Cost-effectiveness Analysis of Demand- and Supply-side Education Interventions: the Case of PROGRESA in Mexico David P. Coady and Susan W. Parker* Abstract The paper is concerned with the issue of the most cost-effective way of improving access to education for poor households in developing countries. The authors consider two alternatives: extensive expansion of the school system (i.e., bringing education to the poor), and subsidizing investment in education by the poor (i.e., bringing the poor to the education system). To this end, the authors evaluate PROGRESA, a large poverty-alleviation program recently introduced in Mexico, which subsidizes education. Using double- difference regression estimators on data collected before and after the program for randomly selected “control” and “treatment” households, the relative impacts of the demand- and supply-side program com- ponents are estimated. Combining these estimates with cost information, it is found that the demand-side subsidies are substantially more cost-effective than supply-side expansions. 1. Introduction There is a vast body of literature that identifies the expansion of formal education as a key component of successful development strategies. However, there is still much disagreement about how best to allocate scarce public resources within the education sector. The policy debate is typically couched in terms of the relative importance of improved school quality vis-à-vis improved school access, with different researchers drawing very different policy conclusions from the same body of empirical evidence (Hanushek, 1995; Kremer, 1995). The quality versus access debate addresses the issue of the most cost-effective way of achieving a given total years of education. Yet concern for equity (i.e., the distri- bution of this education across different income groups) is a strong motivating factor underlying government intervention in the education sector. Since economies of scale imply that it is generally more cost-effective to locate schools in relatively densely populated areas, poorer households, who tend to be disproportionately located in remote areas, may face substantially higher private costs and, as a result, tend to acquire lower education levels.This may be further exacerbated by the relative impor- tance of credit market failures for the poor. In this paper we are specifically concerned with the issue of the most cost-effective way of improving access to education for poor households in developing countries.We consider two alternatives, namely: (i) extensive expansion of the school system (i.e., bringing education to the poor), and (ii) subsidizing investment in education by the poor (i.e., bringing the poor to the education system).To our knowledge, this is one of Review of Development Economics, 8(3), 440–451, 2004 *Coady: International Food Policy Research Institute, 2033 K Street NW,Washington, DC 20006-1002, USA. Tel: 202-862-6463; Fax: 202-467-4439; Email: d.coady@ifpri.org. Parker: División de Económia, Centro de Invesigación y Docencia Económicas, Carretera Mexico-Toluca No. 3655, Col. Lomas de Santa Fe, 01210 Mexico DF, Mexico; E-mail: susan.parker@cide.edu. Both authors would like to thank John Maluccio, an anonymous referee, and staff and colleagues at PROGRESA, CIDE, and IFPRI for helpful comments. Special thanks to Daniela Sotres and Mari Carmen Huerta for help will collecting the supply-side data. © Blackwell Publishing Ltd 2004, 9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA