Application

School places

Which procedures best meet preferences, what parents actually do and what early tracking does.

under 3%

instead of 15% placed only at their fourth-ranked school or lower (Budapest)

5 km

gain from joint allocation, measured as a school that much closer

540

parents and students in the school choice experiment

Findings Materials Publications

Starting point

The problem

At the transition to secondary school, popular schools are often oversubscribed. How places are then allocated determines how many families get their preferred school, whether strategic behaviour pays off and who ends up disadvantaged.

Approach

What the research contributes

A comparison of allocation mechanisms using administrative data and experiments, and an analysis of what tracking into school types means for achievement and opportunity.

What the research shows

Mechanisms

Which mechanism best meets preferences?

The widely used Deferred Acceptance mechanism makes truthful reporting the best strategy, but in Budapest places almost one child in seven at their fourth-ranked school or lower. A mechanism that explicitly minimises poor placements reduces this share to under three per cent.

Games and Economic Behavior, 2023

Experiment

What parents actually do

In an experiment with 324 parents and 216 students, all mechanisms were frequently manipulated, including the one considered strategy-proof. The advantages of the more efficient mechanism nonetheless largely persisted.

Working paper, 2026

Districts

Joint allocation across districts

When school districts merge their allocation, everyone gains on average, especially families from small districts with sought-after schools. In Hungary the gain is equivalent to a school five kilometres closer.

Games and Economic Behavior, 2024

Tracking

Early tracking

Attending an academic-track school improves achievement and university aspirations, at least as much for disadvantaged children as for others. These children are less likely to get there; early and rigid tracking can therefore amplify inequality.

Working papers, 2024 and 2025