Market reform, contestability and determinants of the Maize Board-Miller marketing margin in the South African maize industry.

Abstract

The dissertation analyses market reform in the South African Maize Industry at two levels. Firstly, aspects of the theory of contestable markets are used to analyze maize grain marketing reform and identify what measures are appropriate to promote contestability (ease of entry and exit, and potential rather than actual competition as a means of constraining uncompetitive behaviour). A brief history of the South African Maize industry since 1931 shows a highly regulated grain marketing system which lacked contestability. Recent amendments to the Maize Marketing Scheme such as abolition of controlled prices for maize products at retail/merchant (1960) and wholesale/milling (1971) levels; movement away from limited registration of maize processors (1977); and abolition of the statutory single channel fixed price pool scheme and storage control allowing "free" trade within the domestic market (1995), promote market contestability. Contestability can be further enhanced by reduction of sunk costs through possible on-farm storage and handling facilities (bunker storage, plastic tunnels, steel and concrete silos) and the leasing of existing silo space (possible excess capacity and alternative uses) by producers and maize traders. The maintenance of some statutory powers for the Maize Board (single channel exports, compulsory registration and levy payments) still limits contestability. Secondly, the determinants of the Maize Board - Miller (MBM) marketing margin between 1977 - 1993 (period defined by data limitations) are identified using Ordinary Least Squares (OLS), Three Stage Least Squares (3SLS) and Principal Component Analysis. The MBM margin was positively related to miller market power (proxied by industry conjectural elasticity), the real miller maize meal selling price, real variable processing costs, and a change in Maize Board maize grain pricing policy after 1987 (export losses reflected in low real net producer maize price). The estimated conjectural elasticity was low, indicating competitive conditions, although concentration ratios indicate entrenched, but falling, market power. The main component of the Maize Board-Miller (MBM) marketing margin for 1977 - 1993 was variable processing costs. The real consumer price of maize could be reduced via lower real processing costs of maize meal, possibly with the removal of fixed administered prices of inputs (like electricity) and moderation of real wage demands in negotiations between trade unions and millers. Foreign exchange and import controls may, however, raise input costs if the Rand should continue to weaken. The increased number of "bosmeulens" (small mills not registered with the Maize Board and using relatively inexpensive technology not having substantial sunk investment) entering the market means that mill sunk costs may be less of a deterrent to entry in future. The 1987 Maize Board pricing policy change captured the effect of input price risk on the MBM margin indicating a significant effect of past maize pricing policies on this margin. The Maize Milling Industry appears to be competitive (low industry conjectural elasticity) over the study period, although the oligopoly component still contributes significantly to the MBM margin. Miller market power may possibly be exerted on other products (e.g. wheat) as white maize may be seen as a loss leader. This interrelationship between maize and other grains in processing is an area for future research.