Gains from crop diversification in the Sudan Gezira scheme

Land allocation in Gezira irrigation scheme (the Gezira) has, for the most part, been based on the political judgement of the government rather than economic criteria. This paper determined optimum crop combinations for the Gezira using the mean-variance (M-V) model. This large-scale gravity-irrigated project which was started for export cotton cultivation on 27,372 feddan (11,496 ha) in 1927 has now expanded to over two million feddan, and has a multi crop production system, Sources of production uncertainty include government control of prices, often announced after planting, and crop area in favor of cotton. Diversification was analyzed under various levels of risk using, in contrast to what is in the reviewed literature, both cross sectional and time series farm level data. The cross-sectional data comprised of net farm returns of 100 farmers for the 1987/1988 season, and the time series data were average net farm returns from the same area from 1971 to 1988. Cropping decisions in the Gezira are mostly based on cross-sectional data from a sample of tenants in the preceding season. This paper estimated variance and covariance parameters of observed net margins of both sets of data and found significant differences in the results. The certainty equivalence results used to evaluate crop diversification showed that current policies of controlling crop acreage lead to inefficient land use; and that farmers diversify into wheat and groundnuts out of cotton when cross-section data were used; and from groundnuts to cotton and wheat when time series data were used. (C) 2001 Elsevier Science Ltd. All rights reserved.