Certainly the inability to run a test or set a control is a constant problem for macroeconomics, but I don't think that deprives us of all information. Piketty

Certainly the inability to run a test or set a control is a constant problem for macroeconomics, but I don't think that deprives us of all information. Piketty, at least, is content to compare the relationship between top marginal tax rate, pre-tax income share, and GDP in Germany, Denmark, the UK, and the USA since the 1970s. He finds that higher top marginal tax rate correlates to (and precedes) decreased top pre-tax incomes, and doesn't correlate with GDP. Inequality numbers might be distorted by other changes, but it would be interesting to at least see the naive comparison between top marginal tax rate and change in wealth inequality for those same countries.