Printer Friendly

Statutory speed bumps: the roles third parties play in tax compliance.

 A. Horizontal Equity and Parallelism
 B. Exclusions and Deductions: Where's the Equivalency?
 C. The Importance of Enforceability to Tax Policy
 A. Clark v. Commissioner: A Tax Refund by Any Other Name
 B. Excludible Personal Receipts
 C. Excludible Employment-Related Receipts
 2. The self-employment comparison
 3. Moving expenses
 A. The Zero-Sum Constraint and the Unselfish Exception
 B. Manufactured Surplus Minimized: The Example of Wages
 C. Identifying Accommodation and Cooperating Parties
 1. Tax-indifferent parties
 2. Familiar parties
 3. Soliciting parties
 4. Parties playing two roles


Numerous laws prohibit and punish behavior the government seeks to prevent, such as running red lights or driving faster than the posted speed limit. However, recent legal and economic scholarship has recognized that the government has an important alternative mechanism it can use to reduce prohibited behavior: "structural" systems. (1) Structural systems either facilitate enforcement, as red light cameras do, (2) or actually help constrain behavior. For example, if the government seeks to reduce speeding in a residential neighborhood, instead of (or in addition to) imposing fines and ticketing speeders, it can construct roads in ways that help reduce speeding, (3) such as making them narrow or winding, or including speed bumps. The structure of the road can thus actually help prevent the behavior the government seeks to reduce.

This insight can apply in an array of areas in addition to traffic laws. For example, why people pay taxes is sometimes described as a puzzle. (4) From an economic perspective, it appears that penalties and enforcement rates are too low to deter cheating with respect to such taxes as the federal income tax in the United States. (5) Yet, the federal government estimates that 84% of federal income taxes due are timely and voluntarily paid. (6) An essential missing piece of this seeming puzzle is that the federal income tax law benefits from structural mechanisms that constrain payment with respect to the major sources of income for many people, including wages and salaries. (7)

The structural mechanisms the federal income tax uses, unlike red light cameras and speed bumps, make use of third parties to the taxpayer/government relationship. As is well known, in a variety of situations, the federal government requires third parties to report to the government, with a copy to the taxpayer, amounts the payor transferred to the taxpayer. (8) This "information reporting," like red light cameras, provides information to the government, and it is information that the taxpayer knows the government is receiving. (9) Moreover, in some situations, the payor, such as an employer, must also withhold taxes from the payment and remit those taxes to the government. Withholding taxes, like speed bumps, constrain compliance with the law. However, unlike speed bumps, withholding taxes are effective largely because they essentially make a third party responsible for paying the taxpayer's taxes. (10)

Information reporting and withholding extend to a variety of types of income in the U.S, and are highly successful at securing compliance. (11) A comparison of the estimated "voluntary compliance" rates under the federal income tax with respect to various types of income suggests just how effective these systems are. (12)
 Amounts subject to withholding (e.g., wages and salaries) have a
 net misreporting percentage of only 1.2 percent. Amounts subject to
 third party information reporting, but not to withholding (e.g.,
 interest and dividend income) have a slightly higher net
 misreporting percentage of 4.5 percent. Amounts subject to partial
 third-party reporting (e.g., capital gains) have a still higher net
 misreporting percentage of 8.6 percent. Amounts not subject to
 withholding or other information reporting (e.g., Schedule C income
 or other income) are the least visible, with a much higher net
 misreporting percentage of 53.9 percent. (13)

Structural systems that engage third parties to help facilitate compliance with the federal income tax are thus highly successful. (14) The use of such mechanisms need not necessarily be limited to tax administration, however. Substantive tax law could incorporate structural systems that benefit from the use of third parties. In fact, although it has gone unnoticed until now, federal income tax law already implicitly takes account of the structural benefit arm's length third parties can offer. Specifically, as discussed below, the tax law often fails to extend the favorable tax treatment afforded particular reimbursed expenses or losses to similar but unreimbursed items. (15) This distinction does not reflect different tax treatment of equivalent events, as prior scholarship suggests. (16) Instead, it reflects the enforcement benefits that a reimbursement provides--including the presence of a third party who implicitly has "vouched" for the bona tides of the taxpayer's claim. (17) In fact, the more favorable tax treatment of amounts verified by third parties is not only rational, it is consistent with the important tax policy goal of an administrable tax system. (18)

Although third parties can thus provide a type of "friction" that reduces tax avoidance, (19) they do not do so in all contexts. In fact, transaction counterparties have also been known to participate in abusive tax-reduction strategies in return for a portion of the tax savings generated. (20) The government thus needs to be able to identify the types of situations in which counterparties will tend to act as verifiers of taxpayers' claims, as well as those in which they are more likely to collude in noncompliance. Aspects of the structure of each situation can help the government make this distinction. It then generally can "free ride" in situations in which third parties perform a verification function while scrutinizing more closely other transactions. Moreover, the government can consciously seek to use legislation to create more of the beneficial structures and reduce the number of structures that lend themselves to exploitation.

In elaborating on these distinct but connected points relating to the bridge that third parties provide between substantive income tax laws and enforcement of those laws, this Article proceeds in three major parts. Part I develops the insight of this Article that the substantive role third parties play in federal income tax law helps explain numerous instances in which an exclusion of a reimbursed or compensated expense or loss is not matched by a deduction in full of a comparable but uncompensated expense or loss. It first discusses the concept of "parallelism," which Professor Jeffrey Kahn describes in a recent article as a concept related to but narrower than horizontal equity, the concept that similarly situated taxpayers should be similarly taxed. (21) This Part then demonstrates that the exclusion of a reimbursement is not economically equivalent to the deduction of an unreimbursed amount. Finally, it connects the economic distinction between reimbursed and unreimbursed amounts to the important tax policy principle of administrability of the tax system.

In Part II, the Article explores the use of third parties as a compliance tool, as currently reflected in substantive federal income tax law. It discusses several situations in which more favorable tax treatment is accorded to a transaction that involves an arm's length third party than to an otherwise similar transaction that does not involve a third party, and argues that the distinct treatment is justified by the policy considerations discussed in Part I.

Because third parties are not a panacea for tax compliance, Part III of the Article explores how the government can distinguish contexts in which third parties tend to foster compliance from settings in which they may tend to undermine it. This Part first discusses the underlying economic distinction between contexts in which third parties, in acting out of their own self-interest, will verify the taxpayer's claim, and situations in which third parties' self-interest will instead incline them to collude with the taxpayer to exploit the tax system. Next, it examines these issues in the context of employers' payments of wages to employees. Finally, the Article identifies four types of transaction counterparties for whom the government should be watchful because transactions in which they are involved provide opportunities for collaboration in abusive tax-minimization strategies.


A. Horizontal Equity and Parallelism

Equity in a tax system is an important tax policy consideration. (22) "[H]orizontal equity," which "demands that similarly situated individuals face similar tax burdens[,] ... is universally accepted as one of the more significant criteria of a 'good tax.'" (23) Horizontal equity implicitly refers to taxpayers' overall tax burdens. (24) In a recent article, Professor Jeffrey Kahn discussed a related, narrower principle, which he terms "parallelism"--the notion that "the same or equivalent receipts, expenditures or losses should be treated the same by the tax law." (25)

The focus of Professor Kahn's analysis "is the notion that if the reimbursement of an expenditure or loss is excluded from the recipient's income, the same type of expenditure or loss that is not reimbursed should be fully deductible." (26) His analysis rests on the principle, discussed further below, that an exclusion is mathematically equivalent to an inclusion coupled with a deduction. (27) Professor Kahn argues:
 Since the exclusion and deduction approaches generally are
 identical for tax purposes, one might expect there to be parallel
 treatment of reimbursed and unreimbursed expenditures and losses.
 That is, one might expect a taxpayer who incurs an expenditure or
 loss to be treated the same by the tax law whether the item is
 reimbursed or not. (28)

Yet, as Professor Kahn explains, there are numerous examples in which the federal income tax law allows an exclusion for a reimbursed or compensated expense or loss but disallows or limits any deduction for a similar but unreimbursed, uncompensated item. (29) Professor Kahn discusses seven examples of this phenomenon, arising in the following general contexts: (1) damage to property; (2) personal injury damages; (3) forgone or unavailable tax refunds; (4) employee business expenses; (5) life insurance proceeds; (6) employee meals and lodging; and (7) job interview expenses. (30)

The following frequently discussed example, which involves the tax treatment of victims of personal injuries, is illustrative of the phenomenon in the federal income tax of allowing an exclusion for a compensated loss but no deduction for an uncompensated one. (31) Assume that Ann and Bob are identically situated, and each is identically physically injured by different tortfeasors. Ann was injured by a wealthy tortfeasor and receives a large sum in settlement of a personal injury suit--say $2 million. Bob, by contrast, was injured by a poor, unidentifiable, or judgment-proof tortfeasor and receives nothing.

Under current law, Ann can exclude the damages she received, (32) but Bob cannot claim a deduction comparable to the amount Ann excluded from income. (33) Ann will thus exclude the $2 million from income, while Bob gets no such tax benefit. As a result, assuming that their tax profiles are otherwise the same, Ann and Bob will pay the same amount of tax despite Ann's additional $2 million receipt. (34) This result appears troubling. (35) In fact, not only does it appear that Bob is treated worse than Ann, but Bob seems to be the one more in need of a tax benefit, as he did not receive anything from the tortfeasor. (36)

Professor Kahn's article does not seek to resolve the seeming puzzle of this and the six other examples he discusses of lack of parallelism between exclusions and tax deductions. (37) Instead, his article focuses on the role of parallelism as a policy principle. Thus, he takes a case-by-case approach to each of his examples, arguing that "[e]ach instance of nonparallel treatment of the tax law should be examined separately to determine whether there are competing principles that outweigh the goal for parallel treatment." (38) Accordingly, Professor Kahn's article generally analyzes the rationales supporting each exclusion and those in favor of disallowing or limiting a corresponding deduction, and also considers the role, if any, that parallelism should play.

Professor Kahn's approach has the virtue of examining, in several distinct areas of federal income tax law, whether an existing exclusion is justified; some of them, though well-entrenched, may not be. (39) It also draws valuable attention to a number of areas in which the tax law does not treat reimbursed and unreimbursed amounts alike. Yet, his approach ignores two important issues that allow development of a larger framework that resolves the seemingly puzzling inconsistency that Professor Kahn highlights.

First, as further explained below, although every exclusion from income implicitly contains a deduction in full, such an implicit deduction inherently is paired with an implicit inclusion in income in the same amount; it is the combination of inclusion and deduction that is equivalent to an exclusion. (40) The inclusion is not only critical, it generally does not occur in the absence of a receipt. Second, as a matter of tax policy, the substantive distinction between reimbursed and unreimbursed amounts is actually highly meaningful, although the literature previously has not recognized this fact. Thus, it is not the case, as Professor Kahn's analysis suggests, that because there is a deduction implicit in an exclusion of a reimbursed amount, economic equivalency calls for a deduction of a comparable, unreimbursed amount (41) that should be allowed unless situation-specific policies triumph. (42) These points are discussed in the next two Sections.

B. Exclusions and Deductions: Where's the Equivalency?

The federal income tax is imposed on "taxable income." (43) Taxable income generally means "gross income" minus deductions, except that, for individuals, deductions are separated into categories, with only "above-the-line" deductions--those taken off the top--being fully deductible. (44) Thus, in isolation, an inclusion in gross income increases taxable income and thereby increases tax liability. Conversely, a deduction, if not disallowed, reduces taxable income, and thus generally reduces taxation.

Because taxable income reflects both gross income and deductions, a receipt constituting gross income that is paired with an offsetting deduction (one in the same amount as the receipt, and deductible in full) is equivalent to exclusion of the receipt from gross income and no deduction. (45) For example, assume that Carla, a corporate employee, purchases office supplies for use this year in her job, at a cost of $100, and her employer reimburses her. (46) The supplies are a business expense, so Carla can deduct them. (47) Assuming that Carla's employer has an appropriate reimbursement arrangement, the deduction is above the line, (48) so the full $100 is deductible. (49) If Carla deducts the cost of the supplies, she then takes the $100 receipt into gross income. (50) Her taxable income from the transaction would be as follows:

$100 gross income

<$100> above-the-line deduction


$0 taxable income.

This result is equivalent for federal income tax purposes to allowing Carla to exclude the $100 receipt from gross income but not take a deduction for the expense; she simply would have no gross income from the transaction, no deduction, and thus no taxable income from it. In fact, unlike the Internal Revenue Code (Code), which provides for an inclusion in gross income coupled with an above-the-line deduction, (51) Treasury regulations take this simplified approach. (52)

Note that the equivalency holds only if the deduction is allowed in full. If a portion of the deduction is disallowed or not usable, exclusion is more beneficial than the receipt coupled with the deduction because an exclusion does not give rise to taxable income. (53) An individual's deduction may be unusable, in whole or in part, if it is an "itemized deduction." To use an itemized deduction, an individual must elect to itemize in lieu of taking the standard deduction, so itemized deductions, in the aggregate, must exceed the standard deduction in order for that election to be warranted. (54) In addition, some itemized deductions are "miscellaneous itemized deductions" that only are included as itemized deductions to the extent that, in the aggregate, they exceed two percent of the taxpayer's adjusted gross income. (55) Furthermore, there is an overall limitation on itemized deductions for higher-income individuals. (56)

Where the deduction is allowed in full, however, it fully offsets the receipt, as shown above, and thus is equivalent to an exclusion. Accordingly, an exclusion is equivalent to an inclusion coupled with a full deduction for the same amount. (57) Federal income tax law contains a number of exclusions from income, such as the exclusion for life insurance proceeds received by reason of the death of the insured, (58) personal physical injury damages, (59) and qualified scholarships. (60) The equivalency principle means, for example, that no change in federal income tax would result to a taxpayer who, instead of excluding personal physical injury damages from gross income, as provided by current law, (61) took the receipt into income but took an above-the-line deduction for that amount. The mechanics in this example would be somewhat odd--the taxpayer would be taking a deduction not based on an expense or tax loss--but the economic result would be the same as the exclusion. (62)

By contrast, an exclusion of a personal physical injury damages receipt and the deduction of uncompensated personal physical injury damages do not have the same effect on tax liability. For example, assume that taxable income is taxed at a flat rate of 20%. If Dan has $100,000 of taxable income apart from $70,000 of excludible personal physical injury damages he received, his tax liability is $20,00063 because of the exclusion of the $70,000 in damages. If Dan instead had no such receipt but were allowed to deduct $70,000 worth of uncompensated personal injuries, rather than facing a $20,000 tax liability, the deduction would shelter $70,000 of income, lowering his taxable income to $30,000 and thereby reducing his federal income tax liability to $6,000. (64)

Thus, in the example above involving Ann and Bob, (65) Ann's exclusion of her personal injury damages receipt is economically equivalent to a deduction in full for her loss coupled with inclusion of the damages she received. There is no such equivalency for Bob because, unlike Ann, he received nothing from the tortfeasor. A deduction for Bob would actually offset other income, such as salary, whereas a deduction for Ann would only shelter the related receipt. (66) Exclusions thus implicitly have built-in "basketing" (67) in that reconceiving the exclusion as a deduction would allow the hypothetical deduction only in the presence of the hypothetical gross income that necessarily accompanies it. (68) By contrast, a deduction without an inclusion shelters other, otherwise-taxed amounts--actually lowering tax liability rather than keeping it the same.

This reality raises important compliance issues that would arise were uncompensated personal injuries allowed to be deducted. Under current law, a taxpayer has no tax incentive to fabricate the receipt of a personal injury damages payment because such a receipt, even if it is excludible from income, would not reduce her taxes (it would merely keep them from increasing). Ann, accordingly, has no incentive to lie to the federal government about the existence or magnitude of her injuries or about the amount of damages she received for those injuries. (69) Moreover, the payment to Ann means that the government can have confidence in the magnitude and value of her injuries. The tortfeasor who pays $2 million in damages has no self-interest in disgorging funds, so the bona tides of the transaction generally can be respected. (70) That is, the payment by the tortfeasor eliminates questions about the existence of the injury and the magnitude of the damage. (71)

There is no such structural constraint on Bob. Because a deduction can offset other income, if the value of uncompensated personal injuries were deductible, taxpayers would have an incentive to exaggerate the extent of personal injuries on their tax returns (72)--and even to fabricate them entirely, because deductions actually reduce tax liability. (73) Accordingly, a deduction for uncompensated personal injuries, particularly one not limited by the amount of actual medical expenses, (74) would amount to an invitation to commit tax evasion. Without a requirement of the receipt of funds as a prerequisite to a tax benefit, a taxpayer could fabricate an injury or exaggerate the consequences of an injury in order to reduce tax liability and gamble on the small chance of losing the "audit lottery." (75) Moreover, if such a taxpayer were audited, his claim of uncompensated injury would be hard for the government to counter. (76) He could claim that the tortfeasor injured him in a hit-and-run accident, for example, so that even the identity of the supposed tortfeasor was unknown.

