January 21, 2011
In our prior post on these cases, we compared the different factual findings made by the courts in analyzing penalty exposure under section 6664 and discussed the very factual nature of a reasonable cause and good faith penalty defense. Both cases were subsequently appealed. Canal Corp.looks like it is going to settle with the Fourth Circuit granting a motion to hold the appeal in abeyance pending finalization of that settlement (the company is in bankruptcy). Thus, those hoping for an appellate smack-down of the penalty supporting opinion from the Tax Court will be disappointed. NPR Investments is a different story. Briefs are due in that appeal (which lies in the Fifth Circuit) starting on February 19th. So Government hopes of another Son-of-Boss penalty success live on. If we see anything interesting, we will report on it.
September 13, 2010
We have been promising a post on the application of the section 6664 reasonable cause and good faith defense to tax penalties as it relates to reliance on tax advisers. Here it is.
There has been much activity in this area in the district courts and the Tax Court and not much winnowing or rule setting in the circuits. This is understandable; the application of the standards is highly factual and is well-placed in the hands of trial judges. We will analyze here some potential inconsistencies in two recent high-profile section 6664 decisions, Canal Corp. v. Commissioner, (Slip Op. attached) (August 5, 2010) (which found reasonable cause and good faith lacking) and NPR Invs., LLC v. United States, (Slip Op. attached) (E.D. Tex. Aug. 10, 2010) (which found reasonable cause and good faith met).
In Canal Corp., the Tax Court considered the application of section 6664 to a should-level PricewaterhouseCoopers opinion. (In the parlance, a “should-level” opinion means that the transaction “should” be upheld; it is a higher standard than more-likely-than-not, which means only that the transaction is more, perhaps only 51% more, likely to be upheld than not.) The Canal Corp. transaction emerged from the decision of a predecessor of Canal Corp., Chesapeake Corporation, to dispose of its tissue business, WISCO. After seeking advice from PwC and others, Chesapeake decided to dispose of the business by forming a partnership with Georgia Pacific to which WISCO would contribute its assets and liabilities and from which WISCO would receive a distribution of cash. The cash was funded by the new partnership borrowing money, and that debt was indemnified by WISCO. In essence, the substantive question presented to the court was whether the contribution/distribution amounted to the formation of a partnership (which would not trigger the built-in gain on the WISCO assets) or, rather, a sale of those assets to GP (which would).
In addition to helping structure and advise on the transaction, PwC was asked to prepare the aforementioned opinion. The partner writing the opinion was not the historic PwC engagement partner but rather an expert from the Washington National Tax group of PwC. PwC charged a flat fee of $800,000 for the opinion. Because the area of the law was relatively unclear, the opinion relied on analogy and analytics to reach its conclusions (including a withdrawn revenue procedure that set out tests to apply for advance rulings in a different area); there was apparently little direct authority available to cite. The parties effectuated the transaction on the day that PwC issued the opinion.
The Tax Court determined that the transaction was a disguised sale. This was based largely on the court’s conclusion that the indemnity by WISCO was illusory and thus that WISCO should not be allocated any amount of the partnership’s liabilities. If WISCO had been allocated these partnership liabilities then the transaction would be viewed as a financing transaction and not a sale. After all, you can’t call something a sale if the seller gets left holding the bag for the purchase price. But the court found WISCO’s bag empty and proceeded to penalties.
The Canal Corp. court began its analysis of section 6664 by recognizing that “[r]easonable cause has been found when a taxpayer selects a competent tax adviser, supplies the adviser with all relevant information and, in a manner consistent with ordinary business care and prudence, relies on the adviser’s professional judgment as to the taxpayer’s tax obligations.” Slip Op. at 31. However, the court noted that such advice “must not be based on unreasonable factual or legal assumptions” and cannot be relied upon when given by an advisor “tainted by an inherent conflict of interest.” Id. at 32. Citing Pasternak v. Commissioner, 990 F.2d 893 (6th Cir. 1993), a case involving promoters of master recording leasing programs, the Court concluded that a “professional tax adviser with a stake in the outcome has such a conflict of interest.” Id.
Applying these conceptual standards to the PwC opinion, the Tax Court found it lacking. At the outset, the court thought it incredible that significant time had been spent on an opinion so “littered with typographical errors, disorganized and incomplete.” Id. at 33. The Court’s confidence in the opinion was further undermined by the fact that only a draft could be found and the author (even after presumably being prepared for trial) did not recognize parts of the opinion when asked about them in court. On the question of assumptions generally, the lack of specific citation in support of the opinion’s premises and the frequent use of terms such as “it appears” in the place of hard analysis was also troubling for the court, which found it unreasonable that anyone would issue a should-level opinion on analogy and analytics with no direct support for the position. The Court found the author’s testimonial responses to challenges on these points unsettling and bluntly concluded that the only reason the opinion was issued at the should level was that “no lesser level of comfort would have commanded the $800,000 fixed fee that Chesapeake paid for the opinion.” Id. at 35.
On the question of a conflict of interest, the Court found a large one. Commenting that it “would be hard pressed to identify which of his hats” the author was wearing when he rendered the opinion, the Court concluded that the author’s work in developing, planning, structuring, and implementing the transaction took away too much of his independence (which the Court found to be “sacrosanct to good faith reliance”) to allow him to objectively analyze the merits of the transaction. Slip Op. at 36-38. Given that the only hurdle to closing the transaction was, in the end, the $800,000 flat fee opinion, the Court found that Chesapeake was attempting to buy “an insurance policy as to the taxability of the transaction.” Id. at 37. The Court voided the policy.
