Document(s) 7 of 10

“Better use of better statistics leads to better policy and better development outcomes.”

— Christopher Scott, London School of Economics, January 2005

The previous chapter briefly described the implementation of a number of CBMS initiatives in Asia and Africa. The scope and maturity of these programs vary, as do their prospects for institutionalization and sustainability. Most are ongoing and more continue to be launched, further adding to the CBMS experience.

The growth of CBMS has demonstrated that the need for disaggregated, community-level information is acute in many parts of the world. And demand is growing, in part to meet local-level planning requirements in support of decentralization and to implement targeted poverty interventions, and in part to monitor progress toward meeting benchmarks in anti-poverty and other public investment programs (such as PRSPs). The need to measure progress toward meeting MDG commitments creates additional demand.

The overarching lesson from the research over the past decade is that designing and sustaining community-based poverty monitoring systems is an essential component of national poverty monitoring systems. The usual system based on national sample surveys does not provide all the information needed by policymakers. Our experience has shown that CBMS is feasible. With training, communities and local governments are able to implement the systems at an affordable cost.

Carried out at regular intervals, CBMS provides accurate disaggre-gated data that enables local government planners to effectively and efficiently target investments — with attention to gender and social equity considerations — and monitor progress. It also enables them to do so relatively easily in partnership with their constituencies. Furthermore, it gives communities a simple tool for holding governments accountable, thereby fostering good governance.

However, CBMS is not a panacea for statistical and planning challenges. It should be viewed as a complement, not a substitute, to national-level sample surveys. Nor should it be seen as a turnkey solution. As shown in Part 3, indicators, data-collection methodology, analysis, and validation must be adapted to local conditions and developed over time. Enabling conditions also need to exist for CBMS to be effectively institutionalized. These include decentralization, public participation, political will, and adequate human and financial resources.

The challenges of institutionalizing CBMS also need to be recognized. These include the difficulties of integration within well-established systems of national agencies and local governments, the accuracy of data, the sustainability of CBMS at the community level and resource constraints, and the management of tensions between different interest groups competing for limited public resources.

Lessons learned about the enabling conditions

Decentralization facilitates the adoption of CBMS

Local government autonomy in delivering development programs is a common denominator in all countries where CBMS is most successfully used. CBMS is intended to be implemented by local governments in partnership with communities to provide detailed data on all households in those communities. Decentralized planning, budgeting, resource allocation, and program management are important elements for CBMS to take root and develop organically as part of local public sector program planning. Officials — both national and local in most countries where CBMS has been tested — have made it clear that the system is highly suited for decentralized government functions and devolved fiscal and political authority.

CBMS can strengthen the capacity of local governments to meet the challenge of improving the lives of their constituents. As has been demonstrated in a number of countries, CBMS allows local governments to use easy-to-understand and appropriate information and analysis to formulate policies and programs that are responsive to people’s needs. This promotes poverty reduction and helps governments to realistically move forward to meeting the MDGs.

Political commitment is key to sustainability

To succeed, CBMS requires strong political commitment on the part of local government. To be sustainable, it also requires the commitment of higher levels of government, both to implement the system and to use the data generated.

The Philippines team found that a formal agreement (a Memorandum of Agreement) between government and partners, setting out objectives and outputs, scope of work, allocation of responsibilities, timetables, and resource requirements, was essential.

Commitment is expressed in other ways in other countries. In Cambodia, for instance, the Royal Government’s Statistics Law calls for a statistics/planning officer to be placed in every commune to work with communities and officials in collecting information. Plans are now underway for the National Institute of Statistics to establish a CBMS bureau within the ministry to coordinate CBMS activities country-wide.

Local governments committed to evidenced-based planning and budgeting create an ongoing demand for the timely, relevant data that CBMS can provide. A vibrant civil society also helps to implement CBMS and increases the use of the data. Research shows that choosing the right kind of partners is critical: a strong corps of CBMS researchers and local government officials willing to work hand-in-hand with local communities not only ensures that the right indicators are developed to monitor poverty, but also builds local capacity and empowers communities.