Thus, when Congress enacts an exclusion for a reimbursed item, whether or not the exclusion is justifiable from a tax policy perspective, the cost of the provision to the federal government--aside from issues of misallocation of receipts to the excludible category, which are discussed below (77)--will be limited to the forgone tax revenue from those entitled to benefit from the exclusion. However, if Congress enacts a deduction for a similar but unreimbursed item, the likely costs will include not only the forgone tax revenue from those entitled to the deduction but also the forgone tax revenue from those who falsely claim the deduction but escape detection.

C. The Importance of Enforceability to Tax Policy

When the appropriateness of a tax proposal or provision is evaluated as a policy matter, the concerns usually referenced are efficiency, equity, and "simplicity" or "administrability." (78) The simplicity or administrability of a tax system has two major components: the ease of government enforcement and the ease of taxpayer compliance. Administrability may appear comparatively superficial or more concerned with form than efficiency and equity do, which resonate as matters of substance. A simple or easily administered tax may even seem like a luxury if those values are in tension with the efficiency and/or equity of the tax system. Yet, administrability of a tax is key to its effectiveness; "[t]he best tax policy in the world is worth little if it cannot be implemented effectively." (79) In fact, Milka Casanegra de Jantscher famously stated that "tax administration is tax policy." (80)

Similarly, in 1965, when Judge Joseph Sneed developed a list of seven attributes that have shaped the federal income tax, he included "[p]racticality," which he defined as "a practical and workable tax system." (81) Included in Judge Sneed's concept of practicality is the government's ease of collection of the tax. (82) Although he ranked practicality and equity (83) as the most important of his seven criteria, he noted that practicality generally is given the most weight. (84)

In part, the importance of the administrability or practicality of a tax system reflects the notion that a tax system that appears equitable is not so if it is not enforceable in a manner that reaches equitable results. (85) For example, assume that, in a two-person economy populated by Ellen and Fred, the most equitable tax system designed to raise $100 needed by the government would tax Ellen somewhat more than Fred. Tax A, which would tax Ellen $60 and Fred $40, is the best such tax the government can design. Unfortunately, the structure of the tax is such that Ellen can easily evade it and pay nothing. The government then faces the possibility of collecting only $40 (from Fred), which is insufficient; to meet its revenue needs, it would have to raise tax rates so as to collect $100 from Fred (Tax A'), a very inequitable result in that it actually taxes Fred much more than Ellen (who is not taxed).

Assume that the government could instead impose a different tax (Tax B) that will tax Ellen and Fred in the amount of $50 each, and that this tax is designed to be easily enforced, so the government will collect its full $100. Tax B is somewhat less equitable than Tax A in theory because it taxes Ellen and Fred equally, rather than taxing Ellen somewhat more than Fred. However, in practice, Tax B will be more equitable than Tax A, because, given the constraint that the government needs $100, Tax A will become Tax A' in practice, and Tax A' does not tax Ellen at all. (86)

In addition, enforcement of the tax laws is an important issue because uncollected tax dollars generally must be recouped by the government through higher taxes or reduced spending. It has been estimated that "the tax gap increases the tax burden on every compliant individual taxpayer by $2,000." (87) As discussed above, (88) compliance with the federal income tax is estimated at about 84%, (89) but the gap between what is due and what is paid was estimated at $345 billion for tax year 2001 (90)--an amount "approximately equal to the amount that the federal government pays each year for Medicare or the 2005 federal budget deficit." (91)

The federal income tax often is said to require "voluntary compliance." That is, those liable for the federal income tax are required to calculate and report their tax liabilities. (92) More specifically, taxpayers must file annual federal income tax returns that report income, deductions, credits, and other information. Audit rates are quite low, (93) so tax provisions that are effectively self-enforcing are much more administrable than provisions that are not self-enforcing.

The example of Ann and Bob, the hypothetical personal injury victims, (94) helped illustrate that an exclusion for a reimbursement has two structural limitations that a deduction for an unreimbursed item does not have: (1) it inherently involves a third party willing to transfer funds to the taxpayer, and (2) it does not reduce existing tax liability; it merely keeps tax liability from increasing, (95) which limits the possible benefit of false claims. Accordingly, the disallowance or restriction of deductions for various unreimbursed amounts makes the federal income tax much more administrable than it would be were such deductions allowed in full. (96)


Part I helped show that the tax system can be structured such that a taxpayer inclined to make certain false claims must enlist the cooperation of a third party in order to obtain tax benefits. Where the third party has an economic incentive to vet the taxpayer's claim, as the tortfeasor who compensated Ann did, the government can free ride on that verification. The following examples further illuminate this role that third parties implicitly play in the federal income tax system.

A. Clark v. Commissioner: A Tax Refund by Any Other Name

An interesting example often cited as one involving an inevitable absence of similar tax treatment for similarly situated taxpayers involves tax overpayments. In Clark v. Commissioner, tax counsel prepared a joint return for a married couple. (97) After an audit, new computations disclosed that if the taxpayers had filed separate returns, they would have saved almost $20,000. (98) The return preparer, acknowledging that he had erred, transferred that amount to Mr. Clark, and, following an audit, the Board of Tax Appeals analyzed the federal income tax consequences of that payment. (99) Professor Kahn states, "[W]hichever way the court ruled, the policy would fail in some manner the principle of horizontal equity." (100) That is because a taxpayer who receives a tax refund from the government can exclude it. (101) By contrast, a taxpayer who overpays federal income tax but does not timely claim a refund and receives no reimbursement from a return preparer is not entitled to a deduction for the overpayment, even if the overpayment was due to preparer error. (102)

In Clark, the Board of Tax Appeals allowed the exclusion of the return preparer's payment, and the government acquiesced. (103) Although this case is regarded as troubling because of the competing analogies that would lead to different outcomes, (104) the result makes sense once enforcement considerations are taken into account. In fact, this example is particularly helpful in highlighting the third-party verification role on which the government can free ride because one of the examples to which it is typically compared for parallelism purposes--the tax refund granted by the government--involves opportunity by the government itself for scrutiny of the taxpayer's claim.

If a return preparer pays the taxpayer the amount the taxpayer overpaid due to preparer error (for which a refund can no longer be claimed, or because, as in Clark, the law at the time did not allow married taxpayers to change their filing status from joint to separate after filing a return for the year), the return preparer is in a situation much like the tortfeasor paying the victim. The fact that the preparer is making the payment suggests that there was, in fact, an error on the initial return. (105) The third-party payment replacing the lost overpayment can thus be treated the way a refund would have been--as excludible. As in the case of the payment by the tortfeasor, the taxpayer has no incentive to fabricate receipt of such a payment because the payment, though excludible, would not reduce the taxpayer's tax liability.

By contrast, a deduction for amounts allegedly overpaid with respect to a year in which the statute of limitations on refund claims has expired and for which no preparer is compensating the taxpayer would open the door to false claims that could not be verified without an audit. Professor Jeffrey Kahn argues, in part, that the disallowance of a deduction for overpaid federal income taxes is justifiable because the "loss" from the overpayment was not business or profit-seeking in nature:
 Why should Y be denied a deduction for the overpayment of his
 taxes? One answer is that a deduction is allowed for a loss only if
 there are compelling reasons for it. Losses incurred in a business
 or profit-seeking activity are generally deductible. Personal
 losses are not deductible unless they are the product of a theft or
 casualty. Y's loss is not a business or profit-oriented loss, and
 is not a casualty or theft loss. (106)

The compliance perspective demonstrates that even if such an overpayment error (107) were made by a business--say, by a corporate taxpayer--a deduction should nonetheless not be allowed.

B. Excludible Personal Receipts

The tort victim example, discussed above, (108) provides an example of excludible compensation for purely personal losses unmatched by a deduction for uncompensated injuries. Another example of such a distinction between the treatment of a personal item for exclusion and deduction purposes involves child support. Although child support receipts are excludible from gross income, (109) there is no general deduction for the costs of supporting children. (110) In fact, despite the sympathetic nature of the claim, no deduction is authorized for the taxpayer who fails to receive child support from the other parent, even when the payment of child support is required by court order and the taxpayer expends his or her own funds to support the child. (111) If the taxpayer's expenditures on the children were deductible in this context (for example, as a bad debt loss (112)), that could open the door to false claims consisting of exaggerated expenditures, false allegations of nonpayment of support, and fabricated child support agreements, all of which would require audits for the government to check. By contrast, the excludibility of child support does not itself give rise to an enforcement issue for the same reasons that allowing an exclusion for personal injury damages does not. (113)

More favorable tax treatment also is accorded with respect to medical expenses for which third-party monitoring exists. Health insurance reimbursements are excludible, regardless of the taxpayer's basis in the insurance policy. (114) Similarly, "flexible spending accounts," which are vehicles established by employers and voluntarily funded by employees with pre-tax dollars, allow complete exclusion from income of reimbursement of a participating employee's otherwise unreimbursed medical expenses, without a floor. (115) In addition, non-prescription drugs, which generally are not deductible under the medical expenses provision of Code section 213, (116) are reimbursable from the pre-tax dollars of a flexible spending account. (117)

On the deduction side, under section 213, a taxpayer may deduct as itemized deductions (118) unreimbursed medical expenses only to the extent they exceed a high floor--7.5% of adjusted gross income (AGI) under the regular tax (119) and 10% under the alternative minimum tax. (120) These high floors effectively limit the deduction of unreimbursed medical expenses to extraordinary medical expenses, which, by definition, are large relative to AGI. (121)

Flexible spending accounts and accident or health insurance are alike in that reimbursement requires submission of receipts to an administrator, who has no incentive to disburse funds for nonqualifying items. The bona tides of a medical expense deduction, on the other hand, can only be verified through an audit. The more generous substantive provisions on the reimbursement side thus make sense. (122) In fact, it is interesting that Congress chose to allow at all the deduction of medical expenses that exceed certain (AGI-sensitive) amounts. Floors allow Congress to provide a deduction it deems appropriate, while both effectively eliminating small claims that are not worthwhile to audit and increasing the salience of these deductions because, when claimed, they will necessarily be sizeable relative to AGI.

Personal casualty losses provide another, similar context. Professor Kahn points out that "[i]f a taxpayer receives compensation for damaged property, the taxpayer includes the recovery in income only to the extent that it exceeds the basis of the property" and "it does not matter whether the payor is the person who damaged the property or a third party insurer." (123) Yet, if the taxpayer experiences an unreimbursed casualty loss of a personal-use item, the deduction is limited in several ways. It faces a $100-per-event floor and requires basketing with personal casualty gains. In addition, any net casualty loss faces a 10%-of-AGI floor and is classified as an itemized deduction, (124) and thus subject to reduction. (125) The different treatment in the two contexts implicitly reflects concern about the difficulty in verifying a claim of an unreimbursed casualty loss, in comparison to the absence of compliance concerns reflected in reimbursed amounts, (126) as discussed above. (127) The floors imposed on personal casualty losses reflect much the same concerns that arise with respect to medical expenses (128) and with respect to unreimbursed business expenses, discussed below. (129)

Similarly, the Internal Revenue Service (IRS) has ruled in a number of contexts that reimbursements for expenses made on behalf of another, outside of the context of an employer-employee relationship, are excludible, even though the expense itself would not be deductible (typically because it is considered personal in nature). (130) Such contexts include travel expenses for a state government official to attend a political fundraiser, reimbursed by the political organization sponsoring the event; (131) the expenses of caring for a foster child, reimbursed by a child placement agency; (132) and reimbursement by car pool members for the expenses of the car pool's driver. (133) Again, whether or not each exclusion is justifiable as a policy matter, the exclusion is much less costly for the government than a deduction for a similar but unreimbursed item would be because of the special enforcement risks that deductions present.

C. Excludible Employment-Related Receipts

A number of situations in which one taxpayer may be reimbursed or subsidized, while another, similarly situated taxpayer may not be, involve employees' expenses. For example, the money saved by an employee benefitting from a "qualified employee discount" on purchases of the property or services the employer offers for sale "in the ordinary course of the line of business" for which the employee works is excludible from gross income, (134) while an employee who does not benefit from such a discount cannot deduct a percentage of the costs of such items purchased for personal use. (135)

An example of this type discussed by Professor Kahn involves meals or lodging furnished "for the convenience of the employer." (136) Code section 119 allows the exclusion of meals and lodging furnished on the employer's business premises to an employee, as well as to the employee's spouse and dependents, so long as they are provided for the convenience of the employer, (137) and, "in the case of lodging, the employee is required to accept such lodging on the business premises of his employer as a condition of his employment." (138) Yet, employees may not deduct personal meals and lodging expenses. (139)

In each of these cases, the exclusion, whether or not normatively justifiable, does not raise the specter of numerous false claims because the exclusion does not reduce the employee's taxes on other income. (140) If an argument were made for a deduction, it would likely be based largely on parallelism concerns. A deduction for personal expenses, such as with respect to personal-use items purchased from the employer or personal meals, would offset other income, reducing overall tax liability and thus providing an incentive for fabrication or exaggeration of expenses.

1. Employee business expenses

A standard example of lack of parallelism with respect to employees involves business expenses. An employee whose business expenses are reimbursed by his or her employer under a "reimbursement or other expense allowance arrangement" is allowed to deduct those expenses in full (above the line), (141) and Treasury Regulations convert the income and above-the-line deduction to an exclusion. (142) By contrast, an employee with unreimbursed business expenses is limited to taking a miscellaneous itemized deduction, (143) which generally will reduce the amount that is deductible or even eliminate any deduction for the expenses. (144)

The limitation on the deductibility of employees' unreimbursed business expenses has been criticized. (145) For example, Professor Jeffrey Kahn has argued:
 Another category of deductions that should be reclassified [as
 above-the-line deductions] is unreimbursed employee trade or
 business expenses. These are the only trade or business expenses
 that are classified as miscellaneous itemized deductions, and even
 several specified classes of employee expenses are excluded from
 itemized treatment. Singling out such expenses for limitation is
 unfair. (146)

Yet, the legislative history of the provision imposing the 2% floor explains, in part:
 The Congress concluded that the prior-law treatment of employee
 business expenses, investment expenses, and other miscellaneous
 itemized deductions fostered significant complexity, and that some
 of these expenses have characteristics of voluntary personal
 expenditures.... Moreover, the fact that small amounts typically
 were involved presented significant administrative and enforcement
 problems for the Internal Revenue Service. These problems were
 exacerbated by the fact that taxpayers frequently made errors of
 law regarding what types of expenditures were properly
 allowable.... (147)

Thus, the use of the 2% floor intentionally disallows at least some amount of deductions in an area in which the government found enforcement difficult. (148) Once such a floor exists, any amounts deducted as miscellaneous itemized deductions are, by definition, large enough in the aggregate to exceed two percent of the taxpayer's AGI. As with medical expenses, (149) that should make erroneous miscellaneous itemized deduction claims both easier to detect and more financially worthwhile for the government to pursue. (150)

Putting employers in charge of monitoring the business legitimacy of employment-related expenditures facilitates the IRS's enforcement responsibilities, particularly for small expenses. While the IRS could not pursue many taxpayers taking deductions of a few hundred dollars for employment-related expenses, employers have a financial incentive to seek substantiation of expenses for which employees seek reimbursement. In addition, employers are better situated to know whether a particular type of expense, such as an entertainment or travel expense, really is business-related. Employers therefore verify both that (1) the taxpayer incurred an expense in the amount indicated and (2) the expense was for the purpose claimed. (151) The tax law's compliance focus in this context is underscored by the requirement that the employer maintain what the regulations term an "accountable plan" (152)--one "that meets the requirements of ... business connection, ... substantiation ..., and ... returning amounts in excess of expenses" (153)--for the employee to benefit from the exclusion.

The IRS cannot mandate that employers reimburse all business-related expenses, however. The reality that some employers may be less financially sound or less generous means that there will likely always be some unreimbursed employee business expenses. The current tax system, then, relies on the fact of reimbursement where it does occur to allow an exclusion, but imposes a taxpayer-sensitive bar (varying with AGI) on such unreimbursed employee business expenses (once aggregated with other miscellaneous itemized deductions). (154)

2. The self-employment comparison

The imposition of the 2% floor on unreimbursed business expenses has raised questions because of the natural comparison to the self-employed, who are entitled to deduct their business expenses above the line (155) without regard to any oversight of their expenses by a third party. (156) The self-employed are not inherently less likely to cheat on their taxes. In fact, they generally pose tax compliance problems (though a substantial portion of the cheating likely consists of omission of income). (157)

The problem of how to address erroneous deductions by the self-employed is evident when considering this from the perspective of tax system design. In designing an income tax, legislators could opt to allow full deduction of the business expenses of the self-employed, a partial deduction, or no deduction. Unfortunately for the tax system, the self-employed--unlike those employed by another--do not routinely have present a third party that the IRS can leverage to foster compliance. Although self-employed individuals may present some expenses to clients for reimbursement, many business expenses, such as rent, supplies, and equipment, may not be directly charged to particular clients but instead may be factored into the rates charged for services. (158) Thus, it is not easy to harness the interests of a third party to foster compliance by the self-employed.

Yet, disallowing or limiting the business deductions of the self-employed would be inconsistent with a normative income tax. (159) Moreover, allowing anything less than a full deduction for the business expenses of the self-employed could stifle entrepreneurship, (160) and, as a result, probably would not be politically viable. While, for employees, lack of reimbursement for an expense signals that the expense may be suspect, (161) that is not the case for independent contractors.

