Capital rates preference, 1913-2012.

Mitt Romney's Cayman Island hideouts and unusually stuffed IRAs get most of the attention, but the main explanation for his low tax rate isn't at all nefarious: 80 percent of his income comes from investments, and the top tax rate on capital gain and investment income is 15 percent. That's compared to a top tax bracket of 35 percent on ordinary income. Why does investment income get this big bonus?

On 60 Minutes, Romney gave one reason. "Capital has already been taxed once at the corporate level, as high as 35 percent," he said, which justifies the lower rate when the income is taxed again. This doesn't really fly, most obviously because the "double taxation" charge can also apply to wage income. Almost all economists agree that while investors like Romney bear the brunt of the corporate income tax, employees of the taxed company pay too. The Tax Policy Center estimates that 20 percent of the tax is ultimately paid by workers. So if Romney gets to plead double-taxation, so do typical workers.

The rationale economists – even liberal ones – give is quite different. Take Emmanuel Saez and Peter Diamond. Saez is best known for the work he's done with Thomas Piketty detailing the rise in inequality over the last century. Diamond is best known for winning a Nobel prize even as congressional Republicans blocked his appointment to the Federal Reserve's Board of Governors. All three have advocated marginal tax rates far above those being considered by either Democrats or Republicans right now. And yet, even they think savings and investment income should be taxes at a lower rate.

The basic idea here is that you investing is a form of savings. And economists don't really want to tax savings, in part because the effects of taxing savings can be a little weird. Saez and Diamond imagine that there's a 30 percent tax on income, whether or not it's saved. They then imagine you save that money for 40 years, and earn 5 percent interest every year. If savings weren't taxed at all, then you could take that money out after 40 years and pay the 30 percent rate. But if the savings were taxed before it was saved and after it's pulled out, the total rate comes to a staggering 60.6 percent. So there is, in effect, a massive incentive to spend money now rather than save it and spend it later on.

Some economists argue that this means there shouldn't be any tax on capital, because as the number of years you save approaches infinity, the tax rate on savings starts to become truly exorbitant. But Saez and Diamond (as well as Saez and Thomas Piketty in another paper) note that this only makes sense if you assume people live as though they're immortal. It's not as crazy an assumption as you might think. Many people earn income not just for themselves to spend, but for their children to inherit. If people keep thinking like that, then treating families as a single, immortal, money-maximizing individual isn't so nuts.

The problem, as Saez and Diamond argue, is that people don't actually treat their children like themselves. They're more selfish than that. And when people actually save to help themselves more than their children, there's a limit to how much time they can save money, and thus how much taxes can punish that savings.

There are other problems too. It's hard, in practice, to distinguish wage and investment income. Romney is a great example of this. His income as head of Bain Capital counted, technically, as investment income, but he got it for doing a workaday job. A zero tax rate on investment income would provide a huge incentive to pretend wage income is investment income.

That said, Saez and Piketty conclude that there should probably be some advantage for investment income, because of the savings disincentive. And it's worth noting that while Saez, Piketty and Diamond support taxing investment, many other economists disagree. Greg Mankiw, Matthew Weinzierl and Danny Yagan make the case for zero taxation here.

So the disagreement among economists isn't about whether people like Romney are paying too little. It's about whether or not they're paying too much.

Correction - This post originally said the 40-year savings tax was 73.8 percent. That was an arithmetic error. It's actually 60.6 percent - still extremely large.