No taxation without misrepresentation

A review of The Triumph of Injustice.

Emmanuel Saez & Gabriel Zucman, The Triumph of Injustice. How The Rich Dodge Taxes And How To Make Them Pay, New York, WW Norton, 2019, 232 pages, ISBN 9781324002727, URL

For nothing is worthy of man as man unless he can pursue it with passionate devotion. Yet it is a fact that no amount of such enthusiasm, however sincere and profound it may be, can compel a problem to yield scientific results.”

Max Weber, Science as a vocation, Speech at Munich University, 1918.

The industrious and frugal United States of Benjamin Franklin and Theodore Roosevelt are no more. The America of Donald Trump and Jeff Bezos, of the Koch Brothers and Jersey Shore, has taken their place, leaving far behind it a country where wealth was slowly and strenuously accumulated, but cautiously and modestly displayed. Income and wealth inequality not seen since the Gilded Age of the robber barons are now a central feature of the American economy, corrupting its politics, ruining its sense of fairness, distorting its incentives. In the coming of age of this American dystopia, one factor played a major role : a tax system that no longer plays its intended role of keeping large incomes and big fortunes in check, and has given up on redistributing wealth and opportunities.

That is the thesis put forward by two Berkeley economists, Pr. Emmanuel Saez and Pr. Gabriel Zucman, in a provocative new book, The Triumph of Injustice. To these public finance economists specializing in the study of inequality, taxes, and tax avoidance, what they describe as a tax schedule slowly losing its redistributive character is crucial to any understanding of the long-term rise in income and wealth inequality after the 1980s, documented – to varying degrees – by several economists over the last fifteen years. Reduced top federal income tax rates on high earners, exemptions on capital income, along with a growing reliance on proportional levies such as sales, payroll, or state income taxes to fund public spending, have made the US tax system less progressive than it used to be, according to Saez and Zucman. These policy choices have enabled the richest, and especially the most agile at tax-dodging, to escape paying their fair share of the « price of civilization ». Now is the triumph of the voluptuous, the cheaters and the confidence-men ; the era of conspicuous consumption and shameful tax evasion, of banksters and fraudsters, of tax heavens and health insurance hell, has come.

To drive their point home, Saez and Zucman resort to the most effective and boring weapon an economist can draw : numbers. They build on some of their earlier academic work to elaborate a new system of « distributional national accounts », a way to reconcile two different concepts of income : the one measured by national accounts statisticians when establishing aggregate measures of GDP, investment or consumption ; and the one recorded by the IRS when assessing individual or household fiscal income and tax liability. This allows them, with what sometimes seems to be extreme confidence in the precision of necessarily noisy estimates, to allocate the share of all taxes paid in the economy to each income group over time, and thus compute an estimated « effective tax rate », total taxes paid divided by total pre-tax income, for each income level in the American economy since the early twentieth century. Their takeaway is so surprising it deserves closer scrutiny : the wealthiest Americans, according to their calculations, nowadays pay a lower share of their income in taxes than the poorest twenty percent. The book then takes these novel and roughly estimated numbers at face value to describe a radical reform of the US tax system that the authors claim would drastically reduce inequality, promote growth, provide free healthcare and education to all, improve fairness, cleanse the political system, all while costing nothing to anywhere from 95 to 99 % of Americans. What’s not to like ?

If this result strikes the reader as extremely implausible, it’s because it probably should. It is most likely untrue : the richest Americans do not pay a lower income tax rate than everybody else in the economy. The statement is highly dependent on extreme assumptions made in the book, that have been questioned across the board by senior economists, using both detailed individual data and standard theories. First, methodological choices at odds with the data systematically attribute excessively high income and wealth shares to the highest earners in society. This would already mechanically lower the authors’ estimated progressivity of the tax system ; but, partly because of the book’s treatment of income and taxes as a static, single-year process, it also allocates too low a share of overall taxes paid to those same high earners. Finally, the planned reforms, from the now famous Warren wealth tax to a doubling of the corporate income tax, are scored with optimistic revenue estimates, that far exceed receipts raised by similar measures in other advanced economies. Evaluating these reforms in isolation tends to overlook the interactions across all the new and higher taxes the authors suggest : this makes it difficult to obtain a clear intuition of how much additional public and welfare spending they could truly finance, if implemented.

