One more question remains after my recent discussions about U.S. income inequality: Does reducing income inequality via government taxes and transfers hold back economic growth?
The answer may surprise you.
As you may recall, over the past two weeks, we examined trends over the past 90 years and discussed the causes of rising income inequality. I also pointed out that economic growth is the primary driver of income inequality, but that increases in income inequality do not increase growth.
Income inequality: before and after government
To summarize income distributions, economists use Gini coefficients, which range from 0 (perfect income equality) to 1 (complete income inequality). The Organization for Economic Cooperation and Development (OECD) provides Gini coefficients for industrialized countries before and after governments levy taxes and make transfer payments.
Let’s take a look at the numbers and see what they tell us. (Note: My thanks to Andrew Twite, graduate student at the Hubert H. Humphrey School of Public Affairs, for writing to me after reading my Nov. 16 column and reminding me of these data. He kindly put together the figures that appear below.)
The chart below shows two Gini coefficients for each OECD country: one computed before taxes and transfers (in red) and the other after taxes and transfers (in blue).
Click on chart to enlarge
The countries here are ranked from least pre-tax inequality to highest pre-tax inequality. By this measure, the United States is about average in terms of income inequality, with the OECD average Gini at 0.45 and the U.S. Gini equal to 0.46.
Now, let’s take the same data and rank the countries by their post-tax inequality from least to most:
Click on chart to enlarge
The U.S. now ranks at the top, tied with Portugal for the highest income inequality in the OECD. Further, if we compare the red and blue bars, we find that the U.S. is near the bottom in terms of reducing income inequality via taxes and transfers. Only Korea does less to reduce income inequality, and they already start with a more equal income distribution than does the U.S.
Does Big Government hurt economic growth?
One quick response to the charts is: “Sure, the U.S. has higher income inequality, but it probably grows faster, too.” But that view is not true.
Peter H. Lindert, a professor of economics at the University of California, Davis, addressed this question in his 2004 book, “Growing Public: Social Spending and Economic Growth since the Eighteen Century.” He summarized his findings in his 2004 Clemens Lecture (PDF) at the College of St. Benedict and St. John’s University.
Lindert analyzed data on taxes, transfers and growth in the industrialized countries for three different periods: 1880-1930, 1962-1981 and 1978-1995. His central conclusion:
“Larger welfare states have not had any net cost, either in terms of economic growth or in terms of budget deficits. Had our politics permitted it, we Americans could have had the same economic growth of GDP — with more security, more equality, and longer lives (Clemens Lecture, p. 3).”
You might find this a pretty startling conclusion.
Lindert writes: “Facing facts like these, if you have always believed that the social programs should be very costly, you have a tough choice to make. You can choose to be strong. Stand by your beliefs. Don’t let the facts push you around. Or you can be a wimp. I am a wimp. I let the facts push me around. I react to them by asking ‘How could that be?’ ”
He provides a number of intriguing answers to that question. One that I find particularly interesting is that “high-budget welfare states, while taxing heavily overall, actually favor types of taxation that mainstream economists think are better for economic growth (Clemens Lecture, p. 4).” In particular, high-budget welfare states tax corporations at very low rates and rely more heavily on consumption taxes than on income taxes.
A Minnesota connection
Reducing income inequality via the welfare state does not slow economic growth at the national level. How about at the state level? Lindert addresses this issue, too:
“The costs are also invisible among the states of the U.S. States like Connecticut and California [that] have higher taxes and more generous social benefits. Idaho and Alabama don’t believe in taxes or social programs. That has been true for a century. So why haven’t Connecticut and California become as poor as the national average? Why haven’t Idaho and Alabama become as rich as the national average? Of course, places differ in other ways than just their views of taxes and welfare. So we need a deeper statistical analysis that gives many forces their due. I have done that, and so have other economists. The net cost just isn’t there. It’s zero. The welfare state looks like a free lunch, for the nation as a whole (Clemens Lecture, p. 4).”
Substitute Minnesota for Connecticut or California and the statement remains true. We can put together a reformed tax system that provides a strong social safety net, reduces income inequality and promotes economic growth. State policymakers should stop being so strong — and let the facts push them around a bit more.
A follow-up to last week’s column
Last week, I mentioned a Congressional Budget Office study of lifetime tax rates, but I forgot to include the companion study of income inequality that applies these data. The latter study is available here and contains the evidence for my statement that “tax policy was probably a wash in terms of income inequality.” I regret the omission.
Interestingly, a new paper on this very topic appeared on Monday. The paper (PDF), by Piketty, Saez, and Stantcheva uses the income data I discussed in my last two columns to more carefully parse the relationship between top marginal tax rates and the share of income going to the top 1 percent of income earners. They find that cuts in top marginal tax rates increase the income share of the top 1 percent in both the U.S. and in the U.K. Further, they find that higher marginal tax rates across countries are associated with lower shares of income received by the top 1 percent.
So, perhaps tax policy contributed more to income inequality than I thought. We’ll have to wait and see where the research leads us.