Thus, a rational income tax design determining how to treat business expenses could start with a baseline of full deduction of business expenses for those who typically have no third party to reimburse those expenses--the self-employed. For employees, who by definition have a third party who might reimburse some expenses, such a system could allow similarly favorable treatment only for those expenses accounted to the employer under a plan structured to alleviate tax compliance concerns. Conversely, the system would look askance on employees' unreimbursed expenses, though perhaps without disallowing them entirely, in recognition of the importance of allowing expenses that support a flow of income, (162) given that some of these unreimbursed expenses may be entirely legitimate. That would result in something like the current approach to this issue under the federal income tax.

The system is imperfect--among other things, the increased opportunity to evade taxes available to the self-employed encourages over-investment in that sector of the economy. (163) Yet it does employ compliance constraints where they are available--for employees--while generally continuing to allow the deductions that support entrepreneurship.

3. Moving expenses

A seeming counter-example to the general federal income tax treatment of employee business expenses involves job-related moving expenses. Qualifying employer-reimbursed moving expenses are allowed as an exclusion, (164) which is comparable to the treatment of other qualifying employer-reimbursed expenses. However, rather than being limited in their deductibility, an employee's unreimbursed qualified moving expenses are allowed in full, as an above-the-line deduction. (165) Thus, in this area, there is no cutback on the deduction side.

This seeming anomaly may exist in part because, due to the way the provision is structured, the taxpayer need not have an employer (from which to seek reimbursement) in the new location at the time of the move in order to qualify for the deduction. Moves to obtain work qualify if the taxpayer becomes employed fairly soon after the move, (166) which may serve the policy goal of facilitating moves to areas where the taxpayer has better prospects. (167) In addition, compliance-oriented cutbacks on the moving expense deduction may be unnecessary. Moving expenses may be less likely to be fabricated than other unreimbursed expenses because the provision requires a new principal place of work "at least 50 miles farther from his former residence than was his former principal place of work." (168) The W-2 Forms that an employee is required to include with the return when it is filed the following spring (and that the employer must send to the IRS) should tip off the IRS as to likely violations of this requirement, and thus help constrain compliance in this area.

In sum, in numerous ways, substantive federal income tax law takes into account whether a third party is available to verify the taxpayer's claims. Employers are routinely used for that purpose, but the principle applies in other contexts as well. Congress has thus implicitly recognized that third parties can be used by the government to protect the federal treasury so long as their own self-interest provides an incentive to verify claimed amounts.


The discussion above demonstrated the important role that third parties play in fostering tax compliance, which benefits the tax system. (169) Third parties implicitly verify two facts: (1) the taxpayer incurred an expense or loss in the claimed amount, and (2) the expense or loss was for the claimed reason. (170) Thus, for example, an employer's reimbursement of a business expense claimed by an employee indicates to the government that the employee actually incurred the expense and did so in connection with the employer's business.

Third parties do not always behave in such benign ways vis-a-vis the federal fist, however. (171) In a multitude of contexts, third parties may actually foster tax evasion, colluding with the taxpayer in abusive transactions. Therefore, one of the challenges of a tax system that, implicitly or explicitly, relies to some extent on third parties, is to distinguish between compliance-fostering situations and others. (172)

A. The Zero-Sum Constraint and the Unselfish Exception

The reason that the government generally can rely on the third party's act of reimbursement as verification of the bona tides of the taxpayer's claim in each of the situations discussed in Part II is that, in each instance, that party disgorged funds, transferring them to an unrelated taxpayer. Such a transfer is zero-sum: whatever the taxpayer gains, the other party loses. Thus, structurally, the third party generally has no motive for making the payment other than to make good on the taxpayer's claim. (173)

In zero-sum contexts, the government generally can free ride on the third party's verification. There is an important exception to this point, however, which is any circumstance in which parties are willing to share with each other despite the zero-sum nature of the game. That is, if parties form an economic unit or are gratuitously willing to relinquish assets, the government cannot rely on the taxpayer's counterparty to act as a verifier.

Professor Ted Seto has explained, "[T]he [Internal Revenue] Code's general rules are written on the assumption that taxpayers are self-interested, unaffiliated individuals--the atomistic rationalists of the classic economic model. In general, we are unwilling to transfer property or rights to income to others simply to avoid tax...." (174) Thus, for example, if a taxpayer starting a business has more expenses than income in the first year, the government, relying on the self-interest of the taxpayer, generally can presume that the taxpayer intends to make a profit, and thus can allow a deduction for the net loss. (175)

Taxpayers may not be strictly self-interested, but instead willing to give away income or property to others with whom they share a close relationship, such as family members. (176) The willingness to do so, which exists for non-tax reasons, allows such taxpayers flexibility to structure transactions in ways they otherwise would not. (177) For example, a mother might be willing to transfer an income-earning property to her son if that would lower the tax rate applicable to that income and thus provide the family with a larger after-tax return.

Congress has recognized the types of special opportunities related parties may have to structure transactions so as to obtain tax benefits. Professor Ted Seto explains that "where [the] assumption of selfishness proves or is likely to prove incorrect, the Code makes adjustments to its otherwise applicable rules." (178) Thus, the Code includes an array of anti-abuse rules applicable to transactions involving related parties. (179) The rules incorporate the notion that related parties often do not act at arm's length. (180) For example, Code section 267 disallows a deduction for losses on sales between related parties--even sales at fair market value (181)--presumably because related parties cannot be relied on to engage in arm's length bargaining. (182)

In general, these anti-abuse rules are designed to prohibit a taxpayer from lowering tax liability while retaining control of the underlying asset or continuing to receive the economic benefits of owning the property, such as by claiming a tax loss on the sale of property to the taxpayer's parent or child. (183) Of course, these rules do not perfectly target the relationships that may result in opportunities for taxpayers to claim tax benefits while leaving their economic situations largely unchanged. (184) The underlying principle is nonetheless sound, however. (185)

The Code's anti-abuse rules also do not address every situation of this type. Judicial doctrines help fill gaps in the Code, however. For example, because the Code contains progressive tax rates, a taxpayer who does not mind sharing income with someone else, such as a family member, might try to direct the payment of income, such as salary or interest, to that person. (186) The assignment of income doctrine generally addresses such gratuitous transfers by taxing the income to the person who earned it (or who owns the property that gave rise to the income), rather than to the person who received the income. (187)

B. Manufactured Surplus Minimized: The Example of Wages

Transactions involving zero-sum transfers are quite different from contexts in which the transaction involves an opportunity to enlarge the pie, creating a "surplus" the parties can share by bringing in the government as an unwitting participant to subsidize the transaction--an amount this Article refers to as "manufactured surplus." As discussed above, in zero-sum contexts, the government can free ride on the third party's own incentives so long as the third party does not have an incentive to act unselfishly. The Code addresses the latter context through related-party rules.

By contrast, the government cannot free ride on the third party's incentives if the transaction involves the possibility of manufactured surplus. A simple example illustrates this point. Part I discussed the example of a tortfeasor compensating a tort victim, Ann, for injuries. (188) If Ann and the tortfeasor negotiate in the absence of tax consequences, as Part I implicitly assumes, the game is zero-sum. (189) However, when, as under current law, certain personal injury damages received by the victim are excludible from gross income (190) without affecting the tax consequences to the tortfeasor, the parties' allocation of an increased amount of the damages to the excludible category creates a surplus in which the tortfeasor can share through a reduced payment of damages.

For example, a tortfeasor might be willing to compensate a victim $1 million for a combination of injuries, only half of which are entitled to exclusion of damages from income. Assume that the victim will pay a 30% tax on any includible amounts. The proposed $1 million payment would result in $150,000 of tax for the victim, (191) yielding only $850,000 after tax. The tortfeasor could offer to cooperate in claiming that the full payment is allocable to excludible damages, so long as the victim accepts $900,000 instead. The tortfeasor would thus benefit from making a smaller payment, without affecting his or her own tax treatment, and the victim would benefit from receiving $50,000 more after taxes. This reality means that the government cannot simply respect the parties' allocation. Instead, not surprisingly, courts generally examine whether the allocation was made at arm's length. (192)

The government can act to eliminate manufactured surplus by linking an inclusion by one party with a deduction by the other. The payment of wages and salaries by employers provides an example of how this works. Manufactured surplus would exist, regardless of any withholding requirement, in a situation in which employees were taxed on their compensation but the employer were not entitled to a deduction. In that situation, the employee's taxes, if evaded, would create a pool the parties could divide through the payment of a lower, unreported salary. (193)

Because employers typically can deduct the wages they pay their employees, (194) the federal income tax imposes a structural limitation on this type of cheating. Paying an employee off the books makes it more difficult for the employer to claim a deduction because of the possibility that the IRS will match its return with the employee's. (195) The existence of a deduction thus effectively limits the pool of unpaid federal income taxes the parties can share to the amount, if any, by which the employee's taxes will exceed the value of the employer's deduction. If the employee's tax rate is only slightly higher than the employer's, the manufactured surplus will be small, and probably not worth the risks of collusion. (196) Moreover, if the employer faces a higher federal income tax rate than the employee does, the employer has no incentive to underreport the employee's wages in order to cheat on federal income taxes. (197) Thus, as this example demonstrates, allowing a deduction for amounts paid to a payee that will be taxed on those amounts can reduce or eliminate manufactured surplus. This is already the case in some areas of substantive federal income tax law. (198)

The compliance context is different with respect to employment taxes, however, because the employee's tax liability is not matched with similar tax savings by the employer. Instead, the, employer and the employee generally must each pay 7.65% of the employee s salary as employment taxes. (199) If the employer and employee collude not to report any of the employee's wages, then, after considering the value of the deduction an employer in the 35% tax bracket would have gotten on its share had the salary been reported, (200) that leaves approximately 12.6% of the employee's salary as surplus for the employer and employee to divide. (201) Large companies that are highly regulated cannot easily pay employees off the books. (202) A small company struggling financially might be inclined to collude with employees to evade employment taxes, however, particularly because a company that is unprofitable and thus has no corporate income tax liability cannot substitute cheating on its corporate taxes for cheating that requires collusion with employees. (203)

Thus far, this discussion has not focused on withholding because withholding is irrelevant to the creation or elimination of manufactured surplus under the federal income tax. (204) As indicated above, withholding is highly effective at obtaining compliance from employees with respect to the taxes on their wages and salaries. (205) It thus addresses an important source of evasion that can occur in the absence of collusion--employee cheating. Yet, once the tax is withheld, there is a possibility that the employer will fail to turn it over to the government. (206) That is a real problem in the U.S., and one that is not reflected in voluntary compliance statistics that show the high compliance rates of payees.

The opportunity for the employer to fail to remit funds withheld occurs not through collusion but as a result of the actual possession of funds belonging to the government. (207) As the phenomenon of embezzlement demonstrates, theft can occur when one party has access to another's funds. However, this non-collusive form of cheating generally is easier for the IRS to detect without an audit than collusion is because the former lacks the employee's collaboration. (208) The employee not only has little incentive to help hide the employer's theft, the employee's incentive is to report the withholding because the employee will get credit towards his or her taxes due for the amounts withheld even if the employer fails to remit those amounts. (209) Many employees are entitled to tax refunds based on withholding (210) and thus have a strong incentive to file and report the withheld amounts. (211)

C. Identifying Accommodation and Cooperating Parties

As the discussion above suggests, in order to be sufficient for government use as a tool for tax enforcement, structural mechanisms that rely on the actions of third parties require that the parties generally have actual arm's length relationships. Situations in which the third party and the taxpayer are related are straightforward situations that the law already recognizes as likely presenting a lack of independent action. (212) Of course, parties who are not technically related may also act in concert to obtain tax benefits that they divide in some way between them. In some contexts, the law may expressly allow for, and even encourage, such coordination. (213) Outside those contexts, such coordination merely provides a costly and unintended government subsidy.

Contexts that involve manufactured surplus inherently share the characteristic that a tax benefit, such as an exclusion, deduction, or rate reduction, is created without an offsetting inclusion to the other party. Accordingly, either (1) the parties differ in such a way that their tax consequences do not offset each other, (214) as in the case of "tax-indifferent parties," which are discussed in the next section, and/or (2) substantive tax law is such that it at least arguably provides a loophole that can be exploited. (215) In either context, the presence of certain types of parties may be a red flag suggesting the possibility of abuse. These are contexts the government should address either through legislation--ideally, limiting structurally the opportunities for abuse--or through the allocation of its enforcement resources.

1. Tax-indifferent parties

A party that will not be subject to federal income tax--such as a nonresident alien, foreign corporation, or tax-exempt entity--is well-situated to participate in a transaction creating manufactured surplus in which it can share. Typically, the way this works is that the tax-indifferent party is allocated the income (without paying federal income tax on it), while the U.S. taxpayer receives the benefit of an offsetting deduction or other favorable tax attribute. (216) The tax-indifferent party can then receive a share of the tax benefits in the form of a fee. The corporate tax shelter literature well recognizes this phenomenon. (217) Professor James Eustice has referred to the presence of a tax-indifferent party as one of the "'big-time badges' of an abusive tax shelter." (218)

The presence of a tax-indifferent party in a transaction is thus a red flag for the IRS as to transactions that may warrant a closer look. In addition, some situations in which tax-indifferent parties have helped exploit the tax system could be structurally altered by explicitly hinging the U.S. taxpayer's tax benefit on U.S. tax liability of the counterparty. For example, in the "CINS" transaction at issue in ACM Partnership v. Commissioner, (219) a contingent installment sale was entered into by a partnership, and, during the first year, in which a substantial gain was recognized, a tax-indifferent party had a large interest in the partnership and was allocated most of the gain. The partnership interest of the tax-indifferent party was then redeemed, leaving the offsetting losses recognized in later years to be deducted by the U.S. partner, Colgate-Palmolive. (220) This ploy could be undermined by limiting the deductions of a U.S. taxpayer engaged in a contingent installment sale to the amount of gross income on the sale that gave rise to actual federal income tax liability. (221)

The prospect of manufactured surplus also arises if the taxpayer is a U.S. taxpayer but is simply tax-indifferent in the transaction because its tax consequences will not vary regardless of whether the other party obtains a tax benefit. An example of this latter situation is the allocation of a personal injury settlement between excludible and nonexcludible damages. (222) That is, although the tortfeasor and victim are adverse, arm's length parties with respect to whether the tortfeasor compensates the victim, they are not adverse as to the allocation for tax purposes of any such payment, because allocation of amounts to excludible personal physical injuries will not affect deductibility by the payor but will reduce the payee's income. Moreover, any tax benefit the victim receives may be shared with the payor through adjustment of the amount paid. (223)

If, contrary to current law, any deduction allowed the tortfeasor were tied to income inclusion by the victim, (224) case-specific examination of personal injury settlement agreements (225) would be needed far less frequently. More generally, much like the employer's deduction for its salary obligations reduces or eliminates any incentive to collude with the employee in income tax evasion, (226) allowing a deduction to a payor only for amounts includible in a payee's gross income eliminates transaction-specific tax indifference.

This linkage of the payor and payee sides of the transaction is a powerful tool. Although its application is not appropriate to situations in which strong policy reasons support both the payor's deduction and the payee's exclusion, its effectiveness in reducing manufactured surplus means that it should be consciously considered by Congress as a mechanism for limiting abuses.

2. Familiar parties

A close relationship that does not fall within the related-party rules can provide opportunities to coordinate on tax reduction. There are two distinct situations of this type. The first is where the parties act unselfishly rather than at arm's length. That can be the case with family or other relationships that do not fall within the related-party rules. (227) These situations are a cost of bright-line rules that cannot, without being overly broad, capture the full range of relationships within which a particular taxpayer will act unselfishly. (228)

The other context involves a transaction that can be exploited to create manufactured surplus. A party with close connections to the taxpayer but who is technically unrelated within the meaning of the Code may be an ideal accommodation party. The close relationship lowers transaction costs and reduces the documentation needed to accomplish collaboration because contacts can be less formal. (229) Such a party may be a former employee of the taxpayer, for example, (230) or another shareholder in the same closely held corporation. (231) The participation in a critical step of the transaction of what might be termed a "familiar" party may thus be a sign of possible structuring of an abusive transaction.

Of course, there are obvious non-tax reasons to transact with trusted, familiar parties rather than strangers. Transacting with a familiar party is not prima facie evidence of tax evasion. It is nonetheless a red flag that suggests that the parties might not be acting independently of one another, and, therefore that the government cannot free ride on their independence for tax purposes. It should therefore be recognized, particularly by courts, as an important fact in cases in which, for example, the taxation of the transaction depends on actions (such as a purchase) having been taken in good faith, (232) or the substantive bona tides of the form claimed.

Coordination between parties can also take place from the top, as where both parties are advised by the same tax attorney (233) or other relevant advisor. (234) Thus, as a general matter, a situation in which parties have a relationship beyond the transaction in question may suggest that the parties might not be acting independently of one another, and, therefore, that scrutiny of such a transaction that has been uncovered by the government may be warranted.

