Daniel Kammer

Income shocks and education: How coffee price booms affect schooling in Rwanda (currently under review at the Journal of Development Economics)

This  paper provides novel evidence on the short and long term effects of coffee price booms on schooling outcomes in Rwanda. Exploiting the exogeneity of international coffee prices and spatial variation in coffee cultivation intensity across Rwanda, I show that children are more likely to attend school when coffee prices are high and complete more years of schooling by the age of 18. I provide evidence that income effects are the main mechanism and dominate increased opportunity costs of schooling. However, educational gains are concentrated among households near privately operated washing stations, rather than households near cooperatives, suggesting that privately managed washing stations transmit income gains differently to local households than cooperatives. For privately operated stations, smaller station sizes, rather than differences in coffee quality, explains the observed effects. Overall, my findings highlight important heterogeneity in how commodity booms translate into educational gains in exporting countries. 

Keywords: Commodity booms, income shocks, schooling, coffee, cooperatives 

JEL codes: O13, O12, I20

Genocide, Women’s Empowerment, and Intergenerational Transmission of Violent Attitudes (with Alina Greiner-Filsinger, R& R at World Development)

In this paper, we explore how mass violence shapes attitudes on violence against children, and we examine how these attitudes are transmitted across generations in the context of the Rwandan genocide. Leveraging a natural experiment, we find that local genocide violence causes younger women from regions more affected by genocide to hold less violent attitudes compared to older women from the same regions, and to women in the same age cohorts from less genocide-affected regions. Using an instrumental variable approach to estimate the transmission effect, we show that descendants of these younger women from genocide-affected regions are similarly less likely to adopt violent attitudes. We provide evidence that genocideinduced empowerment is the underlying mechanism. As such, our findings underscore previous evidence on the conflict–prosociality link and demonstrate persistent effects across generations, but call for a more detailed investigation of the underlying mechanisms for the second generation. 

Keywords: Women’s Empowerment, Mass Violence, Attitudes, Intergenerational Transmission 

JEL codes: J13, D74, D79

The Human Capital Costs of Violent Teachers (R&R at the Economics of Education Review)

This paper provides evidence on the effect of violent teachers on human capital formation. I combine panel data from multiple cohorts of primary students in rural Malawi with detailed information on their teachers’ violent behavior during class, including physical, emotional and sexual violence. To tackle the potential endogeneity of teacher violence, I apply a difference-in-differences strategy with staggered treatment and estimate a value-added model that includes teacher fixed effects and lagged student's test scores. My results suggest that violent teachers reduce successful grade progression by 30 percentage points and math scores by 15 percentage points over an academic year. Early secondary students show the highest effects. I do not find any significant effects on English and Chichewa test scores.

Keywords: Teacher Violence, Education, Test Scores, Grade Progression, Malawi 

JEL codes: I25, O12, I20

Data and Methods in Development Research (Undergraduate Seminar - Spring 2024)

Syllabus

Education and Development (Undergraduate Seminar - Fall 2023)

Syllabus