Social Development Papers, No. 102 (Manila: World Bank, 2007). 13 Analysis of the KALAHI-CIDSS program in the Philippines provides a useful example. In villages that were successful in receiving local development funds from this project, trust in local government officials increased by 10.7 percent. See Julien Labonne and Robert S. Chase, Do Community-Driven Development Projects Enhance Social Capital? Evidence from the Philippines, World Bank Policy Research Working Paper No. 4678 (Washington, DC: World Bank, 2008), http://go.worldbank.org/AZ31FASA50. Accessed July 2010. 5 Box 1: Keeping it in the community: Defining LDP through a local lens In a small, urban aldeia in Lautem, local political identities played a crucial role in shaping the implementation of an LDP project. In 2007, the aldeia received a project to upgrade the small access road linking it with neighboring communities. The project was conceived through consensus decision making by local leaders, in which citizens were consulted while an executive decision was made by the chefe aldeia and other elders (lia na'in). The leaders also created a design for the project that they believed was most suited to local conditions and that would make most use of local resources. In this sense, customary leadership processes were practiced, and a sentiment of belonging to a local, bounded community was expressed, with compatibility to LDP procedures. However, from this initial stage, decisions over project selection, design, and implementation became more and more removed from the community, and local dissatisfaction grew in parallel. The project underwent so many changes as it progressed through the sub-district assembly to the district assembly, where its design and budget were eventually finalized, that upon implementation, the community felt betrayed by its perceived lack of quality. According to senior district public servants and the contracting company, the project was correctly implemented per the design produced by the district technical staff and agreed to by the assembly. However, the local population, including the chefe aldeia, was not informed of these technical changes, and the completed project did not meet their expectations. Community mistrust of external government was also apparent in their unwillingness to express complaints to the district monitoring and evaluation team, instead preferring to organize a community workday to improve project quality themselves. Recommendations The lessons on enhancing local ownership from analysis of the LDP program might be taken on board as the GoTL moves forward with local development planning under decentralization. In particular, the government might consider that: The LDP and Youth Development Program (YDP), implemented by the Ministry of State Administration and Territorial Management (MSATM), focus on transferring responsibility for local development planning to the lowest level at which capacity exists, indicating a strong understanding of both community capacities and local beliefs about ownership. These community-based approaches warrant long-term support from the government. The model for local development planning under the draft decentralization law, however, risks cutting out sucos and, by extension, communities, from direct decision-making authority. The YDP approach differs from LDP in that it places funding, decision-making power, and responsibility for project implementation directly in the hands of community members. If piloting of YDP is successful, the government might consider expanding and scaling up this community-driven approach. By tapping into local channels of communication (for example, sharing information at churches, markets, sporting events, and other gathering spots), public information offices within the municipality will have the potential to reach broader and more diverse audiences than those captured in community meetings. The community role in local development can be strengthened through participation throughout the planning and implementation cycle. When communities are involved in project planning, design, tendering, and monitoring, they are likely to show more ownership over community resources. Evidence from many similar contexts has also shown that community-based project tendering is often more cost-efficient. Of course, community contracting is not a fix-all--community capacity to implement development projects varies 6 widely--but experience from similar environments has shown that where communities are provided with adequate support, community contracting has clear benefits. There may be room for an enhanced role for civil society; in particular, monitoring municipal assembly meetings, participating in community-development planning, evaluating and supporting municipal service delivery, addressing grievances that may arise, and supporting improved public information mechanisms could prove useful.14 Research in similar programs has shown that civil society involvement at only the latter stages of project planning or review is not sufficient; to be effective, engagement must continue from the earliest stages of project identification through to technical assistance and monitoring (Peterson and Muzzini 2005). Where possible, coordination with civil society on local development will also be a priority. At present, civil society organizations are implementing projects that are often similar to those that are in place under LDP (and by extension, to the types of local development activities that will be implemented according to municipal development plans). Differences in planning and selection processes and poor coordination could contribute to confusion, gaps, and duplication in local development planning. Targeted civic education campaigns for suco councils and citizens, tied to the community- development planning process, could help to build citizens` awareness and participation. In particular, these campaigns could focus on topics such as how project selections are made (both through community and municipal development plans), the roles of civil servants, community leaders, and citizens within development planning, the value of participatory processes, and so on. The intent of these campaigns would be to help community leaders and citizens understand the reasons behind and value of the steps in the participatory development process, so that they are not seen only as burdens. 14 While civil society groups in Timor-Leste experience substantial capacity constraints, particularly outside Dili and the major urban areas, they can potentially play an important role in local development planning. During research, the team observed civil society representatives taking part in a district-integration workshop, and civil society representatives were also very vocal in project selection under YDP. The church plays a substantial role in development in Timor-Leste, and several national NGOs monitor implementation of development programs as part of their mandate. 7 3. Cooperation and competition for development benefits Intricate networks of affiliation, exchange and alliance within and between Timorese houses of origin represent the historical and continuing basis for the reproduction of Timorese society (McWilliam 2005, 34). All in the community want to be part of the projects, but because the projects are small scale it is not possible for everyone to be involved equally ... and so we use a rotation system to avoid problems, those who don`t get work this time will work on next year`s projects (chefe suco, Lautem district). We understand the limitations of the budget, and we have to give the chance to other suco, because they also have needs (chefe suco, Aileu district). Local conceptions of authority and equity impact the ways in which programs and policies are viewed and implemented at the local level. Conversely, it is also the case that through program design, development actors can impact and transform the ways in which such local conceptions are manifested. In this section we discuss how individuals` and communities` incentives to cooperate or compete for development benefits are patterned by the interplay between local realities and development initiatives. In particular, we emphasize that the choice to implement a one-time, rather than a repetitive, benefit or program can produce confusion and conflict at local levels. Communities in Timor-Leste engage in complex relationships, where tension between states of competition and cooperation are the norm. Cooperation between houses allied by marriage is expressed through exchange relationships, in which bride-wealth goods and services are circulated among houses--a process that can help sustain livelihoods in difficult conditions. These exchange relationships are primarily characterized by asymmetric reciprocity, in which the house one gives to is different from the house one receives from (see Diagram 1, below). But exchanges can also involve discrete moments of direct reciprocity, such as when bride wealth is exchanged between two houses.15 Alongside these cooperative relationships, competition in a customary sense is primarily manifested as either structural contestation over orders of precedence or political and economic contestation over scarce resources. Diagram 1: Asymmetric Exchange Relationships A D B C 15 The asymmetry of exchange is underpinned by the prescriptive asymmetry of marriage, in which a particular house cannot give and take wives (or husbands if the system is matrilineal) from the same house. 