Improving the efficiency targeting of Malawi's farm input subsidy programme: Big pain, small gain?

The Farm Input Subsidy Programme (FISP) in Malawi was introduced in the 2005/2006 season against a background of bad weather affecting production and prolonged food shortages. Vouchers are distributed empowering eligible farmers to exchange them for fixed quantities of inputs at subsidized prices. Since its inception, there has been a debate at national level about whether the FISP's potential has been fully exploited, with policy makers exploring options to improve the programme. Proposals include targeting efficient and productive farmers to maximize returns. In this paper, we evaluate the effectiveness of these proposed changes to the existing FISP design by utilizing two waves of the LSMS-ISA survey merged with climatic data. We estimate how the national demand for agricultural inputs varies according to a variation in the targeting criteria by means of a two-stage demand system. Then, we identify more efficient farmers by means of a stochastic frontier approach. We observe a mismatch between voucher recipients and efficiency, with approximately 60% of vouchers being allocated to the three bottom quintiles of efficiency. This mismatch is observed also at the spatial level with more vouchers going to districts characterized by less efficient production. While concerns on the distributional impacts of the new criteria are discussed together with some suggestions for spatially diversifying the structuring of the policy and incentivizing crop diversification, our results highlight a high substitutability of commercial with subsidized inputs by new eligible farmers. Consequently, simulating the targeting policy variation we obtain an outcome that would lead only to a limited increases in predicted food expenditure ranging from 0.27% to 0.8% and maize production from 0.2% to 1.3. Scope for analysing different adjustments in the functioning of FISP are, thus, proposed to policy makers