What Can Be Done If Tax Restitution Exceeds the Tax Due (9/2/13)

I previously expressed concern as to whether tax restitution ordered incident to sentencing could be fine-tuned in subsequent IRS administrative proceedings if it appeared to the amount of the tax restitution were too high.  See Tax Restitution and Doubt As to Amount (Federal Tax Crimes Blog 7/10/13), here. I was concerned that the new statute permitting immediate assessment of tax restitution and precluding contesting same would foreclose adjustments downward.  As a consequence, I urged, in imposing tax restitution the sentencing court should err on the side of caution to insure that the tax restitution amount is not too high.  In the case discussed, the IRS agent had indicated during the sentencing phase that adjustments might be made later in the civil phases.

In PMTA 2012-027 (10/22/12), here, reprinted at 2013 TNT 168-24, the author concluded:

A taxpayer cannot challenge the amount of court-ordered restitution at a CDP [Collection Due Process] hearing. A district court's final order cannot be modified by challenging the amount of ordered restitution at a CDP hearing. Also, I.R.C. § 6201(a)(4) prohibits collateral attacks on a restitution order in a subsequent legal and administrative proceeding under the Internal Revenue Code, of which a CDP hearing is an example. Furthermore, a challenge to the amount of restitution in a CDP hearing is prohibited under I.R.C. § 6330(c)(4) because the criminal tax case itself is considered a prior judicial hearing on that issue in which a taxpayer meaningfully participated.

I won't attempt now to dissect the reasoning of the author.  I just note that the conclusion is a major warning about dangers that may lurk in restitution proceedings.  I would hope that counsel -- both government and defense -- and the courts will be sensitive to the issue and do the work necessary to insure that tax restitution does not overstate the tax liability.

The IRS is not hurt by caution in the restitution amount at the sentencing proceeding because the IRS can assert any additional amounts due through the regular tax determination and assessment procedures.  The author of the PMTA thus reasons:

The Service's assessment of the amount of restitution ordered by a federal district court does not, however, prevent the Service from assessing civil penalties and tax liabilities on top of the criminally ordered restitution if the amount of restitution is less than the defendant's total tax liabilities for that same period. See Helvering v. Mitchell, 303 U.S. 391(1938) (holding that Congress may impose both a criminal and a civil sanction in respect to the same act or omission); Morse v. Commissioner, 419 F.3d 829, 833-35 (8th Cir. 2005) (holding that despite a federal criminal case against the same taxpayer resulting in a sentence the taxpayer pay a fine and make restitution to the Service, the doctrine of res judicata did not apply to preclude a civil fraud penalty assessment on a tax deficiency because criminal prosecution for filing false income tax returns did not involve same cause of action as civil tax deficiency case). Cf. United States v. Helmsley, 941 F.2d 71, 102 (2d Cir. 1991) ("It is true that the government may pursue a tax evader for unpaid taxes, penalties and interest in a civil proceeding. However, we believe it is self-evident that any amount paid as restitution for taxes owed must be deducted from any judgment entered for unpaid taxes in such a civil proceeding."). In other words, the restitution amount is a floor and not a ceiling with respect to the tax period at issue. n4 Although the amount of a restitution may not be the actual amount of the taxpayer's full tax liability for that tax period under Title 26, the Service will effectively treat the amount of restitution as the minimum tax liability for the relevant tax period by assessing it "as if such amount were such tax" for that period. Section 6201(a)(4)(A).
   n4 The tax or penalty liabilities determined by the Service in excess of the amount ordered as restitution are not an assessment of the restitution ordered by a federal district court. Accordingly, section 6201(a)(4)(C) does not prohibit a taxpayer from challenging in any judicial or administrative proceeding under the Code those tax or penalty determinations that are in excess of the amount ordered as restitution. For example, if a taxpayer is ordered to pay $100,000 in restitution for the tax period ending 20XX and the Service determines pursuant to a subsequent examination that the taxpayer has an additional unpaid tax liability of $20,000 and is subject to a section 6663 fraud penalty of $90,000 for that same period, section 6201(a)(4)(C) does not prohibit the taxpayer from challenging the $20,000 additional tax liability and the $90,000 fraud penalty, but does prohibit the taxpayer from challenging the $100,000 amount of restitution determined by the federal district court and assessed by the Service.