8 The ideology of cooperation and competition that is grounded in customary social organization plays out at various points in the implementation of LDP (particularly as this organization maps onto aldeia and suco identity). Cooperation between sucos is an important part of subproject selection in the sub-district and district assemblies. In interviews with suco council members, over one-quarter independently made reference to the practice of lobbying or swapping projects--that is, offering support for a particular suco's subproject one year, with the expectation that one`s own subprojects will receive support in subsequent years. While not envisioned by the LDP project-implementation guidelines, this practice is analogous to customary exchange relationships. Importantly, this practice is enabled by the multiyear implementation of LDP, which allows for communities to engage in reciprocal relationships with the expectation of benefitting in future project cycles. Cooperation in LDP can have both positive and negative impacts on sub-project implementation and community development. It can prepare communities for models of democratic decision making and community development, where benefits are not necessarily distributed equally -- the decision-making model under LDP requires parties to evaluate the merits of interventions and make trade-offs (though the selection process does not necessarily conform to legal-rational arguments or assessments of merit). Cooperation in the form of swapping may reduce the number of very small budget (and likely small impact) projects that would normally be selected, while leading to prioritization among aldeias of subprojects that are seen to benefit the larger community or suco--for example, health centers, schools, and water and sanitation facilities. In a resource-constrained, recent post-conflict state such as Timor-Leste, the sharing of project benefits likely has a conflict-mitigating effect. However, the desire to share benefits has also led to inefficiencies in project processes--for example, an informal arrangement in Lautem district whereby all eligible contractors would receive at least one LDP contract per year. 16 While this arrangement may prevent allegations of favoritism in the distribution of LDP contracts, it also undermines the free market model of tendering and can weaken contractors` incentives for implementing high-quality, cost-effective projects. Because project resources are not sufficient to fund the full range of community needs, some degree of competition has been built into the LDP design. However, community dynamics and some elements of program design can prevent the healthy competition envisaged by LDP. For example, the considerable power given to ostensibly non-voting civil servants to evaluate, plan, and cost subprojects can allow them to lead assembly voting members to select the projects they find appropriate, thus undermining the role of community leaders (and by extension, the participation of citizens) in community development. This occurred during a district assembly meeting attended by J4P field researchers, where the district development officer (DDO) stated that of the eight projects eligible for selection, only four could be considered as technically feasible. The DDO provided no explanation for the elimination of the other four projects and expressed a clear preference for two particular projects, stating that only two projects in a certain combination could be selected because of budgetary constraints. Though other public servants voiced their dissension, the projects favored by the DDO were eventually selected. We can see that new patterns of cooperation and competition between aldeias and sucos are emerging, as local values and identities interact with LDP`s democratic model. A comparison between LDP and TIM Works serves to draw out the implications of program design that is responsive to local values and identities.17 By creating a reliable annual funding cycle that allows for the development of culturally relevant, reciprocal exchange relationships, LDP 16 Every contractor and civil servant interviewed in Lautem district knew of this arrangement. 17 LDP and TIM Works operate under different models of planning and project implementation. For more information on both of these programs, please see Annex 3. 9 prevents competition over these development resources from arising. In contrast, programs that are seen as one-time (or short-term) benefits can lead particular communities or individuals to attempt to hoard resources. The case study of a TIM Works project highlights this scenario. Box 2: How project design can turn cooperation into competition: experiences from TIM Works A TIM Works maintenance project to improve an existing road in the heartland of Mambai traditional culture, where customary clan affiliation and political identity are strong, first triggered a novel expression of cooperation and then transformed it into competition. While the original project design--told to the chefe suco and chefes aldeia at a district meeting-- required 20 full-time workers for five months, eventually a decision was made by the local parties to employ 100 workers over the same time period. In order to share employment between the five aldeias through which the project would run, it was agreed that work would proceed in each aldeia for one month and that each aldeia would provide 20 workers for work conducted only in their own aldeia. However, after the project had begun in one aldeia, the TIM Works district team changed the original direction of the project after discovering that the roadway initially slated for maintenance included two small bridges that could not be budgeted for. Thus, the project was redirected away from the road servicing all five aldeias onto a road that services only two. Through unclear communication between TIM Works staff and local actors, the aldeias that would now miss out mistakenly understood that while the change was necessary, the project would eventually continue into their communities. By the time of the field research, the redesigned project had been completed and tensions between aldeias were rising. While the members of the non-beneficiary aldeias did not begrudge the others their opportunity for work, they were increasingly concerned that the project would not return and they would miss out altogether. Recommendations Though customary exchange relationships have adapted somewhat to fit new project models (for example, through the lobbying process), they remain an important feature in the interactions between people and communities in Timor-Leste. The GoTL faces a significant challenge in designing local development programs that provide individuals and communities with the appropriate incentives to cooperate rather than compete for resources, while also instituting transparent and effective processes. Although finding this balance will require continued thought and discussion, we preliminarily recommend that: Emphasis is and should be placed on long-term community-development planning, which allows for predictable and transparent funding and the maintenance of reciprocal relationships. The development of reciprocal relationships could be important to prevent competition from manifesting in potentially destabilizing ways. Suco officials and civil society should be actively engaged in planning and monitoring development resources at the municipal level, so that they are able to understand and explain to communities the budget constraints and prioritization processes. In addition to preventing allegations of corruption or bias in projects, access to information can help citizens take an informed role in building the state and identifying their priorities. Ideally, suco officials and civil society could participate in decision making through selected representatives of these bodies. At minimum, observer status on municipal planning committees would enable these actors to play a role in ensuring a consistent feedback loop. Involving civil society and community representatives in project tendering could also increase transparency in the competitive selection of project contractors. Civil