Is the economics profession guilty of ignoring inequality? This charge of scholarly negligence is at the center of a growing literature of academic arguments from other disciplines, usually associated with self-described studies of “neoliberalism” and the increasingly popular “New History of Capitalism” genre of historical writing.
As the basic story goes, distributional questions such as inequality dominated economics throughout the classical period of the 19th century and into the Progressive Era works of figures such as Richard T. Ely and John R. Commons. Around the mid-20th century, however, the emergence of an empirically minded neoclassical school of economics gradually abandoned scholarly interest in questions of equity and fairness to focus instead on growth — even if that growth only benefited the few at the expense of the many.
Only in the past decade, according to this thesis, have economists redirected their attention to the problem of inequality, most notably as a result of Thomas Piketty’s 2014 book Capital in the 21st Century. The history of economics from the mid-20th century to 2000 (or 2014 in some versions, where it is more closely associated with Piketty), then, is accordingly one of a “neoliberal” takeover of the profession that distorted and diverted its attention away from ever-pressing questions of equity and fairness.
Despite its growing popularity, the core argument of this inequality-rediscovery thesis appears to be built upon a mistaken premise and an ensuing misreading of the literature. Economists never abandoned the study of inequality — on the contrary, it remained a persistent theme in both top economic journals and classroom textbooks throughout the alleged period of neoclassical or “neoliberal” dominance in the profession. The emergent body of works on this subject have simply failed to do an adequate literature review, and therefore misattribute the recent surge of interest in inequality studies to the comparatively politicized and, at times, empirically suspect contributions of Piketty and associated scholars.
Versions of the inequality-rediscovery thesis may be found in several recent works. The more nuanced takes appear in a working paper by sociologist Daniel Hirschman on the alleged “rediscovery of the 1%,” and in a newly published essay by historian Eli Cook on “historicizing Piketty.”
Both scholars allege that inequality research effectively dropped off of economists’ radar after roughly 1954, when Simon Kuznets laid out a version of the famous Kuznets curve on the relationship between inequality and growth during his presidential address to the American Economics Association. Kuznets himself pioneered the long-term measurement of top income distributions in the United States in the first half of the century — a subject that was revitalized by the work of Piketty and his frequent collaborator Emmanuel Saez in an updated time series in 2003. In Hirschman’s telling, the Piketty-Saez empirical updates “surprised” the economics discipline, essentially catching it off guard after purportedly ignoring decades of mounting evidence of a surge in inequality through neglect of the subject. Cook goes even further, contending that the entire study of inequality was “marginalized” by economists “for much of the second half of the twentieth century, only to return with a vengeance in the twenty-first.”
A non-scholarly version of the same argument may be found in a 2017 Boston Review article by Marshall Steinbaum, which alleges somewhat incredibly that economists “are giving Piketty the cold shoulder” in retaliation for reportedly showing that “the workings of the capitalist economy are malfunctioning.” Steinbaum’s version also takes the inequality-rediscovery argument deep into conspiracy-theory territory, implying a plot to deny Piketty’s thesis a full hearing before the economics profession.
Although he provides no proof for these exclusionary designs, Steinbaum’s argument also chafes against clear signs of Piketty’s reach and warm reception by other economists. Piketty’s work has been featured multiple times in the Quarterly Journal of Economics — a top journal of the discipline — and his 2003 article with Saez has garnered over 3,000 citations, making it one of the most influential economics papers of the last 20 years.
These and other signs suggest right away that something is amiss with the inequality-rediscovery thesis. The contention that inequality was neglected and shunted to the margins of the profession, only to be revived by Piketty, however, remains popular and appears to be growing — especially among non-economists who study the history of the discipline from the outside.
So how does the inequality-rediscovery thesis compare against the evidence from published scholarship in economics between 1950 and 2000 (or 2014)? As we show in the subsequent section, not very well. Not only was inequality a consistent and sustained research agenda within economics throughout this period, but it regularly attracted the attention of top scholars, including Nobel laureates, and generated a large published literature in the discipline’s most exclusive journals.
Far from proving their case that inequality disappeared in the neoclassical or “neoliberal” era, only to be rediscovered in modern times, the main variants of this thesis share a common neglect of that same published body of scholarly works. In short, they failed to do an adequate literature review before reaching their conclusions.
Measuring Inequality Research From 1950 to 2000
One simple test of the inequality-rediscovery thesis is to take the words “inequality” and “distribution” together and see how frequently they appeared in top economics journals during the second half of the 20th century. We pair the two terms in order to avoid accidental inflation of the numbers. For example, the word “inequality” could apply to the topic of optimal control theory, which is unrelated to the study of the income distribution. Taking only the instances where inequality is mentioned in the context of distribution therefore restricts our measurement to articles that overlap with the research questions of both Kuznets, in the preceding decades, and Piketty and Saez in the present day.