The converse is not true, however. Parties with no previous or other relationship may certainly have informal understandings that are not memorialized, so as to evade taxes that would otherwise apply. (235) For example, the parties may precommit that Step B will follow Step A but not memorialize that agreement if better tax consequences would result from the tax law's treatment of Step A and Step B as two separate transactions. (236) Nonetheless, the parties' risks in engaging in such informal agreements generally will be lower when they have a previous relationship or an advisor coordinating their activities, so the government should recognize this factor as a red flag that may call for closer examination of the transaction.

3. Soliciting parties

Abusive transactions require coordination between the parties to make the transaction take on the guise in which it will be presented to the IRS. This coordination requires either (1) contact of one party by the other, directly or through an agent, or (2) a preexisting relationship between the parties, as is the case with familiar parties, discussed immediately above. With respect to the former category, the contact may provide evidence that the transaction is designed to exploit an apparent loophole in the tax system. (237)

A transaction that is already structured (238) and marketed to the taxpayer by a third-party as tax-advantaged (239) is a prime example of such a contact. (240) The contact suggests that the transaction is motivated by tax-reduction motives, rather than non-tax business purposes. Such a contact, if uncovered by the IRS, is thus a red flag suggesting that the transaction--and similar transactions engaged in by other taxpayers--warrants closer scrutiny to determine its substantive content. (241)

4. Parties playing two roles

Another signal of possible cooperation to create manufactured surplus is where a third party has two roles in the transaction. The two hats worn by the third party may allow that party to operate at arm's length to the taxpayer with respect to only one of those roles while cooperating in a tax play in the other role.

An example of this phenomenon involves certain individual tax shelters of the 1970s and 1980s involving inflated basis amounts claimed by a purchaser (through the use of debt) so as to create large depreciation and interest deductions. Seller-provided financing was important to those shelters, because a conventional loan would be capped by creditworthiness and would likely be enforced, whereas the financing provided by a tax shelter promoter was largely artificial, allowing purchasers to obtain inflated basis at no additional risk to the seller/lender. (242) Professor George Cooper explained:
 [I]nternal leverage differs from the external variety in a critical
 non-tax respect. The party providing the leverage is also the
 seller of the property. Once he has gotten as much as he can for it
 in the form of cash or monies borrowed from external sources, it
 costs the seller nothing to offer additional internal leverage. He
 does not have to put up cash; he has only to raise the price of his
 asset. If the internal loan is eventually paid off, it is a
 windfall premium to him; if not, so what. A seller can blithely,
 even enthusiastically, increase internal leverage to whatever level
 is needed to please a buyer .... The availability of internal
 leverage is a remarkable boon to the structuring of tax shelters.

In other words, the seller of the property was able to function as an accommodation party in the tax shelter. The seller likely acted at arm's length with respect to the cash portion of the sale, but the loan, like an overly favorable settlement allocation, (244) was structured simply to maximize the buyer's tax savings, which the seller had no structural incentive to limit.

The more recent basis-shifting "FLIP" shelter shared the characteristic of using internal financing to create basis, though the basis was used to create a tax loss on sale, rather than inflated depreciation deductions. (245) As Professor Calvin Johnson has explained, the structure of the deal involved a Cayman Islands entity not subject to U.S. tax that "borrowed $100 million in cash from UBS in mid-1998 to buy UBS shares." (246) The shares were later redeemed in a transaction designed to be treated as a dividend for federal income tax purposes (so that the basis would not be recovered). (247) That unused basis was said to shift to a U.S. taxpayer who owned a small amount of UBS shares and was "related" to the Cayman Islands entity through options to buy ownership interests in that entity. (248) The U.S. taxpayer then sold the UBS shares for their fair market value, arguably recognizing a sizable loss for tax purposes. (249)

This area is one in which legislation could have brought about structural change that would have benefited the government. A statutory restriction on the extent to which the proceeds of a seller-provided loan are included in basis would have made the FLIP shelter more difficult to accomplish because it would have required a third party who was willing to advance the funds used to purchase the UBS stock. Professor Johnson points out that "[t]he $100 million cash never in fact left UBS's hands, from loan through repayment, but there were electronic entries for loan, issuance of stock, redemption back, and repayment of the loan." (250) UBS, therefore, unlike the situation an outside lender would have faced, took no risk of loss on the $100 million loan despite the fact that the borrower was a newly established Cayman Islands entity with few or no other assets. (251)

 The individual income tax is used throughout the world because of
 its appeal as a "modern" tax: it is based on ability to pay, can
 take into account taxpayers' personal circumstances, is adaptable
 to progressive rates, is more neutral in its economic effects than
 most indirect taxes, and has a highly elastic yield.... The
 individual income tax, however, can present formidable enforcement
 problems. (252)

The enforceability of the income tax is a critical policy issue. (253) In fact, the administrability of a tax system may be key to its longevity because enforceability will tend to foster compliance, while a lack of enforceability allows an ethos of noncompliance to fester. (254)

Laws that are largely self-policing will reduce the cost of overseeing compliance with the laws--much as speed bumps do for speed limits. One of the ways in which tax administrators ease enforcement difficulties is by using third parties as verifiers. The government generally can free ride on the action of a third party who transfers funds to the taxpayer in a zero-sum context because, structurally, the third party has its own incentive to verify the bona tides of the taxpayer's claim.

Fortunately, this reality already implicitly is reflected in the design of substantive federal income tax law. Numerous provisions that provide an exclusion for a reimbursed or compensated amount are unmatched by a deduction in full for a similar amount that is not reimbursed or otherwise compensated by a third party. The insight that the reimbursing payors serve as verifiers in these contexts, facilitating enforcement of the tax laws, thus helps explain a variety of instances of seeming lack of parallelism, and helps justify the distinct treatment as a policy matter.

Yet, third parties are not saints, and they have certainly been known to collaborate in tax evasion. Because parties do not always act independently of each other, one of the difficulties the government must face is to ferret out the types of situations in which transaction counterparties will not tend to act as verifiers. The various related party rules, though imperfect, reflect the reality that even in zero-sum contexts, taxpayers may not act out of pure self-interest when they have close connections, such as a family relationship.

Other contexts in which parties may cooperate to avoid or evade taxes involve situations that are not zero-sum because the transactional "pie" in which the parties will share can be enlarged through exploitation of an unintended tax subsidy. The presence of any of four types of parties--a tax-indifferent party, a familiar party, a soliciting party, or a party playing two roles in the transaction--presents a red flag that this type of cooperation may occur. These situations are thus ones that the government either needs to change structurally through legislation or other lawmaking or monitor closely in the audit process.

The recognition of the roles third parties play in fostering or undermining compliance with the law has important implications that extend beyond the tax law. Moreover, within federal income tax law, it holds promise as a mechanism to help decrease the tax gap. Thus, lawmakers should expressly consider its possible application when considering amending existing laws. Undermining existing structural constraints within the tax law would only increase opportunities for tax evasion. By contrast, increasing the tax law's structural constraints would help foster tax compliance at reduced enforcement cost to the government and little compliance cost to taxpayers.

(1.) See, e.g., Edward K. Cheng, Structural Laws and the Puzzle of Regulating Behavior, 100 NW. U. L. REV. 655, 662 (2006); Jon D. Hanson & Kyle D. Logue, The Costs of Cigarettes: The Economic Case for Ex Post Incentive-Based Regulation, 107 YALE L.J. 1163, 1173-74 (1998); Piroska Soos, Self-Employed Evasion and Tax Withholding: A Comparative Study and Analysis of the Issues, 24 U.C. DAVIS L. REV. 107, 121 (1990).

(2.) See William Neuman, City to Double the Number of Cameras at Traffic Lights, N.Y. TIMES, Sept. 14, 2006, at B3; cf. Cheng, supra note 1, at 690 n.205.

(3.) See Road Design Push, HERALD SUN (Melbourne), May 11, 1992.

(4.) See Eric A. Posner, Law and Social Norms: The Case of Tax Compliance, 86 VA. L. REV. 1781, 1782 (2000) ("A widespread view among tax scholars holds that law enforcement does not explain why people pay taxes.").

(5.) See id. ("The penalty for ordinary tax convictions is small; the probability of detection is trivial; so the expected sanction is small. Yet large numbers of Americans pay their taxes. This pattern contradicts the standard economic model of law enforcement, which holds that people violate a law if the benefit exceeds the expected sanction."). But cf. Leandra Lederman, Tax Compliance and the Reformed IRS, 51 U. KAN. L. REV. 971, 974 (2003) ("In fact, this simple comparison of relatively high rates of voluntary compliance rates with relatively low audit rates and penalties is flawed because it does not account for the role of information reporting and withholding in constraining the opportunity to evade tax.").

(6.) Dustin Stamper, Everson Pledges to Narrow Growing Tax Gap, 110 TAX NOTES 807, 807 (2006).

(7.) See Lederman, supra note 5, at 974-75; see also Cheng, supra note 1, at 676.

(8.) Information reporting thus helps to correct the information asymmetry between the government and the taxpayer by enlisting the assistance of a third party. Cf. Wendy E. Wagner, Stormy Regulation: The Problems that Result when Stormwater (and Other) Regulatory Programs Neglect to Account for Limitations in Scientific and Technical Information, 9 CHAP. L. REV. 191, 200-01 (2006) (discussing, in pollution control context, information asymmetries as to compliance by regulated entities).

(9.) As mentioned in the text, the taxpayer receives copies of the forms. See 2007 INTERNAL REVENUE SERV., FORM W-2 WAGE AND TAX STATEMENT 11 (2007), available at; INTERNAL REVENUE SERV., GENERAL INSTRUCTIONS FOR FORMS 1099, 1098, 5498, AND W-2G at GEN-9 (2007), available at

(10.) See JOHN BRAITHWAITE, MARKETS IN VICE, MARKETS IN VIRTUE 162 (2005) ("[T]he design genius of ... deduction at source ... [is that it] is controlled by employers rather than employees who suffer the consequences of the tax bill."). Withholding has other advantages for the government, as well. See infra note 205.

Although withholding is a highly effective tax collection mechanism, see Soos, supra note 1, at 126-30, it is not a panacea for all of the ills of a tax system and will be ineffective if it cannot be administered, see Richard M. Bird, Administrative Dimensions of Tax Reform, 10 ASIA-PAC. TAX BULL. 134, 136 (2004) ("The tax administration must be able to control withholders to make sure they hand over to the Treasury the amounts withheld, and it must also be able to check whether the amounts taxpayers credit against their liabilities have in fact been withheld.").

(11.) See Cheng, supra note 1, at 675-76.

(12.) See Karen Setze, Taxpayers Honest When Someone's Checking, Say IRS Officials, 111 TAX NOTES 1216, 1216 (2006) ("[R]esults from the recently completed individual reporting compliance study for 2001 ... showed that only 1.2 percent of wage income was underreported, 57 percent of nonfarm proprietor income was misreported ... and 72 percent of farm income was misreported."); see also Lederman, supra note 5, at 974-76.

(13.) Charles P. Rettig, Nonfilers Beware: Who's That Knocking at Your Door?, J. TAX PRAC. & PROC., Oct.-Nov. 2006, at 15, 15-16. The dramatic effectiveness of withholding and information reporting in obtaining compliance by payees is notable given that income is particularly tempting to falsify because unreported income, unlike falsified deductions or credits, need not show up on the taxpayer's tax return at all.

Of course, not all reported wage income is collected; withholding agents can fail to remit the taxes withheld. See infra notes 206-08 and accompanying text. Nonetheless, withholding is quite successful as a tax collection mechanism, in part because it centralizes collection. See Cheng, supra note 1, at 666 ("[A] chief advantage of structural laws is that they regulate centralized institutions rather than individuals. Institutions, usually in the form of corporations, are easier to regulate because they are smaller in number, have known locations, and have significant economic incentives to comply with government mandates."); Soos, supra note 1, at 126-30 (discussing the success of withholding as a tax collection device).

(14.) In this connection, Professor Edward Zelinsky has noted that realization offers an underutilized advantage over accretion taxation because realization events involve third parties that could be used as withholding agents. See Edward A. Zelinsky, For Realization: Income Taxation, Sectoral Accretionism, and the Virtue of Attainable Virtues, 19 CARDOZO L. REV. 861,901-03 (1997).

(15.) For example, as discussed below, a victim of personal physical injuries can exclude most resulting damage awards from income under Internal Revenue Code (Code) section 104(a)(2), I.R.C. [section] 104(a)(2) (2000), and thus receive them free of federal income tax, but a similarly injured victim who receives no award gets no tax deduction for the value of the injury. See infra text accompanying notes 31-34.

(16.) See, e.g., Jeffrey H. Kahn, The Mirage of Equivalence and the Ethereal Principles of Parallelism and Horizontal Equity, 57 HASTINGS L.J. 645, 645 (2006); Sophia Hudson, An Argument for Untidiness: Non-Parallel Treatment of Exclusions and Deductions in Federal Income Taxation, MICH. TAX LAW., Winter 2006, at 34, 34; cf. Joseph M. Dodge, Taxes and Torts, 77 CORNELL L. REV. 143, 148-49 (1992) (discussing context of personal injuries).

(17.) See infra Part II.

(18.) See infra Part I.C.

(19.) Cf. David M. Schizer, Frictions as a Constraint on Tax Planning, 101 COLUM. L. REV. 1312, 1315 (2001) ("This Article borrows from the economics literature, and specifically from Professors Scholes and Wolfson, in using the term 'frictions' to describe constraints on tax planning external to the tax law." (citing MYRON S. SCHOLES & MARK A. WOLFSON, TAXES AND BUSINESS STRATEGY: A PLANNING APPROACH 7 (1st ed. 1992))).

(20.) See Linda M. Beale, Putting SEC Heat on Audit Firms and Corporate Tax Shelters: Responding to Tax Risk with Sunshine, Shame and Strict Liability, 29 J. CORP. L. 219, 228 n.40 (2004) ("Accommodation parties such as foreign banks or tax-exempt pension funds often receive fee payments for participating in a shelter transaction in a role that is necessary to achieve the purported tax benefits.").

(21.) See Kahn, supra note 16, at 690; infra text accompanying notes 23-24.

(22.) See, e.g., David Elkins, Horizontal Equity as a Principle of Tax Theory, 24 YALE L. & POL'Y REV. 43, 43 (2006); Joseph T. Sneed, The Criteria of Federal Income Tax Policy, 17 SWAN. L. REV. 567, 568 (1965).

(23.) Elkins, supra note 22, at 43.

(24.) See Richard A. Musgrave, Horizontal Equity, Once More, 43 NAT'L TAX J. 113, 113 (1990); see also Elkins, supra note 22, at 43 ("The principle of horizontal equity demands that similarly situated individuals face similar tax burdens.").

(25.) Kahn, supra note 16, at 645; see also Hudson, supra note 16, at 34 (also discussing parallel and non-parallel treatment in this context).

Because Professor Kahn argues that "nonparallel treatment will result in unequal treatment of some persons," Kahn, supra note 16, at 652, his notion of parallelism can be understood as a component of horizontal equity. See id. at 690 ("[P]arallelism is merely one aspect of the broader concept of horizontal equity...."). Professor Kahn also argues that parallelism can operate as an independent policy norm. See id. at 652 ("For those few who consider horizontal equality to be irrelevant, the application of this Article's reasoning to that principle is of no consequence. However, many persons do give weight to horizontal equity, and even those who do not frown on unequal treatment of the same item.").

(26.) Kahn, supra note 16, at 645; see also Hudson, supra note 16, at 34 ("Because the effect of an exclusion for a reimbursement is equivalent to allowing a deduction for the expenditure or loss that was reimbursed, one might expect the tax law to provide parallel treatment by granting a deduction for such expenditures or losses when they are not reimbursed.").

(27.) See Kahn, supra note 16, at 646; infra text accompanying notes 43-52.

(28.) Kahn, supra note 16, at 647.

(29.) See generally id.

(30.) See id. at 655-90. Each of these examples, as well as others, is discussed at least briefly below. See infra note 71; infra Part II.

(31.) Professor Kahn mentions this context in the introduction to his article, see Kahn, supra note 16 at 648-49, and discusses it in more detail in Part III of his article, see id. at 660-64. This example is also a focus of Sophia Hudson's article. See Hudson, supra note 16, at 34-35 (referring to personal injury damages as "[p]erhaps the most well-known example of non-parallel treatment" and analyzing the personal injury damages context).

(32.) See I.R.C. [section] 104(a)(2) (2000).

(33.) See Kahn, supra note 16, at 648. Bob may be entitled to deduct his unreimbursed medical expenses, but medical expenses are subject to a floor that will reduce or eliminate otherwise deductible amounts. See infra note 74.

(34.) The reason for this is the exclusion provided in Code section 104(a)(2), which renders Ann's tax liability the same as Bob's despite her higher pre-tax income. Section 104(a)(2) will also cause Ann to owe less tax than someone with an otherwise identical tax profile except for a $2 million receipt that is includible (say damages received for a dignitary tort, rather than a physical injury). In this situation, the tax law fails to provide the same tax liability for taxpayers with the same amount of pre-tax income. Both of these disparities would no longer exist if section 104(a)(2) were eliminated. That would not be the case if a deduction were enacted for uncompensated personal injuries; it would eliminate the disparity in the former situation but not the latter. Moreover, it would give rise to the enforcement problems discussed in the text. See infra text accompanying notes 72-76.

