A recent wave of randomized trials in developing countries has shown demand for public services to be highly price elastic, with even low user fees dramatically reducing utilization (Michael Kremer & Alaka Holla 2009). Critics have questioned the relevance of these partial-equilibrium exercises for policymaking on the grounds that they ignore potential general equilibrium effects (Daron Acemoglu 2010). In the education context, a large literature on school segregation and sorting has shown that relatively privileged parents may seek to disassociate themselves from marginal social groups (Thomas J. Nechyba 2006, Sarah Reber 2011). In this paper we explore whether these social interactions generate general equilibrium effects, which might obscure or reverse the downward-sloping demand curves anticipated by partial equilibrium analysis of fee abolition.
In January 2003, the Kenyan government announced the abolition of all school fees in public primary schools. We estimate the effect of this nationwide reform by comparing public and private schools at the primary and secondary levels, before and after the reform— analogous to a triple-differences estimator. Confirming anecdotal reports by James Tooley (2009), we find that free primary education (FPE) was an ineffective tool to increase school participation. Instead, net enrollment in government primary schools remained unchanged over the 10-year period from 1997 to 2006. Meanwhile, both net enrollment and fee rates in private schools more than doubled.
We interpret this perverse demand response to a fall in price as prima facie evidence that FPE led to a (perceived) decline in public school quality. There are at least three potential mechanisms that may explain this shift in perceived quality, all of which would result only from a price decrease affecting a large number of children and might elude detection in small-scale experiments: (i) an influx of new students and concomitant deterioration in the composition of pupils; (ii) a change in underlying school productivity or value added, possibly due to a change in the accountability framework under free service provision; and (iii) changes in overall financial resources due to lost fee revenue. By linking FPE to reduced school quality, each of these mechanisms may obscure the downward-sloping demand curves predicted by partial equilibrium analysis.
