This article originally appeared in the May 10, 2021, issue of Tax Notes Federal.
]Jeffrey N. Pennell is the Richard H. Clark Professor of Law at Emory University. He thanks Eduardo Fernández for his research and editorial assistance.
In this article, Pennell explains why a tax on naked wealth would likely fail, even if enacted, and he outlines a far less problematic concept for Congress to consider in any effort to address wealth inequality.
Copyright 2021 Jeffrey N. Pennell.
All rights reserved.
Introduction
Asking the rich to pay more tax has been a consistent concern for politicians. Rising inequality and concentrations of wealth in the United States1 have caused some policymakers to question whether to reduce tax benefits that significantly lower the tax bills of high-net-worth (HNW) individuals. Suggested changes include raising the moderately low income tax rate,2 decreasing the federal estate tax exclusion amount, and altering or repealing the section 1014 new-basis-at-death rule. More dramatically, scholars3 and political candidates propose to impose a “wealth tax” on HNW Americans who often own substantial wealth but do not realize significant taxable income (as defined by the code). The most widely recognized of these proposals were advocated by Senate Finance Committee member Elizabeth Warren, D-Mass.,4 and Sen. Bernie Sanders, I-Vt.,5 during the 2020 presidential election season. Each advocated an annual wealth tax.6 Because the negative attributes of these proposals are likely to preclude their success, even if enacted, this article offers in skeletal detail an alternative, aimed at the same taxpayers but more consistent with historical taxation and less likely to fail.
A Wealth Tax
Most countries that experimented with wealth taxes ultimately repealed them.7 Most of the European wealth taxes were repealed because they raised little revenue, entailed increased administrative costs, and induced a migration of wealthy individuals and their wealth. Wealth taxes also may cause economic distortion, such as altering investments8 and forms of saving. Despite calls to address wealth inequality through the tax system, the historical evidence suggests that a wealth tax is not an effective way to do so.
Some observers contend that wealth taxation in the United States would not suffer the three main weaknesses of the European wealth taxes because (1) offshore tax evasion can be fought more effectively, (2) no exemptions permit taxpayers to shift wealth into preferred categories, and (3) modern information technology can be leveraged by our tax authorities.9 Even if these predictions prove to be true, other weaknesses also must be considered in the design of a wealth tax.
For example, a wealth tax likely would generate a constitutional challenge in the United States. Under Article I, sections 2 and 9, of the U.S. Constitution, Congress may not impose a direct tax (on property) that is not apportioned based on population. The unapportioned federal income tax is specifically authorized by the 16th Amendment, and the wealth transfer taxes (estate, gift, and generation-skipping transfer taxes) are justified as an impost on the transfer of wealth rather than a naked tax on the wealth itself.10 The wealth taxes recently proposed by Warren and Sanders are akin to state and local property taxes. Scholars debate whether these proposals would thus be regarded as unconstitutional. Without a specific constitutional amendment similar to the 16th Amendment, a wealth tax may be regarded as a direct tax that must be apportioned based on population. My proposal avoids that question.
A wealth tax also would have high implementation costs, lack an effective enforcement mechanism, increase procedural costs, and (most especially) generate administrative difficulties concerning the valuation of illiquid assets. One observer opined:
Enforcement of the tax would be cumbersome on both the taxpayer and the IRS. Since the tax is based on the net assets of the individual, it would require the individual to seek an independent valuation of assets in order to determine the fair market value of all of his or her assets on an annual basis. If all the individual’s assets were in cash and/or publicly traded securities, then the calculation would not be too difficult. However, it would be expected many of the wealthy also own interests in privately held companies with no liquidity or interests in various real estate projects or farmland. Determining the fair market value of privately held company interests or real estate could prove difficult and would certainly be subjective, potentially leading to tax planning opportunities, such as discount valuation strategies.11
Based on IRS Statistics of Income data from federal estate tax returns filed in 2019, liquid assets (marketable securities, bonds and notes, cash, and retirement funds) were just 54 percent of reported estate wealth.12 Some undetermined portion of that wealth is held offshore. Thus, predictions about the effective administration and reach of proposed wealth taxes likely are correct.
A wealth tax would fail in the United States, as it has in other developed countries, if these critical issues cannot effectively be addressed.13 I analyze several of them below.
