ArticlesVol. 3, No. 1

Realization and Progressivity

Ilan Benshalom1,

Kendra Stead2

1 Associate Professor Hebrew University,
2 J.D. Northwestern University School of Law
Citation:3 Colum. J. Tax. L. 43


The realization requirement is the income tax’s original sin. Although long-standing, it is widely considered the main source of tax complexity, inequity, and economic distortion. Despite these problems, realization is also considered a fundamental element of modern income tax regimes. It is explained early in most federal income tax courses as necessitated by problems of asset valuation and taxpayer liquidity. To the dismay of certain professors, this explanation usually generates little class discussion. More worrisome, it is also widely accepted outside the classroom—prompting few political objections or normative academic inquiries.

The goal of this article is to provide a normative framework that allows policymakers to better understand the role of the realization requirement. It makes two related arguments. First, with respect to certain emotionally non-fungible (personal) assets, the realization requirement is normatively justified because the market price is not a good indication of the assets’ value to their owners. Second, contrary to the traditional view of realization as a regressive element, taxing only these personal assets upon realization would promote income tax progressivity. This article’s normative approach provides a basis for developing a more effective and coherent redistributive income tax policy. This analysis contributes to the broader tax reform debate and opens a novel theoretical inquiry with respect to the distributive impact of different types of errors.




Imagine a legal arrangement that transfers a huge percentage of the federal budget to wealthy Americans, is inefficient, results in complexity, and encourages dishonesty. Further imagine that when called upon to explain the arrangement’s objectives, policymakers, professionals, and academics merely shrug and reply that it is technical in nature and driven by administrative convenience. Would you expect the arrangement to last for even a day? Certainly not if it were part of the federal spending budget. However, the realization requirement has created exactly such a situation in the federal income tax context for over a century. This article critiques the academic literature dealing with realization and offers a new understanding of the requirement’s function: realization is another way to promote distributive justice within the tax regime. This insight provides policymakers a normative basis from which to combat the realization requirement’s negative effects in a principled way. It also raises the broader question of the role that the market prices of various assets should have in determining distributive policies.

The realization requirement means that tax liability is assessed only when assets are exchanged on the market, and not, as an “ideal” income tax would dictate, when the market values of assets change.[1] This seemingly simple requirement is the most “well established, and yet so widely criticized” attribute of our income tax regime.[2] It has been called the Achilles’ Heel of the income tax[3] and is widely considered to be the primary source of distributional inequity and complexity in the tax system.[4] Our main argument is that, contrary to the widely held view that the realization requirement is merely an administrative rule, realization is actually normatively justified with respect to certain assets—emotionally non-fungible (personal) assets. Personal assets are those assets for which there is a high probability that the subjective value individuals attribute to them differs significantly from the market value. We argue that taxing these assets only upon realization helps the income tax to better achieve its professed objective of progressivity.

A preliminary example demonstrates the realization requirement’s normative basis. Consider two family vacation albums, the first made by a talented professional photographer and the second created by her spouse, a lawyer. The photographer’s album is filled with well-composed landscape shots and insightful portraits that many galleries would love to display, whereas the lawyer’s is comprised mostly of poorly framed shots of distracted children that even the grandparents tire of. Nevertheless, both the photographer and the lawyer seem to derive the same amount of subjective value from their albums.

It seems intuitively wrong to increase the tax liability of the photographer only because her album has considerable market value. That is, as long as both albums are sitting on coffee tables in the family’s living rooms, it seems unreasonable to increase the photographer’s tax liability for the year simply because her personal album could fetch thousands of dollars, especially if she has no interest in selling it. However, under what is often described as an ideal income tax, the photographer would pay taxes in the year she compiled the album, because the album has the effect of increasing her net worth.

But taxing her for the value of the album’s photos becomes completely reasonable if she sells the exclusive rights to the photos, in which case it would be difficult to distinguish it from any other work she sells.[5] With respect to the album, realization does not increase or diminish the value or the nature of the photos—at least to the photographer. It only changes the type of the asset from a personal to a commercial one. This change signals that the asset’s value to the taxpayer is not different from its market value.

We expand the example slightly to demonstrate why the realization requirement cannot be normatively justified with respect to all assets. Imagine that the photographer and the professor each hold a hundred shares of Apple stock and that the stock has increased significantly in value over the past year. The fundamental (and intuitive) difference between the owners’ perceptions of their photo albums (personal assets that will likely never be sold) and their stocks (investment assets that they hold solely for financial reasons) suggests that it would not be wrong to tax them for any wealth increase associated with the stock, even if they chose not to sell this year.

Currently, the realization requirement shields both types of wealth accumulation from taxation until the asset is sold, but we argue that this delay is only justifiable with respect to the category of personal assets, such as the photo album. Recognizing this overlooked function of the realization requirement—allowing taxpayers to defer taxes on unique personal assets—provides policymakers with a valuable point of departure when deciding which assets should be subject to realization and which should not.

Articulating a normative basis for the requirement is certainly timely, as recent policy proposals claim the problems of realization prevent the income tax from meeting its distributive and efficiency goals, arguing that policymakers should abandon the income tax altogether and shift to a (less progressive)[6] consumption tax system.[7]

Scholars have debated the merits of realization for some time but mostly from the standpoint of assessing the practicalities of moving away from it. Some, such as Daniel Shaviro and Edward Zelinsky, view realization as a necessary and manageable weakness.[8] Others, such as Noel Cunningham, Deborah Schenck, Daniel Halperin, David Shakow, and David Weisbach, think that the income tax should gradually shift away from realization.[9] As policymakers consider decreasing the set of assets subject to realization, they should be guided by some commonly accepted baseline principles.

Although scholars disagree about the proper role of realization moving forward, there is consensus on three main issues. First, most of the income tax’s shortcomings are a direct result of the realization requirement.[10] Therefore, as long as realization is part of the income tax regime, problems of complexity, planning, inequity, and inefficiency are inevitable.[11] Second, there can be no practical income taxation without realization, so at least some assets will always be taxed only upon realization.[12] Third, all tax scholars, with the exception of David Schizer,[13] seem to accept that realization has no theoretical or policy underpinnings and that it is merely a second-best solution driven primarily by concerns about asset valuation and taxpayer liquidity.[14]

This article challenges the third of these views and raises some important implications regarding the other two. It calls for applying the realization requirement in a fundamentally different way and suggests some substantial, yet feasible, policy reforms that would allow the tax system to better achieve its distributional objectives. It expands upon a previous article that deals with the taxation of earning capacity, examining how tax policy can be advanced with respect to the realization principle.[15]

Admittedly, from a “pure” income tax perspective, taxing personal assets only upon realization rather than as their market value changes provides their owners with deferral and other tax benefits.[16] However, we argue that these benefits may be justified because they are progressively distributed.[17] The intuition behind this argument is that granting a tax benefit to personal assets requires imposing higher effective tax rates on other assets—namely, investment capital assets. Hence, if the tax system shifts to taxing personal assets upon realization and non-personal investment assets upon market-price fluctuation, the system would become more progressive. This is because data about the distribution of assets strongly suggests that personal assets comprise most of low and medium income households’ assets while investment and capital assets are disproportionately owned by affluent taxpayers.[18] By addressing realization as a question of distributive justice, this article connects existing tax practices to the theoretical bases of redistribution: the questions of what policymakers should seek to redistribute and how they should best measure it.

It is worth emphasizing the difference between our approach and the more canonic arguments with respect to realization. The bulk of the literature focuses on the difficulty of determining the market value of illiquid assets and describes the arbitrary tax assessments that would result from valuation error. This article makes the point that the problem of assessment errors would still be relevant with respect to personal assets even if all assets had a verifiable market price. We are concerned less with the consequences of inaccurate tax assessments and more with how different types of inaccuracies impact distribution. Accordingly, the article highlights a consequence of errors that has not yet been addressed. We contend that even if we had efficient markets for all assets, this improved accuracy would not be free of errors. Including personal assets within the income tax base would decrease mistakes with respect to the market price valuations, but increase the probability of error with respect to the subjective value taxpayers attach to different assets.[19] The increased probability of that type of error would not be distributionally neutral, increasing the relative tax burden on low and middle income taxpayers.

This analysis leads to the conclusion that realization may be justified with respect to personal assets but not with respect to investment assets. The implication of this conclusion is that policymakers should consider applying the requirement very differently by taxing investment assets as they appreciate or depreciate in value rather than when they are sold. Such a reform would significantly reduce the income tax’s current problems by making it less distortive, more consistent, and more equitable. This article explains why such a system would not be any more difficult to maintain than the current tax regime, which also applies special rules to investment assets. Hence, while our proposal raises some concerns, it also offers an administratively and politically plausible alternative that could be used to maintain the income tax as an effective wealth redistribution tool.

Before launching into a full discussion of realization, we provide a brief overview of the broad issues reached in this article. Part I describes the function and impact of the realization requirement, explains why it is the major source of income tax difficulties, and clarifies why it is viewed as solely the byproduct of tax administrative constraints. Part II explores the academic literature with respect to realization. It shows that all leading approaches are similarly incomplete because all view realization only as a necessary evil. It then argues that no reasonable income tax regime can be devised without understanding that realization can be justified in its own right.

Parts III and IV develop this article’s analysis—the former develops the normative framework, and the latter extracts its policy implications. Part III begins by explaining the difficulty of making interpersonal comparisons in a redistributive tax regime. It then establishes two normative pillars of realization: the particular difficulty of making interpersonal comparisons with respect to personal assets and the progressivity gained by taxing only personal assets on a realization basis. It ends by discussing a possible theoretical criticism of this article’s proposal—namely whether there are alternative ways to better attain tax progressivity. Part IV explains the policy reform implications of this article’s normative analysis. It also elaborates upon some empirical questions and points to some potential avenues of future research. We close with several brief conclusions that highlight this article’s policy and theoretical contributions.

I. The Realization Requirement—Costs and Endurances of an Achilles’ Heel

The realization requirement emerged in the early twentieth century and soon became a foundational attribute of all income tax regimes.[20] In the United States, the requirement got an early endorsement from the Supreme Court, which found that imposition of the federal income tax was constitutionally limited to only realized gains.[21] The Court soon retreated from that position, however, and scholars widely agree that realization is not constitutionally mandated.[22] In fact, Congress has successfully adopted many tax arrangements that impose taxes on unrealized profits.[23] This part briefly explains the realization requirement’s operation, the tension it creates within the income tax, and its significant social costs. It then presents the canonic explanation for how this “Achilles’ heel” of the income tax has endured in spite of its shortcomings.[24]

A. The Function and Social Costs of the Realization Requirement

This subpart describes the realization requirement as a transaction (rather than an income) tax and explains the enormous social costs this transactional element imposes upon the current regime.[25]

A simple example may help illustrate how the realization requirement is implemented. In 1991, a hypothetical relative of ours invested his small inheritance and pension savings in one company: AIG. In 2001, the price of his shares had increased fifteen fold, and the hypothetical relative was doing great. Not feeling any obligation to save, he was vacationing in the Caribbean, avoiding cheap hotels, taking big mortgages on his New York apartment and Florida mansion, considering early retirement, and planning to buy his kids new cars. At family parties, we would all gossip about his wife, saying how lucky she was to marry such a clever investor. By 2011, however, things have radically changed for the hypothetical relative—the Caribbean seems as distant as the moon, the kids have to take out huge student loans, he rarely attends family festivities, and when his ex-wife does, we all gossip about how unfortunate she is to have married such a reckless speculator.

Over the last twenty years, the hypothetical relative’s life has been a roller coaster, yet one thing has remained stable: his tax liability. Because of the realization requirement, he paid no income taxes and received no deductions with respect to his holdings in AIG throughout those years. In contrast, under a “mark-to-market” tax regime, the unfortunate relative would have been required to pay taxes (and been able to deduct losses) as the value of his investments changed. Mark-to-market tax treatment may not seem intuitive, but it is sensible because the gains and losses of the hypothetical relative reflected real changes in his economic purchasing power.[26] True, he could not have paid his grocery bills with AIG stock, but he could have sold any amount of the stock at any point in order to do so. More importantly, he borrowed against the stock and used the loan proceeds for consumption.[27] Simply put, while his stock was not equivalent to money, it was viewed by this relative, and basically everyone else, as being almost as good.

It is widely accepted among economists and tax professionals that taxpayers’ economic incomes increase as their assets appreciate, not just when those assets are sold.[28] To the extent that individuals’ increase in wealth is a good proxy for their well-being—and therefore a relevant benchmark for taxation—taxpayers’ realized gains and cash flows are (from a pure income tax perspective) beside the point.[29]

The principle that all accretions to wealth should be measured by an income tax creates an interesting tension. If policymakers wish to use the income tax to generate revenue, they should try to avoid giving it a transactional dimension, meaning that taxation should not depend on when taxpayers choose to put assets on the market.[30] Transactional elements are problematic because taxpayers can time and structure exchanges and sales. That is, transactional taxation gives taxpayers considerable ability to control how and when to pay the taxes.

The current realization-based income tax system aims to levy tax according to individuals’ incomes, but in fact it relies on transactional realization triggers that may have nothing to do with income.[31] This tax regime, however, is not truly transactional either, because it requires that assets appreciate in value for the tax to be triggered. Under the current system, taxes are only paid if a transaction occurred and gains were realized. This dual requirement grants taxpayers the ability to structure their transactions to attain certain tax advantages that would not be available to them in either a pure income tax regime or a pure transactional tax regime.[32]

The inherent conflict between the measurement of actual income and the transactional element of realization makes the current realization-based income tax unworkable. Policymakers find it difficult to generate broad rules that reconcile the two notions and are therefore unable to generate coherent rules explaining what types of transactions amount to a realization event.[33]

In terms of efficiency, the realization requirement leads well-informed and self-interested taxpayers to sub-optimally invest their resources in order to attain higher after-tax returns.[34] The realization requirement also distorts the timing of asset dispositions.[35] Taxpayers can also defer the recognition of gains but selectively sell depreciated assets to enjoy the value of their losses—a practice typically referred to as strategic trading.[36] The potential for manipulation introduced by the realization requirement compels the tax authority to introduce complex loss-limitation rules to protect the revenue base. These rules disallow the deduction of losses incurred with respect to certain investments, thus making them inherently riskier and deterring some (potentially less well-advised) taxpayers from making those investments.[37] Finally, the realization requirement results in significant revenue loss that has to be compensated for—either by inefficient higher marginal tax rates[38] or lower (sub-optimal) public investment.[39]

In terms of complexity and uncertainty, the transactional element introduced by the realization requirement materially complicates the income tax’s implementation.[40] Policymakers’ attempts to rein in problems stemming from the realization requirement are the hidden drivers of many notoriously complex provisions and regulations. In terms of deferral, the attempt to tax individuals only on their realized gains and losses raises the question of when investors in companies actually realize the profits of their investments. In fact, two of the most complex areas of law, corporate and partnership taxation, would be almost completely unnecessary absent the realization requirement.[41] As mentioned, any realization-based income tax regime needs loss-limitation rules to protect its revenue base against strategic trading.[42] Over the years, many loss-limitation rules have been enacted to curb tax planning with respect to income generated from passive investments and financial assets.[43] The rules formulated, however, are so staggeringly complex that commentators have voiced skepticism as to whether they can be enforced in even a marginally coherent way.[44] As difficult as these issues are, the above examples are just the tip of the iceberg in terms of realization’s costs.

