Focus on Tax Controversy


IRS Updates Administrative Appeals Process for Cases Docketed in Tax Court

Joshua M. Ellenberg

In Notice 2015-72, the Internal Revenue Service (IRS) provided a proposed revenue procedure to update Rev. Proc. 87-24, 1987-1 C.B. 720, which describes the procedures for handling docketed cases in furtherance of the IRS Office of Appeals’ mission to resolve tax controversies without litigation in a manner that is fair and impartial to both the government and the taxpayer. If finalized, this proposed revenue procedure would supersede Rev. Proc. 87-24.

The IRS noted that since the 1987 issuance of Rev. Proc. 87-24, the IRS has been reorganized several times, the volume of litigation in the Tax Court has increased, and the IRS has adopted new policies and procedures to more efficiently manage its workload. Therefore, the notice states, the old revenue procedure should be updated to more accurately reflect the procedures utilized in managing the flow of docketed cases between the Office of Appeals (Appeals) and the Office of Chief Counsel (Counsel), and also to ensure that docketed cases are handled consistently throughout the United States.

Although the notice asserts that the proposed update “is not intended to materially modify the current practice of referring docketed cases to Appeals for settlement currently utilized in the vast majority of cases,” it does in fact contain some significant changes from Rev. Proc. 87-24.

For example, the updated Rev. Proc. stresses that, except in rare circumstances, Counsel will refer docketed cases to Appeals for settlement consideration. Under the old Rev. Proc., it was not entirely clear that Counsel was required to refer docketed cases back to Appeals. Although in practice Counsel often did refer docketed cases to Appeals, the language of the new Rev. Proc. should give more leverage to taxpayers arguing for a referral back to Appeals.

Rev. Proc. 87-24 also explicitly states that for cases involving deficiencies over $10,000, Appeals must promptly return the case to Counsel when no progress is being made towards settling the case. The proposed update omits this statement, perhaps indicating a willingness to give Appeals more leeway in deciding when to send a case back to Counsel.

Further, the proposed Rev. Proc. addresses the frequent occurrence of cases in which Appeals is forced to issue a notice of deficiency or make a determination without having fully considered the issues because of an impending expiration of the statute of limitations on assessment. The notice states that if Appeals requests that the case be returned to it for full consideration once docketed, it will be treated as if Appeals did not issue the notice of deficiency or make the determination.

An additional, somewhat puzzling, revision in the updated Rev. Proc. is sure to be a source of many comments. The notice explains that those “rare circumstances” in which Counsel will not refer docketed cases for settlement include instances where an “issue has been designated for litigation by Counsel” and where “Division Counsel or a higher level Counsel official determines that referral is not in the interest of sound tax administration.” This provision seems to contradict Rev. Proc. 87-24, which only allowed Counsel to make such a decision in consultation with Appeals. The IRS will go through a long process, with input from Appeals and opportunities for the taxpayer to argue against designation, before it designates an issue for litigation, but ultimately IRS chief counsel has final say on whether to designate the case for litigation. If the proposed Rev. Proc. in fact intends to take Appeals—an independent organization—out of this process and give sole discretion to the IRS in deciding when to refer cases to Appeals, taxpayers should be concerned.

Another change in the proposed Rev. Proc., however, should be quite welcome to taxpayers. In an effort to preserve Appeals’ independence, the notice clarifies that even in docketed cases Appeals may exclude Counsel from settlement conferences with the taxpayer if, after considering the views of both Counsel and the taxpayer, Appeals determines Counsel’s participation in the settlement conference will not further settlement of the case. The proposed Rev. Proc. also addresses the coordination between Appeals and Counsel when a taxpayer raises an issue for the first time while the docketed case is with Appeals for settlement consideration.

Lastly, the proposed Rev. Proc. removes the prior exclusion for cases governed by rulings by the National Office in employee plans and exempt organizations to reflect recent organization changes in the Tax Exempt and Government Entities Division.