In the American Economic Review — the flagship journal of the American Economic Association — the two terms appeared together in 497 articles between 1950 and 2000. This accounts for roughly 5 percent of the articles published by the journal during the period. To provide contrast, the term “monetary policy” — a consistently prominent topic among economists — appeared in 637 articles (7 percent of all articles). When taken separately, the terms “monetary” and “policy” yielded 1,200 articles (13 percent of all articles). Given the central importance that monetary economics occupied in the discipline, particularly between 1960 and 1990, the topic of inequality was pulling its weight quite well.
To be sure, the level of attention to inequality since 2000 has increased markedly. The increase may simply be due to the lowering of computing costs, which makes it easier to generate information and test hypotheses about inequality. That more information now exists and is being produced does not mean that top economists were not developing high-level contributions on the topic.
Summary statistics run directly counter to the inequality-rediscovery thesis by showing the topic’s recurring prominence. They do not tell us as much about the influence or nature of the work that scholars were doing between Kuznets and Piketty-Saez. For a better indicator we may look at the major contributors to the literature during this period.
Consider the cases of Jeffrey Williamson, Peter Lindert, and Robert Fogel. Fogel, who earned his fame through his work on the economics of slavery, had, by the late 1970s, moved on to new grounds regarding health economics. More precisely, he began to invest time and effort into anthropometric measurements of living standards (heights, nutrition, death rates). The primary focus of this new research agenda was to consider the role of health inequalities.
For example, one of his main findings was that the distribution of calories consumed in many pre-industrial countries during the 18th and 19th centuries suggested that the poorest were so poorly fed that they could hardly provide more than a few hours of work per day. In fact, this evidence is a large part of what became his Nobel lecture in 1993. Fogel argued that this offered a link between economic growth and health inequality — something that textbooks like Greg Mankiw’s Principles of Macroeconomics do discuss (even if briefly and tangentially).
Jeffrey Williamson and Peter Lindert are also good examples of the sustained research interest in inequality measurement across the period of its alleged neglect. Between 1960 and 1990 (and even since then), these two economists have been leading scholars of the economic history of inequality. They have consistently deployed efforts to measure income and wealth inequality in countries like Britain and the United States before, during, and after the industrial revolution. Their main interest was to extend the work of Simon Kuznets from its mid-century mark. Kuznets had proposed in 1955 that inequality rises with the beginning of urbanization (the process that accompanies the industrial revolution) but then begins to falls when urbanization rates have reached a certain threshold. Lindert and Williamson developed empirical evidence over the very long run to test this idea. They showed that income inequality rose in the 19th century before beginning a long fall to the mid-1970s (following the pattern proposed by Kuznets). However, after the 1970s, it began to rise again.
When one concentrates his gaze only on the 20th century, their work also pre-empted the big finding of Thomas Piketty and Emmanuel Saez. With their big breakthrough article in 2003, Piketty and Saez popularized the idea of a U-curve of inequality whereby inequality fell between 1913 and the mid-1960s only to rise thereafter. However, Lindert and Williamson had been discussing this 20th-century empirical pattern throughout the period from 1960 to 1990. They did not stop at income inequality only. They also extended their work to the measurement of inequality in the cost of living, adjustments for regional differences, and the effect of immigration on inequality.
Lindert and Williamson published a multitude of books and articles on the topic. These include works in top venues such as the American Economic Review and the Review of Economics and Statistics, not to mention their recurrent appearances in top field journals such as the Journal of Economic History, Explorations in Economic History, Economic History Review, and Economic Development and Cultural Change. Their 1980 classic book, American Inequality: A Macroeconomic History, was heavily cited despite being a somewhat technical work, and remains the standard text on the subject to appear between Kuznets and Piketty-Saez.
Lindert and Williamson’s work also served to motivate the next wave of inequality research that began in the 1990s, starting with the research of Robert Margo and Claudia Goldin on the “Great Compression” (the reduction in wage inequality below the 90th percentile during the 1940s), which was published in the Quarterly Journal of Economics (one of the top journals of the profession). More important to investigating the inequality-rediscovery thesis, we should note how the economics profession treated these scholars. Both were hired at top universities — Harvard University for Williamson and UC-Davis for Lindert. Among the next generation, Margo currently teaches at Boston University and Goldin teaches at Harvard. Such rewards — just like Fogel’s Nobel Prize — would not have been bestowed had their primary topic of interest been deemed “marginal.”
These examples of scholars who worked on inequality during the second half of the 20th century are not exhaustive. Other pioneers include top scholars such as Morton Paglin, Peter Gottschalk, Barry Chiswick, Samuel Bowles, and even Albert Hirschman.