I have a concern about this as well.  One key component in its reasoning that the taxpayer should be denied a CDP remedy to lower the tax restitution is the prohibition in Section 6330(c)(4) of CDP consideration for issues where the taxpayer previously had a meaningful opportunity to participate in a prior proceeding.  The taxpayer surely does have a right to participate at sentencing in the quantification of the restitution.  But, the Government has that right also.  So, why is not the quantification at sentencing preclusive in later proceedings as to what the amount of the tax liability is?  Why is it preclusive only as a minimum for the taxpayer but not a maximum for the IRS?

I have previously argued for collateral estoppel on the related issue of the tax loss amount which is the principal driver for sentencing.  John A. Townsend, Collateral Estoppel in Civil Cases Following Criminal Convictions, 2005 TNT 4-28 (1/6/05).  Here is how I illustrate the issue in my current draft of the Federal Tax Crimes text:(some footnotes omitted)

Let’s test our understanding of collateral estoppel. 

Example 1:  For tax year 1, taxpayer fraudulently claims large deductions, thus wiping out his year 1 tax liability and creating a net operating loss on the tax year 1 tax return filed 4/15 of year 2.  Taxpayer elects not to carryback his losses, and therefore carries them forward.  He claims some portion of the NOL carryforward in each of the years 2 through 4 on 4/15 of each succeeding year (e.g., the year 2 tax return is filed on April 15 of year 3, etc.).  On 5/1 of year 5 (shortly after filing the year 4 tax return), he is indicted for tax evasion for years 1 through 3, each year being a separate count.  He is subsequently convicted on all counts.  Based on the foregoing, the conviction of tax evasion for years 1 through 3 is collateral estoppel as to civil fraud in years 1 through 3.  But, is the year 1 conviction collateral estoppel on the civil fraud issue as to the year 4 return?  One court has so held, reasoning that the claiming of a fraudulent NOL carryforward perforce renders the return fraudulent and, since the fraudulent claiming of the NOL loss itself has already been litigated between the parties in the NOL year (year 1 in this example and, by extrapolation, years 2 and 3 for which he was convicted for carrying the fraudulent loss to those years), it is perforce collateral estoppel in the subsequent nonconviction year (year 4 in the example).  What would you argue in response? 

You would argue, of course, that the convictions in the prior years are not preclusive on the issue of whether the taxpayer had the requisite intent in the fourth year.  It is at least conceivable that he would not.  For example, what if the carryforward to year 4 were relatively small in amount, and the taxpayer perhaps just did not think about the issue in year 4 (his return preparer just carried it forward from last year’s return without discussing it with the taxpayer)?  By analogy, in the criminal proceeding involving tax evasion charges for years 1 through 3, could the judge have instructed the jury that, if it found the taxpayer guilty for year 1, it must find him guilty for years 2 and 3?  I think not, because even though the fraudulent tax benefit traced to year 1, the issue is the taxpayer’s guilt at the time he filed his year 2 or 3 return.  If that finding of guilt is not preclusive in the criminal proceeding, why should a finding of guilt be preclusive in a noncharged year? 

Example 2: Assume the same example.  In the sentencing phase, the quality of the evidence is as follows: (1) the tax loss number established by a preponderance of the evidence for each of the years 1-3 is $100,000; and (2) the sentencing judge is in equipoise as to an additional $25,000 and thus cannot include that amount in the tax loss number.  In the ensuing civil fraud case, the parties agree that the total underpayment in each of the first 3 years was $200,000.  Assume that the Government establishes by collateral estoppel its initial burden to show fraud by clear and convincing evidence (i.e., the evasion conviction for years 1-3 establishes beyond a reasonable doubt that some amount of tax was evaded, hence it is collateral estoppel as to the Government’s ability to show fraud by clear and convincing evidence).  Then, assume that the evidence in the civil trial shows by a preponderance of the evidence that $75,000 of the $200,000 understatement is not due to fraud and that $100,000 of the $200,000 understatement is due to fraud (the latter being the same as at the sentencing phase).    Just as in the sentencing phase, however, the civil trial finder of fact is in equipoise as to the remaining $25,000.  Under§ 6663, the civil fraud penalty base is $125,000 for each of the years 1-3. 

