Wells Fargo – the Federal Circuit’s opinion was helpful but . . .

The Federal Circuit issued its decision in the Wells Fargo “interest netting” case on June 29th.  It’s a welcome decision, because the Federal Circuit rejected  the government’s extreme interpretation of the law and ruled for the taxpayer for one of the two scenarios in dispute.  But I found the court’s reasoning unpersuasive with respect to the other scenario.  Arguably, it framed the analysis incorrectly and reached the wrong decision in ruling for the government in that scenario.  As a result, taxpayers will have more opportunities to claim the benefits of interest netting, but less than they would have had under the Court of Federal Claims decision.  Corporate taxpayers should review past interest computations for overpayments and underpayments to identify opportunities for refund claims.


The “interest netting” issue arises when a corporation has an overpayment for one tax period and an underpayment for another tax period, both of which are outstanding for an overlapping period of time.  The Code requires the government to pay interest (with some exceptions) to the taxpayer when refunding an overpayment and also requires the taxpayer to pay interest to the government on an underpayment.  But overlapping overpayments and underpayments create a problem for taxpayers, because corporations have to pay a higher rate to the government for underpayments than the rate they receive from the government for overpayments.  The rate differential ranges from 1% – 4.5%, depending on circumstances.  (For those masochists interested in more details, see this article/outline I prepared for a presentation two years ago; the “same taxpayer” issue is discussed on pp. 16-19.)  This differential only applies to corporations; the interest rates for individuals are the same for underpayments as for overpayments.

This obviously could lead to situations where, for example, a corporation had a $1M overpayment for the 2005 tax year and a $1M underpayment for the 2006 tax year both outstanding for the period from 2007 through 2012.  Although the corporation did not owe a net tax amount during that period, it would still owe interest because the overpayment for 2005 would earn interest at a rate lower than the underpayment for 2006.  Congress considered this an inequitable result and, after 10 years of asking the IRS to implement procedures to net the two balances, enacted IRC § 6621(d), which provided for the elimination of interest on overlapping periods of tax overpayments and underpayments:

To the extent that, for any period, interest is payable under subchapter A and allowable under subchapter B on equivalent underpayments and overpayments by the same taxpayer of tax imposed by this title, the net rate of interest under this section on such amounts shall be zero for such period.

That lead to questions about exactly what “same taxpayer” meant in a corporate context.  Corporations liable to have significant interest netting issues also tend to have large corporate structures, often with consolidated returns and often with occasional acquisitions or mergers over the years.  The IRS initially put forth fairly reasonable interpretations of IRC § 6621(d) in these contexts.  But in recent years the government has retreated to an extremely narrow interpretation while taxpayers have pushed for broader interpretation.

The three scenarios

The Federal Circuit identified three different typical scenarios in this case, for which the results would not necessarily be – and eventually were not – identical.  For example:

  • Scenario 1 – Old Wachovia has an overpayment for the 1993 tax year and First Union had an underpayment for the 1999 tax year.  In 2001, First Union and Old Wachovia merged into New Wachovia.
  • Scenario 2 – For the 1993 tax year, First Union had an overpayment.  Between 1993 and 1999, First Union underwent four statutory mergers and was the surviving corporation in each merger.  Then, First Union had an underpayment for the 1999 tax year.
  • Scenario 3 – CoreStates had an overpayment for the 1992 tax year.  In 1998, CoreStates merged with First Union, with First Union as the surviving corporation.  Then, First Union had an underpayment for the 1999 tax year.

The overpayments and underpayments in issue were outstanding, accruing interest, for long periods of time.  In all three scenarios, the overpayment and the underpayment were both outstanding during an overlapping period and Wells Fargo argued that interest netting should be allowed in all three scenarios.  The Court of Federal Claims ruled in Wells Fargo’s favor for all three scenarios, without really distinguishing them, because under principles of merger law, merged entities are the “same taxpayer” for purposes of IRC § 6621(d).

Enter the Federal Circuit

The Federal Circuit took a slightly different approach.  Scenario 2 was not in dispute after oral arguments, as the government had conceded.  The government initially argued that the “same taxpayer” meant not only that the tax returns in issue were filed under the same tax identification number (TIN) but also that the taxpayer (First Union, in that scenario) was identical in all respects in 1993 and 1999.  Because four other companies had merged into First Union during that period of time, it was not the “same taxpayer” in 1999 as it was in 1993.  As courts had previously recognized, large corporations undergo regular changes and this argument would virtually eliminate “interest netting” for the very taxpayers most likely to be disadvantaged by the interest rates differentials.  By the time of argument in the Federal Circuit, the government had retreated to a slightly less extreme argument: the taxpayer with the overpayment and the taxpayer with the underpayment must have the same TIN and be the “same in relevant essentials.”  In retrospect, what is most surprising about Scenario 2 is that the government had asserted such an extreme position in the first place, in the Court of Federal Claims.

