Saturday, January 7, 2012

Conviction for Hokey Insurance Tax Shelter Affirmed; Backdating Bad and No Relief for Government's Failure to Prosecute Others (1/7/12)

In United States v. Rozin, ___ F.3d ___, 2012 U.S. App. LEXIS 230 (6th Cir. 2012), here, the Sixth Circuit affirmed a business owner's conviction for tax shenanigans related to a tax shelter promotion known as Loss of Income insurance.  Essentially, the taxpayer wrongfully claimed a deduction for the so-called insurance premium when it was nothing more than an investment through a U.S. Virgin Islands supposed insurance company.  Whether or not the insurance company was bona fide, the insurance policy as written was not at least as an insurance policy.  A number of people from the promoters through Rozin and a co-owner were indicted.  All were charge with the Klein defraud conspiracy.  Rozin was also charged with tax perjury and tax evasion related to the scheme.  Some defendants pled.  One died before trial.  Rozin and an in-house attorney for his business went to trial.  Rozin lost and was not happy with the result.  He moved for acquittal and, failing on that motion and being still unhappy, appealed.

I found the  appeal to be a fairly routine-type of appeal in a tax shelter oriented criminal case -- i.e., standard arguments for Cheek willfulness, reliance on tax professionals, etc.  As usual, they did not work on appeal.

Two aspects of the opinion, while not particularly exceptional, are noteworthy as reminders for students and practitioners.

First, there was backdating to claim a tax benefit.  All of us should know that backdating is highly suspect with the high risk is illegality if the date is critical to tax consequences.  The Court of Appeals said:
The district court also properly concluded that backdating the LOI policies showed willfulness, because there was no reason for such backdating other than to claim the improper tax deductions. The district court regarded the practice as highly suspicious, noting at sentencing that "[a]t some point in time, [Rozin] should have known something was wrong. Probably when there was the backdating." According to the testimony of Steve Rowe, an IRS Revenue Agent called by the Government, "[a]ny evidence of backdating results in a badge of fraud, meaning the IRS would question the validity of the policy itself and the deduction." Though the district court did not elaborate, the court found "ample evidence" that the policies were backdated and that Rozin was instrumental in this decision. Liss testified that Rozin knew that the policies were going to be backdated and that as a result he would receive "no economic benefit for that period of time." Backdating served no legitimate business purpose; when the  policies were purchased in August 1999, but backdated to December 1998, only five months were left on the policy coverage and Rozin had triple coverage during this time. Though Rozin argues that businesses will backdate policies for a "myriad of legitimate reasons," he provides no viable rationale for his actions. Instead,  Rozin refers to the testimony of Mary Blanton, the former administrative assistant of Cohen Insurance, to support his claim. However, none of the reasons mentioned in her testimony is applicable to this case. Rozin's backdating and moving up the policy date permitted Rozin to have access to his funds even earlier than otherwise, and thus reflected improper motivation on his part.
Second, the Court of Appeals rejected Rozin's argument that the Government's failure to prosecute others purportedly similarly situated somehow conceded that willfulness was not involved.  The Court said:
There is also no merit to Rozin's argument that the Government "conce[ded]" that "purchasing and deducting the LOI policies on the advice of qualified professionals does not constitute willful tax evasion or the willful filing of a false tax return" because others who purchased the policies were not prosecuted. First, it is not possible to tell from the record whether the individuals mentioned by Rozin — Lawley, Sparnell, and Cohen's daughter — could have legitimately claimed good faith reliance. Just because Rozin was not able to meet the standard for a good faith defense does not mean that others might not meet it. Second, the facts undermine Rozin's allegations. For instance, Rozin claims that Lawley, an individual to whom Rozin marketed the LOI policies, bought the policies but was never prosecuted. However, Lawley never claimed a deduction on his 1999 tax return, and thus did not partake in an illegal activity. Though Rozin tries to argue that he also did not claim a tax deduction on his 1999 tax return, Count Two and Count Six of the Indictment were for the deductions taken during the 1998 calendar year. Rozin's situation was not analogous to Lawley's.
I suppose that was some type of selective prosecution claim but they too usually fail.  The truth is that the Government knows about many more tax cheats than it prosecutes and therefore is selective in the sense that some are selected and some are not.  But there are usually subtle issues involve that go into the exercise of prosecutorial discretion, including the truism that, systemically, only a relatively limited number of taxpayers can be prosecuted annually.  (Having said that, in many cases, it does seem almost as random as lightning striking to determine why some particular taxpayers are selected and some are not; and, having said that, the truth is that lightning is not always random.)

Finally, just one anecdotal point, Rozin was contemporaneously considering a "Basis Boost" shelter being promoted to Altheimer & Gray, where Paul Daugerdas was a partner prior to joining Jenkens & Gilchrist (which Daugerdas succeeding in destroying with shelters of similar genre).  The Daugerdas saga, of course, has contributed much content for this blog.  See here.

No comments:

Post a Comment

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.