Robin who? Using indivisible labor and systemic misapprehension to solve the Robin Hood paradox of the Meltzer and Richard model on government size.

Abstract: To better understand why "history reveals a 'Robin Hood paradox,' in which redistribution from rich to poor is least present when and where it seems to be most needed." (Lindert, 2004, p.15), I make two modifications to the model of Meltzer and Richard (1981) on government size. Their model predicts that a lower productivity of the median voter, relative to the mean, leads to higher taxation, the opposite to what is observed in empirical studies. First, I assume indivisible labor, a binary choice of employment instead of the original assumption of a continuous choice, and find that for plausible parameterization, this solves the Robin Hood paradox. Then, I examine the potential effects on government size of voters that overestimate others productivity, using evidence from behavioral research (Karadja et al., 2014). Systemic overvaluation of others productivity increases the demand for redistribution, thus if the misapprehension is larger where the productivity distribution is more unequal, the Robin Hood paradox is solved.