Tax Court Order Reaffirms that State of Mind Defense Waives Attorney-Client Privilege

Joshua M. Ellenberg

In Eaton Corp. v. Commissioner, No. 5576-12 (2015), Special Trial Judge Daniel A. Guy, Jr. reinforced the U.S. Tax Court’s controversial opinion from AD Investment 2000 Fund LLC v. Commissioner, 142 T.C. 248 (2014), which held that a taxpayer implicitly waives its attorney-client privilege simply by asserting a reasonable belief defense to accuracy-related penalties.

AD Investment

In AD Investment, the Tax Court left many tax practitioners surprised and dismayed when it held that taxpayers asserting good-faith defenses to accuracy-related penalties had waived the attorney-client privilege by putting their state of mind at issue. The case involved two partnerships engaging in what the Internal Revenue Service (IRS) described as Son-of-Boss tax shelter transactions designed to create artificial tax losses. Based on these transactions, the IRS adjusted partnership items and determined that various section 6662 accuracy-related penalties should apply to any resulting underpayments of tax.

The partnerships defended against the penalties by claiming two commonly pled affirmative defenses under section 6664: the reasonable belief defense and the reasonable cause/good faith defense. Although the taxpayers had received six opinion letters regarding the transaction from the law firm Brown & Wood LLP, they did not claim that their “reasonable belief” and “good faith” centered on that professional advice. Instead, they stated that they had relied on their own self-determination, and thus they claimed that those tax opinions were not relevant to their defenses and should therefore be protected by the attorney-client privilege.

In an opinion by Judge James S. Halpern that stunned the tax bar, the Tax Court held that the taxpayers had implicitly waived the attorney-client privilege by raising the reasonable belief defense. Judge Halpern wrote that the taxpayers’ defense “placed the partnerships’ legal knowledge, understanding, and beliefs into contention, and those are topics upon which the opinions may bear.” He further opined that if the partnerships had relied on the legal knowledge of their lawyers in forming their reasonable and good faith belief that the tax treatment of the items in question was more likely than not the proper treatment, then “it is only fair that respondent be allowed to inquire into the bases of that person’s knowledge, understanding, and beliefs including the opinions (if considered).” Thus, Judge Halpern ordered production of the once privileged documents.

Eaton Corp.

Many tax lawyers were cautiously optimistic that the Tax Court’s ruling would be limited to the area of tax shelters, where courts may be less inclined to allow the attorney-client privilege. These hopes were dashed by the order in Eaton Corp., which revolved around the IRS’s motion to compel the production of certain documents held by the taxpayer. These documents comprised internal e-mails, memos and data compilations exchanged between the taxpayer and the taxpayer’s legal counsel at Mayer Brown LLP, and tax practitioners at PricewaterhouseCoopers and KPMG. The documents were generated in support of the taxpayer’s negotiation of an advanced pricing agreement (APA) with the IRS.

Special Trial Judge Daniel A. Guy, Jr. first rejected the IRS’s argument that these documents weren’t at all privileged. Finding that the documents under review were prepared because of the prospect of litigation and that the communications in question were intended to be confidential, the court concluded that the documents were theoretically protected from discovery based on the attorney-client and tax practitioner privileges, as well as under the work product doctrine.

Next, the judge turned to the question of whether the taxpayer had implicitly waived those privileges by asserting the reasonable cause/good faith defense. The taxpayer tried to differentiate its facts from those of AD Investment by arguing that the AD Investment court had only properly analyzed the reasonable belief defense and not the reasonable cause/good faith defense that was being utilized in the taxpayer’s case. Here, however, the judge agreed with the IRS that AD Investment was controlling, concluding that the taxpayer’s “reasonable cause/good faith defense puts into contention the subjective intent and state of mind of those who acted for [the taxpayer] and [taxpayer]’s good-faith efforts to comply with the tax law.” Thus, the judge stated, “it would be unfair to deprive [the IRS] of knowledge of the legal and tax advice that [taxpayer] received in the course of requesting and negotiating the APA.” Accordingly, the court held that by raising the reasonable cause/good faith defense, the taxpayer had waived its right to proclaim the documents privileged.


It is disheartening that the protections offered by the attorney client privilege are eroding. It remains to be seen whether and to what extent the full Tax Court will continue to apply AD Investment. In the meantime, taxpayers should weigh their options and closely evaluate alternatives before invoking a 6664 state of mind defense to an accuracy-related penalty.

