By DANIEL HEMEL
Review of Taxing the Rich: A History of Fiscal Fairness in the United States and Europe, by Kenneth Scheve and David Stasavage
Princeton: Princeton University Press, 2016
Senator Bernie Sanders has attracted the support of millions of Democratic primary voters with his call for redistribution of wealth on the home front and his opposition to military action abroad. These two policies are standard planks in virtually any progressive platform. Yet according to political scientists Kenneth Scheve and David Stasavage, the two policies also work at cross-purposes. To really make the tax system more progressive, what we need is an all-out war.
Scheve and Stasavage are not the first to draw a connection between war and the redistribution of wealth: Thomas Piketty likewise observes in Capital in the Twenty-First Century that World War I and World War II produced a dramatic—though temporary—narrowing of the gap between the rich and the poor in Europe (and, to a lesser extent, North America). And Scheve and Stasavage are not, of course, arguing that progressives should switch their votes from the dovish Sanders to a more hawkish alternative. War might be good for wealth redistribution, but it’s not good for much else. Moreover, limited foreign forays such as the invasions of Afghanistan and Iraq won’t do the trick: according to Scheve and Stasavage, it takes a mass mobilization on the order of the First or Second World War to generate significant support for high tax rates on the rich. With developments in military technology over the last several decades, such mass mobilizations may be a thing of the past: future conflicts, according to Scheve and Stasavage, are less likely to require large armies and thus less likely to result in high top tax rates. So while Sanders’s tax-the-rich rhetoric has won over many millennials, Scheve and Stasavage’s analysis implies that his tax policy playbook may be better suited to the last century than the current one.
Scheve and Stasavage never mention Sanders by name in their new book, Taxing the Rich: A History of Fiscal Fairness in the United States and Europe, but the Vermont senator’s remarkable run of primary victories makes the subject of their book all the more relevant. The timeliness of their topic would have been difficult to predict nearly a decade ago when they began their project. Back then, a major party candidate in the United States who proposed raising the top tax rate to 65% would have been laughed off the debate stage; income inequality had not yet been named “the defining challenge of our time” by the President or “the root of all evil” by the Pope; France had yet to impose (and rescind) its 75% “supertax” on millionaires; and no one had heard of (much less read) Capital in the Twenty-First Century. (Some things never change . . . .)
In that very different political climate, Scheve and Stasavage set out to learn when and why many countries have imposed high tax rates on the rich in the past and why very few still do. Their answer to the “when” question is that governments have imposed heavy taxes on the rich primarily “during and in the wake of mass mobilization for war” (p. 53). Their answer to the “why” question is: “because war mobilization . . . created an opportunity for new and compelling compensatory arguments that increased support for taxing the rich” (p. 135). They end with a prediction that taxes on the rich will rise only if proponents of a more progressive rate structure can come up with “compensatory arguments compatible with an era of peace” (p. 214).
Scheve and Stasavage’s greatest contribution is their answer to the “when” question. The “dominant narrative,” they write, is that countries tax the rich more heavily as they extend the right to vote more broadly (p. 63). Contrary to the conventional wisdom, Scheve and Stasavage show that shifts toward progressive taxation cannot be attributed to universal male suffrage. In 1900, for example, the top income tax rate in advanced economies with universal male suffrage was on average lower than the rate in advanced economies without universal male suffrage (p. 64). Scheve and Stasavage also find no significant increase in the top income tax or inheritance tax rates in the 10 years after a country moves to universal male suffrage (pp. 65-66). And focusing specifically on the United Kingdom, Scheve and Stasavage demonstrate that top rates did not rise in the immediate aftermath of any of the Reform Acts, which expanded suffrage among men in increments. (Unfortunately, the authors do not consider the effect of female suffrage on top tax rates, despite work by several others indicating that the spread of progressive economic policies in the United States and Western Europe coincided with the extension of voting rights to women.)
Just as the spread of democracy does not appear to answer the “when” question, the rise of inequality does not do so either: Scheve and Stasavage find little evidence that industrialized countries respond to rising wealth inequality by taxing the rich more. They do find evidence of a relationship in the other direction: countries exhibit lower levels of inequality after they impose higher tax rates on the rich. The latter result is unsurprising: policies that redistribute wealth from the rich to the poor do, in fact, narrow the gap between the rich and the poor. (Indeed, it would be surprising if that were not the case.) But that fact does not answer Scheve and Stasavage’s first motivating question: when do countries impose higher tax rates on the rich in the first place?