Death of a taxman

A tax loophole is « something that benefits the other guy. If it benefits you, it is tax reform. »

Russell B. Long, US Senator.

Before embarking on a long journey describing some of the limitations of both the empirical investigation in the book and its policy implications, it is worth stressing where it succeeds in convincing the reader. Many of the arguments made in The Triumph of Injustice are compelling, especially when the authors stress the role of a lawyering-accounting industrial complex in supplying tools to game the tax system, and when they document and propose solutions to deal with the long-term rise in artificial profit shifting by the largest multinational corporations.

The authors first walk us through an – admittedly politically slanted – narrative of the buildup of progressive taxation as foundational to the establishment of modern American institutions. Claims that taxation grew hand in hand with American democracy may sound slightly exaggerated – even without resorting to clichés like the Boston Tea Party starting the American revolution, there is no denying that a strong anti-government encroaching streak was also a key part of the American political identity, and to some extent, a driver of the strong protection of private property rights that has made the American economy the most prosperous ever seen. Nevertheless, the book precisely and clearly documents the reduced progressivity of the tax system over the last thirty years, even though claiming it is the major factor driving income and wealth inequality may sound extreme, given the plethora of alternative explanations, from trade and financial integration, to skill-biased technological change, reduced market competition, and rising prices and incomes in the largest cities due to housing supply constraints.

The book also introduces to a broader audience a major research agenda championed by Saez and Zucman at Berkeley, along with Thomas Piketty at the Paris School of Economics : the « distributional national accounts », an effort to measure an income concept at the individual level that is consistent with the way statisticians elaborate national accounts. Distributing the entire national income has many appealing properties. It ensures that measures of income growth by percentile indeed add up to the salient concept of national income growth. This important methodological innovation is not a small part of the book’s appeal : by making it possible to reconcile standard measures of growth at the national level with individual and group-level income dynamics, it is able to « distribute growth » in an intuitive and internally consistent way, although it has limitations when trying to estimate yearly tax burdens, as we show in more detail below.

Saez and Zucman justifiably emphasize the role of an established complex of law practices, accounting firms, and lobbying shops in creating, disseminating and protecting tax avoidance mechanisms. Indeed, a crucial point made by the Triumph of Injustice lies in documenting the role of the supply side of the tax avoidance and outright tax evasions « markets ». Analysts have long explored the many ways and tricks used by the wealthy – and the less wealthy – to evade their tax bill, from outright non-reporting, to shifting income towards non-taxable sources, charitable giving in name only, or trusts and shell corporations incorporated on the sandy beaches of the Caiman Islands. The supply of tax avoidance mechanisms, however, accounts for a large part of the growth in the possibilities offered to corporations and individuals to escape part of their nominal tax liability. Transfer pricing practices in the largest accounting firms have made a fortune certifying the validity of inflated prices paid by corporations for intellectual property they artificially located in low-tax jurisdictions. Tax lawyers have earned a living devising cascades of shell corporations to mask the identity of the final owners of tax shelters located in tax heavens. The book does a great job at documenting the emergence and rise of an array of support activities entirely designed to game the tax system, although it may take too strong a leap forward, relative to available estimates, when attributing almost the entire rewards of these tax gaming schemes to the richest Americans.