society and 10 community engagement in project evaluation, as well as more effective social auditing mechanisms, could allow for consideration of past performance to evaluate tenders. Again, the strength and capacity of civil society organizations in Timor-Leste varies substantially, and working relationships between suco representatives and civil society representatives may likewise vary between districts, but the active involvement of civil society representatives provides communities with an information source outside of traditional power structures. An active and well-funded public information office at the sub-district (or sub-municipality) level, which makes use of a wide range of communication techniques (radio, information booths at markets, easily accessible and understandable printed material), can keep communities informed about the development programs that are planned and implemented in a given year. Beyond helping citizens take a role in monitoring service delivery, this would enable communities to develop additional ownership over the projects and the democratic decision-making processes around their selection. The role of the public information office could include the distribution and socialization of suco development plans. Effective complaints/grievance mechanisms, which link sucos to the municipal planning board, would allow for a more effective response to potentially destabilizing allegations that development benefits are distributed unfairly or preferentially. To promote positive competition between municipalities, citizen-perception data on municipal performance could be collected and widely disseminated. Perception data could include at minimum key indicators, such as citizens` views on official corruption and responsiveness of both municipal and local government. 11 4. Defining relationships between civil servants and suco authorities There needs to be more civic education, not just among civil servants, but also among the community, so that they can become more involved in decentralization (senior civil servant, Aileu district). The relationships between chefe suco and the DDO and CDO [Community Development Officer] are very important in the selection of projects: There are political processes at work (chefe suco, Aileu district). As discussed in the preceding sections, our research findings confirm the existence of strongly localized identities within communities and deeply held beliefs on responsibilities for project implementation. These findings suggest that community-driven development models, with their emphasis on community participation in all aspects of planning, implementation, and monitoring of development projects, are an appropriate model of engagement between the state and the citizens. While direct participation of citizens in development planning would be the ideal result, the authors recognize that community authorities continue to play a crucial role, and are an important interface between citizens and the state in wider development-planning initiatives. Looking down from the perspective of the state, district- and sub-district-level civil servants are the closest formal state authority to the population. However, from the standpoint of the majority of the citizenry, particularly in rural areas, aldeia- and suco-level authorities are the most visible and accessible form of governance, and relationships with the formal state are often mediated through these authorities. This relationship is a key intersection between the citizens and the state, and is therefore crucial to the design of effective local development models and to the government`s decentralization agenda. Thus, in this section, the authors discuss the important role of the suco council members (as representatives of the community) in facilitating relationships between the citizens and the state, and propose ways to create greater synergy between civil servants (as representatives of the state) and suco authorities. During research into the implementation of LDP, the team consistently noted disparities between civil servants and suco authorities in their knowledge of LDP (and community-development resources more broadly). While civil servants universally had a strong appreciation of the goals, processes, and implementation of LDP, suco leaders in both Aileu and Lautem indicated that their limited understanding of LDP had led to problems, such as presenting proposals incorrectly, confusion about subproject monitoring and evaluation procedures, and difficulty in finding the proper channels to express complaints. At its worst, this imbalance in information creates an inequality in power that can lead the more knowledgeable party to take over decision making. If the municipal assemblies and local-level development planners are not anchored by strong downward accountability requirements, it is possible for community voices to go unheard in development planning. Currently, district-level officials are above all responsible to the central government. Under the proposed decentralization model, this would continue largely unchanged, although civil servants would have additional (horizontal) accountabilities to the municipal administration and assembly (who are presumably accountable to citizens).18 Thus, while 18 We recognize that the accountability of civil servants is to the government, rather than the population directly. It is the elected governments who ensure that civil servants perform their roles in a way that is responsive to the 12 information sharing and consultation with sucos is required, formal structures for downward accountability in local development planning remain weak. Imbalances such as these are further emphasized by the current draft Law on Local Government, which requires that Municipal Development Plans align with National Development Plans, but does not include similar obligations for plans to adhere to or incorporate Community Development Plans. In Indonesia, where a comparable model of district planning and development budgets exists, the World Bank`s Local Level Institutions Study showed that concentrating project selection and decision making in district officials led to just 13 percent of community proposals being incorporated in the District Development Budget in the 1997­98 fiscal year (Evers 2000).19 Lack of attention to community priorities can undermine incentives for local participation in community development; one local Indonesian leader asked, What`s the use if the proposals can be unilaterally overruled without consultation? (Evers 2000, 24). In addition to having greater access to information, civil servants` proximity to the state is also reflected in their superior access to financial resources and technical expertise. Under the LDP program, while project selection takes place at the community level, technical aspects of design, costing, tendering, and implementation are largely at the discretion of the sub-district- and district-level civil servants. The concentration of both resources and information in the civil service can contribute to the marginalization of suco-level authorities (and, by extension, communities) from full and active participation in local development planning.20 According to the 2009 Law on Community Authorities and their Election, the suco is a community organization and community leaders, though elected, are not part of the public administration. While the suco council has specific responsibilities21 (including those related to economic development), and a limited budget is provided in support of these functions, the role of the sucos in governance is restricted. However, despite the legal and policy limitations on the formal roles of the suco council, council members remain the key source of external relations, information, and dispute resolution for communities. To a considerable extent, while direct relationships between the citizens and the formal state are being developed, the way the state engages with suco representatives will help determine its legitimacy in the eyes of citizens. Contradiction between the official and unofficial roles of suco officials has contributed to some confusion over LDP responsibilities. In cases such as that highlighted in the box below, suco representatives have not fully appreciated their responsibilities and have thus not fulfilled their obligations. This was particularly so in areas such as program monitoring and evaluation, where the roles of civil servants and suco councils had some overlap. population, thus ensuring downward accountability. However, given the role of civil servants in selecting, planning, and implementing community-development priorities, the accountabilities of civil servants to the central government and municipal government (in implementing municipal development plans) might overshadow community priorities. Further, expectations of municipal assemblies` ability to genuinely and effectively represent community interests in the initial phase of decentralization should be managed. 