The most problematic issue in the design of a comprehensive wealth tax system is the valuation of illiquid assets.14 The potential for undervaluation creates a challenge for the IRS, which would need to develop valuation protocols for unique assets. Unlike the process undertaken for the one-time wealth transfer tax, an annual wealth tax would require annual reevaluation of difficult-to-value assets. Imagine, for example, the annual assessment of a 10,000-bottle wine cellar, a yacht, a private airplane, a family farm, or a minority interest in a closely held business.15 The IRS and the Tax Court would require more resources to resolve the multitude of potential valuation cases.
Venture capital would be hard hit by a wealth tax that magnified the risks associated with successful start-up entities that appreciate in value but do not generate an income flow,16 subjecting investors to wealth taxes without liquidity with which to pay. Worse would be start-ups whose value tanks after an initial valuation that generated a wealth tax. This could cause a shift for many Americans toward investment only in publicly traded assets, undermining investment in ventures that benefit the economy in general. The tax also would imperil conservative buy-and-hold investment strategies unless dividend policies changed to provide cash flows to wealthy taxpayers who need liquidity to pay the annual wealth tax.17
A regime (similar to the passive foreign investment company rules) could impose a wealth tax only upon a realization event, with a deferral charge that reflected the delay in payment of an otherwise annual wealth tax. But even a deferred tax would imperil investments with significant growth and might harm innovation, risk-taking, and entrepreneurship.
Thus, an annual wealth tax is challenged by valuation issues and the need to convert enough wealth into liquidity to pay the tax.18 If measured by wealth alone, the need for liquidity may require liquidation of investments, or borrowing against the value of that wealth, to finance payment of the tax — all to the detriment of the investment itself. Consider owners of real estate development projects that have high value but negative cash flow and zero profits before completion. Selling an interest in that project could be nearly impossible without a significant discount to reflect lack of liquidity, and borrowing might be entirely impossible if the project itself is highly leveraged already. Moreover, adding debt could disrupt operations and the ultimate success of the venture from a working capital perspective, and indirectly affect owners as well as employees.
An even more difficult example would be the valuation and payment of tax on agriculture, whether a corporate enterprise or a family farm. Given the United States’ proclivity to protect farmers and ranchers, some advocates for the wealth tax might favor an exemption to this sector of the economy. But disparate treatment of different investments (or worse, different landowners) is horizontally inequitable and could encourage wealthy taxpayers to invest in forms of favored property in ways that artificially affect the market.
Finally, the tax would be a godsend for tax specialists and valuation experts, who would devise strategies to circumvent or minimize the impact of the tax.19 Those strategies might not be the best economic use of valuable resources.
An Alternate Proposal
Each of the problems identified above (which are just some of the potential challenges of a wealth tax) could be solved by the concept discussed next. It offers an alternative to the wealth tax that could avoid constitutional concerns, eliminate liquidity issues, and minimize administrative challenges.20 This approach would avoid the need to value assets while still accomplishing the objective of raising significant added revenue from taxpayers whose tax burden is less than seems appropriate, given the wealth gap in the United States. This alternative would tax income, not capital/wealth,21 and builds off the concept of a progressive impost as found in the federal gift tax.
Historically, the fundamental notion embodied in the gift tax was that a higher progressive tax rate would apply in calculating the tax liability as more wealth is transferred during life.22 To illustrate (using hypothetical numbers), imagine that the first $100,000 of taxable gifts might be taxed at 5 percent, that a second $100,000 of lifetime transfers might push that taxpayer into a 10 percent bracket on that second gift, that a third $100,000 transfer might be taxed at 15 percent, and so on. The gift tax does not determine the tax rate on the basis of annual gifts (each $100,000 of taxable gifts in this example) taxed separately. That would allow a taxpayer to begin at the lowest marginal rate every year and effectively avoid the higher gift tax brackets by spreading their lifetime transfers over multiple years. Instead, lifetime aggregate gifts are determined by consulting prior-year gift tax returns (thus, recordkeeping is not difficult), and the same aggregate gifts would incur the same gift tax liability whether made all in one year or spread over a lifetime. At death, under the unified gift and estate tax system, the lifetime aggregate gifts are a base on which the taxpayer’s remaining wealth (estate) is added to determine how much higher in the progressive rate tables the estate would be taxed. The same aggregation of prior taxable amounts could be the foundation of a surrogate wealth tax.23
As an income tax rather than a wealth tax, the analog discussed here would tax extraordinary amounts of income at a progressive rate on an aggregated (lifetime) income basis. For example, a taxpayer who had a single year of extraordinary income might not exceed a threshold amount and would not incur this special tax. But a taxpayer who had multiple years of extraordinary income would eventually exceed a lifetime threshold (based on the aggregate of those amounts) and begin to progress up a rate table for taxation of those extraordinary amounts, year after year.
