Rejection without falsification on the history of testing the homogeneity condition in the theory of consumer demand

Five-a-day, a price to pay: An evaluation of the UK program impact accounting for market forces

“…Considering models by income quartile only, rejection is less frequent (both restrictions are met in 13 models of 16) and the R-square is generally higher. These rejection rates with large samples as those of the current study are relatively low compared to the literature, where rejection of homogeneity is a common finding (Keuzenkamp and Barten, 1995). Theoretical constraints were imposed in all models at the simulation stage.…”

Living up to expectations: Estimating direct and indirect rebound effects for UK households

“…Alternatively, an unrestricted model can be estimated for both the first and second stage and the homogeneity and symmetry restrictions tested. It is common for these restrictions to be rejected in empirical studies (Keuzenkamp and Barten, 1995). For example, the foundational LAIDS study by Deaton and Muellbauer (1980) rejected these restrictions.…”

Revisiting Aggregate U.S. Meat Demand with a Bayesian Averaging of Classical Estimates Approach: Do We Need a More General Theory?

“…The data strongly discount the homogeneity restriction, with a posterior probability and odds ratio of 0.01, and strongly favor the symmetry restriction, with a posterior odds ratio of 499. 7 This homogeneity rejection and symmetry acceptance seems to be an unresolved puzzle of demand systems (see Pope, Green, and Eales 1980;Keuzenkamp and Barten 1995;Deaton and Muellbauer 1996;Buse 1998). We also checked the negativity conditions for all 1,048,576 models evaluated at the earliest, mean, and latest budget shares and 99%, 100%, and 100% of the models satisfied the negativity condition.…”