The fungibility of health aid: reconsidering the reconsidered

Abstract

Lu et al. found that health aid displaces domestically-raised government health expenditure, which renders health aid at least partially fungible. These findings are questioned in The Fungibility of Health Aid Reconsidered. Van de Sijpe’s emphasis on disaggregating on- and off-budget aid is a valid contribution, although his empirical conclusions are overstated. We re-evaluate the data he criticises and find they sufficiently capture on-budget aid. To re-measure health aid fungibility, we update the Lu et al data, adding 23 countries and four years of data. Despite the confidence we have in these data, we employ two estimation specifications, each of which addresses the measurement error discussed by Van de Sijpe. The extended data and alternative methods show that development assistance for health channelled to governments remains significantly fungible.