Again, to take a simplistic example, imagine a taxpayer who had $10 million of income in year 1 and paid income tax only under the normal income tax rate system. In year 2 the same taxpayer has another $10 million of income, and together, the two years’ aggregate income might push this taxpayer into the special HNW income tax system. On year 2 income of $10 million, the taxpayer might incur a 1 percent HNW income tax ($100,000) in addition to the regular income tax under the existing system. In year 3 the same taxpayer might have $15 million of extraordinary income, and that $15 million, added on top of the prior $20 million, would push the taxpayer into a 1.5 percent bracket for the HNW income tax, imposed on the year 3 income of $15 million (an added $225,000), again in addition to the regular income tax imposed in that year on the taxpayer’s total income for year 3.
The notion behind this suggestion is that:
A single year of extraordinary income might not subject a taxpayer to this HNW income tax. The threshold would avoid imposition on many one-time windfalls (such as the sale of a business or the recovery of a large tort judgment).24
As an income tax, this impost would be calculated on income versus capital — and would be no harder to administer than the normal income tax.
As an income tax on current income, this impost would not require liquidation of capital investments; liquidity would exist in the form of the current income realized.
As an income tax on current income, this impost would be easier to administer because the valuation of difficult-to-value assets would not be required, and the filing of a special return could be avoided (an additional schedule to accompany the taxpayer’s traditional Form 1040 income tax return could suffice).
As an income tax, there would be no question about the constitutionality of this impost.
As a progressive tax, this impost would better comport with the American belief that those with a greater ability to pay should pay a greater amount of tax.25 Given that basic living or lifestyle needs are not paid from extraordinary income received after prior years of extraordinary income, the bite of this tax also would not be as severe as other alternatives. For example, simply increasing the tax rates applicable to all taxpayers would impose more pain across the board, while this tax would reduce only the extraordinary income of taxpayers, presumably with enormous wealth accumulated from their after-tax income from a series of prior years.
Consistent with other current tax reform proposals, this construct should be coupled with modifications of the section 1014 new-basis-at-death rule. Otherwise, taxpayers would be inclined to accumulate income within an entity if taxation was lower inside the entity than the tax that would be incurred at the owner level (which is the case now). Thus, to avoid accumulations that would escape the HNW income tax through basis step-up, section 1014 would need to be amended to provide for carryover basis at death or to deem death a realization event.26
Consideration also should be given to whether special tax rates should apply to capital gain versus ordinary income in the calculation of the HNW income tax. An illustration involves taxpayers who are wealthy enough that they do not need to rely on current income to finance their lifestyle. These individuals can sell capital assets as the need for cash flow exists, and they should not benefit from favorable capital gain tax rates. There is a serious issue, however, if such a taxpayer generates neither income nor capital gain to finance their lifestyle and instead borrows against the collateral of their significant wealth (perhaps in a manner that effectively puts off repayment of that debt for years, decades, or a lifetime). If the interest incurred is cheaper than the HNW tax that would apply if assets were liquidated to repay those loans, a taxpayer who is comfortable with debt can again avoid the HNW tax.27 At a minimum, no deduction should be allowed for interest paid on those loans, and other mechanisms might be used to impede this avenue to avoid or defer the HNW tax.
Further, any tax reform proposal to increase the fair-share tax burden on extraordinarily wealthy taxpayers must have strong provisions (such as an exit tax) that address wealth or income that is shifted offshore. This is not the best forum in which to explore those options.