What I want this example to illustrate is that the only difference between the sentencing finding as to the tax loss number and the civil fraud base is with respect to a possible inclusion or exclusion of the amount, if any, as to which the trier in the civil case is in equipoise.   In both proceedings (sentencing and civil trial), the base is the same (i.e., at sentencing it is the tax loss number and in the civil trial it is the civil fraud penalty base which define the same base).   Yet, as illustrated in the example, the base is a different amount at sentencing than at the civil trial solely because of the phenomenon of equipoise.  Equipoise is a useful tool for analyzing burden of proof but, I submit, not helpful in real world cases where it is only the rare case that has key findings turn upon a state of equipoise. n1083 In other words, for the real world, the example illustrates a phenomenon not likely to occur. n For this reason, I have argued that, for judicial estoppel and prudential reasons, the sentencing findings as to the tax loss number should determine the civil penalty base. n1084  The case authority, however, rejects collateral estoppel, although the circumstances may be distinguishable. n1085
   n1083 See Cigaran v. Heston, 159 F.3d 355, 357 (8th Cir. 1998) (“The shifting of an evidentiary burden of preponderance is of practical consequence only in the rare event of an evidentiary tie . . . .”) (emphasis supplied); see also Polack v. Commissioner, 366 F.3d 608, 613 (8th Cir. 2003) (citing the Cigaran case).
   n1084 Kosinski v. Commissioner, 541 F.3d 671 (6th Cir. 2008), a well reasoned and nuanced opinion citing, inter alia, the Booker shift in binding effect of the Sentencing Guidelines and Maciel v. Commissioner, 489 F.3d 1013 (9th Cir. 2007).   Two points may be worth consideration for those willing to not accept these holdings at face.  First, the Kosinski opinion makes clear that, although the scope of its holding might suggest that any sentencing finding would not be preclusive, it was in fact holding only that tax loss findings were not preclusive.  Second, in Maciel the court applied a presumption against collateral estoppel for sentencing findings on but seemed to allow the possibility that the presumption could be overcome.  Maciel involved a situation where the Court found, under the facts in the sentencing proceeding, the Government has little incentive to litigate the issue of evasion.  The Maciel Court had earlier quoted 18 Charles Alan Wright et al., Federal Practice and Procedure § 4423, at 612 (2d ed. 2002) as follows:  “The most general independent concern reflected in the limitation of issue preclusion by the full and fair opportunity requirement goes to the incentive to litigate vigorously in the first action.”  So, at least arguably, this holding may not apply where the finding of no fraud or the tax loss number was in the context of an incentive to litigate – i.e., real sentencing consequences could turn upon the dispute.  In Morse v. Commissioner, 419 F.3d 829 (8th Cir. 2005), the taxpayer was convicted for tax perjury (§ 7206(1)).  The conviction itself is not preclusive as to either the amount of the deficiency or the civil fraud penalty, because neither is an element of the offense.  The taxpayer argued, however, that the order of restitution at sentencing was preclusive.  Focusing only on the conviction (and not the judicial determinations at sentencing), the Court held that “An order for criminal restitution is not essential to the judgment of conviction against a criminal defendant ‘because it [is] not an element of the crime of conviction.’” [Citation omitted.] Finally, it is interesting that, although the trend in authority is to reject collateral estoppel for sentencing findings, do not assume that that is always the case.  To use the old saying that consistency is the hobgoblin of small minds, at the same time as denying preclusive effect to tax loss findings when the taxpayer attempts to invoke the doctrine, courts appear willing to permit the doctrine in other cases.  See Arguelles-Olivares v. Mukasey, 526 F.3d 171 (5th Cir. 2008) (accepting a PSR determination of the tax loss number that was adopted by the sentencing court as a finding of the tax loss for purposes of determining that in excess of $10,000 was involved for sentencing purposes).

I am presently in the process of conceptualizing and preparing an article on this and related subjects in conjunction with the Villanova Seminar I blogged recently, Villanova Seminar on Selective Issues in Tax Administration (Federal Tax Crimes Blog 8/31/13), here.  The summary of the article is as follows:

Title (Preliminary):  Tax Liability at Sentencing and Beyond 

In this article I will discuss the role that the tax liability at the center of the criminal tax prosecution plays out at and after sentencing.  Tax crimes, even those that do not require a tax due and owing as an element of the crime, are usually not prosecuted unless there is a material tax due and owing.  So, tax liability is at the center of the initial decision to prosecute.  Then, at sentencing, the "tax loss" is the principal driver for the Sentencing Guidelines calculations for tax crimes.  The article will discuss how the tax loss is determine (including concepts such as the criminal tax loss and relevant conduct).  The article will also discuss how any unpaid tax might be reflected in restitution (either statutory or contractual restitution).  The article will finally discuss other tax matters related to the crime but which arise after sentencing. 

In the process of discussing this, I will cover the recent amendment to the Sentencing Guidelines for unclaimed deductions, credits, etc., and the amendment that requires expedited assessment of tax restitution.

I plan to update the foregoing excerpted discussion in my book more or less contemporaneously with completing that article and may have further postings on the blog.