Essays in education and health economics
- Autores: Cristina Bellés Obrero
- Directores de la Tesis: Robin Hogarth (dir. tes.), Marc Vorsatz (codir. tes.)
- Lectura: En la Universitat Pompeu Fabra ( España ) en 2017
- Idioma: español
- Tribunal Calificador de la Tesis: Ghazala Azmat (presid.), Joan Costa Font (secret.), Caterina Calsamiglia Costa (voc.)
- Programa de doctorado: Programa de Doctorado en Economía, Finanzas y Empresa por la Universidad Pompeu Fabra
- Materias:
- Enlaces
- Tesis en acceso abierto en: TDX
- Resumen
This doctoral thesis brings together three independent research projects that involve the study of incentive programs in the educational context, and the empirical analysis of a labor market policy over fertility and infant health outcomes.
Even though there are substantial socioeconomic returns to education, underinvestment and low educational performance are still important problems in many developed and developing countries. As a consequence, many governments and policy makers have implemented monetary incentive programs, targeting teachers and students, to boost educational attainment. Students' incentive programs aim to correct for students' insufficient motivation, excessive discount rates for the future, or underestimation of the returns to education, which are considered the be main reasons for their suboptimal study effort. Given that teacher quality is a key factor determining student achievement, teachers' incentive programs, on the other hand, intent to compensate for the incentives generated by teachers' compensation policies (flat salary progression and lifetime job tenure) that create weak incentives for teachers to exert high levels of efforts. However, evidence on the effectiveness of these programs is still scant and inconclusive. Moreover, previous studies have primarily focused on the question of whether a particular type of incentive works in a specific educational environment, overlooking the role of the incentive design. The first two chapters of this thesis aim at shedding some light on the impact of these incentive programs and their design on students' attainment.
In the first chapter, I investigate which features of these incentive programs are effective at increasing students' attainment and how these interact with the characteristics of the students that are being incentivized. For this purpose, I conduct a randomized control trial at a distance learning university in Spain to compare three monetary incentive schemes with different performance targets for students. The distance learning setting is particularly advantageous for this intervention as it guarantees independent observations (no spillovers between treatments and the control group) and no effect on teachers, at a very low cost. The first treatment (Threshold) provides a cash reward for students who achieve a grade threshold, the second (Top percentile) for students in the top of their class, and the third (Improvement) for those that improve their expected grade. From a theoretical point of view, incentive schemes with different performance targets should increase attainment of students with different characteristics. I focus on the interaction of the incentive schemes with three characteristics of students that have been widely discussed in the literature on education as determinant for incentives to work: intrinsic motivation, experience with the incentivized task, and ability. I find that for the "Top percentile'' treatment, the effect of the monetary reward is positive for students with a high intrinsic motivation and negative for students with a low intrinsic motivation. On the other hand, the "Threshold'' and "Improvement'' incentives have positive effects on students with more experience with the incentivized task and negative effects on those with less experience. Finally, a novel finding of this paper is that incentives foster students' strategic behavior, that is triggered by the way performance is measured (multiple choice exam with penalties for incorrect answers). The study emphasizes the importance of incentive design. The question of whether monetary incentives are effective in increasing students' attainment is too narrow and further research is needed to identify the features of incentive design that matter in practice as well as how different design features interact.
In the second chapter, María Lombardi and I focus on incentive programs for teachers. Specifically, we investigate how tying teachers' pay to the performance of students affects the latters' achievements. We conduct an evaluation of "Bono Escuela", a nationwide program implemented in public secondary schools in Peru that gives monetary rewards to teachers conditional on their students' performance in a standardized test. The program takes the form of a tournament, awarding a bonus of over a month's salary to the principal and every teacher from schools in the top 20 percent within a group of comparable schools. The main measure used to rank schools in the tournament is the average score of 8th graders at the standardized test. Using a novel administrative database covering the universe of Peruvian students, we perform a difference-in-differences estimation comparing the change in the internal grades of 8th graders before and after the incentive was introduced to that of 9th graders from the same school. We find that the program had a precisely estimated zero impact on students' grades. We argue that this zero effect can be explained by some aspects of the program's design (students' low stakes, teachers' inexperience with the standardized test, and the program's timing), which may have hindered teachers' ability to improve the incentivized outcome or infer their probability of winning. The study sheds some light over the scarce literature on scaled-up teachers pay-for-performance programs in developing countries and highlights the need of a deeper understanding about the role played by the different characteristics of these programs in their success.
My third chapter, co-authored with Sergi Jiménez Martín and Judit Vall Castelló, studies the effect of a child labor regulation on fertility and infant health outcomes. For this purpose, we examine a child labor reform that took place in Spain in 1980, which increased the minimum legal age to work from 14 to 16 years old while the compulsory school age was maintained at 14. Contrary to previous literature, we exploit the interaction between the compulsory schooling age and the minimum legal age to work, as both age thresholds are important to determine an individual's decision to remain in the educational system. We perform a difference-in-differences strategy to identify the reform's within-cohort effects, where treated and control individuals only differ in their month of birth. We find that the reform increased educational attainment, but decreased womens' completed fertility and reduced their probability of ever marrying. In addition, we show that the reform was detrimental for the health of the offspring at the moment of delivery. Newborns of women affected by the reform had a higher probability of being premature and having low birth weight. We identify three different channels contributing to this detrimental effect: the postponement of fertility, the change in the maternal marital status, and the improvement in the labor market conditions of more educated women, which increases the likelihood of engaging in unhealthy behaviors such as smoking for the affected cohorts. This last surprising result is a direct consequence of the socioeconomic situation of Spain when the reform took place. Women in these cohorts grew up during the early post-Franco era and experienced the process of gender equalization. As a consequence, these more educated women had more access to and social acceptance of smoking compared with pre-reform cohorts. This study is informative, from a policy perspective, for developing countries whose educational system, child labor market participation rates, and womens' social development are similar to the levels that Spain was experiencing around 1980.
