TaxProf Blog op-ed: Marinello – Curbing Abusive Exercise of Prosecutorial Discretion in Tax Crimes Cases, by Steve R. Johnson (Florida State):
On Wednesday, March 21, by a vote of seven to two, the U.S. Supreme Court decided Marinello v. United States, No. 16-1144. Marinello is an important tax crimes decision. It also is instructive at many points with respect to statutory interpretation, an enterprise fundamental to all areas of tax, civil as well as criminal.
But the core of the case transcends the particular statute at issue and the respective merits of various canons of statutory interpretation. The questions at the base of Marinello are as fundamental as one can get: the proper relationship between the people and their government and the proper relationships among the three branches of government. In my view, the Marinello dissent had the better of the technical argument as to statutory construction but missed much of the larger picture. The Marinello majority had a better sense of the whole and moved the needle in the right direction, but it produced an ill defined “rule” through dubious reasoning. So, two — not three — cheers for the Marinello decision.
I. The Statute
I.R.C. section 7212(a) proscribes either “corruptly or by threats of force … endeavor[ing] to intimidate or impede any officer or employee of the United States acting in an official capacity under this title, or in any other way corruptly or by force or threats of force … obstruct[ing] or imped[ing], or endeavor[ing] to obstruct or impede, the due administration of this title.” The “or in any other way …” portion of the section is known as the Omnibus Clause. “This title,” of course, means Title 26 of the United States Code, i.e., the Internal Revenue Code. Under the case law, “corruptly” means acting with the intent to obtain an unlawful advantage for oneself or another. Violation of the section is a felony punishable by a fine and up to three years of imprisonment.
II. The Dispute
Carlo Marinello was tried and convicted on nine counts of tax-related offenses. Eight counts involved failing to file personal and corporate returns, misdemeanors under I.R.C. sec. 7203. The ninth count involved corruptly endeavoring to impede tax administration under the Omnibus Clause of section 7212(a). The government identified eight practices of Marinello’s that it viewed as violations of the Omnibus Clause, including failing to maintain books and records, failing to give his account complete information, destroying records, hiding income, and paying employees with cash. Many of these practices occurred before the IRS began to audit Marinello civilly or investigate him criminally.
On appeal, a panel of the Second Circuit affirmed the sec. 7212(a) conviction. The Second Circuit, over a sharp two-judge dissent, declined to review the decision en banc. The Supreme Court granted certiorari to resolve a circuit split. The Second Circuit’s Marinello decision, 839 F.3d 209 (2016), conflicted with the Sixth Circuit’s decision in United States v. Kassouf, 144 F.3d 952 (1998)(holding that conviction under the Omnibus Clause requires proof that the acts occurred while an IRS proceeding was pending).
III. The Technical Interpretive Issues
Generally agreeing with the Sixth Circuit, the Court reversed the Second Circuit’s decision. The majority opinion was penned by Justice Breyer and was joined by the Chief Justice and Justices Kennedy, Ginsburg, Sotomayor, Kagan, and Gorsuch. The majority opinion advanced a number of statutory interpretation arguments. First, the majority invoked something like the “in pari materia” canon, based on three of its precedents narrowly reading arguably comparable language in non-tax criminal statutes. Second, it found the language of section 7212(a) to be “neutral,” i.e., unhelpful. Third, it concluded that a “words in context” approach suggested a narrow interpretation. Fourth, for “[t]hose who find legislative history helpful,” it offered snippets from the admittedly sparse committee reports. Fifth, it noted that various of Marinello’s acts are crimes, typically misdemeanors, under other statutes as well. For example, section 7203 is violated by the willful failure to keep records. The majority maintained that a broad reading of section 7212(a) would render such other sections unnecessary, violating the “no surplusage” canon. Sixth, the majority invoked something like the rule of lenity, based on the fear that, read broadly, the Omnibus Clause would not give citizens adequate notice of what conduct of theirs could violate the Clause.
Justice Thomas dissented and was joined by Justice Alito. The dissent addressed each of the above interpretive rationales in the majority opinion. Styles of interpretation vary, of course. By my lights, though, the dissent had the better of the technical debate on most of the points.
IV. The Broader Issues
Much more persuasive, in my estimation, were the broader policy arguments advanced by the majority. First, the majority offered the amorphous but nonetheless powerful notion that there must be some limit on the sweep of the Omnibus Clause. For example, the majority hypothesized that a broad reading of the Omnibus Clause could render a person guilty of a felony in the following situation. “A taxpayer may know with a fair degree of certainty that her babysitter will not declare a cash payment as income — and, if so, a jury could readily find that the taxpayer [by nonetheless paying in cash] acted to obtain an unlawful benefit for another.” Prosecuting the parent/payor under these circumstances would not be so silly as to fall within the “absurd results” exception to textual interpretation, but it would make many uncomfortable with reading of the Omnibus Clause expansively.