In fact, and quite tellingly, the interest cut across the different schools of economic thought. For example, during the 1980s and early 1990s, Gerald Scully published works that brought public choice insights to the topic of inequality in journals like the Southern Economic Journal, the Journal of Econometrics, and Public Choice. By the early 2000s, before the publication of the breakthrough article of Piketty and Saez, Scully had published additional (and now well-cited) works on the topic of trade-offs between growth and inequality.
Significant contributions on the empirical measurement of top income shares also traverse the era in which inequality was allegedly marginalized. In the 1960s, the NBER commissioned a now well-cited book from Robert Lampman estimating wealth inequality in the United States between 1922 and 1956. That work served to motivate Francis Pryor to do his own work on the topic, which ended up being published in the American Economic Review. The Review of Income and Wealth also devoted considerable attention during those decades to the topic.
Historical measurements were also enriched in the early 1970s when Lee Soltow extended distributional measurements back to the American Revolution in a pair of papers for the Journal of Economic History, while also publishing an influential book on top wealth accumulation in the 19th century. Since 1989, a team of Federal Reserve statisticians and economists led by Arthur Kennickell have administered the periodic Survey of Consumer Finances — a highly regarded survey instrument that measures top wealth concentrations by sampling the ultra-wealthy.
Inequality measurement also regularly appeared in top journals and scholarly presses throughout this period. Anthony Atkinson’s book Wealth, Income, and Inequality went through several editions in the 1970s and 1980s, and included pioneering work on the measurement of top wealth shares in the United Kingdom that he continued to extend until his death in 2017.
Although often portrayed as a “lone voice” in a negligent discipline by modern historians, Atkinson held an academic post at the University of Oxford, developed an inequality-measurement index that bears his name, and received a knighthood for his scholarly contributions.
Edward N. Wolff, who is often associated with heterodox approaches that are more politically inclined to do inequality research, published widely cited estimates of wealth concentration in the American Economic Review, using survey data between the mid-century and the late 1980s. The 1990s saw a similar continuation of empirical work on estimating top incomes by Daniel Feenberg and James Poterba, as well as continued revisions to historical income distributions from the early 20th century by Gene Smiley. Tax economist Joel Slemrod edited an entire volume of papers on the relationship between income inequality and tax progressivity in 1996, and Slemrod’s student Wojciech Kopczuk has been publishing empirical work on inequality measurement since the late 1990s.
Several recipients of the Nobel Prize for economics also maintained an interest in inequality throughout this period. In addition to Fogel, Angus Deaton received the 2015 award for his work on inequality and poverty — a sizable share of which was conducted during the time of its erroneously alleged marginalization and was published in top field journals or top general-interest journals such as the Journal of Political Economy. And while they received the award primarily for other contributions, both Joseph Stiglitz and Paul Krugman have published on inequality in top venues. To this list, one can also add the important works of Amartya Sen on distributive justice and hunger, which required complex connections between economic theory and philosophical concepts about redistribution.
Turning to classroom instruction, we also find that inequality has remained a constant theme in economics-textbook instruction throughout the period of its alleged neglect. Paul Samuelson’s Economics became the most widely used textbook of this period. Chapter 4 of the first edition from 1948 covered the “Distribution of Income in the United States” including “The Inequality of Income.” The section was retained and expanded over the next half-century. The 1970 edition, for example, has a section on “How to Measure Inequality Among Income Classes” and an accompanying discussion of anti-poverty policy.
The more recent standard textbook, by Harvard’s N. Gregory Mankiw, was published in 1998 and is often maligned by ideologically minded academics as the public face of “neoliberalism” in economics. Yet it too contains an entire chapter on “The Distribution of Income.” The chapter surveyed data on the gap between the richest and poorest fifths of the U.S. income distribution and tied the discussion directly with a survey of poverty-alleviation policies.
The plain reality of both Samuelson’s and Mankiw’s textbooks stands in sharp contrast with the inequality-rediscovery thesis, and similarly belies a common complaint about economics instruction from persons who work outside of the discipline. One of the major talking points used to promote the recently published CORE Open Source economics textbook is its self-depiction as a corrective to a deficit of classroom instruction about inequality by including a chapter devoted to the subject.
“Most introductory courses in economics do not equip students with the tools they need to explore complex issues such as … growing inequality,” reads a promotional website for the project. And yet as we can plainly see, entire chapter-length discussions of inequality are nothing new to principles textbooks. They’ve been a mainstay of the discipline since at least 1948.
As can be plainly ascertained from the foregoing evidence, the study of inequality was alive and thriving throughout the period of its alleged 50-year neglect by the economics discipline. It consistently attracted the attention of top scholars, including future Nobel laureates. It consistently appeared in published research in top-tier journals such as the American Economic Review as well as in multiple top field journals and books at top university presses. The topic also received prominent and consistent treatment in two of the main classroom textbooks in use throughout this period.