(35.) See Kahn, supra note 16, at 648-49 (referring to this situation as involving a lack of parallelism and horizontal equity); cf. Dodge, supra note 16, at 148-49 ("A ... possible explanation for section 104 is that Congress wants to provide a federal subsidy to personal injury plaintiffs.... If Congress was so motivated, it acted arbitrarily by rewarding plaintiffs who receive recoveries, while precluding loss deductions for nonrecovering injured parties." (footnotes omitted)).

(36.) See Kahn, supra note 16, at 652 ("Indeed, there is a perverseness in the tax law's more favorable treatment of the reimbursed party than is provided to the one who is not compensated for his loss or expenditure since the latter is more deserving of sympathy.").

(37.) Professor Kahn does, however, argue that a "framing effect" that leads the public to perceive exclusions and deductions as different "is a factor in the existence of many of the nonparallel treatments of the tax law." Id. at 663. Exclusions and deductions are substantively, not just perceptively, different, however. See infra text accompanying notes 40-42.

(38.) Kahn, supra note 16, at 645. Sophia Hudson takes a similar approach. See Hudson, supra note 16, at 35 ("If there are more compelling reasons for an exclusion than for a deduction, then an exclusion may be warranted even though a deduction is not. The merits of each treatment weighed separately should drive tax policy rather than an inclination for tidy parallel treatment.").

(39.) For example, section 104(a)(2), the provision excluding personal physical injury damages, has been criticized. See I.R.C. [section] 104(a)(2) (2000); Mark W. Cochran, Should Personal Injury Damage Awards Be Taxed?, 38 CASE W. RES. L. REV. 43, 45 (1987) ("While the existence of section 104(a)(2) traditionally has been justified as a humanitarian gesture, more logical explanations occasionally have been offered. As illustrated below the proffered explanations either rest on erroneous assumptions or do not justify a blanket exclusion." (footnotes omitted)); cf Dodge, supra note 16, at 188 ("The analysis of existing section 104 with respect to recoveries for lost earning capacity yields the interesting conclusion that section 104 is neither categorically right nor wrong with respect to plaintiffs. Rather, plaintiffs end up in the right position after taxes under current section 104 if the damages are calculated in certain ways.").

(40.) See infra text accompanying notes 43-52 (discussing this equivalency in more detail); see also supra text accompanying note 27 (mentioning this equivalency).

(41.) See Kahn, supra note 16, at 647 ("Since the exclusion and deduction approaches generally are identical for tax purposes, one might expect there to be parallel treatment of reimbursed and unreimbursed expenditures and losses. That is, one might expect a taxpayer who incurs an expenditure or loss to be treated the same by the tax law whether the item is reimbursed or not."); cf. id. at 648 ("[T]he Code effectively provides a deduction for the taxpayer who happens to be compensated for a physical injury, but provides no corresponding relief for a taxpayer who is not compensated." (footnote omitted)).

(42.) See id, at 650 ("[T]he apparent equivalence of the deduction and exclusion is deceptive because different policy considerations can apply to each. So, the crucial question in such cases is whether the goal of parallel treatment is sufficiently strong to outweigh the other considerations.").

(43.) See, e.g., I.R.C. [section] 1 (2000) (tax imposed on individuals); id. [section] 11(a) (tax imposed on corporations).

(44.) See id. [section] 63.

(45.) See Kahn, supra note 16, at 646.

(46.) This example is analogous to Professor Kahn's example of employee "G" whose employer reimburses her for travel expenses for a trip she made for her employer. See id.

(47.) See I.R.C. [section] 162 (2000).

(48.) Id. 8 62(a)(Z)(A), (c).

(49.) See supra text accompanying note 44.

(50.) I.R.C. [section] 61 (2000).

(51.) See id. [section][section] 61, 62(a)(2)(A), 162.

(52.) Treas. Reg. [section] 1.62-2(c)(4) (as amended in 2003). Reimbursed amounts qualifying under this provision are also excluded for employment tax purposes. See id.

(53.) For example, if $100 is included in gross income but only $70 is deductible, taxable income is $30 ($100 minus $70), which is higher than the $0 of taxable income resulting from an exclusion.

(54.) See I.R.C. [section] 63(a), (b) (2000).

(55.) See id. [section] 67. Miscellaneous itemized deductions are also disallowed entirely in computing alternative minimum tax. See id. [section] 56(b)(1)(A)(i).

(56.) See id. [section] 68.

(57.) See Kahn, supra note 16, at 646.

(58.) I.R.C. [section] 101(a) (2000).

(59.) Id. [section] 104(a)(2).

(60.) Id. [section] 117.

(61.) See id. [section] 104(a)(2).

(62.) An example of this type of situation is the corporate dividends received deduction with respect to qualifying dividends. A corporation receiving a qualifying dividend takes it into gross income under Code section 61 but is allowed a deduction under section 243 for 100% of the receipt--which, combined, is tantamount to an exclusion. See id. [section][section] 61(a)(7), 243(a)(3).

(63.) Twenty percent of $100,000.

(64.) Twenty percent of $30,000. Of course, the lack of compensation for the personal injury would reduce pre-tax income from $170,000 to $100,000.

(65.) See supra text accompanying notes 31-34.

(66.) See Lawrence Zelenak, The Taxation of Tax Indemnity Payments: Recovery of Capital and the Contours of Gross Income, 46 TAX L. REV. 381,387 (1991) ("Allowing the loss [on a transaction] to offset the recovery, so as to prevent the creation of taxable income from a transaction which was an economic wash, is not necessarily inconsistent with a refusal to allow a deduction of the loss against unrelated income.").

(67.) "Basketing" is the conceptual grouping together of similar items, typically so that a taxpayer may deduct certain expenses or losses only to the extent of income of the same type. See Daniel N. Shaviro, Selective Limitations on Tax Benefits, 56 U. CHI. L. REV. 1189, 1189 (1989).

(68.) Cf Zelenak, supra note 66, at 387 ("[A]llowing an exclusion for a recovery of a nondeductible loss is much narrower than allowing a deduction for the original loss. The former allows the loss to offset only a recovery of the loss, while the latter would allow the loss to offset any income.").

(69.) The incentive to mischaracterize the nature of the receipt for tax purposes--which is distinct from the incentive to misrepresent the amount of the receipt--is discussed below, in connection with allocation issues that arise when a payment may be partially allocable to includible amounts. See infra text accompanying notes 190-92.

(70.) More accurately, Ann's claim that she received a payment of a particular amount is not suspicious. Ann's characterization of the payment for tax purposes cannot be respected in the same way, however, because she and the tortfeasor are not adverse with respect to allocation of the payment. This issue is discussed below. See infra text accompanying notes 190-92.

(71.) A similar analysis applies with respect to Professor Kahn's example of the exclusion of life insurance proceeds "paid by reason of the death of the insured." I.R.C. [section] 101(a) (2000); see Kahn, supra note 16, at 681-83. That is, whether or not it is normatively justifiable, an exclusion for life insurance proceeds does not raise compliance concerns because the third party vouches for the claimed amount, the entitlement to it, and who receives it. Moreover, the existence of the exclusion does not provide an incentive for fabricating receipts because an exclusion does not offset other income. See supra text accompanying notes 67-68. A deduction arising upon the death of an uninsured individual, by contrast, would raise difficult enforcement questions that would tempt taxpayers to cheat.

(72.) Such a deduction also could discourage a victim from trying to collect from the tortfeasor. Because there is no market-determined valuation of uncompensated injuries, the victim could opt to exaggerate the value of the claim and then deduct it, so as to receive from the government something close to (or even in excess of) the actual value of the injuries. The incentive created would be analogous to the situation of someone donating non-cash property to a charity in a situation in which its value is deductible and claiming an exaggerated valuation for tax purposes, rather than selling it for its true market value. Cf Marc Kaufman, Big-Game Hunting Brings Big Tax Breaks; Trophy Donations Raise Questions in Congress, WASH. POST, Apr. 5, 2005, at A1 (discussing phenomenon of appraisals of big game trophies reflecting many times their fair market value, so that hunters can obtain large tax deductions by donating them to museums).

(73.) See supra text accompanying note 64. Note that this type of cheating does not entail the costs of colluding with a counterparty. See infra note 196 and accompanying text (discussing the transaction costs of colluding in tax evasion).

(74.) Medical expenses generally are deductible to the extent they exceed 7.5% of adjusted gross income (AGI). See I.R.C. [section] 213 (2000); infra text accompanying note 119. This high floor greatly reduces the number of taxpayers that are entitled to medical expense deductions and thus makes medical expense deduction claims more salient. See infra text accompanying note 121.

(75.) See infra note 93; cf James S. Eustice, Abusive Corporate Tax Shelters: Old "Brine" in New Bottles, 55 TAX L. REV. 135, 161 (2002) ("The Service's shockingly low audit coverage makes the audit lottery an irresistible attraction; it is not even a lottery, but rather a virtually sure thing." (footnote omitted)).

The problem discussed in the text would remain an issue even if a deduction were only allowed for uncollected personal injury damages for which the taxpayer had a court judgment. The government could not readily determine that fact without an audit, and the existence of the deduction would mean that dishonest taxpayers could falsely claim such a deduction, gambling on the audit lottery. An exclusion poses no such risk because a false claim of receipt of damages does not reduce tax liability.

(76.) This is not meant to imply that the government would bear the burden of proof. Typically, the taxpayer would, unless the government sought to establish fraud. See I.R.C. [section][section] 7454, 7491 (2000); TAX CT. R. 142; Welch v. Helvering, 290 U.S. 111, 115 (1933).

(77.) See infra text accompanying notes 190-92; infra text accompanying notes 222-23. In some contexts, misallocation problems can be limited structurally by allowing the payor to deduct the payment, but only to the extent that it is included in the payee's gross income. See infra text accompanying notes 192-98; infra text accompanying notes 224-26.

(78.) See, e.g., Cheryl D. Block, Pathologies at the Intersection of the Budget and Tax Legislative Processes, 43 B.C.L. REV. 863, 865 (2002); Samuel A. Donaldson, The Easy Case Against Tax Simplification, 22 VA. TAX REV. 645, 732 (2003); Mona L. Hymel, Consumerism, Advertising, and the Role of Tax Policy, 20 VA. TAX REV. 347, 423 (2000); Edward J. McCaffery, The Missing Links in Tax Reform, 2 CHAP. L. REV. 233,239 (1999). Professor Sam Donaldson has argued that "simplicity is perhaps best understood as a component of efficiency." Donaldson, supra, at 652.

(79.) Bird, supra note 10, at 134; see also David M. Schizer, Enlisting the Tax Bar, 59 TAX L. REV. 331,334 (2006) ("The U.S. tax law is filled with complex rules that aspire to capture economic nuance. Yet these rules can be effective only if they can be implemented, not by some idealized tax administrator, but by the flesh-and-blood administrators we have.").

(80.) Milka Casanegra de Jantscher, Administering the VAT, in VALUE ADDED TAXATION IN DEVELOPING COUNTRIES 171,179 (Malcolm Gillis et al. eds., 1990).

(81.) Sneed, supra note 22, at 568 (emphasis removed).

(82.) See id. at 573 (referring, among other things, to "convenience and ease in assessment and collection from the standpoint of both government and taxpayer ... [and] adequate powers in government to deal effectively with the recalcitrant and fraudulent").

(83.) Equity as a tax policy goal generally is understood to have two components. "The first, horizontal equity, requires the equal treatment of equals. The second, vertical equity, calls for the unequal treatment of unequals." Rebecca S. Rudnick, Who Should Pay the Corporate Tax in a Flat Tax World?, 39 CASE W. RES. L. REV. 965, 1200 (1989) (footnotes omitted). Judge Sneed's discussion of equity generally focuses on horizontal equity. See Sneed, supra note 22, at 568 (listing the goal of "impos[ing] equal taxes upon those who enjoy equal incomes" (emphasis removed)). He also discusses vertical equity, however. See id. at 577, 581.

(84.) Sneed, supra note 22, at 601-02.

(85.) Cf. Roy Bahi & Jorge Martinez-Vasquez, The Nexus of Tax Administration and Tax Policy in Jamaica and Guatemala, in IMPROVING TAX ADMINISTRATION IN DEVELOPING COUNTRIES 66, 66 (Richard M. Bird & Milka Casanegra de Jantscher eds., 1992) ("Because low-income countries do not efficiently administer the systems they have in place, they fail to collect the true amount of revenue due, the efficiency objectives of the tax structure are not realized, and both the horizontal and vertical equity intent of the nominal tax structure are compromised."); Bird, supra note 10, at 135 ("How a tax system is administered affects its yield, its incidence, and its efficiency.").

(86.) Cf. Bird, supra note 10, at 135 ("Revenue outcomes may not always be the most appropriate basis for assessing administrative performance. How revenue is raised, i.e. the effect of revenue generation effort on equity, the political fortunes of the government, and the level of economic welfare, may be equally (or more) important as how much revenue is raised." (footnote omitted)).

(87.) Danshera Cords, Tax Protestors and Penalties: Ensuring Perceived Fairness and Mitigating Systemic Costs, 2005 BYU L. REV. 1515, 1522 (citing AM. INST. OF CERTIFIED PUB. ACCOUNTANTS, UNDERSTANDING TAX REFORM: A GUIDE TO 21ST CENTURY ALTERNATIVES 6, 29 (2005)).

(88.) See supra text accompanying note 6.

(89.) Stamper, supra note 6, at 807.

(90.) Id.

(91.) Cords, supra note 87, at 1521-22 (citing George K. Yin, JCT Chief Discusses the Tax Gap, 107 TAX NOTES 1449, 1449 (2005)); see CONG. BUDGET OFFICE, MONTHLY BUDGET REVIEW (2005)).

(92.) See I.R.C. [section] 6011(a) (2000).

(93.) The overall audit rate for individuals for the 2006 fiscal year was 1.0%. For all returns, it was 0.80%. See INTERNAL REVENUE SERV., PUBL'N NO. 55B, DATA BOOK 2006, at 23 tbl.9 (2007), available at

(94.) See supra text accompanying notes 31-34.

(95.) That is the implicit "basketing" nature of an exclusion, which, unlike a deduction, necessarily involves a receipt. See supra text accompanying notes 67-68.

(96.) In fact, the more puzzling question may not be why some unreimbursed expenses or losses are disallowed while their reimbursed counterparts are excludible, but rather why some unreimbursed expenses--such as personal casualty losses, medical expenses, and employee business expenses--are partially allowed. This issue is discussed below. See infra text accompanying notes 122, 147-50.

(97.) 40 B.T.A. 333 (1939), acq., 1957-2 C.B. 3.

(98.) Id. at 334.

(99.) Id. The Board of Tax Appeals is a predecessor of what is now the United States Tax Court. See Leandra Lederman, "Civil"izing Tax Procedure: Applying General Federal Learning to Statutory Notices of Deficiency, 30 U.C. DAVIS L. REV. 183, 191 n.36 (1996).

(100.) See Kahn, supra note 16, at 665; see also WILLIAM A. KLEIN ET AL., FEDERAL INCOME TAXATION 124 (14th ed. 2006) ("We cannot avoid committing one or the other of these two 'errors' (of the overall tax system ...) given that [the two hypothetical individuals] are not being taxed correctly relative to each other."); cf Zelenak, supra note 66, at 389. Professor Zelenak explains:
 If the Clarks are allowed the exclusion [for a payment from their
 negligent return preparer], they are treated equitably compared to
 some taxpayers and too favorably compared to other taxpayers. If
 they are not allowed the exclusion, they are treated equitably
 compared to some taxpayers and too harshly compared to other
 taxpayers. The basic point is that, once a decision has been made
 by Congress not to allow a deduction for a particular kind of loss,
 a taxpayer who later recovers such a loss is going to be taxed
 unfairly as compared with someone else, no matter how the recovery
 is treated.


(101.) See Kahn, supra note 16, at 653.

(102.) See id.

(103.) 40 B.T.A. 333 (1939), acq., 1957-2 C.B 3; see also Rev. Rul. 57-47, 1957-1 C.B. 23; Zelenak, supra note 66 (arguing, in part, that Clark was correctly decided).

(104.) See supra note 100 and accompanying text.

(105.) It is possible that the return preparer and client could collude to falsify a preparer error through the mechanism of an excessive payment for tax preparation services and return of the extra amount (purportedly because of preparer error). However, the principal benefit of this ploy would be an inflated deduction for the client in the amount of the preparer's fee, which is a miscellaneous itemized deduction if not connected to the taxpayer's trade or business. See I.R.C. [section][section] 62(a), 67, 212(3) (2000). The inflated deduction would have to appear on the taxpayer's return, inviting detection. Moreover, feigning error could be costly for the preparer as a reputational matter.

(106.) Kahn, supra note 16, at 653-54 (footnotes omitted).

(107.) That is, an overpayment made in a year with respect to which the taxpayer cannot claim a refund, typically because the statute of limitations on refund claims has expired.

(108.) See supra text accompanying notes 31-34.

(109.) See Alan Gunn, Basis and the Bad-Debt Deduction, 77 TAX NOTES 337, 341 & n.27 (1997); cf I.R.C. [section] 71(c)(1) (2000) (generally excluding child support receipts from the definition of alimony).