In terms of equity, this distortion and complexity significantly reduce the ability of the income tax to effectively promote fairness objectives. If the income tax is expected to tax individuals according to changes in their economic well-being, then the realization requirement is an obvious obstacle. First, it provides a tax deferral benefit to successful investors who enjoy real, unrealized gains and (through loss-restriction rules) denies deductions to investors who suffer and realize actual losses. Second, since unrealized gains are frequently associated with capital assets, realization reduces the effective tax rate on capital.[45] Because the majority of capital assets are owned by affluent taxpayers, the realization requirement provides a tax benefit that is primarily skewed toward the wealthy.[46] Consequently, in the current tax regime, the realization requirement is justly viewed as an inherently regressive feature that hinders the income tax’s wealth redistribution objectives.[47]

The realization requirement and its byproducts undermine the integrity of the income tax regime in almost every respect. Instead of promoting distributive justice and ensuring efficient allocation of resources, the realization requirement and its host of problems create the perception that the tax code is a wasteful and arbitrary body of law. The discontinuities created by the requirement have given rise to a socially wasteful tax-planning industry.[48] Given realization’s obvious and harmful faults, one can only wonder why policymakers have not seriously considered any meaningful alternative.

B. Technical in Nature—The Valuation Liquidity Consensus

Tax academics widely perceive the realization requirement as having little, if any, theoretical or policy justification. Nevertheless, contemporary tax scholarship rarely takes on the requirement, adopting a “cross we must bear” approach to its problems. Realization’s endurance is generally explained in terms of two tax-administration concerns: problems of low taxpayer liquidity and the technical difficulty of valuing assets.[49] This subpart briefly explains and evaluates these views.

With respect to liquidity, realization makes the tax collection process relatively easy because it avoids the tax-without-cash problem. Assets are typically exchanged for cash, so taxpayers have the necessary cash to cover the tax liability at the time of sale.[50] In contrast, a mark-to-market regime may force cash-poor taxpayers to sell assets to pay their tax liabilities on unrealized profits. Hence, a mark-to-market system would require policymakers and tax authorities to undertake a politically unpopular and legally complicated position of dealing with some taxpayers’ inability to produce the cash necessary to pay taxes. Much of the political traction of the income tax relates to the way it considers taxpayers’ ability to pay and, through the realization requirement, does not disrupt taxpayers’ affairs by requiring them to sell assets or borrow money to pay their taxes.

Problems associated with taxpayers’ liquidity could be dealt with in a mark-to-market system. While low liquidity is not a completely manufactured problem, leading scholars have convincingly argued that problems associated with liquidity are not substantial enough to justify the broad imposition of the realization requirement.[51] For example, under a mark-to-market regime, taxpayers would be likely to structure their investment and borrowing transactions in ways that addressed liquidity concerns by making sure that they had sufficient funds to pay their taxes. [52] Additionally, lawmakers could provide exceptions with respect to certain assets (e.g., residential homes, closely held corporations). Such provisions would assure that liquidity-constrained taxpayers who owned certain appreciated assets were not required to pay the tax until they sold the asset.[53] These exceptions could be structured so that when the assets are eventually sold, the gains are taxed at nominal rates (as occurs under the current tax regime). Alternatively, they could be taxed at higher effective tax rates that tried to capture the value of deferral.[54]

The stronger argument in support of the realization requirement is that a mark-to-market regime would require burdensome annual asset appraisal. A coherent tax that accounted for accrued gains and losses would involve huge administrative costs associated with valuation.[55] Even the scholars who are most optimistic about the prospect of abolishing the realization requirement and shifting to a mark-to-market regime agree that certain types of intangible assets are very difficult, perhaps impossible, to value with sufficient accuracy.[56] This difficulty raises significant questions about whether valuation provided by taxpayers could be trustworthy—a serious concern in a tax regime dependent on self-assessment.[57] If the realization requirement were abolished, one could expect a significant increase in the amount of fact intensive, valuation related litigation between tax authorities and taxpayers.[58]

The realization requirement avoids these valuation problems because the price an unrelated buyer pays for the asset in a market exchange provides a strong indication of the asset’s actual value. By focusing only on the price garnered in actual transactions, a realization-based income tax allows tax authorities and taxpayers to avoid lengthy, costly, and inaccurate valuation processes. However, this type of audit accuracy comes at a cost. In order to rely on unrelated parties’ willingness to pay, tax authorities must wait, often ignoring significant gains for many years. Unlike problems of liquidity, problems of valuation remain a major obstacle to implementing a mark-to-market income tax regime.

While not denying their salience, we note that valuation problems are likely to play an increasingly smaller role in tax policy moving forward. First, the problem of valuation is constantly decreasing because the scope of information markets is rapidly growing, and asset valuation methods are improving over time. In fact, because of these two phenomena, many assets that once were considered impossible to value now have established market prices.[59] Second, the investment patterns of society have shifted.[60] Rather than investing in concentrated private ventures, individuals now invest large shares of their wealth in diversified portfolios of publicly traded assets.[61] This shift means that far more of the wealth held by individuals has an established market price. Furthermore, certain taxpayers, especially firms, already engage in fair market valuation of their assets for tax and non-tax reporting purposes.[62] This does not suggest that a shift to a mark-to-market regime would not involve significant problems and administrative costs related to valuation difficulties. It does signify, however, that the marginal costs of producing the extra valuations necessary for such a shift may be considerably lower today than they were in the past and, if the above trends continue, are likely to be still lower in the future.

The mitigation of liquidity and valuation problems is undoubtedly the strongest explanation for the presence of the realization requirement in all income tax regimes. However, this subpart questions whether problems of valuation and liquidity justify the requirement’s broad application.[63] This inquiry is necessary because the administrative concerns surrounding valuation and liquidity currently foreclose any normative or policy discussion about the appropriate role of the realization requirement.

That realization undermines the redistributive function of the income tax regime, which is one of the most effective redistributive tools of modern liberal democracies,[64] should be a point of concern. In our view, policymakers may find it beneficial to seriously reconsider the much deeper normative discussion of how realization promotes or impedes the income tax’s distributive objectives.[65]

II. With It or Without It? Existing Approaches to Realization

Given the consensus that there is no normative justification for the realization requirement,[66] the current theoretical discourse has centered on whether there is a way to remove the requirement while keeping the income tax. This part identifies and evaluates two general approaches within contemporary literature. The first advances the notion that the income tax regime is so irreparably broken because of the realization requirement that a shift toward a mark-to-market regime is desirable. The second argues that such a shift would be unwise because the comparative costs of the alternatives to realization may be higher than initially meets the eye.

This part reviews and critically assesses both approaches, observing that their proponents disagree about one issue: the comparative costs of the realization requirement and various mark-to-market regimes. Both approaches accept the notion that the realization requirement merely represents an answer to administrative difficulties associated with liquidity and valuation. Yet, as the final subpart argues, this acceptance is problematic because there are social costs to a century of stagnation in the tax discourse with respect to the realization requirement.

A. Without It: A Less Realization-Based Income Tax Regime

Commentators have advanced many proposals to reduce the tax distortions and inequities of the realization requirement.[67] While they differ significantly, each proposal contends that realization cannot be justified while simultaneously arguing that it should not be eliminated altogether.

The argument underlying proposals for a shift to a mark-to-market regime is relatively straightforward and compelling. Because they see no normative justification for the realization requirement, the commentators argue that the income tax regime should minimize realization’s social costs by limiting its role. To do this, policymakers should engage in a cost–benefit analysis to determine whether it is possible to tax the value fluctuation of certain assets independent of realization. This approach emphasizes that the administrative costs associated with valuation are often overstated. Many assets, particularly publicly traded instruments, are actually quite easy to value. This argument becomes stronger as public markets continue to grow in both scope and scale,[68] and the relative size of portfolio investment continues to grow as well.[69]

Advocates of broader mark-to-market taxation also emphasize that a tax based on changes in asset value would not require annual valuation of every asset. There are a number of ways to indirectly reduce the tax benefit of deferral and the problem of lock-in without annual valuation of all assets.[70] Further, inaccuracies in any one year would correct themselves over time and even out when the asset was ultimately sold, and since more assets would be valued, the impact of mispricing any specific asset would be greatly diminished.[71]

Diminishing the costs of the realization requirement would reduce some of its social costs and would result in a broader, more comprehensive tax base,[72] which would allow policymakers to maintain the same level of revenues while reducing high marginal tax rates (and their associated work disincentives).[73]

The argument for a shift to a mark-to-market regime is undeniably strong,[74] but also materially incomplete. It fails to give sufficient weight to two important points. First, none of the proposals advocates a shift to a comprehensive mark-to-market regime, and the claim that comprehensive mark-to-market taxation would be beneficial says little, if anything, about whether a shift to a partial regime would be positive. A partial mark-to-market regime covering fifty percent of assets would not be likely to eliminate fifty percent of the problems associated with the realization requirement. In fact, as described in the next part, it would likely give rise to new problems. The overall advantages of a partial mark-to-market regime are therefore somewhat speculative. Second, none of these approaches explains why the realization requirement should not be eliminated altogether through mark-to-market auditing when there is no problem of valuation.

B. With It: Some Hidden Justifications for Realization

Most advocates of the realization requirement support it only on a second best basis.[75] They advance two lines of argument: the costs of the requirement are not as high as they seem and the alternative to the realization requirement, a partial mark-to-market regime, would introduce new, deeper problems. Both of these arguments engage in a hypothetical cost–benefit analysis in an effort to determine whether a shift from the current system to a partial mark-to-market regime is justified. Much like the arguments in favor of a mark-to-market regime, this analysis provides little insight into the policy goals the realization requirement may promote. After discussing these two lines of argument, this subpart surveys and evaluates the relatively recent policy argument made by David Schizer in favor of realization.[76]

The first set of arguments stresses that there are two types of efficiency considerations—pre- and post-investment.[77] The realization requirement obviously distorts post-investment considerations of whether to engage in a sale or exchange of an asset. However, if imposed uniformly on all assets, it does not necessarily reduce the efficiency of pre-investment decisions between different investment opportunities and consumption.[78] Proponents of the pre-versus-post-investment arguments contend that because the realization requirement does not distort pre-investment decisions, the problems with it are overstated.[79] This analysis has two main weaknesses: it assumes that the realization requirement is the only rule that could be broadly applied and that the realization requirement’s distortion of post-investment decisions is low. In recent years, both assumptions have been called into question. First, experimentation with partial mark-to-market regimes shows that these regimes are administrable and that they could thus be systematically and coherently applied.[80] Second, the development of new financial instruments has introduced new pressures on the current tax regime by allowing aggressive tax planners to push the boundaries of realization-based rules to new frontiers.[81]

A different, and perhaps stronger, argument emphasizes that realization’s social costs may be lower than those of any feasible mark-to-market regime.[82] The advantages of a full mark-to-market regime would not flow to a partial one, which would remedy some problems only at the cost of aggravating others.[83] Because no reform proposal suggests applying a comprehensive mark-to-market regime,[84] taxpayers would have obvious incentives to be in one category,[85] and complex line-drawing rules would be required to reduce manipulation.[86]

Another argument for maintaining the requirement contends that most individuals do not consider unrealized gains as real.[87] According to some research in behavioral economics, individuals seem to distinguish between realized (real) and unrealized (paper) gains and significantly discount the latter for purposes of evaluating their wealth or welfare.[88] This process of “mental accounting” goes a long way toward explaining the requirement’s endurance.[89] These common perceptions about the realization requirement are important in terms of legitimacy and compliance, two crucial elements in the current self-reporting regime.[90] Yet the popular perception observation fails to explain why, and more importantly, how this vague notion of popularity should guide policymakers.[91] Moreover, since paper gains (and losses) are only discounted and not disregarded altogether, mental accounting at best suggests that unrealized gains should not be fully taxed—not that they should be completely ignored. For example, if the general public discounts paper gains by 30%, then 70% of the change in value should be taxed on a mark-to-market basis.

To be clear, we do not reject the public perception argument. As Parts III and IV demonstrate, we explain and, to a certain extent, justify it. Our only point here is that simply claiming the public wants a rule provides policymakers with little if any guidance as to how to structure good tax policy.

The observation that the realization requirement corresponds with public preferences underlies the only recent piece suggesting a policy justification for the requirement. In an insightful article, David Schizer argues that the realization requirement should be understood as a politically credible investment subsidy.[92] He suggests that, by allowing deferral and strategic trading, the realization requirement reduces the effective tax on investments.[93] It therefore should be viewed as one of many tax incentives that Congress uses to promote investment. Realization is unique among these incentives because of its political credibility. Unlike other subsidies for investment, the realization requirement has a relatively stable history,[94] has no significant political opposition,[95] and complements administrative needs of tax authorities.[96] Hence, Schizer contends that investors perceive this subsidy as credible and stable, and unlike other subsidies, it is unlikely to be changed by future legislation. Since investors see little likelihood of realization’s repeal, they are not expected to discount the value of its tax benefits. Despite the imperfections of the realization subsidy,[97] the lack of discounting by taxpayers makes the subsidy more efficient (from the government’s perspective) in achieving the goal of changing taxpayers’ investment behavior.[98]

Schizer offers a compelling argument that realization is a subsidy for investment. We are, however, skeptical, as to whether the case can be made that realization is a good or unique subsidy for investment. Schizer illustrates his claim that the realization requirement is a credible and legitimate subsidy by comparing it to the fluctuating policy of lower tax rates on capital gains. However, this comparison is only one of the many possible. First, it is important to note that the favorable capital gains rate applies only to individuals (and not to corporations), while the realization requirement applies to all taxpayers.[99] There seem to be other equally credible ways of directing a subsidy toward individuals—e.g., encouraging pension-saving vehicles,[100] encouraging home equity savings through the home interest deduction, and providing certain exemptions for gains on the sale of residential homes and closely held small businesses.[101] History informs us that these targeted saving incentives are credible and stable[102] and unlike a broad realization requirement, they carry more concrete and narrowly tailored social benefits. Furthermore, their costs in terms of regressivity and dead-weight loss may be easier to control than those of the realization requirement.

Second, and more notably, Schizer also overlooks the possibility that a partial mark-to-market regime could offer a more credible incentive for investment and risk-taking because of the way such a tax system would treat losses.[103] As mentioned,[104] the various loss-limitation rules create significant disincentives for risk-taking and investment and generate huge compliance costs.[105] Even a partial mark-to-market regime would make many loss-limitation rules unnecessary because whatever was taxed on a mark-to-market basis would take into account all unrealized gains and all unrealized losses—so taxpayers would not even have to dispose of an asset to deduct their losses.[106] This fundamental characteristic of mark-to-market taxation is perhaps the most credible investment and risk-taking subsidy imaginable.

C. The Policy Costs of a Theoretical Debate

The standstill in the debate about the realization requirement—with most commentators accepting that it is unjustified but virtually impossible to replace—has produced some troubling outcomes. Tax policy has stagnated as the intellectual resources devoted to its development have been channeled almost entirely into preventing the abuses made possible by realization. It would be naïve to expect that every aspect of a tax regime would be anchored to an explicit, principled policy objective. However, realization is not just another aspect of the tax regime. As the above analysis suggests, it is the most significant and problematic feature of the current income tax. The costs associated with the inability to link realization to any policy objective cannot, therefore, be overstated.

Within tax scholarship, disagreement over how to accommodate a transactional realization requirement has enhanced support for the idea of abandoning the income tax altogether. The notion that the realization requirement is indispensable to an income tax,[107] in spite of its serious costs, has prompted calls that favor shifting to a consumption–wage tax that would exempt all the income derived from investments and savings .[108] Such a tax focuses on certain wage-paying and/or consumption transactions that are easy to observe and define while completely exempting investment and savings from taxation. Since affluent individuals hold most of the capital assets that would become exempt under a consumption–wage tax, such a shift would reduce tax complexity at the expense of reducing its progressivity.[109] Because of the realization requirement, the current tax regime’s treatment of investment returns is arbitrary.[110] Thus, shifting to a consumption tax would actually sacrifice very little in terms of progressivity.

In the United States today, and in the developed world more generally,[111] the notion of tax progressivity has strong policy justifications and garners considerable political support.[112] If tax progressivity is a worthy goal, then the consumption tax is an undesirable answer to the predicament of realization.[113] Instead of throwing the baby out with the bathwater, it is time to critically reconsider the role of the realization requirement in contemporary income tax policy and theory.