U.S. Tax Court Upholds Favorable Definition of Insurance

Kristen E. Hazel

On September 21, 2015, the U.S. Tax Court released its decision in R.V.I. Guaranty Co. v. Commissioner (RVI), 145 T.C. No. 9, and ruled that the taxpayer’s residual value insurance contracts constituted insurance for U.S. federal income tax purposes. As a result, the taxpayer was able to more closely match premium inclusions with loss deductions. This case represents a third victory for taxpayers with respect to the definition of insurance for tax purposes, following Rent-A-Center v. Commissioner, 142 T.C. 1 (2014), and Securitas Holdings, Inc. v. Commissioner, T.C. Memo. 2014-225.


R.V.I. Guaranty Co. Ltd. (RVIG), is a Bermuda corporation registered and regulated as an insurance company under the Bermuda Insurance Act of 1978. As an electing domestic taxpayer under Internal Revenue Code (Code) § 953(d), RVIG is the common parent of an affiliated group of corporations that includes R.V.I. American Insurance Company (RVIA), a property and casualty insurance company domiciled in Connecticut. RVIG and RVIA are together referred to as the “taxpayer.”

During the years at issue, the taxpayer sold residual value insurance. Residual value insurance policies are generally offered to leasing companies, manufacturers and financial institutions, and cover assets such as passenger vehicles, commercial real estate and commercial equipment.

The policies operate to protect the insured against the risk that the value of the insured asset at the end of the lease term will be lower than the expected value. For example, an insured may be a vehicle leasing company. In setting the periodic lease payments, the insured must estimate the residual value of the vehicles on termination of the lease. A residual value insurer covers against the risk that the actual value of the vehicles upon termination of the lease will be lower than the expected value.

During RVIG’s 2006 audit, the Internal Revenue Service (IRS) concluded that the policies were not “insurance” for U.S. federal income tax purposes, based largely on the IRS’s determination that the policyholders were purchasing protection against investment or business risk rather than insurance risk. The IRS assessed a deficiency of more than $55 million, and the taxpayer timely petitioned the Tax Court for redetermination of this deficiency.

U.S. Federal Tax Definition of Insurance

The fundamental issue addressed by the Tax Court was whether the residual value insurance contracts protected the insured against an investment risk or an insurance risk. Neither the Code nor the Treasury Regulations define the term “insurance,” but over the years a body of law has developed, and the following guiding principles have emerged:

  • Insurance involves both risk shifting and risk distribution. The risk of loss must shift from the insured to the insurer, and the insurer must pool multiple risks of multiple insureds in order to diversify its exposure; this is known as the “law of large numbers.”
  • The transaction must constitute insurance in its commonly accepted sense.
  • Especially relevant to the R.V.I. decision, the risk transferred must be an “insurance risk.”

Against that framework, and notwithstanding the fact that commercial insurance companies have offered residual value insurance for more than 80 years, the IRS concluded that the contracts protected the insureds against investment risk and thus were not insurance for tax purposes.


At trial, the taxpayer’s experts concluded that the policies covered an insurance risk, much like mortgage guaranty insurance. The taxpayer’s experts also concluded that the risks were distributed in the same way as any other property and casualty carrier, that the taxpayer was subject to underwriting risk and that the risk was transferred.

The IRS countered with three experts who concluded that the policies covered a speculative risk, much like a stock investment. While the IRS’s experts acknowledged that the risks were distributed, they also concluded that the risks were highly correlated, thus challenging the notion that the aggregate risk was truly distributed.

After admitting to a methodological error, the IRS’s experts seemingly conceded that the taxpayer had a significant risk of loss. Nevertheless, the experts concluded that the policies were not typical insurance policies, i.e., not insurance in the commonly accepted sense, because they did not insure against a fortuitous event and the insurer did not face any timing risk.

The Tax Court dispensed with the IRS’s contentions regarding risk shifting and risk distribution by concluding that the taxpayer’s actual loss experience demonstrated that it bore a significant risk of loss (thus, risk had been shifted) and that there was meaningful risk distribution. There were more than two million separate risk units of varying types (passenger vehicles, real estate properties and commercial equipment), and the risk units were distributed over varying lease terms.