Among the advantages and criticisms of any form of tax on wealthy Americans, consider that this is a tax on income, not a tax on capital or on capital appreciation. As such, it would not suffer the critique offered by some economists that taxing capital works as a drag on the mobility of investment dollars and a restraint on the flow of money from lower- to higher-risk investments. This tax would not impose a brake on investment decisions or the allocation of dollars.
It could, however, encourage taxpayers to invest in ways that minimize annual returns of income in favor of capital appreciation — which would avoid the annual tax and defer liability until a realization event. That effect is common to the current differential between ordinary income and capital gain and feeds into the same debate that suggests28 (1) a mark-to-market system of periodic taxation of appreciation without realization, (2) taxation of capital gain and ordinary income at the same rate, and (3) repeal of new basis at death.29 Only the last of those proposals needs to be linked to or adopted in conjunction with the proposal here,30 although attention to the basis issue concerning accumulated income inside an entity is needed.
Any tax on individuals with significant wealth is likely to cause imaginative advisers to devise avoidance schemes. An important consideration in the construct of an HNW tax is the avoidance techniques that might be available, which include:
income shifting or the assignment of income;31
allowing income to accumulate inside an entity to avoid paying dividends or interest;
capital flight — wealth relocation — to offshore tax havens;
deductible transfers (for example, to charity);
uncapped deductions (for example, depreciation); and
investment in assets that generate no income (such as precious metals).32
Each of those would require some consideration in light of whatever alternative Congress adopts for taxing wealth.
Conclusion
The critique that “many Americans have vast assets but little income” raises the question of why wealth itself, invested without consumption, is an evil to be discouraged through taxation. There are only two benefits of wealth: personal consumption (either of the asset itself or of income generated by that asset) and control over that personal consumption by some other individual. As a matter of tax policy, the tax laws need not punish the possession of wealth that no one is enjoying through consumption. Merely owning the wealth does nothing to prevent others from creating their own wealth, because there is no finite amount of wealth that can be created or owned.
A wealth tax encourages taxpayers to consume their wealth in ways that have no residual value — such as by travel, literal consumption of food or beverages, or entertainment — none of which benefits society as much as investment of the wealth (which is a form of saving). It also should not matter what the form of investment is: income-producing assets, or assets such as land or art that alone produce no income.
An HNW income tax would be constitutional, and it would avoid valuation and liquidity concerns. It could be implemented immediately, administered easily under the existing IRS infrastructure, and would require little added reporting. It would need no special exemptions and could apply to any form of income (although it might be coupled with refinement of the capital gain rules). Its only significant downside is that it would generate less revenue, even if the rate imposed was high and the threshold at which it began was low.33 A workable tax that produces some revenue (half a loaf) is better than a proposal that in theory would generate more revenue but would not be viable in practice.
FOOTNOTES
1 The inordinate share of wealth owned by the richest Americans is an accepted reality. See, e.g., Emmanuel Saez and Gabriel Zucman, “The Rise of Income and Wealth Inequality in America: Evidence From Distributional Macroeconomic Accounts,” 34 J. Econ. Pers. 3 (Fall 2020) (claiming that wealth owned by the top 0.1 percent has increased “from about 7 percent in the late 1970s to around 20 percent in recent years”).
2 Since 1913 the highest marginal income tax rate in the United States was 94 percent (during the waning years of World War II). The primary cause of low tax on income of wealthy taxpayers is the gain and loss realization rules, which leave capital wealth to grow without taxation until there is a sale or exchange. And then special capital gain tax rates cause most capital gain to be taxed at a mere 20 percent when there is a realization event.
3 See, e.g., David Shakow and Reed Shuldiner, “A Comprehensive Wealth Tax,” 53 Tax L. Rev. 499 (2000); and Catherine Clifford, “Top Economists Stiglitz and Piketty: The U.S. Needs a Wealth Tax on Millionaires and Billionaires,” CNBC, Sept. 17, 2020.
4 The Ultra Millionaire Tax Act of 2021 (S. 510), cosponsored by Warren, would address wealth concentration with a proposed annual federal tax on real property, personal property, and financial assets, less debt. Warren’s proposal would impose a tax of 2 percent on net wealth above $50 million, rising to 3 percent on net wealth above $1 billion. It is estimated to affect up to 100,000 households. See Warren Democrats, “Ultra-Millionaire Tax.”