III. A Normative Defense for Realization—Promoting Income Tax Progressivity

This article advances the notion that realization is tied to the progressivity of the income tax base and can therefore be seen as part of a broader attempt to promote wealth redistribution. It approaches the normative arguments for realization by asking what should happen if policymakers could accurately and costlessly observe the market price of all assets. That is, if valuation were no obstacle, would we want to subject all assets to mark-to-market taxation and, if not, why? The main argument is that while errors will occur under any system, not all errors are equal. If a primary goal of the system is redistribution, we should consider the distributive impact of different types of errors.

The first subpart explains the fundamental relationship between tax policy and redistribution. The second subpart argues that the realization requirement is normatively justified, but only with respect to certain emotionally non-fungible (personal) assets. Personal assets are broadly defined as assets for which the market price provides an inconsistent indication of the value taxpayers assign to them. With respect to these assets, the market price offers but a poor proxy for well-being and therefore should not be used by the tax regime for interpersonal comparisons. The third subpart asks what course of action policymakers should take knowing the high likelihood of error in measuring economic well-being by the market price of personal assets. It suggests that policymakers committed to income tax progressivity may choose to tax personal assets only upon realization. This analysis acknowledges that the taxation of personal assets only upon realization should be seen as a tax benefit. However, we argue that unlike most tax benefits, this benefit is relatively progressive in nature, favoring primarily the middle (rather than the upper) class. This is because personal assets comprise a disproportionately high percentage of less affluent taxpayers’ assets but a considerably smaller percentage of affluent individuals’ assets.[114] The fourth subpart replies to the major potential criticisms of our approach. It argues that making the income tax base more progressive by applying the realization requirement to personal assets has some significant advantages over other means of promoting tax progressivity.

A. Tax and Redistribution: At the Crossroads of Philosophy and Public Finance

It is necessary to clarify the scope of this article because questions of redistribution involve complicated notions of philosophy and public finance. Any type of redistributive regime is based on two fundamental policy decisions: what should be distributed and how it should be measured by the tax regime. The two are obviously related, with the latter becoming virtually impossible to directly address when the currency of justice (i.e., what should be distributed) is intangible and difficult to observe (e.g., utility). Hence, policymakers have to rely on proxies rather than try to measure the currency of justice directly.[115]

Unsurprisingly, there is strong disagreement among scholars about the proper distributive justice benchmark for measuring and remedying disadvantages. Political philosophers have suggested a number of such “currencies” through which well-being should be measured, including opportunities,[116] primary goods,[117] and capabilities.[118] Even though this article cannot resolve this question, it adopts the terminology of the literature that focuses on the distribution of welfare as a subjective measurement of well-being.[119] It does so because the welfarist terminology is consistent with most of the tax literature.[120]

When policymakers wish to formulate an effective distributive scheme, there is an obvious mismatch between what they seek to redistribute (e.g., welfare) and what they are actually measuring, taxing, and distributing (e.g., income). Hence, the current income tax regime is a result of the difficulty of directly observing how much economic well-being (e.g., in terms of welfare) individuals have and of the belief that income is considered a good proxy for economic well-being.

This article considers whether the market prices of different assets are good proxies for economic well-being. From a philosophical perspective, it is clear that there can be many cases in which the objective market price may not provide a good indicator for a subjective benchmark such as welfare. In these cases, relying on assets’ market prices for tax purposes becomes problematic because it leads to redistributive errors.

This article’s analysis adopts a public finance, rather than a philosophical, approach to this problem. It assumes that if large-scale redistribution is warranted, making some type of assessment of individuals’ well-being is inevitable. It also acknowledges that market price is policymakers’ main source of information, so that some errors are inevitable as well. We then identify categories of assets for which one could expect to find a significant mismatch between market price and what the tax system is aiming to redistribute. In other words, this article considers when market price is not an adequate proxy for well-being (and other potential currencies of justice) and evaluates the implications of that mismatch for the realization requirement.

B. Market Price, Personal Assets, and Interpersonal Comparisons

This subpart advances the claim that the realization requirement is normatively justified with respect to certain types of personal assets. For those assets, the market price may be an inherently inaccurate proxy for value and therefore should not be relied upon from a tax perspective.

The analysis begins with the observation that there is one point upon which everyone seems to agree: personal assets could not be taxed on a mark-to-market basis under any politically feasible tax regime.[121] In many cases personal assets may be difficult to value—e.g., old family furniture, real estate in places with low exchange activity. However, many personal assets are very easy to value—for example, real estate in certain cities, old cars, horses in the family stables, certain collectable items (e.g., stamp collections). The intuitive case against taxing these assets on a mark-to-market basis derives from the notion that market price may not adequately reflect their true value to taxpayers.

Advocates of shifting to a mark-to-market regime focus on publicly traded instruments, arguing that exceptions could be made for hard-to-value personal assets. Advocates of the realization requirement stress that valuation and liquidity concerns would make it difficult to tax closely held companies, certain farms, and real estate under a mark-to-market regime.

This article takes a different approach and questions what makes the mark-to-market taxation of certain assets so unappealing and asks if the answer helps explain the normative base of the realization requirement. This type of inquiry may help policymakers start succeeding in what they so far have failed in doing—promoting consistent general rules for applying the realization requirement.

There are only two pieces that touch upon the above inquiry. The first is by Louis Kaplow, who examined how human capital would be taxed under an ideal income tax, which would include a tax on unrealized earnings.[122] Kaplow did not make this inquiry to determine whether there is something unique about human capital that justifies applying the realization requirement to it. Instead, he argued that those advocating for a shift to an ideal income tax, which like a mark-to-market regime would tax changes in value, should account for how human capital would be treated under such a regime. In his opinion, if human capital received special realization-based treatment,[123] the advocates for the mark-to-market regime would have the burden of explaining why.[124] Some commentators have read Kaplow’s argument not as an inquiry into whether the realization requirement is justified but more as a question as to whether an income tax should be imposed on capital assets.[125]

The second is a recent article by the authors of this article that deals with the taxation of earning capacity as a point of intersection between public finance and political philosophy theories.[126] That article suggests that the market price for earning-capacity endowment is not an adequate benchmark for a tax-transfer regime. Earning capacity should not be perceived as a cash asset that individuals can simply use, because the decision of how to work is multilayered and also driven by preferences, moral convictions, and needs. In essence, we argued that the market value of earning capacity was insufficient as a normative foundation for the tax-transfer regime.

In this article we expand upon that preliminary analysis to determine how it can advance tax policy with respect to the realization principle in general. We agree with Kaplow that a pure mark-to-market income tax regime would require taxing the value fluctuation of all assets, including human capital.[127] We also agree with him that a rigorous mark-to-market regime would include other types of assets, such as personal belongings, residential homes, and family businesses. We broadly classify personal assets as assets in which there is a high probability that taxpayers’ subjective value is radically different (that is, significantly higher)[128] than the assets’ market value.[129] The higher personal value attached to these assets is prohibitive—that is, it renders the possibility of trading them on the market very unlikely. Even though it is impossible to comprehensively define emotionally non-fungible assets, one can think of them as assets of personal use or those that involve significant human capital contributions of the taxpayer or of his relatives.[130] This definition is somewhat crude and overinclusive, since not every asset that comprises private consumption or has human capital embedded in it has prohibitive personal trading costs. For example, the New York Times is a publicly traded company, yet it is controlled by the Ochs-Sulzberger family. Should the shares of the family be considered personal while all other shares are non-personal? Should all members of the family be considered as involved with the corporation or only those that have invested their human capital? This issue is of crucial importance, since much of the wealth owned by affluent taxpayers is in the form of equity investment in closely held businesses, which arguably could be considered personal assets.[131]

The above questions are indeed vexing, but we believe that a more precise definition is not necessary at this stage. As we discuss in Part IV, with a few refinements, the broad definition we provide for personal assets suffices for the purposes of promoting real-world reform recommendations.[132] Hence, to develop the normative argument, this part assumes it is possible to distinguish between emotionally non-fungible (personal) assets and other assets. While the task of translating this argument and its conclusions to concrete policy measures is taken up in detail only in Part IV, it is worthwhile to note a few of our conclusions to assure the skeptical reader of the practical value of our argument. First, investment assets, which are already defined in the code for many purposes, and are (almost by definition) non-personal, could be taxed on a mark-to-market basis. This partial shift to a mark-to-market regime would be a significant and administrable reform, which would not require a full definition of personal assets. Second, to accommodate revenue, distributional considerations, and concerns over tax avoidance, we suggest that tax authorities cap the amount of personal assets that should be taxed only upon realization at a certain fair market value.[133]

Our analysis departs from the notion that even though there are significant differences from person-to-person as to what assets are emotionally non-fungible, there are categories of assets whose values are very unlikely to have such idiosyncratic psychological components (e.g., publicly traded assets held by portfolio investors).[134] Similarly, there are broad categories of items—such as inherited family jewelry, homes, and self-created artwork—for which we can reasonably assume a substantial gap between taxpayers’ subjective value and the market price. With respect to these emotionally non-fungible personal assets, the existence of this value gap can be assumed with relatively great certainty, but the size of this gap varies heterogeneously and is not observable. While many people may view their grandmother’s wedding veil as a precious heirloom, only few would consider it priceless, most would be willing to sell it eventually if offered enough money, and some would sell it immediately for the highest bid a Craigslist post could drum up. Any attempt to base tax treatment on the market value of these assets would carry a serious risk of error.

An example may help to show why the market price is a poor indication of the value of personal assets. Imagine two individuals, both of whom value their kidneys, family homes, and employment as law professors as priceless.[135] The market, however, values the kidney of one professor as worthless (not because she is not healthy but simply because there are very few, mostly poor people who could receive a transplant from her) and the other’s as worth one million dollars (again not because she is particularly healthy but because she happens to have a kidney that is in high demand by Silicon Valley CEOs). In the same manner, the market assesses a negative value to one’s home (because a correctional facility is being built nearby) and a value of one million dollars to the other’s home (because a new technological park is being built fifteen minutes away). Furthermore, their human capital is priced dramatically differently by the market: while the former has the skills to be a successful human and labor rights lawyer, earning close to minimum wage, the other possesses the capacity to be a generously compensated tax planner to high net worth individuals. While the well-being of these two individuals seems to be rather similar, their tax liabilities under a full mark-to-market tax regime would be radically different because the personal assets they own, which neither of them intends to put on the market, are quite differently priced. Most people would probably consider this type of interpersonal comparison arbitrary and unreasonable for tax purposes, because the essence of the price signal is what others will pay for assets if they were put on the market.

In contrast, the market price would offer a valid proxy for well-being if each of the professors chose to sell these personal assets. When the professors choose to sell their family homes, they signal that they accept that the market valuation of the properties is greater or equal to the subjective values they assign to them. This acceptance of the market equilibrium validates the market price as a good proxy for their economic well-being, and this validation makes it a relevant benchmark for tax redistribution. This analysis does not mean to suggest that, with respect to all personal assets, a voluntary transaction is always an indicator of well-being.[136] On aggregate, however, the market return for the actual use or sale of personal assets offers a good proxy for economic well-being, which is not the case when the market price is hypothetical.

For tax purposes, it is difficult to draw a meaningful interpersonal comparison of well-being by focusing only on the market value of personal assets. In the context of these assets, the market value offers only a poor proxy for individual well-being because the subjective value of how to use these personal assets is determined by a complicated metric. In contrast, market pricing is a one-dimensional process that aggregates the maximum cash flow a certain asset could be expected to yield if put on the market. Simply put, the market values of assets such as our human capital are poor indicators of our well-being and therefore should not be used by the tax system for interpersonal comparisons.

Welfare, as mentioned, is not the only possible currency of justice,[137] and some important theories indeed rely on more objective measurements of economic well-being.[138] Our example assumes that both professors derive similar welfare from their personal assets, but this assumption is not necessary. One can reasonably argue that both professors are similarly positioned, in objective terms, regardless of how they value their personal assets. Each of them has a primary residence that she was able to choose, a job that allows her to meet her financial objectives and promote her intellectual interests, and a healthy body. To accept our analysis, one has only to accept that the market price does not capture all of the objective value associated with certain assets—not that redistribution of welfare is the only benchmark of equality. Put differently, if one agrees that both professors are similarly situated in objective terms, then the conclusion must follow that the market price is an objective measure of their economic ability but not necessarily the only or the best objective measurement policymakers can or should employ to determine that ability.[139]

To be clear on this point, the market price is a social construct and therefore not “correct” when it assesses Google stock or “wrong” when evaluating more personal assets. The difference is not that Google stock has a price and personal assets are “beyond pricing.”[140] Instead, the claim is that the tax authorities’ ability to make meaningful interpersonal comparisons is much greater when considering the price of Google stock than it is for the market price of personal assets. Individuals may invest in Google stock for a number of different reasons—speculation, risk-diversification, or belief in the company’s corporate values. However, for most shareholders, Google shares are primarily investment assets—and as such are not very different from money because they primarily represent consumption power.[141]

Investment assets, of course, are not the only type of assets that could be valued independently from the decision to use them. Money, for example, is in most cases simply a piece of paper that enumerates fungible and homogeneous units that represent individuals’ comparative abilities to participate in market exchanges. It is very unlikely that certain individuals would attach special immeasurable utilities to homogeneous and fungible cash assets.[142] Accordingly, the market assessment of the worth of monetary assets seems valid when making interpersonal comparisons of individuals’ well-being for tax purposes. Money and human organs lie on two ends of the continuum—everything else falls in between. When an asset is closer to the money side of the continuum, it is easier to see how its market value can serve as a proxy for well-being. It is therefore reasonable to assign tax liability for that asset’s fluctuation in value according to a mark-to-market regime. The assignment of mark-to-market tax liability seems less justifiable for assets closer to the personal asset end of the continuum. There may be a lot of very difficult cases in between, but it is possible to envision some preliminary classification rules. Primary residences could be categorically treated as personal assets as could small family businesses, which are likely to embody a great deal of psychological value and non-diversified risk taking. Other categories of emotionally non-fungible assets may include assets employed in active trade or businesses with respect to which taxpayers invest a considerable amount of their time. In contrast, portfolio holdings in publicly traded corporations and other entities with a large number of owners are generally emotionally fungible. These types of distinctions may seem too vague and impractical at first, but, as we discuss later, they align well with various distinctions on which the current tax regime already relies.[143]

No liberal egalitarian regime aspiring to engage in redistribution can avoid interpersonal comparisons. That interpersonal comparisons are unavoidable, however, does not mean that all interpersonal comparisons are alike. In Part I.A, we discussed why it is problematic for an income tax regime to base tax assessments upon interpersonal comparisons made only with respect to realized gains. In this subpart we modify this observation by arguing that, when it comes to personal assets, the market price of an unrealized asset is not a good indicator of well-being. In this respect, it is worth emphasizing the following points. First, taxpayers can attach higher (subjective) value to all assets—and not necessarily just to personal assets. However, from a public finance perspective, the important thing is that gaps between subjective and market value are more likely to occur with respect to personal assets.

Second, if subjective value is unobservable, policymakers cannot know how heterogeneous it is.[144] So if two taxpayers own residential homes with the same market price, tax authorities will be unable to distinguish between the taxpayer who values her home as priceless and the taxpayer who values his home at $1,000 over the market price. Hence, with respect to personal assets that are likely to have higher subjective than market values, tax policymakers operate in a scenario of extreme uncertainty.

Third, some commentators may argue that since all assets will eventually be taxed on their market price, the difference between a mark-to-market and a realization-based income tax regime is merely one of timing.[145] Hence, if policymakers tax all personal assets on a mark-to-market basis, the only difference between this regime and a realization-based regime would be that the former would eliminate the arbitrary advantage of deferral. This objection overlooks that many (and maybe even most) personal assets would never be realized. A great many personal assets, from family farms to jewelry, are often transferred by gift or bequest. Hence, the market is not the ultimate appraiser of many unrealized personal assets, which leaves policymakers with the difficulties associated with the unobservable varied subjective values attached to those assets.

