2004 Bridges Summer Research Report Frederico Finan Economics "Evaluating the Impact of Decentralized Conditional Cash Transfer Programs: A Study of Brazil’s Bolsa Escola Program"

Incentive-based poverty reduction programs have generated substantial interest within the international development community. Also known as conditional cash transfers, these social programs can help alleviate short-term poverty (through the provision of a cash transfer), while providing households the incentives to invest in human capital and thereby reduce poverty in the long-run. Central to the success of these programs is the ability to effectively reach and engage the poor. This has led several countries to experiment with devolving to local communities the targeting of program recipients.

Devolution of decision-making power from the central government to the local community has several potential advantages that can lead to better targeting outcomes and improved project performance. Local authorities tend to have more information about the community and can better identify the poor, which should allow for fewer targeting errors. With better information on local costs and fewer levels of bureaucracy, the local community can deliver goods and administer the program more efficiently than a central government that must rely on monitoring devices. Because local institutions are potentially more accountable to local citizens, this creates further incentives for both better program targeting and program outcomes.

The decentralized experience has, however, varied significantly and with far too many failures. The advantages of decentralization, ultimately, depend on its design and the institutional arrangements governing its implementation. Without laws, rules and oversight over local government operations, or mechanisms to express views to government bodies or change public service providers, local governments may actually be less accountable to local citizens; in which case, a decentralized program creates the potential for rent-seeking and corruption. Capture of the local government by interest groups may then aggravate targeting failures and worsen project performance. The debate over decentralization centers, for this reason, on the tradeoff between efficiency gains and the extent of program capture by local elites.

With a few exceptions, progress in analyzing decentralized poverty programs has, to this date, been mostly theoretical, while empirical evidence is both sparse and inconclusive. With a clear need for further empirical work, the objective of my research is to evaluate the effects of decentralized beneficiary selection on a conditional cash transfer program aimed at improving the schooling of children of poor families throughout Brazil.

Bolsa Escola is a demand-driven education program that provides cash transfers to mothers of poor children throughout Brazil, conditional on their children’s continued attendance in school. While initiated in 1995 on the outskirts of Brasília, Bolsa Escola became a nationwide federal program in 2001. By the end of 2001, Bolsa Escola had been implemented in 98 percent of the 5,561 Brazilian municipalities, providing stipends to over 8.2 million children from 4.8 million families, at a cost of over US$700 million. Having benefited millions of Brazilian school-aged children, the program has served as a source of inspiration and a point of comparison with similar educational programs throughout the world. And yet, surprisingly, even with these levels of investment and its world-wide acclaim, there exists little quantitative evidence on the impact of the program. This research will analyze the impact of Bolsa Escola on education and targeting outcomes.

Bolsa Escola was implemented in two stages. First, the Federal government allocated, based on a determination of need, the number of federally-financed stipends that a municipality could provide to its population. Secondly, given this number of stipends, the municipality selected which households receive the program. This devolution of the selection process allowed each municipality to target the program, within the general guidelines, according to its own local objectives and preferences. As a result, the program’s impact on schooling and targeting outcomes may vary considerably across municipalities and expectedly along a number of observable dimensions. This two-stage design thus provides a unique laboratory to analyze how differences in institutional settings and program implementation affect the program’s impact on school attendance and achievement and the targeting of children who are at risk of discontinuing school.

With these research objectives, the purpose of my recent trip to Brazil was to better understand, through a series of case studies, how differences in the local institutional settings may affect the implementation and performance of this program. In the end, I conducted case studies in nine municipalities located in states of Bahia, Ceará, and Paraíba.

To capture a variety of perspectives, these case studies generally consisted of interviews with the program coordinator, beneficiaries of the program, school teachers, local politicians and some other key member of the community; such as the local pastor, or president of the agricultural worker’s union. These interviews were structured to help understand the process and criteria by which the municipalities selected the program recipients given that the number of qualified families typically exceeded the municipality’s quota.

I quickly discovered that there was substantial variation not only in the manner in which municipalities registered and selected its beneficiaries but also in the transparency of these processes. A municipality in Ceará, for example, used school teachers to enroll all the eligible families and then relied on a social council to verify the household’s information and select beneficiaries. Among the qualified families, the municipality prioritized households with single mothers and those that lived in the more distant rural areas. What made this municipality unique was that the names of the recipients along with the selection criteria were published in its local newspaper. This municipality provides stark contrast to one in Paraiba where families and community did not know how beneficiaries were selected and most assumed the selection was done in Brasilia by the federal government, when in reality the mayor had decided the list of beneficiaries. Interestingly, these municipalities shared other differences in the transparency of their budgeting process, the degree of nepotism in the local administration, and other aspects of local governance.

Given Brazil’s long history of clientelistic practices and patronage, it is perhaps not too surprising that in some municipalities, local politicians used Bolsa Escola as a political instrument. In a few of these municipalities, local politicians either selected families on the basis of their political support and/or threatened to remove families if political support was not given. In addition to these clientelistic practices, there were also indications of blatant fraud, as in one municipality where a local politician actually participated in the program.

These case studies have been central in the design of a municipal survey that will be applied in approximately 300 randomly selected municipalities in the Northeast region of Brazil. The survey instrument will try to measure the level of transparency of government actions, citizen participation in service delivery, formation of social councils and the level of vote buying. Also, in many municipalities where social councils (conselhos) participate in the selection of beneficiaries, it is important to measure the effectiveness of these councils and the degree of their community representation. Existence of a functioning oversight committee may also prove useful in identifying checks and balances on the mayor and municipal council members.

In many respects, the Northeast provides the ideal laboratory. The Northeast, representing one of the poorest regions of Brazil, is where we expect Bolsa Escola to have its most significant impact. It is also a region with a long-standing tradition of clientelistic and oligarchic politics and where the local government plays an immensely important role in the economic livelihood of its population. The Northeast is also highly diverse and should present a range of institutional settings in which to analyze the program.

An immediate evaluation of the Bolsa Escola program comes at a very opportune moment. It was announced, in 2003, that Bolsa Escola and Brazil’s other cash transfer programs will be unified into a single program called Bolsa Familia. The program, which is being supported by the World Bank, represents an integral part of President Lula’s antipoverty campaign and a major investment in Brazil’s future. Since the design and implementation of the Bolsa Familia program are vital to its future success, the benefits of a rigorous evaluation exploring the impact of different institutional arrangements on a program’s performance are high. Although several conditional cash transfer programs in various countries have been evaluated (see for example, Colombia, Costa Rica, Honduras, Jamaica, Mexico, Nicaragua and Turkey), few are as decentralized as Bolsa Escola. An analysis of Bolsa Escola’s rich variety of implementation experiences and decentralized design could potentially serve as important inputs for the design of the new Bolsa Familia program, as well as similar programs around the world.