Skip to content

Grasping at Straws: A Response to Paul Krugman

5 min readBy: William McBride

Paul Krugman has challenged us once again on the veracity of our analysis, this time regarding income inequality:

“So today I was informed that the TaxA tax is a mandatory payment or charge collected by local, state, and national governments from individuals or businesses to cover the costs of general government services, goods, and activities. Foundation was claiming that there has been no long-run upward trend in income inequality. My immediate question was, how did they fudge the data to get that result? Well, the answer is here. Actually, the whole post is kind of a classic of disinformation: pretense that income taxes are the only taxes? Check. Bemoaning the rising share paid by the rich, while ignoring their rising share of income? Check. Whining that the lucky duckies with low incomes don’t pay enough? Check.

First of all, they’re using CBO data, which only go up to 2009; we do, as it happens, have estimates from Piketty and Saez that go up to 2010, and they show inequality already rebounding.

Rebounding from what? Well, what major income source does the top 1 percent have that is major for them but minor for everyone else? Capital gains. And what was happening to capital gains in 2009? Financial crash, anyone?”

First of all, Krugman links to my blog post rather than my more recent and more detailed analysis. In that, I note (footnote 8) that the story is much the same when using CBO’s tally of all federal taxes, i.e. taxes on personal income, corporate income, payroll, etc.:

“The federal tax code as a whole is progressive and becoming more so. For instance, the CBO finds that the top 1 percent has a total combined federal tax rate of 28.9 percent versus a 1.0 percent tax rate for the bottom quintile. The top 1 percent pays 22.3 percent of all federal taxes versus 1.0 percent for the bottom quintile. Likewise, the CBO finds that the progressivity of the entire federal tax system is at a record high.”

Second, in both analyses I clearly point to capital gains as the main driver of income inequality over the last two decades:

“Here again the evidence points to no long-term trend in inequality since the 1980s. Rather, there were two large cycles up and down corresponding to historic stock market bubbles in the late 1990s and mid-2000s. High-income earners receive much of their income from capital gains, so their fortunes are particularly tied to the stock market.”

Krugman points to the Piketty and Saez chart of income shares with and without capital gains. That is very useful for sure, to strip out capital gains so as to see the remaining income sources, i.e. largely wages and business income. That chart does indicate increasing income inequality, from the late 1980s to 2010 (shown below in Figure 1, but for the top 1 percent).

Now what happens when we strip out business income as well and just look at wage inequality? This is shown in Figure 2, also from Piketty and Saez. It indicates that the long-term trend of increasing wage inequality, beginning in the 1970s, comes to a screeching halt in 2000, when the share of wages earned by the top 1 percent peaked at 12.26 percent. Since then it has fluctuated dramatically along with the economy. In 2010, the top 1 percent’s share of wages was 10.9 percent – roughly where it was in 1997-1998. Much of what Piketty and Saez identify as growth in income inequality is actually growth in the share of businesses filing under the individual code. These so-called passthrough businesses, such as S-corporations and partnerships, now represent the majority of business income, and this income mainly accrues to high-income earners.



Lastly, in addition to this, the Piketty and Saez methodology, as compared to that of the CBO, greatly overstates income inequality for three main reasons:

1) It excludes public transfer payments, such as Social Security, Medicare, and Medicaid benefits. These are a large and growing source of income for low- and middle-income households, as the baby-boomers have aged and healthcare costs have increased. The CBO includes this income, and our forthcoming analysis of that data indicates, for example, that transfer payments make up 75 percent of total income for low-income earners, i.e. the bottom 20 percent of households. This is up from about 60 percent in 1979.

2) It excludes fringe benefits, particularly employer sponsored health insurance, which also have grown to be a disproportionately large share of total compensation for mainly middle-income households. CBO includes these benefits.

3) It ignores the changing size and structure of households. Piketty and Saez use “tax units” rather than households, where tax units are basically tax filers or potential tax filers. The number of tax units per household has increased in recent decades due to factors such as increased cohabitation rather than marriage, and increased number of adult relatives living together. Households share economic resources to some extent, making households the proper unit of observation for income inequality research, which is what CBO does. When high- and low-income tax units are combined in one household, measured income inequality goes down.

Burkhauser, Larrimore and Simon take into account these three factors plus they adjust for household size. They find that income inequality actually decreased over the course of both the 1990s and 2000s (see the Gini indexes in Table 4). From their conclusion:

“Much of the previous research on income and its distribution is based on the types of income captured in the data and the sharing unit over which it was summed, without fully considering the implications of those choices. In this paper, we demonstrated that such choices can substantially change the view of how the average American has fared over the past three business cycles (1979-2007) and who benefits from public policy choices going forward. When using the most restrictive income definition – pre-tax, pre-transfer tax unit cash (market) income—the resources available to the middle class have stagnated over the past three business cycles. In contrast, once broadening the income definition to post-tax, post-transfer size-adjusted household cash income, middle class Americans are found to have made substantial gains, and these increases are even larger when including non-cash income such as the ex-ante value of health insurance.”

Follow William McBride on Twitter @EconoWill