Subscribe (Free) to
Daily or Weekly Newsletters
Post a Job

Featured Jobs

401(k) Distribution Manager

Nova 401(k) Associates

Nova 401(k) Associates logo

Retirement Plan Administrator

Nicholas Pension Consultants
(Remote / Corona CA / Rancho Cordova CA)

Nicholas Pension Consultants logo

Client Service Manager

United 401(k) Plans, Inc.

United 401(k) Plans, Inc. logo

Defined Benefit Calculation Specialist/Actuary

The Angell Pension Group, Inc.
(Remote / East Providence RI)

The Angell Pension Group, Inc. logo

Associate General Counsel

Central Pension Fund of the IUOE
(Washington DC)

Central Pension Fund of the IUOE logo

Compliance Analyst

(Remote / West Harrison NY)

Pentegra logo


Central Pension Fund of the IUOE
(Washington DC)

Central Pension Fund of the IUOE logo

Retirement Plan Administrator (TPA)

Retirement Plan Consultants

Retirement Plan Consultants logo

Retirement Plan Documents Specialist

Loren D. Stark Company

Loren D. Stark Company logo

Quality Assurance Manager

Bates & Company
(Remote / Winter Park FL)

Bates & Company logo

Retirement Plan Consultant

Carpenter Morse Group

Carpenter Morse Group logo

Retirement Plan Legal Specialist

(Remote / West Harrison NY)

Pentegra logo

Distribution \ Loan Clerk

Retirement, LLC
(Remote / Oklahoma City OK)

Retirement, LLC logo

Relationship Manager

Compass Retirement Consulting Group, Inc.
(Remote / NH)

Compass Retirement Consulting Group, Inc. logo

Retirement Plan Relationship Manager

ERISA Services, Inc.

ERISA Services, Inc. logo

Jr Retirement Plan Administrator/ Administrative Assistant

Hochheiser Deutsch & Co, Inc.
(Melville NY)

Hochheiser Deutsch & Co, Inc. logo

Executive Compensation Lawyer

Trucker Huss, A Professional Corporation
(Remote / San Francisco CA)

Trucker Huss, A Professional Corporation logo

View More Employee Benefits Jobs

Free Newsletters

“BenefitsLink continues to be the most valuable resource we have at the firm.”

-- An attorney subscriber

Mobile App image LinkedIn icon
Twitter icon
Facebook icon

Guest Article

To Foil Filching from Federal Fisc -- Test to Thwart (Some) Shelters

By Alvin D. Lurie
May 28, 2004

Is the verb "filch" too strong to describe a raid on the fisc by way of abusive tax shelters? In normal usage an "abuse" connotes a corrupt practice. If we can identify corruption, in reference to tax practice, as frustrating collection of one's tax liability by unlawful means, I submit we have fairly described a filching. If anything, the verb is not strong enough.

So, all we have to do now is decide what are unlawful means, in the context of abusive shelters. Unlawful, in this context, cannot mean everything that is against the tax law, for that would advance none of the objectives for applying special rules to tax shelters, not least the specific reporting and other requirements now imposed; so clearly a narrowing of the concept is essential.

Is it fraud that marks a shelter as abusive? Fraud may attend the utilization of a shelter, but it is not a necessary element. Besides, there is no need to struggle, as courts have, with the abuse concept where fraud is present, because it has its own markers and sanctions. What about like engaging in a sterile act that has the outward appearance of something that it is not, "skim milk masquerading as cream." Is that fraud or just sham? Either way it does not stand much chance of recognition for tax purposes, so again we need not bother to fix it with the label "abusive tax shelter" to deny it the sought after tax results. More pointedly, a sham act is not prerequisite to the engaging in shelter abuse. All the steps can be real (if not meaningful) without divesting a transaction of its abuse of the tax laws.