Public participation is important

Involving communities and other stakeholders in discussions and decisions on local public sector investments has proven to be important. Through the process of data validation and analysis, communities have worked closely with public planners, officials, and policymakers in helping to prioritize investments for poverty reduction. As important, the knowledge gained through CBMS has empowered local communities and fostered participation in the public policy process. It has also improved systems for local accountability by facilitating greater transparency in public investments, and stimulated demand by local communities for better governance and accountability.

CBMS is cost-effective

Like all initiatives that involve primary data collection, CBMS entails costs in collecting and processing data. And because it is a census instrument, scaling-up province- or nation-wide poses financial challenges for governments. But as Carmelita Ericta, administrator of the Philippine National Statistics Office, points out, governments should recognize that “Information gathering is not a cost but an investment. It has high payoffs in terms of better allocation of resources, better designed programs, and improved targeting of beneficiaries.”

Financing for CBMS should come largely from those who will benefit from the data generated, not external donors. IDRC’s contribution has been designed to support the pilots of CBMS initiatives, in partnership with others, and has been directed to providing the technical guidance and capacity building needed for sustainability. Since local governments are the primary users of data, they should provide the bulk of the resources for implementing the system. In the Philippines, the provincial, municipal/city, and barangay governments provide the funds to implement CBMS in their localities.

CBMS can also generate savings and attract additional funding. Local governments have learned that the system helps to make best use of scarce public resources by better targeting programs and reducing discretionary spending. For example, the municipality of Sta. Elena of Camarines Norte in the Philippines was planning to allocate 220 000 pesos (US$4 500) for a feeding program to address child malnutrition. The CBMS survey showed, however, that there were only 97 malnourished children. As a result, they needed to allocate less than 20% of the proposed budget to address this problem. This saving was possible only because those who really needed entitlements were identified through the survey.

In Vietnam, local governments and communities have recognized the value of information that CBMS provides and have invested scarce local finances and human resources into its implementation. In Ha Tay province, for example, officials pointed out that many people from the communes willingly volunteer their time to collect data, which they see as socially important.

While local financing is critical for CBMS sustainability, the survey results can attract additional funding from international donors to finance complementary, larger projects to meet identified needs. Donors tend to favour proposals when they can see that the projects respond directly to an actual problem and they can see who will benefit. This has certainly been the case in Cambodia.

Lessons learned about CBMS design and implementation

Partnerships between researchers, government officials, and communities are essential

The research team should include members with varied experience and expertise. As Bangladesh and the Philippines have shown, researchers need to be knowledgeable about information systems and have a keen interest in, understanding of, and patience for working with communities. Also important, the team should be familiar with the policy environment and understand the workings of administrative institutions and the local political establishment. Experience has shown that where there is a purely academic orientation, the application of data for public action is greatly constrained and the chances of institutionalizing the system are severely reduced.

To ensure that the CBMS can survive changes in local and national governments, the CBMS needs a stable partner at all levels. Notes Vietnam’s Vu Tuan Anh: “To institutionalize the CBMS, there should be a close partnership between researchers and governmental authorities who are in charge of poverty reduction, and non-governmental organizations in local areas.”

Enlisting and orienting the community determines success from the outset

As a report by the Senegal team notes: “The value of an information system is a given.… Rather it is more useful to reflect on whether or not the local community is ready to accommodate the monitoring system” (Sylla 2004).

The communities and other stakeholders should participate in CBMS planning from the outset. In all countries where CBMS has been tested, a series of meetings and workshops with local officials, NGOs, and civil society ensured that they understood the project, enlisted their support, and identified data needs. This local advocacy can take various forms. In Bangladesh, for instance, the team first carried out a thorough participatory rural appraisal to better understand the communities and to develop relationships with the villagers. The research team noted that this exercise considerably changed the attitudes of local government officials. Discussions also made residents more aware of their problems and the role service providers play in their locality. They were asked to report inefficient providers. “That way,” says the report, “the feedback mechanism gives the poor a voice.”

In Burkina Faso, the pilot test convinced the research team to carry out an awareness-raising activity before conducting the survey. The researchers also found that the collection methodology needed to be clearly explained to all involved to avoid misunderstandings.

Selecting indicators and developing survey tools require research

The utility and success of any CBMS hinge on the selection of indicators. Developing this set requires first analyzing and comparing poverty-monitoring systems in use, determining how poverty can be defined in the community, and developing those indicators.

