Controversy of the week: Is Thomas Piketty’s data flawed?

It’s the “50 Shades of Grey of economics books”, according to The Guardian’s Larry Elliott. The best-selling Capital in the Twenty-First Century by Thomas Piketty (pictured), which argues that rising wealth inequality is an inevitable, disastrous consequence of capitalism, has been hailed by the left as one of the most important contributions to economics since Karl Marx published Das Kapital.

And while those on the right might not like or agree with its conclusions, reviewers of all stripes have praised Piketty’s depth of research.

But this week, a new row erupted when Chris Giles, economics editor of the Financial Times, questioned Piketty’s data. Giles cites problems ranging from simple transcription errors and selective quoting of material, to the more serious issue of making unilateral adjustments to raw data from patchy sources without providing justification.

Overall, “there is little evidence in Prof Piketty’s original sources to bear out the thesis that an increasing share of total wealth is held by the richest few”. “Apocalyptic visions of a return to Dickensian social division are over-heated,” agrees The Independent’s John Rentoul.

Income distribution has “remained broadly unchanged” since the 1980s. Recent data from the Office for National Statistics suggests that wealth inequality is also peaking. “If you take a 100-year view, Piketty is clearly wrong,” says the economist Kitty Ussher, in The Guardian.

For instance, “one in five people in the early 20th century owned property; today two thirds do”. The short-term evidence may be more “nuanced”, but “there is little to show a marked trend”.

Others have been more sympathetic to Piketty. The Economist reckons that Giles’ analysis does not “seem to support many of the allegations made by the FT, or the conclusion that the book’s argument is wrong”.

Paul Mason in The Guardian takes a more conspiratorial line: “if Giles is right, then all the gross designer bling advertised in the FT… can be morally justified”.

Piketty himself accepts that “the available data sources on wealth are much less systematic than for income”. However, responding to the FT, he says his adjustments are simply an attempt to make the data ”more homogenous over time and across countries”.

He also says he would be “very surprised if any of the substantive conclusions about the long-run evolution of wealth distributions” were altered by future data improvements, noting that recent reports by University of California, Berkeley and the London School of Economics “confirm and reinforce my own findings”.