Drought and flood relief programs distributing free seed and fertilizer are common in southern Africa, but little is known about their efficacy. This study  mmarizes the impacts of input relief programs in Zimbabwe, based on data from surveys conducted in 2004, following two consecutive drought years.

The analysis reveals substantial opportunities for improving these programs. First, targeting of beneficiary households must be improved. There was little difference between recipients and nonrecipients in terms of household characteristics, composition, poverty level etc. Many householdsreceived inputs from more than one NGO. Targeting can be improved through better sharing of information, and by using simpler selection criteria (eg, ownership of livestock) to identify beneficiaries. Contrary to common perceptions, farm communities tend to be reasonably successful at maintainingseed stocks even after multiple years of drought. Correspondingly, the delivery of free seed did not contribute to an increase in planted area. If seed i  provided, more emphasis is needed on quality   control and proper labeling. Also contrary to common perceptions, distribution of small quantities of   fertilizer offered substantially higher returns than distribution of seed. The application of as little as 10 kg of nitrogen per hectare contributed subst ntially to food security in drought-prone regions. This study also compared three alternative input distrib ution methods: direct handouts of seed andfertilizer, seed fairs, and the use of vouchers redeemable at retail shops. While direct handouts are logistically the easiest method (and the most widely used), voucher-based programs linked with retail shops potentially offer the greatest development impacts.