Now suppose the actions purported to occur actually do occur, but they are largely formalistic, lacking real significance, albeit not rising to the level of sham? For example, suppose they are taken as a means of appearing to deconstruct a unitary transaction into discrete parts that separately pass muster under a particular Code provision, but, if viewed in their entirety, would not achieve the desired tax result under any of several theories, most obviously the "step" doctrine, or because the intervening events lack economic substance. Now we are getting closer to an abusive tax shelter; but even here the tax law is capable of denying recognition to the transaction without resort to a shelter approach, as it has for many years, going back at least as early as the Gregory decision.

Statement of The Problem

What is the point of this exercise? Why labor so hard to deny the shelter designation and analysis to transactions, like those above, that involve self-evident tax avoidance? My answer is that if the term can mean almost anything, it means almost nothing. Its ubiquitous application deprives the concept of any utility, as courts come down all over the landscape on similar transactions, as practitioners become unable to advise clients where they are entering dangerous territory, and as taxpayers, and advisers and promoters lack the necessary guidelines for complying with the various rules and requirements that have developed in the past couple of years in respect of abusive tax shelters (such as the obligation to report participation in a shelter on one's tax returns, or to maintain lists of shelters on the part of promoters, advisers and others).

In the more exotic forms of tax planning that have emerged in the recent past, where the Code provisions appearing to bear on the transactions are deftly manipulated to lend support to an outcome that facially fits within the law but which produces a tax result that is obviously unreasonable and presumably unintended, the government and the courts strain to give the fisc its due, under various theories, sometimes successfully, sometimes not. Resort to those war horses of the shelter battles, Economic Substance and Business Purpose, with their amorphous criteria and susceptibility to idiosyncratic application, has proved as undependable as Humpty Dumpty's standard for word usage.

The Problem With BOSS

The consequence is that promoters have felt free to concoct transactions like BOSS -- a scheme developed to manufacture losses on the disposition of stock -- by exploiting rules relating to the disposition of encumbered property to shareholders, in order to create artificially high basis in the stock, which translates into tax losses for investors without corresponding economic loss. The BOSS transaction was devilishly ingenious, implicating the basis rule of Crane (331 U.S. 1, 1947) regarding mortgaged property, as bearing on the tax treatment of contributions and distributions of property between a corporation and its shareholder (a partnership), and between the partnership and its partners, which in turn affects the realization of capital loss by the partnership (and, derivatively, by its partners) on a deemed liquidation of the corporation triggered by a check-the-box election. Also implicated in the analysis argued for by the promoter is the issue of gain recognition to the corporation.

The object of BOSS was very simple, to create capital losses for an investor group for application against gains they had realized from transactions entirely unrelated to those taken in consummation of the strategy. The steps taken are almost as complicated as the legal analysis: an encumbrance is imposed colorably, but not meaningfully, on a portfolio of securities acquired with a loan, following which the encumbered securities are transferred successively among related parties consisting of the investor group, a special-purpose foreign entity formed by them, and their investment partnership. Each such transfer of the securities cum encumbrance purportedly modifies the parties responsible for payment of the loan, with allegedly great significance to the tax consequences to the parties resulting from the prescribed chain of events comprising the BOSS design. For example, the obligation was claimed to degrade the value of the SPE's distribution of the encumbered securities in the hands of its stockholder, but at the same time recognized as debt of the corporation to strip the corporation itself of value, and, yet again, considered as stockholder debt to give the investment partnership, qua stockholder, and its partners stepped up bases to support their claimed capital losses, the manufacture of which was the object of the whole elaborate scheme. In a final switch on the treatment of the debt, the promoter ignores it in order to contend that the investor group (the taxpayers using BOSS) will not have a constructive dividend on satisfaction by the corporation of the asset-encumbering liability for which the partnership and its partners were secondarily liable, notwithstanding that at an earlier point in the programmed sequence of events that debt was the foundation for giving the investors a stepped up basis.

