Bribing votes: A new explanation to the "inequality-redistribution'' puzzle in LDC's

The recent empirical literature on redistribution and development emphasize s two main evidences: (i) more redistribution generally induces higher grow th rates and (ii) more inequality does not necessarily increase the politic al demand for redistribution. These stylized facts are at odds with the cor relations observed in developed countries. Several theoretical arguments ca n be advanced to explain these puzzles. In this paper, it is shown that ``v ote purchases'' may be seen as an additional argument to explain puzzle (ii ). We formalize this idea and examine the conditions under which vote bribe s may be an obstacle to redistribution (and thus to growth) in a developing economy.