Yesterday at the colloquium, Kim Clausing presented Capital Taxation and Market Power, which focuses on the importance of extra-normal returns that are earned these days, especially by big multinational companies such as the FAANG crew (Facebook, Apple, Amazon, Netflix, Google).
Extra-normal returns are important not just to rising high-end inequality in the US and around the world, but also analytically. For example, as I discuss here, they can reverse both the efficiency and the incidence analysis of entity-level corporate income taxes. They can also reduce the need for income rather than consumption taxation at the entity level, even if one is pro-income tax as to the taxation of individuals.
I see the paper as raising 4 main types of issues, as set forth below.
1) What are extra-normal returns, and what gives rise to them?
Their most pertinent features are:
(a) at the risk of belaboring the obvious, their being by definition high, hence something to care about.
(b) their being generally non-scalable, unlike risk-bearing or making investments that merely earn the normal return. Thus, Jeff Bezos presumably could not have responded to the existence of the tax system by making 2 Amazons, rather than just one.
(c) their being potentially highly efficient to tax. But this depends on understanding what gives rise to them.
Suppose we call them rents, and think of rents as free money that falls from the sky. Then taxing them at even a 99% rate does not induce substitution and is perfectly efficient. But this may not be an entirely credible way to think about them.
Rents are also sometimes defined as amounts in excess of one's reservation price. Thus, consider LeBron James, whose salary for playing this year is nearly $50 million. If he would play for just $5 million, then the extra $45 million is a rent and could be taxed away without affecting his behavior. (Note that this perhaps unrealistically ignores the scalability of LeBron's effort level - he works very hard, all year round, to be ready to play.) But in any event it is hard to know people's reservation prices.
The paper mentions the distinction between rents and quasi-rents. The latter, like the former, are defined in the literature in multiple ways. The paper notes that observed (ex post) extra-normal returns can reflect compensation for risk, implying the possibility of a normal return (any risk premia aside) ex ante. One wouldn't judge rates of return on the New York State lottery by just looking at the winners' return on investment. Quasi-rents are also sometimes defined as being merely temporary, or as being enjoyed ex post (with no further effort being required) once they have been successfully created. And there, I think, we are on to something more general.
I think of seeming rents as generally quasi-rents that are conceptually the products of labor income. Which doesn't mean that they're generally either deserved or the products of toil and "sweat equity." But they generally involve one's having done something or made some particular choice in response to one's opportunities. This could involve something that's socially productive, or rent-seeking efforts to create and defend monopoly power, or anything in between. But good or bad, deserved or not, I'd say they're geneally not entirely free money, and therefore that, when we're taxing them, we are still in the world of comparing distortions from the use of alternative instruments.
2. Implications of taxing extra-normal returns
This is not to dispute that taxing extra-normal returns that are conceptually labor income (even if viewed by the tax system as capital income) may have efficiency as well as distributional advantages. But it's no longer a slam dunk. The issues that one might think about here include the following:
Time: How should we evaluate the appeal of taxing returns that are ex post rents? It's the familiar time consistency issue, but in a distinctive setting. As Keynes said, in the long run we're all dead, and one also might ask to what extent (and how generally) one is actually going to discourage new investment and labor effort if the setting for the ex post taxation is a distinctive one.
Elasticity of labor supply: How tax-discouragable are the would-be Jeff Bezoses of the future? I am inclined to think: not very, but admittedly this calls for empirical investigation.
Externalities: As the paper discusses, these are at least plausibly rife, on both the positive and negative sides. E.g., positive externalities (or the creation of consumer surplus) from innovation, versus inducements to rent-seeking and monopolistic behavior.
Political economy: There are theoretical reasons for positing that companies like those in the FAANG group would be prone to being overtaxed. For example, insofar as they earn non-rival, location-specific rents that are actually quasi-rents, each country might have an incentive to grab (once it is repeated everywhere) "too much." But if the companies are politically very powerful - including US companies outside the US - then the main problem may lie the other way, in terms of their securing favors that include low taxation. I am inclined to think that, in practice, the latter problem significantly outweighs the former one, but again this may be context-dependent and in need of verification.
3. Tax versus non-tax instruments
Both tax and other scholars (e.g., in antitrust and IP) are, to an undesirable degree, siloed in their own disciplines. This happens for good reasons - it's not that easy to become an expert in something - but it can have bad effects when the alternative instruments are interchangeable and interact with each other.
The paper notes that some of the problems it proposes addressing through taxation (e.g., higher taxes on those who may have monopoly power) can potentially be addressed through such alternative instruments as antitrust or IP rules. There may be both substantive and political economy differences between the alternative approaches. E.g., tax and antitrust may raise distinct political and administrative issues; tax may be more continuous and antitrust more of an on-off switch; and a tax approach to monopoly merely seizes some of the profits earned by limiting supply rather than directly preventing it. While clearly this paper could not sensibly have taken on the alternative-silo issues in any detail, they do merit being more fully addressed somehow.
4. Tax proposals in the paper
The paper notes several interesting tax law issues as to which its analysis might support changes in the rules (e.g., pertaining to tax-free reorganizations, international taxation, and the tax treatment of large pass-through entities that may be earning extra-normal returns). But the most novel one is to restore graduated rates in the corporate tax, on the view that companies with very large amounts of income may tend to have a higher proportion of extra-normal returns. This would be accompanied by rules requiring consolidation, at least for tax rate purposes, when commonly owned companies were nominally separate. While there are many possible objections to such a rule, I agree that it should be on the agenda of current debate (including for political economy reasons, if smaller companies have especial political clout in Congress).
No comments:
Post a Comment