19 While this was from the predecentralization period in Indonesia, the outcome demonstrates how communities can be marginalized from local development-planning processes when discretion over decision making lies with district governments. 20 For a more detailed discussion of the impact of information and resource disparities on the Local Development Programme, please see World Bank, Redefining Local Governance through Development Initiatives: Design Contributions for Community-Driven Development Programs in Timor-Leste. 21 Per Chapter 2, Section 1, Article 10, suco leaders may develop activities in areas such as peace and social harmony, census and registration of population, civic education, promotion of official languages, economic development, food security, environmental protection, education, culture and sports, and assisting in the maintenance of social infrastructure. 13 Box 3: How disconnections between civil servants and local authorities undermine project success During a project to provide clean water to rural schools and surrounding houses, the relationship between the chefe suco and sub-district and district civil servants was critical to negotiating contractor misconduct. The local community was dissatisfied with the work of the company contracted to improve two water tanks and replace 163 lengths of pipe because even after being recalled to improve on their original work, the company ultimately replaced just 25 pipes and repaired the tanks with small cement patches and paint. While some houses now receive water, respondents indicate that the schools have yet to feel any benefit. The chefe suco and the sub-district civil servants needed to work closely together to hold the negligent contractor to account. The project site was in a difficult area to reach and direct monitoring by the verification and appraisal team was infrequent. Instead, reporting from the local community to the chefe suco was paramount to informing the civil servants of progress. On the other hand, the contractor hailed from outside the sub-district, and the civil servants were in a better position than the chefe suco to follow up on complaints. While the relationship between the local authority and the civil servants was initially cohesive and they were successful in recalling the company to improve its work, when the chefe suco became ill, the partnership broke down, and with this vital connection fractured, their monitoring efforts were rendered ineffective. Eventually the sub-district could affect only postproject sanctions on the company, with the CDO stating we chased them but could only notify them that they improve their work in September, even though the project should have been finished by March ... [and because of this] ... they can certainly still apply for tenders, but we are not going to give them any more projects. Recommendations Defining the roles of sucos in relation to the civil service and municipal government has been a priority for the GoTL as it debates the decentralization agenda, and it has examined various models of local-national relations to identify lessons that could be applied in the Timorese context. The research findings suggest the need for clearly defined accountabilities between municipal leaders and suco authorities, and active involvement of suco authorities in municipal governance. With this in mind, the J4P team offers the following suggestions: Enhanced communication pathways, through avenues such as observer status on municipal assemblies or the provision of transport costs for council members to attend municipal meetings about issues that impact their suco, could help to ensure that suco leaders are kept informed of developments related to their communities. Additionally, training to prepare community leaders to make evidence-based policy decisions and use available information (for example, from monitoring and evaluation documents) would be useful. Though not without problems, areas such as constituent communication and local-level development planning are currently being undertaken with a substantial suco role. Clarifying suco responsibilities in these areas (including suco involvement in the Public Information Office), and providing resources to match responsibilities could usefully support efforts to build municipal-suco links. Support the role of local media in covering local and municipal politics. Facilitate links between local institutions and national networks to share experiences and strategies. Requiring municipalities to formally report to suco councils on yearly and multiyear development plans could assist sucos to respond to queries from within their communities 14 about the outcomes of local development planning. Further, a requirement that municipal development plans reflect both national and suco development priorities could encourage municipalities to incorporate community projects into their budgeting and planning. In cases where suco council members will be asked to participate in aspects of project implementation within their communities (for example, through community-led monitoring and evaluation efforts), their role could be expanded to include participation in site selection and tendering. Additional training in technical aspects of project implementation may be necessary to facilitate this role. 15 5. Creating equitable grievance-management mechanisms In the short term the rule of law can do little for the poor and disadvantaged` sections of Timorese society who receive minimal protection from the state. In the long term, strengthening the state and its institutions may assist these people but this cannot be guaranteed (Grenfell 2009, 232). According to our Timor traditions, we children do not tell tales to our parents: we work problems out ourselves, and do not complain to the government (chefe suco, Aileu district). Justice is a key pivot on which tensions between community and state play out. Defining equitable and effective mechanisms for resolving disputes requires subtle negotiation between local and state moral values. The resolutions that are legitimate and workable are those that have consensus in and between community and state on basic issues such as defining right and wrong behavior and enacting appropriate sanctions. However, finding that consensus is a challenging task, and this is particularly so when disputes emerge in local-state relationships. As Nixon has pointed out, although local communities have great capacity to dispense justice and resolve conflicts within the village (2006, 94, emphasis in original), this effectiveness diminishes as the sphere of relationships extends beyond village borders. The issue can also be viewed from the top-down, for, as many commentators such as Grenfell (2009) and Babo-Soares (2004) assert, the effectiveness of state law diminishes the further one moves away from the state`s center. Separating community and state might work on certain issues, but when the dispute is between community and state, it could be necessary to formulate syncretic mechanisms. Hybrid models for dispute resolution have been effective, for example in natural resource management (see Meinzen-Dick and Pradhan 2002; Meitzner-Yoder 2007; Palmer and do Amaral de Carvalho 2008). Our research suggests they could be equally useful for protecting citizens` rights under decentralization. Neither LDP nor TIM Works has dedicated grievance-resolution procedures. Instead, in practice both of these programs include strong accountability and monitoring mechanisms that aim to minimize the potential for dispute. If disputes do occur, resolution is sought case-by-case through whichever channel is deemed most appropriate (be that resolution by suco authorities, intervention of civil servants, or application of formal law). However, research indicates that many grievances that do occur in the implementation of LDP and TIM Works--while small-- are not resolved to the satisfaction of the local actors.22 This is primarily because such actors are uncertain of their rights and responsibilities to make or handle grievances. In other words, the hands-off approach taken by the programs to grievance resolution, while accommodating of local conditions, can also be a source of confusion and misunderstanding. Complaints made by citizens on both projects predominantly concerned the quality of subproject design, as well as the quality of construction under LDP. These complaints also refer to many citizens` suggestion that were they more involved in design and construction, fewer issues of 22 While some disputes (for example, about subproject quality or the use of local laborers) were found in both programs, the different implementation models used by the two programs means that there are differences in the disputes experienced. However, the authors have comingled analysis and recommendations for the two programs in this section because many of the underlying causes of disputes--communication challenges, for example--are similar in both programs. 