The procedure used by Van Praag and Kapteyn to test the theory that the individual welfare function of income is a lognormal distribution function is critically evaluated. It is shown that random data give the same test results as the income levels actually collected from real consumers. Thus it is doubtful whether Van Praag and Kapteyn's results, even although they are based on data collected from as many as 12000 consumers, provide any support for the lognormal model. It is concluded that more powerful methods are required to test the lognormal model thoroughly.

longnormal models, welfare economics, welfare functions
dx.doi.org/10.1016/S0014-2921(78)80009-9, hdl.handle.net/1765/12558
ERIM Article Series (EAS)
European Economic Review
Erasmus Research Institute of Management

Wierenga, B. (1978). The individual welfare function of income: a lognormal distribution function?. European Economic Review, 11(4), 387–393. doi:10.1016/S0014-2921(78)80009-9