Several farm sector econometric models are reviewed hitially and the aggregation problem highlighted. A t!&ty-eyuation model of the Australian agricultural sector is specified m which farm output, stocks and exports and the domestic demand for farm products are endogenous, as well as farm., export and retail prices. Disaggregation is into food and non-food componeuts of unprocessed output, and the processing o f food is traced through to final demand. The model is estimated by a modified 2SLS procedure using quarterly data covering the period 1960-1970. In a recent review of the state of the art of econometric modelling of the agricultural sector, King [9] drew attention to the relative paucity of models of the agricultural sector as a whole. A major question confronting builders of econometric models of the agricultural sector has been the choice of a level of aggregation. At one extreme is a model such as that of Egbert [2] which concentrates on explaining annual aggregate supply and demand for the U.S. agricultural sector as a whole without any consideration of the structure of component industry or commodity groups. The endogenous variables in Egbert's model are total agricultural output, domestic consumption of agricultural products, farm prices and the level of farm stocks; imports and exports are assumed to be exogenous. Models such as Egbert's may be relatively straightforward to specify and estimate, but their usefulness for policy analysis and forecasting is limited by their highly aggregative nature.At the other extreme, an agricultural sector model may aggregate upwards from individual commodities, such as in Cromarty's El] pioneering work of the late 1950s. He specified supply equations, and demand equations for consumption, inventories and government stocks, * The research on which this paper is based was supported by a grant from the 157