In a world of near-zero official rate policy in response to the COVID-19 pandemic, there has been increasing discussion of the effectiveness of negative interest rates.1 In fact, US Treasuries daily bills briefly traded into negative territory on March 25, 2020,2 and the Secured Overnight Funding Rate (SOFR) has been hovering near-zero (but not yet negative) since March 19, 20203. Recent reports4 note that the bond market appears to be pricing in negative rates later in 2020. Implicit in this discussion is an assumption that negative interest rates will be an effective stimulus (that is, paying someone to borrow effectively reduces the opportunity cost of cash and creates economic incentive to invest).
However (and as we reasoned in 20165 and as we discuss below), lenders in commercial loan transactions likely never intended that LIBOR could or should be negative in the determination of interest rates for those transactions. Zero percent LIBOR floors had become almost universal by 2016, with many term loan transactions including 1% floors, and, in fact, in recent months lenders in documenting new commercial loans and amendments to existing ones have begun applying 1% or 75 basis point floors to all loan tranches. As we noted in 2016 (when discussing the possibility of negative LIBOR):
Despite the lack of certainty, a court may decide that a lender should be able to charge a higher rate of interest than what the contractual rate is (based on a contractual rate that would subtract negative Libor from the spread) on the basis that there is an implied floor of zero under these credit agreements; lenders never intended to charge an interest rate lower than the spread. If the implied floor is zero, then the contractual rate is actually higher than what the words in the contract say in the event Libor is negative.
And (in discussing whether, even if, there were no implied floor and the “all-in” rate was negative, the lender would have to pay the borrower):
In most cases interest rate margins are large enough that even if Libor were to be negative, the result would be a decrease in spread but still a positive all-in interest rate. However, depending on how far Libor reaches into negative territory, there will be some agreements, particularly for investment grade and other high credit quality borrowers, where margins are low enough that the all-in rate may actually be negative. In these cases, it would be difficult for a borrower to argue that they are actually owed interest by the lenders. Credit agreements are generally written so that the borrower has an obligation to pay interest to the lenders, and there is no contractual mechanic for lenders to pay interest back to the borrower. As a result, in the event that courts decide (or lenders decide to concede) that negative Libor can reduce spreads in certain credit agreements if there is not a zero Libor floor, it seems very likely that at the very least there will be a zero floor on all-in rates and lenders will not be expected to pay borrowers interest for the loans they made to those borrowers.
While there remain no US cases addressing these questions, cases elsewhere are. These cases include a Swiss Federal Supreme Court case and a United Kingdom Court of Appeal case, and, since they are instructive on how courts face these questions, they are briefly described below.
In the Swiss case,6 a loan agreement was entered into on July 20, 2006, in which the parties had agreed on a base interest rate (six-month LIBOR-CHF interest rate) plus a margin of 0.0375% per year. With the introduction of negative interest rates and the SNB's announcement of the cancellation of the CHF-EUR minimum rate in January 2015, the six-month LIBOR-CHF interest rate turned negative. In the case, the borrower requested the lender to freshly calculate the interest rate in accordance with the updated negative six-month LIBOR-CHF interest rate and pay the resulting negative interest. The borrower argued that the base interest rate was negative and that even taking into account the margin owed of 0.0375%, he would still be entitled to interest. The lender rejected the claim and argued that the loan agreement contained no explicit clause for the unexpected case that the six-month LIBOR-CHF interest rate turns negative. Accordingly, the borrower filed a lawsuit against the lender. The first- and second-instance courts adopted the interpretation of the contract of the lender and dismissed the borrower's lawsuit.
As the contracting parties had not expressly agreed on the possibility of negative interest rates and the real intention of the parties could not be established, the loan agreement had to be interpreted objectively under the so-called “principle of trust” under Swiss law. The Swiss Federal Supreme Court referred to different doctrinal views on the consequences if the base interest rate turns negative in loan agreements:
- According to the first doctrine, the interest rate payable can never be lower than the agreed margin, and the interest rate can therefore not fall below 0%. Thus, the margin is always owed to the lender, and the lender must never pay interest rate to the borrower.
- According to the second doctrine, the base interest rate can turn negative but only up to the amount of the margin; a negative interest rate of more than the margin would thus reduce the interest to be paid by the borrower to 0%. Further, the lender must not pay interest rates to the borrower.
