In a blow to the Lehman Chapter 11 estates, the United States Bankruptcy Court for the Southern District of New York held on September 16, 2015 that Intel Corporation’s Loss calculation resulting from a failed transaction under an ISDA Master Agreement was appropriate.1 The decision is significant both because of the dearth of judicial interpretation of the ISDA mechanics regarding the calculation of early termination amounts, and because it affirms the general market understanding that a non-defaulting party has broad discretion in calculating “Loss,” so long as its calculation is reasonable and made in good faith. It also suggests that, in considering Lehman’s valuation disputes with non-settling counterparties that elected the Loss calculation mechanism, the focus of the bankruptcy court’s inquiry will be on whether the counterparty’s calculation was reasonable, not whether Lehman can prove that it has a superior calculation.


Plaintiff Lehman Brothers OTC Derivatives Inc. (“LOTC”) and defendant Intel Corporation (“Intel”) entered into a 1992 ISDA Master Agreement, as modified by a schedule (together, the “ISDA Master”), governing the over-the-counter derivatives relationship between the parties. Under the terms of the schedule, Lehman Brothers Holdings Inc. (“LBHI,” and together with LOTC, “Lehman”) acted as a guarantor for LOTC’s obligations under the ISDA Master.

On or about August 1, 2008, LOTC and Intel entered into a trade confirmation (the “Confirmation”) for a share repurchase transaction whereby LOTC would purchase Intel shares and deliver them to Intel during a “quiet period” in which Intel was prohibited by securities regulations from transacting in its own shares. In accordance with the Confirmation, Intel remitted $1 billion to LOTC on August 29, 2008, and on September 29, 2008, LOTC was obligated to deliver to Intel a number of shares of Intel common stock determined by dividing $1 billion by the value-weighted average price of Intel shares between September 2, 2008 and September 26, 2008, less a forward price adjustment. The parties further agreed that to secure its obligations, LOTC would post $1 billion of collateral to be held by Intel. In the Confirmation, the parties selected “Loss” as the termination payment measure for the transaction agreed to in the Confirmation.2

LBHI (the credit provider under the Confirmation) commenced its Chapter 11 case on September 15, 2008, followed by LOTC on October 3, 2008. After LOTC failed to deliver the shares of Intel stock on September 29, 2008, as contemplated by the Confirmation, Intel declared an “Early Termination Date” of September 29, 2008. Using Loss, Intel calculated an Early Termination Payment of $1,001,966,256, consisting of the $1 billion it delivered to LOTC on August 29, 2008 plus interest, and on September 30, 2008, Intel set off and applied the collateral it was holding.

Lehman subsequently commenced an adversary proceeding against Intel disputing Intel’s Early Termination Payment calculation. Lehman contended that the only reasonable calculation of Intel’s Early Termination Payment was an amount equal to the fair market value of the undelivered Intel shares as of the close of trading on September 29, 2008, or $873 million. Lehman also argued that Intel breached the Confirmation when it seized the portion of the LOTC collateral in excess of the $873 million fair market value. Both parties moved for summary judgment. 

The Bankruptcy Court’s Analysis

Intel, as the non-defaulting party, was responsible for calculating the Early Termination Payment. Lehman argued that the fair market value of the shares at the close of the markets on the Early Termination Date was the only reasonable calculation of “Loss” as that term is interpreted by the ISDA User’s Guide to the 1992 ISDA Master Agreements (the “ISDA User’s Guide”). The bankruptcy court disagreed, finding that nothing in the definition of Loss mandates the use of any particular calculation method. Moreover, the court found that the ISDA User’s Guide makes clear that Loss is intended to provide parties flexibility in selecting a method to calculate Early Termination Payments and thereby functions as an express alternative to the more rigid methodology and procedure of Market Quotation.3 As a result, the court concluded that strong textual support exists for the proposition that non-defaulting parties generally may select any methodology for calculating Loss, so long as the methodology is reasonable and in good faith.

