The Corner

Economy & Business

‘This article does not question Piketty’s integrity.’

That’s from the abstract of a new paper by economist Richard Sutch, a disclaimer that may have been included because Sutch is so critical of the methodology employed by Thomas Piketty in Capital in the Twenty-First Century. Piketty draws from a massive body of historical data to show that the wealth and income distributions in the United States and Europe have skewed toward the rich. (He argues further that since the rate of return on capital tends to exceed the rate of economic growth, rising inequality is an inherent feature of capitalism — the “dynamics of private capital accumulation inevitably lead to the concentration of wealth in ever fewer hands” — and advocates policies like a “wealth tax” to counteract it.)

Sutch doesn’t discuss Piketty’s theorizing. Instead, he focuses on Piketty’s empirical work, finding enough problems with it to make a disclaimer defending Piketty’s integrity necessary.

Piketty’s analysis of the concentration of wealth in the U.S. in the twentieth century uses two data sets: an archive of estate-tax returns and a survey conducted by the Federal Reserve. But the data sets are not perfectly reliable and, what’s more, each diverges from the other, since one measures the wealth share of households and the other measures that of individuals. So Piketty adjusted the estimate of the share of wealth held by the top one percent given by the estate-tax data upward by a factor of 1.2 to reconcile it with the Fed survey. That’s not indefensible, provided the (debatable) assumption that the Fed survey is more reliable than the estate-tax archive. But Sutch says the multiplier Piketty uses “is a bit of a mystery.”

Piketty plots a rolling ten-year average of the wealth concentration of the top one percent to develop his narrative of rising inequality. While smoothing data by using a rolling average is not objectionable, a ten-year average, Sutch says, “makes it difficult to connect public policy changes, stock market swings, and other developments to changes in the distribution of income and wealth.” That’s a problem, given that Piketty makes specific claims about public policy.

Sutch is most critical of Piketty’s analysis of inequality in the nineteenth century, from which he says “very little of value can be salvaged.” Piketty himself admitted that the data from this era must be considered “uncertain,” but nonetheless tried to retrieve wealth-distribution data for 1810, 1870, and 1910. Sutch casts doubt on each of these figures, and concludes: “The heavily manipulated data, the lack of clarity about the procedures used to harmonize and average the data, the insufficient documentation, and the spreadsheet errors are more than annoying.”

Leave aside Piketty’s claims about capitalism or his preferred, fanciful tax on capital. As Alex Tabarrok notes, even scholars who demurred on Piketty’s theorizing lauded his empirical research. But Sutch’s is the second paper to find serious flaws with that research, supporting the conclusion of libertarian economists Philip Magness and Robert Murphy. This is not a case of academic fraud: There’s no evidence that Piketty’s integrity should be in doubt. But the quality of his research certainly is.

Most Popular

Elections

The Highest-Stakes Moment Brings the Worst Debate

Tonight’s debate would have been only marginally less incoherent, noisy, and grating to the ears if CBS had broadcast two hours of static. The last debate before the South Carolina primary featured so much shouting, you would think that the candidates had just been told their microphones weren’t working. ... Read More
Elections

The Highest-Stakes Moment Brings the Worst Debate

Tonight’s debate would have been only marginally less incoherent, noisy, and grating to the ears if CBS had broadcast two hours of static. The last debate before the South Carolina primary featured so much shouting, you would think that the candidates had just been told their microphones weren’t working. ... Read More
Media

‘Undiagnosed Sociopath’

As we abandon moral language for clinical language, we run into technical difficulties. Writing in the New York Times, Thomas Friedman describes the 2020 presidential election as one that may be a contest between “a self-proclaimed socialist and an undiagnosed sociopath.” There is no such thing as an ... Read More
Media

‘Undiagnosed Sociopath’

As we abandon moral language for clinical language, we run into technical difficulties. Writing in the New York Times, Thomas Friedman describes the 2020 presidential election as one that may be a contest between “a self-proclaimed socialist and an undiagnosed sociopath.” There is no such thing as an ... Read More