Another strong point of the book lies in the question of how to adapt to the increased agility of global firms, in an age where the nationality and tax residence of a corporation is becoming ever harder to identify. The astonishing increase in the share of profits global firms now book in tax heavens and quasi-heavens, documented in some of Gabriel Zucman’s earlier work, is a major challenge for the modern social contract. Can firms entirely abstract from their original place of incorporation, or the geography of their sales or expenditures, when moving profits from high- to low-tax jurisdictions ? While there is some credible evidence that cracking down on the use of tax heavens abroad tends to lower investment at home too, by raising the worldwide tax burden of companies, it may be politically and fiscally responsible for large countries to exert some of their market power in the global taxation competition. Saez and Zucman suggest original and innovative solutions to deal with the new constraints imposed by the globalization of multinational firms, and their ability to shift profits across the globe independently of economic substance. Here, the book provides interesting technical formulaic ways to apportion some of the low-taxed global profits based on a firm’s global sales distribution. Such a reform is likely to have somewhat perverse effects, notably by reducing global trade integration, creating incentives to cease fiscal cooperation, and increasing the one capital income tax (corporate taxation) that has been consistently shown to discourage investment. From an international law perspective, it would allow a global coalition of large countries to negate the right of smaller places to democratically choose low tax rates because they benefit from them – an interference in other countries’ policy-making process that even their rant against the “commercialization of sovereignty” cannot conceal. Overall, however, improved cooperation to stem the most blatant cases of profit-shifting will be part of the global agenda in the coming years, and the book contributes elegantly to this endeavor.

Ploutocracy in America ?

“But this proportion must in every nation be regulated by two different circumstances : first, by the skill, dexterity, and judgment with which its labour is generally applied ; and, secondly, by the proportion between the number of those who are employed in useful labour, and that of those who are not so employed.”
Adam Smith, An Inquiry into the Nature and Causes of the Wealth of Nations, 1776.

Notwithstanding these appealing dimensions, some key features of the book fall short of their stated objectives. In attempting the monumental task of documenting the evolution of income shares and tax burdens of all groups along the income distribution, Saez and Zucman have to make a number of simplifying assumptions, each of which may appear negligible, but which, all compounded, paint a distorted picture of the distribution of tax burdens and the revenue impact of potential reforms. Because “estimated tax rate” equals “estimated taxes paid” divided by “estimated income received”, errors that allocate unreported income at the top than there truly is have an oversize effect on estimated tax burdens, lowering the measure of taxes paid at the same time that they blow up income received, artificially making the tax system seem regressive.

The book attempts to distribute national income across all individual Americans, to show that the richest earners have “captured” most of the growth in wealth produced by the country over the last forty years. One could discuss the wisdom of describing as a capture the outcome of a market process, as if some buccaneers had taken away part of a bounty belonging to others. After all, consumers are willingly and happily paying fees to Jeff Bezos’s Amazon for the services it provides, and no one is being forced to use Google’s Youtube services to watch videos of Stephen Curry scoring slam dunks. But even accepting the “accounting” framework which Saez and Zucman adopt, in which all income can simply be shifted around at the will of the all-knowing legislator, many of the choices they make to go from the recorded income of taxpayers on their 1040 forms to their share of the net national product overstate income flowing to the top.

The Triumph of Injustice shocks the reader by first showing that the bottom 40 % of individual Americans only receives $18,500 on average from the market, and that this number has barely budged over the last forty years. This result, presented as fact by Saez and Zucman, is highly disputed, with other economists1 finding a roughly 65 % increase in the income of the bottom 40 % over the same period, using similar data and a different allocation methodology. More importantly (although such a discrepancy is no quibble), one of the reasons for the slow growth of market income at the “bottom” of the income distribution is quite simply that a substantial number of individuals in the bottom four deciles of the pre-tax income distribution have no market income at all. Those labelled “the working class” by Saez and Zucman include unemployed and out-of-the-labor-force prime-age workers, retirees, students, parents taking a break out of their career to raise a child… The share of adults older than 20 years old currently employed hovers around 65 % ; the bottom quintiles of incomes mostly correspond to a “non-working class”, which lives either through their spouse’s incomes, their parents’ support, or government transfers. Overall, it is not very surprising that unemployed people earning no income at all (or working only a few weeks or a few months per year), students (a rising share of the bottom 40 % of the population above 20 years, given the increasing numbers attending college), or retirees receiving pensions based on low contributions they made in the past, all of which represent a larger share of the bottom 40 % than they previously did, have witnessed limited income growth. Sweeping demographic changes (rising college attendance, increased life expectancy and time spent in retirement, lower white male labor force participation rates) explain much more of the slow growth among the bottom half of individual pre-tax and pre-transfer incomes than does any form of “capture” by the wealthiest.