Scheve and Stasavage’s primary answer to the “when” question is that “major increases in the taxation of high incomes coincide[] with mass mobilization for war” (p. 76). World War I, in particular, pushed top tax rates to new heights: the top tax rate on income in advanced economies that mobilized for World War I rose from 4% in 1913 to 63% in 1920, compared to a modest rise from 10% to 17% in counterpart countries that did not mobilize for war (p. 82). The “difference in difference” for World War II was more modest—in part because tax rates started out higher and in part because nonmobilizing countries also increased their rates dramatically during the war (pp. 84-85). But all in all, the two world wars provide powerful evidence that countries tax the rich more heavily when their populations mobilize for war en masse.
Having offered an answer to the “when” question, Scheve and Stasavage next ask “why.” They discuss and dismiss the “Willie Sutton argument” (p. 129): that governments tax the rich during wartime because that’s where the money is (an allusion to the famous bandit’s apocryphal explanation for why he robbed banks). According to Scheve and Stasavage, the need for more money might explain why governments would raise taxes across the board, but it does not explain why countries raised rates on the rich so much more than on everyone else. In their view, “the story of taxing the rich has more to do with politics” than with fiscal constraints (p. 131).
“The Conscription of Wealth”
For Scheve and Stasavage, “politics” specifically means rhetoric: their answer to the “why” question focuses on the types of tax fairness arguments that advocates for redistribution have employed. Scheve and Stasavage direct their attention (and ours) to three particular tax fairness claims. The first is what they call the “equal treatment” argument: “the fairest system involves equal treatment for all” (p. 6). The second is what they describe as “the ability to pay doctrine”: each additional dollar of taxation represents less of a sacrifice for someone earning $10 million a year than for someone earning $10,000, and so a progressive tax system imposes a roughly equal burden on the rich as on the poor even while the rich pay much more in dollar terms. The third type of argument is “compensatory”: “taxing the rich more heavily than the rest serves to correct or compensate for some other inequality in government action” (p. 5). According to Scheve and Stasavage, the last type of argument is the only one that historically has justified highly progressive rate structures.
Scheve and Stasavage’s strongest evidence comes from the United Kingdom during World War I. At that time, members of Parliament and the press argued that the “conscription of men” should be accompanied by the “conscription of income” or the “conscription of wealth” (pp. 140-41). Since the government was already calling on young men to make the ultimate sacrifice for the war effort, then calling for a larger financial sacrifice from the rich (who also tend to be older) served a compensatory function. Scheve and Stasavage collected data on the types of tax fairness arguments made in the U.K. Parliament before and after World War I began, coding speeches as reflecting “equal treatment,” “ability to pay,” or “compensatory” theories. They find that the share of claims in the “equal treatment” and “ability to pay” categories fell dramatically after the beginning of the war in July 1914, while the share of tax fairness claims in the “compensatory” category rose from less than 6% before the war to 62% once the war was underway (pp. 142-43).
Before putting too much stock in these findings, though, several caveats are in order. First, the logical link between the “conscription of men” and higher taxes on the rich is not obvious. If young men are called upon to make the ultimate sacrifice, then requiring everyone else (not just the rich) to make a larger financial sacrifice would serve a compensatory function. In other words, the compensatory argument would seem to justify higher taxes on all income earners back home, not only the highest earners. Scheve and Stasavage’s observation about the Willie Sutton theory thus applies with similar force to the compensatory argument: mass mobilization might explain why the United Kingdom raised rates on everyone, but it does not so easily explain why the United Kingdom raised rates on the rich more than on the rest.
Second, even if the compensatory theory explains why the United Kingdom levied high taxes on the rich during World War I, it does less to explain why rates reached a peak of 60% in 1920 and 1921—after the war’s end (p. 141). The compensatory theory also struggles to explain why high rates persisted in the United Kingdom and other industrialized democracies deep into the second half of the 20th century, well after any mass mobilization. The case of the United States is particularly perplexing: the top marginal income tax rate exceeded 90% for more than a decade after the Korean War armistice, and then fell to 70% just as the deployment of U.S. troops to Vietnam accelerated.
Scheve and Stasavage might respond that the compensatory theory explains why countries raise tax rates on the rich in the first place and that inertia explains why those high rates persisted. (Scheve and Stasavage do not offer this reply themselves; the inertia response is anticipated rather than actual.) The inertia hypothesis is plausible, but Scheve and Stasavage have done little to prove it: they do not comprehensively examine the reasons why countries retained high rates in postwar periods. Another possible explanation is that countries continued to tax the rich at high rates until they had paid down their war debts. But this explanation is difficult to square with Scheve and Stasavage’s compensatory theory. After all, many of the middle-aged male taxpayers subject to 90%-range rates in the 1960s would have been the same people who fought in World War II as young men. These men were asked to sacrifice once by risking their lives and again by forfeiting most of their income. In what way does that double sacrifice advance “compensatory” objectives?
