More money is redistributed to poor children in the U.S. through school spending than
through any other government program such as health care, welfare, subsidized housing, or
food stamps and related programs. If the goal issimply to transfer money from high income
to low income people, then economic theory suggests that school spending is a poor
mechanism for such transfers. The justification for such transfers is more likely the
Efficient Redistribution Hypothesis which implies that, in the absence of redistribution,
returns to the marginal dollar of school spending on a poor child are likely to exceed the
returns to the marginal dollar of school spending on a rich child. Thus, efficiency
considerations alone may justify redistribution, with no need for concerns about inequality
per se. Currently, there is no direct empirical support for the Efficient Redistribution
hypothesis. In this paper, I test whether U.S. states that practiced more redistribution
through school spending ended up producing adults whose incomes were higher on average
and less unequal. I exploit substantial within-state variation generated by changes in school
finance policies. Owing to rather arbitrary implementation of state Supreme Court
judgements, the redistributive consequences of these policies are fairly uncorrelated with
other factors affecting states' income distributions. I find that redistributive school spending
reduces income inequality among adults, but the effects are too weak to justify the use of
school spending (as opposed to more direct mechanisms such as income transfers) for purely
redistributive purposes. I do not find support for the Efficient Redistribution Hypothesis.
Even if the Efficient Redistribution Hypothesis is correct is theory, it could fail to work in
practice if schools that serve low income children make inefficient of resources.