Capital in the Twenty-First Century.

• Author: Levmore, Saul
• Position: Book review

CAPITAL IN THE TWENTY-FIRST CENTURY. By Thomas Piketty. Translated by Arthur Goldhammer. Cambridge and London: The Belknap Press of Harvard University Press. 2014. Pp. viii, 577. $39.95.

INTRODUCTION

Rising inequality in the developed world has become a hot topic, especially in the shadow of the Great Recession in the United States. Social movements ("We are the 99%!"), university courses, documentary films, and best-selling books have capitalized on--and contributed to--the heat. Thomas Piketty's Capital in the Twenty-First Century, (1) the most significant and probably best received of these books, is provocative, data driven, very French, pessimistic, widely reviewed, admirable, and maddening. In contrast to many other works on inequality, it is organized around a single idea. The thesis predicts growing inequality of wealth in the absence of external shocks or interventionist policies. This argument is set forth in lucid fashion and then surrounded by a great deal of evidence from around the world; this evidence dates from the late-1700s to the present. The data, including available technical appendices, provide context and confirmation.

This is a serious book. In its final chapters, the book turns to its eponymous time period and suggests a global wealth tax and other means of reversing the present course. Here, it is more speculative than empirical. These prescriptions have unsurprisingly garnered a large fraction of the attention paid to this book, although Piketty's data choices have hardly gone uncriticized--or undefended. (2) Data collection and analysis have been Piketty's impressive stock-in-trade for many years, but this Review focuses on his central thesis and normative prescription.

Part I sets out the book's central thesis. Whether or not it is completely correct, the thesis may well emerge as one of the great ideas of social science. This is not just another book about inequality, economic history, or the relationship between labor and capital in production functions, although these subjects do find their way into the book. Piketty presents a big idea. It may not be quite as jolting as comparative advantage or deadweight loss or rational expectations, to name a few of economists' truly lasting ideas from several centuries. But it comes close. The number of copies sold and the number of professional reviews suggest that professional and lay readers alike recognize the remarkable potential of Piketty's idea. Part I, therefore, attempts to introduce the thesis in plain terms and to show its counterintuitive qualities. Part II widens the picture by discussing alternative explanations of some of the data, as well as selected objections to the logic advanced in the book. Part III turns to assets that are excluded from Piketty's calculations. Part IV then explores the book's suggestion about wealth taxation and other means of offsetting the march to destabilizing inequality. The discussion uses the occasion of Piketty's great splash to introduce the idea that optimal fiscal policy needs to include considerations of political decisionmaking, or public choice. Concerns about inequality might provide the impetus necessary to make us rethink the way we tax and spend.

1. THE BASIC THESIS: r > g

   The central idea begins with the observation or intuition that the rate of return available to a passive investor generally exceeds the rate of growth in income available to most people in an economy (pp. 25, 571). One who sits back with inherited wealth, reasonably well invested, will have an increasingly large claim on resources compared to the hard-working laborer across town who relies on earned income. Over time, the gap between the two will grow, and, on a larger scale, wealth (as well as income) inequality in the economy will increase. For a variety of technical reasons, Piketty normally states this thesis--or trend--in terms of the inequality, r > g, where r is the rate of return on capital and g is the growth rate of the economy (p. 25). Readers who have shielded themselves from this claim or who are meeting it (or large parts of economics) for the first time would do well to think about r > g before proceeding.

   Some simple observations about r and g may be useful. An individual can flourish economically and increase her earnings faster than the economy grows, but all individuals cannot. There are different ways we could measure national income, but the notion is that average (and aggregate) individual income is tied to national income. This growth in income, g, might depend on population, immigration, and technological change, but it is observable, and Piketty will rely not on theories about g but on observations over time (pp. 16-20).

   It is easy to observe one's own savings stagnate in an era of low interest rates while getting raises at work, and to imagine that g must exceed r, and perhaps by a fair amount. This was especially so when many families' savings took the form of equity in housing and that asset class declined in value. At the same time, when the rate of return from invested savings, or capital, is high, it is often so because the investor has some tolerance for risk. When people imagine what their world would be like if they had bought Google stock when it was first available or when they observe peers in Silicon Valley becoming wealthy through stock options, they are not really accumulating evidence of high r but rather of returns to risk. Myopia is similarly apparent in assessing g. An individual might experience low g, but the population (and even the working-age population) might be increasing, so that per-capita g is low but g is relatively high.

   Piketty is terrific at helping the reader comprehend these things. One feels in particularly safe hands when it comes to the author's specialties: acquiring and explaining available data sets and analyzing evidence about national income accounts and the return on capital, including profits, dividends, rents, and interest. The more one reads Capital in the Twenty-First Century, the more one will become convinced that over the long run--and in many places--r is indeed normally greater than g. If it helps to have actual numbers in mind, then it might be useful to think of the long-run return on capital, r, as 3 or 4% (pp. 208, 358-59) and the long-run growth in income, g, as 1.5% (pp. 73, 93). Higher growth rates are often associated with developing economies, technological advances, and sudden improvements in infrastructure. Once economies mature, however, we can expect some convergence in g, if not in r as well. (3) Almost amusingly, Piketty is quick to assure us that even a g of 1% is formidable when compounded over many years (pp. 95-96). Fifty years of that kind of g yields a complete change in lifestyle. It goes without saying, though it is never quite said, that half a century of sitting back and earning 4% on one's inheritance would produce yet more dramatic material changes.

   It bears emphasis that it is the long run that counts. The book, and perhaps all serious discussion of inequality, is about the long run. If we see hedge fund managers making a great deal of money for several years while school teachers are experiencing stagnant pay, we might have reason for concern, and we can expect passionate discussion of such things as comparable worth and whether markets really work. But seasoned observers know that industries and incomes rise and fall. The starting salaries at large law firms grew about 5% a year for nearly two decades until 2008. (4) No insular law student would have believed that g was really 1 or 2%. But the cost, and even the net cost, of law school was also skyrocketing, the stock market and inflation had their own histories over the same period, and the prospects of law school graduates fell quickly after 2008. At various points during those heady years, one might have made all sorts of claims about growing or shrinking inequality. But inequality is a macroeconomic topic, calling for data over a long period from many or all industries. Even thirty-year snapshots are entirely inadequate, especially when the researcher can pick and choose among periods. Piketty might lead us astray with this tactic here and there, (5) but where the all-important g and r are concerned, his data go back as far as good data are available, and the reader is treated to discussions of income, capital stocks, and inequality trends over generations and even centuries, and across many countries. If r exceeds g, then those who inherit wealth will outperform--if that is the word--those who work and who live off earned income rather than income from capital. Over time, inequality will become more dramatic. The haves will thrive while the have-nots will need to hustle to get a share of the mere 1 or 1.5%. To be sure, r is not always greater than g, and there can be reversals. If it were not so, Piketty's inequality would be obvious. The data reveal that 1914-1945 was an unusual period (pp. 274-78, 284, 293-94). The two world wars and the Depression were trend busters. They altered the value of assets and ownership structures; they caused population shifts; and, perhaps as important, they facilitated dramatic changes in tax rates and social-welfare policies. In many ways, Piketty's insight is to show that the inequality trends of the last couple of decades are of a piece with hundreds of years of economic history.

   Is there a problem? If r> g were embedded in a larger pattern in which g was relatively impressive--or even perhaps where g increased with the inequality--then for many observers there would be no problem to solve. Inequality for most people is a way of thinking about the well-being of those at the bottom of the income distribution. (6) If their lot is improving rapidly, few begrudge the wealth at the top, and that would probably be so even if the growth at the top were yet faster than that at the bottom. For example, if...