Working Papers

  • "Pre-school Education and Long-Run Human Capital Formation"
    with Elisabetta Aurino, Katherina Thomas and Sharon Wolf
    ▾
    Abstract

    While early childhood education is rapidly expanding globally, concerns about quality and long-term effects remain. This paper leverages rich longitudinal and experimental data to study the effects of improved-quality early childhood education on skill development through mid-childhood, and the mechanisms underlying longer-term treatment effects. In 2015, preschools in Greater Accra, Ghana, were randomized to an intervention designed to enhance child school readiness through teacher professional development and increased parental awareness about early childhood education. Following children from preschool through age 8, we estimate both reduced-form effects on cognitive and social-emotional skills and structural models of skill formation to understand how initial gains from the intervention evolve over time. A key contribution is that we extend the technology of skill formation to explicitly incorporate teacher inputs, alongside home inputs, allowing us to quantify the role of classroom-level investments and reassess treatment mechanisms in a teacher-focused intervention. Preliminary results show that the fade-out commonly observed in simple reduced-form specifications largely disappears once we properly account for measurement error in cognitive and social-emotional skills, highlighting the critical importance of latent factor models for policy evaluation. Our structural model decomposes the mechanisms driving treatment effect evolution through the technology of skill formation, parental investments, and school/classroom quality. Results highlight that improving preschool quality provided initial gains that exhibited self-productivity, reinforcing themselves over subsequent periods.

  • "Reshuffling the Desks? Affirmative Action, Segregation, and Educational Attainment"
    with Clemens Oberhuemer and Juan Pascual
    ▾
    Abstract

    This paper presents the first causal evaluation of quota-based desegregation policies, exploiting Catalonia's rollout of reserved quotas for socioeconomically vulnerable students. Drawing on exhaustive administrative data, we estimate effects on peer composition, educational outcomes, and parental school-choice behavior. Preliminary results show that the policy lead to an increase in the share of students with special educational needs derived from socio-economic conditions through improved detection and significantly decreased school segregation. At the same time, first evidence points to a decline in demand, driven mainly by schools that began in the lowest and middle tertiles of vulnerable-student share.

Work in Progress