Abstract

Query: Please synthesise information and lessons learned from the Indonesian National Program for Community Empowerment (PNPM) and where relevant other community-driven development (CDD) programmes in Indonesia. Specifically, please highlight available information on: 1. To what extent have CDD programmes in Indonesia contributed to women's equality and empowerment? What are the challenges to promoting gender equality in CDD processes? 2. To what extent have civil society organisations (CSOs) supported communities to get value from CDD programmes? What are the strengths and weaknesses of CSOs in CDD programmes? 3. What are the most effective types of monitoring and evaluation activities that can inform programme improvement in CDD? Enquirer: East Timor Section AusAID Author: Huma Haider (huma@gsdrc.org) Contents 1. Overview 2. General lessons learned 3. Women's equality and empowerment 4. Civil society organisations 5. Monitoring and evaluation 6. References Overview Programme overview The National Program for Community Empowerment Mandiri (PNPM Mandiri) was launched by the Government of Indonesia in 2007. It is a national programme within the overall policy framework established to implement programmes aimed at poverty alleviation, strengthening local governance and service delivery, such as PNPM Rural which began in 1998 as Kecamatan Development Program (KDP) and PNPM Urban which began in 1999 as the Urban Poverty Program (UPP). PNPM adopts a community-driven development (CDD) approach, providing direct block grants to local 2 communities at the sub-district level to finance an open menu of local development priorities (typically small-scale social/economic infrastructure, education and health activities, and micro-loans to women's savings groups). The aims are to improve the economic and social welfare of the poor and to empower communities in decision-making and the management of development activities. 1 A review of evaluations of PNPM programmes highlights that PNPM Rural is contributing to improvements in poverty reduction, household welfare, access to health services resulting from better roads and bridges, and long-term employment. The impacts have been stronger in poor and remote areas where large infrastructure gaps are the primary factor limiting movement out of poverty. In such cases, the interests of the community and the needs of the poor are aligned (PNPM Support Facility, 2011). The PNPM also carries out various 'strengthening programmes'. These are pilot programmes, implemented in selected areas in Indonesia with different focuses. They include the following.  PNPM Generasi: This programme aims to address certain lagging human development outcomes. It adapts the PNPM Rural programme delivery model to achieve improvements in 12 areas of health and education. 2 The block grant mechanism differs from PNPM Rural though in that the government adopts a performance incentive approach. The size of the village's block grant for the subsequent year is based on the village's performance on each of the targeted indicators. An evaluation of the Generasi programme found that it had a positive on average across all 12 indicators it was designed to address The pilot phase of this new programme took place from July 2011 to June 2012. Unlike other PNPM programmes, Peduli is driven by civil society organisations (CSOs) and community groups in Indonesia. It targets those who do not ordinarily benefit from existing poverty reduction programmes and aims to improve their access to public services, justice and economic opportunities. Grants support not only these activities but also the strengthening of CSO organisational capacities. 3 General lessons learned There are general lessons learned that have emerged from the implementation of PNPM.  Target areas: Programmes have had a greater effect in areas that are poor, remote and with limited infrastructure (PNPM Rural); and in areas with low health and education indicators (Generasi).  Target sectors: There is debate over whether PNPM Rural should focus on community infrastructure, considered a key strength (and popular with the open menu approach), or whether it should try to promote agricultural sector development that may have a greater benefit to the lives of the poor.  Target populations: While PNPM has been considered beneficial to communities, there are many cases where it did not directly cater to the needs of the poor -due in large part to the open menu 1 http://pnpm-support.org/about-pnpm 2 http://pnpm-support.org/pnpm-generasi 3 http://pnpm-support.org/pnpm-peduli; http://psflibrary.org/catalog/repository/REDESIGN%20BROCURE%20ENGLISH.pdf 3 programme, which resulted in community choices that did not match with the needs of the poor. Marginalised groups have limited participation in decision-making processes which remain dominated by elites and interest groups. Programme design could be adjusted to better reward/require participation of marginalised groups. Separate projects could also be implemented to strengthen the capacity of marginalised groups to participate.  Social accountability: In order to successfully achieve social accountability, the following elements are necessary: strong information sharing and transparency; good quality participation, including women's participation; and well-functioning mechanisms for ordinary people to be able to hold the programme to account, including complaints handling. Despite gains in social accountability achieved by PNPM, there are still weaknesses in all of these areas.  Village level governance: There has been little spillover of PNPM core values into planning and implementation of other village and local government development activities. Existing power structures that favour traditional elites remain unaffected by the programme. In order to address this, it is necessary to identify constraints to spillover and to adapt programme design to address these constraints.  