Price formation in the Brazilian soybean industry 1982/1989

Este trabalho procurou analisar a formação de preços em vários níveis de mercado de soja. O sentido de causalidade foi testado pelo teste de Sims e as elasticidades de transmissão de preços foram estimadas. Os preços externos lideraram os internos. No mercado domestico o atacado liderou os outros níveis. Relação bicausal foi encontrada entre preços de farelo no mercado externo e no atacado, possivelmente devido a importância do Brasil no mercado mundial de farelo. Os fatores determinantes dessa liderança parecem ser o poder de mercado e o acesso as informações. O comportamento do varejo sugere o uso de uma politica de markup. As elasticidades de transmissão de preços não diferiram de um com exceção das relações entre preço de farelo no mercado externo e no atacado e entre preço de grão no mercado externo e ao produtor rural, ambas maior que um. Isso deve-se a rigidez dos custos de exportação. O teste de assimetria de houck indicou que os vários níveis de mercado respondem mais aos acréscimos de preços do que aos decréscimosThe objective of this paper is to analyse the price formation process at vairous market levels in the Brazilian soybean industry. The period analysed was 1982 to 1989. Initially, the causality directions were analysed, through the Granger-Smis's test, and then the elasticities of price transmission were estimated. The internacional prices were show to lead the domestic prices. At the domestic markets, the wholesale prices tend to lead the other prices. Bi-directional causality directions were found between the wholesale meal price and the international meal price, possibly due to the importance of Brazilian exports to the meal international market. The price leadership by international markets and wholesale level may arise from arket power and availability of market information. The retail behavior suggests that this market level may use a markup pricing policy. The elasticities of price transmission showed integral transmission (unitary elasticity)between the various market levels, except between the international meal price and the wholesale meal price, and between the international grain price and the farm grain price, both greater than one. These high elasticities arise, possibly, from rigidity of the exports cost. The number of lags for price transmission was between one and four (months). The Houck's asimmetry teste indicated that the various market levels respond more to price increases than to price decrease