Lehman also argued that where a terminated transaction does not call for deliveries beyond the close-out date, the definition of Loss requires limiting the non-defaulting party’s loss incurred on the undelivered property as of the close-out date. According to Lehman, because such “Unpaid Amounts” are defined in the ISDA Master as the fair market value of the undelivered property, Intel was required to calculate its Loss as the fair market value of the shares Intel would have received had the transaction settled. The bankruptcy court again disagreed. Noting that the term “Unpaid Amounts” does not appear in the definition of Loss (but instead was a provision relevant only to the Market Quotation method), nor is it referenced in the ISDA Master provisions governing the calculation of an Early Termination Payment using Loss, the court was not persuaded by Lehman’s argument that the Loss calculation must be so limited.4

Lehman argued that further support for its position could be found in the fact that its calculation of Loss led to the same result as applying either the Market Quotation method of calculating an Early Termination Payment under the ISDA or New York law on damages. The court was not persuaded. With respect to the Market Quotation method, the Court found that the text and drafting history of the ISDA Master Agreement and the ISDA User’s Guide support the conclusion that Market Quotation and Loss can and should produce different results in certain circumstances. With respect to New York law, which provides that loss in connection with undelivered securities is limited to the fair market value of those securities on the date they were to be delivered, the Court noted that the parties did not document the transaction at issue as a simple contract for the purchase of shares governed by New York law; instead, they documented the transaction under the Confirmation and the incorporated ISDA Master. Accordingly, the Court held that Lehman could not, in hindsight, look to New York law to receive a more favorable outcome.

The court emphasized that there is no single correct methodology for calculating Loss. Rather, non-defaulting parties are afforded broad discretion in choosing a method to calculate Loss, so long as the calculation is performed reasonably and in good faith. As Lehman did not challenge Intel’s good faith in its calculation of Loss, the court went on to consider whether there was a disputed issue of material fact as to the reasonableness of Intel’s calculation of its Loss. After considering various methodologies advanced by Intel, including that its upfront costs were $1 billion, the court concluded that there was no such issue, and that Intel’s Loss calculation was reasonable.


The bankruptcy court’s conclusion that Loss generally permits non-defaulting parties such as Intel to select any calculation methodology, so long as that methodology is reasonable and in good faith, is, in the court’s own words “rather unremarkable,” given the broad definition of Loss in ISDA master agreements. Nonetheless, the decision is significant given the scarcity of judicial guidance on ISDA close-outs, as evidenced by the fact that the only other cases on close-out calculations discussed by the court in the decision were two factually distinguishable cases decided by courts in England. Furthermore, an opposite determination by the court would have been contrary to the general market understanding and could have led to market uncertainty. Concern in that respect had prompted ISDA to file an amicus brief in support of Intel’s motion for summary judgment.

With regard to the impact of the decision on the numerous valuation disputes that Lehman has outstanding with non-settling derivatives counterparties, it suggests that the focus of the court’s inquiry, at least for those counterparties that elected Loss and Second Method, will be on the reasonableness of the counterparty’s calculation, and that the court will not be receptive to efforts by Lehman to argue for a calculation approach that would have been preferable from its perspective with the benefit of hindsight. That said, the decision does not speak directly to many of the specific lines of attack that Lehman has advanced with respect to “big bank” early termination amount calculations such as: (i) emphasis on discrepancies between net mark-to-market values on September 12, 2008 (which was the last trading day prior to Lehman’s filing) and close-out values on September 15, 2008; (ii) the so-called “phantom loss” argument that if no replacement trades were entered into, there is no loss; (iii) allegations regarding the failure to apply portfolio aggregation to determine the economic equivalent of material terms of transactions as a group (vs. closing out trades in isolation, which would have the effect of amplifying the number of loss charges); and (iv) alleged inconsistent pricing across desks in closing out identical or similar trades. Accordingly, the decision leaves many key issues unresolved. It is noteworthy, however, that although the decision did not directly address one of the other issues involved in the “big bank” disputes—namely whether the calculation of close-out values for trades after the close-out date is appropriate—the bankruptcy court’s quotation of Professor Jeffrey Bruce Golden, one of the principal drafters of the ISDA Master Agreement, as stating that “[s]etting specific fixing times or prices was not the game,” could be viewed as potentially undermining Lehman’s position on that issue.