Finally, we do not wish to deny that fluctuation in personal assets’ market values provides some relevant information about individuals’ economic well-being. Recall the example of the two professors. It is difficult to deny that the professor with the more valuable home and higher earning capacity is at least in some ways better off than her peer.[146] We argue, however, that more information is not necessarily better if it affects policymakers’ tax decisions in a way that does not seem to correspond with what the tax system is trying to achieve.

The relevant tax policy inquiry, therefore, is neither whether the market value of personal assets is equivalent to well-being (it obviously is not) nor whether it is a proxy for well-being (it obviously is). The relevant question is whether market value is a good proxy for well-being—and this subpart has established why the proxy value of the market price is poor with respect to emotionally non-fungible assets.

If the realization principle is justified only with respect to personal assets, it follows that this normative constraint does not apply, or applies with much less force, with respect to other assets. The upshot of this analysis is that if policymakers consider income to be a good benchmark for redistributive taxation, then there may be a case for adopting a hybrid mark-to-market-realization tax regime. This regime would tax certain assets, namely investment assets, under a mark-to-market system while taxing personal assets on a realization basis. The next subpart argues that in the context of such a regime, the realization requirement offers a reasonable policy tool to assure the progressivity of the income tax base. Part IV explains the fundamental attributes of such a hybrid income tax regime and suggests a realistic proposal for reform that offers improvements over the current realization-based regime in terms of complexity, equity, and revenue.

C. Realization as a Progressively Distributed Tax Benefit

This subpart establishes why a hybrid tax regime that taxes personal assets on a realization basis and other (namely, investment) assets on a mark-to-market basis would help promote a more progressive redistribution of income in society. The key point is that personal assets represent a larger portion of the wealth of low and (especially) medium-income taxpayers than of the wealthy.[147] Put differently, such a regime would promote progressivity because investment assets, which categorically are not personal assets, are disproportionally held by affluent taxpayers. Therefore, providing a tax benefit to personal assets, and not to other assets, would make the tax system as a whole more progressive. This claim should be integrated with the claim of the above subpart rather than understood as free standing. Having a progressive income tax does not require giving realization-based treatment to personal assets. However, given the difficulty of using the market value of personal assets as a proxy for well-being, taxing only those assets upon realization should be viewed as a progressive move. If confined to personal assets, the realization treatment would still be a tax benefit and would continue to cause some distortions and inequities. However, given that personal assets comprise a much greater component of non-affluent individuals’ resources, it would be, overall, a progressively distributed tax benefit. This subpart only argues that limiting the application of the realization requirement to personal assets makes the tax base more progressive. The more complicated task of discussing why tax progressivity is warranted, and whether taxing personal assets only upon realization comprises a sound policy to achieve progressivity, is left for the next subpart.

At the outset it is clear that even though personal assets are not unique to less affluent taxpayers, they comprise a much larger proportion of those taxpayers’ wealth. For example, in a pure mark-to-market system both Joe Six Pack and Bill Gates would be taxed on the change in market value of their identical golden wedding rings. Assume that Gates values his ring at $500,000, Joe values his at $5,000, and each ring has a market value of $500. The strength of the mark-to-market approach is that it avoids inquiries about subjective value and focuses solely on the market price as an objective and value free way to measure economic ability.

The important point, however, is that even if both rings were identical, exempting them from mark-to-market treatment would not result in a distributional wash. The truth is that Gates’s ring represents a much smaller percentage of his overall wealth than Joe’s ring does for him. Hence, in a revenue-neutral setting, where taxes forgone have to be raised somewhere else within the system, granting a tax benefit to wedding rings so that both would be taxed only upon realization would increase Gates’s effective tax burden. If the government wants to maintain a constant level of spending without adding new taxes, and is not able to tax wedding rings and other personal assets that both Joe and Gates have, it would have to increase tax rates on other types of assets—namely investment assets—that only Gates has.[148]

The costs of detachment from one’s personal assets are not unique to wedding rings. For example, if instead of a wedding ring one thinks about identical baseball card collections, gold watches, or old family farms, one will get to precisely the same intuitive results. High detachment costs characterize basically all types of personal assets including human capital, homes, and small family businesses. A tax professor values his job partly because he values the freedom and flexibility that many tax planners do not have. Many elderly people do not wish to move to different apartments or retirement homes because leaving their homes is emotionally costly to them. From a neo-classical economic perspective, these costs represent (some out of many) common factors that impact the supply curve.

In terms of progressivity, the example of humble golden wedding rings is not very representative. The case for the progressivity of applying the realization requirement to personal assets may be much more complicated with respect to other types of assets—e.g., human capital, family farms, residential real estate. Generally, wealthy taxpayers tend to have a higher number of (more valuable) personal assets than poor and middle-class taxpayers. Hence, the progressivity tax benefit is questionable, since much of it would not be directed towards the lower economic quintile of society. Middle class individuals have much of their wealth concentrated in home equity, private closely held businesses, and human capital. Allowing all or some of these assets to be taxed only upon realization would provide a tax benefit that is less progressive than what Gates’s and Joe’s wedding bands may suggest.

The above observation is evidently true, but it fails to capture the core of the argument presented. First, when discussing income tax progressivity, it is important to bear in mind that low-income taxpayers play a relatively minor part in this debate. Personal exemptions, tax credits, and low marginal tax rates mean that low-income taxpayers pay little or no taxes.[149] Even though assessing what people have may also be important for transfer purposes,[150] when it comes to taxes, the question of allocating the tax burden is primarily a tradeoff between the middle class (whose wealth is primarily in the form of human capital and other personal assets) and the upper class (who also have significant amount of investment assets).

Second, policymakers should question whether, given the difficulty of taxing personal assets on a mark-to-market basis, the advantages of a mark-to-market regime should be abandoned altogether. More specifically, they should inquire whether there are reasonable ways to achieve the distributional advantages of the mark-to-market regime without subjecting personal assets to it. If savings and investment assets were taxed on a mark-to-market basis, the effective tax rate on them would rise. Therefore, the option of subjecting these assets to mark-to-market taxation promotes a relatively progressive alternative to the current regime.

Indeed, under our proposal the realization treatment of personal assets would continue to benefit the middle class, but by the same token, mark-to-market taxation of investment assets would increase the tax burden on wealthy individuals.[151] Such a mark-to-market regime would eliminate the benefits of tax planning, strategic trading, and deferral opportunities with respect to capital assets. It would therefore result in a higher effective tax rate on investment assets (if tax rates on them remain the same as on personal assets) as well as more transparent, equitable, and efficient taxation of such assets. Implementing this type of selective realization requirement for personal assets would position the income tax more as an elite tax—a notion that is not in any way new.[152]

By exploring the unique case of certain personal assets, we argue that not all errors are distributionally equal and that policymakers should take errors other than incorrect market valuations into account. Tax systems rely on market valuation because it is a proxy for economic well-being. However, in certain cases the market price offers only a weak proxy, so relying on it is an error in itself. By highlighting the different distributional impact of various types of errors, this article hopes to add a new analytic consideration to the discussion of realization.

This argument deviates from traditional tax analysis, which associates base progressivity with a broad and comprehensive base—one that employs few exceptions and exemptions.[153] The ability of the income tax base to promote wealth redistribution is a combination of two major factors: the tax base (what is taxed) and the rate structure (how much is taxed). Any income tax regime can make a change in one of these factors and remain distributionally neutral if it modifies the other factor. For example, if the top marginal tax rates are increased for households earning more than $250,000 a year, the income tax would become more progressive unless the tax base were adjusted. This adjustment could take a number of forms—exempting certain types of savings (e.g., college savings) or providing benefits to certain consumption activities (e.g., new, hybrid cars) that affluent households are more likely to engage in. In this respect, this article offers an intellectual supplement to the traditional perception of tax base progressivity. Rather than arguing for a comprehensive benefit-free tax base, it contends that a progressive tax base could be attained by endorsing a tax benefit that is progressively allocated. This is precisely what we have argued with respect to applying the realization requirement only to personal assets. Such a deviation from a pure mark-to-market income tax regime would make the tax base more progressive because the tax benefit would be allocated to the owners of personal assets— assets that represent a larger portion of the wealth of less affluent individuals than of the rich.

As mentioned, taxing individuals according to their economic well-being requires certain interpersonal comparisons, which may be inherently inaccurate.[154] However, not all tax assessment inaccuracies are created equal. This article introduces this insight to the realization-mistake literature, showing that the costs of certain realization inaccuracies may disproportionally fall on low and medium-income taxpayers in a way that undermines the notion of tax progressivity.

D. Advantages and Limitations of the Proposal

A few lines of criticism may be deployed against this argument. First, on a very basic level, some may contend that the income tax regime should not promote values of progressivity and wealth redistribution among taxpayers. Second, even for those who accept that tax progressivity is a worthy policy goal, there is still a question of whether the tax system should become more progressive. Third, it is not clear that granting a realization tax benefit to personal assets is the best way to promote tax progressivity. The realization requirement features some fundamental arbitrariness, so embracing it makes redistribution inherently inaccurate. Similarly, other mechanisms (e.g., allowing higher personal exemptions or employing higher marginal tax rates) could prove to be as effective. Fourth, if personal assets are held in a progressive manner, there is the question of why their taxation should be deferred by the realization principle rather than completely exempt or simply taxed at lower rates.

In capsule form, our answers to these criticisms are that with respect to the first and second lines of criticism, we point out that while the desirability of tax redistribution goes beyond the scope of this paper, there is strong theoretical and popular support for a certain level of tax progressivity. The third and fourth potential criticisms question whether this article’s proposal is the proper way to structure income tax progressivity to effectively promote redistributive objectives. The reply to these questions stresses one core idea. This article does not try to propose the adoption of a theoretically clean yet impractical model. Instead, it seeks to suggest how tax policymakers can try to improve the current redistributive tax arrangements in the messy reality in which they operate. In this context, applying the realization requirement to personal assets is a realistic measure that reasonably balances distributional objectives with efficiency, revenue, and tax administration concerns.

The first line of criticism highlights two different questions inherent in the debate over tax progressivity: whether wealth redistribution is a desired objective and whether the tax system is the proper way to promote redistribution. These questions are extremely broad, so this article can merely point to the relevant political philosophy and tax theories addressing them. Most contemporary liberal political theories are explicitly egalitarian to some degree.[155] This means that they do not categorically accept the fairness of the distribution of assets by the market and that they justify coercive state action to attain wealth redistribution. Progressive wealth distribution is typically not a goal in itself, but only a proxy for meeting a distribution of something more fundamental (e.g., welfare or resources, as explained in Part III.A). Even though the distributional question involves many factors and nuances,[156] it is generally accepted that promoting goals of wealth redistribution justifies coercive state action.[157]

Philosophers and tax scholars tend to agree that, if wealth redistribution is explicitly incorporated into the social welfare function, the tax regime is an effective and efficient way of promoting it.[158] This overall consensus does not downplay the serious normative questions as to whether this means that a just tax system has to be progressive.[159] It also does not claim that achieving distributive justice is or should be the only or the primary objective of the tax regime. It does however, reflect the notion that the tax system comprises a key, and some would say the prominent, element within the arsenal of tools that policymakers have to promote redistribution of wealth.[160] This suggests that modern liberal states should consider employing a progressive tax regime to meet their wealth-redistribution objectives.[161] The use of progressive taxation also seems to carry significant support among the American public.[162]

The second type of criticism asks whether more progressivity is warranted. The reply to this criticism is that the current proposal does not contribute to the discussion of how progressive the tax system should be, but rather to the discussion of how a progressive tax system should be designed. Policymakers could use our proposal to make the tax system more progressive, or they could use it to maintain a distributionally neutral tax regime—by, for example, reducing marginal tax rates, which would make the tax rate structure less steeply progressive.[163]

The third line of potential criticism against our argument questions whether providing a realization-based income tax treatment to personal assets is the best way to promote income tax progressivity. This criticism would stress that there are other, potentially more accurate ways to assure that the income tax regime has an overall progressive effect. One of the biggest concerns is that giving a special realization treatment to personal assets, such as family farms and closely held family corporations, would encourage taxpayers to overinvest in them.

The brief reply is that policymakers should not be asking whether providing realization tax treatment to personal assets would be the perfect way to promote tax progressivity. Instead, the key inquiry is what course of action policymakers should take given the normative difficulty of drawing interpersonal comparisons with respect to the market price of personal assets.[164] The more difficult it becomes to observe what we are trying to distribute, the more inaccurate and susceptible to abuse and the less redistributive the tax regime becomes. Even if policymakers can observe the market price of personal assets, there would be questions of how much it could help them to promote redistribution. The high probability that taxpayers attribute higher, yet heterogeneous, subjective values to personal assets raises questions of whether and how the market price signal of personal assets should be used for redistributive tax purposes.

In the context of our argument, policymakers who want to use the tax regime to redistribute wealth and have only the market price of assets to do so face four alternatives.[165] One is to maintain the current realization-based income tax regime and endure the enormous social costs associated with taxing all assets on a realization basis.[166] The second alternative is to tax all assets on a mark-to-market basis—thus paying the costs of inaccuracy and perhaps regressivity of taxing personal assets based on their market value. The third alternative would be to limit taxation to observable assets, to exempt personal assets altogether, or to shift to a more easily observed tax base—e.g., wages or consumption.[167] This, however, would narrow the tax base and require very high marginal tax rates to attain redistribution.[168] The fourth alternative, which this subpart explores, is to provide realization-based tax treatment only to those (personal) assets that are difficult to tax on a mark-to-market basis. As established, if the realization requirement were applied only to personal assets, the tax base would become more progressive overall. We concede that this method of attaining progressivity raises some concerns. In its crude form, the realization requirement is not sensitive to the amount of personal assets individuals may have—allowing the middle class to relish the majority of the tax benefit. Furthermore, as discussed in the next part, it would create incentives to over-invest in personal assets, which would result in tax-avoidance opportunities and complex administrative countermeasures. Most importantly, by adopting realization as a way to promote progressivity, this article’s proposal inevitably accepts some of its inherent timing arbitrariness.

Given these difficulties, it is important to examine whether higher personal exemptions or higher marginal tax rates can (by themselves) better achieve income tax progressivity. Policymakers should therefore compare our proposal with a universal mark-to-market regime, which attains progressivity by adjusting the tax rate and level of personal exemptions rather than by providing realization treatment to personal assets.

Tax progressivity is typically associated with the progressive tax rate structure rather than with the progressivity of the tax base.[169] In fact, many of the arguments against progressivity relate to the social costs associated with the progressive rate structure as a method rather than to its underlying objective of promoting economic equality.[170] It has been convincingly argued that progressive tax rate structures with high marginal tax rates cause or aggravate many of the notorious income tax complexities[171] and that they generate significant work and investment disincentives.[172]

The actual behavioral impact of tax rate progressivity has been disputed,[173] but it is difficult to deny that high progressive tax rates produce distortive incentives and tax-planning opportunities. It is therefore widely accepted that tax rate progressivity’s major drawback is the unavoidable tradeoff between allocating the tax burden equitably and reducing overall social productivity.[174] The high social costs of attaining overall progressivity of the tax system through high marginal tax rates suggest that policymakers would be wise to explore other means of achieving this goal. In the context of this inquiry, tax policy analysis should focus on how to make the tax base more progressive. A more progressive base would allow policymakers to decrease high marginal tax rates and thereby reduce the overall costs of employing a tax-driven wealth redistribution scheme.[175]

Taxing personal assets on a realization basis is not the only way to make the income tax base more progressive. Allowing for personal exemptions that are sensitive to economic hardship would also make the tax base more progressive as a whole.[176] If this is true, a general mark-to-market regime could cover all assets (including personal assets), while personal exemptions could make the income tax more progressive. It seems, however, that even a carefully structured mark-to-market regime that provides relatively generous personal exemptions would result in problematic interpersonal comparisons. For example, as long as it relies upon the market price of personal assets as a benchmark, such a regime would impose a higher effective tax rate on the professor with the valuable assets than on her colleague with the less valuable ones.

