Although food aid may have important medium- to long-term effects, there is a glaring absence of empirical research on food aid dynamics. This paper applies vector autoregression methods to data from 18 countries over the period 1961-95. We find evidence that food aid has a pronounced J-curve effect on recipient country per capita commercial food imports but only negligible negative effects on recipient country per capita food production. The commercial export gains are primarily enjoyed, however, by the donors' competitors, revealing heretofore unrecognized positive pecuniary trade externalities associated with foreign assistance. Copyright 1999 by Oxford University Press.