For example, colonialism, which Picketty and Acemoglu base a lot of their work on, would be hard to reimplement for experimentation. Hence the debate. The data and conclusions are interesting and compelling, but not yet decisive.