Saturday, March 20, 2010

Altria # 1 - Frank Lyon and tax shelters (3/20/10)

I write today about a decision in a refund suit because it involves a jury verdict in a highly structured tax shelter. I have spent some words in this blog discussing concerns about the task a criminal jury faces in determining criminal liability for structured tax shelters, so I think this case is notable for this blog. In Altria Group, Inc. v. United States, 2010 U.S. Dist. LEXIS 25160 (S.D.N.Y. 2010), the jury was tasked to determine whether the substance over form or the economic substance doctrines (or both) applied to deny the taxpayer the tax benefits of ownership.

As introduction to the case and today's topic, I quote Judge Holwell's introductory summary of the case:
This case concerns federal income tax deductions plaintiff Altria Group Inc. and its subsidiary Phillip Morris Capital Corp. ("PMCC" or "Altria, " collectively with plaintiff) generated by leasing big pieces of infrastructure from tax-indifferent counterparties. These tax shelter transactions are known in the leasing industry as SILOs ("Sale-In-Lease-Out") and LILOs ("Lease-In-Lease-Out"). Following a two-week trial, the jury concluded on the facts presented to it that plaintiff's SILOs and LILOs lacked economic substance and failed to transfer tax ownership of the properties to Altria, thereby justifying disallowance by the IRS of certain deductions claimed by Altria. Several courts and another jury have reached similar conclusions in other jurisdictions. See BB&T Corp. v. United States, 523 F.3d 461 (4th Cir. 2008); AWG Leasing Trust v. United States, 592 F. Supp. 2d 953 (N.D. Ohio 2008); Fifth Third Bancorp & Subs. v. United States, 05 Civ. 350 (S.D. Ohio, April 18, 2008) (jury verdict); Wells Fargo & Co. v. United States, 91 Fed. Cl. 35 (Fed. Cl. 2010). But see Consolidated Edison Co. v. United States, 90 Fed. Cl. 228 (Fed. Cl. 2009).

The four transactions in the present case relate to Altria's acquisition of leasehold interests in the Long Island Railroad's primary maintenance facility, a Dutch wastewater treatment plant, and two power plants in Georgia and Florida. Each of the counterparties was indifferent to U.S. federal income tax, in the sense that the ability to depreciate or amortize the assets would not substantially affect the entities' tax liability. In each transaction, Altria immediately leased the asset back to its original owner using agreements with a number of unusual features, including complete defeasance (prepayment, in essence) of the lessee's rent and an owner's option to repurchase the asset. Altria then claimed depreciation, amortization, interest expense, and transaction expense deductions on its 1996 and 1997 corporate tax return based on its newly acquired assets, even though (i) its purchase money immediately was invested in securities that the nominal lessees could not access without providing substitute collateral, and (ii) the lessees could reacquire the assets without incurring any out-of-pocket costs.

From the perspective of the U.S. Treasury, these and similar transactions entered into by Altria created billions of dollars of tax-deferral benefits out of thin air. The Commissioner of Internal Revenue concluded that the transactions had no purpose, substance, or utility apart from their anticipated tax consequences and disallowed the deductions. Altria filed this suit, arguing, in substance, that because it complied with certain standards developed for traditional leveraged leasing transactions, it was entitled to the challenged deductions.

From June 23 to July 9, 2009, the Court held a jury trial to determine Altria's entitlement to the deductions. At the close of evidence, Altria moved for judgment as a matter of law, and the Court reserved decision on the motion. After the jury returned a verdict for the Government on all of Altria's claims, Altria renewed its motion and moved in the alternative for a new trial.

For the reasons that follow, the motions will be denied.
The opinion itself is quite excellent and long, so I will try to break it down in smaller pieces and devote separate blogs to each key component. In this blog, I address the role of Frank Lyon Co. v. United States, 435 U.S. 561 (1978). The Supreme Court's decision in Frank Lyon is a real piece of work, the legacy of which proves the truth of the aphorism that tax cases are too important to turn loose before the Supreme Court.

I am reminded of Charles Kingson's comment that few tax shelters are "shoddier" than the one the Supreme Court approved in Frank Lyon. Charles I. Kingson, How Tax Thinks, 37 Suffolk U. L. Rev. 1031, 1035 (2004); see also Bernard Wolfman, The Supreme Court in the Lyon's Den: A Failure of Judicial Process, 66 Cornell L. Rev. 1075, 1098 (1981) and Leandra Lederman, W(h)ither Economic Substance?, 95 Iowa L. Rev. 389, 409-416 (2010). Bottom-line, my cut on the Supreme Court decision in Frank Lyon is that the Supreme Court gave the taxpayer the victory but junked out the decision in order that few other taxpayers could really gain anything other than superficial wishful-thinking support from it and it would not hurt the fisc too much. In any event, Frank Lyon is there for taxpayers to give taxpayers some hope that their shoddy tax shelters too will work and to give the courts something to do to apply the decision in a case with a different set of facts.

