The Effects of Competing Definitions of Income on Measured Trends in Income Inequality

• In Event: Inequality, Poverty, and Mobility Section Roundtables
  In Refereed Roundtable (60 minute): 17. Heterogeneity

Abstract

Trends in U.S. income inequality are surprisingly dependent on the definition of income
used. While sociologists have made contributions to our understanding of the causes of inequality;
the proper mathematical function used to measure it; and the limits of survey data as opposed to
administrative data, the definition of income has received comparatively little attention. At the same
time, the debate over income’s definition — happening largely in economics — remains muddled,
even as it exercises a significant influence on the degree of observed income inequality. This paper
reviews the extant literature and discusses the benefits, drawbacks, and empirical consequences of
several prominent definitions of income. It highlights two major problems common to most extant
definitions: the tendency to impute forms of income (such as housing services) in a way that double-counts
income; and, the tendency to count capital income as it accrues, rather than when it is
realized, which also overstates income, especially in periods in which stocks are overvalued before
a market correction. In contrast, the present paper puts forward a rigorous definition of income
as cash wages, profits, and realized capital gains: the money equivalent of new production, net of
the costs to society of producing it. Finally, this paper compares the empirical consequences of
this definition to those common in the economics literature. It demonstrates that many definitions of
income put forward by economists which ostensibly lower measured trends in inequality in fact raise
it, a finding that challenges a recent consensus

Author

• Griffin Joseph Bur, University of Wisconsin-Madison