One might be forgiven for viewing the whole scenario as more like a game of cubic 3-card-monte (a metaphor I presume I have invented) than a tax plan. The Service must have thought so too. "No way," it announced in Notice 99-59. It called it "a series of contrived steps [under which] taxpayers claim tax losses for capital outlays that they have in fact recovered." The reasoning was slight, but the message was clear. The promoter promptly stopped marketing the program.

It remains to be seen what will happen to the taxpayers who bought into the BOSS before the Service spoke, should they challenge their tax liabilities in the courts. For the reader who would like to know what I think, I refer you to my article in Tax Notes, "I Know Crane and BOSS Isn't Crane," March 27, 2000, page 1932. But, of course, what I think is of very limited value. What matters is what the courts think.

Conflict In The Courts

Because no generally accepted, definitive test of an abusive shelter exists, it is perilous to predict how courts will react. One finds mythic cases such as Compaq (277 F.3d 778, 5th Cir. 2001) where the taxpayer wins, and ACM (157 F.3d 231, 3d Cir. 1998) where the taxpayer loses -- each involving a program enabling taxable income arising from a business or investment activity of the taxpayer to be conveniently offset by a deduction constructed out of an extraneous transaction ( i.e., not connected with the source of the income) entered into for the sole purpose of establishing the deduction. One is hard pressed to find the principle or distinguishing element that accounts for the different results, other than that different judges with different gestaltic responses were sitting on their respective benches.

Such outcomes are inevitable given the absence of general agreement as to the defining principle and critical facts that separate an abusive shelter from all the rest of tax planning. Just having a broad concept like Business Purpose or Economic Substance is not helpful in achieving consistency, absent concrete indicia generally recognized as establishing the requisite "purpose" or "substance." Realistically, it is doubtful that such a unifying approach can be fashioned to deal with all the varied forms of abuse that have been -- and are yet to be -- designed. In fact, the elasticity of these concepts for application in such variegated settings is their strength; but at the same time it is their weakness.

A Test To Catch Abuse

It would thus be appealing to be able to fasten on a core element typically found in certain of the most abusive of schemes that have surfaced, and to formulate an objective set of markers that point unfailingly to its presence, a litmus test if you will. It is the burden of this paper that such a core element is readily observable and that such a test exists, as I will momentarily demonstrate. But, first, let us consider what such a test would accomplish.

Flunking the test -- by which I mean that the core element is found to be present -- would create, at the least, a strong presumption of abuse. Passing the test, on the other hand, would not free the strategy of further scrutiny, because not all shelters would possess the incriminating markers that I shall identify; and its challengers would have to resort to traditional doctrines such as Sham, and Step Transactions and Business Purpose. The difference is that shelters that flunk the test could be taken down without the need for such analysis. It would be no defense to show that the activity giving rise to the favorable tax outcome that is challenged has a potential for making a modest profit, or that it even has yielded such a profit; although it should be possible to build into the test certain safe harbors that would relieve from the harsh verdict of the test taxpayers whose business or investment activities have certain redeeming characteristics.

What is that core element that can be said to characterize many of the shelters that have been found to be most abusive? Look at the very transactions involved in the cases I have already cited and in the BOSS program. It is the existence of a gratuitous structure, separate and distinct from the taxpayer's business or investment activities (let me call the latter "basic activities," for convenience of reference), having no demonstrable purpose other than the saving of taxes attributable to the basic activities. Thus, it is the conjunction of two separate activities, one, a basic business or investment activity engaged in for legitimate purposes and producing the normal tax consequences flowing from such activity, and, two, a separate business or investment activity entered into for the purpose of avoiding the tax liability or other unwanted tax consequences (e.g., expiring loss carry forwards) associated with the former activity. Let's call the entire arrangement a linked tax construct ("LTC").

