Home > Uncategorized > Piketty’s data set. Criticisms and 3 graphs.

Piketty’s data set. Criticisms and 3 graphs.

When I was a student, Angus Maddison was the dominant intellectual force in the economic department of Groningen University (sorry, Jan). This taught me among other things (1) that science is (also) about continuously updating and improving your datasets, (2) being totally open about your data and how you ‘handle’ your data or these of others while (3) sometimes you will have to make heroic assumptions about data, for instance because the data you have are the lengths of Roman femurs while you want to say something about the long term economic development of the Roman Empire (look here for an excellent application of Ockham’s razor blade in that discussion, there are by the way 35.000 measurements of Roman ‘long bones’).

Which means that as far as I’m concerned Chris Giles does an indispensable job by close reading the sources of Piketty and the way Piketty uses these sources, which is possible because Piketty passes the (2) test. While Piketty reacts in the right way, passing the (1) test: open to criticism and willing to change and learn (i.e. not like Rogoff reacted to criticisms of the Reinhart and Rogoff dataset). Heya Chris, allez allez Piketty!

This, however, leads to two questions:

(a) What to think of the data-criticism

(b) What does this mean for the ideas of Piketty

Ad (a)

Giles points out some perceived flaws in the data on (ownership of) wealth – the 1% stuff.

1. He shows mistakes and flaws in some data. Piketty has to correct these, ‘when data can be improved they have to be improved’. Mind that there are always mistakes in large datasets, economists have to get used to this (and take not of it, when talking about data).

2a. Piketty has to answer the Giles questions about data for which Giles could not find sources

2b. Most importantly, Giles shows that Piketty did not use some obvious recent sources, like wealth data from the ONS (the english Statistics Bureau). Piketty has to either explain this better than he did (he might have done so in some of his background articles, I didn’t check) or he has to incorporate these data into his main data set.

3. He shows that Piketty uses some heroic assumptions (which are clear from the text and especially the Excel sheets). Fine, as long as Giles does not present a better alternative this criticism is not valid (Giles owever shows alternative series, good!).

Ad (b)

Giles presents some new estimates of ‘1% data’, differences with Piketty are mainly based upon alternative data sets. According to me, these data still show a (limited) recent increase in wealth inequality, according to Giles they don’t. Piketty should follow Giles by calculating ‘European’ inequality by using a population weighted average of Swedish, French and UK inequality instead of the arithmetical average. Generally, in my view, differences between Giles and PIketty are limited. Modern inequality according to Giles starts to rise a little later than in the Piketty series. Caveat: Giles mentions the ONS wealth data but as these show, for the most recent period, by far the lowest wealth inequality of the entire period, compared with all other estimates, it does not seem right to splice these data (this data?) to the other series without thorough discussion. Giles understands this and shows two distinct paths for post 2000 British inequality  (mr. Giles, please do use different colors for different series in your graphs!). But when push comes to play the main differences between the Giles and the Piketty data are caused by inclusion of these ONS data (which may be right, of course, but which are remarkably low).

Piketty reacts in the right way to these remarks by pointing out new data by Zaes and Zuckman (published after the book was printed). These/this data (the latest and the best for the USA) show that (1) rising inequality, (2) that this was driven by, believe it or not, the 0,01%, while (3) return on investment did not decrease (despite the increase in total wealth, including pension fund wealth, and low interest rates), see their graphs below. Piketty will however have to explain the low ONS data, or change his files.

Graph 1. Wealth inequality (in the USA) continues to increase. And not just a notch.


Graph 2. The 0,01%. I’m living in a city of about 100.000 persons. In such a city 0,01% is equal to 2 or 3 households which control 12% of the entire wealth of the city. Do they ‘deserve’ this wealth? Do they still deserve it when they get old, get Alzheimer-disease and are of no ‘utility’ whatsoever to ‘society’ anymore? Does Paris Hilton deserve it?


