Thomas Piketty’s Capital in The Twenty First Century, has spawned a cottage industry of dissent.  Piketty uses masses of data to illuminate a growth in inequality, that he surmises is an inevitable result of capitalism and can only be resolved by painfully high taxes on the rich. For the left it is a pivotal work that brings data and credentialism to their ideology that capitalism is so flawed that it requires constant and strong control from the state.

Anti-Piketty is a collection of noted economists and political thinkers that find significant flaws with Piketty’s work.  These critiques include serious flaws with the data itself and how it is used, the difficulty of measuring the forms of income and inequality itself, conclusions that are not supported by the data, and a philosophically flawed concept of wealth, growth and capitalism.

While most of the critics find some praise for the volume of his research and the effort, the sum effect of the collection is to cast such serious doubt on his work and conclusions that it is intellectually worthless, in spite of its political utility.  This leaves a question of how a credentialed and respected economist like Piketty can deliver such a flawed work.

This excerpt from Anti-Piketty contains part of an answer:

Part 2. Criticizing the Empirical Strength of Capital in the 21st Century

“We have data which prove that….” This type of phrase is usually the punchline to authority in economic debates. The opposition can only keep silent: “They have the data….” When faced with theoreticians presented as too abstract, having data seems to allow one to rise above the fray. That implies theory is necessarily detached from reality. The evolution of economic science in the 20th century has indeed imposed levels of theoretical abstraction— or rather types of abstraction, imported from physics— which are problematic in their application in social sciences. Work that relies on data appears refreshing.

However, empirical work is not without a degree of abstraction. The word data is somewhat misleading: although its Latin origin means “something given,” the numbers in question are more likely “picked” than simply “given.” A mechanism of selection is inevitably at work: the collection of data is not impartial, and the resulting statistics can be tainted with a certain vision. Furthermore, data processing is another opportunity to practice “smoothing” and “averaging” that can strengthen even further the trends that the researcher would like to demonstrate.

The “art of measurement” of a social scientist can thus be very subtle, with fairly broad room for maneuver in methodological choices. The data can quite easily be made to “speak” in a desired manner. Thomas Piketty is accused of such maneuvering in his empirical analysis of inequality— especially since the book claims that its high empirical content is its major strength.