Facilitation: Strong facilitators are considered essential to the governance and success of PNPM. However, facilitators are increasingly in short supply and overworked due to pressures of scale-up and administrative demands. Facilitators are often poorly trained and lacking in support. Facilitators need to be better equipped with the knowledge and skills to promote participation, transparency and accountability.  Performance incentives: The performance incentive model adopted in the Generasi programme was effective in the area of health but not education.  Scale-up: The scale-up of programs resulted in an increasingly mechanical implementation process, focused on administration, at the expense of effective facilitation that empowers communities and proper monitoring and evaluation. Scale-up has resulted in project monitoring being spread across too many areas and in an emphasis solely on quantitative monitoring.  Financial management: The legal requirement to disburse funds within particular time-frames has led to immense time pressure. Allegations of misuse may be overlooked and accountability steps rushed. In some cases, delays in disbursements instead have resulted in frustration among project actors and villagers. Delays can also result in is little time or capacity to deepen the participation process. In some cases, separate meetings for women have been cancelled. Women's equality and empowerment Women's participation PNPM has produced positive impacts on social accountability/ transparency within the programme. Compared to the pre-PNPM era, more women have attended decision-making processes at the village level. High rates of community participation in PNPM, including the participation of women, have not, however, translated into quality participation. There are many indications that their participation is merely instrumental, to fulfil the programme requirements. Their presence has not undermined the dominance of males, the village elite and activists in the decision-making process. Within PNPM, there are women-only forums, in order to encourage women to actively voice their ideas and opinions and participate in decision-making. Decision-making still tends to be limited to the elite and activists among the women. In addition, although women dominate discussion at the women-only forums, men can still influence the decision-making process. In some cases, women's suggestions are re-negotiated at the village level. 4 Akatiga (2010) argues that the participation of women and marginalised groups can be improved when there is special facilitation support that focuses on these groups and when pro-poor village institutions are available. Such support can help marginalised groups that lack resources, access to information and confidence to organise, develop skills and to represent themselves. Women's access to credit and income generation PNPM-Rural provides loans only for women (SPP). Civil society organisations There is minimal discussion of the role of civil society organisations (CSOs) in the literature on PNPM and CDD programming in Indonesia, more generally. The only place where CSOs receive attention is in relation to the PNPM Peduli, the new strengthening programme piloted in 2012. PNPM Peduli partners directly with Indonesian CSOs through a grant mechanism to target the marginalised. In a recent workshop conducted on PFPM Peduli, the programme was described by CSOs as a 'breakthrough' in collaboration between government and CSOs, improving levels of trust and cooperation. The engagement of local government, CSOs, the private sector and other development partner programmes is considered to have strengthened implementation. CSO capacity building is an integral part of the Peduli. It was reported that training and support provided to CSOs has improved their skills, knowledge and understanding of working with marginalised groups. In addition, some CSOs are demonstrating better programme management. Further, the involvement of executing organisations and CSOs, and the development of bottom-up structures have provided CSOs with a sense of programme ownership A key challenge of Peduli is targeting. The programme is not always working with the most marginalised. When it was originally set up, there was often no community mapping to identify the most marginalised. Instead, executing organisations partnered with CSOs that were already working with particular groups. To identify the most marginalised, CSOs will need to identify who is not being reached and where they are located, develop strategies to reach them and design programmes that target specific needs that are not currently being met. Monitoring and evaluation There is very limited discussion in the literature on M&E activities in PNPM. The following challenges and recommendations have been identified.  The need for substantive monitoring and incorporation of feedback. The scale-up of PNPM Rural has resulted in a mechanical form of monitoring and evaluation focused on administration and based solely on quantitative assessment. Akatiga (2010) advises that for large programmes, it would be better to focus on certain key issues deemed essential to the programme, such as quality of participation. Alongside this kind of substantive monitoring, there needs to be 5 consideration of how to address specific aspects of marginalisation and how the participation process can be deepened.  The use of independent monitoring groups. Akatiga (2010) advises that regular monitoring by the Government and donors should be complemented by an independent monitoring group (such as NGOs), particularly to provide a more ongoing, qualitative evaluation of the programme process.  