Thursday, June 14, 2007
Asking the right question
In all my discussion of the state income tax I never did ask the right question. Greg Mankiw asks it today, How much progressivity is optimal?
In short, responding to a paper buzzing around the economics blogs right now written by economists who have worked or advised Democrats, Mankiw asks them to spell out what the model is in their heads. How heavy a weight do you put on the aversion to inequality or poverty? Should you actually specify a "social welfare function"? How distortionary are income taxes? The first two are matters of political philosophy, Mankiw argues, while the last is an economic debate that is unsettled (see the discussion we had last month, for example.)
When I wrote many years ago on inflation as a tax (see this old paper for instance), I was always discomfited by the word optimal. Optimal to whom? The one number I could focus on was "revenue maximizing", which I took to be optimal to Leviathan. That's more defensible in my view (the paper shows that, since seigniorage gains from inflation are not very large, it takes only a small output loss from inflation to make using the inflation tax a loser to government.)
We tend to avoid the political philosophical debate and go for the harder case to make -- that the tax increase is a money loser -- because as economists we don't have any special insight into the philosophical questions. But it's worth asking someone "if we could show the tax increase raised huge sums of money, would you be for it?" Many times, my answer is still no, because I have a different philosophy than the questioner. And we should be honest about that.
UPDATE: And sometimes tax policy is just a bribe to let progress continue.
In short, responding to a paper buzzing around the economics blogs right now written by economists who have worked or advised Democrats, Mankiw asks them to spell out what the model is in their heads. How heavy a weight do you put on the aversion to inequality or poverty? Should you actually specify a "social welfare function"? How distortionary are income taxes? The first two are matters of political philosophy, Mankiw argues, while the last is an economic debate that is unsettled (see the discussion we had last month, for example.)
When I wrote many years ago on inflation as a tax (see this old paper for instance), I was always discomfited by the word optimal. Optimal to whom? The one number I could focus on was "revenue maximizing", which I took to be optimal to Leviathan. That's more defensible in my view (the paper shows that, since seigniorage gains from inflation are not very large, it takes only a small output loss from inflation to make using the inflation tax a loser to government.)
We tend to avoid the political philosophical debate and go for the harder case to make -- that the tax increase is a money loser -- because as economists we don't have any special insight into the philosophical questions. But it's worth asking someone "if we could show the tax increase raised huge sums of money, would you be for it?" Many times, my answer is still no, because I have a different philosophy than the questioner. And we should be honest about that.
UPDATE: And sometimes tax policy is just a bribe to let progress continue.