(110.) See I.R.C. [section] 262 (2000) (disallowing any deduction for "personal, living, or family expenses," except as expressly provided elsewhere). Some personal expenses of children are deductible. See, e.g., id. [section] 213(a) (authorizing a deduction for medical expenses of a dependent).

(111.) See Rev. Rul. 93-27, 1993-1 C.B. 32 (also discussing cases on this issue).

(112.) See I.R.C. [section] 166 (2000). Revenue Ruling 93-27 considered the bad debt loss provision as a possible basis for a deduction, and ruled that it did not apply. Rev. Rul. 93-27, 1993-1 C.B. 32. Professor Alan Gunn points out that "the fact that someone is liable to reimburse a taxpayer for an outlay does not make that liability a 'debt' within the meaning of section 166. This principle, not just difficulties about basis, justifies denying a deduction for worthless child support." Gunn, supra note 109, at 345.

(113.) See supra text accompanying notes 69-71. Issues of allocation between excludible and includible amounts can arise; although child support payments are exeludible, alimony payments are not. See I.R.C. [section] 71(a) (2000). However, the tax law contains a structural constraint in this regard; child support payments are not deductible by the payee, while alimony is. See id. [section] 215(a); Gunn, supra note 109, at 346. The allocation thus matters to both parties. By contrast, allocation issues pose a structural enforcement problem in contexts in which the allocation does not affect one of the parties' tax consequences. That issue is discussed below with respect to personal injury damages. See infra text accompanying notes 190-92.

(114.) The Code allows exclusion, without regard to cost recovery principles, both for amounts paid by self-funded health and accident insurance plans, under Code section 104(a)(3), and medical expense reimbursements from employer-funded plans, under section 105(b), so long as the reimbursed amounts were not previously deducted by the taxpayer. See I.R.C. [section][section] 104(a), 105(b) (2000). The inclusion for previously deducted amounts precludes the double tax benefit of both a deduction and exclusion with respect to the same amount.

(115.) See Robert A. Hillman, The Rhetoric of Legal Backfire, 43 B.C.L. REV. 819, 846 (2002); see also I.R.C. [section] 105(b).

(116.) Id. [section] 213(b).

(117.) Rev. Rul. 2003-102, 2003-2 C.B. 559.

(118.) They are thus only deducted if the taxpayer's itemized deductions (those deductions not listed in section 62(b)), in the aggregate, exceed the standard deduction. See I.R.C. [section] 63 (2000).

(119.) Id. [section] 213(a).

(120.) Id. [section] 56(b)(1)(B).

(121.) For a discussion of the enforcement benefits of AGI-based floors, in the context of the deduction of unreimbursed business expenses by employees, see infra text accompanying notes 147-50.

(122.) The verification role that third parties can play does not explain why employers can deduct health insurance premiums for their employees in full (under Code section 162) while employees cannot (because of the 7.5% floor on medical expenses). This disparity does, however, subsidize employers who provide health insurance for their employees, thus serving the non-tax goal of broad health insurance coverage. Employer provision of health insurance also allows some centralization of audits of this expense. Cf. Cheng, supra note 1, at 666 ("Institutions, usually in the form of corporations, are easier to regulate because they are smaller in number [and] have known locations....").

(123.) Kahn, supra note 16, at 655.

(124.) See I.R.C. [section] 165(c)(3), (h) (2000); Kahn, supra note 16, at 655.

(125.) See supra text accompanying notes 54-56.

(126.) Interestingly, casualty losses of business- and investment-use property are not subject to the limitations placed on personal casualty losses. See I.R.C. [section] 165(c)(1)-(2) (allowing deductions for business and investment-related losses, including casualty losses); cf id. [section] 165(c)(3), (h) (imposing limitations on deductibility of other casualty losses). As with independent contractors, discussed below, the liberal allowance of casualty losses in the business and investment context opens an opportunity for cheating, but one that apparently yields to the importance of entrepreneurship. See infra text accompanying notes 156-61. This analysis suggests that casualties of investment property should be subject to the same restrictions that personal casualty losses are.

(127.) See supra text accompanying notes 95-96. The exclusion of reimbursed amounts up to basis does give taxpayers an incentive to inflate basis. See Joseph M. Dodge & Jay A. Soled, Debunking the Basis Myth Under the Income Tax, 81 IND. L.J. 539, 540 (2006). The exclusion of returns of capital is entrenched and required as a matter of tax consistency, however. See Kahn, supra note 16, at 659. Moreover, the involvement of a third party in a disposition transaction poses little constraint on the taxpayer's subsequent basis claim, unfortunately. See Dodge & Soled, supra, at 556-60.

(128.) See supra text accompanying note 121.

(129.) See infra text accompanying notes 147-50.

(130.) See Rev. Rul. 80-99, 1980-1 C.B. 10, 10 ("[I]t is... a well-established position of the Internal Revenue Service that reimbursements for expenses incurred by a taxpayer on behalf of another in a nonemployment context are not includible in the taxpayer's gross income.").

(131.) See id.; see also Hudson, supra note 16, at 34 (mentioning this example).

(132.) See Rev. Rul. 77-280, 1977-2 C.B. 14.

(133.) See Rev. Rul. 55-555, 1955-2 C.B. 20; see also Hudson, supra note 16, at 34 (mentioning this example).

(134.) I.R.C. [section] 132(a)(2), (c) (2000).

(135.) See id. [section] 262 (disallowing a deduction for personal expenditures).

(136.) Id. [section] 119(a); see Kahn, supra note 16, at 683-88.

(137.) See I.R.C. [section] 119(a).

(138.) Id. [section] 119(a)(2).

(139.) See id. [section] 262. Similarly, the working-condition fringe provision of Code section 132 allows an exclusion for amounts that the employee could have deducted as a business expense without reducing the exclusion by any reduction to the deduction the employee would experience. See id. [section] 132(a)(3), (d). Unreimbursed employee business expenses are miscellaneous itemized deductions that, accordingly, often face reduction or disallowance. See infra notes 143-44 and accompanying text. In their federal income tax casebook, Professors Schmalbeck and Zelenak point out that this is an example of exclusions being treated more favorably than deductions, in that exclusions typically do not face phase-outs and floors. See RICHARD L. SCHMALBECK & LAWRENCE ZELENAK, FEDERAL INCOME TAXATION 493 (2d ed. 2007).

(140.) It would lower the employee's tax liability overall if the provision of meals and lodging lowers wages as an economic matter (despite Treasury Regulations' categorization of meals provided as "additional compensation" as nonqualifying, see Treas. Reg. [section] 1.119-1(a)(2) (as amended in 1985)). However, that is a separate issue from the concern about false and exaggerated claims that a deduction raises.

(141.) I.R.C. [section] 62(a)(2)(A) (2000).

(142.) Treas. Reg. [section] 1.62-2(c) (as amended in 2003); see also supra text accompanying notes 51-52. An inclusion coupled with an above-the-line deduction gives rise to the same amount of taxable income as an exclusion from gross income. See I.R.C. [section] 63 (2000); see also supra text accompanying notes 45-50.

(143.) See I.R.C. [section][section] 62, 67(b) (2000).

(144.) See id. [section][section] 55(b)(1)(A)(i) (disallowing miscellaneous itemized deductions for purposes of the alternative minimum tax), 67(a) (imposing 2% floor on aggregate miscellaneous itemized deductions), 68(a) (imposing overall limitation on itemized deductions).

(145.) See, e.g., Jeffrey H. Kahn, Beyond the Little Dutch Boy: An Argument for Structural Change in Tax Deduction Classification, 80 WASH. L. REV, 1, 62-63 (2005); Robert J. Peroni, Reform in the Use of Phase-outs and Floors in the Individual Income Tax System, 91 TAX NOTES 1415, 1423-25 (2001).

(146.) Kahn, supra note 145, at 62 (footnote omitted).

(147.) STAFF OF JOINT COMM. ON TAXATION, 99TH CONG., GENERAL EXPLANATION OF THE TAX REFORM ACT OF 1986, at 78 (Comm. Print 1987) (emphasis added).

(148.) See Deborah A. Geier, Some Meandering Thoughts on Plaintiffs and Their Attorneys' Fees and Costs, 88 TAX NOTES 531, 533 (2000) ("Such unjustifiable deductions were difficult, if not impossible, for the IRS to monitor in any effective way."); see also Daniel S. Nagin, Policy Options for Combatting Tax Noncompliance, 9 J. POL'Y ANALYSIS & MGMT. 7, 18-19 (1990) (referring to the miscellaneous itemized deduction provision as an example of "legislating away an attractive noncompliance opportunity").

(149.) See supra text accompanying note 121.

(150.) See Nagin, supra note 148, at 19-20.

(151.) A similar analysis holds for Professor Kahn's example of interview expenses by job seekers. See Kahn, supra note 16, at 688-90. The IRS has ruled that reimbursements by a prospective employer of interview expenses are excludible by the prospective employee, apparently without limiting its holding to situations in which the taxpayer is already working in the trade or business in which he is interviewing. See Rev. Rul. 63-77, 1963-1 C.B. 177. However, unreimbursed interview expenses are not deductible in instances in which the taxpayer is not already carrying on a trade or business. Moreover, if the taxpayer is carrying on the trade or business for which the employee is interviewing but is an employee, unreimbursed expenses are categorized as miscellaneous itemized deductions. See Kahn, supra note 16, at 688-89.

Professor Kahn persuasively explains the enforcement benefits of disallowing a deduction in full for an unreimbursed interviewing expense:
 If a full deduction were allowed, a taxpayer might arrange an
 interview with a firm in a resort town in order to qualify his
 travel for a deduction. While that may also occur when the
 prospective employer reimburses the applicant for his expenses, the
 difference in that latter case is that an independent party made a
 judgment that the applicant has a serious interest in the job, and
 the strength of that judgment is evidenced by the fact that the
 prospective employer expended its own funds to bring the applicant
 to the interview. There is merit to the government's accepting the
 bona tides of a taxpayer's action where a third party has
 demonstrated its belief that the action is business related.

Id. at 690.

(152.) See Treas. Reg. [section] 1.62-2(c)(2)-(4) (as amended in 2003).

(153.) Id. [section] 1.62-2(c)(1).

(154.) There are other situations in which the tax system makes use of the employer to impose a limitation on tax benefits, such as the exclusion under Code section 119 for meals and lodging furnished "for the convenience of the employer," discussed above. See supra text accompanying notes 136-39.

(155.) See I.R.C. [section] 62(a)(1) (2000).

(156.) See Kahn, supra note 145, at 50 ("[T]here is no 'independent' third party reviewing the expenses [of an independent contractor] to determine if they are legitimate (as there is with reimbursed employee expenses). It is difficult to see why Congress should trust independent contractors more than employees with these expenses." (footnotes omitted)); Peroni, supra note 145, at 1421 ("[A]n independent contractor has no employer to account to with respect to business expenses; yet, such a taxpayer is able to deduct his or her expenses in full in arriving at adjusted gross income and is not subject to the section 67 floor. It is unclear why Congress believes that independent contractor taxpayers are less likely to try to disguise personal consumption expenditures as deductible business expenses than are employees...." (footnote omitted)).

(157.) See supra notes 12-13 and accompanying text.

(158.) Cf Kahn, supra note 145, at 51 ("An independent contractor may set out a flat or hourly fee for services and personally absorb any expenses, an allowance for which is incorporated in the size of the fee charged.").

(159.) See J. Clifton Fleming, Jr., The Deceptively Disparate Treatment of Business and Investment Interest Expense Under a Cash-Flow Consumption Tax and a Schanz-Haig-Simons Income Tax, 3 FLA. TAX REV. 544, 545 (1997) ("[A] theoretically correct Schanz-Haig-Simons (SHS) income tax ... require[s] that dollars paid out as business or investment expenses be eliminated from the [tax] base." (footnotes omitted)).

(160.) See Leandra Lederman, The Entrepreneurship Effect: An Accidental Externality in the Federal Income Tax, 65 OHIO ST. L.J. 1401, 1444-46 (2004).

(161.) Of course it could simply mean that the employer is struggling financially or has a reimbursement policy that covers only a subset of expenses that constitute business expenses within the meaning of Code section 162.

(162.) See supra note 159 and accompanying text.

(163.) See Leandra Lederman, The Interplay Between Norms and Enforcement in Tax Compliance, 64 OHIO ST. L.J. 1453, 1506-07 (2003).

(164.) I.R.C. [section] 132(a)(6), (g) (2000).

(165.) Id. [section] 62(a)(15).

(166.) See id. [section] 217(c)(2).

(167.) See Hearing on H.R. 2264 Before the H. Comm. on Ways and Means, 103d Cong. 1064-65 (1993) (statement of H. Cris Collie, Executive Vice President, Employee Relocation Council).

(168.) I.R.C. [section] 217(c)(1).

(169.) Not only does the presence of third party verifiers result in increased compliance by the affected taxpayers, it should have a spillover effect. The perception that other taxpayers are compliant tends to reinforce compliance norms. See Lederman, supra note 163, at 1469-76. Even if compliance appears to be coerced (which it might not be when third parties are involved in a form of verification so subtle it has escaped notice even in the academic literature), that should not undermine the development compliance norms. See id. at 1484-99 (discussing the experimental and other literature); id. at 1499 ("If enforcement keeps at least some people in line, it may help retain a critical mass of compliant taxpayers. Enforcement may therefore have the effect of deterring some people and increasing the robustness of a compliance norm for others by minimizing their exposure to tax evasion." (footnotes omitted)).

(170.) See supra text accompanying notes 151-53 (discussing employer reimbursement of employee business expenses).

(171.) In Frank Lyon Co. v. United States, 435 U.S. 561 (1978), a case involving a sale-leaseback by a bank, Worthen Bank & Trust Company, from the taxpayer, the Supreme Court said that the involvement of three parties in the case distinguished it from prior cases involving two parties. The distinction in terms of this Article is that Frank Lyon involved not just a third party to the government-taxpayer relationship but also a fourth party insurance company providing financing. The presence of the additional party is irrelevant, as critics have pointed out. See Bernard Wolfman, The Supreme Court in the Lyon's Den: A Failure of Judicial Process, 66 CORNELL L. REV. 1075, 1099-1100 (1981). In fact, the Frank Lyon case involves the red flag of likely coordination between the parties by their shared tax attorney. See id. at 1098; infra note 233 and accompanying text.

(172.) Markets cannot necessarily be relied on to foster tax compliance. In fact, competition may foster tax evasion. For example, small businesses experiencing market pressures may resort to tax evasion as a way to cut prices. See Lederman, supra note 163, at 1505-06. Similarly, publicly held companies face pressure to lower their effective tax rates. See, e.g., Robert L. Nelson & Laura Beth Nielsen, Cops, Counsel, and Entrepreneurs: Constructing the Role of Inside Counsel in Large Corporations, 34 LAW & SOC'Y REV. 457, 475 (2000) (quoting a tax lawyer as stating, "Last year we ... had one very specific objective in mind for the corporation. We had to lower our effective tax rates; they had just gotten too high. We had a lot of pressure and we brought [them] way down."); Joel Slemrod, Tax Minimization and Corporate Responsibility, Address Before Council on State Taxation (Aug. l, 2002), in 96 TAX NOTES 1523, 1523 (2002) ("One hears of the increasing pressure on tax departments to lower the company's effective tax rate, to convert the tax department from a cost center to a profit center."). Those pressures can spur the use of abusive tax-minimization strategies. See Developments in the Law--Corporations and Society, 117 HARV. L. REV. 2169, 2249 n.4 (2004) ("The marketplace rewards companies with lower effective tax rates than their peers, creating a powerful competitive pressure for executives to manage tax liabilities aggressively.").

These market pressures can give rise to norms of noncompliance. See Lederman, supra note 163, at 1508. Enforcement can be key to tipping norms to ones of compliance. See id. at 1508-09 (making this argument in the context of small business); cf Susan Cleary Morse, The How and Why of the New Public Corporation Tax Shelter Compliance Norm, 75 FORDHAM L. REV. 961 (2006) (discussing development of an anti-promoted tax shelter group norm in public companies following the Sarbanes Oxley Act, but not discussing the market's role in reinforcing or undermining that norm).

(173.) In some contexts, an incentive to mischaracterize the nature of the payment nonetheless arises. This issue and a structural response to it are discussed infra in text accompanying notes 222-26.

(174.) Theodore P. Seto, The Assumption of Selfishness in the Internal Revenue Code: Reframing the Unintended Tax Advantages of Gay Marriage 6 (Loyola Law Sch. L.A., Legal Studies Paper No. 2005-33, 2005), available at Professor Calvin Johnson has explained, "If you leave a couple of hundred dollars on a park bench, I suspect that you will eventually get a theft loss. But that's not the point." Calvin H. Johnson, What's a Tax Shelter?, 68 TAX NOTES 879, 881 (1995).

