An Economic Evaluation of Alternative Maize Policies

Abstract

MAIZE IS THE SINGLE MOST IMPORTANT CROP in South Africa and the second biggest agricultural industry. It contributes 20 per cent to the gross value of agriculture. It employs an estimated 32 000 farmers and feeds about 25 million people, apart from being an essential input to the livestock industry. Maize is currently marketed through a single-channel fixed price scheme but there has been much debate about an alternative scheme, on account of the costliness of the current programme. In 1983, the Jacobs commission of enquiry into the maize industry suggested a two-tier marketing system with quotas to restrict local supply. But, on 13 September 1985 the Minister of Agriculture rejected the Jacobs recommendations and requested that the National Marketing Council formulate an alternative policy (Mielies/Maize, November 1985, p.11). Various policies for the maize industry are evaluated in this study which aims to be descriptive rather than prescriptive. No one policy is singled out as the best policy. Rather, a wide range of alternative policies are analysed and their effects on various aspects of the economy are given. Estimates are made of prices, quantities, rents, and social costs for each policy. No claim is made for the accuracy of these figures. The intention is rather to provide a means of comparison. The purpose of the study is to evaluate the effect of different policies on the current situation in the maize industry against the background of oversupplied world markets. The present export demand situation was incorporated in the model. TechniqueA linear programming model was developed to determine the aggregate and regional effect of alternative policies. The model incorporates negative sloping demand curves and allowance is made for producer risk aversion. The matrix is comprised of 268 rows and 822 columns with 3 335 non-zero elements. Norton and Solis (1983) have used such an approach in analysing Mexican Agricultural policies. Nieuwoudt et al. (1976), used it for comparing US peanut policies, and Ortmann (1984) used the technique to investigate the viability of ethanol production in South Africa. The 1987 SAJE v55(4) p356model for the competitive case, assuming consumer and producer surplus are maximized, can be expressed as followsThe first term measures total welfare, which is the integral of the linear product demand function P = A -BWX, where W is an n x n diagonal matrix of average yields, A and B vectors of coefficients and X is an n x 1 vector of aggregate hectares generated endogenously within the model. Production costs excluding labour are deducted in the second term [C'X] where Ω is a vector of production costs. Costs associated with risk are deducted in the last term, where Ù is a variance-covariance matrix of gross incomes per hectare, is an aggregate of risk aversion coefficients of all farms for region i and N the number of major regions. SupplyThe maize producing area of the country was divided into 96 homogeneous production districts. The four main alternative crops in these districts a...

Analytical Tools in Production Economics

“…For example, Hazell et al (1983), Simmons & Pomareda (1975) and Kutcher Scandizzo (1981) reported 0 values ranging from 0,5 to 1,5 for sector models in Mexico and Brazil. Nieuwoudt et al (1976) used a value of 2 because it gave the best solution in simulating peanut production in the USA, while Ortmann & Nieuwoudt (1987) and Nieuwoudt and Frank (1987) reported 0 values of 0,25 and 0,5 in simulating regional sugar-cane and maize production, respectively, in South Africa. In these sector models 0 is theoretically an aggregate of risk aversion coefficients of all farmers in the region.…”

“…crop insurance) and different objective functions (Haze11, 1982:386;Young, 1979Young, :1066. Nieuwoudt et al (1976), Ortmann & Nieuwoudt (1987) and Nieuwoudt & Frank (1987) attached little significance to the value of 0 , which was used only to fine-tune the predictive ability of their models.…”