The Return of Capital

Thomas Piketty, Capital in the Twenty-First Century (2014).

If I had nothing more specific in mind, it would verge on being trite–or perhaps achieve triteness with margin to spare–to identify Thomas Piketty’s Capital in the Twenty-First Century as a 2014 publication worth noting at Jotwell. At least as of early spring, the reviews–predictably laudatory from the left, and cautious or critical from the right–have been sprouting like dandelions in Central Park, and there have also been regular (indeed, almost daily) features about Piketty and his opus, appearing in periodicals of all kinds. Most readers no doubt already know that Piketty has combined a compilation of groundbreaking empirical research about wealth distribution in multiple countries over the last few centuries with an important and provocative thesis about the likely (or at least a possible) future.

Piketty argues that high-end wealth concentration has a tendency to keep on augmenting itself in modern capitalist societies, at least for an indefinite time. He views the mid-twentieth century’s “Great Easing” of this process as reflecting distinctive and anomalous factors that make it unlikely to be repeated. He attributes it mainly to the era’s enormous shocks–in particular, the Great Depression and two calamitous world wars–and secondarily to the pursuit of economic and regulatory policies that deliberately sacrificed neoclassical market efficiency in pursuit of other objectives, or else in response to concerns about market failure that came to be dismissed with the rise of such political leaders as Ronald Reagan and Margaret Thatcher.

The catchphrase that everyone has come to know is “r > g”–that is, the rate of return to capital exceeds the growth rate of the economy, an empirical relationship that he views as highly resilient during most periods and as inevitably accentuating economic inequality given how the ownership of capital is concentrated at the top. He views this as leading to political plutocracy or oligarchy, and to the rise of a “rentier” society like that depicted in the works of Jane Austen and Honore de Balzac, in which potential returns to labor are dwarfed by those to inherited wealth. He also argues that rising wage inequality in the U.S. driven by the rise of managerial compensation, both (a) reflects corporate governance failures that deprive the amounts earned at the top of any strong relationship to marginal private (much less social) value provided, and (b) will end up mainly fueling, rather than significantly modifying, the relative rise of capital income over labor income.

Again, this much is already well-known and being vigorously debated. What I want to add here concerns the sizeable, but as yet little discussed, disjuncture between the frameworks used by Piketty on the one hand, and in much of the last three decades’ tax policy literature on the other hand. One can see the disjuncture on both sides. Piketty, for his part, appears not to have drawn much from, say, the literature concerning optimal income taxation or the income tax versus consumption tax debate. On the tax policy side, many lawyers but also economists have tended to view vertical distribution issues through a very different lens than the one he employs. It is important to evaluate the extent to which the different lenses are inherently competing as opposed to complementary, and in either case to evaluate their relative usefulness.

The main disjuncture relates to Piketty’s use of two key concepts: “capital” and the “return to capital.” While I have already noted the role that both play in Piketty’s account of the past and our possible future, how do things differ in, say, the legal academic literature concerning optimal income taxation and the income tax versus consumption tax choice?

“Capital,” to which Piketty assigns such a central role, has in a sense been banished from much recent tax policy literature. We know that, while deployed jointly with labor in economic production, it is considerably more mobile. But we sometimes think of it as earning a normal, real rate of return that is little distinguishable from zero. This conclusion reflects teasing out, as conceptually distinct from the normal rate of return, the elements of risk-taking and of inframarginal returns that may generally be labor income in substance (e.g., the entrepreneurial profits of a Bill Gates or Mark Zuckerberg). Thus, the importance of “capital” in the Piketty story raises questions that need further examination.

Turning from the macro level of economic production to the micro level of individual choice under scarcity, James Mirrlees’ foundational optimal income tax model does not even have capital. Instead, it is a one-period model in which people earn labor income and simultaneously spend it on market consumption. While the model’s omitting the element of time is a conscious simplification, adding it back can make surprisingly little difference to the analysis. Suppose we think of saving, which leaves one with capital to invest, as merely deferred consumption in what is mainly a lifetime income (and consumption) model. Even adding inheritance may not change things very significantly. Where not purely accidental (reflecting under-annuitization), it may be thought of as reflecting altruistic or warm-glow utility functions in which bequests yield bonus utility–that of the donor and the donee alike, from the former’s enjoying the prospect that the latter will derive enjoyment from spending the inheritance. Once again, we are clearly very far here from Piketty’s world.

Finally, how can one reconcile Piketty’s historical evidence, showing r > g, with viewing the normal real return to saving as scarcely distinguishable from zero? Here, the key point is that Piketty is measuring historically observed ex post rates of return, without addressing ex ante expectations or the risky versus risk-free element.1 By contrast, recent legal literature addressing the risk issue frequently assigns a large component to risk, viewed as conceptually distinct from the “normal” return to saving, with the breakdown being the two being evaluated based on the observed rate of return on specific, apparently risk-free assets, such as short-term Treasury notes. Here too, the relationship between the two literatures leaves us with much to think about.

One last point that could possibly lead to a narrowing of the gulf is as follows. Suppose one agrees with Piketty’s empirical prediction that, barring some disruption to observed trends, high-end wealth inequality will keep on growing significantly. It is a separate question whether one should consider this a problem. However, there is plenty of room, including in a purely utilitarian framework, to consider this highly regrettable (or at least potentially so) for reasons that are not limited to observing that poorer individuals might tend to have higher marginal utilities for a dollar than richer individuals. For example, one might believe that extreme high-end wealth concentration can have broader adverse effects on individuals’ welfare, including through its effects on the functioning of various social and political institutions. If one accepts all this, then Piketty’s work might be taken as suggesting that saving, and high ex post rates of return on saving (even if reflecting risk), may have significantly negative net externalities, even if they might also yield positive externalities of various kinds. This, too, however, serves merely to suggest possible directions for further inquiry, the outcome of which should not be regarded as preordained.


Editor’s note: For other Jotwell reviews of Thomas Piketty’s Capital in the Twenty-First Century see:

  1. Piketty recognizes that observed capital income may have an intermingled element of labor income, when entrepreneurs work for themselves, but argues that as an empirical matter this issue has only limited importance. []