(175.) See I.R.C. [section][section] 162, 165 (2000). This statement calls for a caveat. Congress has recognized that the ability to deduct business losses may lead some taxpayers to claim as a business an activity that actually is not engaged in for profit. Code section 183 generally denies a deduction for net losses from such activities. See id. [section] 183(b). Activities that involve strong elements of consumption and do not actually produce a profit will tend to fall within section 183. See Treas. Reg. [section] 1.183-2(b)(9), (c) ex. 1-3 (1972); cf. Portland Golf Club v. Comm'r, 497 U.S. 154, 166 n.16 (1990) ("The general rule that losses incurred in a not-for-profit activity may not be used to offset unrelated income rests on the recognition that one who incurs expenses without an intent to profit presumably derives some intrinsic pleasure or benefit from the activity.").

(176.) See Alex Raskolnikov, The Cost of Norms: Tax Effects of Tacit Understandings, 74 U. CHI. L. REV. 601,633 (2007) (referring to "the norm of mutual cooperation and support among family members (the 'family commitment norm')"); Seto, supra note 174, at 7-8 ("Not all marriages or parent/child relationships are unselfish, of course. But the assumption of selfishness fails commonly enough in such relationships....").

(177.) See Raskolnikov, supra note 176, at 633 ("[T]he 'family commitment norm' ... did not arise to reduce family members' taxes.... However, as the proliferation of the related party rules in the Internal Revenue Code amply demonstrates, Congress has realized that the family commitment norm has a strong potential to be used in tax planning."). See generally Seto, supra note 174.

(178.) Seto, supra note 174, at 4.

(179.) See id.; see, e.g., I.R.C. [section] 267 (2000) (disallowing loss on sale to related party). Of course, transactions can also be designed to exploit related-party rules. See Calvin H. Johnson, Tales from the KPMG Skunk Works: The Basis-Shift or Defective-Redemption Shelter, 108 TAX NOTES 431,434-36 (2005) (describing FLIP/OPIS shelter).

(180.) The tax law can apply rules or standards to distinguish between arm's length and non-arm's length transactions. In certain situations, the Code chooses rules, such as in Code section 267. Standards are more flexible but operate ex post, and the results may be more variable because of the lack of bright-line rules. For example, Professor Brant Hellwig has argued that the Supreme Court has extended the assignment of income doctrine beyond its original scope, reaching results in the context of commercial transactions that should instead have been reached other ways. See Brant J. Hellwig, The Supreme Court's Casual Use of the Assignment of Income Doctrine, 2006 U. ILL. L. REV. 751, 753-54.

(181.) See I.R.C. [section] 267 (lacking a "fair market value" exception). Code section 267 thus generally establishes an irrebuttable presumption that a taxpayer cannot claim a tax loss on property sold to a member of the taxpayer's family, for example. See Miller v. Comm'r, 75 T.C. 182 (1980) (applying Code section 267 to disallow losses on sales between hostile brothers required under decision of arbitrator). However, Treasury Regulations provide an exception for a sale at fair market value between members of a controlled group of corporations, if the sale involves a receivable received from a non-member of the controlled group, and the selling member recognized income on the transaction that gave rise to the receivable. See Treas. Reg. [section] 1.267(f)-l(f) (as amended in 2006).

(182.) A rule that allowed a deduction for sales between related parties at fair market value (but not otherwise) would be very expensive to enforce because of its fact-sensitive nature.

(183.) Professor Seto terms this the "benefits and burdens" approach. Seto, supra note 174, at 5.

(184.) Typically, each such rule sets forth a group of relationships to which it applies. See id. at 4-5. These groups differ; for example, siblings are included for some purposes but not others. Compare I.R.C. [section] 267(c)(4) (including siblings and half-siblings in the definition of "family" in that section) with id. [section] 318(a)(1) (failing to include siblings in group with heading "[m]embers of family"). Professor Seto has pointed out that these differences make little sense. See Seto, supra note 174, at 5-6. The groupings are also both overinclusive and underinclusive. See id. at 6 ("[I]n many contexts the approach's underlying factual premise is simply false."); cf Raskolnikov, supra note 176, at 633 & n.147 ("Overstating only somewhat, the same implicit norm that would jolt my stepmother into action if I needed her help because I became sick or injured would also guide her decisions if I asked her to hold some stock that I 'sold' to her to realize a tax loss.... [Yet] the related party rules of [Code section 267] ... do not apply to stepparents."). Some brothers, for example, have no personal relationship and do not seek to protect each others' economic interests. See Miller, 75 T.C. at 184 ("During the ... period [in question], the brothers did not see each other socially and rarely spoke. Their strained relationship continued. They did not trust each other. Although petitioner is an attorney specializing in real estate and [his brother] Marvin is in the real estate business, neither has referred any business to the other for several years."). Conversely, some cousins may act as close as most siblings, yet they generally are treated as unrelated by the Code's related-party provisions. See, e.g., I.R.C. [section][section] 267(b)(1), (c)(4)-(5), 318(a)(1).

(185.) Cf. Seto, supra note 174, at 7 ("Some proxy for an expected failure of the assumption of selfishness is ... necessary to invoke the special [anti-abuse] rule [that applies regardless of whether transactions are at arm's length]; specified formal relationships serve this proxy role.").

(186.) See I. Richard Gershon, Teaching Federal Income Taxation Using Socioeconomics, 41 SAN DIEGO L. REV. 201, 206 n. 19 (2004).

(187.) See Hellwig, supra note 180, at 754.

(188.) See supra text accompanying notes 31-34.

(189.) See supra text accompanying note 70.

(190.) See I.R.C. [section] 104(a)(2) (2000).

(191.) This is 30% of the $500,000 taxable portion.

(192.) See, e.g., Bagley v. Comm'r, 105 T.C. 396, 409 (1995) (allocating a portion of award to punitive damages and noting that "it was clearly in the interest of both parties not to show an amount allocated to punitive damages"); Robinson v. Comm'r, 102 T.C. 116, 129 (1994) ("[T]his Court will not blindly accept the terms contained in a settlement agreement, especially when the circumstances behind the agreement indicate that the allocation of the amounts contained therein was uncontested, nonadversarial, and entirely tax motivated."); McKay v. Comm'r, 102 T.C. 465, 484 (1994) (respecting allocation where the "record ... establishes that petitioner and [the payor] were hostile adversaries with respect to the allocations made in the settlement agreement"), vacated, 84 F.3d 433 (5th Cir. 1996).

(193.) For example, if an employee's market salary were $50,000 per year, and the employee were taxed at a flat 20% rate, the employee's tax liability on a reported salary of $50,000 would be $10,000. If the employer were allowed no deduction for that salary, then, considering just the effect of federal income tax, the parties could benefit by failing to report the salary payments and dividing the $10,000. The parties could agree, for example, to a salary of $45,000 paid under the table. That salary would save the employer $5000, and would also increase the employee's take-home pay by $5000 because he would only receive $40,000 net of taxes were he paid $50,000 but taxed $10,000.

(194.) I.R.C. [section] 162 (2000). The incentives are different for employers of domestic help because that wage expense is considered a personal expense and thus is not deductible. See id. [section] 262. However, certain dependent care expenses give rise to a tax credit. See id. [section] 21.

(195.) Cf Gideon Yaniv, Withholding and Non-Withheld Tax Evasion, 35 J. PUB. ECON. 183, 185 (1988) ("[U]nderstatement of wage payments results in overpayments of profit taxes....").

(196.) For example, if the employee's marginal rate is 28% and the employer's marginal rate is 25%, the federal income tax that will be saved by colluding is only 3% of the unreported salary. For example, if the employee's salary is $50,000, the federal income tax savings from colluding to report the entire amount would be only $1500. The small amount by which each party could benefit may not warrant the risks of cheating in this way, particularly because there are other, possibly less risky, ways in which each party can cheat. For example, the employer can underreport business profits for tax purposes.

Collusion increases the transaction costs of cheating and thus reduces its likelihood. Cf Gideon Yaniv, Collaborated Employee-Employer Tax Evasion, 47 PUB. FIN. 312, 314 (1992) ("Assuming that all employees have identical tastes, collaboration with their employer would be desirable either to all or to none. The employer, on his part, would collaborate either with them all or with none, fearing that those uncollaborated with (although wishing to be) might inform the tax authorities."). Whether or not employees have identical tastes, there is always the risk of defection, both on the part of employees not participating in the collusion, and on the part of those who participate but, for example, decide that they want more money to keep their end of the bargain.

The IRS is authorized to pay a bounty for "(1) detecting underpayments of tax, or (2) detecting and bringing to trial and punishment persons guilty of violating the internal revenue laws or conniving at the same." I.R.C. [section] 7623(a) (2000). "Any amount payable under the preceding sentence shall be paid from the proceeds of amounts collected by reason of the information provided...." Id. Former employees of a taxpayer are among the most likely people to report to the IRS tax cheating by a taxpayer. See Richard A. Carpenter, Practical Guide to Understanding Criminal Tax Matters, 27 TAX'N FOR LAW. 41, 45 (1998) ("Ex-spouses, ex-lovers, and ex-employees, seeking a form of revenge---or possibly a reward under Section 7623--are some of the best sources of suspected tax crimes for CID [the IRS's Criminal Investigation Division].").

(197.) For example, assume that (1) the employer is a corporation paying income taxes at the top marginal federal rate of 35%, (2) the employee's marginal income tax rate is 20%, and (3) there are no relevant taxes other than federal income taxes. Assume also that the employee's market wage is $30,000 per year. The employee would owe $6000 of taxes in this example, for a net return of $24,000. If the employer were not to obtain any tax benefit from paying labor costs, the employer could try to share with the employee the surplus cheating would create, at the limit paying the employee entirely off the books at an amount above $24,000 but below $30,000, such as $25,000. The $25,000 would not be taxed to the employee in this example, because it would be paid under the table.

However, employers are entitled to a federal income tax deduction for wages paid. At a $30,000 wage, the deduction is worth $10,500 to the employer in the 35% bracket. The employer nominally paying a $30,000 salary would thus actually bear only a $19,500 cost, net of federal income tax. The employer is therefore better off paying a $30,000, on-the-books, deductible salary than a $25,000, off-the-books, non-deductible salary because the employer's alter-tax wage cost is $5500 lower in the former scenario than in the latter. The employer's financial incentive is thus to pay the higher, market wage and report it, not to collude with the employee. An employee determined to increase his alter-tax salary from $24,000 to $25,000 could either negotiate with the employer to share some of the employer's tax savings, or cheat on his own income taxes (such as by inflating deductions), without collusion with his employer.

(198.) See I.R.C. [section][section] 83(h) (2000) (property received for performance of services), 215(b) (2000) (alimony).

(199.) See id. [section][section] 3101, 3111; see also id. [section][section] 1401(a) (providing a rate of 12.4% for the general self-employment tax), (b) (setting the hospital insurance tax rate for the self-employed at 2.9%). The 12.4% tax is capped based on a wage base of $97,500 for 2007, but the 2.9% tax is not capped. See id. [section][section] 3101(b), 3111(b), 3121(a)(1); INTERNAL REVENUE SERV., PUBL'N 15, (CIRCULAR E), EMPLOYER'S TAX GUIDE (INCLUDING 2007 WAGE WITHHOLDING AND ADVANCE EARNED INCOME CREDIT PAYMENT TABLES) (revised Jan. 2007), at 16 (providing 2007 wage base), available at pub/irs-pdf/p15.pdf.

(200.) See I.R.C. [section] 162(a) (2000).

(201.) That is, 65% of the employer's 7.65% share added to the employee's 7.65% share.

The likelihood of collusion is somewhat mitigated by the employee's prospects of receiving Social Security benefits tied to contributions. Cf James Alm et al., Tax Structure and Tax Compliance, 72 REV. ECON. & STAT. 603, 613 (1990) (finding, in study of workers in Jamaica, that individuals increased their compliance when benefits related to payroll taxes increased).

(202.) Cf Cheng, supra note 1, at 666 ("Institutions, usually in the form of corporations, are easier to regulate because they are smaller in number, have known locations, and have significant economic incentives to comply with government mandates."); Yaniv, supra note 196, at 313 ("Collaborated evasion ... may not only operate in small enterprises with fairly homogeneous employees. It may also be relevant to certain groups of workers in large enterprises ... where tax avoidance, in the form of paying part of the total compensation in non-taxable fringe benefits, is a very common practice. Avoidance and evasion are closely related decisions....").

(203.) See supra note 196; cf Christopher Bergin, CID to Employment Tax Evaders: "We Will Catch You," 2001 TAX NOTES TODAY 94-9 ("On the civil enforcement side ... [the IRS is] training agents to deal with cash-pay schemes."); Edward D. Urquhart & Susan Schwyn Martinez, Handling Investigations Involving Civil and Criminal Tax Cases, 45 S. TEX. L. REV. 193, 232 n.204 (2003) ("Preparing false payroll tax returns understating the amount of wages on which taxes are owed, or failing to file employment tax returns are methods commonly used to evade employment taxes." (citing INTERNAL REVENUE SERV., CRIMINAL INVESTIGATION, EMPLOYMENT TAX ENFORCEMENT PROGRAM)).

A company with no income, and thus no benefit from a deduction for employee compensation, would also face an unreduced employment tax pie to divide through collusion and would benefit with respect to income taxes as well by paying the employee under the table. See supra note 193 and accompanying text.

(204.) A withholding requirement could provide an incentive for the employer and employee to collude in tax evasion in a system in which employees are not required to file a return when they have no other income and their taxes are entirely withheld at source. See Yaniv, supra note 196, at 320 ("[I]f the tax and law enforcement parameters allow for the existence of bargains mutually beneficial to the parties (which they do if they generate incentives for underreporting in the absence of tax withholdings), a withholding system would result in increased tax evasion regardless of the finally agreed-upon bargain on the contract curve." (emphasis added)); id. at 314 n.2 ("Being single-job holders with no other income, whose entire tax liability must be deducted at source, employees will be assumed, as is the case in Israel, for example, to be exempt from the obligation to declare their earnings themselves through the filing of an income tax return."). The U.S. federal income tax is not an exact withholding system, however, and its filing requirement does not depend on whether all of the taxpayer's tax liability is covered by withholding. See I.R.C. [section] 6012(2000).

(205.) See supra text accompanying notes 12-13. There are a number of reasons for the success of withholding. First and foremost, bringing in a third party responsible for the payment of the taxes reduces cheating by the payee, the employee. See supra note 10 and accompanying text. It also eliminates reliance on the employee's ability to set aside enough money to pay the taxes. See Cheng, supra note 1, at 676. In addition, because withholding tax tables are designed to slightly overwithhold and thus put many taxpayers in a "refund" posture, the employee may be less inclined to cheat to avoid paying taxes on other income because the tax due will lower the employee's refund rather than result in additional tax owed. See Lederman, supra note 5, at 974-75 & n.23. Although a $200 smaller refund, for example, is equivalent to $200 of tax due, the $200 of tax due is more salient. In addition, the employee may not have funds to pay taxes due (an issue that does not arise with a smaller refund), and framing effects are such that a smaller refund may not be viewed in the same way as an increased amount owed, though taxpayers' expectations as to how much they will owe or receive may matter more. See John S. Carroll, How Taxpayers Think About Their Taxes: Frames and Values, in WHY PEOPLE PAY TAXES 43, 49, 60 (Joel Slemrod ed., 1992). Withholding also provides the government the use of the funds more quickly than if it had to wait until after the year closed.

(206.) See ABA Comm'n on Taxpayer Compliance, Report and Recommendations on Taxpayer Compliance, 41 TAX LAW. 329, 367 (1988).

(207.) See id. at 367-68 ("Noncompliant taxpayers are usually failing businesses.... The withholding problem exists because unprofitable businesses have access to a tempting pool of money at a time when money is desperately needed.").

(208.) This type of noncompliance also is subject to much higher penalties than apply to ordinary noncompliance. Instead of the typical 20% penalty applicable to just the delinquent taxpayer with respect to underreported income taxes, see I.R.C. [section] 6662 (2000), a 100% penalty applies to "[a]ny person required to collect, truthfully account for, and pay over any tax" who fails to do so. Id. [section] 6672(a). The Code, in turn, provides a right of contribution for a party who pays more than his share of the penalty. See id. [section] 6672(d). The IRS uses the penalty "merely as a tool to collect the full amount of the taxes withheld but not paid rather than to impose a penalty." Bryan T. Camp, Avoiding the Ex Post Facto Slippery Slope of Deer Park, 3 AM. BANKR. INST. L. REV. 329, 331 (1995). Delinquent "trust fund taxes" can nonetheless be difficult to collect. See ABA Comm'n on Taxpayer Compliance, supra note 206, at 367 ("Since at least 1981, 35-40% of all tax-delinquent accounts and 40-45% of the total amounts due on such accounts have involved employee taxes that have been withheld but not yet paid over to the Internal Revenue Service.").

(209.) See ABA Comm'n on Taxpayer Compliance, supra note 206, at 367 ("When this withheld tax money is not paid, the Service is not entitled to assess the tax payments against the employees, and moreover, must give refunds to over-withheld employees.").

Professor Yaniv assumes for purposes of his economic model of employer-employee tax evasion, which finds that withholding taxes will increase collusive tax evasion, see supra note 204, that the employee is not required to file a tax return. See Yaniv, supra note 196, at 314 n.2; supra note 204. The filing requirement contained in U.S. law, see I.R.C. [section] 6012 (2000), has an enforcement advantage in that the employee, in effect, has an incentive to report on the employer.