16 low quality would arise. For example, on several of the projects studied, beneficiaries judged the design to be inadequate or unsuitable for local conditions. In these cases, issues included the wrong choice of building materials (for example, poor quality wood), wrong choice of tools (such as inferior quality roadwork machinery), incorrect timing for construction (such as a rushed job during the rainy season), and under-scaling of project (for example, too small or restricted to make a real difference). Furthermore, allegations of corruption, collusion, and nepotism surrounding the programs were often raised, but without effective channels for judgment, such allegations remain uninvestigated. While these cases did not escalate to violent conflict (although in one case, thievery on the project site was attributed to frustration over project quality), they were nonetheless triggers for tensions between community members, contractors, and civil servants (see Box 4 below for a specific example). Box 4: Disputes Arising from Overlapping Responsibilities An LDP project to upgrade a primary school classroom carried out under the aegis of the district office of the Department of Education illustrates the potential hazards of implementing projects without well-defined channels for resolving grievances. While the community, school, and education department all had a common interest in the project, the school headmaster and the chefe suco were marginalized during project selection and implementation. The headmaster told the J4P research team that we just sent the project idea to the district department of education and after that it was all in their hands until it finished, and the chefe suco stated that there was no coordination between the education department and our suco council. The chefe felt that the project had been usurped by the civil servants, and he stated that he did not feel responsible for monitoring and evaluating the project quality because of the involvement of a third party, namely the education department. In the meantime, several quality issues noted by the headmaster--a broken lock and a broken door--were left unaddressed, as the sub-district administration, education department, and contractor attempted to understand their respective responsibilities. The confusion is exemplified in the words of the superintendent of the education department, who asserted that the door was definitely broken, but we don`t know if it is our responsibility to fix because the inspection was carried out by the LDP technical team, and they have not told us whether the breakage occurred within the maintenance period or not. Dispute-prevention mechanisms that require the ideal functioning of all relevant actors in project monitoring were ineffective because of irregular and often unclear communication between project beneficiaries, suco authorities, sector departments, contractors, and civil servants. Without any other avenue to address grievances, the school`s dissatisfaction with project quality and the suco council`s dissatisfaction with their marginalization were left unchecked. Similar problems to those identified in the preceding paragraphs and the box above can be expected under decentralization. Especially in its initial stages, decentralization could lead to greater confusion and perhaps competition between suco officials, municipal governments, and the central government over their roles. Experience in other countries undergoing decentralization suggests that it can be accompanied by an increase in demands for informal payments, theft or misuse of local development funds, inefficiencies or inequality in public service delivery, and allegations of state capture, favoritism, and cronyism in the distribution of development resources (Campos and Hellman 2005). The transfer of authority over resources to community members, bypassing local authorities, also has the potential to increase conflict, though recent evidence shows that effective, community-driven development in postconflict situations can reduce conflict within communities (World Bank 2006). 17 Under LDP, citizens who held complaints about the program were generally unwilling to approach civil servants (such as those acting on monitoring and evaluation committees), preferring to consult their chefe aldeia or chefe suco. However, some chefes were themselves either unsure of how to resolve the complaint (especially when they involve external actors such as contractors) or pessimistic about the impact of referring the complaint to civil servants. For example, in one case examined by J4P, the chefe took the matter into his own hands by tearing down the construction and demanding it be rebuilt. This, and the potential for new and more difficult state-community disputes under decentralization, points to the need for clearer grievance-resolution mechanisms. Recommendations Recognizing that complaints are inevitable in state extensions into local communities and that supporting communities to resolve complaints helps to build empowerment and accountability, it is incumbent upon the state to provide accessible channels for complaints to be heard and judged, in response to community-development programs and more broadly. With this in mind, the J4P team welcomes the MSATM`s request for diagnostic work and technical assistance in the development of grievance mechanisms for local development programs. While local and customary resolution systems alone can become unanchored (such as when chefes do not know how to handle complaints involving an external contractor), these systems must not be disregarded, as respondents unanimously preferred to direct complaints through their chefe. In this light, we make the following recommendations: Incorporate a well-defined grievance-resolution mechanism into decentralization processes. Such a mechanism would build on culturally valid dispute-resolution practices, and procedures could be widely publicized through the Public Information Office. This mechanism would also contain alternative avenues for redress in cases where local means of dispute resolution prove ineffective. Effective social accountability mechanisms can also play a key role in preventing grievances from arising. For example, contractor quality assessments through social auditing and evaluation of past performance could be considered during the bidding process for projects, thus preventing the awarding of contracts to consistently poor performers. 18 6. Conclusions In this policy paper, we have analyzed four aspects of local governance that impact upon the extent to which the state`s extension into local communities for development purposes is effective, relevant, and sustainable. The core of our argument is that successful local development in a decentralized Timor-Leste will feature community-led efforts to identify, design, and implement local development priorities, supported by a well-defined and prominent role for community representatives in municipal government. With this in mind, we have made an effort to provide actionable recommendations on ways in which the state can support communities to engage actively in local development. Many of these recommendations relate to the need for improved channels of communication between the citizens and the state (often, in recognition of the role played by traditional authority structures, through community leaders). Others discuss how development planning under decentralized authorities might integrate the voices of community leaders, media, and local organizations. We have also discussed the importance of grievance-prevention and resolution mechanisms--a need that has been recognized by the MSATM itself. We hope that these and other recommendations are useful to the ministry staff as they plan for local development under decentralization. Beyond the specific recommendations, however, we believe it is important to acknowledge a broader point about state-society engagement for local development planning and beyond: relationships between levels of government, as well as between citizens, communities, and state institutions, will continue to change and be changed by political processes and the imposition of new rules and practices.23 While the experience of other nations is informative, no best model for local development under decentralization in Timor-Leste exists. And while questions of the form and function of decentralized institutions of local development are of course pertinent, it is essential to remain cognizant of the political and