Judge Holwell does as good a job as can be done with Frank Lyon. Indeed, he emphasizes a key facet of Frank Lyon that will take down many tax shelters without further ado. Judge Holwell repeatedly empahsizes that Frank Lyon involved -- and the Supreme Court prominently emphasized -- a zero sum game for treasury, regardless of who is the owner for tax purposes.  The Supreme Court said that either the nominal lessee or the nominal lessor in the sale-leaseback transaction is entitled to the depreciation and, so long as they are subject to equal tax rates, it truly is zero sum to the fisc. These are a few snippets from the Altria opinion to give the flavor:
[In Frank Lyon], [t]he Court also emphasized, in contrast to this case, that the transaction did not create any tax deductions, because Lyon and Worthen paid taxes at the same rate. Id. at 580. But see Wolfman, supra, at 1095-98.

Frank Lyon * * * did not involve a loss of revenue to the U.S. treasury.

Fourth, because each counterparty was indifferent to U.S. federal income tax, the transactions had the effect of creating tax benefits rather than transferring a tax benefit between taxpayers that paid comparable tax rates. Logically, one would not expect this to affect whether a nominal owner acquired a depreciable interest in leased property. See Bernard Wolfman, The Supreme Court in the Lyon's Den: A Failure of Judicial Process, 66 Cornell L. Rev. 1075, 1098 (1981). The Supreme Court, however, has expressly indicated that a transaction's effect on the U.S. treasury must inform a federal court's analysis of whether a transactional form chosen selected by a taxpayer should be respected for federal tax purposes. See Frank Lyon Co. v. United States, 435 U.S. 561, 580, 98 S. Ct. 1291, 55 L. Ed. 2d 550.
I mentioned specifically Judge Holwell's citation to Wolfman's article that, while intemperate in some of his comments, Wolfman does nail down the error in that the Supreme Court simply assumed away in Frank Lyon -- the parties were not subject to zero rates and hence the game was not zero sum for the fisc. The parties involved in the transaction -- Worthen Bank and Frank Lyon Co. -- were the winners because of the tax arbitrage they shared and the fisc was the loser. Still, Judge Holwell had to deal with the Supreme Court's assumption and found it dramatically absent in the case before him.

And, there is more in Frank Lyon, for the Supreme Court said the ownership issue was just an indeterminate mix of facts and circumstances. As Judge Holwell said:
In reaching this conclusion [in Frank Lyon], it is difficult to overstate the scope of the evidence the Supreme Court relied on. As a provocative commentary notes, the Court assayed at least twenty-five factors in concluding Lyon was entitled to the deductions. Michael H. Simonson, Determining Tax Ownership of Leased Property, 38 Tax Lawyer 1, 17 (1984); see also id. at 18 ("[T]he only enduring principles that can be derived from the majority opinion [in Frank Lyon] are that no single test is determinative of ownership and that each case must be decided on its own complex and numerous facts and circumstances."). The factors included Lyon's status as the party "whose capital was committed to the building," Frank Lyon, 435 U.S. at 581; Lyon's primary liability on the note to New York Life, Frank Lyon, 435 U.S. at 576; the fact that Lyon's return depended on the rent it would receive during the final ten years of its ground lease if Worthen put Lyon to the available extensions of the lease, id. at 579; the "regulatory realities" that prompted Worthen to structure the transaction in the manner it did, id. at 583; and the competitive bidding between firms that sought to invest in the building, id. at 576. The Court also emphasized, in contrast to this case, that the transaction did not create any tax deductions, because Lyon and Worthen paid taxes at the same rate. Id. at 580. But see Wolfman, supra, at 1095-98. It disclaimed any intention to identify a set of factors that would determine tax ownership in all cases, noting that the "significant and genuine attributes of the traditional lessor status . . . in any particular case will necessarily depend upon its facts." Frank Lyon, 435 U.S. at 584.
Realizing that Frank Lyon provided an indeterminate test that in most cases such as this had facets that cut both ways, the taxpayer tried to side step the troublesome facets by asserting that the real list was shorter -- as evidenced by an IRS Rev. Proc. And decisions of the Tax Court -- and that it passed the shorter list.  Judge Holwell said:
Given that Altria declined to seek an advance ruling as to the tax consequences of these transactions, and that it was free to file this action in the Tax Court, see 26 U.S.C. § 6214, there is a through-the-looking-glass quality to its argument that this Court should disregard Frank Lyon and limit its consideration to indicia of ownership noted in Revenue Procedure 75-21. But in any case, Altria's argument reflects an understanding of the relationship between the Supreme Court, lower Article III courts, and the Tax Court that is at the very least strange. This Court respects the Tax Court's views and for that reason charged the jury to consider each of the indicia of ownership relied on by Altria. (Compare Altria Mem. 42 with Charge to the Jury, at 33-34.) To say, however, that the Tax Court's decisions identify the exclusive criteria for determining which taxpayer is entitled to a depreciation deduction would be to ignore the essential holding of Frank Lyon, that whether a taxpayer possesses a depreciable interest in a leased asset must be determined through a fact-intensive analysis focused on the "substance and economic realities" of the challenged transaction. Frank Lyon, 435 U.S. at 582. Altria does not simply ask this Court to find that the Tax Court has "underruled" the Supreme Court, but to hold that as a result of the Tax Court's "clarification" of the law, the jury was precluded from giving weight to a number of material features of the transactions. In the Court's view, the law does not require such an anomalous result.