Regular review of the appropriateness of targets. Target indicators should be relevant to communities, while also reflecting the development priorities of the government. A lesson from PNPM Generasi is that the programme should regularly review the appropriateness of its selected target indicators and assess if existing ones should be replaced and/or new indicators added.  The possibility of follow-up evaluation. A follow-up evaluation can help to determine whether the impacts of the programme are indeed sustainable over time and if further lessons can be learned.  Burdensome reporting and information challenges. A common finding in much of the literature is that reporting requirements for PNPM programmes are burdensome. Not only does this detract facilitators from engaging in their core community work, but also leaves little time to verify information. KDP adopted a broad mix of approaches to monitoring and evaluation. Community participatory monitoring is a feature of KDP that has been highlighted in the literature as a good example of monitoring. Such monitoring was facilitated by CSOs. Organisations provided oral and written reports at monthly coordination meetings that included descriptions of problems encountered and recommendations for corrective action. Although there were some critiques that reports were sometimes late or unclear in the presentation of results, the majority of provinces incorporated the NGO reports and recommendations into their monthly meetings and action plans. General lessons learned The following are general lessons learned from PNPM and recommendations cited in the literature. Target areas PNPM Rural was considered most effective in poor, remote and infrastructure deficient contexts, where infrastructure gaps contributed significantly to poverty. In such cases, the needs of the general community and the poor were aligned, and infrastructure projects resulted in higher returns and greater poverty/welfare effects (PNPM Support Facility, 2011). Similarly, PNPM Generasi had the greatest impact in areas with low health and education indicators. Target sectors Akatiga (2010) argues that PNPM Rural should focus on community infrastructure which is its key strength, having successfully implemented infrastructure projects in a cost-effective manner. Adopting a single focus would help ease the task of programme facilitation. While recognising that infrastructure does not yet proportionally benefit marginalised groups, the organisation finds that such public goods can still be benefit them. A review of PNPM evaluations states that the programme should continue delivering infrastructure, particularly in areas with large infrastructure gaps, where such projects are most effective (PNPM Support Facility, 2011). Syukri et al. Target populations A qualitative study of PNPM Rural found that while the programme was considered beneficial to the community, there were many cases where it did not directly cater to the needs of the poor. This was due largely to the open menu programme, which resulted primarily in decisions to fund general infrastructure. The main needs of the poor, however (in areas that were not the most poor, remote and infrastructure deficient), were job opportunities, additional capital and skills training Evidence from PNPM Rural indicates further that marginalised groups 4 have yet to share in the benefits of the programme in comparison with other groups, including the poor in general. These groups have limited participation in decision-making processes which remain dominated by elites and interest groups (PNPM Support Facility, 2011). In order to improve the reach of PNPM to poor people and marginalised groups, programme design could be adjusted to better reward/require participation of marginalised groups. This requires a stronger role of facilitators, particularly during the stages of formulating needs and programme implementation. Separate projects could also be implemented to strengthen the capacity of marginalised groups to participate in decision-making activities, including organising capacity, access to information and negotiation skills (PNPM Support facility, 2011; Social accountability Social accountability is a crucial component of PNPM. Woodhouse (2012) defines the heart of social accountability as empowerment and civic engagement. In order to successfully achieve social accountability, the following elements are necessary: strong information sharing and transparency; good quality participation, including women's participation; and well-functioning mechanisms for ordinary people to be able to hold the program to account, including complaints handing. In addition, community members must have a full understanding of their rights and responsibilities in relation to PNPM. Facilitators are very important to this process and must provide excellent training at the program 'socialisation' stage (Woodhouse, 2012). The programme has contributed to gains in social accountability. Participation rates are high and community members have good knowledge of PNPM and its processes. However, the quality of participation is mixed. While community members attended meetings, this did not necessarily mean that they spoke, debated or participated in any other meaningful way. There are also weaknesses in information sharing, transparency and accountability. Government leaders have continued to keep planning processes internal and budget documents privileged (PNPM Support Facility, 2011). Community members tend to know little about oversight structures, project finances or where and how 4 Akatiga (2010: 3) defines marginalised groups as 'having no (valuable) assets, living in outlying areas with limited basic infrastructure, having limited income with a large number of dependents, and originating from an ethnic/religious minority'. 