social realities that shape power and decision making, whether through formal rules or through custom and practice. Thus, as formal institutions of governance in the new state expand their reach through the current decentralization agenda, many important questions on citizen expectations of the state, the form of state-nonstate engagement, and sources of authority and accountability at the local level will need attention. How these questions are answered will influence the success and legitimacy of the government`s state-building--and more specifically, local development-- efforts at the local level. We have argued for an interpretation of governance in Timor-Leste that has a recognition of sociopolitical transformation at its core. Our recommendations are directed towards helping the state understand the patterns of these transformations, and thereby utilize and adapt those aspects of local governance that can positively serve the goals of decentralized local development. It is well recognized in political science and anthropological literature that customary authorities in Timor-Leste continue to serve as the first, most relevant, and often only form of government in rural communities (see, for example, Brown 2009; Fox and Soares 2003; Hicks 2007; Nixon 2006). For better or worse, the state`s interaction with citizens in rural areas is primarily mediated through local institutions. Engaging with local and customary governance institutions and values will thus be a priority for the GoTL as it seeks to establish a stronger local 23 For more on the dynamic nature of institution building, please see Caroline Sage, Nicholas Menzies, and Michael Woolcock, Taking the Rules of the Game Seriously: Mainstreaming Justice in Development. The World Bank`s Justice for the Poor Program, Justice and Development Working Paper Series (Washington, DC: World Bank, 2009), http://go.worldbank.org/S0NWFV3LL0. Accessed July 2010. 19 development presence within the districts. Without transgressing democratic political values-- and, indeed, to also protect such values--many aspects of local governance can be combined with state institutions. And then, at the very least, government and donor programs that seek to understand local perspectives can more effectively predict how policies will be implemented in practice, and thus provide equitable, relevant, and workable services by incorporating local perspectives from the outset. 20 Bibliography Athmer, Bas. 2009."TIM Works Project, Investment Budget Execution Support for Rural Infrastructure Development and Employment Generation, Independent Mid-Term Evaluation (Timing of Evaluation Mission in Timor-Leste: 20 September ­ 10 October 2009). Final Report, ILO, Dili. Babo-Soares, Dionisio. 2004. Nahe Biti: The Philosophy and Process of Grassroots Reconciliation (and Justice) in East Timor. The Asia-Pacific Journal of Anthropology 5 (1): 15­33. Brown, M. Anne. 2009. Security, Development and the Nation-Building Agenda - East Timor. Conflict, Security and Development 9 (2): 141­164. Butterworth, David, and Pamela Dale. Forthcoming. Redefining Local Governance through Development Initiatives: Design Contributions for Community Development Programs in Timor-Leste. World Bank, Washington, DC. Campos, Jose Edgardo, and Joel S. Hellman. 2005. Governance Gone Local: Does Decentralization Improve Accountability? In East Asia Decentralizes: Making Local Government Work, 237­252. Washington, DC: World Bank, http://siteresources.worldbank.org/INTEAPDECEN/Resources/Chapter-11.pdf. Accessed July 2010. Democratic Republic of Timor-Leste, Ministry of State Administration and Territorial Management. 2008. Policy Orientation Guidelines for Decentralization and Local Governance in Timor-Leste. Dili, http://www.estatal.gov.tl/Documents/Ministry/2008Policy/LG%20Policy%20and%20DS F%20-%20English.pdf. Accessed July 2010. Evers, Pieter. 2000. Resourceful Villagers, Powerless Communities: Rural Village Government in Indonesia. Local Level Institutions Study Working Paper, World Bank, Jakarta. Fox, James J. 2009. Repaying the Debt to Mau Kiak: Reflections on Timor`s Cultural Traditions and the Obligations of Citizenship in an Independent East Timor. In Democratic Governance in Timor-Leste: Reconciling the Local and the National, ed. David Mearns. Darwin: CDU Press. ------. 2003. Tracing the Path, Recounting the Past: Historical Perspectives on Timor. In Out of the Ashes: Destruction and Reconstruction of East Timor, ed. James Fox and Babo Soares. Canberra: Australian National University E-Press, http://epress.anu.edu.au/oota/c1.pdf. Accessed July 2010. ------, ed. 1993. Inside Austronesian Houses: Perspectives on Domestic Design for Living. Canberra: Australian National University E-Press, http://epress.anu.edu.au/austronesians/inside/html/frames.php. Accessed July 2010. ------. 1989. Category and Complement: Binary Ideologies and the Organization of Dualism in Eastern Indonesia. In The Attraction of Opposites: Thought and Society in a Dualistic 21 Mode, ed. David Maybury-Lewis and Uri Almagor, 33­56. Ann Arbor: University of Michigan Press. Fox, James J., and Dionisio Babo Soares, eds. 2003. Out of the Ashes: Destruction and Reconstruction of East Timor. Canberra: Australian National University E-Press. Grenfell, Laura. 2009. Promoting the Rule of Law in Timor-Leste. Conflict, Security and Development 9 (2): 213­238. Guggenheim, Scott, Tatag Wiranto, Yogana Prasta, and Susan Wong. 2004. Indonesia`s Kecamatan Development Program: A Large-Scale Use of Community Development to Reduce Poverty. Paper presented at Scaling Up Poverty Reduction: A Global Learning Process and Conference, Shanghai, May 25­27, http://info.worldbank.org/etools/docs/reducingpoverty/case/88/fullcase/Indonesia%20K DP%20Full%20Case.pdf. Accessed July 2010. Hicks, David. 2007. Community and the Nation State in East Timor: A View from the Periphery. Anthropology Today 23 (1): 13­16, http://www.cultura.gov.tl/sites/default/files/DHicks_Community_and_nation_state_2007 .pdf. Accessed July 2010. Hohe, Tanja. 2002. The Clash of Paradigms: International Administration and Local Political Legitimacy in East Timor. Contemporary Southeast Asia 24 (3): 569­589, http://info.worldbank.org/etools/docs/library/109847/SD_Communication/epublish/zip_f iles/tslg/pdf/clash.pdf. Accessed July 2010. Lutz, George, and Wolf Linder. 2004. Traditional Structures in Local Governance for Local Development. Switzerland: University of Berne. McWilliam, Andrew. 2009. Customary Governance in Timor-Leste. In Democratic Governance in Timor-Leste: Reconciling the Local and the National, ed. David Mearns. Darwin: CDU Press. ------. 2005. Houses of Resistance: Structuring Sociality in the New Nation. Anthropological Forum 15 (1): 27­44. Mearns, David, ed. 2009. Democratic Governance in Timor-Leste: Reconciling the Local and the National. Darwin: CDU Press. Meinzen-Dick, Ruth S., and Rajendra Pradhan. 2002. Legal Pluralism and Dynamic Property Rights. CAPRi Working Paper No. 22, International Food Policy Research Institute, Washington, DC, http://www.capri.cgiar.org/pdf/CAPRIWP22.pdf. Accessed July 2010. Meitzner-Yoder, Laura S. 2007. Hybridising Justice: State-Customary Interactions over Forest Crime and Punishment in Oecusse, East Timor. The Asia Pacific Journal of Anthropology 8 (1): 43­57. Molnar, Andrea. 2006. Died in the Service of Portugal`: Legitimacy of Authority and Dynamics of Group Identity among the Atsabe Kemak in East Timor. Journal of Southeast Asian Studies 37 (2): 335­355. 22 Nixon, Rod. 2006. The Crisis of Governance in New Subsistence States. Journal of Contemporary Asia 36 (1): 75­101. Palmer, Lisa, and Demetrio do Amaral de Carvalho. 2008. Nation Building and Resource Management: The Politics of Nature` in Timor-Leste. Geoforum 39 (3): 1321­1332. Peterson, George E., and Elisa Muzzini. 2005. Decentralizing Basic Infrastructure Services. In East Asia Decentralizes: Making Local Government Work, 209­236. Washington, DC: World Bank, http://siteresources.worldbank.org/INTEAPDECEN/Resources/Chapter- 10.pdf. Accessed July 2010. Siegle, Joseph, and Patrick O`Mahony. 2006. Assessing the Merits of Decentralization as a Conflict Mitigation Strategy. U.S. Agency for International Development, Washington, DC, http://www.dai.com/pdf/Decentralization_as_a_Conflict_Mitigation_Strategy.pdf. Accessed July 2010. United Cities and Local Governments (UCLG). 2009. Decentralization and Local Democracy in the World: First Global Report 2008. Barcelona: UCLG, http://www.cities- localgovernments.org/gold/Upload/gold_report/01_introduction_en.pdf. Accessed July 2010. Vischer, Michael, ed. 2009. Precedence: Social differentiation in the Austronesian World. Canberra: Australian National University E-Press. World Bank. 2006. Community-Driven Development in the Context of Conflict-Affected Countries: Challenges and Opportunities. World Bank Social Development Report 36425, World Bank, Washington, DC, http://go.worldbank.org/MF7S18B840. Accessed July 2010. ------. 