* * * *

Legally, Altria could justifiably rely on the Tax Court's decisions only if they considered transactions that are materially similar to the ones at issue. But that is hardly the case here. While the decisions Altria cites concededly consider leasing transactions, not one of them considers the tax consequences of the sale and leaseback of a critical piece of municipal infrastructure from a tax-indifferent counterparty in a transaction in which the "lessee's" rental obligations and purchase options are fully defeased, and the lessor is insulated from any meaningful economic risk of loss or potential for gain. Of course, that is unsurprising, for the LILO/SILO transaction structure's heyday was not until the 1990s. See Maxim Shvedov, CRS Report for Congress: Tax Implications of SILOs, QTEs, and Other Leasing Transactions with Tax-Exempt Entities 8 (Nov. 30, 2004). Altria's reliance argument also depends on the manifestly false factual premises that Altria is a naive investor that did not appreciate the novelty of the LILO/SILO structure or the tax risks inherent in the structure. But it did. For example, an agenda from PMCC's 1996 "closing meeting," dated January 21, 1997, notes that while the LILO structure "is now 'popular' as a tax shelter," it "is being scrutinized by the IRS." (GX 15, at 1.) The agenda continues to note that that the "[r]isk of IRS attack grows should PMCC increase its portfolio of such assets." (Id.) These hardly are the sort of comments one would expect from an innocent investor relying on "well-established and well-understood" principles of tax law.
I do note that the taxpayer may have been enamored of the purported holdings of the Tax Court precedent, but if so it is strange that the taxpayer did not take this case to the Tax Court. The reason is obvious.

Finally, I wonder why cases like this -- whether in a civil context or in a criminal context -- are submitted to a jury at all at least in the form it was here. The ownership issue in Altria is nothing more than a legal conclusion that is based on facts. Traditionally, the jury determines the facts and the court determines the law (specifically in this case, that legal conclusion that is required by the facts as found). If there really were facts in dispute in Altria (I suspect most of the facts were really undisputed and it was the proper characterization of the facts that was in issue), the jury as is common in civil cases could have simply been given special interrogatories as to the disputed facts and then the court could have made the legal conclusion required by the facts. Hence, in a highly structured transaction where the pertinent facts as in the legal documentation and the relationships created by the documentation, the conclusion as to whether those documents add up to ownership really ought to be by the judge and not by the jury. (I think Judge Holwell's opinion makes clear how he would make that determination, but the question I raise here is whether it should be submitted to a jury at all.)

If the reader perceives some bias in how I present Judge Holwell's consideration of Frank Lyon, I must admit that I am biased for reasons that would be distracting to develop here. But, I think I fairly set forth Judge Holwell's consideration of Frank Lyon. As to my bias, I believe the Supreme Court improvidently granted cert in Frank Lyon (I think there is a story there also) and then, rather than dismissing the case when it realized that the decision it could provide little meaningful guidance to the lower courts for future cases, crafted an opinion that articulated so many factors in the mix of resolving the case that the Court must have imagined / hoped that it would not create too much mischief in the future). The Court was, of course, wrong, but the Supreme Court is unlikely to wade in that pool to clean up the mischief.

Finally, just for fun, I drafted what I call a teaching indictment (my term) based on the facts of Frank Lyon and the allegations made in United States v. Stein, the KPMG criminal case where 13 of 19 defendants defendants were dismissed (United States v. Stein, 541 F.3d 130 (2d Cir. 2008)), 2 pled guilty, 3 were convicted and 1 was acquitted.  The draft indictment may be downloaded here.


  1. Excellent blog, informative, balanced and insightful. How does the teachings of Altria apply in the context of what I will characterize as "income shelters" (by contrast to the "loss generating shelters" presented in Altria)? The typical set up I have heard described at investment fund conferences involves a US taxpayer/hedge fund manager setting up a tax haven corporation, with offshore nominee offices (Mr. Obama referenced Ugland House on campaign), nominee directors supplied by offshore providers including, regrettably, law firms. The shares are held by the law firm (or its special purpose subsidiary) and the corporation is controlled by the US taxpayer through a consulting/investment management contract. The US taxpayer "defers" his fees and avoids paying enormous amounts of taxes. How does this work when the corporation does nothing substantive except act as a conduit for non-US investors to channel money to the US tax advisor. I keep hoping that Treasury would go after this pillage of our fisc but no joy so far.

