Structural change and the demand for meat and fish in Norway

Structural change in the consumption of four meats and fish is investigated. A GARP test does not reject the hypothesis that a stable set of consumers' preferences could have generated the data. However, since the power of this test may not be sufficient to detect structural change, a dynamic switching almost ideal demand system is estimated. Prices appear to be endogenous in the system and instrumental variables are used. A gradual structural change during the 1980s is found. This change is biased towards chicken and fish and against beef