As the U.S. economy has been recovering from the 2001 recession, there has been a growing level of chatter in the media about the “rich getting richer and the poor getting poorer.” To reach this conclusion, these reports rely on “income distribution analysis” that uses income data published by the Internal Revenue Service. As a result, these studies are really examining the distribution of tax returns and not of individuals. The problem is that tax returns are vastly different from one another and not controlling for those differences can lead to erroneous conclusions.


The basic tools used by economist to study income distribution were created over 50 years ago. In the intervening years, these tools have essentially remained unchanged. In a nutshell, economists sort the IRS data by income and then divide them into five equal parts–referred to as “quintiles.” By taking the difference between the average income in the first quintile and the top quintile can tell you whether or not income disparities are shrinking or growing. Or does it?
You may or may not be surprised to know that most income distribution analysis do not adjust for basic socio-economic differences. Many of these difference show up in our tax returns. If one is single you file as a “Single,” but if you are married you file as “Married Filing Jointly”–your filing status determines your standard deduction and your tax brackets to name a few items affected.
A recent book published by the non-partisan Tax Foundation sheds light on the extent of this problem that plagues income distribution analysis. The book finds that a person in the lowest quintile is more likley to be a young student working part-time to pay tuition while a person in the top quintile is more likely to be a dual-income family with children in their peak earning years. Is there a difference in income? I certainly hope so!
There have been many long-term trends that lead to a number of biases in many income distribution studies. For example, the Tax Foundation study found that in 1960 35 percent of tax filers were single while 65 were married filing jointly. By 2002 these percentages nearly flipped with singles accounting for 59 percent of all tax filers and only 41 percent were married filing jointly. Needless to say, that is a major change in the underlying socio-economic picture.
The implicit assumption in many news reports is that a growing income inequality is inherently a bad thing. But, in reality, that is an open question–especially if the perceived inequality is really just a statistical illusion.