The second major challenge in allocating national income lies in distributing all forms of income not properly recorded in tax returns. Here, Saez and Zucman make choices that tend to overestimate income and wealth flowing to the top, leading them to suggest that the top 1 % of Americans, 2.4 million people, earn on average $1.5 million from the market. These estimates are far and above all those available in the literature. This is because Saez and Zucman tend to allocate almost all of the income not seen by the IRS, especially under-reported self-employment income, as well as profits of corporations not distributed as dividends, to the very top. For example, estimates of tax fraud and under-reporting of self-employment income based on randomized audits by the IRS suggest much less under-reporting at the top than used by Saez and Zucman to build their measure of top incomes. More importantly, by allocating all the yearly profits not distributed by firms in proportion to the dividends declared by individuals, they blow up true inequality at the very top. To the contrary, it is widely documented2 that dividends and retained earnings tend to be negatively correlated across companies, which makes intuitive sense, since companies tend to smooth dividends relative to their overall profits3. When an individual receives higher dividends, this should shade down, not up, our estimates of how much income that individual left inside the companies she owns. While seemingly technical, this difference has massive implications for the overstatement of inequality in The Triumph of Injustice.

A subtler issue with the allocation of national income is one of terminology. Comparing income groups, such as the top 1 % or the bottom 40 %, and the evolution of their average income over time, masks substantial changes in the composition of these groups, and mobility in an out of top and bottom percentiles. This difference between “anonymous” growth rates (how rich are all those in the top 1 % in 2018 relative to all those in the top 1 % in 1975) and more intuitive individual growth rates is not always made clear, and for good publicity reasons. Throughout the book, the reader unfamiliar with the literature on inequality could be forgiven for thinking that, when Saez and Zucman tell him that “the rich – adults within the top 1 %, have received big raises”, or that “over the same forty uears, the working class has received almost no raise”, they are actually talking about individual income growth, when they are aggregating very different people over time. While estimates of income mobility in the United States vary substantially depending on methodologies used, an individual whose earnings are in the bottom 20 % of the income distribution is extremely likely to move upwards over time, and conversely for the top 1 %4.

Do the poor really pay so much…

“Unquestionably, there is progress. The average American now pays out twice as much in taxes as he formerly got in wages.”

H.L. Mencken

The second shocking fact documented by Saez and Zucman, beyond their alleged “capture” of growth at the very top, is the massive tax burden falling on the bottom 50 % of income earners, and its long-run increase since the 1970’s. According to the authors, this is a feature of deliberate policy choices designed to soak the poor : the creation and vast increase in the slightly regressive payroll tax on wages ; a rising reliance on proportional taxes, such as most state income taxes ; and the existence of consumption taxes that fall primarily on goods consumed by the poorest. For each of these three factors, however, methodological choices made in the book tend to overstate the burden of taxes on the bottom half of earners, especially among the poorest twenty percent.