7 to complain about misuse of funds. There is also a lack of capacity in following up complaints. In addition, community notice boards are poorly maintained and their purpose misunderstood (Woodhouse, 2012; see also World Bank, n.d.). Village level governance There has been little spill-over of PNPM core values into planning and implementation of other village and local government development activities. Existing power structures that favour traditional elites remain unaffected by the programme. Although such structures can be accommodating and at times supportive of the PNPM system, they do not necessarily have the will or incentives to extend PNPM procedures to other development activities (PNPM Support Facility, 2011). An evaluation of the PNPM Rural programme finds that in all villages, except for one, participation and transparency applied in PNPM were seen as special features of that particular programme that did not have to be applied in other programmes (Syukri et al., 2011). A review of PNPM evaluations recommends that it is essential to identify constraints to spill-over and to adapt programme design to address these constraints (PNPM Support Facility, 2011). Woodhouse Facilitation Strong facilitators are considered essential to the governance and success of PNPM However, facilitators are increasingly in short supply with high change-over of staff, resulting in little continuity for the groups in development relationships (World Bank, n.d.). They are overworked due to pressures of scale-up, the proliferation of PNPM offshoot programmes and demands of reporting and other administrative tasks, which hinder their ability to perform the task of facilitation effectively. In addition, facilitators are often poorly trained and lacking in support. Training tends to focus on the details of project procedures and rules rather than on the concepts of empowerment and how to encourage and facilitate inclusive meaningful participation and accountability In order to sustain and improve positive outcomes of PNPM, the role of facilitators need to be strengthened with an emphasis on improving quality and quantity of community facilitation. Facilitators need to be better equipped with the knowledge and skills to promote participation, transparency and accountability. There is an initiative underway to create a certification organisation for facilitators to ensure proper training for and quality of facilitators. It will be necessary at a later point to evaluate whether the school actually increases the competency of the facilitators. It is also important to ensure that their workload is balanced and allows for adequate time to empower the community (PNPM Support Facility, 2011; 8 Performance incentives The performance incentive model adopted in the Generasi programme was effective in the area of health but not education. Between 50 to 75 per cent of the total impact of the block grant programme on heath indicators can be attributed to the performance incentives (Olken et al., 2011: 4). There was little difference, however, in the performance of education indicators between incentivised and nonincentivised groups. Scale-up (facilitation and M&E issues) Akatiga (2010) finds that project design and institutions did not keep pace with the scale-up of PNPM. In order to carry out the scale-up, the implementation process became increasingly mechanical and focused on administration, at the expense of effective facilitation that empowers communities and proper monitoring and evaluation. While PNPM Rural has improved some community member's capacity with project administration, in general facilitators still lack the capacity to promote the interests of marginalised groups. In some cases, it has no longer been possible to monitor the quality of participation, beyond quantitative assessment. Scale-up has also resulted in project monitoring being spread across too many areas. Akatiga (2010 recommends that it may be better to focus on certain issues considered particularly important to the programme (e.g. quality of participation). Financial management (women's empowerment issue) Woodhouse (2012) finds that problems with financial management have decreased the quality of local governance of PNPM. The legal requirement to disburse funds within particular time frames has led to immense time pressure and created a 'culture of disbursement' where the needs of disbursement trump empowerment goals. Allegations of misuse may be overlooked in the need for speed, and accountability steps may be rushed. This is exacerbated by time-consuming manual bookkeeping, with little transparency, and by the high number of cash transactions (Woodhouse, 2012). Akatiga (2010) highlights instead problems with delays in disbursement in PNPM Rural, which creates frustration among project actors and villagers. Such delays are due to insufficient institutional capacity which extends beyond the PNPM Rural programme. Delays can also be attributed to the view of PNPM Rural as a project by the implementing agency rather than a programme with long-term support. Delayed fund disbursement can also have negative effects on the goals of PNPM. Under such circumstances, there is little time or capacity to deepen the participation process. There are cases where two meetings were collapsed into one, with the separate meeting for women being cancelled, in order to make up for the delay. Women's equality and empowerment Review of gender in community-driven development projects in Indonesia prior to PNPM In the lead-up to the implementation of PNPM Manduri, Wong et al. 9 Women's economic empowerment The review finds that support to credit groups in the CDD programmes have generally had a weak impact on poverty reduction and the economic empowerment of women. The credit groups rarely included the poor unless this was a project rule. In addition, the types of enterprises that were financed were few and focused on women's traditional roles (cooking, sewing, kiosks) rather than opening new opportunities. The review lists some limited examples of good practice in the support to credit groups.  