All of the aforementioned work preceded Piketty’s Capital and the 21st Century by over a decade, as well as Piketty and Saez’s 2003 article — the two events that allegedly reversed the discipline’s attention according to the inequality-rediscovery thesis. To their credit, Piketty and Saez also cited and acknowledged that they were building on this half-century of work. We may ask ourselves then how the more recent commentators got the story so wrong. The answer may be found in their misidentification of the Piketty-Saez contribution.
How Piketty and Saez Advanced the Inequality Literature
When Piketty and Saez published their new time series on top U.S. income distributions in 2003, they built upon an existing literature that traced back to Kuznets a half-century earlier. The most important original contribution of the piece was not its alleged “rediscovery” of a neglected and marginalized topic — indeed their literature review cited several of the very same authors and contributions we note above, including Feenberg and Poterba, Lindert and Williamson, Goldin and Margo, Atkinson, and Slemrod, among others. Rather, Piketty and Saez’s novelty arose from their attempt to construct a single century-long measure of top income shares from a continuous data source.
Prior works since the time of Kuznets had in fact measured income distributions over the intervening half-century. Feenberg and Poterba, for example, used a different methodology to estimate the top 0.5 percent income share, and other authors had pieced together a long-term series from other sources such as surveys or closely related metrics from wages. Piketty and Saez, however, adapted a single common source — the IRS’s annual Statistics of Income report — to derive a contiguous series covering the entire period between the first income tax in 1913 to the present. The result gave a complete century-long depiction of top income distributions taken from the same data source.
Having a common series is advantageous as it avoids the challenges posed by piecemeal assembly from disparate sources and different levels of comprehensiveness. In that sense, Piketty and Saez contributed a new and standardized approach to the inequality-measurement problem that has since been adapted to dozens of other countries, where tax data permit.
The new approach had a clear effect upon the inequality literature, as proponents of the inequality-rediscovery thesis appear to recognize. Those proponents however misidentify the empirical contribution of a single-source series, and interpret it instead through its uses as a political advocacy tool for progressive taxation. This leads them to overlook the earlier literature, which generally avoided taking strong political advocacy stances.
Indeed, Piketty and Saez’s own interpretations of their work have moved in a much more overtly political direction since it first appeared. Piketty’s 2014 book, for example, asserts strong causality between high taxation and lessened inequality. He makes an overt political argument for global wealth taxation as a prescriptive correction. Saez currently advises Elizabeth Warren’s presidential campaign, and designed the main tenets of her proposal to enact the same in the United States.
Compare that to the much more circumspect interpretation of their results offered in the 2003 paper: “We have tentatively suggested that steep progressive taxation, by reducing the rate of wealth accumulation, has yet prevented the large fortunes to recover fully from these [leveling] shocks,” of the Great Depression and World War II.
Interestingly, subsequent developments within the economic literature on the Piketty-Saez series have retained their long-term outlook and methodology but also directed attention to several areas of weakness that arise from the approach. IRS statistics, though they have the advantage of dating back a century, have highly inconsistent levels of economic quality due to changes in the way taxes were collected and enforced, and due to selectivity issues in the way that the rich structure their tax reporting.
Indeed, even before Piketty and Saez’s work, Slemrod along with Roger H. Gordon drew attention to the distortive effects of income shifting in response to tax-rate changes. After all, taxpayers engage in legal tax planning to insulate their income from a heavy statutory tax burden but may relax such practices under lower rates, or shift income streams to different reporting structures to take advantage of rate differentials.
For the Piketty-Saez series, this means almost every major tax-code overhaul in the past century has the potential to affect the economic quality of IRS data and alter the amount of money that gets reported as taxable personal income. Several recent works have called attention to these problems and attempted to correct for them in the Piketty-Saez series since the mid-century mark. Our own work, which revises the entire Piketty-Saez series between 1917 and 1962, addresses similar issues in the first half of the century, when IRS data problems were more severe. The thrust of this newer literature indicates that the empirical approach by Piketty and Saez remains a live issue of inquiry, with most of the subsequent revisions pointing in a downward direction.
But that’s a very different story than the one that the scholars invested in the inequality-rediscovery thesis wish to tell. It also reveals that the measurement of inequality is not some neglected and then recovered subject of investigation, but rather a continuously evolving line of research with a vibrant and uninterrupted genealogy going all the way back to Kuznets (and earlier, as Kuznets admitted).
The trajectory of that research is a process of constantly revising and strengthening our measurements, with an eye toward improving the accuracy of the results. Unfortunately, that story of scholarly research evolution is much less exciting than a fairytale of neglect and rediscovery in which several of its proponents also evince explicit interests in politically justifying an upward revision to U.S. tax rates.
Trying to force the history of economic thought to fit a particular narrative out of the need for ideological congeniality is the greatest disservice that the inequality-rediscovery thesis does to the topic of inequality.