Thomas Piketty’s Motte and Bailey
26th May 2025
This radical bailey position is made untenable in three ways. First, my work with Phil Magness, John Moore, and Phil Schlosser (published in Economic Journal and Economic Inquiry) suggests that Piketty was careless in his usage of historical source materials and made numerous important errors in estimating inequality pre-1960. One involved his application of a rough estimation of income, though original sources contained data that could have enabled a more precise calculation. My coauthors and I tried to refine these pre-1960 estimates. Further, Piketty’s work contained significant historical inaccuracies (such as overlooking the exemption of state and local government employees from federal taxes before 1938) and misrepresented several steps in his methodology. We addressed these errors, too. Overall, we discovered that Piketty overstated inequality levels before the 1960s by about 20 percent.
Second, our work in Economic Inquiry showed that most of the levelling in the 1930s occurred as a result of the wiping out of capital gains. Roughly four-fifths of the “golden age” of equality (between 1950 and 1980) owed to the Great Depression, not tax policy. This finding is hard to celebrate because it means that greater equality was achieved while everyone was getting poorer. It also eliminates most of the purported influence of higher tax rates in generating the “golden age” of equality.
Third, the work of Gerald Auten and David Splinter shows that the golden age was not so golden. Once they corrected for how tax policy often encouraged changes in how taxpayers organized their income sources according to corporate or personal identities, they found that inequality started from a higher floor in the 1960s than Piketty and his colleagues presume. They also find a milder increase in inequality since the 1980s.