'Ensuring demand from women for support for economic activities and small savings and loans schemes is heard and considered in the decision-making;  Where possible, linking with other resources such as training and capacity building, or with existing cooperatives, credit unions, or other providers;  If credit is provided through the project the financial management systems, procedures, and training modules which have been developed over the years need to be further improved and adapted ' (Wong et al., 2007: 20). Women's political empowerment The review finds that while CDD projects increased the participation of women in these projects and the project's responsiveness to women, the impact outside the projects has been limited. Women's participation in project planning has not spilled over into separate village planning processes. In addition, election process bias against women's selection remains. Women also still lack the confidence and experience to compete against men. The review advises that selection processes are improved and that project and village planning processes be brought together in order to ensure spillover. There have been some attempts to do this. The following are examples of good practice in the CDD projects under review.  'Including quotas for women's participation in meetings;  Holding separate women's meetings at key stages in planning and decision-making process;  Including targets for women's participation in decision-making bodies;  Opening up a range of positions on implementing and monitoring committees at community level and encouraging women's selection for these committees so that they can demonstrate their skills and capabilities ' (Wong et al., 2007: 21). Women's empowerment -social Progress in empowering women economically and politically is constrained by norms and attitudes, in particular the attitudes of male leaders in the villages. It is thus important to include male leaders in gender training exercises. M&E systems that focus on quantity can also be problematic as they constrain the identification of gaps and disparities in the participation of women. In addition, women's involvement in community management tends to be encouraged on a volunteer basis, while paid positions go to men. The following are practices that have been helpful in improving the enabling environment for women to access economic and political opportunities.  'Women staff and facilitators are role models, especially for women, and can give women the confidence to follow in their footsteps and take on challenging positions.  Organizing meetings at times convenient for women, and encouraging them to bring their children, means that more women can attend.  Providing space in separate women-only meetings for women to discuss issues important to them without men around gives them confidence and helps to reach consensus before facing a mixed group. 10  Including specific activities in the socialization and planning that help both men and women analyze and discuss gender roles.  Requiring women's attendance at meetings, or participation in committees or as facilitators, through quotas and targets helps to normalize this and makes it easier for women to attend other meetings and participate in other activities ' (Wong et al., 2007: 22-23). Kacematan Development Program (KDP) As noted, KDP began in 1998 and has since evolved into PNPM Rural (from 2007). The design of KDP incorporated various tactics to improve the participation of women. Activities throughout the project cycle of KDP, from planning to implementation and monitoring all require the participation of women (often a minimum of 40 per cent). In addition, it is required that Village Facilitators comprise of one man and one woman. There are also separate planning meetings for women and one proposal from each village must come from a women's group. Further, there is a specific microfinance component that funds women-only credit groups (Voss, 2008). Voss (2008: 13) finds that although KDP has benefited poor households, disadvantaged groups (other than the poor), such as female-headed households and households where the head is lacking primary education, are less likely to benefit from the programme. This is largely due to continued difficulties in reaching various pockets of highly vulnerable groups. Often, members of marginalised groups are less likely to attend project meetings as they are not invited or the costs in time and lost income of meeting attendance is high. In addition, the practice of majority voting sometimes has the effect of discouraging smaller, often outlying groups from participating. They may also not see any benefit to projects that are centred in more populated parts of the village. The programme sought to address this weakness by creating a separate programme specifically targeted at female heads of households (PEKKA). A field guide developed based on the KDP outlines specific tactics to promote the participation of women in meetings. These include:  specifically invite the women by name, as opposed to sending an invitation to the head of the household (which will be deemed to be the male only);  make announcements at venues which women frequent, such as at mosques, or over the radio;  ensure that meeting times take into consideration women's routine activities;  seat women in the front rows and have the facilitator actively involve women in discussions;  develop specific criteria for women representatives attending socialisation meetings to ensure that the village head does not simply appoint the women representatives and that the women have the capabilities and relations in the community to disseminate KDP information (World Bank and Programme Pengembangan Kecamatan, 2003). PNPM The PNPM is considered a key instrument to improve gender equality.  'Respond to women's practical needs: by funding, for example, water supply, health and education facilities, which help to remove practical barriers of time and capacity that constrain women's involvement in development.  