Trials in all countries have shown that a set of core indicators on the multi-dimensional aspects of poverty is needed. These indicators need to be simple enough for the enumerators to collect and community members to understand. In fact, field tests in a number of countries, such as Burkina Faso, led to a review and simplification of the indicators originally selected where it was found that lengthy questionnaires reduced the quality of responses.

Using the core set of indicators in all localities allows for aggregation at higher levels, thus increasing the usefulness of the data. Adding locally specific indicators increases utility at the community and local government levels. In Senegal, Vietnam, and the Philippines, for instance, the CBMS researchers found that different indicators needed to be introduced for different geographic regions.

The indicators also need to be revised periodically and evolve as local conditions change. Indicators related to emerging concerns may need to be incorporated in the system. Moreover, further use and analysis of the data to come up with new composite indices or targeting mechanisms may be necessary to respond to the needs of decision-makers. All these require ongoing research.

The design of survey tools — questionnaires — must take into account the knowledge levels of the local population, as well as the availability of data-processing equipment and software. Everywhere, recruiting local enumerators was found to increase community buy-in, build capacity, and increase the prospects for sustainability. But, as in Burkina Faso, this often requires reviewing and amending the questionnaire, rendering it into a user-friendly medium, and translating it into local languages.

Adequate training should not be underestimated

Training is critical to successfully implementing a CBMS. This is particularly so in communities with low literacy skills. In almost all countries where CBMS has been introduced, the teams initially underestimated the time needed for training, particularly for data processing and analysis.

Computerized data processing helps to speed up analysis, and an electronic database is easier and quicker to consult. But it is a viable option only where local human, technical, and financial resources exist.

Data collection and processing must be done in a timely manner

Time is of the essence in carrying out the survey, in processing, and in feeding results back to the community.

In the Philippines, it was determined that surveys should start soon after the training. The team has found that a month is usually sufficient time for the survey to be completed. Experience there shows that an enumerator can interview, on average, 10 households a day. And, of course, enumerators must be literate.

In a number of countries, it was found that qualitative and quantitative methods of data collection should be used in tandem. This was particularly obvious in Vietnam and Bangladesh. Teams in these countries supplemented the questionnaire with group discussions and interviews with key community informants.

Accurate processing is critical. Whether this is done manually or by computer — or a mixture of both — depends on the capacity and resources of the local government implementing CBMS.

Computerized processing has evolved through the years and from on-the-ground experiences. In the Philippines, local government partners experienced difficulties in using Microsoft Excel and the initial encoding sheets developed by the CBMS team. A more efficient, user-friendly encoding system — based on the Census and Survey Processing (CSPro System) — was therefore adopted. This freeware from the US Bureau of Census is designed to process household censuses and surveys. The solution to processing problems is the CBMS Indicator Simulator, developed by the team and made available free of charge.

In Cambodia, the lack of electricity did not deter the CBMS research team from exploring ways to use computers in processing the data — they used a car battery to power a laptop computer and a printer. This technology can be promoted in other CBMS areas with limited or no access to electricity.

Validating the data is essential

Community validation not only allows the CBMS team to test the reliability of data, but it also builds understanding within the communities of the dimensions of poverty and development. In so doing, it helps to empower communities, increase local ownership, and to mobilize communities for development programs.

In Senegal, for example, it was found that enumerators and their supervisors were best placed to lead the validation process. Experiences described in Part 3 of this book demonstrate the importance of bringing together government, civil society, and the community to validate results. Not only does this ensure that all the community learns about survey results, but it also provides an avenue to verify the accuracy of the findings and to provide explanations for unexpected outcomes.

The validation meeting also serves as a venue for identifying major problems and possible interventions, and facilitating the integration of CBMS results in the community’s development and investment plans and socio-economic profile.

Dissemination is crucial

Returning the data to all stakeholders is central to CBMS. Meetings, workshops, community scorecards and Village Books, CD-ROMs, and newsletters — even traveling caravans — have all been used to present the results to communities, enabling them to set priorities and seek solutions to problems. In Bangladesh, for example, a village information book was prepared and planning workshops organized for service-delivery agencies, local government representatives, and villagers. The data book enabled the local government to identify needy households and record actions taken, thus helping to avoid duplication. The data also allowed local government to say no to demands not based on evidence.