Finally, while we recognize that granting realization treatment only to personal assets would undoubtedly distort ownership patterns, there are reasons to believe that the impact of these distortions would not be huge. Because of the high subjective value taxpayers attach to personal assets (and the costs of detaching from them), one can view most taxpayers as being neither short nor long on those assets.[177] In other words, the ability of taxpayers to manipulate their holdings of these assets to achieve a tax advantage exists but is limited. Taxpayers may not be as tax efficient when it comes to holding of personal assets—for example, if the taxpayer sells a residential real estate property, she becomes short on that property. While she can rent for a while, the costs of shifting in and out from a residential property are high, which drives down the likelihood of manipulation. Additionally, while granting realization treatment only to personal assets may result in overinvestment in businesses that fall under the personal asset category, there are other ways to address these concerns.[178]

The fourth criticism, that exempting personal assets altogether or subjecting them to a lower rate is more consistent with this article’s claim that they are progressively distributed, is the most difficult objection. One may wonder why a sale of a personal asset should trigger a tax at all. If a taxpayer sells a personal asset, which he initially valued significantly more than the market price, it may be because he is actually less well-off. For example, a taxpayer may have bought a residential home, enjoyed it very much, and been unwilling to sell it. After a while, however, he may have found out that he was not that fond of the house anymore and may have sold it for its fair market value. In other words, the taxpayer sold his personal asset because he experienced a loss of utility with respect to it. Because the realization event is a byproduct of a utility-loss (rather than gain), making it a trigger of taxation seems utterly unfair.

We have two replies to this concern. The first is that in public finance the focus is on rough justice rather than upon particular circumstances. Hence, as mentioned in Part III.A, we focus on the distribution of personal assets as a class of assets rather than on the changes in the specific price of an asset. Since personal assets comprise a bigger share of low to medium-income taxpayers’ assets, then giving them an advantageous tax treatment would result in increased progressivity, even if in the case of a specific asset it may be difficult to see why the sale justifies triggering a tax.

The second reply is that, as noted, in some extreme cases there may be a good reason to exempt the returns for personal assets altogether.[179] However, in most cases there seems to be a strong case for taxing the income generated from the sale and use of personal assets. For example, it seems unreasonable to tax individuals according to the market value of their human capital. Although many healthy women can work as gestational surrogates, the market valuation of this ability is a poor indication of their economic well-being.[180] The argument changes if a specific individual makes a voluntary decision to work as a surrogate.[181] The monetary returns for surrogacy are not fundamentally different from those received through other employment options. If society chooses income as the relevant benchmark for measuring economic well-being, different sources of income should be treated the same.[182] In this context, the realization principle operates to avoid problematic interpersonal comparisons with respect to human capital, while making sensible comparisons with respect to income actually earned.

It is important to stress that this article’s analysis neither advocates nor rejects the notion that personal assets should be subject to a different tax rate—or even to a form of retrospective taxation.[183] It merely suggests that even if the market price of assets is easy to observe, a mark-to-market tax regime should not apply to personal assets. To accept our argument, one need not subscribe to the notion that realization-based treatment of personal assets provides an a priori superior way of promoting income tax progressivity. Instead, one need only accept that a mark-to-market regime has important benefits and that these benefits are not completely erased if some assets are exempt from it.

The skeptical reader should further agree that, given the difficulties of drawing interpersonal comparisons with respect to the market price of personal assets, there are some important benefits of excluding them from the mark-to-market regime. Additionally, one should accept that there are high social costs to progressive tax rate structures with high marginal rates. Given these costs, policymakers should be willing to consider achieving income tax progressivity through a combination of measures—including the enhancement of the progressivity of the base. A realization requirement that is confined to personal assets would provide a tax benefit, which would erode the theoretical ideal income tax base but would do so in an overall progressive way.

The conclusion of this analysis is that there is a strong normative argument for shifting the current realization-based income tax system towards a decidedly more hybrid regime. This hybrid system would distinguish between personal assets, which would be taxed upon realization, and emotionally fungible assets (namely, investment assets), which would be taxed under a mark-to-market regime. Such a tax regime would broaden the tax base to include almost all investment assets and would make the income tax more progressive, efficient, and resistant to tax planning.

Despite the above, critics may legitimately argue that there are other, theoretically superior, ways of promoting tax progressivity. Nevertheless, we argue two things. First, as we stress in this subpart, every tax (and transfer) distributive scheme involves certain social costs,[184] so that the optimal redistributive policy may require using different elements rather than relying on a few of them extensively. Second, as the next part suggests, tax policy should be pragmatic and not merely theoretically correct.

The potential shift to a more hybrid income tax regime involves a number of complicated issues. Hence, the normative justification for a hybrid tax regime does not automatically justify an actual shift to such a system. The next part discusses how our proposal could help establish a better, more progressive tax system that would not require a complete overhaul of the existing income tax.

IV. The Road Ahead—Preliminary Policy Implications

This part examines how this article’s normative analysis could be meaningfully integrated into current tax arrangements. Taking a radically different view of realization, this article offers only the foundational normative argument for a hybrid income tax regime. It is part of a much broader project that seeks to re-examine how a more progressive and efficient income tax base should be structured, so a detailed policy proposal lies beyond the scope of the current analysis. Instead, the discussion below seeks to draw attention to two major points. First, it briefly explains how our argument supports a concrete and immediate change in the current tax regime—the taxation of all investment assets on a mark-to-market basis. This proposal’s main benefits are that it broadens the income tax base while expanding its progressivity, it does not increase complexity because our current income tax regime already distinguishes investment assets and applies special rules to them, and it reduces much of the manipulation and economic dead weight loss associated with the realization requirement. Put differently, this article’s proposal builds on a distinction already made by the current realization-based income tax regime but derives more value from it without adding much complexity to the system.

Second, proposing mark-to-market taxation of investment assets is only the first and most straightforward option for reform—but not, by any means, the only policy implication of this article’s analysis. Offering the novel category of emotionally non-fungible (personal) assets as a way to determine the tax treatment of assets suggests many more avenues of research. This part outlines the main trajectories of likely future research and positions them within the scholarly literature.

A. Taking the First Step: A Modest Real-World Reform Proposal

The category of emotionally non-fungible personal assets may seem vague and difficult to monitor. While this may be true, there is at least one category of assets that clearly cannot be identified as personal: investment assets. Therefore, at the very minimum, our analysis suggests that investment assets should be taxed on a mark-to-market basis.

The advantages of such a system are straightforward—with respect to investment assets, all the problems mentioned in Part I.A would be eliminated. Investment assets are the easiest vehicle for manipulating the realization requirement, particularly through inefficient practices such as strategic trading.[185] Mark-to-market treatment would eliminate tax barriers to the efficient allocation of investment assets and allow taxpayers to attain certain investment goals—such as risk diversification—without triggering a tax liability. Furthermore, investment assets are disproportionately held by affluent taxpayers, so subjecting them to mark-to-market taxation would deny them the tax benefit of realization for these assets, thereby increasing the progressivity of the income tax base.[186]

We can anticipate two main difficulties with such a proposal, neither of which is insurmountable. The first is how to define investment assets, and the second is how our proposal would manage issues of valuation and liquidity. With respect to the first problem, the category of investment assets should include individuals’ portfolio investment in financial markets, debt instruments, and any non-control equity holdings in entities.[187] It is important to note that the current tax regime already employs rules that distinguish investment assets from other types of assets.[188] These rules are biased against investment assets because they prevent taxpayers from using deductions generated by investment activity to offset the income derived from ordinary trade and business or as compensation for services.[189] Congress and the U.S. Treasury enacted these rules as a response to one of the most basic tax planning strategies in a realization-based income tax regime:[190] the acceleration of deductions generated by investment activity to reduce tax liability on income derived from other sources.[191] The upshot of this discussion is that while it can certainly be difficult to make distinctions between types of holdings, this difficulty is not fatal to our proposal. In fact, the current tax regime makes such distinctions all the time.

With respect to the problems of valuation and liquidity, we concede that they might remain a source of concern under this proposal. Instead, we argue that policymakers can address those problems much more easily in the context of investment assets than with respect to all assets. Many investment assets are traded and therefore are relatively liquid and easy to value. As noted earlier, we believe that liquidity problems are not detrimental to this proposal because the current literature has identified the need to make exceptions in such limited cases of low taxpayer liquidity.[192]

Valuation is the main explanation for why the realization requirement has endured as a second best alternative[193] and probably remains the main impediment to adopting a mark-to-market regime.[194] Nevertheless, the notion that valuation will always render the mark-to-market alternatives impractical is likely to eventually fail. If one assumes that public markets will continue to grow in both scope and scale[195] and advancements in computation and data analysis will continue to grow at an increasing rate, then the implication of this is that the market prices of a growing number of assets will be more easily determined and observed in the foreseeable future.[196] Hence, the notion that tax authorities would be able to value and tax assets under a mark-to-market income tax regime is realistic for an increasingly large proportion of investment assets.[197]

Because current income arrangements rely on the ability to distinguish between investment assets and other assets, using these same distinctions to promote a hybrid mark-to-market regime would not add significant complexity to the system. In other words, the costs of distinguishing between investment assets and other assets are already present in the current income tax regime. While our proposal does not reduce these costs, it achieves many advantages of the mark-to-market regime, appears to be administratively feasible (if not now, then in the near future), and reduces the complexity of many other rules,[198] along with much of the revenue loss and waste associated with tax planning.

In terms of future research, we believe that the above proposal for taxing all investment assets on a mark-to-market basis would be a significant improvement over the current regime but is not necessarily the ideal solution. Even though it is difficult to determine the precise efficiency and distributional outcomes of a hybrid regime, we believe that a more nuanced distinction between emotionally non-fungible assets may be possible.[199] This article’s analysis establishes why limiting realization benefits to personal assets would increase distributional fairness and calls for expanding mark-to-market treatment to non-personal assets. However, the remainder of this part stresses that the benefit of such an expansion should be weighed against its social costs.[200]

B. Future Avenues of Research

In suggesting a new category of personal assets, we lack empirics to propose any bolder tax reform proposal than what we have outlined with respect to investment assets. However, we believe that it is worthwhile to consider some potential areas in which this article’s normative analysis could be developed. Earlier, we described the realization literature as dealing with the consequences of tax assessment mistakes.[201]

However, as we explained, market valuation issues should become less of a problem over time—a fact that only highlights the problem with personal assets. If the long feared mistakes in market price assessment are indeed becoming less common, then the relative role of mistakes with respect to taxpayers’ subjective values of personal assets would increase. Put differently, we believe that as market valuation of both personal and non-personal assets becomes more accurate, policymakers will more frequently have to confront the problem we identified: taxpayers’ economic well-being from personal assets is not accurately reflected in the assets’ market prices. Future research on this topic requires empirical inquiry about the relative proportions that personal assets represent in the overall pool of assets. However, it should also focus on theoretical inquiries with respect to tax policy. Another important issue in shifting to a hybrid regime is how tax authorities should distinguish between personal and non-personal assets. We first explain the main theoretical threads of the topic and then elaborate on how they affect our argument for a hybrid mark-to-market realization regime.

By their very nature, partial mark-to-market regimes differentiate the tax treatment of certain assets and taxpayers. This differentiation may result in discontinuities that carry potential for distortions and inequities.[202] The key element in determining whether a shift to a partial mark-to-market regime would be beneficial is whether tax authorities could easily distinguish between types of assets. This is because such a shift would require tax authorities to develop a set of coherent and consistent recognition rules to distinguish between assets that should be taxed on a mark-to-market basis and those that should not. The benefits and administrability of a partial mark-to-market tax regime would turn on the precision of the recognition rules and the elasticity of demand for specific assets. Applying mark-to-market taxation to assets that are very elastic—perhaps because there are many non-mark-to-market assets that can easily substitute for them—would be difficult or counterproductive. If taxpayers could costlessly transition to a substitute, most would try to do so, which would result in no social gain in terms of revenue, efficiency, or equity.[203] In this type of situation it may be advisable to either avoid mark-to-market taxation of the asset or include some of its substitutes as well.[204]

An example could help identify some of the main difficulties of drawing a distinction between assets in a hybrid income tax regime, which taxes all non-personal assets on a mark-to-market basis. Consider Gill Bates, a young businesswoman who created a start-up internet company that captured a significant market share. Even though the company made an initial public offering, Bates continues to manage the company and maintains significant holdings in its stock. Bates could argue that even though the shares are liquid and easy to value, their (substantial) appreciation should only be taxed upon realization. For years, she has invested her human capital in the corporation and plans to continue doing so. Accordingly, there seems to be a lot of truth in the claim that her holding in the corporation’s stock is an emotionally non-fungible asset.

Allowing Bates’s holdings to be taxed only upon realization would, nevertheless, be problematic from a number of perspectives. First, taxing her holdings in the corporation only upon realization would have significant costs in terms of revenue and progressivity.[205] Second, if her holdings were not taxed on a mark-to-market basis, she would have incentives to try and manipulate her tax liabilities. For example, she could overinvest in the assets that could be labeled as emotionally non-fungible and defer the tax payment on their appreciation. She could use the corporation to attain risk diversification by retaining the corporation’s earnings and using them to buy a diversified portfolio so that her holdings in the corporation partially reflected the value of this portfolio.[206] This, as mentioned, would result in misallocation of resources, less risk diversification, more administrative and compliance costs, and all the flaws of the current realization regime.

It is important to note that this example is not an indication that a hybrid tax regime is impossible to implement but rather a demonstration of the fact that pushing a single principle to an extreme can present complex questions that may not apply across the board. Admittedly, our analysis suggests that emotionally non-fungible assets should be taxed only upon realization in order to prevent errors related to subjective valuation. The analysis does not in any way suggest that preventing this type of error is the only objective of the tax system. Revenue, distributive, and tax enforcement objectives may require that this goal be balanced against other values. In the case of Bates, these considerations may require that her ability to claim realization treatment for emotionally non-fungible assets be capped at a certain dollar amount. This capping, however, would still recognize the basic point advanced by this analysis: realization is a normative component of how a liberal society values individual well-being and promotes redistribution.

As one commentator has noted, the weaknesses of both pure realization and pure mark-to-market regimes leave policymakers with no option but to balance between them.[207] A hybrid regime requires line drawing between similar assets—a practice that is always difficult and somewhat arbitrary.[208] Wherever the line is drawn, assets on each side will be assigned different tax treatment in spite of their strong similarities.[209] Nevertheless, a frictionless tax regime is not a possibility, so line drawing will always be a significant component of tax policy debates.[210] Accordingly, any future research exploring the nature of a hybrid mark-to-market regime will have to evaluate whether the line drawing associated with the research is comparatively easier or less distorting than those associated with the one it seeks to replace.

We wish to end by noting a set of broader policy issues to which this article’s argument relates. First, with respect to the income-tax-versus-consumption-tax debate, if an effective hybrid income tax regime could be developed along the lines suggested by this article, then recent calls of some leading tax scholars to abandon the income tax are premature and misleading.[211] Although the realization requirement is indeed inherent to income taxation, limiting its role to personal assets can reduce many of the problems of the income tax and sustain it as a much more progressive and effective fiscal instrument.

Second, a shift to a more hybrid mark-to-market tax regime highlights other vexing tax policy questions, particularly with respect to the budgeting process. A mark-to-market tax regime would have to take account of more gains during economic booms and more losses during economic downturns, and would therefore require a very strong political commitment to prudent countercyclical spending.[212] Additionally, a shift to a hybrid mark-to-market tax regime would call into question the desirability of taxing income rather than wealth.

Third, from a historical perspective, it seems as though much of the justification for taxing income in the first place was the necessity of relying on the realization event to determine the actual value of assets. There are, of course, independent arguments for why income rather than other tax bases such as consumption or wealth should be taxed.[213] However, once the historical connection between income and the realization requirement is weakened, the question of whether income tax is the most appropriate tax base merits re-examination.

