Capital taxation: What is it good for?

President Obama tonight will announce a proposed change to the U.S. tax system that would make it much more progressive. The plan would reduce the amount of federal taxes paid by middle-income earners while increasing taxes for those at the top of the wealth and income ladder.

The proposal has three main changes to the tax system for those at the top. The first would eliminate the so-called step-up in basis for taxes on capital gains. When a person inherits, say, a large amount of stock holdings from a parent, the inheritor is only taxed on the gains made after they inherit the stocks. So if a parent bought a stock at $1 and it appreciates to $99 before the child receives the stock, then the child would only be taxed on the gains over $99. This is how large amounts of wealth are passed untaxed from generation to generation.

The second proposal would levy a small tax on large banks when they borrow money. This proposed reform is designed to reduce bank borrowing to increase the stability of the financial system. In modern finance, banks no longer rely solely on deposits to amass the capital needed to lend money, instead borrowing increasing amounts of money for lending purposes, which in modest amounts is not risky but threatens financial stability when overdone.

The third would increase the tax rate on long-term capital gains. This last proposed reform is particularly important for policymakers to consider, but first let’s examine the economic thinking around taxing capital gains.

The most famous way economists think about capital taxation is known as Chamley-Judd model, named after several papers first penned in the mid-1980s by economists Christopher Chamley of Boston University and Stanford University’s Kenneth L. Judd. The general thrust of these papers is that individuals in the economy are extremely forward looking when it comes to their savings decisions and therefore capital formation. Capital gains taxes would have a negative impact on the long-run growth potential of the economy by reducing savings-induced investments and capital formation, therefore lowering the potential living standards of workers.

In short, the model finds that the supply of capital is infinitely elastic to changes in the capital taxation rate, which means the long-run supply of savings and capital would change drastically depending on the tax rate. And when something has an infinite elasticity, it makes no sense to tax that good or activity at all because the reduction would be, well, infinite.

But policymakers and economists alike need to consider whether we are really in an infinite elasticity world. First, a paper by economists Thomas Piketty at the Paris School of Economics and Emmanuel Saez at the University of California-Berkeley builds a model of capital taxation that accounts for a variety of real-world factors and accurately accounts for a several empirical facts. They find an optimal capital tax rate that is larger than zero. So Chamley-Judd’s supremacy in discussions about capital taxation may not be warranted.

What’s more, a few quick glances at several types of data also cast doubt on the assumptions that inform the Chamley-Judd model. As Piketty and Saez point out in their paper, if those assumptions are correct, then we should see large swings in the capital-income ratio as tax rates change. But we don’t see that in the data.

Finally, based on the Chamley-Judd model, the savings rates of the rich (those who hold most of the capital) should be quite sensitive changes in tax rates. But as Slate’s Jordan Weissmann points out, there doesn’t seem to be much of a relationship. Similarly, economist Len Burman, the director of the Tax Policy Center, finds no relationship with economic growth.

All of this isn’t to say that capital taxation is a free lunch. At high enough levels it certainly would have negative consequences. But the idea that any level of capital taxation would be an immense problem for long-run economic growth seems to be a misplaced concern.