When school-age children work, their education competes for their time and effort, which may lead to lower educational attainment and academic achievement. This paper develops and estimates a model of student achievement in Mexico, in which students make decisions on school enrollment, study effort, and labor supply, taking into account locally available schooling options and wages. All of these decisions can affect their academic achievement, which is modelled using a value-added framework. The model is a random utility model over discrete school-work alternatives, where study effort is determined as the outcome of an optimization problem under each of these alternatives. The model is estimated using a large administrative test score database on Mexican 6th grade students combined with survey data on students, parents, and schools, geocode data on school locations, and wage data from the Mexican census. The empirical results show that if child labor bans were enforced, dropout between Grade 6 and Grade 7 would only decrease by 6.6%. To minimize dropout, access to school is most important, and policies that decrease travel costs or increase perceptions of rural schools are effective. 

Millions of students around the world are retained and repeat a grade each year, yet we may not fully understand the consequences of this choice. Especially in low-income countries, where the majority of repeaters reside, and the decision is often at the teacher's discretion, it is important to understand how parents, students, and schools react to this negative information shock. In this paper, we use rich matched household-child-school panel data from Pakistan to study the dynamics of the beliefs, investments, and outcomes of these three agents following a student not being promoted to the next grade. We find that following grade repetition, parents revise downward their expectations, beliefs, and investments. Students are discouraged by the retention, and decrease their beliefs in the value of study effort. Conversely, schools do not play a large role, and teachers have no negative bias towards students who are repeating. Overall, we find negative effects of repetition: repeaters score -0.27 to -0.44 standard deviations worse in math, English, and Urdu and are 7.1 percentage points more likely to drop out than their peers.

In many countries, conditional cash transfers offered by the government have been shown to increase school enrollment for the beneficiaries. However, these policies’ potential to increase overall human capital depends on the schools’ capacity to expand enrollment with sufficiently high-quality instruction, ensuring that larger classes do not negatively affect peers. We develop a structural model in which students first face an enrolment decision, and then an effort decision which will impact their achievement and the effort of their teacher. Class composition and effort choices are determined endogenously via a strategic game, which takes into consideration peer effects within the classroom. To estimate the model, we combine administrative data on test performance and cash transfers, socioeconomic surveys, and spatial data on child wages. Our model allows for heterogeneous endowments and teacher ability, and with it we can evaluate the impact of a conditional cash transfer on not only beneficiary enrollment choices and achievement, but also on their classmates.