First, the federal income tax in the United States includes refundable tax credits, such as the Earned Income Tax Credit, or the Child Tax Credit, that are directly and specifically designed to offset the burden of the payroll tax on people making less than the median wage. These correspond to direct refunds that are part of the tax code and administered by the IRS, much like the one Americans receive at the end of the year when they overpaid withholding taxes. But Saez and Zucman, in a statistical choice at odds with the methodology of all government agencies and think-tanks, choose to exclude these refundable tax credits from their computation, mechanically overstating the tax rates at the bottom by substantial amounts (close to 5 %, according to some estimates). Similarly, all but six State income taxes also include refundable credits, which Saez and Zucman, again uniquely among available analyses, choose to not take into account when estimating the burden of these proportional levies on income.

Secondly, Saez and Zucman estimate extremely high payroll taxes for the bottom 20 % of income earners (more than 11 % of income, according to the book). While this sounds like a reasonable estimate – although on the high end, given the extensive definition of pre-tax income – if all income at the bottom were in the form of wages, we know that most income for the bottom 20 % corresponds to the operations of private and public pensions, old-age insurance, and disability insurance (according to the latest published Saez-Zucman methodology5), all of which are exempt from the payroll tax. This exemption makes the number hard to reconcile with known estimates of the distributive incidence of the payroll tax, which show numbers closer to 7 % in the bottom two deciles of income.

Finally, sales taxes on the poor as a share of income are overestimated in the book, because of an accounting issue. To allocate sales taxes, they need to estimate each individual’s consumption. However, an individual will usually consume a lesser share of her income when she earns a lot (for example, at the peak of her active life), but a much larger share of her income when she has very little income (for example, as a student, when she retires, or when she becomes unemployed for part of the year). Therefore, sales taxes will mechanically be over-estimated on the “poorest” taxpayers, and under-estimated on the richest ones, not because the rich “escape” the sales tax, but because the same household will pay more sales taxes as a share of her income in years when she runs down her assets than in the years when she saved them.

… and do the rich really pay so little ? DINA without dynamics

“In short, the Way to Wealth, if you desire it, is as plain as the Way to Market. It depends chiefly on two Words, Industry and Frugality ; i.e. Waste neither Time nor Money, but make the best Use of both”

Benjamin Franklin, “Advice to a young tradesman”, Philadelphia, July 21st 1748

The most publicized finding of The Triumph of Injustice is its emphasis on the very low tax rates paid by the wealthiest in society. Here, the book confirms and amplifies a true intuition, and one that is widely held : some income at the very top tends to escape taxation, making a few thousands of the richest 0.001 % subject to a lesser tax burden than their immediate counterparts in the top 0.01 %. In particular, the existence of a step-up in basis at death for capital gains, which allows the share of capital gains accrued from the purchase of an asset up to the death of the owner to go untaxed once it’s inherited, is a problematic feature of the tax code long lamented by economists of all stripes, and one that exempts some of the most well-off in society without any substantial justification. Similarly, excessively broad and uncapped exemptions for charitable giving, or the availability of trust funds or exempt assets to escape the estate tax, feed into the injustice of multi-billionaires not paying a similar share of their income as multi-millionaires.

Nonetheless, some of the findings on the low tax burden of the rich stem from the peculiar way in which “Distributional national accounts”, or DINA, handle the accumulation of income and wealth over the medium-run. Among the very top incomes, those making tens or even hundreds of millions in a given year, there is very little persistence : this mostly corresponds to individuals obtaining a windfall from the sale of their business, finding shale oil under their house, or successfully betting on an extremely speculative asset. Therefore, given the limited number of households – a few hundred – they represent, one must handle with extreme care the issue of allocating income and taxes over several years. Unfortunately, DINA isn’t great at dynamics.

In particular, the timing of income received and taxes paid can sometimes be out of sync, generating under-estimates of the progressivity of the tax code, and over-estimates of top income shares. While such timing issues may sound mundane or be discarded as mere technicalities, they can generate crucial discrepancies between measured and actual tax burden. We have already mentioned how sales and other consumption taxes fail at taking into account savings. Being paid out of consumption, which may not track income, these taxes lead to systematic underestimates of true tax burdens at the top of the income distribution, and overestimates at the bottom6.

