Accounting for Changes in Tastes

Health concerns are thought by many to have shifted consumption away from red meats, though econometric evidence is mixed. Testing for structural change is difficult, especially when one time series is used for both estimating demand equations and testing their stability. Specification errors may suggest a shift where none has occurred. Using nonparametric demand analysis, we find that meat consumption patterns in the United States and Australia can be explained using only relative prices and expenditures. Only imposing particular functional forms can reverse the conclusion, suggesting that specification errors in econometric demand studies can account for findings of taste changes.