Research

PUBLICATIONS

Outsourcing Education: Experimental Evidence from Liberia
(with Mauricio Romero and Justin Sandefur)
American Economic Review, Volume 110, No. 2, 2020
AEA registry

Abstract
In 2016, the Liberian government delegated management of 93 randomly selected public schools to private providers. Providers received US$50 per pupil, on top of US$50 per pupil annual expenditure in control schools. After one academic year, students in outsourced schools scored 0.18σ higher in English and mathematics. We do not find heterogeneity in learning gains or enrollment by student characteristics, but there is significant heterogeneity across providers. While outsourcing appears to be a cost-effective way to use new resources to improve test scores, some providers engaged in unforeseen and potentially harmful behavior, complicating any assessment of welfare gains.

Sibling Spillovers and Free Schooling
(with João Ferreira)
Review of Economics and Statistics, 2026
PDF — Supplemental Appendix — CESifo WP 11436 — IZA DP 17228 — EdWorkingPaper 25-1320

Abstract
We use administrative data to measure sibling spillovers on academic performance before and after the introduction of Free Secondary Education (FSE) in Tanzania. Prior to FSE, students whose older siblings narrowly passed the secondary school entrance exam were less likely to go to secondary school themselves; with FSE, the effect became positive. A triple-differences analysis, using geographic variation in FSE exposure, shows that FSE caused the reversal. Mechanism analyses suggest that changes in parental investments were a more likely channel for this reversal than direct sibling interactions. By alleviating financial constraints, FSE allowed households to invest in more children.

The Leaky College Pipeline: Evidence from Administrative Data in Uganda
(with Isaac Ahimbisibwe, Adrien Dautheville, Lenka Fiala, and Saint Kizito Omala)
AEA Papers & Proceedings, Volume 116, May 2026
PDF — Supplemental Appendix

Abstract
We use linked administrative data from Uganda to track student progression from primary school through university scholarship admission, documenting large leaks in the education pipeline at every stage. Many high-performing students from poorer districts exit before upper secondary school, so the pool eligible for selective public university admission is highly socioeconomically stratified. Merit scholarships are heavily concentrated in a small number of elite secondary schools in the Kampala metropolitan area, implying that financial aid primarily rewards earlier advantages. The results highlight that tertiary merit aid alone cannot equalize access without upstream reforms to broaden basic and secondary education access.

WORKING PAPERS

Secondary school access raises primary school achievement
Conditionally Accepted, American Economic Journal: Economic Policy
CESifo WP 11343

Abstract
I study how the introduction of Free Secondary Education (FSE) affected primary students in Tanzania. I first show descriptive evidence that FSE increased secondary access: secondary enrollments rose, household spending on secondary school fees fell, and elites’ secondary transition advantage disappeared. Then, using variation in treatment exposure from fee payment microdata, I show that FSE increased primary exam pass rates by 4–6% and secondary transition rates by 8–19%. This appears to have been driven by demand-side investments rather than supply inputs: students selected into better primary schools, attended more, and worked less.

Tax morale, public goods, and politics: Experimental evidence from Mozambique
(with Pedro C. Vicente)
Revised & Resubmitted, Economic Development & Cultural Change
IGC WP MOZ-22067

Abstract
Tax revenue is vital for development, but governments must balance raising revenues with maintaining political support. Partnering with a city government in Mozambique, we experimentally test a grounded hypothesis: that tax morale and political support are increased by information highlighting 1) public good provision and 2) local political autonomy. We find that the treatments have different strengths. Public goods information raises tax morale, driven by areas of low baseline public good provision, but has no effect on voting. The political message increases electoral support but raises tax morale only among co-partisans. Balancing tax morale and political support requires careful communication.

The politics of public service reform
Revise & Resubmit, Journal of the European Economic Association
VoxDev column — Medium post — CESifo WP 10633 — SSRN — IZA DP 18346

Abstract
This paper provides experimental evidence on the electoral effect of a large education reform in a developing democracy. Despite significantly improving school quality, the reform reduced the incumbent party’s presidential vote share by 3 percentage points (9%). This does not imply that ordinary voters opposed the policy’s aims: electoral effects were positively correlated with school quality effects, and household surveys showed strong support. However, the reform also reduced teachers’ job satisfaction, support for the incumbent government, and political activity. The reform lost the most votes where it caused the greatest political disengagement of teachers.

The Political Consequences of Controversial Education Reform: Lessons from Wisconsin’s Act 10
(with Barbara Biasi)
Under Review
NBER WP 33666 — IZA DP 17836 — CESifo WP 11817

Abstract
We study the electoral consequences of Wisconsin’s Act 10, a controversial law that weakened teachers’ unions and enabled flexible teacher pay. Exploiting variation in the timing of implementation across districts, we first show that it raised student test scores, reduced union revenues, and created winners and losers among teachers in terms of pay. We then show that the reform increased the vote share for the incumbent GOP governor by about 2 pp and lowered Democratic campaign contributions from teachers and unions. The electoral gains were driven by districts with ex ante stronger unions and more potential winners among teachers and students.

On the Political Economy of Urbanization: Experimental Evidence from Mozambique
(with Alex Armand, Frederica Mendonça, and Pedro C. Vicente)
Under Review
CEPR DP 21450

Abstract
Urbanization can generate large economic gains, but it presents electoral risks for incumbents. This paper studies the economic and political effects of a program to integrate rural migrants in a growing Mozambican city. We randomized city blocks into three arms: program delivery with or without the involvement of local leaders, and a no-program control. The program increased rural-to-urban migration across both treatment arms, but improved labor market integration and local incumbent electoral outcomes only with leader involvement. We also observe electoral spillovers in migrants’ origin areas, no clientelistic responses, and no resident backlash. These findings show that city-level integration policies can deliver both economic and political returns.

Won by a mile: Electoral effects of the Interstate Highway System
(with Daniel Leff Yaffe and Alejandro Nakab)

Abstract
We measure the electoral effects of the construction of the US Interstate Highway System (IHS). We show descriptively that construction follows state electoral cycles. To identify causal effects, we use a shift-share instrument based on temporal variation in federal funding formulas and geographic variation in planned mileage. Completing one extra highway mile in a county during an election year increases vote share for incumbent governors’ and senators’ parties by 2.9 and 2.7 percentage points respectively, with positive spillover effects on unconnected neighboring counties. Electoral effects for incumbent presidents are negative, driven by states governed by the opposing party.

Priming the pump: Can upfront interest payments increase savings?
(with Peter Carroll, Flora Myamba, Daniel Nielson, Joseph Price, and Phillip Roessler)
IPA summary

Abstract
We encouraged an NGO’s cash transfer recipients in Tanzania to keep 10% of the transfer in their mobile money account until a certain date. We randomized whether doing so would permit them to keep a bonus provided upfront, receive an equivalent amount of “interest” at the end, or confer no benefit. Mobile wallet administrative data show that the Upfront group were 14pp (74%) more likely to reach the savings target (net of the bonus). This was driven by participants with higher balances ex-ante. Offering interest payments at the end had no effect on saving, despite a significantly above-market interest rate.

WORK IN PROGRESS