Finally, and most importantly, we have argued that current tax literature relies on tax administration difficulties with respect to valuation of assets to justify the realization requirement. This article’s argument is a byproduct of tax authorities’ inability to use the market price as a good proxy to determine the well-being (or other currencies of justice) associated with an asset. One may observe that our argument can also be viewed as an administrative constraint. After all, we start by explaining that the main problem with realization, as it is perceived today, is the difficulty of making accurate market price valuations. Our claim is that this ability to make accurate market valuations may change over time and that the real problem is the inability of tax authorities to measure the subjective value (or utility) that individuals derive from personal assets. One can claim that policymakers’ capacities to observe utility may change over time. These two types of constraints should be distinguished. The current market valuation constraint is something that technology can change, and is actually in the process of changing. As a field in finance, asset pricing constantly advances, so that an increasing number of assets can be consistently and cheaply priced. This is not the case with the gap-in-value argument because technology cannot help us measure welfare and other benchmarks of economic well-being. For this reason, policymakers cannot avoid making some normative assumptions about how to measure well-being for tax purposes.[214] This article cannot explore this point, but it is important to note that, ideologically, we would welcome a discussion about the limits of market price as a proxy. Such a discussion would depart from what seems to be the current orthodoxy within tax law, which focuses almost exclusively on the virtues of the market price as an objective, value neutral measurement of economic well-being.


The above analysis explains why realization should play a normative role in modern income tax regimes. This role intimately relates to the way liberal egalitarian societies should promote notions of distributive justice.

Rather than summarizing our points, we wish to highlight the article’s main contribution to legal literature discussing tax-distributive schemes. First, the dominant role of the realization requirement in the income tax framework indicates that administrative difficulties will not stifle a normative-distributive analysis of its function. The income tax regime is a vital instrument for achieving redistribution in liberal democracies, so academics and policymakers should see a concrete distributive analysis of its attributes as a necessity.

Second, by emphasizing the importance of the difference between market and subjective valuation with respect to certain personal assets, this article encourages a new paradigmatic thinking about realization. The enormous social costs associated with realization should prompt serious consideration of how to advance a progressive hybrid income tax regime. Such a regime would not be flawless but would attain the preferred level of tax-redistribution in a much more efficient and equitable way.

Finally, this article calls for a re-examination of the role that market price plays in contemporary tax theory. This is not to suggest that market price and its related efficiency implications offer no value but simply that reality is far more nuanced. Market price is just one way to describe individuals’ economic well-being for distributive purposes, but there are other objective and subjective ways. Good policymaking requires trying, to the extent possible, to make the best account of them all.


Appendix: Asset Distribution tables by income level 2001: Demonstrating the proportional holdings of personal, investment and business asset classes by different groups of taxpayers.


Distribution of all Assets: assets and liabilities as a share of total assets by percentile groups of distribution of wealth (%)[215]
all 0-50 50-90 90-95 95-99 99-100
#1 Total financial assets 42.2 19.1 35.9 50.7 47.9 45.1
#2 Total of non-financial assets(home equity+ “other assets” + investment real estate) 57.8 80.9 64.1 49.3 52.1 54.9
#3 Investment in non-residential real estate 4.2 1.6 4.2 4.7 6.1 4.6
#4 Investment in residential real estate 4.7 0.5 2.3 3.7 7.1 6.5
#5 Investment in the equity of a (closely held) business 16.9 1.1 5.6 9.5 18 33.5
Distribution by category
#6 Total of investment assets (1+3+4) 51.1 21.2 42.4 59.1 61.1 56.2
#7 Total of non-business personal assets (2-3-4-5) 32 77.7 52 31.4 20.9 10.3
#8 Total of business assets (5) 16.9 1.1 5.6 9.5 18 33.5
Total assets (6+7+8) 100 100 100 100 100 100


[1] Michael J. Graetz & Deborah H. Schenk, Federal Income Taxation: Principles and Policies 154 (6th ed. 2009); Mary Louise Fellows, A Comprehensive Attack on Tax Deferral, 88 Mich. L. Rev. 722, 723-29 (1990); David M. Hasen, A Realization-Based Approach to the Taxation of Financial Instruments, 57 Tax L. Rev. 397, 400 (2004); David M. Schizer, Realization as Subsidy, 73 N.Y.U. L. Rev. 1549, 1551 (1998); Daniel N. Shaviro, An Efficiency Analysis of Realization and Recognition Rules under the Federal Income Tax, 48 Tax L. Rev. 1, 12 (1992); David A. Weisbach, A Partial Mark-to-Market Tax System, 53 Tax L. Rev. 95, 95 (1999).

[2] Schizer, supra note 1, at 1551.

[3] See William D. Andrews, The Achilles’ Heel of the Comprehensive Income Tax, in New Directions in Federal Tax Policy for the 1980s, at 278, 280 (Charles E. Walker & Mark A. Bloomfield eds., 1983) (calling the realization requirement the Achilles’ Heel of income taxation).

[4] See infra Part I.A.

[5] See infra Part III.B (arguing that her acceptance of the market value would be a signal that the market price is a good indicator for measuring her economic well-being).

[6] Chris W. Sanchirico, A Critical Look at the Economic Argument for Taxing Only Labor Income, 63 Tax L. Rev. 867 (2010). However, some scholars claim that the inherent regressivity of the consumption tax base could be balanced, and perhaps even reversed, by a much more progressive rate structure. See Edward J. McCaffery & James R. Hines, The Last Best Hope for Progressivity in Tax, 83 S. Cal. L. Rev. 1031 (2010).

[7] See Joseph Bankman & David Weisbach, Consumption Taxation Is Still Superior to Income Taxation, 60 Stan. L. Rev. 789, 789–91 (2007); Daniel N. Shaviro, Replacing the Income Tax with a Progressive Consumption Tax, 103 Tax Notes 91 (2004). See also William D. Andrews, A Consumption-Type or Cash Flow Personal Income Tax, 87 Harv. L. Rev. 1113, 1115–16 (1974).

[8] See Shaviro, supra note 1, at 66; Edward A. Zelinsky, For Realization: Income Taxation, Sectoral Accretionism, and the Virtue of Attainable Virtues, 19 Cardozo L. Rev. 861, 862 (1997).

[9] See Noël B. Cunningham & Deborah H. Schenk, Taxation without Realization: A “Revolutionary” Approach to Ownership, 47 Tax L. Rev. 725, 799–800 (1992); Daniel Halperin, Saving the Income Tax: An Agenda for Research, 24 Ohio N.U. L. Rev. 493, 501 (1998); David J. Shakow, Taxation without Realization: A Proposal for Accrual Taxation, 134 U. Pa. L. Rev. 1111, 1114-15 (1986); Weisbach, supra note 1, at 99.

[10] See Cunningham & Schenk, supra note 9, at 728; Edward D. Kleinbard & Thomas L. Evans, The Role of Mark-to-Market Accounting in a Realization Based Tax System, 75 Taxes 788, 789 (1997) (identifying realization as the root of many tax evils). See also Joseph Bankman, What Can We Say About a Wealth Tax?, 53 Tax L. Rev. 477, 477 (2000); Fred B. Brown, “Complete” Accrual Taxation, 33 San Diego L. Rev. 1559, 1559 (1996); Deborah H. Schenk, Taxation of Equity Derivatives: A Partial Integration Proposal, 50 Tax L. Rev. 571, 631–32 (1995); Shakow, supra note 9, at 111–13.

[11] For example, the realization requirement is often cited as the main reason for capital gains. See Terrence R. Chorvat, Ambiguity and Income Taxation, 23 Cardozo L. Rev. 617, 647 (2002); Edward J. McCaffery, A New Understanding of Tax, 103 Mich. L. Rev. 807, 896 (2005) (arguing that under an income-with-realization tax, some preference for capital gains is needed). A recent realization-related debate concerns the profit interest compensation of hedge fund managers. The main issue was whether the managers received something of value before cashing their profits. See Victor Fleischer, Two and Twenty: Taxing Partnership Profits in Private Equity Funds, 83 N.Y.U. L. Rev. 1, 5 (2008).

[12] See Halperin, supra note 9, at 503; Deborah H. Schenk, A Positive Account of the Realization Rule, 57 Tax L. Rev. 355, 364-65 (2004); Shakow, supra note 9, at 1144; Shaviro, supra note 1, at 7; Zelinsky, supra note 8, at 876.

[13] See Schizer, supra note 1, at 1552–53.

[14] See, e.g., Halperin, supra note 9, at 499; Schenk, supra note 12, at 355–56; Schenk, supra note 10, at 629; Shakow, supra note 9, at 1114; Weisbach, supra note 1, at 95.

[15] See Ilan Benshalom & Kendra Stead, Values and (Market) Valuations: A Critique of the Endowment Tax Consensus, 104 Nw. U. L. Rev. 1511 (2011).

[16] See supra note 1. For a comprehensive analysis of this point, see Part I.A.

[17] This is true even if the market value of family albums (and other personal assets) owned by high net worth individuals may be greater than the value of albums owned by most taxpayers. See infra Part III.C–D.

[18] See infra Appendix and notes 46, 114.

[19] In other words, this suggests that market price is a tool—a proxy—for getting at individuals’ economic well-being. However, as a proxy it should only be used when there are good reasons to believe it correlates well with well-being.

[20] See Helvering v. Horst, 311 U.S. 112, 115 (1940); Cottage Sav. Ass’n v. Comm’r, 499 U.S. 554, 559 (1991). See also I.R.C. § 1001(a) (2006).

[21] Eisner v. Macomber, 252 U.S. 189 (1920).

[22] See Marvin A. Chirelstein, Federal Income Taxation 73 (11th ed. 2009); Terrence R. Chorvat, Perception and Income: The Behavioral Economics of the Realization Doctrine, 36 Conn. L. Rev. 75, 82 (2003) (providing a short survey of the erosion of the realization requirement’s constitutional protection); Cunningham & Schenk, supra note 9, at 742.

[23] This includes the tax arrangements that governed the taxation of pass-through entities in the 1950s (Subchapters K and S in the Internal Revenue Code) and the rules that govern the taxation of certain traded commodity futures and financial institutions. Chorvat, supra note 22, at 83-86 (describing these examples).

[24] See Andrews, supra note 3, at 280.

[25] See Bankman, supra note 10, at 479; Cynthia Blum, New Role for the Treasury: Charging Interest on Tax Deferral Loans, 25 Harv. J. on Legis. 1, 93–94 (1988); Weisbach, supra note 1, at 97.

[26] There may be some question over how sensible it is to collect taxes and then turn them back when losses appear if there is significant value fluctuation and a progressive tax rate is employed. See Jeffrey B. Liebman, Should Taxes Be Based on Lifetime Income? Vickrey Taxation Revisited (December 2003) (unpublished manuscript), available at

[27] See McCaffery, supra note 11, at 888.

[28] This concept calculates taxpayers’ income in a given period by measuring their overall consumption and net increase in the fair market value of their assets. Haig-Simons taxation focuses on asset appreciation and depreciation, not on whether the assets were sold. It would therefore not include a realization requirement. See Henry C. Simons, Personal Income Taxation: the Definition of Income as a Problem of Fiscal Policy 50 (1938); Robert M. Haig, The Concept of Income—Economics and Legal Aspects, in The Federal Income Tax 1, 7 (Robert M. Haig ed., 1921).

[29] There is a wide consensus that an ideal income tax would adopt the Haig-Simons concept of the income tax base. See, e.g., Graetz & Schenk, supra note 1, at 97; Fellows, supra note 1, at 723–24; Stephen B. Land, Defeating Deferral: A Proposal for Retrospective Taxation, 52 Tax L. Rev. 45, 48 (1996); Schenk, supra note 10, at 631–32; Michael J. Stepek, The Tax Reform Act of 1986: Simplification and the Future Viability of Accrual Taxation, 62 Notre Dame L. Rev. 779, 785–86 (1987); Weisbach, supra note 1, at 95. This concept calculates taxpayers’ income in a given period by measuring their overall consumption and net increase in the fair market value of their assets.

[30] See Shaviro, supra note 1, at 1–2.

[31] See, e.g., David M. Schizer, Balance in the Taxation of Derivative Securities: An Agenda for Reform, 104 Colum. L. Rev. 1886, 1893 (2004) (describing one method by which wealthy taxpayers can reduce their tax liability that is not available to those with less planning advice and fewer assets).

[32] Unless there are strong non-tax considerations involved, taxpayers will exercise these tax options in a way that increases their overall profit. Schizer, supra note 1, at 1555–63 (adding that realization allows taxpayers to blunt the penalty of inflation in a tax regime that assigns tax liabilities according to nominal amounts). See, e.g., Myron S. Scholes et al., Taxes and Business Strategy: A Planning Approach 185–89 (4th ed. 2009) (discussing the ability of businesses to choose between two inventory accounting methods in order to minimize their tax liability without affecting their actual business practices).

[33] David A. Weisbach, Line Drawing, Doctrine, and Efficiency in the Tax Law, 84 Cornell L. Rev. 1627, 1633–45 (1999).

[34] Scholes et al., supra note 32, at 70–73, 107–08; Fellows, supra note 1, at 727; Weisbach, supra note 1, at 100.

[35] See Scholes et al., supra note 32, at 5.

[36] Shaviro, supra note 1, at 4.

[37] Land, supra note 29, at 51.

[38] See infra notes 170–172 and accompanying text.

[39] See Ilan Benshalom, The Dual Subsidy Theory of Charitable Deductions, 84 Ind. L.J. 1047, 1064–70 (2009) (discussing what comprises an optimal allocation of public goods).

[40] Stepek, supra note 29, at 779.

[41] Fellows, supra note 1, at 728; Jeffrey Kwall, The Uncertain Case Against the Double Taxation of Corporate Income, 68 N.C. L. Rev. 613, 628–630 (1990); Land, supra note 29, at 54; Schenk, supra note 12, at 369 (“If C corporations and pass-through entities were valued annually, almost all of subchapters C, S, and K could be repealed.”).

[42] See supra note 36 and accompanying text.

[43] The mobility and fungibility of those assets, along with financers’ unlimited ability to infinitely slice and dice them, lead to fears that tax planners can structure cash flow positions without realizing income. Graetz & Schenk, supra note 1, at 623–26; Ilan Benshalom, How to Live with a Tax Code with Which You Disagree: Doctrine, Optimal Tax, Common Sense, and the Debt Equity Distinction, 88 N.C. L. Rev. 1217, 1219–20 (2010).

[44] The passive loss restrictions and various constructive ownership and sale requirements with respect to financial instruments are additional examples of the remarkably complex rules that aim to control selective loss realization. Graetz & Schenk, supra note 1, at 391–93.

[45] Noël B. Cunningham, The Taxation of Capital Income and the Choice of Tax Base, 52 Tax L. Rev. 17, 41 (1996).

[46] See G. William Domhoff, Wealth, Income, and Power, Who Rules America? (last updated Nov. 2011),, for distributional tables showing that, in 2004, the top 1% of U.S. citizens in terms of net worth controlled 34.3% of the country’s overall wealth; George R. Zodrow, Economic Analyses of Capital Gains Taxation: Realizations, Revenues, Efficiency and Equity, 48 Tax L. Rev. 419, 492–93 (1993).

[47] Chorvat, supra note 22, at 91. See generally David Kamin, What Is a Progressive Tax Change?: Unmasking Hidden Values in Distributional Debates, 83 N.Y.U. L. Rev. 241 (2008) (discussing different criteria for what may be considered a progressive change in the tax system).

[48] Weisbach, supra note 1, at 131; David A. Weisbach, Ten Truths About Tax Shelters, 55 Tax L. Rev. 215, 222–26 (2002).

[49] See Chorvat, supra note 22, at 91–92; Schenk, supra note 12, at 360–70 (exploring these justifications and concluding that neither completely explains the requirement); Shakow, supra note 9, at 1118; Shaviro, supra note 1, at 5, 12–13; Weisbach, supra note 1, at 95–96; Zelinsky, supra note 8, at 879.

[50] Schenk, supra note 10, at 630.

[51] See, e.g., Schenk, supra note 12, at 360–65; Shakow, supra note 9, at 1167–76; Weisbach, supra note 1, at 96.