Scenario 1 was disposed of relatively easily.  The Federal Circuit decided that an earlier decision, the Energy East case, governed and that interest netting should be allowed only if Old Wachovia and First Union were the “same taxpayer” when the overpayment arose (in 1993) and the underpayment arose (in 1999).  Because the merger between the two did not take place until 2001, “at the respective times of the overpayment and underpayment, there were two distinct taxpayers.”  Thus, Wells Fargo was not entitled to the benefits of IRC § 6621(d) for those balances.

But the taxpayer won under Scenario 3.  The TIN for CoreStates (1992 overpayment) and the TIN for First Union (1999 underpayment) were not the same.  The court pointed out that Scenario 2 and Scenario 3 differed “only in the choice of who is the named surviving corporation.”  The court concluded that, in the context of a statutory merger, a post-merger entity (First Union) is the “same taxpayer” as a pre-merger acquired entity (CoreStates).  The analysis was based on the purpose of the statute, legislative history, the parallel with IRC § 6402(a), and how the Code treats mergers.

The court’s analysis of interest netting in the context of mergers, and its rejection of the government’s “identical TIN/same in relevant essentials” argument, are welcome clarifications of the proper application of IRC § 6621(d).  Although there was already some non-precedential support, in the IRS’s initial guidance on the issue, a decision by the Federal Circuit to that effect will be very helpful to taxpayers.

But are Scenario 1 and Scenario 3 really distinguishable?

The court’s analysis of Scenario 3, although I believe correctly decided, is a little bit shaky.  The court states that “it is the identify of the corporation at the time of the payments that matters. . . . Nothing in Energy East suggests that later changes in corporate structure can retroactively change a taxpayer’s status as to earlier payments.”  But in fact that’s exactly what the court did in Scenario 3.  CoreStates was not the “same taxpayer” as First Union in 1992; that happened only when the two corporations merged in 1998.  Effectively, what the court was really doing – although it did not state this explicitly – was evaluating the identity of the corporation at the first date both the underpayment and overpayment were outstanding.

That is certainly a reasonable approach to IRC § 6621(d).  Bu it raises another question.  In Scenario 3, the court can retroactively change the identity of CoreStates (with an overpayment for the 1992 tax year) in 1998.  Suppose in Scenario 1 that Old Wachovia’s 1993 overpayment was not refunded, and First Union’s 1999 underpayment was not paid, until 2005.  Can the identity of both corporations be changed in 2001, so that interest netting is not available initially but is available from 2001 through 2005?  If the merger can change the identity of one corporation with a pre-merger balance, why can’t it change the identity of both corporations with pre-merger balances, effective as of the date of the merger?  (I think this is Stephen Olsen’s point at the end of his excellent post on Wells Fargo over at the Procedurally Taxing blog.)  There may be a legitimate argument that the first date both the underpayment and overpayment were outstanding should govern the determination, rather than the date of the merger, but the court didn’t really explain it well.

Framing the question

The analysis might also change if the question had been framed as the IRS did in its 2002 FSA.  There, instead of focusing on two different taxpayers and deciding whether (and when) they had become the “same taxpayer,” the IRS effectively just focused on one of the taxpayers and asked whether it would “both be liable for the underpayment of tax, and entitled to the overpayment of tax.”  And the determination was different for consolidated returns (situations 1-4, 8-9) and mergers (situations 5-7).  Generally speaking:

  • For consolidated returns, a subsidiary is jointly and severally liable for an underpayment, but is not necessarily entitled to an overpayment, by the consolidated group.  Thus, the date of acquisition is key because the newly-acquired subsidiary is not liable for the consolidated group’s underpayments for periods before it was a member of the affiliated group.  The acquisition doesn’t change the status of pre-merger underpayments and overpayments.  But a pre-acquisition overpayment by the subsidiary can be netted with a post-acquisition underpayment by the consolidated group because the subsidiary is both entitled to a refund of its own overpayment and liable for the consolidated group’s underpayment.
  • For mergers, the surviving corporation is both entitled to the refunds of any overpayments by either merging company and liable for any underpayments by either merging company.  The merger does change the status of pre-merger underpayments and overpayments.  But in an acquisition in which both companies survive (situation 6), this does not apply because there is no assumption of assets and liabilities.

Under this way of framing the question, Energy East (involving consolidated subsidiaries) was decided correctly because the acquisition of the subsidiaries did not make the parent company either entitled to one subsidiary’s pre-acquisition overpayment or liable for the other subsidiary’s pre-acquisition underpayment.  Scenarios 2 and 3 in Wells Fargo were also decided correctly.  Scenario 1 is less clear.  In fact, Scenario 1 seems indistinguishable from situations 5 and 7 in the 2002 FSA, for both of which the IRS concluded (correctly I believe) interest netting was permissible.


Although there was a reasonable argument for broader application of IRC § 6621(d), the Federal Circuit wasn’t willing to go quite that far.  But overall, this is still a good result for taxpayers.  And another illustration of just how complex and confusing simple interest questions are.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s