There’s now a pretty active cottage industry of analysts pointing out that Piketty’s inequality analysis is incomplete because it leaves out the impact of taxes and transfers. It’s an important point; one I’ve made as well. Including the impact of taxes and transfers is particularly important when you’re analyzing poverty or the well-being of low-income households, since safety net benefits and social insurance programs, such as Social Security, Medicare, or nutritional support, make an obvious and positive difference in the economic lives of their recipients.
But here’s the rub: including taxes and transfers doesn’t much change the trajectory of income inequality over the past few decades. It’s grown considerably anyway you cut it.
The most commonly cited data source for comprehensive income data is from the Congressional Budget Office. The first figure below shows the trend, 1979-2010, in the Gini coefficient, a measure of income dispersion, both pre and post-tax and transfer.
As you’d expect, the post-tax series is below the pre-tax one: our system of taxes and transfers is progressive and it reduces the extent of inequality in the income distribution at any point in time. But turning to trends, both series go up about the same amount over these years: 19% for the pre-tax series and 21% post-tax.
Or, look at another measure of income concentration: the share of income going to the top 1% of households. The table shows these shares for all households and those with children, along with the change in the shares over these 31 years. The post-tax changes in percentage point terms are slightly less than that pre-tax ones: by this metric, inequality rose a bit less after taxes and transfers than it did before. But rise it did!
Two more points.
First, let’s be clear: in analyzing economic inequality, Piketty is by no means mistaken to examine market incomes. If, as has been the case over most of past three decades, market-based inequality generally rises, we will have to continuously ratchet up redistribution through the tax and transfer system to offset the ever more skewed primary distribution of income, aka market outcomes. That is neither good policy nor plausible politics.
In this regard, the increases in income and wealth concentration shown by Piketty are both worrisome and emblematic of a set of inequality-inducing forces firmly embedded in our economy. I can’t imagine that anyone concerned about these distributional dynamics and their impact on opportunity, growth, and the economic future of those on the have-not side of the divide will take any solace from the fact—and I’m certain it is a fact—that once you include transfers, these concentration levels are not as high as they were in the 1920s, as Robert Samuelson points out based on work by Gary Burtless (see links above)
To be clear, both of these authors recognize the trend-based point made in the figure and table above, but their arguments have confused others who interpret them as implying that there’s no increasing trend in inequality in recent decades once you account for taxes and transfers. And that’s wrong.
Second, there are those (not Burtless) who raise this “transfer defense”—the idea that the damage done to middle and low-income households by market outcomes is largely repaired by taxes and transfers—on Monday, only to turn around and argue that we can’t afford such offsets on Tuesday. Here I cite someone from the Cato Institute criticizing earlier work by Piketty and Saez based on these same market-based data for leaving out the subsidies in Obamacare! As I noted, “it is odd to favorably cite the impacts of transfers in the inequality debate while trying to significantly shrink them in the fiscal debate.”
Let’s definitely examine the most comprehensive income sources we can, but in doing so, let’s neither lose sight of the more recent inequality trends in those sources as well, nor the fundamental importance of the distribution of market-based outcomes.