[52] Schenk, supra note 12, at 362.

[53] Shakow, supra note 9, at 1167–76. But see Zelinsky, supra note 8, at 892 (noting that the Internal Revenue Service may find it very difficult to evaluate taxpayers’ liquidity).

[54] As mentioned earlier, the deferred payment grants taxpayers the benefit of enjoying the time value of money. See supra notes 45–47 and accompanying text. The economic value of this benefit to the taxpayer could easily be computed through basic finance theory principles. While the tax assessment could be made on an annual basis, the taxes could be paid only when the asset was sold. The tax liability would be calculated with reference to the nominal gain and the economic value of deferring the tax payment. Shakow, supra note 9, at 1176.

[55] See Schenk, supra note 10, at 630.

[56] See Brown, supra note 10, at 1587 (discussing the difficulties of determining the value of human capital assets). See also Shakow, supra note 9, at 1157–58 (suggesting that intangibles should not fall within the scope of his proposed mark-to-market regime).

[57] See Zelinsky, supra note 8, at 887–88.

[58] Schizer, supra note 1, at 1594–95. See also Zelinsky, supra note 8, at 881–82.

[59] See David Schmudde, Responding To the Subprime Mess: The New Regulatory Landscape, 14 Fordham J. Corp. & Fin. L. 709, 711 (2009) (describing the rise of the mortgage-backed securities market, which was once considered illiquid). See also Cass R. Sunstein, Group Judgments: Statistical Means, Deliberation, and Information Markets, 80 N.Y.U. L. Rev. 962, 971–74 (2005) (describing valuation strategies in information markets).

[60] See Michael J. Graetz & Itai Grinberg, Taxing International Portfolio Income, 56 Tax L. Rev. 537, 542–45 (2003). See also Katherine Pratt, The Debt-Equity Distinction in a Second-Best World, 53 Vand. L. Rev. 1055, 1057 (2000) (describing the development of two alternative theories of the firm as a result of the dispersion between public ownership of debt and stock).

[61] See Graetz & Grinberg, supra note 60, at 547–54.

[62] For example, under the Securities Exchange Act (1934), the Securities and Exchange Commission requires all public corporations to report financial statements according to the general accounting principles. 15 U.S.C. § 78(m) (2006). Additionally, Treasury Regulation § 1.861–9T(h) gives tax benefits to U.S. multinationals that allocate their interest deductions—which requires them to undertake annual fair market valuation of all their domestic and foreign assets. Treas. Reg. § 1.861-9T(h) (2009).

[63] See Halperin, supra note 9, at 499.

[64] See Linda Sugin, Theories of Distributive Justice and Limitations on Taxation: What Rawls Demands from Tax Systems, 72 Fordham L. Rev. 1991, 2013–14 (2004).

[65] Many prominent scholars have argued that the realization requirement has no normative basis. See, e.g., Cunningham & Schenk, supra note 9, at 742; Halperin, supra note 9, at 499; Schenk, supra note 10, at 629; Shakow, supra note 9, at 1114.

[66] See supra note 14.

[67] See generally Blum, supra note 25, at 93–94 (assessing the desirability and feasibility of using an interest charge to compensate taxpayers’ deferral); Cunningham & Schenk, supra note 9 (advocating for the application of a presumptive tax on imputed income at the risk-free treasury rate to split ownership interests); Fellows, supra note 1, at 728 (offering a proposal for retrospective accrual taxation based on the assessment of when the asset appreciates in value); Land, supra note 29, at 47 (proposing a form of retrospective taxation that grants the government an equity share in the time-adjusted gains at the time of the sale); David S. Miller, A Progressive System of Mark-to-Market Taxation, 109 Tax Notes 1047 (2005) (arguing that affluent individuals, public companies, and medium-sized private companies would be required to mark-to-market their publicly traded property and derivatives); Weisbach, supra note 1 (examining whether it is feasible to expand mark-to-market taxation to cover most liquid assets while leaving hard-to-value assets such as real estate and small businesses on the realization system).

[68] Benshalom & Stead, supra note 15, at 1557.

[69] See Graetz & Grinberg, supra note 60, at 542–45.

[70] Periodic assessment and reliance on presumptive statistical appraisals should provide a good proxy for changes in taxpayers’ portfolios. Furthermore, exemptions for low value assets (e.g., consumer durables), could also reduce administrative costs without significant revenue implications. See Shakow, supra note 9, at 1121–23. A mark-to-market regime could also be administered through special arrangements for assets for which valuation seemed completely implausible. Extremely difficult to value assets—such as artwork—could be taxed on a realization basis with measures to correct for deferral. Another possibility would be to use an ex post retrospective approach that taxed the value of deferral—meaning that when the assets were sold the tax liability would be calculated based on certain assumptions with respect to when the value accumulated. For example, tax authorities could operate under a rebuttable assumption that assets had straight-line appreciation. Taxpayers who realized profits later would face higher nominal tax liabilities, because they would be taxed on the assumed tax value of money they attained by deferring realization. Alternatively, investments could face presumptive tax on their expected return—investors would have to pay a tax on their expected benefits, and the total tax liability could later be modified according to the actual price at which the asset was realized. For such a proposal see Edward D. Kleinbard, Designing an Income Tax on Capital, in Taxing Capital Income 165 (Henry J. Aaron et al. eds., 2007).

[71] See Shakow, supra note 9, at 1118.

[72] See Weisbach, supra note 1, at 122. For example, expanded use of mark-to-market methods would call into question the need for non-recognition transactions and favorable capital gains rates.

[73] See, e.g., Joseph Bankman & Thomas Griffith, Social Welfare and the Rate Structure: A New Look at Progressive Taxation, 75 Calif. L. Rev. 1905, 1921 (1987).

[74] See Schizer, supra note 1, at 1601 (acknowledging that a repeal of realization is not totally un-administrable).

[75] See Shaviro, supra note 1, at 5. See also Schizer, supra note 1, at 1564–65 (referring to all arguments, other than his own, justifying realization).

[76] See infra notes 92–102 and accompanying text.

[77] See Shaviro, supra note 1, at 24–26.

[78] In fact, to the extent the realization requirement provides tax advantages for investment, it may also help correct a bias of the income tax for saving and against consumption. See Shaviro, supra note 1, at 27.

[79] See Chorvat, supra note 22, at 101–12 (surveying the behavioral economics literature, which points out that taxpayers may not take full advantage of the realization requirement’s incentives because of other non-tax considerations and behavioral attributes).

[80] See I.R.C. §§ 475 (2006), 1256 (West Supp. 2010) (enacted in the 1990s and imposing mark-to-market taxation on segments of the financial sector).

[81] These rules allow taxpayers to diversify their risks and cash out their investments without selling their assets and realizing the associated gains. See generally Daniel Shaviro, Risk-Based Rules and the Taxation of Capital Income, 50 Tax L. Rev. 643 (1995) (explaining how financial derivatives allow taxpayers to attain certain tax arbitrages).

[82] Zelinsky, supra note 8, at 939.

[83] These new problems would include computation complexity (in the case of retrospective taxation); valuation cost for non-traded assets; and, most importantly, vexing line-drawing problems between assets taxed only upon realization and those taxed only upon appreciation.

[84] Zelinsky, supra note 8, at 863. See also infra note 121.

[85] Informed taxpayers would seek to be in the realization category when they had assets with unrealized profits (so that they could defer the tax payment) and in the mark-to-market category when they had depreciated assets (so that they would be able to deduct their losses without waiting for realization). The difference between realization assets and mark-to-market assets would result in significant inefficiencies because it would create huge pre-investment distortions in the allocation of resources. Zelinsky, supra note 8, at 915.

[86] Zelinsky, supra note 8, at 904–05.

[87] Chorvat, supra note 22, at 893.

[88] Id. at 77.

[89] Id. at 77; Schenk, supra note 12, at 355–56.

[90] Zelinsky, supra note 8, at 903.

[91] For example, it is not clear to us that public perception would favor a situation in which affluent individuals reduced their tax liabilities by realizing losses while borrowing against their appreciated assets.

[92] Schizer, supra note 1, at 1554 (assuming the political will to use the tax system to tax subsidize private investment).

93 Schizer, supra note 1, at 1552.

[94] Id. at 1580–82.

[95] Id. at 1601–02, 1606–08.

[96] Id. at 1592–95.

[97] Schizer claims that the criticism against realization could be applied against any tax subsidy for saving. Id. at 1574, 1617, 1624.

[98] Id. at 1574. If the value of the tax benefit were not discounted, the government would have to provide less of it to encourage the desired change in investment behavior.

[99] Sections 1(a) and 1(h) of the Internal Revenue Code prescribe different individual tax rates on income and on capital gains while I.R.C. § 11 lays a uniform tax rate on all corporate earnings capital.

[100] Charlotte Crane, Honoring Expectations About Taxes: Are Roth IRAs Different?, Social Science Research Network, at 2–4 (Nov. 12, 2009), available at (explaining why the pension benefits are typically politically stable).

[101] I.R.C. §§ 121, 163(h) (West Supp. 2010).

[102] Lawrence Zelenak, The Federal Retail Sales Tax That Wasn’t: An Actual History and an Alternate History, Social Science Research Network 82, 88 (Oct. 12, 2009), (observing that both the home mortgage and the pension saving tax benefits have always been part of the income tax regime in the United States).

[103] Terrence R. Chorvat, Apologia for the Double Taxation of Corporate Income, 38 Wake Forest L. Rev. 239, 242, 283 (2003) (explaining that by allowing deduction of losses the income tax functions as a type of insurance); Deborah H. Schenk, Saving the Income Tax with a Wealth Tax, 53 Tax L. Rev. 423, 430–35 (2000) (arguing that loss limitation rules are more likely to have a regressive impact because taxpayers with less diversified portfolios and without tax advice will tend to suffer capital losses without being able to offset them); David A. Weisbach, The (Non)Taxation of Risk, 58 Tax L. Rev. 1 (2004); Lawrence Zelenak, The Sometimes-Taxation of the Returns to Risk-Bearing under a Progressive Income Tax, 59 SMU L. Rev. 879, 891–95 (2006).

[104] Land, supra note 29, and accompanying text.

[105] See Ilan Benshalom, supra note 39 (discussing what comprises an optimal allocation of public goods); Walter Blum & Harry Kalven, The Uneasy Case for Progressive Taxation, 19 U. Chi. L. Rev. 417, 427–31, 437–39 (1952);Land, supra note 29, at 51; Shaviro, supra note 1, at 4.

[106] Leandra Lederman, The Entrepreneurship Effect: An Accidental Externality in the Federal Income Tax, 65 Ohio St. L.J. 1401, 1435–44 (2004).

[107] David F. Bradford, Fixing Realization Accounting: Symmetry, Consistency and Correctness in the Taxation of Financial Instruments, 50 Tax L. Rev. 731, 769 (1995); David M. Schizer, Frictions as a Constraint on Tax Planning, 101 Colum. L. Rev. 1312, 1314–15 (2001).

[108] Andrews, supra note 7, at 1115–16. See generally Bankman & Weisbach, supra note 7, at 789-91 (supporting Dan Shaviro’s arguments in favor of the consumption tax).

[109] Advocates of the consumption tax recognize that exempting the yields for saving and investment would have a somewhat regressive impact but argue that all distributional and revenue effects of the exemption could be corrected through higher marginal tax rates. Bankman & Weisbach, supra note 7, at 791–94. The notion that a consumption tax system could be completely revenue and distributionally neutral has been convincingly challenged. Daniel Halperin, Concluding Comment, in Taxing Capital Income, supra note 70, at 309, 310–11; Sanchirico, supra note 6.

[110] See analysis supra Part I.A.

[111] See infra note 162.

[112] See infra notes 156–163 and accompanying text.

[113] See infra notes 167–170 and accompanying text.

[114] Arthur B. Kennickell, A Rolling Tide: Changes in the Distribution of Wealth in the US, 1989–2001, in Int’l Perspectives on Household Wealth 19, 53–54 (Edward N. Wolff ed., 2006) (showing the distribution of assets by personal wealth in 2001: while personal assets consisted of 78% of total assets held by households in the bottom half of the wealth distribution, they comprised only about 10% of the total wealth of the top 1% of households).

[115] Daniel N. Shaviro, Inequality, Wealth, and Endowment, 53 Tax L. Rev. 397, 398–403 (2000).

[116] See generally Ronald Dworkin, What is Equality? Part 2: Equality of Resources, 10 Phil. & Pub. Aff. 283 (1981) (providing leading article on equality of resources).

[117] See John Rawls, A Theory of Justice 54 (1999) (arguing that justice requires equal distribution of primary goods such as rights, liberties, income, and wealth).

[118] For the leading texts on capabilities, see generally Martha C. Nussbaum, Frontiers of Justice: Disability, Nationality, Species Membership (2006); Amartya Sen, Development as Freedom (1999).

[119] Welfarism evaluates social choices solely according to how they affect the well-being of society’s members. Matthew D. Adler & Chris William Sanchirico, Inequality and Uncertainty: Theory and Legal Applications, 155 U. Pa. L. Rev. 279, 282 (2006). In contrast to utilitarianism, welfarism may take into account inequality of well-being across individuals. Id. at 296–97. Like utilitarianism, it focuses on individual utility/welfare as a measurement of well-being and does not take into consideration concepts of “rights” or “fairness.” Id. at 282.

[120] Adler & Sanchirico, supra note 119, at 282–83; Louis Kaplow, The Income Tax Versus the Consumption Tax and the Tax Treatment of Human Capital, 51 Tax L. Rev. 35, 39 (1995); Shaviro, supra note 115, at 398–403; David A Weisbach, A Welfarist Approach to Disabilities, Social Science Research Network 9, 43–44 (Aug. 23, 2007), Nevertheless, it is important to stress that this article’s analysis is not limited to the redistribution of welfare and could be extended to other objective egalitarian benchmarks.

[121] Cunningham & Schenk, supra note 9, at 801–02; Fellows, supra note 1, at 781; Halperin, supra note 9, at 503; Schenk, supra note 12, at 364; Shakow, supra note 9, at 1144, 1153–54, 1158–60; Shaviro, supra note 1, at 7; Zelinsky, supra note 8, at 876, 889.

[122] Louis Kaplow, Human Capital Under an Ideal Income Tax, 80 Va. L. Rev. 1477 (1994) [hereinafter Kaplow, Ideal Income Tax]. See also Louis Kaplow, On the Divergence Between “Ideal” And Conventional Income-Tax Treatment of Human Capital, 86 Am. Econ. Ass’n Papers & Proceedings 347 (1996).

[123] Kaplow, Ideal Income Tax, supra note 122, at 1512–14 (arguing that this is a consumption tax treatment). For a different view, see Benshalom & Stead, supra note 15, at 1542–49.

[124] Kaplow, supra note 120, at 36–37 (clarifying what his objectives were after claiming that they were misunderstood).

[125] Lawrence Zelenak, The Reification of Metaphor: Income Taxes, Consumption Taxes and Human Capital, 51 Tax L. Rev. 1, 5 (1995); Kaplow, supra note 120, at 38–39 (firmly disclaiming Zelenak’s view).

[126] Benshalom & Stead, supra note 15, at 1527–29.

[127] Kaplow, Ideal Income Tax, supra note 122, at 1504.

[128] It is important to note that the revealed preference of owners is that they value all of their assets (not just their personal assets) as worth more than the market price—since otherwise they would sell them. For this reason we define personal assets as those assets were there is a high probability that the subjective value is significantly higher than the market value.

[129] Some may find this argument similar to a line of property law literature developed by Margaret Radin, which claims that personal assets constitute part of individual personhood. See Margaret J. Radin, Property and Personhood, 34 Stan. L. Rev. 957, 958 (1982). However, we think that this article’s argument is different and easier to accept. Rather than arguing that personal assets are inherently different, we merely observe that people attach different values to them so that the trading costs (or the costs of detachment) with respect to these assets are often very high. While there seems to be strong recognition that these costs exist, there is some literature suggesting that they are often given excessive weight. See Stephanie M. Stern, Residential Protectionism and the Legal Mythology of Home, 107 Mich. L. Rev. 1093, 1110–24 (2009) (making such an argument with respect to residential homes).