Other taxes, in a slightly subtler way, suffer from the same issue. Saez and Zucman, in line with the methods suggested by Thomas Piketty, suggest a focus on distributing net national income, rather than gross domestic product, to evaluate changes in absolute and relative standards of living of various groups over time. The main difference between the two concept is the exclusion of capital depreciation. Capital income, however, is distributed by firms out of their cash-flows, which do not incorporate depreciation ; while investment in capital formation, for the most part, is not immediately deducible from income. Therefore, whenever the actual, physical wear-and-tear of assets differs from fiscal allowances for depreciation, the timing of capital income distribution may diverge substantially from the estimated national account aggregates, leading to over-estimates of top incomes, and under-estimates of their tax burden.

Moreoever, when a company retains some of its profits inside the corporation in a given year, without distributing them to shareholders, they are only subject to the corporate income tax ; but they will be subject to additional federal and state income taxes in a later year, when the corporation will see fit to distribute these taxes as dividends or when they will be realized as capital gains. Therefore, even though taxes on undistributed profits may appear low in a given year, it is because they will tend to be levied at a later data ; therefore, effective tax rates at the top will appear lower in periods when corporate retained earnings are rising over the medium-term, as they did in the last twenty years. This force also makes the “regressivity” at the top partly a business-cycle phenomenon : in “good years”, when lots of benefits are put in reserve to be distributed in years of lean cows, the “DINA” income is lower than it truly is, and taxes higher, while the reverse occurs in bad years, when estimated income is too high, and taxes too low.

What are taxes for ?

“Those who oppose reform will do well to remember that ruin in its worst form is inevitable if our national life brings us nothing better than swollen fortunes for the few and the triumph in both politics and business of a sordid and selfish materialism.”

Theodore Roosevelt, “New nationalism” speech, Osawatomie, Aug. 31st 1910

One of the most startling sections of the book is titled “Beyond Laffer”, to suggest that economists should go beyond the intuition of the (in)famous Laffer’s curve, which implied that tax rates should never be raised to a point where they reduce overall tax receipts. Indeed, according to Saez and Zucman, there may be good reasons to tax the highest net-worth individuals even if it ends up reducing resources available to the government to transfer to the neediest in society. The sincerity of candidly admitting that one wants to reduce the income of the richest at the top, not to help the poor, but for the sheer reason that “nobody should have so much money”, is to be admired, although it seems at odds with most of their past academic work.

Their rationale is mostly based on the assumption that excessive income and wealth concentration translates into excessive influence in the political arena, and a capture of the policy-making process. While this argument is not based on documented facts in the book, it rings true to observers of US political life, where lobbyists have been given almost entirely free rein in politics since the Supreme Court’s Citizens United decision. It does seem, however, that directly targeting the influence of money, by reforming campaign finance, capping contributions, or dealing with gerrymandering, is a simpler way of protecting the democratic process than a tax reform which would reduce the government’s ability to provide for the poorest in society. Most people, indeed, think of taxes not as such an unmitigated good, but rather as a necessary evil to fund welfare programs, public goods, and the operations of the State. In these circumstances, it becomes hard to accept the option of going “beyond Laffer”. If a reform has as its only objective to punitively reduce incomes at the very top, but actually decreases resources available to the rest of the population, it does not seem likely to be favored by the vast majority. Herein comes the difficulty of thinking about the distribution of tax burdens without including what taxes are used for. Most government spending, from public education to Medicare, from SNAP benefits, disability or unemployment insurance, to social security or infrastructure, is either common to all, or, more frequently highly progressive. Therefore, one should acknowledge that the very existence of taxes (and their rising share of the economy over the last sixty years) makes the distribution of actual, post-tax outcomes much more equal than it would be in its absence. Saying payroll taxes are “regressive” because they are capped at a given maximal wage does nothing to the fact that that, in their absence, the consumption and well-being of the poorest in society would be substantially lower.