Validation and dissemination go hand in hand in empowering the community. The process provides them with information and a process through which they can actively participate in planning. It enables them to develop a keen sense of their priorities and — on the basis of evidence — to articulate their needs to city planning officers. Armed with hard information on their condition, they are able to have more of a say in the allocation of public resources. And they can demand accountability and transparency on the part of government officials.

Obviously, CBMS can raise expectations that are difficult to meet. This points to the need for CBMS to provide communities and local governments with realistic information about local conditions, and to develop strategies for linking research to policy and practice. It also underlines the need for patience.

CBMS is a two-way process

These experiences have also yielded another lesson: the capacity of local governments to draw upon knowledge within communities, and for communities to learn about local government processes and governance. The CBMS process is not just about the flow of community and household information to local government planners, but also of information about policies and programs to communities, about government capacities and limitations, and about accountability. Communication often involves innovations to make this process better understood, and to take CBMS further than originally conceived. Burkina Faso’s rendering of CBMS data in the form of pictures is one such innovation.

Among other innovations, the Philippine CBMS team prepares poverty maps from spot maps — drawings of the location of households and major infrastructure facilities in the community. These maps show relative locations of households and service facilities and are adequate to meet the objectives of CBMS. Some of the more advanced local governments have started to use global positioning systems to determine the location of their major infrastructure facilities and then situate housing units. Geo-referencing some of the major landmarks in the community can be useful, particularly for rescue operations following natural disasters.

Lessons learned about the benefits of CBMS

The experimentation with and implementation of CBMS around the world has also yielded information about the benefits of the system, some obvious, others less so.

Throughout this volume, it has been shown that CBMS

• empowers the community by building its capacity to participate in diagnosing the problem, offering solutions, and monitoring the impact of these solutions;
  
  
• enriches local government databases as well as those at higher levels;
  
  
• enhances the preparation of socio-economic profiles, as well as development and investment plans;
  
  
• improves the allocation of resources by making it easier to prioritize interventions based on the local poverty diagnosis;
  
  
• increases equity in resource allocation between communities and households, as well as between men and women; and
  
  
• helps to monitor the impact of projects and programs, thus contributing to poverty-reduction efforts.

CBMS pilots have also created strong research-to-policy links. Policymakers — especially those at the local level — have adopted CBMS as a powerful tool to inform policy decisions. Not only has micro data provided the contextual basis that helps local authorities respond effectively to development needs, but policy-makers themselves have often said that the system makes good political sense.

CBMS also increases transparency and accountability of local governments in resource allocation, thereby improving governance. In this way, it is a support mechanism for implementing decentralization policies being pursued around the world. And as this book has shown, CBMS data yields real benefits for communities, including new facilities such as schools, roads, and markets. It has also helped mobilize communities to address some of their own problems.

Extending CBMS uses and benefits

While this book has focused on the implementation of CBMS to monitor poverty at the local level, benefits also accrue at higher geopolitical levels. At a national level, for instance, CBMS data can help validate poverty analyses by offering a disaggregated look at household or community poverty profiles. It can also trace the evolution of poverty over time in a given community and at household levels, thereby providing insights into the dynamics of poverty transitions. This can help design safety nets and redistributive mechanisms and inform policies to better target programs and interventions.

The monitoring systems are being used in a number of other ways, to meet different needs. For instance, UNDP is exploring CBMS as an alternative to weak national information systems in West Africa to better understand energy use and energy poverty at community and household levels, and to monitor access to services. CBMS is being tested in the Philippines to facilitate local-level gender-responsive budgeting. And in both Cambodia and Vietnam, the systems are being used to target households for social protection programs, such as health entitlements and cash transfers.

The data is also proving useful to the private sector. In December 2007, for instance, Fenway International, Inc. — a Canadian cement producer investing in Palawan province in the Philippines — used CBMS results for the municipalities of Sofronio Española, Quezon, and Narra to identify areas for its various social development projects, including schools,barangay clinics, and farm-to-market roads.

Some of the ways in which CBMS is evolving and the challenges this poses are explored in Part 5.

Document(s) 7 of 10