> This Economist article points out some of the many small academic works that quibble over details with Piketty and Saez. But that's not anything new
These are not nitpicks. What you measure drastically changes the results. Piketty and Saez presented their own revision to their data in 2018 that shows starkly different results than the 2013 data everyone cites: http://gabriel-zucman.eu/files/PSZ2018QJE.pdf
For example figure 3 shows that when you account for tax transfers, the bottom 50%’s share of income is only a few points lower than in 1960, though significantly lower than the peak of about 25% in 1970. It also shows that, contrary to common assertions, the bottom 50% income has grown consistently since 1960 adjusted for inflation. Figure 1 shows that the labor share of income has been flat since 1940 when you account for benefits. (Put another way, capital’s share of income is flat.)
The data about returns to capital are particularly important. Supporters of wealth taxes and capital taxes tend to point to rising inequality as the justification for such approaches. But if the returns to capital aren’t increasing as a fraction of the economy, that argument gets harder. Of course there may be other arguments in favor of such taxes, but those seem less well developed.
The point the Economist article makes well is that inequality is hard to measure, and the simple definition used by Piketty and Saez’s initial analysis is inadequate. For example, there is a whole school of libertarian thought that, even if you believe the government should be in the business of redistribution, the best approach is to maximize growth through a free, deregulated economy and then tax individual incomes to provide generous welfare benefits. Most of the developed world has moved in that direction over the last 30 years. Under Piketty and Saez’s 2013 model, which excludes tax transfers, you can’t accurately describe inequality in a country that uses such a model.
Moreover, there are additional analyses showing that, when you look at consumption rather than income (which accounts for both tax transfers and falling prices from globalization) inequality is basically unchanged since the 1970s: https://voxeu.org/article/consumption-and-income-inequality-.... That is to say, the bottom 10% get to consume about the same fraction of total production as they did half a century ago. Of course there are downsides to consumption measures too, insofar as the data is sparser for incomes at the top. (The measure above includes money spent on mortgages for mansions, but not private yachts, for example.)
But just because something is easy to measure (pre-tax income) doesn’t mean it’s the right thing to measure!