Plucking the golden goose

“The art of taxation consists in so plucking the goose as to procure the largest quantity of feathers with the least possible amount of hissing.”

Jean-Baptiste Colbert

The book ends with a call to radically reform the United States’ tax system. Among many proposed reforms, is Elizabeth Warren’s “2-cents” tax (a two percent tax on wealth from $50 million to one billion, and a three percent tax above $1 billion) ; but also a dazzling array of other new taxes, ranging from 60 % top marginal tax rates in the federal income tax, to the inclusion of capital gains in the normal federal tax schedule, an additional 6 % tax on all forms of value added, a doubling of the corporate income tax, and a 25 % minimal tax on profits in all other countries. Such measures, combined, would amount to marginal tax rates on income from capital well above 100 %, especially if one adds to the inclusion of capital gains in taxable income the proposed integration between corporate and personal income taxes, which would result in a double taxation of retained earnings, when they are earned, and when shareholders sell their assets. Wealth at the top would be eroded each year by much more than the income it earns from investing in productive endeavors. Enough to pluck the proverbial golden goose ?

The Triumph of Injustice provides only very sparse details on the way it computes the estimated revenue impact of such reforms for the Treasury. To arrive at an estimate of total additional receipts of 10 % of national income, the authors assume responses to taxation in the lowest range of commonly accepted values in the economic literature. Strangely, this is at odds with their own long-term description of the US tax system, which shows that incomes at the top have surged concomitantly with and, according to them, as a result of, lower top tax rates. This suggests much larger responses of top incomes to tax rates in the long run, more in line with the recent economic literature7. Assuming such heightened responses of income and wealth to taxation would drastically reduce the revenue that can be raised from an overhaul of taxes at the top. Besides, optimistic estimates of wealth tax revenues have already been widely criticized8, notably in the light of the actual experience of countries like France, which failed to raise much revenue from such a wealth levy, despite not being usually shy when it comes to collecting taxes.

More surprisingly, some of these estimates can be shown to be even inconsistent across themselves. For example, according to Saez and Zucman, a wealth tax of 1 % reduces reported wealth by 8 % on average ; but they then only take into account the 2 % rate in the lower bracket and assume a 16 % (2 times 8) avoidance rate, forgetting that the tax goes up to 3 % in their plan. Similarly, the plan doubles corporate income taxes up to 41 % from 21 %, a move likely to reduce reported profits, as companies invest less in productive factories and patents, and spend more on private jets and lavish meals for their executives. But since Saez and Zucman’s estimates of wealth and capital income at the top are mostly based on imputing it from these observed profits, this should substantially reduce the revenue not only from the corporate income tax itself, but also from the federal and state income tax and the Warren wealth tax. Overall, the interaction between all of these proposed new taxes is a blind spot of the book, and one that looms large when evaluating the plausibility of funding universal public healthcare, public child care, public early education, free tuition, and the elimination of sales taxes and tariffs that they affirm these reforms will cover.

Most of the welfare states Saez and Zucman want the US to emulate, such as Sweden or France, actually have relatively flat, but high taxes throughout the income distribution, due to broad and high Value Added Taxes, elevated payroll tax rates, top income tax brackets applied to much broader swaths of earners, and a variety of local property and excise taxes.

Reforming exemptions such as the step-up in basis at death, uncapped deductible charitable giving, the undeserved tax-exempt status of some private institutions, or the loopholes in the estate tax, is probably necessary to deal with the sense of unfairness most Americans feel about their tax system. Increasing taxes on inherited wealth, taxing capital gains as they accrue and not at realization, and improving global cooperation against excessive profit shifting by multinational firms, may even be required to curb inequalities of opportunity and raise some revenue. But it seems highly improbable that the United States can make its way into Scandinavian levels of welfare spending while only resorting to higher taxes on the top 1 to 5 %.

