Implicit Support in the US and Canada: Perrigo and GE Capital Canada

General Electric Capital Canada, Inc. v. The Queen (2009) introduced the use of implicit support by tax authorities to challenge multinational’s assertions of high interest rates for intercompany loans and high intercompany guarantee fees. The IRS has recently asserted implicit support in several situations where a foreign based affiliate extended an intercompany loan to a US borrowing affiliate. In this piece, we'll discuss an intercompany loan involving Perrigo, and compare the IRS settlement to the litigation involving GE Capital Canada before we turn to another potential IRS issue.

The Perrigo Issue

My May 31, 2021 post discussed a $7.5 billion intercompany loan that was incurred in connection with the merger when Perrigo purchased Elan in late 2013 for approximately $8.5 billion. This discussion speculated on the terms of the intercompany loan based on limited information noted in Perrigo’s 10-K filing for fiscal year ended December 31, 2020. More recent filings have provided more information with respect to this issue. These filings also provide more information with respect to the positions of the IRS and Perrigo and note an eventual Appeals decision was reached. This note updates my original discussion, which noted that the IRS proposed reduction was originally based on 130% of the Applicable Federal Rate, which was 2.57% for medium terms loans in February 2014. My original discussion criticized this position as suggesting an interest rate that would have clearly been less than the arm’s length rate. The IRS ultimately dropped this position, but did insist on the consideration of implicit support. The key portions of Perrigo’s more recent filings state:

On January 20, 2022, the IRS responded to our protest with its rebuttal in which it revised its position on this interest rate issue by reasserting that implicit parental support considerations are necessary to determine the arm's length interest rates and proposed revised interest rates that are higher than the interest rates proposed under its 130.0% of AFR assertion. The blended interest rate proposed by the IRS rebuttal was 4.36%, an increase from the blended interest rate in the RAR of 2.57% but lower than the stated blended interest rate of the loans of 6.8%. An IRS Appeals conference for the interest rate issue was held during March 7, 2023 through March 9, 2023. On May 5, 2023, we finalized an agreement with IRS Appeals resulting in settlement of the May 7, 2020 NOPA of $153.4 million of gross interest expense reduction for the 2013-2015 tax years. This implies a blended interest rate of 5.44%.

Our update assumes a five-year intercompany loan with the date being February 12, 2014 when the five-year government bond rate was almost 1.6%. Five-year government bond rates were lower than 1.6% for the other dates during this month. As such, the credit spread implied by the original intercompany interest rate was 5.2%, which would be consistent with a B standalone credit rating. Table 1 presents this scenario as well as three possible IRS scenarios.

Table 1: Intercompany Interest Rates and Interest Expenses Under Four Scenarios (Millions)

The first scenario assumes a BBB credit rating, which was Perrigo’s group credit rating at the time.  Under the taxpayer’s position that the credit rating should be the B standalone rating, interest expenses would be $510 million per year. Under the BBB group credit rating, the interest expense would be only $232.5 million per year. The difference between these two credit ratings represents six notches.

The IRS revised position on January 20, 2022 was consistent with starting with the group credit rating and lowering the credit rating by two notches to BB+. At a 4.36% interest rate, interest expenses would be $327 million per year. The ultimate settlement agreed to a 5.44% interest rate, which would imply interest rates equal to $408 million per year. This interest rate is consistent with a BB-, which can be seen as the standalone credit rating increased by two notches to account for implicit support.

Comparison to the Testimony in GE Capital Canada

Paul Wilmshurst of Charles Rivers Associated provided his views on implicit support in a discussion of the GE Capital Canada litigation:

The case dealt with an explicit financial guarantee from a GE company in the US to its Canadian subsidiary, which was at the time primarily engaged in the business of commercial financing outside of the GE group, with debt that it raised on the Canadian bond market. The Canadian tax authority disallowed the subsidiary’s deduction of the guarantee fee it paid to its US parent on the basis that even without the explicit guarantee, the parent would never allow it to default on its external debt given the problems this would create for the parent. That is, the explicit guarantee had no value since the subsidiary’s creditors would assume that the US parent would always step in. The tax authority considered that the likelihood of this was so great that the subsidiary should have the same credit rating as its parent (which was AAA on the S&P scale) … Although the Court agreed with the tax authority’s arm’s length framework, it ruled in favour of GE Capital on the pricing. (The explicit guarantee fee was 100 basis points.) This was because the Court decided that the explicit guarantee provided an incremental benefit over and above the implicit level and that the fee charged was within a reasonable range. The level of implicit support was assessed as being much lower than that implied by the tax authority’s position (three notches above the subsidiary’s stand-alone rating compared to 12 or 13 notches). Given its conclusion on the low impact of the implicit support, the Court indicated its preference for the credit rating methodology set out by one of the expert witnesses, William John Chambers.

