Sunday, May 20, 2018

Excellent Presentation on Restitution in Tax Crimes Cases (5/20/18)

At the ABA Tax Section May Meeting, a panel discussed restitution in criminal tax cases and presented a slide presentation titled "The Gathering Storm: Grappling with the Impact of Restitution Orders on Civil Tax Litigation," here.  The slide presentation was prepared by Caroline Caraolo, here.

The Topics covered are:
  • Tax Loss in a Criminal Proceeding
  • Restitution
  • How is a Restitution Calculated?
  • How can a Restitution Order be Challenged?
  • How is Restitution Collected?
  • Restitution-Based Assessments
  • Can Restitution be Avoided?
  • Can a Civil Examination be Avoided?
  • Impact of Restitution-Based Assessments in Civil Tax Proceedings

I add some points that are inherent in the presentation, but I would state differently to make the points:
  • Practitioners should understand the difference between tax loss for Sentencing Guidelines purposes and restitution.  Restitution is the actual tax loss for the count(s) of conviction.  Tax loss for Sentencing Guidelines purposes can include, in addition to the tax loss for the count(s) of conviction, the tax loss attributable to "relevant conduct" -- which may include related criminal conduct not included in the count(s) of conviction, whether not charged, dismissed, or acquitted.  Further, tax loss for Sentencing Guidelines purposes can overstate the actual tax loss if calculated without consideration of unclaimed deductions or credits.  And tax loss calculations are, as the slide presentation notes, subject to certain estimating conventions if it is otherwise not reasonably estimable.  
  • I don't think those Sentencing Guidelines conventions for estimation apply to the tax restitution calculation, so at sentencing it is important to argue that the Government should not be able to get by on such estimations.  Estimations work for Sentencing Guidelines purposes because of the broad bands of tax loss in the Tax Table (S.G. §2T4.1., here).  Most of the time even significant differences between the estimated tax loss and the real tax loss will not affect the Guidelines calculations and will not affect sentencing because of Booker downward variances so common in tax cases.  But, restitution calculations will affect the real world.  The taxpayer (or other defendant liable for restitution) will have to pay any excess of restitution over the real tax loss (or at least be liable to pay such excess).  So, in my judgment, estimations should not work unless they resolve all reasonable doubt within the preponderance of the evidence standard in favor of the defendant.  As the presentation notes:  "To estimate loss [for restitution], there must be a reasonable basis to approximate loss, and impossible to determine actual loss. United States v. Futrell, 209 F.3d 1286, 1291-92 (11th Cir. 2000)."  I don't think that is inconsistent with saying that reasonable doubts as to the restitution amount should be resolved in favor of the defendant.  
  • Keep in mind that for tax restitution, the amount can be assessed immediately and cannot be reduced by the IRS, so that really serious consequences can apply if the restitution is higher than the real tax loss.  In this regard, if the IRS believes that the real tax loss is higher than restitution, it can use the standard deficiency procedures to assess and collect the tax, so foregoing asserting restitution amounts that may be too high will not impair the fisc.
Another thanks to Caroline.  (Thanks to Caroline are very common on this blog; or at least not uncommon.)

Prior Blogs on restitution and tax loss are (reverse chronological order):
  • Relevant Conduct Raised by the Probation Officer and Used by the Sentencing Judge in Calculating the Guidelines (Federal Tax Crimes Blog 3/10/15), here.
  • On Restitution, Count of Conviction and Tax Loss (Federal Tax Crimes Blog 10/24/13), here.
  • Restitution, Relevant Conduct, Counts of Conviction (Federal Tax Crimes Blog 4/11/13), 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.