- According to the third doctrine, the interest rate could fall under 0% by applying the contractually agreed formula, with the consequence that the borrower could demand interest payments from the lender as soon as the negative interest rate is higher than the margin. Accordingly, the lender must pay interest rate to the borrower.
The Swiss Federal Supreme Court did not favor one doctrine over another. Rather, it held that the loan agreement in dispute did not contain an explicit clause for the event that the six-month LIBOR-CHF interest rate turned negative, nor did it explicitly guarantee an interest rate of 0.0375% in favor of the lender. Further, the loan agreement did not expressly contain a clause which would provide the possibility of reversing the interest payment obligation. In fact, several clauses of the loan agreement explicitly referred to the borrower's interest payment obligation. Moreover, according to the Court, it is evident neither that the parties had expected negative interest rates when the loan agreement was concluded in 2006 nor that they intended that the borrower should be able to refinance himself using negative interest rates. Therefore, it cannot be inferred from an objective interpretation of the loan agreement that the borrower would receive negative interest payments in good faith. Due to the absence of a counterclaim by the defendant, the Court had to decide only whether negative interest was owed, and it could therefore leave open the question of whether the lender—in connection with the interest formula—would in any case receive the margin of 0.0375% (the first doctrine) or whether this margin would fall to 0% due to the negative base interest rate (the second doctrine).
In the UK case,7 the State of the Netherlands (State) and Deutsche Bank AG (Bank) had entered into a number of derivative transactions pursuant to an ISDA Master Agreement (Agreement) and Credit Support Annex (CSA) dated March 14, 2001, both governed by English law; the Agreement was the 1992 version of the ISDA Master Agreement, and the CSA was the 1995 version, though it had been amended in 2010, deleting and replacing paragraph 11 (which contains information about the credit support to be provided, including eligible collateral and interest rate). This meant that the Agreement and CSA predated the ISDA 2014 Collateral Agreement Negative Interest Protocol (Protocol), and the parties had not amended them in light of that Protocol.
The State and the Bank entered into various derivative transactions on the terms of the Agreement and CSA. If there was a net credit exposure of the State to the Bank, then the CSA required the Bank to provide credit support to the State—in this case, cash collateral. The CSA provided for interest to be paid on that cash collateral credit support at the rate of EONIA minus 0.04% (where EONIA is the Euro Over-Night Index Average).
However, the interest rate had been less than zero for the “larger part of the time since 13 June 2014.” The question before the court was, therefore, “whether the parties’ agreement, as made using the ISDA documentation concerned, requires the Bank to pay ‘negative interest’, i.e. interest from the party who provides a principal sum for a period of time, rather than from the party who receives it and has the use of it for a period of time.”8
The key provision of the CSA relating to interest is Paragraph 5(c)(ii):
Interest Amount. Unless otherwise specified in Paragraph 11(f)(iii), the Transferee will transfer to the Transferor at the times specified in Paragraph 11(f)(ii) the relevant Interest Amount to the extent that a Delivery Amount would not be created or increased by the transfer, as calculated by the Valuation Agent (and the date of calculation will be deemed a Valuation Date for this purpose).
Paragraph 11 of the CSA provided that “Transferee” was to be read as a reference to the State and “Transferor” as a reference to the Bank.
Paragraph 10 of the CSA defined Interest Amount as being calculated for each day as follows:
(x) the amount of cash in such [relevant] currency on that day; multiplied by
(y) the relevant Interest Rate in effect for that day; divided by
(z) 360 (or, in the case of pounds sterling, 365).
Paragraph 11(f) of the CSA stated that (as we have already seen) the relevant Interest Rate was “EONIA minus four (4) basis points,” i.e., EONIA minus 0.04%.
The court held that the State had not succeeded in meeting “the central point”: to show that there was an obligation in the CSA in respect of negative interest. The court accepted that the definition of “Interest Amount” was in principle capable of allowing a negative interest figure. However, it held this was simply “a starting point” and that the Agreement needed to be looked at as a whole.
The State recognized that paragraph 5(c)(ii) of the CSA provided only for the State to transfer Interest Amounts to the Bank and not vice versa. However it argued that negative interest rates should be taken into account in the calculation of the Credit Support Balance.