5 Sanders would impose a tax of 1 percent on amounts exceeding $32 million, and it would rise to 8 percent on amounts exceeding $10 billion. It is estimated to affect up to 180,000 households. See Sanders, “Tax on Extreme Wealth.”
6 Each proposal stated that it targeted the top one-tenth of 1 percent (0.1 percent) of U.S. taxpayers — which is the same slim slice of the decedent population with wealth exceeding the basic exclusion amount (now $11.7 million) for federal estate tax purposes.
7 The number of European countries with a wealth tax has fallen from 12 in 1990 to just three (Norway, Spain, and Switzerland) today. See OECD, “The Role and Design of Net Wealth Taxes in the OECD,” OECD Tax Policy Studies No. 26, at 76 (2018).
8 Chris Edwards, “Taxing Wealth and Capital Income,” Cato Institute Tax & Budget Bulletin No. 85 (Aug. 1, 2019) (“Targeting wealth for higher taxation is misguided. Wealth is simply accumulated savings that the economies need for investment. The fortunes of the richest Americans mainly consist of active business assets that generate jobs and income. Increasing taxes on wealth would not help workers, but instead would undermine productivity and wage growth.”).
9 See Saez and Zucman, “Progressive Wealth Taxation,” Brookings Papers on Econ. Activity 437, 440 (Fall 2019).
10 The constitutionality of the federal estate tax was upheld in New York Trust Co. v. Eisner, 256 U.S. 345 (1921), as a tax on the transfer of property, not on its ownership. The Court held that the estate tax was an indirect tax that need not be apportioned. Using the same rationale, the Court in Bromley v. McCaughn, 280 U.S. 124 (1929), upheld the federal gift tax.
11 Adam Bergman, “Warren’s 2% Wealth Tax Would Cause a Wealth of Problems for the Economy,” Forbes Finance Council, Mar. 15, 2019.
12 The statistics can be found at the IRS Statistics of Income web page, the chart concerning 2019 net estate taxes.
13 For example, some analysts question whether a wealth tax is optimal to raise revenue. See, e.g., Howard Gleckman, “Can a Wealth Tax Raise the Revenue Its Sponsors Hope?” Urban-Brookings Tax Policy Center TaxVox, Sept. 24, 2019. Gleckman contends that (1) estimates of the amount of money that a wealth tax can raise are significantly higher than what will be the reality; (2) there are administrative issues (such as underreporting, difficulties in valuation — especially because the government lacks access to better third-party data — and a lack of well-trained audit and enforcement staff); and (3) there is a corresponding need to impose tougher taxpayer penalties to minimize avoidance of the tax.
14 Asset valuation is probably the single most litigated wealth transfer tax issue for assets that are not publicly traded. See, e.g., Gary Burtless, “Putting a Tax on Wealth Means We First Must Measure It,” Brookings, June 5, 2019 (op-ed); and Estate of Mitchell v. Commissioner, T.C. Memo. 2011-94 (stating that valuing art can be an “ambitious task” because value “often lies in the proverbial eye of the beholder” and that ascertaining fair market value is the “quintessential fact question”).
15 See, e.g., Brian Raub, Barry Johnson, and Joseph Newcomb, “A Comparison of Wealth Estimates for America’s Wealthiest Decedents Using Tax Data and Data From the Forbes 400,” Nat’l Tax Assn. Proc. 103rd Annual Conference on Taxation (2010).
16 “Under a wealth tax, individuals who own the same amount of wealth pay equal taxes, regardless of how much income their wealth generates. You use your wealth productively or you lose it.” Fatih Guvenen et al., “Use It or Lose It: Efficiency Gains From Wealth Taxation,” National Bureau of Economic Research working paper No. 26284 (Sept. 2019).
17 These issues are similar to those that affect a mark-to-market taxation regime, including the need to measure any annual change in value and generate liquidity to pay tax on unrealized gains. See David Kamin and Lily Batchelder, “Policy Options for Taxing the Rich,” in Maintaining the Strength of American Capitalism (2019).
