A Tullock Index for assessing the effectiveness of redistribution

I propose two alternative versions of a "Tullock Index" for assessing the effectiveness of income or wealth redistribution. In the spirit of Atkinson's (J Econ Theory 2:244-263, 1970) inequality index, the Tullock Index is constructed with reference to either (A) the maximum inequality reduction attainable with current transfer spending or (B) the minimum transfer spending necessary to achieve current post-transfer inequality. Using Current Population Survey (CPS) microdata from 1988 to 2014, I construct annual estimates of the Tullock Index at the national level for the United States. The Tullock Index is increasing over that period, suggesting that redistribution has become less effective in reducing inequality. State-level panel fixed-effects estimates show that ineffective redistribution is related to higher state-level poverty rates, lower employment-to-population ratios, and lower levels of overall employment.