摘要：Appeals for targeting agricultural R&D in developing countries more explicitly to the needs of poor smallholders in marginal areas are often countered by arguments of efficiency. Others, however, taking a political economy perspective, argue that there is a bias in the selection of agricultural R&D projects towards commercial farmers in the better agricultural areas. In this paper, we try to bring the two perspectives together and illustrate how they interact. We analyze R&D budget allocation assuming two distinct R&D opportunity curves one for poor smallholders and one for large commercial farmers. We find that, in contrast to the actual allocation of research resources, an efficient allocation (i.e., equalizing the marginal rate of return between the two portfolios) would increase the budget for smallholder projects by 29% and decrease that of commercial farmer projects by 26%. Total economic welfare would increase with 2.1% and there is a modest distributional effect: smallholder welfare increases by 14% and commercial farmer welfare decreases by 11%. In other words, more efficiency in the selection of agricultural R&D projects favors poor smallholders. We also analyze the effects of a premium on research benefits accruing to smallholders and find them to be limited. Rather than trying to shift the R&D opportunity curve artificially, we conclude that policies that can help to increase the profitability of research investments across the board (such as investment in infrastructure, development of markets, and education) and hence shift the R&D opportunity curve for poor smallholders outwards should be pursued more vigorously.
