Merijn Knibbe — Piketty’s data set. Criticisms and 3 graphs

If you are following the Piketty phenomenon, this is a must read. Merijn Knibbe provides a short analysis of Chris Giles's criticism of the data and Piketty's response in terms of a scientific debate, which he sees as being on track.

There are huge difficulties with aggregated historical data, especially in a area that is notorious opaque by design, where many if not most wealthy people take all steps possible to avoid transparency in the name of preserving personal privacy. In fact, Piketty makes the point that the problems with the data show the need for increasing transparency and if that is a result, the work will have accomplished a major part of its purpose, since then a reliable data set will be obtainable.

Piketty's research accords with what one would expect intuitively about a system in which power and the institutional privilege it involves. However, it's unlikely that the essentially neoclassical methodological approach that Piketty adopts in creating his model, along with the questionable data, will provide the strongest arguments for the conclusions he comes to. His work will cheer on the choir but not convince those who are convinced otherwise.

Purely economic models that satisfy the mainstream are unlikely to be conclusive since assumptions and data are open to attack. Moreover, mainstream economists seem to be unimpressed with data.

Only models that include institutional power are likely to provide unassailable conclusions. But mainstream economists do not consider such models to be economic models and are unlikely to give them credence. So we are back to the orthodox versus heterodox debate, which is really a political one rather than an economic one.


Real-World Economics Review Blog
Piketty’s data set. Criticisms and 3 graphs
Merijn Knibbe