NCPA - National Center for Policy Analysis

Flaws in Using Income Data to Measure Inequality

August 18, 2014

Income data is flawed, because it provides a snapshot of just one year in a person's life, writes Alan Cole, economist at the Tax Foundation. Moreover, prices and costs of living vary considerably among regions, making income a poor predictor of a person's actual standard of living.

Cole delves into some of the realities that make income a poor measurement of inequality:

  • Because income is reported annually, it is not necessarily an indication of a person's wellbeing. Income tends to move in an arc over the course of a person's life; it starts low but rises until a person reaches retirement age. Analyzing just one year of income data obscures the gains that are likely to come to many earners.
  • Sometimes, income data is used to identify poor areas. But because young students tend not to be earners, these studies often show that college towns are the poorest cities in America.
  • A study of prices across geographic areas makes clear that some regions are far more expensive than others; San Francisco, for example, is 23.5 percent more expensive than the United States as a whole, according to the Bureau of Economic Analysis.
  • Areas that may seem low-income are not actually worse off after adjusting for purchasing power. For example, Kansas has much lower incomes than New York, but it also has a considerably lower cost of living. After adjusting for those differences, Kansas actually has higher average incomes.

Additionally, Cole explains that not all income is counted in IRS income data. For example, the IRS only measures capital gains as income when realized. This distorts income figures because assets may be held for many years before being sold and, when sold, can create misleading numbers. For example, a study by Robert Carroll determined that 50 percent of millionaires over a nine-year income sample were millionaires for just one year. He attributed that fact partly to the realization of capital gains.

Thus, looking at income data over one year, rather than over time, can paint a very different picture of income -- and consequently income inequality -- than is the case.

NCPA Senior Fellow Richard McKenzie discussed these realities in a recent paper. He noted that official inequality measures are misleading, because a person's income gains are reflected in the upper income bracket that he reaches, not the lower-income bracket that he leaves behind.

Source: Alan Cole, "Income Data is a Poor Measure of Inequality," Tax Foundation, August 13, 2014. 


Browse more articles on Tax and Spending Issues