Essays in development economics : incumbency disadvantage, political competition, and remedial education in India

Abstract

(cont.) The remedial education program hires young women from the community to provide remedial assistance to third and fourth grade children who have fallen behind their peers. The program is extremely cheap (five dollars per child per year), and is easily replicable. We find the program to be very effective, increasing learning by 0.15 standard deviations in the first year, and 0.25 in the second year. The results are similar in the two grade levels, and in the two cities.

This dissertation comprises three separate empirical studies. Using a non-parametric regression discontinuity design that compares candidates who barely win an election to those who barely lose, the first study estimates the effect of incumbency on a candidate's electoral prospects in India. Starting in 1991, I estimate that, rather than being at an advantage, incumbents are actually fourteen percent less likely to win an election than similar non-incumbents. While the available data prevent a formal test, the dominance of a single political party (the Indian National Congress) before 1991 may have provided a framework in which experience was valuable because incumbents who gained experience under the Congress system would interact with the same system when reelected. Starting in 1991, however, no party could be counted on to control parliament, making experience under the previous regime potentially less valuable. The second study estimates the effects of new competitors on existing candidates in India by taking advantage of a change in the election laws in 1996 that made it more difficult for candidates to run for office. The law affected constituencies differently, allowing the use of both across time and between constituency variation in the number of candidates to estimate the impact of restricting the number of new candidates in an election. The resulting estimates suggest that the reduction in the number of new candidates had a small, but measurable effect on the probability that the average existing candidates would win election. However, there is evidence of heterogeneity in the effect across candidates. Finally, the third study presents the results of a two-year randomized evaluation of a remedial education program in India.