While the court acknowledged that the risks could be correlated to, for example, a recession, the court noted that the diversification achieved within the asset pool and lease terms mitigated any systemic risk. Moreover, the court observed that many insurers face systemically correlated risks. The taxpayer’s business model was not materially different than the business model of those insurers.

Investment or Insurance Risk

The court then rejected the IRS’s contention that the policies covered an uninsurable “investment risk.” The court considered first whether the contracts were insurance in the commonly understood sense of the word.

It framed its analysis by considering five factors:

  • Whether the insurer was organized and operated as an insurance company by the states in which it conducted business (taxpayer was)
  • Whether the insurer was adequately capitalized (taxpayer was)
  • Whether the insurance policies were valid and binding (taxpayer’s contracts were)
  • Whether the premiums were reasonable in relation to the risk of loss (premiums were negotiated at arm’s length between taxpayer and its insureds)
  • Whether premiums were duly paid and loss claims were duly satisfied (when losses occurred, insureds filed claims and taxpayer paid those claims)

Even though the taxpayer readily met the five requirements, the IRS concluded that the policies did not qualify as insurance because they differed from policies with which most people are familiar. The IRS noted that the policies did not pay on the occurrence of a “fortuitous event,” such as a car crash. Rather, the policies paid, if at all, at the end of the lease term, which was not random or fortuitous. The court, however, held that losses under the policies were caused by fortuitous events outside the control of the taxpayer. The fact that a loss must persist to the end of the term of the lease does not make the events that cause the loss (e.g., recession, interest rate spikes, bank failures) any less fortuitous, the court stated. Thus, the Tax Court concluded that the contracts were insurance in the commonly accepted sense.

The court next considered whether the contracts covered insurance risk. The court acknowledged that there was little guidance with respect to the difference between “insurance risk” and “investment risk.” The court determined that the policies involved insurance risk from the taxpayer’s perspective.

The taxpayer’s business model depended upon more than investment returns; its model depended on the ability of its underwriters to price the risks to derive a sufficient pool of premiums to cover the aggregate losses. This is the same pricing model used by insurance companies generally.

The court also determined that the policies were insurance risk from the perspective of the insureds. The court first looked to state law and noted that New York and Connecticut had defined residual value policies as a form of “insurance” since 1989, and in 1991 the Washington Supreme Court had reached the same conclusion.

The taxpayer’s regulators and external auditors uniformly reached the same conclusion. The court then scrutinized the nature of the risk itself. The court declined to accept the narrow definition of risk offered by the IRS and instead looked to the more practical guidance offered by the taxpayer.

The insured simply has to shift to the insurer the risk from a “hazard,” a “specific contingency,” or some “direct or indirect economic loss.” The residual value insurance contracts offered by the taxpayer did just that. The insured shifted to the taxpayer the risk that the covered property would decline in value. The types of events that resulted in a loss under the policies closely resembled the losses under, for example, mortgage guaranty insurance—a product long accepted as insurance.

Having concluded that the policies had the hallmark features of insurance (risk shifting, risk distribution, commonly accepted notions of insurance, and insurance risk) the court determined that the policies were insurance for U.S. federal income tax purposes.

Working for Relief from Retroactivity

Retroactivity is an endemic problem in the state tax world. The past year has seen retroactive repeal of the Multistate Tax Compact (MTC) in Michigan, as well as significant retroactivity issues in New York, New Jersey and Virginia. Relief appeared to be on the way until the Supreme Court of the United States denied certiorari in a Washington estate tax case, Hambleton v. Washington, on October 13, 2015. The Supreme Court’s decision came just two weeks after the Michigan Court of Appeals upheld a retroactive period of almost seven years.