2005. East Asia Decentralizes: Making Local Government Work. Washington, DC: World Bank, http://go.worldbank.org/6K9TTFE4G0. Accessed July 2010. ------. 2004. World Development Report 2004: Making Services Work for Poor People. Washington, DC: World Bank, http://go.worldbank.org/PLF4W7YI81. Accessed July 2010. 23 Appendix1: Justice for the Poor Timor-Leste "State-Building at the Local Level" Project The World Bank`s Justice for the Poor (J4P) program supports intensive analytical and programmatic work in countries where engaging with legal pluralism presents a central development challenge to equitable development. In particular, it aims to help poor and marginalized groups understand and enforce their socioeconomic rights. Justice for the Poor Timor-Leste was launched in July 2008 with the generous support of AusAID, the Australian government`s overseas aid program. The program is oriented around the theme Expanding Citizenship in an Expanding State, with an initial focus on the areas of (i) state-building at the local level and (ii) customary systems of land management and rural development. This discussion note presents findings from the first subproject under the state-building at the local level branch of J4P work in Timor-Leste. Under this area, the program is examining issues of how the state projects itself at the local level, and how citizens understand, access, and claim rights and entitlements; societal expectations of the state; the means of citizen-state and state-nonstate engagement; and how the government`s current methods of expanding state presence impact the state`s legitimacy. Initial research studied the Ministry of State Administration and Territorial Management (MSATM)/UN Capital Development Fund (UNCDF) Local Development Programme grants, and to a much smaller extent, the Secretary of State for Vocational Training and Employment/International Labour Organization TIM Works labor-intensive works program and the World Bank-supported Youth Development Program. Research activities, which took place in Aileu and Lautem districts from June to September 2009, examined how selection, distribution, and dispute-resolution mechanisms within these programs functioned both on paper and in practice, and how local power dynamics and ideas of governance impacted on those outcomes. In total, the research team conducted interviews with 151 respondents, capturing the perspectives of civil servants, program teams, community leaders, contractors, and beneficiaries in 10 project sites. Through analysis of these interviews, the team hopes to inform both the development and implementation of World Bank-supported grant programs and the Timor-Leste government`s decentralization program. For more on Justice for the Poor, please go to www.worldbank.org/justiceforthepoor. 24 Appendix 2: Study methodology Summary To pursue answers to questions related to the practice of local development under a decentralized government, the World Bank Justice for the Poor Timor-Leste (J4P) program examined local-level decision making and community participation in governance in Lautem and Aileu districts. The field research focused on three development and public service delivery programs: (i) the Local Development Programme (LDP) (described in further detail in Annex 3, this program was the primary focus); (ii) the World Bank-supported Youth Development Program, which uses many LDP mechanisms but is primarily focused on youth-supported development priorities; and (iii) the TIM Works cash-for-work program. Each of these programs seeks to improve livelihoods through small, participatory infrastructure projects, though the mechanisms are different.24 LDP in particular emphasized participatory development and has instituted democratic decision-making bodies (assemblies) at sub-district and district levels,25 which authorize suco representatives to help plan, choose, and implement subprojects (a practice that will continue through municipal block grants under decentralization). Primary research into these programs in practice, as well as a review of anthropological literature, has led us to a number of conclusions and recommendations regarding articulations of local governance that have implications for decentralization. Data Sources This research concerns questions of process--that is, a focus on how and why certain pathways are followed and outcomes occur during program implementation, and in particular, the influence of customary values on the choice of pathways. There is no hard and fast rule for determining the extent of the influence of customary values and historical events on the decision making of particular individuals. However, while ultimately each actor`s decision making is individual and subjective, it occurs within a cultural and historical context that can be analyzed to understand the constraints and potentials of subjective actions. Qualitative research, including qualitative research that underpins quantitative measurement (such as survey questions that inform statistical analysis), taps into the subjective reality of individuals by seeking to understand their context of action. In this way, the qualitative research methodology used for this research is directed towards understanding the subjective reality of individuals by seeking to comprehend their context of action. To achieve this end, field research was conducted for a combined total of eight weeks using in situ (that is, respondent houses, villages, project sites), free discussion, and semi-structured interview techniques with over 150 respondents. Qualitative research was supplemented by an analysis of existing program data (including past monitoring and evaluation reports) and a review of relevant literature on topics such as local-level decision making, state formation, and community-driven development. Past surveys, such as the extended Timor-Leste Survey of Living Standards (TLSLSx)26 and the GRM/Asia Foundation survey on 24 For more information on LDP, please visit: www.estatal.gov.tl. Information on the TIM Works program can be accessed on: http://www.ilo.org/asia/info/lang--en/WCMS_100842/index.htm. For short descriptions of these projects, please see Annex 3. 25 Under the original LDP model, implemented in the districts studied by J4P, assemblies were active at both the district and sub-district levels. In keeping with a revised model for decentralization, under which functions will be consolidated in municipalities, new LDP districts have implemented assemblies solely at the district level. 26 The TLSLSx was implemented in the first half of 2008 as an extension of the GoTL`s 2007 living standards survey. The extension added supplemental questions in the areas of (i) shocks and vulnerability; (ii) access to financial services; (iii) agricultural production; and (iv) access to justice. A series of reports on the findings of the TLSLSx justice module were published in late July of 2010, and are available on the Justice for the Poor website (www.worldbank.org/justiceforthepoor). 25 Citizen Perceptions of Law and Justice,27 also proved useful in providing background on access to information, trust, and decision making in Timor-Leste. Research design To develop a comprehensive view of program implementation, the J4P research teams targeted a broad range of respondents, including program staff, district- and sub-district-level government officials, local leaders, contractors, beneficiaries, and nonbeneficiaries. In total, the team conducted 119 interviews as broken down in the table below.28 Position/Role Number of Respondents Number of Interviews Gender Community Leaders 44 41 M: 86% F: 14% Project Beneficiaries 62 39 M: 71% F: 29% Public Servants 31 30 M: 100% Contractors 12 9 M: 75% F: 25% Total 151 119 M: 82% F: 18% Research activities took place in three stages, beginning with a desk review of the two programs to examine their information sharing, beneficiary selection, project implementation, and dispute resolution procedures as specified on paper. The second round, which took place in late June and early July 2009, consisted of initial interviews with respondents in two districts to determine how the programs in question were operating in practice. During the final round of field research in August and September 2009, the team followed up on the interviews conducted in the second round, developing a series of case studies illustrating the progress of specific LDP cases in each district. Throughout, and particularly at the latter stages of field research, the research team has engaged in dialogue with the program teams to discuss findings, clarify contradictions, and share suggestions for addressing issues that have arisen in the programs. Site selection Research Took place in Dili and a total of ten sucos in two districts, as indicated by the table below: District Sub-district Sucos Aileu Aileu Vila Fatubossa Lahae Liurai Remexio Acumau Maumeta Lautem Lautem Ililai Serelau Los Palos Fuiloro Home Muapitine 27 Silas Everett, Law and Justice in Timor-Leste: A Survey of Citizen Awareness and Attitudes Regarding Law and Justice 2008 (Dili, Timor-Leste: the Asia Foundation, 2009), http://www.asiafoundation.org/resources/pdfs/2008LawJusticeSurvey.pdf. Accessed July 2010. 