Libertarians and others from the right-ish side of the ideological spectrum have lately been building a compelling case against worrying about income inequality. Here's Will Wilkinson writing on Cato Unbound in October:
As far as I can tell, when most people are worried about economic inequality, they’re usually worried about inequalities in real standards of living — in the real material conditions of life. Income plays a crucial role in this story, but it’s a supporting role. An individual’s or household’s standard of living is determined rather more directly by the level of consumption than by the level of income...If we can directly inspect measures of consumption, and we can, then we probably should. When we do, we find that the familiar story of income inequality is rather misleading and that trends in consumption inequality are considerably less dramatic.
It is well known that the increase in income inequality in the past three decades has not been mirrored in reported consumption inequality. It is equally well known that the micro data used to measure consumption inequality has become increasingly disconnected from other measures of expenditure and savings, such as those derived from the national income and product accounts. This raises the question of whether the relatively small change in consumption inequality is due, at least in part, to systematic trends in measurement error. ... [Two empirical] exercises suggest that the increase in consumption inequality has been large and of the same magnitude as the change in income inequality.
Specifically, Aguiar and Bils find that there is a big discrepancy in the savings rate they derive from the Labor Department's Consumer Expenditure Survey and the better known figure published by the Commerce Department. While the latter figure tells a story of declining savings, the former has savings increasing over the last 30 years to over 20 percent of income -- something which seems quite implausible.
The discrepancy is most likely the result of people choosing to give out less information on questionnaires (which the Labor Department relies on). This non-response bias isn't the hardest data shortcoming to correct, but you have to know it's there. The Labor Department didn't address the issue until the 2004 survey, and the uncorrected data seem to be at the heart of the confusion here.
Aguiar and Bils created a model to try and correct for the measurement error on the many years of data already collected. They report that, with the adjusted data, consumption inequality most likely grew by about 30 percent between the 1980 to 2007, which is roughly the same as the increase in income inequality over that period.