I wonder whether Piketty's proposal might replace one wealth-correlated inequality by an equally frightening one that would arise through the quasi-mechanical ademocratic devolution of power to a global revenue collector (read state) enabling the increased taxation of capital and its transmission. Are the statistical grounds for the conclusion of his analysis really that strong? Inequality and inequity are neither synonymous nor coextensive. The worst inequity (probably) occurs in the quantiles of the poorest most vulnerable members of society. One needs statistics to capture the skewed distribution and to adjust the measures accordingly. Is one attempting to move the mean to the right or to reduce the variance or other higher order moments? A linear shift in the mean will not change the shape of the curve describing the distribution of wealth. Trusting a global state to redistribute is as sensible as preaching the virtues of vegetarianism to a hungry fox in a well-stocked hen house.
Robert Solow's review is as lucid as ever (even if I dare to diverge): http://www.newrepublic.com/article/117429/capital-twenty-first-century-thomas-piketty-reviewed.
Some posts contain periodically updated listings of my research publications and papers in preparation; others relate to matters of general interest in Mathematical Logic. My page at Cambridge: https://www.dpmms.cam.ac.uk/people/ok261/