The Hambleton petition urged the Supreme Court to take the case in order to resolve the uncertainty of “how long is too long” when it comes to retroactive taxes, citing multiple examples of past and ongoing litigation in which lower courts have taken divergent approaches to the length of permissible retroactivity. For example, the petition cited the ongoing litigation in Michigan over the MTC’s apportionment election. In July 2014, in International Business Machines Corp. v. Michigan Department of Treasury, the Michigan Supreme Court held that IBM could apportion its income using the so-called “MTC election,” which allowed a taxpayer to use a three-factor formula consisting of property, payroll and receipts to apportion income, rather than the state’s standard formula. 852 N.W.2d 865 (Mich. 2014). In September 2014, however, the Michigan legislature retroactively repealed the MTC election and effectively overturned the IBM decision. Fifty taxpayers challenged the retroactive repeal, and those cases were consolidated.

On September 29, 2015, the Michigan Court of Appeals upheld the retroactive repeal of the MTC election in the consolidated cases. Gillette Commercial Operations N. Am. & Subsidiaries v. Dep’t of Treasury, et al, Dkt. No. 325258 (Mich. Ct. Claims, Sep. 29, 2015). While the case included several state and federal constitutional and statutory issues, this article will focus on the due process clause.

The due process clause (theoretically) prohibits retroactive laws, because persons must be able to know what the law is, and retroactive law changes prevent a person from having that knowledge. The Supreme Court of the United States’ primary case regarding when due process prohibits a retroactive law is U.S. v. Carlton, 512 U.S. 26 (1994). In Carlton, the Supreme Court established a two-part test to determine whether the retroactive effect of a law is allowed under the due process clause. First, the legislature’s act must be neither arbitrary nor illegitimate. Second, the legislature must act promptly and only enact a “modest” period of retroactivity.

In Gillette, the Michigan Court of Appeals determined that a six-and-a-half-year period of retroactivity was modest. It is not clear why this length of time was deemed modest, but many other Michigan cases uphold laws with retroactive periods of similar length. Outside of Carlton, which approved a retroactive period of one year, the Supreme Court has given little guidance on the definition of “modest.”

Taxpayers were sorely disappointed when Hambleton was denied certiorari, because it is hard to imagine a more sympathetic situation for a due process retroactivity challenge to a state tax. The Hambleton case involved two widows’ estates. As stated in the petition:

Helen Hambleton died in 2006, and Jessie Macbride died in 2007. Each was the passive lifetime beneficiary of a trust established in her deceased husband’s estate, and neither possessed a power under the trust instrument to dispose of the trust assets. Under the Washington estate tax law at the time of their deaths, the tax did not apply to the value of those trust assets. In 2013, however, the Washington Legislature amended the estate tax statutes retroactively back to 2005, exposing their estates to nearly two million dollars of back taxes.

In 2005, Washington State enacted an estate tax that was intended to operate on a standalone basis, separate from the federal estate tax. In interpreting the new law, the Washington Department of Revenue issued regulations that the transfer of property from the petitioners’ husbands to the petitioners through a Qualified Terminable Interest Property (QTIP) trust was not subject to the Washington estate tax. The Department subsequently reversed its position and assessed tax. Petitioners, along with other estates, challenged the Department’s position and won in Washington Supreme Court (In re Estate of Bracken, 290 P.3d 99 (Wash. 2012)).

In 2013, the Washington legislature amended the estate tax to retroactively adopt the Department’s position, going back to 2005. The petitioners challenged this new law and again fought to the Washington Supreme Court, which this time held in favor of the Department and concluded that the retroactive change satisfied the due process clause under a rational basis standard. This chain of events is inherently unfair and, if allowed, potentially subjects taxpayers to new tax liabilities at any time.

Although disappointing, the Supreme Court of the United States’ denial of certiorari in Hambleton is not surprising. The Supreme Court has declined previous opportunities to review retroactive state tax impositions (see, e.g., Miller v. Johnson Controls, Inc., 296 S.W.3d 392 (Ky. 2009), cert. denied, 560 U.S. 935 (2010)). The Gillette case will continue up the chain in Michigan and likely will be appealed to the Supreme Court of the United States, regardless of who prevails in the Michigan Supreme Court. As explained in the Hambleton certiorari petition and the supporting amicus briefs, the Supreme Court needs to revisit the retroactivity issue and act as a check as states continue to aggressively seek ways to raise additional revenue.