In ACM (the "M" standing for Merrill Lynch) and several cases of the same design, culminating most recently in Boca Investorings (314 F.3d 635, D.C. Cir. 2003), the taxpayer had incurred substantial capital gains in the course of its ordinary business, and adopted a program designed to look like a contingent instalment sale program that had the ostensible effect of offsetting that capital gain. The gratuitous structure was a special-purpose partnership formed with a tax-indifferent foreign bank for the purpose of selling securities and, then, deconstructing the sale proceeds into artificial gain and loss components, with the "gains" shunted to the foreign partner and the "losses" to the domestic corporation (the one utilizing the shelter). There were, in fact, no material losses, and, as held by the D.C. Circuit in Boca, no bona fide "partnership."

Compaq was a so-called dividend-stripping transaction, quite different in design from ACM, but similar in its employment of a gratuitous transaction to offset capital gain realized on the sale of stock of a corporation in a transaction entirely unrelated to the dividend strip, under which the taxpayer bought and held ADRs on foreign shares of stock for one critical day, when the stock went from cum-dividend to ex-dividend. The sole purpose of the procedure was to obtain the related foreign tax credits and capital loss manufactured by the sale ex-dividend, for application against the capital gain realized on the original sale of stock in the basic transaction.

BOSS was yet another variation on the same theme, designed by PriceWaterhouse Coopers to function as an elegant structure superimposed on and entirely unrelated to the capital gains that the taxpayers had previously realized, prompting them to enter into the elaborate construct involving a foreign entity, a domestic corporation owned by a partnership that made a check-the-box election to change its tax treatment, and the passing of encumbered property among the parties to achieve basis shifting.

Similarly, tax erasing strategies were devised by KPMG carrying the names BLIPS, FLIP and OPIS, and by Ernst & Young under names like COSS and COBRA, possessing the same dual-structure characteristics, although each vastly different in design. COBRA, as described in a widely reported lawsuit brought by a Mr. Camferdam and others against E&Y and the opinion-writing law firms, was bought from the accountants because the plaintiffs faced the misfortune of having made a $70 million gain on the sale of their business; so COBRA was no less seductive than the original snake, because it promised to wipe out that otherwise taxable gain. It merely required the plaintiffs to simultaneously buy and sell offsetting long and short currency options, transfer them to an S corporation formed specially for the currency transaction, which supposedly resulted in a greatly inflated basis to the S corporation by counting the amount paid for the long option but not the essentially identical amount received on sale of the short position. Hence, the promoter claimed, and the opinions so opined, that a sale of the assets of the corporation produced a capital loss equal to the amount paid for the long option, which, as stockholders of the S corporation, the plaintiffs claimed on their 1040s. Here again the LTC is the culprit.

Obviously, the signature linked structure was not the exclusive property of any one of the tax powerhouses, and was not confined to just the major accounting firms, but also included among its advocates -- sometimes as promoters themselves, sometimes as the signatories on "should" opinions -- some of the largest law firms, as suits by disgruntled investors reported in the public press have revealed. In all these cases and others that could be named, in one form or another the taxpayer buys (and dearly pays for) a one-shot, special purpose tactic for the predominant (usually exclusive) purpose of draining the normal tax incidence (alright, the tax consequences that one would expect) from the basic business or investment activity.

Defining A Linked Tax Construct

My proposal is to treat these linked tax constructs as per se abusive shelters. Expressed in formulaic terms, the requisite elements of an LTC are: (1) employment of a one-shot tax construct ("TC"); (2) completely unrelated to taxpayer's business or investment activity ("B/IA"); (3) the principal purpose of which is to address unwanted tax incidence deriving from B/IA; and (4) designed to eliminate or ameliorate the effect of the B/IA tax posture. The presence of all four elements must be found, and they can be identified objectively without the need to ascertain Congressional intent, or to establish taxpayer intent or purpose, or to resort to spongy doctrines like Economic Substance or Business Purpose to determine whether the TC had economic substance or profit potential standing alone. Thus, it is no defense that the TC, as a stand-alone activity, might have passed muster, because when used as an integral part of an LTC arrangement, it is a very different structure.

