Abstract
This paper examines Australian retail demand for meat for the post‐war period 1949‐50 to 1978‐79. Several systems of demand equations are estimated, each system (comprising equations for beef, mutton, lamb, pork and chicken) being derived from an underlying static utility function. Model selection procedures are used to determine the preferred specifications). The validity of the utility theory approach is investigated by: testing for the presence of autocorrelation; examining the monotonicity and convexity properties of the utility functions; and comparing the estimated price and expenditure elasticities with those of other Australian studies. Serial correlation does arise as a problem, but other results provide support for the theory.