Politics as a vocation, science as a tool

“The appeal to ‘social justice’ has nevertheless by now become the most widely used and most effective argument in political discussion. Almost every claim for government action on behalf of particular groups is advanced in its name, and if it can be made to appear that a certain measure is demanded by ‘social justice’, opposition to it will rapidly weaken. People may dispute whether or not the particular measure is required by ‘social justice’. But that this is the standard which ought to guide political action, and that the expression has a definite meaning, is hardly ever questioned.”

Friedrich A. Hayek, Law, Legislation, and Liberty, vol.2., “The mirage of social justice”, 1976.

The Triumph of Injustice is at times a political manifesto, and at times a popular-science exposition of Saez and Zucman’s cutting-edge research in establishing a set of distributional national accounts for the United States. The only issue is that the authors don’t make it easy to find out when science speaks, and when politics take it place.

The book’s shortcuts are probably necessary, given the attempt at such a formidable and all-encompassing historical investigation and political justification ; but they tend to systematically estimate too much income and wealth at the top, too high taxes at the bottom, and too elevated receipts from its planned tax reforms. Its description of the army of tax lawyers and accountants that has been emboldened over the last three decades to encourage individuals and corporations to escape their tax liabilities is riveting ; but its proposed reforms, by focusing too much on seizing inflated estimates of shadow income at the very top, are unlikely to do much to fund spending that would truly reduce inequality in standards of living and raise the welfare of the least advantaged.

By departing from the semblance of Weberian impartiality most economists pretend to adopt when talking about politics, The Triumph of Injustice is a welcome and refreshing jump into the arena by two senior economists who are also two close advisors to the Warren techno-populist campaign for the Democratic presidential nomination. While, as we have suggested, readers should take most of the quantitative estimates with caution, the book is likely to drive a conversation long avoided in American politics. Can the social contract upon which existing income taxation is founded survive in the era of international mobility, fast-paced financial flows, trade globalization, and increasingly intangible and virtualized private property ? More importantly : should it ?

  3. See e.g. early evidence in Lintner, J. (1956). Distribution of incomes of corporations among dividends, retained earnings, and taxes. The American economic review, 46(2), 97-113.
  6. To understand the intuition, imagine a a world with only a consumption tax of 10 %, where two individuals live for only two years. In the first year, individual A earns $100,000, while individual B only earns $20,000, for example because he is still transitioning from gig jobs to a full-time position. A only consumes $50,000, paying $5,000 in sales tax (a 5 % tax rate out of her income), and depositing the rest in the bank. Taking advantage of zero interest rates, B borrows $30,000 of the cash deposited by A in the bank, consumes $50,000, paying $5,000 in sales tax (a 25 % tax rate out of her income). The next year, A retires, only earning $20,000 in social security benefits, and consuming those in addition to the $50,000 dollars in her bank account. She pays $7,000 in taxes out of her $20,000 measured pre-tax income, a 35 % tax rate ! B, on the other hand, found a permanent job, earning $100,000. After paying back the $30,000 she owes, she consumes the remaining $70,000, paying a meager 7 % of her income in sales taxes ($7,000). In both years, this tax system is – or appears to be – « highly regressive. In the first year, we observe a 5 % tax rate on A, earning $100,000, and a 25 % tax rate on B who only makes $20,000. In the second year, B, who makes $100,000, pays a 7 % tax rate, but A, earning $20,000 faces an apparent tax burden of 35 %. One may then jump to the conclusion that such a sales tax is highly regressive. It isn’t so : over their « life cycle », both individuals earned $120,000, and paid $12,000 in sales taxes, a 10 % effective tax rate over two years out of their identical incomes.
  7. Mertens, K., & Montiel Olea, J. L. (2018). Marginal tax rates and income : New time series evidence. The Quarterly Journal of Economics, 133(4), 1803-1884.
Le Grand Continent logo