The Marginal Utility of Income.

In both public economics and welfare analysis it is crucial to know how fast the marginal utility ofincome declines as income increases. One needs this parameter for cost-benefit analysis, for optimaltaxation and for the (Atkinson) measurement of inequality. We estimate this parameter using fourcross-sectional surveys of subjective well-being and two panel surveys. Altogether, we use data fromover 50 countries, and in a period extending from 1972 to 2005. In all six surveys we find aremarkably consistent relationship between reported well-being and income. We estimate theelasticity of marginal utility with respect to income at around (minus) 1.25. Thus, marginal utilitydeclines somewhat faster than in the case (assumed by Dalton and others) when well-being is linear inlog income. In the second part of the paper, however, we ask whether true well-being may not have aconvex relationship to reported well-being, making it less concave with respect to income. We findsome evidence of this, so that the correct elasticity of marginal utility with respect to income isroughly (minus) 1.15. These figures show that Dalton's theoretical assumption was not far from thetruth revealed by empirical evidence, and provide scientific estimates which can be used in allbranches of economics applied to public policy