A third reason that Scheve and Stasavage’s answer to the “why” question might fail to persuade arises from their broad definition of “compensatory.” For purposes of their quantitative analysis of U.K. Parliament speeches, they define the “compensatory” category to encompass speeches that “suggest a tax policy is justified because of other inequalities, advantages, or sacrifices due to state policy,” so long as the “orientation of the speech” is “for the income tax or higher rates” (p. 239). Thus the following arguments all count as “compensatory” according to Scheve and Stasavage:
-- (1) The rich should pay higher income taxes to compensate for the fact that the young and the poor have been conscripted to fight the war (p. 141);
-- (2) The rich should pay higher income taxes to compensate for the fact that they earn excess profits during wartime shortages (pp. 144-45);
-- (3) The rich should pay higher income taxes to compensate for the fact that other taxes, such as taxes on consumption, fall disproportionately on the poor (p. 239).
Scheve and Stasavage say that all three arguments belong in the “compensatory” category—and that all three are categorically different from other tax fairness claims. But without a finer-grained breakdown, it is difficult to know which of these three arguments were prevalent during wartime tax debates. If (1) and (2) predominated, then perhaps we would conclude that the most successful arguments about tax fairness were war-specific ones. The fact that these war-specific arguments were also “compensatory” in some sense might have mattered and might have not.
Furthermore, even if we accept that the salient feature of arguments (1), (2), and (3) is their compensatory nature—and even if we accept that compensatory arguments have been uniquely successful in persuading industrialized democracies to tax the rich heavily in wartime—we quickly run into a problem of external validity. Perhaps it is true that compensatory arguments—and not inequality-based arguments or ability-to-pay claims—have persuaded democracies to tax the rich in the wake of mass mobilization for war, but why does that mean that compensatory arguments will be any more successful than inequality-based or ability-to-pay claims during peacetime? And why does it mean that compensatory arguments are the only arguments that will persuade democracies to raise taxes on their wealthiest citizens? Political psychology follows no iron law. The arguments that convinced voters to support high taxes on the rich in earlier eras might not be the same arguments that persuade democracies to raise taxes on the rich today.
Finally, we might wonder whether arguments about tax fairness in parliaments and the public sphere are the main act in the political story or a mere sideshow. Scheve and Stasavage say that their theory prioritizes “politics,” but it seems that their theory really prioritizes persuasion. Missing from Scheve and Stasavage’s analysis is any accounting for power. Their core claim is that advocates for redistribution prevailed in wartime debates because they had the strength of the better argument, when perhaps the redistributionists prevailed because they acquired the strength of the better bargaining position.
Recall that mass mobilization, at least according to Scheve and Stasavage, was a product of the particular constellation of military technologies that existed at the time of World War I and World War II, and that no longer exists today. As the authors put it, “[t]he era of the mass army . . . depended on a state of technology in which men and supplies could be moved en masse by rail yet where the remote delivery of explosive force was not yet advanced enough to avoid the need for mass infantry” (pp. 170-71). Manpower was uncommonly valuable under those conditions, and while the wealthy are the segment of society with the lion’s share of capital, the poor and middle class are the segments with the lion’s share of the bodies.
It should be no surprise, then, that when the masses possessed the factor in higher demand, they were able to extract a higher price. Put differently, the poor and the middle class were in a strong position to bargain for larger capital contributions from the rich because the poor and the middle class had what the rich could not themselves supply: bodies—and, particularly, young able male bodies. Politicians may have dressed these dynamics up in the rhetoric of tax fairness, but the taxation of the rich during the era of the mass army can be explained by price theory as much as moral theory.
This political economy account of wartime taxation is not so much a different explanation than Scheve and Stasavage’s as much as it is the same explanation with a different emphasis. On this account, wartime redistribution was “compensatory,” in the sense that it compensated the poor and middle class for providing the factor in short supply. Concededly, the political economy account still struggles to explain how the poor and the middle class continued to extract this compensation in the form of the high tax rates that persisted for several decades after the world wars ended. But that shortcoming of the political economy account is nonunique: Scheve and Stasavage’s rhetoric-focused theory also fails to explain the remarkable persistence of high rates on the rich through the middle of the 20th century.
Taxation as Compensation in the 21st Century
Ultimately, I cannot confidently say whether rhetoric is causal or epiphenomenal—whether claims about tax fairness actually affect outcomes or whether they are the stories we tell ourselves to justify the results of hard bargaining. And in any event, this is no place to resolve once and for all the idealism-materialism debate. But let’s assume, arguendo, that Scheve and Stasavage are right that rhetoric matters. Would that lead us to conclude, as Scheve and Stasavage do, that taxation of the rich in the 21st century depends on whether proponents of greater progressivity can come up with “compensatory arguments compatible with an era of peace”?