  2. Actually, the Court in Frank Lyons said that the determination of tax ownership was a question of law, not a question of fact, for the court to determine. See Frank Lyons n. 16. It was accordingly reversible error for the trial court to send the case to the jury.

  3. Anonymous:

    1. Frank Lyon was tried to a judge, not to a jury.

    2. Footnote 16 to which you refer said: "The general characterization of a transaction for tax purposes is a question of law subject to review. The particular facts from which the characterization is to be made are not so subject. See American Realty Trust v. United States, 498 F.2d 1194, 1198 (CA4 1974)."

    3. If you just took footnote 16 at face, it is consistent with what the 8th Circuit did in holding for the Government. See Frank Lyon Company v. United States, 536 F.2d 746 (8th Cir. 1976).

    4. It is how the Supreme Court applied the footnote 16 concept that created the problem.

    5. BTW, I handled the briefing of the American Realty Trust case cited by the Supreme Court but had moved from the Appellate Section by the time of the oral argument. That case was a mess because the trial attorney did not timely order the trial transcript in a case tried to a jury (she thought it was a fact issue and the jury's verdict was conclusive, and did not anticipate that the Government wanted to argue that it was a legal issue which should not have been submitted to the jury; indeed at trial, the Government attorney loudly argued to the Judge that it was a fact issue that had to be submitted to the jury whereupon the trial judge said something to the effect that, well, little lady you will get your jury). On appeal, the Government argued that the issue was legal and should not have been submitted to the jury (despite the fact that the Government had insisted that it be submitted to a jury). As a result of not timely ordering the trial transcript, the Fourth Circuit did not have the trial transcript when it decided the case (it could have waited on it, but it chose not to). Because of the Government's error in not timely ordering the transcript, the Fourth Circuit essentially resolved all facts that could have been in the transcript in the taxpayer's favor. But ultimately that should have been irrelevant because the question was whether the documents reflected a lending transaction (via sale-leaseback) or a true purchase transaction. The problem in the case was that it was not just a situation that would have resulted in tax because of the loss of depreciation but would have resulted in disastrous consequences because the trust would not have distributed enough income (at least as I recall it). Accordingly, there were draconian consequences that, I think, caused the court to lean / tilt in the taxpayer's favor.

    Jack Townsend

  4. IMHO, the relevant legal issue in the leasing cases is whether the lessor has a significant residual interest or not. Everything else is pretty much irrelevant, including the existence of credit support or defeasance accounts. In the Altria case, the court dumped a bunch of irrelevant facts on the jury and told them to sort it out without giving any kind of guidance. So, in the middle of the great financial collapse of 2009, they took 20 minutes to decide that big bad cigarette company was not paying enough taxes. I would wager a large sum of money that not one person on that jury understood any of the testimony or arguments (N.b., none of them had a college degree). That is why the Court correctly (although in terse and inadequate language) said in Frank Lyons that the characterization of a lease was a question of law not fact - it's ridiculous to expect a lay jury to analyze tax law questions. Of course, there could be factual questions, such as, did the lessor have a significant residual interest? However, the judge did not submit any such questions to the jury.

  5. To Anonymous 7/15/11 1:56pm

    I agree that the bottom line issue should not be submitted to a jury. But the factual components could. I will exemplify in a civil case first, as was Altria. The jury could be asked whether the taxpayer had a nontax profit motive which is generally considered a requirement to sustain the civil tax benefit. That is a factual issue that the jury could decide. Similarly the jury could be asked to determine what, if any residual value there was based upon discounting or not as appropriate. Then, after the jury determines the required components -- the Eight Circuit in Lyon correctly analogized this as to a bundle of sticks -- the Court could then assess the components and determine whether the law characterizes them as ownership or some type of other transactions (usually a lending or lease transaction). The Supreme Court just screwed it up in Frank Lyon with a lot of nonsense that continues to obfuscate analysis. (In addition to the Eighth Circuit decision, check the dissenting opinion in the Supreme Court case.)

    There is, of course, an more difficult decision in criminal cases as to how to properly apply these concepts in a way that the defendant's life and liberty is not put to risk because of inherent jury confusion. I think the fix for that is for the judge (and the parties) to focus on the presence of the big lie and not how the legal issues should be resolved.

    Thanks for your comment.

    Jack Townsend


Please make sure that your comment is relevant to the blog entry. For those regular commenters on the blog who otherwise do not want to identify by name, readers would find it helpful if you would choose a unique anonymous indentifier other than just Anonymous. This will help readers identify other comments from a trusted source, so to speak.