Matt Yglesias, on Monday:

If Romney wants to cut taxes on the rich without hurting middle-class pocketbooks, he needs to return to the faith of his GOP forefathers and ask, “What would George W. Bush do?” The answer is simply not to pay for the tax cuts. That would mean admitting that the attacks on Obama-era deficits are bogus. But it would also have the virtue of being true. For all that Democrats mewled about deficits during the Bush years, there’s precious little evidence they did the economy any harm. Right now the government can borrow money basically for free, so there’s little reason to pay for anything. That wouldn’t be my ideal economic stimulus plan, but it’s better than doing nothing and certainly better than the middle-class tax hike Romney has backed himself into proposing. But the first step toward improving his proposal would be admitting he has a problem.

In other words: across-the-board tax cuts aren’t the greatest stimulus, but we badly need stimulus, and deficits don’t matter, so better that than nothing. And certainly better that than coupling across-the-board tax cuts with a reform of the code that constitutes a middle-class tax hike. Simply cutting taxes across the board, without the loophole-closing tax reform, would be “improving his proposal” by making it more stimulative.

Matt Yglesias, today:

The good thing about taxes is they raise revenue, which can be used to do useful things. The bad thing about taxes is they may be a drag on economic growth. But here there are two considerations. One is the “incentive effect” of taxes—higher taxes mean less incentive to do economically valuable things. The other is the “income effect”—less money in your pocket means more incentive to do economically valuable things. The genius of Romney’s plan is that by eliminating deductions it leaves middle class families with less money in their pockets (so a pro-growth income effect) while also lowering the tax rate they pay on a marginal dollar of additional earnings (so a pro-growth incentive effect). Basically it’s a huge win. You get a bunch of revenue in a way that bolsters the country’s growth prospects. The voters don’t want to hear that this is a good idea, but it’s a good idea.

Now a separate issue is what do you do with the money. You could use it to reduce the budget deficit. You could use it to build new aircraft carriers. You could use it to pay for early childhood education. Or you could do what Rommey does and use it to finance a reduction in marginal income tax rates for high income people. I don’t think that’s a smart use of the money, but the idea that high-end tax cuts are more socially beneficial than increased public services is a longstanding American conservative position. Romney’s innovation is the higher taxes on the middle class, and it makes sense. Only he won’t admit his plan has this feature so he can’t defend it.

In other words: tax reform that eliminates loopholes, and thereby raises taxes on the middle class, is a great idea. It raises revenues, which we could use to “do useful things,” and it doesn’t reduce growth because while people are poorer (because their taxes are higher), they have more incentives to earn more (because their marginal rates are lower). The only problem with Romney’s proposal is that big across-the-board tax cut that overwhelmingly favors the wealthy. This, Yglesias feels, isn’t a “good use of the money.” Presumably, Romney’s proposal could be improved by eliminating these wasteful tax cuts.

On Monday, we should do the rate cuts without closing the loopholes. On Thursday, we should close the loopholes, but skip the rate cuts.

If Yglesias keeps this up, Andrew Sullivan will have to create a new Romney Award, just to give it to him.

Now, to be serious for a moment:

Income effects are something to worry about when labor markets are tight. If it’s relatively easy to find more work, and you cut somebody’s income, it’s plausible that he or she will work harder to make up for the lost income. When labor markets are slack – as they emphatically are now – and middle-class households are desperately trying to rebuild their balance sheets, it is not logical to say, “let’s cut their incomes so they work harder.” In this environment, cutting middle-class incomes will only make them spend less – which is exactly what we don’t want, since we’re suffering from a substantial shortfall in demand. Yglesias got this part right on Monday: now is not the time to cut middle-class incomes.

By the same token, when investment incentives are wildly distorted by high nominal tax rates, and nominal interest rates are relatively high, it’s plausible to look at the economy and say, “we need to align incentives better so that capital flows to productive investment.” And one way to do that is to lower marginal rates on upper-bracket individuals and on investment income. But marginal rates on investment income are now at historic lows, and interest rates are in the basement. If you cut the taxes of upper-income earners, you won’t dramatically increase demand – because there’s a limit to how much people will spend. You’ll add to the pool of savings (which is currently mostly parked in low-return assets) and you’ll bid up the price of scarce goods like expensive real estate. Those aren’t very stimulative behaviors. Yglesias got this part right on Thursday: now is not the time to cut upper-bracket tax rates.

Cutting loopholes is a good idea generally, because it enables you to raise more revenue at lower marginal rates, and eliminates unproductive distortions of the investment landscape. But making the tax code more regressive is a bad idea, right now particularly. So you want to do tax reform that broadens the base and maintains the progressivity of the code. This is not a complicated thing to explain. Yglesias agrees with this, and is perfectly capable of explaining it. Why isn’t he doing so? Is there something in his Slate contract that requires him to frame his points in as contrarian a fashion as possible, even if it means he contradicts himself from Monday to Thursday?