Horizontal equity

I'm teaching a Tax Policy course this semester, mainly on distribution issues (my other course focuses on efficiency issues). One funky thing about teaching Tax Policy at NYU Law School is that, since it's a required course for tax LLMs, you can get a lot of people who don't really want to be there. This is no fun if you're the teacher, even if you are agnostic rather than self-righteous about whether, from their standpoint, they ought to care. I actually raised this issue in class on the first day, asking anyone who might have been there for that reason to be a good sport & give it a shot. One way or another, my sense has been that it's working, and that a lot of the people in the class are engaged and interested. I've tried to do my bit, both by encouraging discussion and by trying to pick provocative papers rather than those that are ostensibly (or actually) canonical.

One of today's readings is a well-known 1976 article by Martin Feldstein about horizontal equity. Interesting to me to read this piece now. Provocative and surprising though I would think it was when it came out, time has truly passed it by, which is part of what makes it fun to read.

Feldstein goes against the Haig-Simons orthodoxy (at least among lawyers) of the time, by defining horizontal equity in terms of legal continuity rather than, say, comprehensive income taxation. Thus, no HE violation if you don't tax municipal bond interest and the tax benefit is capitalized into the price, causing the after-tax return to be the same as that on taxable bonds. Pretty obvious once stated, although at the time not widely understood. (Boris Bittker had written about it, however.)

Less impressively, Feldstein's 1976 view of the economics of information seems to be that X is considered 100% certain, then there's a total surprise and it is replaced by Y, which now in turn is considered 100% certain. Meanwhile, he doesn't think of people as generally engaged in portfolio choice under uncertainty with reasonably complete financial markets and the aim of maximizing expected utility given a constantly updated set of expectations. (A jargon-laden mouthful, I realize, but it captures the way that an economist with Feldstein's training ought to conceptualize issues automatically, at least as a starting point. And the rational expectations school in macroeconomics had arisen by 1976.) Meanwhile, the fundamental political economy issue of how we ought to define the optimal scope of binding government pre-commitment - which obviously shouldn't be assumed to arise either in all cases or in none - isn't even in sight.

It's also amusing to see Feldstein in 1976 being so cautious about the case for consumption taxation. With an air of being daring, he says that maybe capital income should be taxed at a lower rate than labor income,. Of course, the modern consumption tax view is that the return to waiting (capital income stripped of risk premia and other such conceptually separate elements) should be taxed at zero.

Harold Wilson once said that a week is a long time in politics. Perhaps academics should be relieved that, for us, a long time is measured instead in decades.