Increase potential for women's economic activity: by investment in local infrastructure such as roads and bridges which help to remove some of the obstacles to women's access to markets 11 and resources; and supporting microfinance activities which help women engage in incomegenerating activities and expand their businesses; and  Ensure women are active participants in planning and decision-making: through the emphasis on broad-based participation that helps to break down some of the barriers to women's participation in local planning and decision-making, and ensure that their voice is heard and that they can influence the processes and decision to be more responsive to their concerns' (Wong et al., 2007: 11-12). Women's participation A review of PNPM evaluations finds that PNPM is creating positive impacts on social accountability/ transparency within the programme (PNPM Support Facility, 2011). The PNPM results framework includes the requirement to track whether there has been a 'minimum 40 percent participation rate of women and poorest community members in planning and decision-making meetings' (Yaron et al, 2011, 10). Compared to the pre-PNPM era, more women have attended decision-making processes at the village level High rates of community participation in PNPM, including the participation of women, have not, however, translated into quality participation. There are many indications that their participation is merely instrumental, to fulfil the programme requirements. Their presence has not undermined the dominance of males and the village elite in the decision-making process. Women (in addition to many male community participants) are often passive listeners, abiding by decisions made by village elites and community leaders. There are occasions where women will voice their views but usually only to other women or to other participants, not to the village head. This is due in large part to the dominance of the village elite in local structures (kinship and patronage systems) and the persistent belief (particularly among poor women) that only men and village elite are in the position to speak as they have more leadership qualities and knowledge and skills in relation to development projects In many cases, women rely not only on village elite to represent them but also on activists (who possess knowledge of government projects) and formal institutions. These actors are also deemed to have greater knowledge and skills. Females have been represented by organisations, such as the Family Welfare and Empowerment (PKK) or Bundo Kanduang in West Sumatra. Within PNPM, there are women-only forums, in order to encourage women to actively voice their ideas and opinions and participate in decision-making. Decision-making still tends to be limited to the elite and activists among the women. It is important to note that in both women-only and mixed PNPM forums, it is likely that only women from the more prosperous families participate in the meetings. Some female informants from less prosperous families said that they were never invited, let alone participated in the meetings. In addition, although women dominate discussion at the women-only forums, men can still influence the decision-making process. In some cases, women's suggestions are re-negotiated at the village level, where the decision makers are primarily male, village elites Akatiga Women's access to credit and income generation PNPM-Rural provides loans only for women (SPP). However, poor women have generally had limited access to the loan programme. This is due in large part to the requirement by PNPM implementers that recipients already have their own business. This was to make sure that they were able to return the money as scheduled. Therefore, most of the recipients were from more prosperous group and poor households were deterred from applying for loans Facilitation, scale-up and financial management The section above discusses how problems with facilitation, scale-up and financial management can result in problems with the participation and empowerment of women in PNPM. Civil society organisations Generally, there is minimal discussion of the role of civil society organisations (CSOs) in the literature on PNPM and CDD programming in Indonesia. The only place where CSOs receive attention is in relation to the PNPM Peduli, the new strengthening programme piloted in 2012: 'while the general principles are the same, PNPM Peduli mechanisms are different to PNPM Mandiri. PNPM Peduli partners directly with Indonesian CSOs through a grant mechanism to target the marginalized not easily reached by other government and poverty reduction programs. Through this new partnership, the government aims to develop new knowledge on marginalization that can inform its poverty reduction programs' (PNPM Project Facility, 2012: 4). This mission of Peduli is:  'To respond to marginalized people as change agents.  To ensure that the PNPM Peduli approach responds to the characteristics of different target individuals and groups recognizing the complexity of issues these people face.  To adopt a holistic rights-based approach to working with marginalized groups in order to ensure sustainability.  To strengthen synergies and support from government, private sector and community groups.  