Ed Paisley — Heritage Weighs into the Inequality Discussion with Some Problematic Data Analysis

This piece and its underlying data analysis have three fundamental flaws:

• The Gini coefficient they are referencing is of income and does not factor in the effect of taxes or transfers. Thus, the measure they are using explicitly misses the impact of the policies that they claim are ineffective.
• They are suffering from one of the cardinal sins of data analysis: omitted variable bias. More populous areas also tend to have higher inequality, at least in part because higher density allows for higher incomes. Furthermore, cities and urban areas also tend to elect more progressive leaders for a variety of reasons. Thus population density is the omitted variable. They fundamentally misunderstand (or at the very least ignore) the relationship between inequality and population density.
• Finally, they are factually incorrect to say the 1980s and 1990s are emblematic of the very laudable notion that “a rising tide lifts all boats.” As can be seen in the figure below, median hourly compensation has been essentially flat since 1970 despite the fact that per capita economic growth more than doubled over the same period.

 Freshman mistake or lying with statistics?

WCEG
Heritage Weighs into the Inequality Discussion with Some Problematic Data Analysis
Ed Paisley