Friday, October 03, 2014

My remarks at the Piketty symposium today

This is a hard paper to present.  I’m tempted to say: Why don’t you all just spend 20 minutes looking through it yourself, and then we’d be happy to take questions.  But instead Joe [Bankman] and I will offer a few highlights.

The paper’s motivation is that we were struck by the large intellectual gap between Capital in the 21st Century and a bunch of literatures that influence our work – for example, those on optimal income taxation, fundamental tax reform, and a lot of mainstream public economics.  Obviously, Thomas knows this literature well.  But he has written a popular book, and one that’s engaged in a very different sort of project than most of the literature.  Plus, he objects to certain of the literature’s standards and practices.

We were both bothered and stimulated by the disconnect.  We aim to adjudicate it to a degree, and to examine how each undermines or enriches the other.  But this makes the paper hard to present.  Just discussing any one aspect among many – say, the theory of lifecycle saving, or “ability” in the optimal income tax literature – could take 20 minutes all by itself.

So, what’s the bottom line?  Let’s start with the tax policy literature.  Logically and analytically, it does fine, at least granting assumptions that are useful and reasonable within particular realms.  And that’s important.

You know the old joke, told about Ford’s Theater in April 1865.  “Other than that, Mrs. Lincoln, how did you like the play?”  Well, I for one actually care a lot about the play.

But the book suggests that often important things have been missed, and simplifying assumptions treated as if they were entirely true.  For example, if you over-focus on lifecycle saving relative to bequests, or if you model utility as purely a function of own consumption – leaving declining marginal utility as the only welfare-based motivation for concern about inequality – then you may miss important things.

And suppose the book is correct in attributing rising high-end inequality mainly to the excess of r over g.  If it’s correct, high saving and/or high returns to saving and/or bequests have negative distributional externalities that are important yet have been ignored.

Word choice can tell you a lot.  Consider the terms “saving” as compared to “capital.”  The tax policy literature tends to talk about “saving,” which is a verbal noun, denoting the aftermath of a choice.  It uses “capital” mainly as an adjective – as in capital income, capital asset, or capital gains, though with the all-too-telling exception of “human capital.”  Thomas’ book, of course, is about “capital,” specifically other than human capital, which he objects to amalgamating with the rest.  And the book treats capital not just as a thing, but also as the marker for a social group, as in “capital versus labor.”  In the tax policy literature, by contrast, we typically discuss “high-earners versus low-earners.”

The savings literature is fundamentally ex ante and about individuals’ preferences and decisions.  Such a perspective is important, but it can lead to missing the forest for the trees, and also to unconscious normative identification with savers’ particular interests.  This is not a surprise, perhaps – prominent academics are often pretty well-heeled, even if not all the way at the top.

Capital in the 21st Century, by contrast, is fundamentally ex post, emphasizing the measurement of realized outcomes.  But risk, among other underlying components, is hiding behind the scenes.  Thus, while the years 1815 to 1900 were a lot better for capital than 1914 to 1970, who knows how it would turn out in the “What If?” scenario where you could turn back the clock and let history unfold again.

An ex post approach is also valuable, but can result in amalgamating things that are distinct, and in ignoring important nuances that are relevant to choices between policy instruments.  Consider the tax policy literature’s decomposition of r into multiple elements, including the “normal” risk-free return that surely is below g, even if it’s more than, say, the 3-month rate for U.S. government bonds.  There’s also the risk element, including both the expected risk premium, if any, and the actual risky outcome.

Ex ante risk can affect tax incidence, since investors can adjust it in light of the tax regime.  In particular, the impact of a capital income tax on risk can be addressed by choosing investment positions in light of the tax treatment of gains and losses.  In effect, you can undo at least some of the automatic insurance that results from taxing winners more than losers.

Now consider the tax policy literature on gifts.  Henry Simons famously endorsed double-taxing them, based on their commonly representing consumption by both the donor and the donee.  The logic is strong, in terms of measuring individual welfare, whether or not you accept the conclusion.  Louis Kaplow helps explain why we might want to subsidize gifts, relative to Simons’ baseline, given the altruistic externality when a donor makes double consumption possible.  But you may want a high tax on gifts and bequests if they have big negative distributional externalities, and if you don’t assign much weight to high-end altruistic externalities.

I realize I’m being very summary and cursory here, especially for the students in the audience.  But again, the paper offers a fuller discussion.

One last set of points before I pass the baton to Joe.  At least in the U.S., as the book agrees, the main driver of rising high-end inequality in recent years has not been the relationship between r and g.  Instead, it has been rising wage inequality, suggesting a central role for human capital, or what we call “ability” with deliberate scare quotes.

Now, even in the 19th Century literature that the book so delightfully deploys, we see evidence of ability’s important role.  Is Pere Goriot about a rentier society?  Well, certainly yes to a degree.  But it’s also one of many classic 19th century novels that focuses on an adventurer or arriviste who aims at the highest social heights despite starting out with very little.  Yes, Rastignac accepts Vautrin’s advice against wasting his time with law studies – you see, law firm hiring was really bad back then – but that’s just because the real action was in the salons and opera houses.

I wish I could discuss at length our paper’s twentieth century updating of the rentiers versus adventurers literature to include P.G. Wodehouse and Bertie Wooster.  Bertie, of course, is the rentier par excellence, turned object of mockery, from the period of the Great Easing.  Surely his tribulations are more evocative than any economic study in showing what had and hadn’t happened to rentiers since the turn of the century.  But I suppose I should move on.

We like the book’s critique of self-satisfied moralizing about “ability.”  High-earners often like to think that it means IQ or character or honest toil or helping humanity.  But the ability to generate high earnings is purely about the relationship between a given individual and the environment in which she happens to find herself.  If enough of the people in a society are vicious racists, then having white skin may increase one’s potential earnings.  Math skills help more in some environments, resistance to dysentery in others.

There’s an analogy to evolution.  No set of genes is fit in the abstract – it depends on the environment.  And to moralize evolution’s winners would be silly.  Now, it’s true that economic competition with perfect markets and the invisible hand is somewhat more benign than nature red in tooth and claw.  But that merely supports an efficiency argument against too much downward redistribution.  And even that argument depends on the relationship between high-end wages and marginal social productivity.

We agree with Thomas that high-end wages often don’t reflect marginal social productivity.  But while the book attributes this mainly to corporate governance problems, the really huge salaries of recent years have often been earned at arm’s length – whether we’re talking about hedge fund managers, or other entrepreneurial free agents in the financial sector, or the founders of a wildly successful new business venture.  The gap between high-end wages and social value created is often less about corporate governance problems than about the distinction between marginal private productivity and marginal social productivity, as in the case of rent-seeking and heads-we-win, tails-you-lose bets in the financial sector.

Okay, over to Joe.

No comments: