Income Inequality: Still Not A Big Deal
Andrew Sullivan, whose blog I’ve just started reading again, points to this Economist piece on income inequality in the United States. While usually a great source for economic reporting, this piece falls into some common traps with welfare analysis and sounds any number of useless alarms.
The common perception is that income inequality in the US is bad and only getting worse, however a closer look at the actual numbers doesn’t paint nearly as bleak a picture as we might be led to believe. The US Census Bureau put out Income Inequality in the United States: 2002, and their most recent data do not seem to indicate a huge leap in income inequality.
The United States Gini coefficient is .0462 according to the above paper in 1979 the US Gini was 0.406*. Like golf, low numbers are preferable from a Normative Economic perspective (more on that later). However, methodological changes make data previous to 1993 not exactly comparable to current data. This is sort of a niggling detail, and a fact largely ignored by pretty much everyone. However, it doesn’t seem like the methodological changes were really large enough to have much of an effect, so I suppose it’s not a big deal to go right ahead and ignore them.
Table 7** from the above report shows that the Gini by quintile for before-tax income is 0.448, slightly but not impressively lower than the above. However, the United States has a fairly progressive taxation structure, and people’s welfare is determined not by their gross income but by their disposable income. The same table shows the Gini for after-tax income to be 0.426, which is significantly lower than 0.462 and lower than the 2001 number of 0.434.
But wait! There’s more! Disposable income is not the only thing that contributes to total welfare, which is really the disparity everyone in the Normative community claims to be interested in, so we’ll have to look at yet a more comprehensive number. Once non-cash transfers and the benefits of employer-subsidized health-care are factored in the Gini drops slightly to 0.421. If you factor in the benefits of Medicare and Medicaid, the Gini drops to 0.405. If the value of home equity returns is considered, the Gini declines to 0.400.
Essentially, the more comprehensive a measure of welfare is used, the less inequality we see. What’s more interesting, and really does expose the lie in “tax cuts for the rich”, is that the top quintile of income earners lost statistically significant shares in all income categories except gross income. Further, while the bottom two quinitiles remained largely unchanged (probably because they really don’t pay taxes at all), the shares of aggregate income in the third and fourth quintiles made statistically significant gains over 2001 in all categories except gross income. What does that mean, kids? That middle class Americans have gained a lot over the last couple of years.
That we have an “inequality crisis” in America is a fallacy. This has been a boilerplate issue for liberals and leftists for a long time now, and is unlikely to go away, but it is important to remember that their claims are not really supported by much in the way of evidence.
Later in the week: Marginal tax rates and government outlays, plus why Normative Economics is fundamentally flawed.
*Table A-3: Share of Aggregate Income Recieved By Each Fifth and Top Five Percent of Households 1967-2002
** Percentage of Aggregate Household Income Recieved By Income Quintiles and Gini Index by Income Definition.