To strengthen the capacity of CSOs in empowering marginalized groups' (PNPM Support facility, 2012: 9). Although the programme covers a broad range of areas, including agriculture and rural development, social empowerment, health, education, human rights and advocacy, in the pilot phase 90 per cent of 13 activities focused on livelihoods and two percent on rights. A number of CSOs involved in Peduli have a strong background in rights and in the pilot phase used opportunities and networks to focus on livelihoods activities with beneficiaries. The aim is to develop a more integrated approach to activities and to look at opportunities to share their expertise and learning across the network (PNPM Project Facility, 2012: 6). In a recent workshop conducted on PFPM Peduli, the programme was described by CSOs as a 'breakthrough' in collaboration between government and CSOs, working together on issues of marginalisation. It was felt that the programme would improve the levels of trust and cooperation. The engagement of local government, CSOs, the private sector and other development partner programmes is considered to have strengthened implementation. Executing organisations (national organisations) have noted that CSOs cannot operate in isolation and rely on partnerships, including those with government, to ensure sustainable impacts. It was also noted that partnerships with local government needs to be fostered early on in order to ensure commitment. CSO capacity building is an integral part of the Peduli and ensures sustainable benefits for marginalised groups. It was reported that training and support provided to CSOs has improved their skills, knowledge and understanding of working with marginalised groups. In addition, some CSOs are demonstrating better programme management (e.g. reporting, financial management and bookkeeping). Further, the involvement of executing organisations and CSOs and the development of bottom-up structures have provided CSOs with a sense of programme ownership (PNPM Support facility, 2012). The workshop report attributes these achievements to the following key factors.  'Trust: Trust has been built between the CSO/branch and the marginalized groups they are working with.  Synergies: Linkages have been developed with other local level programs or initiatives (so that projects are not happening in isolation).  Partnerships: Existing partnerships including those with local government have enhanced project activity results.  Community engagement: Marginalized groups are engaged and there is a positive spirit emerging. In some cases, for example in some of Lakpesdam/NU/s projects, people are volunteering time and resources ' (PNPM Project Facility, 2012: 3). The report also identifies some challenges faced by the programme.  Communication: Peduli positioning needs to be communicated better at the local level. Brochures and letters have been distributed to local CSO partners to share with their communities in the start-up phase. Letters have also been sent to local government to inform them of the programme and to encourage them to connect with PNPM Peduli CSO/branch partners. There are some examples where this is taking place; however, more needs to be done to highlight specific opportunities for government/CSO collaboration.  Targeting: Peduli is not always working with the most marginalised. This may be because the most marginalised consider it a risk to join such activities and prefer to observe the benefits prior to committing limited time and resources. In addition, when the programme was originally set up, there was often no community mapping to identify the most marginalised. Instead, executing organisations partnered with CSOs that were already working with particular groups. To identify the most marginalised, CSOs will need to identify who is not being reached and where they are 14 located, develop strategies to reach them and design programmes that target specific needs that are not currently being met.  Connection to Mandiri: there is some confusion at the local level, including from the local government, about how PNPM Peduli relates to PNPM Mandiri core and support programmes (PNPM Project Facility, 2012). * For discussion on the role of CSOs in monitoring and evaluation, see the following section. Monitoring and evaluation PNPM There is very limited discussion in the literature on monitoring and evaluation activities in PNPM. The following are the challenges and recommendations that have been identified. The need for substantive monitoring and incorporation of feedback As discussed in section 2 regarding 'scale-up', the scale-up of PNPM Rural has resulted in a mechanical form of M&E, focused on administration (i.e. whether all stages of the programme have been executed) and based solely on quantitative assessment. The quality of implementation has been neglected, such as quality of participation and whether the programme's aim to empower marginalised groups is being consistently applied during all stages of the programme. Akatiga The use of independent monitoring groups Akatiga (2010) advises that regular monitoring by the Government, World Bank and others should be complemented by an independent monitoring group, particularly to provide a more ongoing, qualitative evaluation of the programme process. PNPM-Rural, for example, has been using provincial non-government organisations to do this kind of work for a number of years, but the quality varies. The programme should review the work and select one or two of the best groups to work with in order to improve the quality of the monitoring Regular review of the appropriateness of targets Target indicators should be relevant to communities, while also reflecting the development priorities of the government. A lesson from PNPM Generasi is that the programme should regularly review the appropriateness of its selected target indicators and assess if existing ones should be replaced and/or new indicators added. Where a particular target has been reached (or is close to being reached), the programme may wish to add other targets to capture other priority areas that are lagging behind. It is important, however, not to overload the project with too many target indicators 15 The possibility of follow-up evaluation The Generasi programme engaged in a final round of evaluation after two and a half to three years of project evaluation. This three-year evaluation series was useful in providing empirical evidence to inform project implementation and learn lessons for the subsequent phase. Burdensome reporting and information challenges A common finding in much of the literature is that reporting requirements for PNPM programmes are burdensome. Routine reports tend to be extremely long and time-consuming and often involve unrealistic deadlines. Not only does this detract facilitators from engaging in their core community work, but also leaves little time to verify information and decreases the quality of information reported. A lack of standardisation of software applications across the country also makes it difficult to extract analysis or to compare information among regions (Woodhouse, 2012). KDP A 2003 report (Wong) outlines challenges faced when developing the M&E system for the KDP.  