Graph 3. How do they do it – despite the fact that wealth increased as a % of GDP, wealth related income continues to be stable as a % of total wealth (oops, the other kind of income is labour income).







  1. May 24, 2014 at 4:56 pm

    My bet is that the Lance Taylor critique will in the end do much more to undercut Piketty’s findings than the Financial Times corrections and recalibrations, as useful as those are. Taylor challenged the widely-touted Piketty’s assertion that r > g (the rate of return on capital exceeds the growth rate of the economy). NC reader Ben Johannson provided a helpful summary of Taylor’s paper:

    1) Taylor makes the point that Picketty’s determinations of the rate of profit and the capitalists’ share of those profits assume a fully employed global labor force due to his use of the neoclassical production function (the one trashed back in the 1950s during the Cambridge capital controversies). This is THE critical error in Picketty’s work, that capital can be aggregated and differences simply assumed away while the reality of effective demand is ignored.
    2)The rate of profit and share of net profits will vary over time depending on the business cycles, employment level, monetary policies, technical changes, etc. The neoclassical production function referenced above does not take this into account.
    3) The accumulation of wealth at the top is not an autonomous product of “capital”, some natural law of economics which states that it will always produce growing inequality, but rather a product of specific policies which can be reversed. Altering the ratio of output/capital and the share of profits taken by the capitalist class is the better and more easily implemented choice for reducing inequality rather than taxation. In other words rising real wages is more effective in sustaining aggregate demand and attenuating capitalist power, while relying on taxation will fail to address stagnating wages and continue the current trends.
    “The accumulation of wealth at the top is not an autonomous product of “capital”, some natural law of economics which states that it will always produce growing inequality, but rather a product of specific policies which can be reversed.”
    Man is not created equal with the ways and means of accumulating wealth.
    It is not the accumulation that is cause for concern, rather it is the size of the gap
    between the accumulation and the amount needed for ..“A living standard to each according to their need;self-care and mutual service expected from each according to their ability.”
    A good social government should have as a mandate ( as expressed in the Preamble) : … promote the general Welfare, and secure the Blessings of Liberty to ourselves and our Posterity,…”.
    “…A product of specific policies which can be reversed”
    What specific policies could we address?
    ENFORCE the mandate ” to promote the general welfare, and secure the Blessings of Liberty to ourselves and our Posterity.”

    Who among you would challenge, or correct then endorse:

    Cure for:
    … unemployment.
    ………..inequitable distribution of income and wealth.

    A Central Bank, under full transparency being the sole issuer of the currency. Currency that is owned by the people, currency that the Central Bank is the caretaker and stores while creating lawful transfers of the currency’s redemption value. The total assets of the Central Bank is the total of the redeemable goods and services of the entire sovereignty.This is the “Capital” of the “Capitalistic” sovereign group. In order ” “to form a more perfect Union, establish Justice, insure domestic Tranquility, provide for the common defense, promote the general Welfare, and secure the Blessings of Liberty to ourselves and our Posterity,…” the Central Bank must work “For the People” while at the same time without appropriating
    the people’s already owned “goods and service.”
    What better way is there than:

    AMEND THE FEDERAL RESERVE CHARTER; TURN THE FED RESERVE INTO THE FEDERAL RESERVE BANK OF AMERICA (FRBA),RESTORE MONETARY POWER BACK TO THE PEOPLE ,OPERATE THE FRBA WITH ABSOLUTE TRANSPARENCY, (“GLINDA,the Good Witch, owns a Great Book of Records that allows her to track everything that goes on in the world from the instant it happens.”_The Road to Oz)

    Form a more perfect “capitalistic monetary ” circle: $100 trillion issued as loans to come back as $200 trillion as payment while at the same time as it returns creates $100 trillion in new loans while spending $100 trillion as Congressional appropriations for the benefit of the people.
    No inflation or deflation for there is zero change in the capital value of the sovereignty.
    There is zero change on the balance sheet of the Central Bank; a true zero net change.

  1. No trackbacks yet.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.