18 David F. Bradford, Untangling the Income Tax 149-151 (1986); and David Shakow and Reed Shuldiner, “A Comprehensive Wealth Tax,” Faculty Scholarship at Penn Law 1264 (2000), opine that a tax on income should be based on ability to pay, and that income is the best measure of ability to pay.
19 For example, if the value of an investment is net of debt, strategies could include increasing leverage for the venture, which would generate profit for lenders rather than revenue for the fisc.
20 There are other methods to minimize these issues. See, e.g., Ari Glogower, “A Constitutional Wealth Tax,” 118 Mich. L. Rev. 717 (2020), which describes methods to adjust a taxpayer’s income tax liability on account of wealth.
21 Issues might arise in defining “wealth” subject to a wealth tax. For example, do an option to buy, an incentive stock option, or a rent-to-own lease constitute wealth for these purposes? How about a life estate or a long-term leasehold (whether at market or below-market rates)? What about the remainder interest following a life estate? A wealth tax raises all these issues in a way that the alternative HNW income tax avoids — to the extent that that there is oodles of experience knowing what constitutes taxable income.
22 In its current form, this progressivity is irrelevant for virtually all taxpayers because there is a 40 percent flat tax imposed once a taxpayer’s taxable transfers exceed a threshold exclusion amount. So the tax (as applied today) has a cliff effect. But the origin of the gift tax progressive impost harkens back to a prior era, as illustrated in the following text.
23 However, unlike section 1015, no basis adjustment to the accumulated income base would be needed to reflect the tax paid on a particular year’s income. Nor would any other adjustment be made to reflect the tax or how a taxpayer spent or used the income that remained after its taxation.
24 Existing exclusions from income would not be altered. For example, an inheritance is excluded under section 102 and would not generate this HNW income tax.
25 A bedrock principle of the federal tax system is progressive taxation, imposing higher rates of tax on increasingly large amounts of wealth or income. Progressive wealth transfer taxation is disguised by the mechanism by which the exclusion amount is taxed, but it is apparent in the income tax and remains operational in each of these systems.
26 Unlike some proposals to tax capital gain at death — which hurts “middle-rich” Americans along with those with extraordinary wealth — this tax would not create an artificial realization event, and it would not affect the middle-rich. Avoiding capital gain at death as a burden on the estates of middle-rich taxpayers could be achieved by providing a basis-increase exemption, but the experience in 2010 with such a tax benefit was not favorable because it generated administrative complexity for the estate of every taxpayer whose net worth was greater than the exemption amount.
27 This is one reason why a wealth tax would work better than an HNW excise tax. The issue could be minimized or avoided with a consumption tax, but that proposal does not appear to have much support.
28 In 2019 Sen. Ron Wyden, D-Ore., now chair of the Finance Committee, made two suggestions (forecasting revenue of $1.5 trillion to $2 trillion over 10 years). Consistent with my proposal, he would exempt individuals with less than a threshold amount of annual income. Inconsistent with my proposal, he would exempt assets such as personal residences, retirement accounts, and family farms. Difficulty would arise as a result of deferral of the tax on gain attributable to “non-tradeable property,” as to which interest would be imposed upon payment when a traditional realization event occurred.
29 A tax on capital gain tends to encourage investors to remain fully invested or locked into various positions — especially later in life, when new basis at death is a greater possibility within a reasonable time horizon. This proposal avoids that lock-in.
30 However, it is fair to suggest that annual realization and taxation of capital appreciation (a mark-to-market system) should not be coupled with an HNW tax on the same income.
31 Any tax on income encourages schemes to assign or shift income to taxpayers who incur tax at a lower tax rate — in this case to those who have not yet reached the HNW threshold. Assignment-of-income principles seem strong enough to confront any abuse. Moreover, spreading income among investors who are less risk-averse may be desirable if they might move the wealth from less innovative investments to more innovative investments.
32 If capital appreciation remains taxable, even after death, these investments pose no threat to the system of taxation, other than deferral until a realization event.
33 Note that I have not suggested the threshold level above which this tax might be appropriate. This question needs some study of the sort that Martin A. Sullivan has been writing about. See Sullivan, “Why Is the Effective Tax Rate of High-Income Taxpayers So Low?” Tax Notes Federal, Sept. 14, 2020, p. 1955 (using 2018 SOI data from 2018 returns).
END FOOTNOTES