Magnus Eldrup

Email: mel@econ.ku.dk
Affiliation: Department of Economics, University of Copenhagen
Fields: Public economics, development, taxation
CV: PDF

Hi there. I am a PhD student at the University of Copenhagen with the Development Economics Research Group (DERG). Through my research I try to help better understand issues of taxation in developing countries, especially as it pertains to the interplay between the formal and the informal sector.

Before embarking on my PhD I completed the Msc. in Economics for Development at the University of Oxford and an MA. in Economics from the University of Copenhagen. In the fall of 2025 I visited the Harvard Kennedy School as a PhD Fellow.

Job Market Paper

Horizontal Inequality in Optimal Taxation: Evidence from a Development Context
Job Market Paper
Presented at CSAE 2026, KIEL-CEPR AEDC 2025 and NPPS 2025

Abstract (click to expand)

Exemptions are a key feature of income tax systems in developing countries, yet standard optimal tax models cannot explain them. By assuming full compliance, these models ignore horizontal inequality between formal and informal workers, understating the social cost of taxing the former. Extending the optimal tax framework, I incorporate informality and horizontal inequality aversion. I conduct a survey experiment in Kenya, finding that respondents place substantially higher welfare weight on formal taxpayers than otherwise identical informal workers. Combining these empirical preferences with administrative tax data yields prescriptions of high exemption thresholds, a result alternative explanations such as extensive margin responses or administrative costs cannot rationalize.

Other Ongoing Research

Optimal Enforcement of Redistributive Taxation
with Claus Thustrup Kreiner
Presented at CPER Public Economics 2026, CESifo Public Economics Conference 2025 and IIPF 2025

The Limits of VAT Neutrality: Theory and Evidence from Kenya
with Cyrus Mutuku and Bjørn Bo Sørensen

Preliminary abstract (click to expand)

The value-added tax is prized for preserving production efficiency, but this rests on full refunding of excess credits. We assemble a new database on VAT refund systems and show that 64.5 pct. of people in VAT countries live under source-based systems that often restrict refund eligibility to zero-rated activity. When firms in these systems supply both zero-rated and taxable goods, tax authorities cannot observe the allocation of inputs across activities and refunds must be determined by an apportionment formula. We prove that no formula can deliver both profit neutrality and wedge neutrality except under knife-edge restrictions on technology. Instead, every formula chooses which distortion to impose. Exploiting a 2019 Kenyan reform that increased refund generosity and differentially shocked both margins, we show that formula-induced distortions have sizable consequences for firm production. Firms exposed to a lower marginal distortion reallocated production towards zero-rated goods, increased exports, and expanded overall production. Within four years, the rise in taxable activity turned the reform's annual fiscal balance positive. The reform was likely welfare-improving, yet a third of misattribution is irreducible under any implementable rule, which points to high returns from measuring input use directly.

A Worker’s Worth
with Paolo Falco and Andreas Menzel
Recipient of IGC grant ETH-24324
Field work in progress

Policy Work

Reforming Excise Taxation on Alcohol Products in Kenya and Reforming Excise Taxation on Tobacco Products in Kenya
with Alex Oguso, James Ochieng’, Nathan Remcho & Hellen Chemnyongoi

Media

Dagbladet Information: Comment on Climate-tied Aid (Danish)