Economists talk a lot about why we shouldn’t tax capital gains. The essential idea is sound – you’ve earned your money, it’s been taxed once, and the market value of that cash is the same whether you consume it now or invest it, get interest, and consume more later. If you tax the returns to investment, you make saving less attractive than it should be, and consuming now more attractive than it should be, which leads to there not being enough investment.
Now, a few economists are more sympathetic to the idea of capital taxes. If elasticities are low – rich people will still save lots anyway – it might be a relatively painless way of reducing inequality and getting revenue. And giving a preferential tax rate to capital gives people an incentive to pretend their wages are really capital (see carried interest loophole, the – or the absurd lengths English football players go to). But overall the consensus in the economics profession is that taxing capital is a really bad idea.
This is fine as far as it goes. But you get the worst of both worlds if one kind of capital is taxed more than another kind. You get less investment, and investment gets shifted towards the kind of capital that gets taxed at lower rates. And there is one kind of capital that gets taxed at exactly the same rate as wages – human capital.
Imagine you’re given £20000, and you’re given the option of investing that money in the stock market or spending it going back to school so that you can earn more from your labour in future. Any gains you get from your investment in stocks will be taxed like capital gains. But any gains you get from education will be taxed like wages, so you’ll probably pay more. And the progressive structure of the tax code makes this even worse because some of the ‘extra’ earnings might be taxed at an even higher rate than the rest of your income. So not only have the gains been taxed as wages but your overall average tax rate is higher than if you hadn’t invested in your education at all. The result of this is that people will invest too much in factories and machines, and not enough in education and training* - and the potential positive spillover effects of human capital only make the problem worse.
This is why we should listen to people like Scott Sumner when they caution us against the potential long-term effects of high marginal tax rates. Raising the top income tax rate might cause doctors today to work more, or it might cause them to work less, or it might cause them to dodge taxes. It might have a whole host of other effects. But one thing it does do is reduce the incentive for kids to put themselves through medical school in the future, because their investment in human capital is going to be taxed at a very high rate.
Of course, one way to fix the screwed-up nature of this system might be a shift towards greater consumption taxation, which might also solve someother problems. But how likely is that to happen?
*Of course, if you believe in the signaling model of education, then maybe this is a good thing.