Yesterday, Gabriel Zucman was announced as the recipient of the 2023 Bates Clark Medal. As a comment on this, I will refer to my Economic Journal article with Phil Magness, John Moore and Phil Schlosser. In this article, my co-authors and I replicated the work of Piketty and Saez (2003) in the Quarterly Journal of Economics and found that they did multiple errors, ignored key sources and made serious misunderstandings of the source materials before 1960 . When we corrected their series, we found major differences between their published series that present a strongly different version of the evolution of American income inequality. The main takeaway graph is below and as can be seen — their level is off and the trends are different (theirs show WW2 and tax policies during the war as leveling — ours show the great depression as the great leveller — something we confirmed in another article in Economic Inquiry on why the IRS data is horrible at getting inequality right).
The reason why this matters is that the vast majority of Zucman’s work builds off the 2003 paper (notably with how they poorly create denominators for total income) by Piketty and Saez. They have never updated that work and Zucman just keeps repeating it. As a result, I admit that I consider most of Zucman’s inequality work (both as a solo author or with Piketty and Saez) as highly dubious because he and they keep treating the data with little care as showed in both our articles.
Maybe we can commend the interest in inequality and the commitment to the question, but there is little else that warrants commendation.
Vincent Geloso 