ENGEL FUNCTIONS, PANEL-DATA, AND LATENT-VARIABLES

A system of consumer expenditure functions is estimated from Norwegian household budget data. Specific features of the approach are: (i) Pan el data from individual households are used, which offer far richer op portunities for identification, estimation, and testing than cross sec tion data. (ii) Measurement errors are carefully modelled. Total consu mption expenditure is modelled as a latent variable; purchase expendit ures on different goods and two income measures are used as indicators of this basic latent variable. The usual assumption of no measurement error in total expenditure is clearly rejected. (iii) The distributio n of latent total expenditure across households, and its evolution ove r time, is estimated and important properties tested. (iv) The distrib ution of individual differences in preferences, represented by individ ual time invariant latent variables, are modelled, estimated, and test ed. (v) We test the hypothesis that preferences are uncorrelated with total consumption expenditure, which is basic to virtually all cross s ection studies of consumer demand functions.