David R. Henderson  

Reynolds on Summers and Piketty

PRINT
Walter Oi on Government Budget... The middle class is doing fine...

My copy of Thomas Piketty's book, Capital in the 21st Century, sits unopened. I plan to read it in June for a review that's due at the end of June. Meanwhile, though, I've been reading lots of reviews and comments on his book. When I think of Piketty, I usually think of him as a part of a duo, the other member being UC Berkeley economist Emmanual Saez. But assuming that the reporting I've seen from all sides about the content of Piketty's book is accurate, and assuming that the data have not changed between 2000 and now, both Piketty's book and Summers' understanding seem at odds with work by Saez.

Alan Reynolds has noted this problem. Reynolds writes:

What about Wealth - the defining topic of Piketty's book? Summers conflates wealth and income - claiming Piketty and Saez have shown "absolutely conclusively, that the share of income and wealth [emphasis added] going to those at the very top ... has risen sharply over the last generation." On the contrary, a 2004 paper by Saez and Wojciech Kopczuk found the top one percent's share of wealth was 38.1% in 1916, 40.29% in 1930, 25.25% in 1960 and 20.79% in 2000.

Reynolds goes on to quote from Saez and Kopczuk's abstract. I'll quote the whole abstract:
This paper presents new homogeneous series on top wealth shares from 1916 to 2000 in the United States using estate tax return data. Top wealth shares were very high at the beginning of the period but have been hit sharply by the Great Depression, the New Deal, and World War II shocks. Those shocks have had permanent effects. Following a decline in the 1970s, top wealth shares recovered in the early 1980s, but they are still much lower in 2000 than in the early decades of the century. Most of the changes we document are concentrated among the very top wealth holders with much smaller movements for groups below the top 0.1%. Consistent with the Survey of Consumer Finances results, top wealth shares estimated from Estate Tax Returns display no significant increase since 1995. Evidence from the Forbes 400 richest Americans suggests that only the super-rich have experienced significant gains relative to the average over the last decade. Our results are consistent with the decreased importance of capital income at the top of the income distribution documented by Piketty and Saez (2003), and suggest that the rentier class of the early century is not yet reconstituted. The most plausible explanations for the facts are perhaps the development of progressive income and estate taxation which has dramatically impaired the ability of large wealth holders to maintain their fortunes, and the democratization of stock ownership which now spreads stock market gains and losses much more widely than in the past.

I'm looking forward to resolving this apparent contradiction. Did Summers misquote Piketty? Did Piketty change his mind or come up with new data? Or are the data on wealth since 2000 a lot different from the data in 2000? It's got to be at least one of the three. Do any of know? If not, that's alright, because by the time I sit down to write my review, I will have figured it out.

BTW, I wanted to link to Summers' review of Piketty but that web site seems to be down. Specifically, I wanted to see if Reynolds is correct in saying that Summers referred to Piketty and Saez and not just to Piketty.

UPDATE: Reynolds did quote Summers accurately enough. Summers had not referred to Piketty and Saez, but had referred to "Piketty, in collaboration with others." Summers surely had Saez in mind. Here's the link to Summers' review.



COMMENTS (2 to date)
Vivian Darkbloom writes:

Here are some slides from Saez that are consistent with the above-referenced abstract:

http://gabriel-zucman.eu/files/SaezZucman2014Slides.pdf

Also, Saez' research has indicated that much of the wealth increase amount the top few percent has been largely the result of greater pension wealth. If in the form of an annuity, this wealth is not inheritable. Even if not, these increases are largely attributable to greater longevity.

See here:

http://www.forbes.com/sites/timworstall/2014/04/09/the-thomas-piketty-smackdown-from-emmanuel-saez/

ed writes:

Reynolds is correct, Summers misstated Piketty's claims.

All the wealth inequality data used by Piketty in chapter 10 is given in the following table at Piketty's website:

http://piketty.pse.ens.fr/files/capital21c/en/pdf/supp/TS10.1.pdf

As you can see, according to the data used by Piketty, all the measures of inequality in all countries were much higher just before WWI than today. (For that matter, they were higher after WWI as well.)

The numbers in this table are a combination of various estimates, including the Zucman & Kopczuk paper, which I believe is based on estate tax records. More recent estimates for the United States come from the Survey of Consumer Finances. The details of how these series were constructed and combined can be found in the technical appendix and spreadsheet at Piketty's website:

http://piketty.pse.ens.fr/en/capital21c2

Comments for this entry have been closed
Return to top