Relatedly, the majority argued that we cannot “rely upon prosecutorial discretion to narrow the statute’s scope.” Why not? Because “to rely upon prosecutorial discretion to narrow the otherwise wide-ranging scope of a criminal statute’s highly abstract general statutory language places great power in the hands of the prosecutor.”
The dissent’s counter to these concerns was in the vein of “yeah, maybe that’s true, but it’s up to Congress, not the courts, to make policy judgments.” In the dissent’s words: “To be sure, sec. 7212(a) is a sweeping obstruction statute. Congress may well have concluded that a broad statute was warranted …. Whether or not we agree with Congress’ judgment, we must leave the ultimate resolution of the pros and cons of whether a statute should sweep broadly or narrowly … for Congress.” (Citation and punctuation omitted).
The dissent’s approach amounts to “a court should not try to stop our going to Hell in a handbasket if Congress created that basket, decided to put us in it, and chose to direct it to the infernal regions.” This is a principled approach to the allocation of powers wrought by the Vesting Clauses of the Constitution, but not one that reflects the main thrust of federal separation-of-powers jurisprudence. Justices Thomas and Alito were in dissent for a reason. They walk a lonely road here.
V. The New “Rule”
What does Marinello mean for future section 7212(a) prosecutions? The majority limited use of the Omnibus Clause in two ways. First, the government will have to show “that there is a ‘nexus’ between the defendant’s conduct and a particular administrative proceeding, such as an investigation, an audit, or other targeted administrative action.” Nexus is “a relationship in time, causation, or logic with the … proceeding,” and targeted “does not include routine, day-to-day work carried out in the ordinary course by the IRS.” Second, the government also must show “that the proceeding was pending at the time the defendant engaged in the obstructive conduct or, at the least, was then reasonably foreseeable by the defendant,” i.e., that “the proceeding must at least [have been] in the offing.”
In most situations, the two limitations imposed by Marinello will not seriously hamper tax enforcement. Omnibus Clause charges will remain viable in the clear majority of cases in which they usually are brought. In any event, the Code contains numerous, often overlapping criminal sections, affording a “deep bench” to substitute for section 7212(a) in the relatively few instances in which, as a result of Marinello, section 7212(a) is no longer a viable charge. For instance, in the case at hand, Mr. Marinello’s section 7203 convictions stand. Moreover, to the extent that his conviction of a felony was important, the government quite probably would have gotten a felony conviction had it charged him under section 7201 (criminalizing tax evasion) rather than under section 7212(a).
This is not to deny that there may be some problematic situations. For example, one year after deciding Kassouf, the Sixth Circuit decided United States v. Bowman, 173 F.3d 595 (1999). Bowman felt that he’d been wrong by several businesses and individuals. In retribution, he filed false information reports with the IRS, to get the IRS to investigate his enemies for failing to report the taxable income Bowman’s false information reports said they’d gotten from him. Actions like Bowman’s threaten to waste IRS resources (to say nothing of their effects on the innocent targets). Such actions corruptly impede tax administration and should be punishable under section 7212(a). Indeed, seemingly limiting Kassouf to its facts, the Sixth Circuit upheld Bowman’s conviction.
Neither the majority nor the dissent in Marinello mention Bowman, and Marinello’s new “rule” would undercut section 7212(a)’s utility in Bowman and other unusual situations. Perhaps Marinello’s ill defined tests could be massaged in future cases to cover such situations. The Marinello majority left room for refinement by stating “we need not here exhaustively” particularize all aspects of the new limitations. Or perhaps prosecution in Bowman-type unusual cases could be shoe-horned into some other criminal section. It may become necessary to address such matters in the future.
Nonetheless, one should not lose sight of the big picture. The two aspects of the new “rule” obviously are vague, and the majority’s statutory interpretation was stretched in many ways. So Marinello is not a work of art. But the case does clearly move in the right direction. The Department of Justice, concerned about the elasticity of the statutory language, initially took a restrained view of when Omnibus Clause prosecutions are appropriate. Over time, however, especially in the last three decades, this restraint eroded.
The majority’s intuition that there must be some limit is right. United States v. Caldwell, 989 F.2d 1056 (9th Cir. 1993), forcefully advanced a similar argument in limiting in the tax context conspiracy prosecutions under 18 U.S.C. sec. 371. It cannot be the case that anything that makes the IRS’s job harder is a felony.
Marinello is important, and salutary, not as carefully chiseled, finely nuanced doctrine, but as a “shot across the prosecutorial bow.” Prosecutors wield great power and, within limits, their exercise of discretion should be respected. Yet Lord Acton’s aphorism about the corrupting effects of power still fairly depicts human nature. We have seen prosecutorial abuse too often in too many contexts. In numerous cases, tax and non-tax alike, the courts have felt it necessary to douse the flames of excessive prosecutorial enthusiasm. I believe that the Court was correct in concluding that situations like Marinello call for the same intervention.