I posted recently about Brad DeLong’s correct criticism of a Wall Street Journal op/ed by Martin Feldstein. In the op/ed, Marty claimed that under reasonable assumptions, the 20% cut in marginal tax rates that Romney proposes would be revenue-neutral for two reasons: Romney would eliminate or limit many deductions for high-income people and many of these high-income people would make more income, creating offsetting revenue for the government. Feldstein’s argument is reasonable but one key number was incorrect. As DeLong pointed out, Feldstein made the mistake of assuming that the additional income that high-income people made in response to the cut in their marginal tax rates would be taxed at their old higher marginal tax rates. This makes no sense.

I was disappointed, therefore, to see Marty’s response. He writes:

While I still believe the assumptions that I used in my analysis, I can modify them as suggested by the critics and still support my original conclusion by broadening the tax base in ways suggested but not developed in my WSJ piece.

But why does he believe the specific assumption that I highlighted above? The closest he gets to answering is this:

One further point on the appropriate marginal tax rate (objection 1 above): although the top statutory rate is 35 percent, the effective top marginal tax rate is higher because of various phase-out provisions that affect high-income taxpayers (PEP, Pease, etc.) so my original assumption of a 30 percent marginal tax rate could be appropriate even with the Romney rate reductions.

Well, sure, it could be. But then Marty needs to be consistent. To phase out deductions as people’s income goes up, the government implicitly makes marginal tax rates higher for people in the phaseout range. So, for example, if someone increasing his income from $100,000 to $120,000 loses $5,000 in deductions, his marginal tax rate is 25 percentage points higher than otherwise. Surely this would discourage people in that income range from making more income, just as the lower tax rates at the top encourage people to make more income. I have no idea how big an offset this is, but Marty probably does. This is his bread and butter. So his answer is unsatisfactory.

Princeton economist Harvey Rosen takes a better shot at answering the critics. The table at the end of his piece is informative. He shows that given what he thinks are reasonable assumptions about added income growth for high-income people due to lower marginal tax rates, the cuts in tax rates and reductions in deductions could be self-financing. So, for example, by assuming that the 20% cut in marginal tax rates induces high-income people to increase their income by 3%, he finds that the tax changes would increase the federal government’s revenue. Here’s one paragraph that sums up many of his findings:

Now let’s sum things up. For the over $100,000 group, the reduction in revenue because of rate cuts is about $144 billion; the increase in revenue due to base broadening is $200 billion; and with a 3 percentage point growth assumption, the additional revenue from a rise in incomes is $25 billion. The net impact is a positive $81 billion. That is, under these assumptions, taxpayers with incomes of $100,000 or more would pay $81 billion more in taxes. The second row shows analogous computations for taxpayers with incomes of $200,000 or more. Again assuming a 3 percent growth rate, members of this group would pay about $29 billion (or 6.5 percent of current revenues) more in taxes. The implication is that $29 billion less of revenue from base broadening would be necessary in order to keep the taxes levied upon these high-income individuals about the same.

I would have ideally liked to see how Rosen came up with the 3% number. Here’s my own back-of-the-envelope stab at it. The marginal tax rate for the highest-income people is roughly 43.9% (35% Federal + 2.9% Medicare + 6% state.) Cutting the highest rate by 20% would give a top rate of 36.9% (28% Federal + 2.9% Medicare + 6% state.) So the person’s after-tax incentive to earn an extra $ rises from 56.1% (100% – 43.9%) to 63.1% (100% – 36.9%), an increase of 12.5%. It seems plausible that this would increase income by 3%. That would be an elasticity of about 0.24. The elasticity is probably lower that for men and higher than that for married women. So it could work.