One could identify another common characteristic possessed by each of the above cited programs and many others that have surfaced: each was presented to the taxpayer by a promoter as an essentially off-the-shelf product created for sale to a wide market, and actually sold to thousands of reluctant taxpayers for fees running to the billions ($50 million for BLIPS alone, reportedly). Professor George Yin, chief of staff of the Joint Committee on Taxation, has recently invited comment from the bar on his suggestion that such "wholesaling" (my term, not his) be treated as a badge of abuse in cases of shelter designs that thread their way through tax statutes in ways that produce tax avoidance that Congress presumably had not anticipated (see "Thoughts on Tax Shelters," Tax Notes, Feb. 16, 2004, p. 931). He did not relate it to linked structures, of course; and while his suggestion has much to commend it, I would not favor including such a limiting factor in the LTC test, since the inherently abusive nature of the linked structure is no less so because hand-crafted by the taxpayer's own advisers. A sounder limitation on the LTC test might be to exclude its application where the tax benefit is below a certain significant threshold number.

Properly seen, LTC is but a genus of tax shelters, although probably accounting for the largest number of potentially abusive shelters and the largest leakage of tax revenues. There are, of course, many shelters not having its distinctive characteristics that consequently would not be reached by the LTC test. Professor Yin, in his remarks noted above, illustrates his concerns by discussing what is known by the name SC2, a scheme for introducing a charity into a temporary ownership position in an S corporation in such a way as to siphon off a sizeable chunk of reportable profits into the tax account of the charity, greatly in excess of the actual pre-programmed, beneficial share of the charity. His comments anticipated by several weeks a Service notice (Not. 2004-30) designating this type of scheme as a listed transaction. There are many vulnerable features of this obviously abusive arrangement -- the foregoing Notice enumerates several -- but a linked superstructure would not appear to be one of them. There is only a single business structure, and the unwanted taxable profits all flow from it. Assuming one even recognizes the interest of the charity as bona fide, the tax law has anti-abuse doctrines for reassigning the profits to their proper accounts, without trying to fit an SC2-type strategy within the LTC test and thereby undercutting its specificity, which is the principal value of the test.

How LTC Test Would Work

The steps of the LTC test are:

  1. Identify taxpayer's B/IA
  2. Ascertain that separate, one-shot TC is utilized by taxpayer
  3. Find that TC not connected with or necessary to conducting the B/IA
  4. Determine tax incidence of conducting the B/IA as if no TC exists
  5. Calculate tax effects of utilization of TC on taxpayer's tax posture, (i) separate from the B/IA, and (ii) in combination with conducting the B/IA
  6. Determine whether tax consequences of utilizing TC affects tax incidence pertaining to B/IA more than de minimis

If answer to Step 6 is "yes," presence of LTC has been established.

Step 3 is critical to the proper working of the test. The conduct of one's business in the most tax-efficent manner, even if otherwise abusive, is not an LTC. For this LTC type of shelter, there must be a separate tax-motivated activity not an integral part of or in any way necessary to the conduct of that business, apart from saving taxes (possibly limited to federal income taxes), to bring the arrangement within the ambit of the test. On the other hand, the mere maintenance of two distinct business or investment activities, one profitable, one losing that offsets tax liabilities attributable to the former, are not intended to come within the LTC penumbra. There would have to be criteria for identifying the proscribed linkage, perhaps based on the tenure of the activity producing the offsetting tax benefits, or the circumstances surrounding its adoption. Note that Step 2 of the test limits its application to a "one-shot" TC, which is designed to prevent flunking the test where the suspect TC is a continuing activity. That is meant to restrict it to constructs designed for a single tax benefit, as distinguished from a continuing business or investment activity; since that is how these strategies are typically designed. But that limitation might inappropriately taint a single investment that has independent validity on its own terms, or conversely might permit ready circumvention of the intent of the LTC approach. That suggests the necessity of a "facts and circumstances" feature for dealing with arrangements that test the limits of the concept.

