skip to content

Keynes Fund


Summary of Project Plan

A substantial fraction of households’ income is devoted to housing. In addition, residential location has also implications for access to a variety of local amenities and public goods, including the public school catchment area.

As a consequence, parents are willing to pay a substantial premium to secure a place for their children in a high performing public school. For instance, exploring random assignment of external markers to revise student test scores in the United Kingdom, Battistin and Neri (2017) show large housing price gains in areas in which there were artificially improvements in English test scores. Such areas also experienced lower unemployment and new homeowners from higher socio- economic backgrounds.

The usual view of effects of school integration (e.g. via legal or financing reforms) on education and inequality is that they have redistributive effects, at least in the short run: those in better off (resp. worse off) neighbourhoods lose (resp. gain). Under suitable conditions, the aggregate macro effect in the long run is positive, as the gains from reducing inequality are positive (Benabou, 1996; Gall, Legros and Newman, 2015; Mookherjee, Napel and Ray, 2010). In Benabou (1996), the long run effects are Pareto improving, as global complementarities between learning of the poor and the rich and rising education of the poor also raise incomes of the rich.

One qualification to this is that these results depend on specific assumptions on the nature of complementarities, and also on the proportion of population that is educated. For instance, Mookherjee, Napel and Ray (2010) show that if the proportion of educated is less than a half and there is a global substitutability between education of the rich and the poor, the macro effects go in the opposite direction.

In this project we will be focusing on another important assumption made by previous literature, that housing prices are exogenous and unaffected by the reform. Extensions of the Mookherjee, Napel and Ray (2010) model to incorporate choice of neighbourhoods by households and endogenous housing prices show that integration can give rise to changes in housing prices which offset the direct effects of the reform. Under specific assumptions, the reform can result in a Pareto inferior outcome. E.g. with a fixed supply of housing in each neighbourhood, house prices in rich (resp. poor) neighbourhoods will fall (resp. rise) following a move towards school integration, which results in a complete offset of the direct effects on education cost. Households that educate their kids are unaffected, while those that do not are worse off. If wages are unaffected by the reform (i.e. no global externality through the labour market), we have a Pareto inferior allocation.

We will therefore investigate the implications of endogenous location decisions and house prices theoretically. We will then seek empirical evidence on indirect effects of integration on house prices and the extent to which these offset the direct effects.



Dr. Tiago Cavalcanti and Prof. Dilip Mookherjee


Dr Tiago Cavalcanti is University Reader in Economics at the University of Cambridge, Fellow of Trinity College, Adjunct Professor at Sao Paulo School of Economics-FGV. His research expertise is Macroeconomics, Growth, Economic Development.


Professor Dilip Mookherjee is Professor of Economics and Director, Institute for Economic Development at Boston University. His research interests include Development Economics, Contract and Organization Theory.


Share this Project