The quote in the title comes from this oped piece by conservative political scientist James Q. Wilson. In terms of a quote, it is hard to beat. I can only imagine his next oped on terrorism: “The problem with terrorists isn’t violence.”
Larry Mishel has already straightened out the muddled economic thinking of dear old Professor Wilson. I’d like to use it to emphasize a point or two that I think bear repeating.
First, as Mishel correctly points out, there is a huge difference between what I like to call 80-20 inequality and 99-50 inequality. The difference between being in the 80th percentile and the 20th percentile can be significant. It’s the difference between an upper-middle class lifestyle in the suburbs and a working class job in the city or in rural America. In 2010, according to the Census, the ratio between the 80th percentile and 20th percentile household incomes was about 5. It had increased since 1967 from about 4, or about 25%. Much of this story has been told, and it’s the one that Wilson focuses on. Returns to education have increased. Families at the top are more likely to have two incomes. Globalization tends to increase wages at the top and decrease them at the bottom. And I don’t want to imply that this increase in inequality isn’t a problem. I just don’t think it is THE problem.
The problem is the increase in inequality between the median family and the top 1%. If you look at the ratio of the average income of the top 1% (much of which is going to the top 0.01%) to the 50th percentile, it has increased from about 8 in 1967, decreased slightly in the 1970s and then increased to about 18 in 2008 (down from the business cycle peak of 18.8 in 2007). That’s an increase of about 112%. That’s the problem. In real dollar terms, the average income of the middle fifth (the middle of the middle) has only increased from $40k to $50k from 1967 to 2010. Meanwhile the real dollar average income of the top 1% has increased from $325k in 1967 to over $900k in 2008 (it was almost a million in 2007).
Wealth, of course, is even worse. Mishel points out that over 80% of the increase in wealth went to the top 5%. The net wealth position of the median household has actually gotten worse over the last 30 years, mainly due to higher student loan, mortgage, and consumer debt.
The other point I would like to make is less well developed. On a superficial level, it does not appear that increasing marginal tax rates on the rich would have any effect on the pre-tax income of the middle. People work in the market, do as well as they can, pay their taxes, and then receive any transfers for which they are eligible. But I’m not sure that’s right.
A modern economy is a complex beast. A factory needs janitors to clean up, workers to man the machines, managers to plan production, and executives to strategize. Putting all of that together successfully creates value. That value needs to be split between workers, managers, executives, and shareholders. In that sense, it’s a zero sum game. Workers tend to have very little bargaining power (outside of a union), while executives tend to have the most. Putting more effort into bargaining will be in the executives interest if he can increase his pay. But increasing his pay will reduce the pay of the workers (and/or the shareholders). Similarly, putting extra effort into reducing labor costs will increase the amount available for executive bonuses.
But if marginal tax rates are high, then it’s possible that those executives will expend less effort in increasing their share of the pie. In this case, there will be more left for the workers.
Is there any evidence for this? If you look at overall GDP growth or GDP per capita growth, there is very little correlation with the top marginal tax rate of the top 1% or the top 0.01% in the post-WWII era. But if you look at the growth in average pre-tax income of the bottom 90% (controlling for the business cycle) there is a modest correlation with the marginal tax rate of the top 1% (r = 0.12) and a much more significant correlation with the top marginal tax rate of the top 0.01% (r = 0.55). This is far from a proof, but it suggests the idea might not be too crazy.