28 The variations in the number of respondents and number of interviews conducted stem from the fact that some respondents were interviewed multiple times, while others were interviewed in a group. 26 District sites were selected based on several criteria, including the desire for geographic diversity and inclusion of both high- and low-capacity districts,29 the presence of an office of the research partner institution, and planned implementation of the Youth Development Program within that site. Within districts, the sub-district and suco research sites were chosen based on the presence of active or recently completed LDP and TIM Works projects. Methodological note on using qualitative tools It should be noted that at the time of research, both of these districts were implementing the original LDP model (with assemblies at both district and sub-district level). Some readers thus might question whether the experiences (both positive and negative) of LDP in these districts differ from sites implementing the new LDP model. Also, given the small number of research sites relative to the size of the LDP program, there are likely questions about whether the findings in this study can apply more broadly than the studied sucos. This question of generalizability is often raised with qualitative research projects. While the research team recognizes these concerns, it should be noted that this research was not intended as a program evaluation, but rather as a piece that allows researchers to discover and understand individual and community experiences with local development. These experiences are necessarily embedded in the sociohistorical and cultural context in which respondents live and interact, and we have thus undertaken qualitative research with the understanding that communities` experiences with local development will depend at least as much on these contexts as on the form of assembly used for project selection in the districts under study. The team has and will continue to discuss research findings with program teams in Dili and other districts in order to test the validity of the findings. To date, however, respondents have noted that the challenges and opportunities identified in the research are found not only in the sucos studied, but also more broadly in the context of development in Timor-Leste. 29 Based on interviews with program staff at headquarters level. 27 Appendix 3: Short description of programs studied Local Development Programme (LDP) Begun in 2005 with the support of UNCDF and the United Nations Development Programme (UNDP), the Local Development Programme (LDP)is intended to (i) reduce poverty through local-level infrastructure and service delivery; and (ii) inform national policy on decentralization by piloting procedures for local-level planning, financing, and implementation of public goods and services. Working with the MSATM, the program provides annual community-development block grants to support financing of small-scale public goods. Block grants are allocated based on population size, with a current allotment of $3.50 per suco resident. In 2010, the government has expanded LDP to each of Timor-Leste`s 13 districts and allocated $2.35 million to fund the program. The community-development model implemented under the program promotes participatory development planning and encourages increased transparency and accountability between civil servants, local leaders, and community members in the use of development funds. LDP has put in place local planning processes that allow for community members, in concert with local authorities, to design and identify suco priority projects. Subprojects from sucos are then submitted to district assemblies, where they are evaluated and cost-assessed, and then voted on by a panel of community leaders from the sucos (voting members). Likewise under decentralization, communities will create development plans, which will then be submitted to the municipal government for potential inclusion in the multiyear municipal development plans. Civil servants may also suggest sectoral projects for inclusion in the list of projects considered by the voting members of the assembly. The number of projects funded in a given year is determined by budget envelope and project cost; as of 2009, the average project cost is $9,000.30 Selected projects can be implemented by contractors or communities themselves. Project- monitoring responsibilities are shared by civil servants and community members, and financial management duties rest with civil servants at the sub-district and district level. Overall program management is performed by project staff based within the MSATM, who support district-based LDP staff. Youth Development Program (YDP) Like LDP, the Youth Development Program (YDP) is a participatory development program based within MSATM. The program, which began in early 2009 with the support of the World Bank`s Social Development unit, builds extensively on LDP processes and is part of a two- pronged approach to promoting youth empowerment and inclusion in development. The program confronts the lack of civic education and meaningful participation, as well as the general disconnect between youth and government, that contributed to the 2006 crisis. Under YDP, the ministry distributes small grants (based on an allocation of $1.80 per suco resident) to finance youth-identified priority projects. These projects, which can benefit either youth specifically or the wider community, are designed, voted on, managed, implemented, and evaluated by youth themselves. Suco youth representatives, who are elected members of the suco councils, are responsible for convening meetings with area youth to identify and plan community projects. The youth representatives and local youth work together with youth facilitators (YFs), who are recruited and trained by YDP to support implementation. YFs are charged with program socialization, 30 Source: Susanne Kuehn, UNCDF, in an e-mail dated April 30, 2010. 28 supporting youth consultation meetings, and (together with LDP technical staff) supporting youth implementation teams (YITs) to implement and monitor community projects. Female youth are encouraged to participate in all stages of the YDP process, and each suco is required to submit at least one female-focused project for consideration in each funding cycle. In order to promote transparency and accountability within YDP, YIT members must post information regarding funds received and expenses, hold public meetings to discuss implementation and spending, and work with suco councils to ensure that implementation is carried out as planned. YDP recently completed its first program cycle in Lautem and Bobonaro districts, and is expected to commence in Aileu, and Manufahi in 2010. Over the life of the program, each participant district is expected to receive a minimum of two grant cycles. TIM Works The TIM Works program, which grew out of and is linked to previous cash-for-work efforts in Timor-Leste, was designed to combat the duel challenges of job creation and infrastructure improvements across the nation. The program aims to support the sustainable rehabilitation and maintenance of rural roads and planned infrastructure using labor-based technology. In addition, it seeks to build the capacity of national and local government to plan, build, and maintain rural infrastructure, while simultaneously improving the capacity of local contractors and community groups. TIM Works, which is implemented in partnership between the International Labour Organization (ILO) and the Secretary of State for Vocational Training and Employment (SEFOPE), began in September 2008 with a budget of $8.14 million. Road rehabilitation activities under the program are being implemented in eight districts (Aileu, Baucau, Dili, Lautem, Liquica, Manatuto, Oecusse, and Viqueque), while maintenance activities are nationwide. The program uses primarily unskilled laborers (with targets of 50 percent youth and 30 percent women) from within project sucos, who are compensated at a rate of $2 per day based on their outputs. On average, workers are employed for 35 days. As of the end of 2009, TIM Works teams had completed 70 kilometers of rural road rehabilitation and 684 kilometers of maintenance, resulting in over 400,000 work days (Athmer 2009). 29