Even if I accept Scheve and Stasavage’s historical claim—that “[t]he reason wartime governments increased taxes on the rich more than the rest was because war mobilization . . . created an opportunity for new and compelling compensatory arguments” (p. 135)—I do not think that their prediction about the future necessarily follows. We might say that the salient feature of those successful claims in earlier eras is that they were compensatory. Or, we might say that the salient feature of those successful claims in earlier eras is that they were compelling, in light of the economic, geopolitical, and social conditions of the time. If we draw the latter inference, then the future of progressive taxation does not depend on whether proponents of greater progressivity can come up with compensatory arguments adapted to a less violent era. Rather, the future of progressive taxation depends on whether proponents of greater progressivity can come up with persuasive arguments of any sort. Perhaps it will turn out that the most compelling arguments are compensatory claims, but there is no reason to presuppose that is so. Indeed, when we look at how well the compensatory claims from the first half of the 20th century fit the first half of the 21st, I think we have reason to suppose the opposite.
The first of the three compensatory claims identified by Scheve and Stasavage—the argument that the rich should pay higher taxes because the poor have been conscripted to fight—is obviously inapplicable to the United States today. Conscription ended in the United States more than four decades ago, and the median family income for military recruits is now higher than the overall national median. The third argument—that we should raise income tax rates on the rich because other taxes fall more heavily on the poor—is inapplicable to the United States as well. The Treasury Department’s Office of Tax Analysis publishes data on the distribution of federal taxes (accounting for individual income, payroll, excise, and estate and gift taxes, and allocating the burden of corporate income taxes between capital and labor according to a method explained here). The Treasury analysis indicates that average federal tax rates increase steadily over income, with families in the bottom decile paying an average rate of negative 8.5%, families in the top decile paying an average rate of 29.0%, and families in the top tenth of the top percentile (i.e., the top 0.1%) paying an average rate of 39.0%. State taxes, which tend to be more regressive, change the numbers but not the bottom line: when one looks at federal, state, and local taxes and transfers combined, the rich pay more (in percentage-of-income terms) than everyone else.
But what about the second compensatory claim that Scheve and Stasavage identify: that the rich should pay higher taxes to compensate for the fact that they earn excess profits during wartime shortages? Might progressives be able to modify and update a version of this argument? Of course, we are not suffering from wartime shortages, but perhaps progressives can argue that the rich should pay higher taxes to compensate for the fact that they have benefitted from government privileges at the expense of the rest. The authors consider this possibility, but they ultimately do not find the argument to be persuasive. Concededly, stakeholders in financial institutions may have benefitted from bailouts in 2008, but this fact would—at most—justify a clawback of Wall Street profits, not a higher tax rate on top earners across the board. Scheve and Stasavage put it well when they write: “it is not clear why Silicon Valley should be taxed just because Wall Street was bailed out” (pp. 213-14).
Redistributionists might have more luck arguing that gross inequality is a bad in itself, regardless of how it came about. Michael Mitchell’s excellent recent review here in the New Rambler of Harry Frankfurt’s On Inequality addresses that normative claim in depth. My view, like Frankfurt’s, is that income inequality is not an intrinsic moral bad, but Harry Frankfurt and I do not constitute a representative sample. In a Gallup poll of American adults last year, 63% said that wealth should be more evenly distributed and 52% said the government should impose “heavy taxes on the rich” to reduce inequality. Even if redistributionists have yet to develop compensatory arguments compatible with an era of peace, they seem to be winning the war for hearts and minds.
To be sure, “redistribution for the sake of redistribution” has not been a winning argument in the past. (Or, at least, it has not been a winning argument in the mostly democratic countries covered by Scheve and Stasavage’s dataset: the USSR under Lenin, China under Mao, and Cuba under Castro are conspicuously absent.) Then again, if Thomas Piketty is right, wealth inequality by the middle of the 21st century will exceed anything observed since the Industrial Revolution. Scheve and Stasavage say that rising inequality has not been sufficient to trigger high tax rates on the rich before. But inequality in the future may look nothing like inequality in earlier eras, and debates over tax fairness may take on a new tone.
On the last page of their book, Scheve and Stasavage write: “In the end, one thing that is certain is that taxation of the rich will continue to be a fundamental source of social conflict, and when we seek to understand this conflict, we can learn a great deal from history” (p. 218). I have learned a great deal from Scheve and Stasavage’s book, but I am not so sure how much patterns of taxation in the past can tell us about what is in store. Someone writing the same book at the beginning of the 20th century might look at the lessons of history up to that point and conclude that democracies are unlikely ever to impose high tax rates on the rich, and that compensatory claims about tax fairness are doomed to fail. If the last century has taught us anything about taxation of the very rich, it is that long-term trends are unstable and that predicting the future based on the past is perilous. I still would put my money on the proposition that taxation of the rich will remain a source of social conflict. But beyond that, I’m reluctant to make any wager—especially if my winnings might be taxed at a 90% rate.
Posted on 25 May 2016
DANIEL HEMEL is Assistant Professor of Law at the University of Chicago Law School.