Ambitious objectives: while KDP's overall goal was poverty alleviation, for many stakeholders issues of process and democratisation (citizen involvement, inclusion, and open and transparent discourse) were of equal, or greater, importance. In order to capture this process, complementary qualitative and quantitative M&E instruments were necessary, as well as concerted efforts to systematically document the process.  Wide variety of activities: the open menu and multi-sectoral nature of KDP also meant that a wide variety of outputs (e.g. infrastructure, water, economic activities) had to be monitored and evaluated for quality and cost-effectiveness. A wide range of control systems for these various activities had to be designed and implemented.  Geographical scope and logistics: the sheer scale and geographical scope of KDP also made monitoring and reporting a challenge. KDP adopted a broad mix of approaches to monitoring and evaluation. It combined quantitative instruments (household surveys and regular monthly field reporting) with qualitative research (case studies). It supported participatory methods such as community participatory monitoring along with more highly specialised monitoring and evaluation activities, such as financial audits and sector studies. Lastly, KDP incorporated both internal and external monitoring from various civil society actors and other donors in order to provide a wide range of perspectives. Despite problems with consistency and quality of reporting, the information provided by the internal monitoring system was generally useful and timely and allowed for programme managers to take corrective action and adjust programme procedures. Sectoral evaluations gave constructive recommendations for the redesign of the economic loan portfolio and the cost effectiveness and use of infrastructure. Financial audits and supervision missions led to further financial training in certain areas, programme suspension in problematic areas and reformulation of design issues Case studies also provided a very useful form of qualitative investigation to determine the reasons why indicator targets may not have been met 16 Community participatory monitoring is a feature of KDP that has been highlighted in the literature as a good example of monitoring (see  Monitoring by village councils. Village representatives (BPDs) had responsibility for monitoring KDP activities at all stages: socialisation, planning, implementation and maintenance. The BPDs selected members to monitor each phase. Results of this monitoring were then discussed at council meetings or fed into larger village meetings.  Monitoring by special community groups or teams. KDP encouraged community monitoring groups in each village that were independent of the village implementation teams. The community team members shared responsibilities for overseeing village finances, checking financial accounts, monitoring bank transactions and material purchases or rentals, visiting suppliers to confirm the costs of goods and monitoring subproject activities, including infrastructure construction.  Community participatory monitoring facilitated by CSOs. The growth of civil society organisations in Indonesia post-Suharto allowed for partnerships with civil society groups to undertake independent monitoring of KDP. Many NGOs have extensive knowledge of local conditions and skills in community development. NGOs helped a number of villages, special groups, and teams conduct community participatory monitoring. They helped villagers decide what questions were important to them about KDP, how to collect data to answer those questions and helped villagers analyse community findings CSO facilitation was considered a success story in several locations. The KDP M&E Unit conducted training sessions for NGOs to provide an orientation toward KDP principles and procedures and their expected roles and responsibilities. NGOs were responsible for monitoring: the participation of villagers (particularly women and the poor); information sharing and transparency; financial reviews; adherence with the general principles and procedures of the programme; quality of technical assistance and programme outputs. The NGOs provided oral and written reports at monthly coordination meetings that included descriptions of problems encountered and recommendations for corrective action. Although there were some critiques that reports were sometimes late or unclear in the presentation of results, the majority of provinces incorporated the NGO reports and recommendations into their monthly meetings and action plans. The KDP M&E Unit also brought NGOs together in workshops to discuss experiences in the field, best practices and lessons learned References Akatiga. About Helpdesk research reports: This helpdesk report is based on 3 days of desk-based research. Helpdesk reports are designed to provide a brief overview of the key issues, and a summary of some of the best literature available. Experts are contacted during the course of the research, and those able to provide input within the short time-frame are acknowledged.