The problem addresses in this article is the bias to income and expenditure elasticity estimates caused by inattention to measurement error and unobserved heterogeneity. We gauge these biases by comparing cross-sectional, pseudo-panel and true panel data from Polish and American expenditure surveys.

Qualitative answers given in the INSEE conjuctural surveys (1972-1994) by approximately 2500 households during two consecutive years aloow us to test the rationality of price and unemployment expectations on panel data.