In support, the State emphasized the final sentence of the definition of “Credit Support Balance,” which envisaged the situation where “Interest Amounts” may not be transferred from the State to the Bank. As the CSA provided that interest accrued from day to day, the State argued that the “Credit Support Balance increases by the amount of positive accrued interest and decreases by the amount of negative accrued interest.”9
The court disagreed, finding that this sentence merely reflected the fact that the CSA (specifically Paragraph 5(c)(ii)) envisaged a situation where the Transferee —the State—was obliged to pay interest but had not yet transferred it. It did not of itself recognize an obligation in relation to negative interest.
The court pointed out that if the State’s arguments were correct, the effect was that negative interest had to be paid via a mechanism other than that set out in Paragraph 5(c)(ii) of the CSA—and the court did not believe that there was any “credible commercial rationale for the parties to have made such a choice.” The court noted that, in contrast, the parties had modified the CSA such that if the Bank paid the credit support in to the wrong bank account, the interest rate on those funds would be zero.
The State argued further that the commercial purpose of the interest provisions of the CSA was “equivalence”—they were intended “to bring about a situation in which neither the [Bank] nor the [State] suffers or benefits from the fact that the [State] holds collateral, over and above the fact that such collateral is to be available in the event of termination for default.”10 The court was not persuaded by this argument, noting that it is not necessarily the case that the State would incur loss by holding cash where interest rates were negative (i.e., that the cash collateral would actually diminish): the parties had agreed that the State remained free to use the cash to earn interest elsewhere.
The State also sought to rely, among other materials, on the Protocol. However, while this was introduced so that negative interest rates should “flow through ISDA collateral agreements under certain circumstances,” the court noted that the Protocol “contemplated the parties would amend paragraph 5(c)” of the CSA to achieve this11—which had not occurred in this case.
It was therefore clear it had been open to the parties to determine what was to occur in the event of negative interest rates. Why they had not done so was between them, but one answer may have been that they wanted simplicity. The court found that if there were any obligation in the CSA in relation to negative interest, it would be spelled out.
While, of course, each case will likely depend on its own facts, these cases suggest that, at least in contracts between sophisticated parties that do not explicitly address negative interest rates, courts will be reluctant to infer obligations to pay negative interest.
1 See, for example, “Negative U.S. Interest Rates?,” Christopher J Neely, Federal Reserve Bank of St. Louis, Economic Research, posted February 28, 2020 (available at: https://research.stlouisfed.org/publications/economic-synopses/2020/02/28/negative-u-s-interest-rates).
2 See: https://www.treasury.gov/resource-center/data-chart-center/interest-rates/Pages/TextView.aspx?data=billRatesYear&year=2020.
3 See: https://apps.newyorkfed.org/markets/autorates/SOFR.
4 See, for example, “ Markets Want Negative Rates Despite Unexpected Fed Gift” by Mohamed A. El-Erian, Bloomberg Opinion, May 8, 2020 (available at: https://www.bloomberg.com/opinion/articles/2020-05-08/coronavirus-markets-want-negative-rates-even-after-fed-gift).
5 See: https://www.mayerbrown.com/-/media/files/news/2016/03/what-negative-libor-would-mean-for-the-lending-mar/files/whatnegativeliborwouldmeanforthelendingmarket/fileattachment/whatnegativeliborwouldmeanforthelendingmarket.pdf.
6 The decision is unpublished, but the description is taken from several articles that describe the case and decision, including: https://www.internationallawoffice.com/Newsletters/Corporate-Commercial/Switzerland/Badertscher-Attorneys-at-Law/Federal-Supreme-Court-negative-interest-in-loan-agreements and https://www.mondaq.com/trials-appeals-compensation/817270/swiss-federal-supreme-court-no-negative-interest-rates-in-loan-agreements.
7 The State of the Netherlands v Deutsche Bank AG  EWCA Civ 771 and discussed in our earlier Perspective at: https://www.mayerbrown.com/-/media/files/perspectives-events/publications/2018/08/is-a-party-obliged-to-pay-negative-interest-on-col/files/aug18-is-a-party-obliged-to-pay-negative-interest/fileattachment/aug18-is-a-party-obliged-to-pay-negative-interest.pdf, from which the within description is taken.
8  EWHC 1935 (Comm), paragraph 4.
9  EWHC 1935 (Comm), paragraph 18.
10  EWHC 1935 (Comm), paragraph 20.
5 月 25 日2023 年Global Financial Markets Podcast
5 月 19 日2023 年