Cash transfers' effect on government support: the case of Fiji

Abstract

While some scholars have found that government post-disaster assistance supports the incumbent, others have shown that incumbent effects among beneficiaries are imperceptible or negative. This article contributes to this debate by using a regression discontinuity design of households affected by Tropical Cyclone Winston in Fiji to show that the type of assistance provided is an important variable in understanding the effects of aid on perceptions of the government. Residents of Fiji who received a post-disaster cash transfer are up to 20 per cent more likely to be very satisfied with the government than are those who did not. The probability further increases if the cash transfer was provided along with in-kind benefits or vouchers, but it is not affected if beneficiaries were also encouraged to use their own pension savings. This paper provides evidence in favour of the 'attentive citizen' theory by demonstrating that beneficiaries actively appraise government responses; it also reveals possible effects of elite capture on the relationship between the government and beneficiaries.