Implementing poverty reduction strategies remains a major challenge in developing countries that often lack the infrastructure to do so effectively. In response, many central governments use decentralized poverty alleviation programs, particularly cash transfer programs . While local officials’ knowledge may increase a program’s effectiveness, there must also be sufficient incentive for proper implementation. Therefore, lack of local governmental accountability may minimize or eliminate the effects of a decentralized poverty reduction program . This project evaluated how local electoral incentives affected the implementation of a decentralized cash transfer program.
In 2001, Brazil implemented Bolsa Escola – now a part of Bolsa Familia –a conditional cash transfer (CCT) program to reduce dropout rates for children legally mandated to attend school (ages 6-15). The program offered 6 USD per child per month stipend for each month of enrollment during the school year to all households with monthly per-capita income less than approximately 40 USD. The central government fully funded the program and set a quota for each municipality, based on earlier census data, which was typically less than the number of eligible households. Officials in local municipalities were responsible for identifying qualified households, selecting program beneficiaries, and program administration. Prior studies of the Bolsa Escola program found that a pilot version was successful, reducing dropout rates by between 5.2 and 7 percentage points. [3,4]
The Brazilian constitution limits mayors to two terms, creating different incentives for first-term and second-term mayors. First-term mayors have a stronger incentive to perform well given the possibility of reelection. Constituents who believe the mayor failed to act effectively can hold them accountable during the election. Second-term mayors, on the other hand, may have less incentive to perform well as they cannot be held accountable by voters during the following election.
Researchers used a difference-in-difference model to determine whether Bolsa Escola had a greater impact in municipalities where mayors faced reelection than in those with second-term mayors. They used data on 290,517 students from 261 randomly selected municipalities in four states from school records, Bolsa Escola payment records, municipal surveys and electoral outcomes. To control for mayor’s individual political ability, the analysis compared Bolsa Escola’s impact under second term mayors to only those first-term mayors who won their subsequent reelection bid. The sample of first-term mayors was also limited to those who had previously served at least two terms in another political office to control for political experience. Finally, in order to test the causal role of election incentives, the researchers compared program results of first-term mayors with second-term mayors who ran for another political office in the subsequent election.
Results and Policy Implications
The results indicate that electoral incentives improve the performance of decentralized public programs. On average, Bolsa Escola led to an eight percent reduction in student dropout rates. There was significant variation in municipalities depending on the mayor’s term. First-term mayors had a 36 percent larger program impact relative to those run by second-term mayors. Additionally, there was no difference between first-term mayors and second-term mayors seeking a different political office, indicating a causal link between electoral incentives and improved program performance.
The results indicate that voters were concerned about and aware of the effectiveness of the local Bolsa Escola CCT. Mayors who achieved among the 25 percent highest reductions in dropout rates were 28 percent more likely to win reelection. In addition, mayors who avoided public denouncement for using funds on unqualified beneficiaries were 26 percent more likely to win reelection. The results suggest that formal local institutions, particularly electoral rules that enable voters to reward and punish locally-elected officials, are key to achieving the benefits of program decentralization.
 Bardhan, P. (2002). Decentralization of Governance and Development. Journal of Economic Perspectives, 16(4), 185-205.
 Seabright, P. (1996). Accountability and Decentralization in Government: An Incomplete Contracts Model. European Economic Review, 40(1), 61-89.
 Abramovay, M., Andrade, C., Jacobo Waiselfisz, J. (1998). Bolsa Escola: Melhoria Educacional e Redução da Pobreza. UNESCO.
 Aguiar, M., Araújo, CH. (2002). Bolsa Escola: Education to Confront Poverty. UNESCO.
Photo Credit: Chico Santiago/Pref.Olinda via Flickr