[130] The tax code employs a somewhat similar definition of which relatives count as related parties and accordingly employs special rules for transactions between them. See I.R.C. § 267 (West Supp. 2010).

[131] See Kennickel, supra note 114, at 53–55. See also infra Appendix.

[132] See infra notes 185–199 and accompanying text.

[133] See infra notes 205–206.

[134] See infra note 142 and accompanying text.

[135] Since for many economically oriented readers the term “priceless” would seem naïve or hypocritical, one can assume for the purposes of this article that priceless equals “anything around 5 million dollars.”

[136] For example, one can assume that if the professors did sell their kidneys they would do so out of dire need—so that any market return on this sale would likely fall short of being a reliable proxy for their economic well-being. This claim is valid, and as we discuss in the next subpart, there may be times in which certain assets should be exempt altogether rather than taxed upon realization. However, such exceptions leave the main premise unshaken.

[137] See Ilan Benshalom, The New Poor at Our Gates: Global Justice Implications for International Trade and Tax Law, 85 N.Y.U. L. Rev. 1, 34–36 (2010) (providing a brief review of the distributive justice literature).

[138] See supra notes 116–118 and accompanying text.

[139] One can say that, even in objective terms, the market does not always adequately reflect the value of certain assets—such as shelter and food security, or mobility. One can further accept that tax authorities have difficulties observing whether individuals attach these values to assets and, more importantly determining the relative “price” of these values. See generally Sen, supra note 118 (providing a set of non-money-equivalent goods that are required for individuals to establish worthwhile lives).

[140] The beyond pricing argument is typically identified with the non-commodification critique. The core of the argument presented in the non-commodification literature is that for some things (such as human organs or sex) voluntary market exchanges are inappropriate because they are dehumanizing, are never really voluntary, and/or have significant negative externalities. See Tsilly Dagan, Itemizing Personhood, Social Science Research Network 7-9 (Apr. 21, 2009) (summarizing this literature). See also Margaret J. Radin, Contested Commodities (1996) (providing one of the key texts on this topic). We believe that our argument has little to do with this literature, the core of which argues that certain things should not be on the market. This article emphasizes the difficulties of making interpersonal comparisons with regard to some emotionally non-fungible assets, many of which could be put on the market without incident or outcry.

[141] See Radin, supra note 129, at 959–60 (providing a somewhat similar argument that personal property is not replaceable).

[142] Experimental behavioral economics suggests that people do not attach special value for assets held primarily for exchange purposes. See Daniel Kahneman, et al., Experimental Tests of the Endowment Effect and the Coase Theorem, 98 The J. of Political Eco. 1325, 1328 (1990).

[143] See infra notes 187–191.

[144] This comment is relevant only to well-being and not to more objective “currencies” of justice.

[145] We wish to thank Jacob Nussim for bringing this forceful argument to our attention.

[146] Most notably, the professor with the more valuable assets can offer them as collateral to creditors and borrow at a lower interest rate or rely on them in cases of emergency.

[147] See Kennickel, supra note 114, at 53–55. See also infra Appendix.

[148] A more formal explanation would stress the following. Income (I) could be defined as the net consumption (C) and savings/investment (S) in a given period of time. Therefore, if I=ΔC+ΔS, an income tax regime that provided a tax benefit for private consumption would shift the burden to savings and investment, and an income tax that provided a tax benefit to savings and investment would shift the tax burden to consumed income.

[149] See Michael J. Graetz, 100 Million Unnecessary Returns: A Fresh Start for the U.S. Tax System, 112 Yale L.J. 261, 270 fig.5 (2002).

[150] The relationship between measurement of economic ability for transfer and tax purposes is not always the same. See Anne L. Alstott, The Earned Income Tax Credit and the Limitations of Tax-Based Welfare Reform, 108 Harv. L. Rev. 533, 535 (1995) (suggesting there may be serious disadvantages to using the same measurement of economic well-being for tax and transfer purposes); Benshalom & Stead, supra note 15, at 1530–31; Lawrence Zelenak, Taxing Endowment, 55 Duke L.J. 1145, 1179–80 (2006).

[151] As mentioned, savings and capital accumulation are characteristics of the upper quintile of society. See supra note 46 and accompanying text.

[152] Prior to World War II, the vast majority of the middle class was exempt from the income tax altogether. See generally Carolyn Jones, Class Tax to a Mass Tax: The Role of Propaganda in the Expansion of the Income Tax During World War II, 37 Buff. L. Rev. 685 (1989) (explaining the dynamic of making the income tax to a mass tax). Lately, there have been some calls to re-establish the income tax as an elite tax while introducing a federal consumption tax to replace the income tax revenues from the middle class. See Graetz, supra note 149, at 281–99.

[153] See Shaviro, supra note 1, at 4; Stanley S. Surrey, Tax Incentives as a Device for Implementing Government Policy: A Comparison with Direct Government Expenditures, 83 Harv. L. Rev. 705, 720–26 (1970).

[154] See supra note 115 and accompanying text.

[155] See Benshalom, supra note137, at 10–18.

[156] Id. at 34–36 (surveying these different layers).

[157] See Robert Nozick, Anarchy, State and Utopia 149–74 (1974) (providing one notable exception).

[158] See Charles O. Galvin & Boris I. Bittker, The Income Tax: How Progressive Should It Be? 38–54 (1969) (arguing that income tax progressivity is necessary to offset certain regressive tendencies in other federal, state, and local taxes); Bankman & Griffith, supra note 73, at 1946-47 (saying that income tax progressivity is typically explained by proportionate sacrifice and the declining marginal utility of money); Thomas D. Griffith, Progressive Taxation and Happiness, 45 B.C. L. Rev. 1363, 1364 (2004) (supporting progressive taxation on account of social science research that tries to estimate the marginal effect of wealth on individual happiness); Kaplow, supra note 120, at 39; Louis Kaplow & Steven Shavell, Should Legal Rules Favor the Poor? Clarifying the Role of Legal Rules and the Income Tax in Redistributing Income, 29 J. Legal Stud. 821 (2000) (arguing that the tax-transfer regime redistributes in a way that results in comparatively less economic distortion than other legal rules); Kamin, supra note 47, at 244 (explaining how different measures of progressivity relate to different theories of distributive justice); Sugin, supra note 64, at 2013–14. But see Kyle Logue & Ronen Avraham, Redistributing Optimally: Of Tax Rules, Legal Rules, and Insurance, 56 Tax L. Rev. 157, 208 (2003) (arguing against Kaplow and Shavell that the distinction between tax rules and legal rules is not always that clear). For an example of regressive tendencies in federal spending, see Julia Lynn Coronado et al., The Progressivity of Social Security, National Bureau of Economic Research (Feb. 2000), (concluding that Social Security taxes are regressive in part because lower income individuals have shorter average life spans than higher income individuals).

[159] See Liam Murphy & Thomas Nagel, The Myth of Ownership 8 (2002).

[160] See Sugin, supra note 64, at 2014

[161] See Kamin, supra note 47, at 253.

[162] See Michael J. Graetz, Taxes That Work: A Simple American Plan, 58 Fla. L. Rev. 1043, 1048–52 (2006).

[163] See supra note 153. See also infra notes 173–175 and accompanying text.

[164] This dilemma is an extension of a more general tradeoff that was identified by Joel Slemrod and Christian Traxler. Joel B. Slemrod & Christian Traxler, Optimal Observability in a Linear Income Tax, Social Science Research Network 8 (Dec. 2010),

[165] There are of course numerous other options if policymakers are willing to rely on factors other than the market price. For a thorough discussion of this topic, see infra note 214 and accompanying text.

[166] See supra Part I.A.

[167] See supra notes 107–110 and accompanying text.

[168] See supra note 153 and accompanying text.

[169] See Graetz & Schenk, supra note 1, at 34–35 (providing an overview the discussion of why a progressive tax rate structure may be justified).

[170] See Walter Blum & Harry Kalven, supra note 105, at 417, 427–31.

[171] Id. at 430–31.

[172] Id. at 437–39 (noting that under a progressive tax regime, gains may be taxed at higher rates than the rates in which losses are given deductions and that high marginal rates significantly reduce the monetary reward of the most productive workers—giving them serious disincentives to work an additional marginal hour of work).

[173] See Bankman & Griffith, supra note 73, at 1944–45 (arguing that the impact of progressive tax rates has been overstated).

[174] Blum & Kalven, supra note 170 105, at 444.

[175] See Bankman & Griffith, supra note 73, at 1909–10 (making a related argument that the progressive tax rates are not sufficient to promote overall tax progressivity if the tax base is not sufficiently progressive).

[176] Blum & Kalven, supra note 170, at 506–11.

[177] We thank Michael Smart for this observation.

[178] For example, promoting risk diversification (e.g., regulated investment vehicles for retirement) and capping the benefit realization so that it only applies to a certain amount of personal assets. See supra note 102. See also infra notes 205–206.

[179] Selling wedding rings and other types of personal assets could signal extreme hardship, so it may be reasonable to exempt them altogether, given that the money received for them is not an adequate benchmark for interpersonal comparisons for well-being. See supra Part III.B.

[180] Benshalom & Stead, supra note 15, at 1542–49.

[181] See generally Bridget J. Crawford, Taxing Surrogacy, Social Science Research Network (Aug. 30, 2009), (making the case for taxing surrogacy). It is true that the decision to work as a surrogate may not be random—but rather an option chosen particularly by non-affluent women. Debora L. Spar, The Baby Business: How Money, Science, and Politics Drive the Commerce of Conception 73 (2006). Furthermore, while personal preferences play a role in every employment decision, certain jobs (e.g., migrant labor) and economic decisions (e.g., selling body organs) are held or made only by very poor people. E.g., Catherine Waldby & Robert Mitchell, Tissue Economies: Blood, Organs, and Cell Lines in Late Capitalism 161 (2006).

[182] See generally Brian Galle, Tax Fairness, 65 Wash. & Lee L. Rev. 1323, 1327 (2008) (arguing that fairness prescribes taxing different sources of income equally).

[183] For discussion of retrospective taxation, see supra note 70. Retrospective taxation would reduce the value of deferral and would de facto impose a higher effective rate on realized personal assets. Therefore, because this article does not discuss what the effective tax rate on personal assets should be, we leave the question of retrospective taxation, its theoretical justifications and political viability, for a future paper.

[184] See generally Lee A. Fennell, Interdependence and Choice in Distributive Justice: The Welfare Conundrum, 1994 Wisc. L. Rev. 235 (1994) (reviewing the difficulties of using welfare distributive programs to reduce poverty and inequality due to labor market disincentives).

[185] See supra note 36 and accompanying text.

[186] See supra note 46 and accompanying text.

[187] Given its limited scope, this article cannot address the imposition of the corporate tax. It is, however, worthwhile to note that a regime that employs a corporate tax on all large corporations would probably be subject to mark-to-market taxation on the entity level. See generally Michael S. Knoll, An Accretion Corporate Income Tax, 49 Stan. L. Rev. 1 (1996) (arguing for computing corporate income tax this way).

[188] One example is the passive loss limitations discussed below. Another set of rules can be found in I.R.C. § 163 (West Supp. 2010), which limits the interest deduction of interest payments made to support investment assets.

[189] See I.R.C. § 469 (2006) (providing broad limitations for losses from “passive” activities). The limitations of I.R.C § 469 and its associated regulations set out mechanical rules for determining whether an investment activity is active or passive, including the number of hours the investor dedicates to the activity. The rules were extremely effective at eliminating tax shelters marketed to individuals. See Graetz & Schenk, supra note 1, at 414–15; McCaffery, supra note 11, at 907.

[190] See Marvin A. Chirelstein & Lawrence A. Zelenak, Tax Shelters and the Search for a Silver Bullet, 105 Colum. L. Rev. 1939, 1951 (2005).

[191] See Lawrence Zelenak, When Good Preferences Go Bad: A Critical Analysis of the Anti-Tax Shelter Provisions of the Tax Reform Act of 1986, 67 Tex. L. Rev. 499, 501 (1989). In the once prominent individual tax shelter industry, high-income earners would purchase non-control shares in businesses solely for the deductions the businesses generated. A dentist in a high tax bracket, for example, might purchase a share in a sham real estate investment using nonrecourse debt solely to take advantage of depreciation deductions generated by the investment. Before the 1986 Tax Reform Act, those deductions could offset (even to the point of wiping out) the taxpayer’s dentistry income. Section 469 of the Internal Revenue Code and the associated regulations virtually eliminated these transactions by creating bright line rules to distinguish between ordinary and passive income and prohibiting passive deductions from being used to offset ordinary gains. Among other requirements, I.R.C. § 469 demands that a taxpayer materially participate in an investment activity in order for proceeds from the activity to be considered ordinary income. In actuality, the provisions establish clear (and arguably arbitrary) tests for what amounts to material participation (based mainly on the amount of time spent in the activity).

[192] See supra notes 51–54 and accompanying text.

[193] See Schenk, supra note 12, at 365-66.

[194] Id. at 369.

[195] See Benshalom & Stead, supra note 15, at 1557.

[196] See Schenk, supra note 12, at 374.

[197] See supra note 70 (discussing how administrative costs of valuation could be reduced by non-annual and statistical valuation).

[198] See supra note 72.

[199] Mark P. Gergen, The Effects of Price Volatility and Strategic Trading Under Realization, Expected Return and Retrospective Taxation, 49 Tax L. Rev. 209, 216 (1994).

[200] Deborah H. Schenk, An Efficiency Approach to Reforming a Realization-Based Tax, 57 Tax L. Rev. 503, 504 (2005) (arguing that not every movement away from realization towards accrual taxation is desirable and that each step should be judged by its revenue gains and efficiency costs).

[201] See supra Part III.B.

[202] Taxpayers have incentives to tax plan against the fractional hybrid regime. Schenk, supra note 200, at 519-24. For example, taxpayers could include much of their debt within the scope of the mark-to-market regime while shifting their appreciated assets outside of it. See Shakow, supra note 9, at 1165–66; Weisbach, supra note 1, at 128.

[203] See Schenk, supra note 200, at 527–28.

[204] See Gergen, supra note 199, at 211 (suggesting that putting only publicly traded assets in the mark-to-market regime would make it administrable but would nevertheless have troubling consequences because it would create a bias for non-publicly traded assets).

[205] This concern is substantial given that much of the wealth of high net worth taxpayers is in the form of closely held business equity. See Kennickel, supra note 114, at 53–55. See also infra Appendix.

[206] In this case, holding a share in a corporation is a close substitute to holding a diversified portfolio.

[207] Schenk, supra note 12, at 396.

[208] Benshalom, supra note 43, at 1259–71; Weisbach, supra note 33, at 1631.

[209] This of course assumes that some attributes of the tax code are politically fixed. See Weisbach, supra note 33, at 1631–32.

[210] See generally Weisbach, supra note 33 (claiming that line drawing problems are inherent to tax law).

[211] See generally Bankman & Weisbach, supra note 7 (arguing that a consumption tax is desirable on the grounds that the income tax cannot meet its distributional objectives). See also McCaffery & Hines, supra note 6.

[212] Needless to say, such a commitment does not exist in the current political arena. Daniel N. Shaviro, Taxes, Spending, and the U.S. Gov’t’s March Toward Bankr. 116–47 (2006).

[213] See, e.g., Alvin Warren, Would a Consumption Tax Be Fairer Than an Income Tax?, 89 Yale L.J. 1081, 1093-94 (1980).

[214] For example, policymakers could say that the value of a taxpayer’s personal assets should be calculated based on the market price of the asset multiplied by a factor that is correlated with the taxpayer’s overall wealth. In this scenario, Bill Gates’s ring would be valued at a higher rate than Joe Six Pack’s—even if they had the same market price.

[215] See Kennickel, supra note 114, at 53–55.