Publications and working papers

Estimating Impact with Surveys versus Digital Traces: Evidence from Randomized Cash Transfers in Togo

with Emily Aiken, Joshua Blumenstock, Dean Karlan and Chris Udry.

Revision requested at Journal of Development Economics.

Abstract

We study whether program impacts can be estimated using a combination of digital trace data and machine learning. In a randomized controlled trial of cash transfers in Togo, endline survey data indicate positive treatment effects on food security, mental health, and perceived economic status. However, estimates of impact based solely on predicted endline outcomes (generated using trace data and machine learning, which do successfully predict baseline poverty) are smaller and noisier, and generally not statistically significant. When post-treatment outcome data are used in conjunction with predictions to estimate treatment effects, predicted impacts are similar to those estimated using surveys.

Residential and Social Mobility: A Quantitative Analysis of Parental Decisions with Social Learning

Submitted. 09/2024.

Abstract

In the United States, less-educated parents tend to allocate little time to parent-child activities, reside in disadvantaged neighborhoods, and underestimate the relevance of parental inputs for later outcomes. This paper introduces a social learning mechanism that can explain socioeconomic differences in parental beliefs and decisions. The mechanism posits that young adults learn by observing older people in their neighborhood but are prone to misinferences as they imperfectly correct for selection induced by residential segregation. I incorporate this social learning mechanism in an overlapping generations model of residential and parental time decisions. Once calibrated to the United States, the model accurately captures both targeted and non-targeted parental behavior across socioeconomic groups. Parental beliefs have sizable effects on the economy, increasing income inequality by 3% (income Gini index) and decreasing social mobility by 15% (income rank-rank coefficient). The general equilibrium effects of housing vouchers hinge on their ability to reduce residential segregation. Greater effectiveness in reducing segregation leads to larger gains in welfare, income, and social mobility, but also extends the transition to a new steady state.

Macroeconomic Consequences of Early Education Tracking

with Lukas Mahler.

03/2024.

Abstract

We study the long-run macroeconomic, distributional, and intergenerational effects of school tracking—the allocation of students to different types of schools—by incorporating school track decisions into a general-equilibrium heterogeneous-agent overlapping-generations model. The key ingredient in the model is the child skill production technology, where a child’s skill development depends on her peers and the instruction pace in her school track. We show analytically that this technology can rationalize reduced-form evidence on the effects of school tracking on the distribution of child skills. We then calibrate the model using representative data from Germany, a country with a very early school tracking policy. Our calibrated model predicts that an education reform that postpones the tracking age from ten to fourteen generates sizable improvements in intergenerational mobility but comes at the cost of modest losses in aggregate human capital and economic output, reducing aggregate welfare. This efficiency-mobility trade-off is rooted in the effects of longer comprehensive schooling on child learning and depends crucially on the presence of general equilibrium effects in the labor market. Finally, our calibrated model predicts that policies reducing the parental influence in the school track choice increase both social mobility and aggregate economic output, improving aggregate welfare.

When Effective Teacher Training Falls Short in the Classroom: Evidence from an Experiment in Primary Schools

with Adrien Bouguen, Marc Gurgand, Valérie Munier and André Tricot.

Economics of Education Review (2024).

J-PAL WP
Abstract

Although in-service teacher training programs are designed to enhance the performance of several cohorts of students, there is little evidence on the persistence of their effects. We present the two-year results of a randomized study of an intensive in-service teacher training program conducted in France during and after the training program’s implementation. Our results highlight the short-run effectiveness of the training program: it successfully improves students’ performance but only during the implementation year. A detailed analysis of teachers’ outcomes indicates that teachers changed their pedagogical vision and practices but afterward struggled to apply skills to contents not directly covered during training.

Machine Learning and Mobile Phone Data Can Improve the Targeting of Humanitarian Assistance

with Emily Aiken, Dean Karlan, Chris Udry and Joshua Blumenstock.

Nature (2022).

CEGA
Abstract

Targeting is a central challenge in the administration of anti-poverty programs: given available data, how does one rapidly identify the individuals and families with the greatest need? Here we show that non-traditional “big” data from satellites and mobile phone networks can improve the targeting of anti-poverty programs. Our analysis compares outcomes – including exclusion errors, total social welfare, and measures of fairness – under different targeting regimes. Relative to other feasible targeting options, the machine learning approach reduces errors of exclusion by 4-21%. These results highlight the potential for new data sources to contribute to humanitarian response efforts, particularly in crisis settings when traditional data are missing or out of date.

Work in progress

Private Information in the Family

with Matthias Doepke and Michèle Tertilt.

Abstract

We theoretically and empirically investigate the hypothesis that private information plays a role within families. While standard theory implies that changes in relative income do not affect relative consumption in couples, we find a positive relationship in Dutch panel data. We collected data on how spouses share information with each other within the same panel and split our sample into those couples that are fully informed about each other’s income and those that are not. We find that the positive relationship between income and consumption shares is entirely driven by those couples that are not fully informed about each other’s income. This holds true when adding couple fixed effects, so that, indeed, a change in relative income within a given couple affects their consumption sharing. These findings are consistent with an optimal contracting model with private information. In the model, to provide incentives to share information with each other, the spouse who receives an income increase gets a larger promised utility, which is isomorphic to an increase in the bargaining weight. These findings suggest that information asymmetries are more relevant for families than previously believed.

Employment Transitions: Mechanisms for Gender Earnings Gap

with Valentina Melentyeva.

Abstract

We use German administrative data to compare women’s and men’s career choices and labor income profiles over their working lives. We find a significant and robust gender gap in job-to-job transitions. When working full-time, women are less likely than men to transition from one employer to another. The first finding is that while commuting preferences and sector selection do not explain this gap, the birth event largely explains it. Second, we want to quantify the consequences of the gender gap in job-to-job transitions for the gender earnings gap. If full-time job-to-job transitions increase wage growth, this phenomenon most likely contributes to the well-documented child penalty.