NPR involved a transaction the IRS characterized as a “Son of BOSS” transaction involving offsetting foreign currency options. As explained by the Court, the IRS’s view of a Son of BOSS transaction is “a series of contrived steps in a partnership interest to generate artificial tax losses designed to offset income from other transactions.” Slip Op. at 2 n.3. It is fair to say that Son of BOSS transactions are considered by the IRS to be one of the “worst of the worst,” so much so that they are the only transaction that is specifically barred from being considered by IRS Appeals. See Announcement 2004-46, Sec. 5 (May 24, 2004). Indeed, NPR conceded the merits of the transactions at issue prior to the decision. Accordingly, the only items considered by the district court were a period of limitations issue (which we will not discuss here) and the penalties.
After working through the background elements of section 6662, the district court in NPR began its analysis in a similar way to the Tax Court in Canal Corp., setting out both the restrictions on relying upon unreasonable assumptions and on a conflicted adviser. Slip Op. at 24-26. The conclusion, however, was quite different. Finding that the taxpayers were “not tax lawyers” and were not “learned in tax law” the court held that their reliance on the more-likely-than-not opinion of R.J. Ruble (who, at the time of the court’s ruling, had already been convicted of tax evasion associated with the rendering of tax opinions) was reasonable based on its findings that the opinion reached “objectively reasonable conclusions” and detailed a “reasonable interpretation of the law” (albeit one that the taxpayers conceded before trial). Id. at 27-28. Critically, the Court found persuasive the taxpayer’s plea that they, as unsophisticated men, sought out the advice of professionals who they did not know were conflicted and followed that advice; “what else could we have done except follow their advice?” Id. at *28.
As shown above, the differences between Canal Corp. and NPR are not differences in legal standards but differences in fact-finding. Both courts invoked and applied the same standards and prior interpretations of those standards; they just applied them to different facts as each judge found those facts. That is exactly what trial courts are supposed to do; take legal standards and do the hard work of applying them to the myriad fact patterns that arise. Viewed from that perspective, there is nothing in conflict between the two rulings; different facts support different results.
In a sense, the “inconsistencies” give a certain comfort in the decisions of both courts. Judges say what the law is, that is true. But more relevant to a trial lawyer, in a bench trial, they say what the facts are. In both Canal Corp. and NPR, the judges reached a conclusion based on their common-sense perceptions of what happened in the courtroom. While they can (and likely will) be second-guessed, that is their job. The NPR court was not swayed by all of the IRS’s anti-Son-of-BOSS rhetoric. Rather, the court evaluated the honesty and integrity of the specific taxpayers before it, their options (not the foreign currency kind) and their knowledge, and decided that no more could reasonably be asked of them. Similarly, the Canal Corp. court wasn’t swayed by the involvement of a major accounting firm in a business transaction between two large, sophisticated companies. Instead, the court looked at the analytics and thoroughness of the opinion, the involvement of the author in the transaction (including what he was paid), and his credibility on the stand, and concluded that it was unreasonable for a sophisticated consumer of tax advice to rely on his opinion. Whether you agree with the fact-finding (which is tough to do if you didn’t sit through both trials), the fact-finding has to be separated from the analytics; the analytics were sound (and consistent).
Viewed from the perspective of the tax planner, however, justifying the different outcomes on the basis of different fact-finding does not provide much comfort. Most tax planners would turn up their nose at a Son of BOSS opinion given to a group of individual investors to generate relatively large foreign currency options losses on a relatively minor investment. Yet a significant number have criticized the Tax Court’s opinion in Canal. Perhaps the distinction is just based on an “I know [a good transaction] when I see it” analysis, but many view what Canal Corp. did as “legitimate” tax planning and believe that a PwC advisor from the esteemed Washington National Tax group should have been viewed as more credible than a convicted felon. However, when a judge looks into the eyes of the adviser and doesn’t like what she sees, the taxpayer is at grave risk on penalties. Similarly, when the written product is capable of being analytically questioned, even undermined, based on sloppiness and lack of support or detail, a judge can be expected to have a negative reaction to that work product. That negative reaction will carry over to the credibility of its author, particularly where a substantial fee was received. On the flip side of the coin, if the judge finds the taxpayer honest and forthcoming about what he believed and what he tried to do to confirm that belief, the judge is likely to find reasonable cause and good faith. In short, the way the judge perceives the facts determines the outcome; that is why they call it a facts and circumstances based test.
Asking for “consistency” in such matters amounts to nothing less than the neutering of the trial court. “The ordinary lawsuit, civil or criminal, normally depends for its resolution on which version of the facts in dispute is accepted by the trier of fact.” NLRB v. Pittsburgh S.S. Co., 337 U.S. 656, 659 (1949). Indeed, rather than chasing the siren song of legal consistency, it is better to accept that fact-based tests like section 6664 belong to the trial lawyers to prove and to the trial judges to find. While that may appear to create a lack of consistency, it doesn’t. As we have shown, the inconsistency some see in Canal Corp. and NPR does not flow from an inconsistency in the law. Rather, there is always unpredictability as to how the facts will be perceived by different decision-makers. That is merely the uncertainty of litigation: the risk that a given judge on a given day may or may not believe your witnesses or your theory of the case. This is necessarily so; “[f]indings as to the design, motive and intent with which men act depend peculiarly upon the credit given to witnesses by those who see and hear them.” United States v. Yellow Cab Co., 338 U.S. 338, 341 (1949). Said differently, what a lawyer (or a client) thinks the facts are doesn’t matter if they can’t convince the judge they draw to perceive the facts as they see them. Making choices between “two permissible views of the weight of evidence” (id.) is precisely what trial judges are supposed to do and precisely what both of the judges in these cases did. Appeals in both cases, if they are filed, will have to take this into account.