The 2009 Tax Court decision noted the testimony of 10 expert witnesses on the issue of the appropriate credit ratings and the implied guaranteed fees. This note will present the testimony of Dr. Chambers as well as a few other expert witnesses.

When the difference between the group rating of the parent and the assumed standalone credit rating of the borrowing affiliate is as wide as 12 notches or more, whether implicit support is approached from a top down approach, which was the approach of the Canadian Revenue Agency (CRA), or a bottom up approach is an important consideration. Dr. Chambers noted that Standard & Poors (S&P) uses a bottom up approach that starts with the standalone credit rating and then notches up depending on what extent the parent would provide the support in the event of financial stress.

Chambers testimony was that the standalone credit rating was BB-/B+ and should be increased by only two notches to account for implicit support, which would have resulted in a BB+/B ultimate rating. Mr. Fidelman of Deloitte testified that the standalone credit rating should be BB+/B but argued that no upward notching was necessary. The Tax Court ultimately settled on a credit rating of BBB-/BB+, which is often seen as Dr. Chambers standalone credit rating increased by three notches for the effect of implicit support.

Dr. John C. Hull used credit default rates to evaluate the appropriate intercompany guarantee fee. The court decision noted:

He valued the guarantee by using the yield approach, which consisted of analyzing the spread between AAA rated bonds and bonds that are an average of single B and BB, the credit rating Dr. Hull was asked to assume for the Appellant in the absence of an explicit guarantee … The witness concluded that the overall spread was about 352 basis points between a AAA rating and the B+ to BB- rating assumed for the Appellant in the absence of an explicit guarantee. As a result, the value of the explicit guarantee is approximately 1.83% based on a BB+ to BBB- credit rating range.

His results were consistent with the difference between the credit spreads of the borrower give the assumed credit rating and the credit spread for a AAA borrower. Table 2 illustrates this contention by presenting the type of credit spreads for credit ratings from AAA to B during tranquil financial times.

Table 2: Credit Spreads by S&P Credit Ratings

Under Dr. Chambers estimate of the standalone credit spread, the credit spread would be near 4%. If the credit spread for a corporate bond with a AAA credit rating were 0.5%, the differential credit rating approach would suggest a guarantee fee near 3.5%. The court chose a credit rating between BBB- and BB+, which would suggest a credit spread near 2.3%. As such the differential credit spread would be near 1.8% under the bottom up version of implicit support.

Dr. Anthony Sanders was the primary expert witness for CRA. Although he initially argued that the guarantee fee should be zero under the CRA’s extreme assumption that the credit rating for the Canadian affiliate should be the group rating of AAA. Dr. Sanders, however, did consider a credit rating of AA, which was two notches below the group credit. Under this assumption, he testified that the guarantee fee should be only 0.2%. Dr. Sanders was asked by counsel for GE Capital about the fact that its subsidiary GE Financial Assurance Holdings Inc. was rated A+ on the basis of implicit support. This subsidiary was likely more strategically more important to the parent than GE Capital Canada as its assets were approximately ten times the assets of GE Capital Canada. As such its credit rating under implicit support would likely be higher than the credit rating for GE Financial Assurance.

One has to wonder why the CRA did not use this evidence and evidence discovered in 2010 to argue for an A credit rating under implicit support, which would suggest that the guarantee fee should be only 0.5%. Following the 2009 court decision, the CRA appealed the Federal Court of Appeals to consider new evidence with respect to the credit ratings of the Australian and Hong Kong affiliates of GE Capital. While a November 1, 2010 ruling denied this request, the evidence that the CRA wished to introduce included the fact that both of these affiliates had A credit ratings:

New evidence may exceptionally be presented on appeal if it can be shown that it could not have been discovered before the end of the trial, and that it is otherwise credible and practically conclusive of an issue on appeal … it demonstrates that the Tax Court Judge should, in evaluating the respondent’s credit rating, have given greater weight to the fact that it is owned 100% by GECUS as well as the implicit support of and its integration with triple “A” GECUS. According to the Crown, the new evidence establishes that the respondent’s credit rating is as high as GECHK’s, if not higher … the new evidence establishes the unreliability of Mr. Chambers’ opinion as to how Standard & Poor’s (S&P) would rate an unguaranteed debt issued by the respondent as it was not based on factors which the actual S&P ratings from GECHK took into account … the Crown has failed to show that the evidence sought to be adduced is practically conclusive of the issue raised on appeal. It certainly is an element which the Tax Court Judge would have had to address and weight in the context of the mass of evidence which he was called upon to review. However, it has not been shown that it would have altered the conclusion that he reached.

The Federal Court of Appeals rejected this request to consider new evidence on what appears to be a suggestion that the A credit rating for the two affiliates were not directly comparable to the appropriate credit rating for GE Capital Canada. Whether an analysis using these two credit ratings would have suggested that the appropriate guarantee fee should be lower than 1% was never explored. 

Application to a 2012 Intercompany Loan to a Subsidiary of Kimberly Clark

Kimberly Clark is a multinational with global sales exceeding $20 billion per year with various products in its professional brands, personal care, and consumer tissue. Kimberly Clark has grown in part by acquisitions of other companies with complimentary product lines. On March 5, 2012, a US subsidiary in one of these acquired segments borrowed $300 million from a foreign affiliate in a 10-year fixed interest rate loan with an interest rate = 7%. At the same time, the multinational issued $300 million in 10-year corporate bonds with an interest rate = 2.4%. The defense of this higher interest rate was based on an analysis under the standalone standard. Given how the IRS has recently insisted that the analysis of the pricing for intercompany loans must consider implicit support, a review of this issue may prove useful. 

Table 3: Kimberly Clark Debt Issuances

Table 3 presents the key information on a $300 million intercompany loan from Kimberly Clark’s Singapore affiliate to a US operating affiliate. The intercompany loan was issued on March 5, 2012 with a 10-year term and a fixed interest rate = 7%. The interest rate on 10-year US government bonds on that date = 2% so the implied credit spread = 5%. The estimated credit rating under the standalone principle for this US affiliate was B, which would be consistent with a 5% credit spread. A similar company located in Canada received a B+ credit rating from Standard & Poor’s despite being profitable with the reasoning that this Canadian company incurred significant debt. The estimated B credit rating was therefore reasonable if the standalone principle were respected.

On the same date as the intercompany loan, Kimberly Clark issued $300 million in corporate bonds with a 10-year fixed interest rate = 2.4%. As such the credit spread on this third party borrowing was only 0.4%, which would be consistent with a AAA credit rating. The group rating was therefore 17 notches above the estimated standalone credit rating. Table 3 also presents the key information on a more recent issuance of corporate bonds where the term was also 10 years and the fixed interest rate was only 2%. The interest rate on 10-year government bonds on the November 17, 2021 issuance was 1.6% so the credit spread again was only 0.4%. The credit rating provided by Moody’s for Kimberly Clark was only A, which is puzzling given the fact that it was able to borrow funds at such a low interest rate. 

With a difference of 17 notches between the estimated standalone credit rating and the group credit rating for the intercompany loan back in early 2012, whether and how implicit support is factored in would be a crucial issue if the intercompany interest rate were challenged. The taxpayer would prefer a bottom-up approach as if it had to accept even a three notch increase, the revised credit rating would be BB. This position would suggest a 3% credit spread and a 5% interest rate. 

Could the IRS, using a top-down approach, argue for a credit spread of only 1% on the premise that the credit rating should be A under implicit support? Under such an aggressive IRS position, the interest rate would be only 3%. Standard & Poor’s recently gave an A- credit rating to Kimberly-Clark de Mexico. While this credit rating for an affiliate of Kimberly Clark might be used as evidence for a top down approach, more analysis would have to be performed to see whether the appropriate credit ratio would be A- or closer to BBB. Such an analysis was what the CRA had hoped to conduct in 2010 during the GE Capital Canada case. 

References

Harold McClure, "Intercompany Financing: Engie State Aid Case and Perrigo’s Battle with the IRS," EdgarStat Blog, May 12, 2021.