We can leave it to another day to determine whether the tagging of something as an LTC results in automatic denial of the sought-after tax effects, or merely an evidentiary shifting of the burden, for purposes of adjudication or administrative disposition on audit. Depending on how that issue is resolved, it might be appropriate to change my characterization of LTCs as per se shelters to presumptive shelters. In any case, the bias should be heavily against exceptions to application of the test, lest the benefits of the fixed-rule approach become mired in exceptions and court-developed rationales for avoiding the "judgment" of the test, putting us back to where we are today.

Relation to Listed Transactions

What is the relationship of the LTC concept to the ever-growing "listed transaction" technique? Listed transactions are described in the shelter regulations as "a transaction that is the same as or substantially similar to one of the types of transactions that the Internal Revenue Service has determined to be a tax avoidance transaction and identified by notice, regulation, or other form of published guidance as a listed transaction." Treas. Reg. sec. 1.6011-4(b)(2). These are individual, fact-specific forms of transaction. LTC is not that, but rather is meant to encompass transactions of greatly varied fact patterns and legal rationales (as claimed by their respective promoters), that have in common the linked structure explained in this piece.

That is the power of the approach. Even as there are many ways to skin a cat, LTC should be able to reach all ways that achieve the skinning distinctive to its genus, without the need to particularize varying fact patterns and designs. In this respect an LTC classification is similar to several of the categories besides listed transactions that are treated in the shelter regulations as "reportable transactions," e.g., confidential transactions, transactions with contractual protection and transactions with book-tax differences, that apply broadly without regard to the specific features of any given strategy. Treas. Reg. sec. 1.6011-4(b)(1). Thus, LTC has the advantage over the listed transaction strategy of obviating the need for the government to keep playing catch-up with new variations on an old theme; while it is, at the same time, specific enough in its target to preclude overreaching by the government in application of the test.

In time LTC could be added as a new category of reportable transactions, but first it would be necessary for the Service to establish clear guides to the what does and does not constitute the requisite linkage between the B/IA and the TC, so that taxpayers can discharge their reporting obligations with relative certainty.


Would adoption of the modest proposal of this piece staunch the flow of abusive shelters? I am neither so prescient nor so presumptuous as to claim that. But it seems to me that it would provide a ready means of separating the wheat from some of the chaff. It has the virtue of being easily understood and capable of being consistently applied by bench, bar, promoter, taxpayer and tax return preparer.

It probably need not be re-emphasized that the LTC approach is not meant as the definition of an "abusive tax shelter," but I will state its reach again. LTCs are only to be seen as one type, albeit a large grouping.

It's a scheme for testing for one of the more corrosive of the classes of abusive tax shelters -- not a particular type of shelter, but a broad range of shelters displaying a common bad gene, that permits tagging a strategy as a presumptive, if not per se, shelter, without the need to apply such amorphous, eye-of-the-beholder concepts as business purpose and economic substance. Strategies passing the test (that is, not within its reach) can still be treated as abusive under existing principles, or under whatever rules emerge from the current attention to the abuse problem in legislative and regulatory circles.

An LTC test will not answer all the questions that conscientious taxpayers ask their accountants to the effect, "Is this a good tax shelter?" But it will provide much greater certainty (that is the key) for professionals and their clients -- and the courts that will judge them -- in spotting the bad ones. It can have the collateral prophylactic benefit of discouraging would-be users from buying into (and being harmed by) such schemes. That alone commends its serious consideration.

Copyright 2004, A.D. Lurie
Alvin Lurie has spent many years as a practicing pension attorney, and was appointed as the first person to administer the IRS' ERISA program in the National office in Washington. He is back in practice and can be contacted at Alvin D. Lurie P.C. in New Rochelle, New York, telephone (914) 235-6575.

BenefitsLink is an independent national employee benefits information provider, not formally affiliated with the firms and companies who kindly provide much of the content and advertisements published on this Web site, including the article shown above.
© 2023, Inc.