(210.) See Peter R. Orszag, Individual Income Tax Refunds, 106 TAX NOTES 599, 599 (2005) ("Refunds are not only substantial in the aggregate, but also very common. Data from the Internal Revenue Service suggest that more than three-quarters of fliers receive an income tax refund."); see also George Guttman, The IRS Is Betting Big on Its Prefiling Strategy, 91 TAX NOTES 24, 24 (2001) ("Our current system of slightly overwithholding on wages encourages individual taxpayers to file their returns to get their refunds."); supra note 205.

(211.) The fact of withholding also is subject to information reporting on Form W-2. Even if the employer fails to issue a W-2, the employee has an incentive to file a return and report the withholding, as discussed in the text.

(212.) See supra Part III.A.

(213.) Payments to an ex-spouse provide a good example. Alimony payments are deductible by the payor and includible by the payee. See I.R.C. [section][section] 61(a)(8) (2000), 71(a) (2000) (including alimony in gross income), 215(a) (2000) (authorizing a deduction for alimony). Property settlements, by contrast, are nondeductible and not includible. See id. [section] 1041 (a) (providing for nonrecognition of gain or loss on property transferred to an ex-spouse incident to divorce), (b) (treating the receipt of property as a gift, and thus excludible from income by the payee). Because payments must meet certain tests to qualify as alimony, see id. [section][section] 71(b), 215(b), they can be structured to fall within either the alimony paradigm or the property settlement paradigm.

The so-called "safe harbor leasing" rules of the early 1980s provide another example of this phenomenon. See, e.g., Michael J. Graetz, Your Tax Dollars at Work: Why U.S. Tax Law Needs to Be Changed, Address at the Emory University School of Law Randolph W. Thrower Symposium (Feb. 18, 1999), in 48 EMORY L.J. 849, 856 (1999) ("In 1981, Congress enacted something called 'safe harbor leasing,' which was known to most people as 'buy a tax break.' Under that legislation, a company that did not owe any taxes could capture the tax breaks that it would have gotten under rapid depreciation and other advantages by selling those tax breaks to another company, just by putting the word 'lease' at the top of a piece of paper." (footnotes omitted)).

(214.) Transactions giving rise to manufactured surplus in this context include those exploiting differences in marginal tax rates between the taxpayers, as well as those designed to transfer tax benefits from a party who cannot use them to one who can, such as "trafficking" in corporate net operating losses. See generally Daniel L. Simmons, Net Operating Losses and Section 382: Searching for a Limitation on Loss Carryovers, 63 TUL. L. REV. 1045 (1989).

(215.) This is typically the case with corporate tax shelters, for example, which often involve the artificial creation of tax benefits--such as a loss for tax purposes without an accompanying economic loss. The basis-shifting "defective-redemption" tax shelter, described by Professor Calvin Johnson in a Tax Notes article, Johnson, supra note 179, provides an example of this. See infra text accompanying notes 245-51.


(217.) See, e.g., Peter C. Canellos, A Tax Practitioner's Perspective on Substance, Form and Business Purpose in Structuring Business Transactions and in Tax Shelters, 54 SMU L. REV. 47, 54 (2001) ("The tax shelter stigma attaches most firmly and justifiably to transactions involving loss generation and/or tax-exempt accommodation parties to whom income is deflected or whose investment is used to generate a loss allocated to the shelter investor but without the tax consequences of debt incurrence."); Eustice, supra note 75, at 159 (including among indicia of a tax shelter the presence of "[a] tax-indifferent party ... inserted into the transaction, whose sole function is to absorb the tax burdens thrown off in the deal").

(218.) Eustice, supra note 75, at 159.

(219.) 157 F.3d 231 (3d Cir. 1998).

(220.) See id. at 239-44; Joseph Bankman, The Economic Substance Doctrine, 74 S. CAL. L. REV. 5, 8-9 (2000).

(221.) Another example is the "FLIP" basis-shifting shelter, discussed below, in which the taxpayer argued that it was entitled to the use of basis that was not recovered by a technically related party upon a theoretically taxable event, although the related party was a foreign taxpayer not liable for U.S. taxes. See Johnson, supra note 179, at 435-36; infra text accompanying notes 246-49. That argument could be limited by amending the language of the Treasury regulation on which the taxpayers relied. In fact, the IRS later stated that the relevant example in the Treasury regulations, which involved spouses, "is premised on the concept that an adjustment is appropriate where the redeemed spouse is required to include the full redemption proceeds as a dividend in gross income that is subject to U.S. tax and such spouse retains no stock to which the basis of the redeemed stock could attach." I.R.S. Notice 2001-45, 2001-2 C.B. 129, 129.

(222.) See supra text accompanying notes 190-92.

(223.) See supra text accompanying notes 191-92.

(224.) If payors tend to be more knowledgeable or have greater bargaining power, linking any deduction to the payee's inclusion might burden payees compared to existing law. Alternatively, Congress could repeal the already limited exclusion of section 104(a)(2), eliminating the allocation issues. Over time, as payees and juries learned that the payments were fully taxable, payments likely would increase to cover a least a portion of the tax liabilities.

(225.) See supra note 192.

(226.) See supra notes 194-97 and accompanying text.

(227.) See Raskolnikov, supra note 176, at 633 (stepmother and stepchild); Seto, supra note 174 (same-sex couples).

(228.) That is not to say that the rules cannot be tailored more precisely. However, tailoring will not eliminate this issue so long as some relationships within which some taxpayers act unselfishly (e.g., close friendships) are omitted.

(229.) But cf. Raskolnikov, supra note 176. Professor Raskolnikov argues that, in certain industries, norms become known and thus do not need to be memorialized in writing. For example, he explains that a market practice with respect to certain forward contracts and stock lending agreements developed to avoid linkage between the contracts for tax purposes; the bank that needed to borrow its clients' shares would postpone the request. Id. at 615. He explains:
 How could the banks take on a risk of not being able to borrow the
 clients' shares? They relied on a contractual norm rather than
 contractual language. Even though clients had no legally binding
 obligation to lend, the expectation was that they would cooperate
 and lend the shares. Banks knew this, many clients knew this, and
 those clients who did not were quickly educated by their bankers,
 their tax lawyers, or even other clients. Just as with the
 confidentiality norm, the market practice of delayed share lending
 ... emerged so that the parties could avoid incorporating the
 agreement to lend into explicit contractual language.

Id. at 616.

(230.) See Karen C. Burke, Black & Decker's Contingent Liability Shelter: "A Thing of Grace and Beauty"?, 106 TAX NOTES 577, 579 (2005) (describing Black & Decker's contingent liability transaction).

(231.) See Treas. Reg. [section] 1.351-1(a)(1)(ii) (as amended in 1996) (addressing the situation of an accommodation transferor in a Section 351 transaction).

(232.) Cf. Comm'r v. Day & Zimmermann, Inc., 151 F.2d 517, 519 (3d Cir. 1945) ("Despite the connection of Katz with Day & Zimmermann there is not the slightest reason arising from the agreed facts for any suspicious inference with respect to his purchase of the stock. The whole background of the transaction is meticulously set out and it all indicates the good faith of both the taxpayer and Katz.").

(233.) For example, with respect to Frank Lyon, 435 U.S. 561 (1978), Professor Wolfman has argued:
 It is not credible that Worthen and Lyon, while sharing the same
 tax lawyer, with both Mr. Lyon and the lawyer sitting on the
 Worthen board, were unaware of their differing tax needs and the
 way each might be helpful to the other at the expense of only the
 United States Treasury.

Wolfman, supra note 171, at 1098.

(234.) See Burke, supra note 230, at 579 n.14 ("In its Enron investigation, the JCT Report recommended that the tax law not 'permit use of accommodation parties such as employees, consultants, or advisors, to serve as a party in a transaction or arrangement' intended to provide delivery of tax benefits to a taxpayer." (quoting STAFF OF THE JOINT COMM. ON TAXATION, REPORT OF INVESTIGATION OF ENRON CORPORATION AND RELATED ENTITIES REGARDING FEDERAL TAX AND COMPENSATION ISSUES, AND POLICY RECOMMENDATIONS 25-26 (2003))).

(235.) See generally Raskolnikov, supra note 176 (discussing the costs of tax-driven social norms).

(236.) For a more specific example, see id. at 614-616 (discussing combination of variable delivery prepaid forward contract and loan of shares); supra note 229 (quoting from Professor Raskolnikov's discussion of this issue).

(237.) Cf. Elena Eracleous, Note, Losing the Audit Lottery: Corporate Tax Shelters and Judicial Doctrine, 5 FORDHAM J. CORP. & FIN. L. 205, 235 (2000) ("Corporations should be wary of transactions marketed as tax saving ideas because the IRS is likely to take that fact into account in determining their viability."). The taxpayer's motive for entering the transaction is relevant because it provides information to the government about how closely it needs to monitor the transaction in order to enforce the tax system.

(238.) See Eustice, supra note 75, at 158-59 (listing among indicia of a tax shelter that "[t]he transaction is prepackaged and fully predetermined from the get go").

(239.) See, e.g., Compaq Computer Corp. v. Comm'r, 113 T.C. 214, 215 (1999) ("[A] broker and account executive with Twenty-First [Securities Corporation] mailed a letter to petitioner soliciting petitioner's business. The letter stated that Twenty-First 'has uncovered a number of strategies that take advantage of a capital gain'...."), rev'd, 277 F.3d 778 (5th Cir. 2001); Janet Novack & Laura Saunders, The Hustling of X Rated Shelters, FORBES, Dec. 14, 1998, at 198 ("Recently, Forbes obtained copies of two different letters.... Each was sent by the accounting firm Deloitte & Touche this fall to a medium-size corporation.... [E]ach letter demands a bounty for zeroing out the company's taxes: a contingency fee of 30% of the tax savings, plus out-of-pocket expenses.").

(240.) Corporate tax shelters were aggressively marketed during the 1990s. See Morse, supra note 172, at 994-95. That no longer seems to be the case, at least with respect to public companies. See id. at 962 & n.4 (citing authorities). Nonetheless, "[o]n the evidence we now have, there is little reason to expect a permanent uptick in corporate tax compliance, even with respect to the relatively narrow issue of tax shelters. Commentators have previously observed a historical cycle of fraud, crackdown, compliance, a shift of focus from enforcement to service, and then more fraud." Id. at 1013 (footnotes omitted). Continued enforcement efforts could help sustain compliance norms, see supra note 172, but vigorous enforcement efforts may be politically difficult to maintain when compliance is high.

(241.) The economic substance doctrine, which generally is regarded as having two prongs, an objective one that considers the economics of the transaction and a subjective one that concerns the taxpayer's subjective business purpose, Developments in the Law--Corporations and Society, supra note 172, at 2254-65, helps distinguish abusive tax shelters from legitimate business deals that should be allowed to go forward so as not to inefficiently reduce productive activity. See David A. Weisbach, Ten Truths About Tax Shelters, 55 TAX L. REV. 215, 252 (2002) ("There is a difference between somebody engaging in a transaction for purely business reasons that happens to have fantastic tax consequences and somebody entering into the transaction solely to reduce taxes. In the former case, where the taxpayer enters into the transaction for business reasons, there is no economic distortion caused by taxes--while the person pays low taxes, behavior is not distorted by this prospect. In the latter case, where the motive is taxes, behavior is distorted, and there are real economic costs."). But cf. Daniel N. Shaviro, Economic Substance, Corporate Tax Shelters, and the Compaq Case, 88 TAX NOTES 221, 223 (2000) ("Leaving aside the institutional reasons why (for courts in particular) economic substance is a particularly suitable tool for deterring undesirable transactions, one might as well condition favorable tax consequences on whether the taxpayer's chief financial officer can execute 20 back-somersaults in the IRS National Office at midnight on April Fool's Day, if such a requirement turns out to achieve a better ratio of successful deterrence to inducing wasteful effort in meeting requirements that are pointless in themselves.").

In theory, taxpayers could falsify economic substance, or at least the subjective prong of the test, but that generally has not proven to be the case. See, e.g., Bankman, supra note 220, at 28; Mark P. Gergen, The Logic of Deterrence: Corporate Tax Shelters, 55 TAX L. REV. 255, 280 (2002) ("My intuition, which is shared by others and is consistent with experience, is that taxpayers often do not respond to fuzzy anti-abuse standards by making a greater effort at the margin to cloak transactions with apparent economic substance." (footnote omitted)). Moreover, there likely are good reasons that tax shelters are not designed with falsified economic substance. See, e.g., Bankman, supra note 220, at 28 ("It may be ... that the creation of a false paper trial [sic] is harder than it sounds.... It may be difficult to generate a plausible paper record to support a transaction that on its face offers only tax benefits. In addition, ... corporate purchasers generally will not purchase a shelter if it carries with it any significant business risk. It may be difficult for a promoter to sell a shelter to a corporate officer if it is accompanied by written materials that emphasize non-tax benefits and risks; difficult for a corporate officer who favors the shelter to sell the shelter to her colleagues if it is so accompanied by such written materials; and so on."); Canellos, supra note 217, at 56 ("The hard part [of tax shelter practice] is not finding the loophole .... Rather, it is cloaking the shelter in the mantle of a real transaction by incorporating the requisite economic return to satisfy a perceived 'economic substance' minimum threshold.").

(242.) See George Cooper, The Taming of the Shrewd: Identifying and Controlling Income Tax Avoidance, 85 COLUM. L. REV. 657, 675-76 (1985) (explaining that seller-provided ("internal") financing is a special problem, beyond that created by the leverage itself, and noting that "[i]nternal leverage ... combines the two critical ingredients of the alchemy, a tax-favored asset and a cooperative lender, in one package"); see also Theodore S. Sims, Debt, Accelerated Depreciation, and the Tale of a Teakettle: Tax Shelter Abuse Reconsidered, 42 UCLA L. REV. 263, 325 (1994). Professor Sims argued, in part:
 Conventional lenders ... expect to be repaid. Section 453 [which
 taxes installment sales over time], however, was used to inflate
 the nominal prices of depreciable assets--so that they were not
 "arm's length" prices in any conventional sense--using
 purchase-money debt that, tacitly at least, it was understood never
 really would be repaid. From a non-tax perspective, an investor in
 a tax shelter financed with purchase-money debt thus could afford
 to be indifferent to the investment's nominal "cost." In terms of
 taxes, however, the inflated price meant more deductions for
 depreciation. Sellers, on the other hand, were more than happy to
 sell for a price in excess of an asset's "real" value. As to the
 inflated portion of the price they were, moreover, equally content
 to accept the purchaser's promise to pay it in the future, however
 unlikely repayment might actually be.

Id. at 335 (footnotes omitted).

(243.) Cooper, supra note 242, at 675-76.

(244.) See supra text accompanying notes 190-92.

(245.) See Johnson, supra note 179, at 435.

(246.) Id.

(247.) See id.; I.R.C. [section][section] 301,302(b), (d) (2000).

(248.) See Johnson, supra note 179, at 435; see also I.R.C. [section] 318(a)(4) (2000); Treas. Reg. [section] 1.302-2(c) Ex. 2 (as amended in 2007).

(249.) See Johnson, supra note 179, at 435.

(250.) Id.

(251.) See id. at 434-35.

(252.) Soos, supra note 1, at 109-11.

(253.) See supra Part I.C.

(254.) See Andrew Walker, The Tax Regime for Individual Expatriates: Whom to Impress?, 58 TAX LAW. 555, 583 (2005) ("An important objective of any tax legislation is to promote among taxpayers a perception that obedience to the tax laws, as well as punishment for violation of the tax laws, is enforced equitably."); see also Lederman, supra note 165, at 1487 ("If auditing will remain low even after a planned increase in audits, the increase may not be sufficient to convince conditionally cooperative taxpayers that they will not be 'chumps' if they pay all of their taxes." (footnote omitted)).

Leandra Lederman, William W. Oliver Professor of Tax Law, Indiana University School of Law-Bloomington; The author is grateful to Brian Galle, Jeffrey Kahn, David Walker, and Larry Zelenak for helpful comments on a prior draft of this Article. For helpful suggestions, the author thanks Bill Black, Leslie Book, Diane Fahey, Lee Fennell, Victor Fleischer, Jonah Gelbach, Brant Hellwig, Anthony Infanti, Jonathan Klick, Sally Lederman, Sagit Leviner, and Charlene Luke; participants in the 2006 Critical Tax Conference, including Steven Bank, David Brennen, Charlotte Crane, Joseph Dodge, Lily Khang, Stephen Mazza, Michael McIntyre, Ajay Mehrotra, Diane Ring, and Kirk Stark; participants in the 2006 Midwest Law and Economics Association meeting; participants in the National Tax Association's 99th Annual Conference on Taxation; and participants in a 2007 faculty workshop at Florida State University College of Law. Stephen Brandenburg, Tanner Coulter, Kristen Fowler, and Chloe Pullman provided valuable research assistance.
COPYRIGHT 2007 Stanford Law School
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2007 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Lederman, Leandra
Publication:Stanford Law Review
Date:Dec 1, 2007
Previous Article:Doubtful duty: physicians' legal obligation to treat during an epidemic.
Next Article:Beyond the pro-consumption tax consensus.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters