CDS

CFTC Delays Reduction in Swap Dealer De Minimis Exception Threshold

On October 13, 2016, the Commodity Futures Trading Commission (the “CFTC”) approved an Order delaying for one year the reduction of the threshold for determining whether an entity constitutes a “swap dealer” for purposes of the U.S. Commodity Exchange Act.[1]  Currently, persons are not considered to be swap dealers unless their swap dealing activity in aggregate gross notional amount measured over the prior 12-month period exceeds a de minimis threshold of $8 billion.  This threshold had been scheduled to automatically decline to $3 billion on December 31, 2017, but the Order extended that date to December 31, 2018, absent further action from the CFTC.

The delay in the threshold decline follows the recent issuance by the CFTC of the Swap Dealer De Minimis Exception Final Staff Report (the “Final Report”).[2]  The Final Report supplemented a preliminary report (the “Preliminary Report”)[3] on the same matters and provided a summary of numerous comment letters the CFTC received in response to that report, as well as further data analysis.  These two reports together comprise the “report” contemplated by CFTC Regulation 1.3(ggg)(4)(ii)(B), which directed the CFTC to issue a report on topics relating to the definition of the term “swap dealer” and the de minimis threshold.

The Preliminary Report analyzed available swap data (primarily from the four swap data repositories (“SDRs”) registered with the CFTC) during the period from April 1, 2014, through March 31, 2015, for five asset classes: interest rate swaps (“IRS”), credit default swaps (“CDS”), non-financial commodities (“Non-Financial Commodities”), foreign exchange derivatives and equity swaps. However, the CFTC noted in the Final Report that it faced numerous challenges in data quality available from the SDRs, including a lack of information regarding whether a swap was entered into for dealing purposes and a lack of reliable notional data for all but IRS and CDS.[4]

The CFTC solicited comments on several topics in the Preliminary Report, including: (i) whether the current de minimis threshold should be maintained, raised or reduced; (ii) whether swaps that are executed on a swap execution facility (“SEF”) or designated contract market (“DCM”) and/or centrally cleared should be excluded from an entity’s de minimis calculation; (iii) whether the de minimis exception should be based on multiple factors (e.g., number of counterparties) instead of only gross notional swap dealing activity; (iv) whether a de minimis threshold should be established for each asset class; and (v) whether the current exclusion available to insured depositary institutions should be expanded.  The CFTC received 24 comment letters from banks, industry groups, legislators and other market participants and interested parties in response to the Preliminary Report.

The Final Report analyzed an additional one-year period of data for the IRS, CDS and Non-Financial Commodity asset classes to the period considered in the Preliminary Report.[5]  The primary conclusion of the Final Report was that “only a substantial increase or decrease in the de minimis threshold would have a significant impact on the amount of IRS and CDS covered by swap dealer regulation, as measured by notional amount, transactions, or unique counterparties.”[6]  The following chart from the Final Report summarized the results leading to this conclusion:[7]

Table 1 – IRS and CDS Potential Dealing Activity Covered by Notional Amount

dir-october-2016-table

Consistent with the Preliminary Report, the Final Report estimated that approximately 84 additional entities (from 145 to 229 entities) trading IRS and CDS might have to register as swap dealers if the de minimis threshold declined to $3 billion.  However, this 58% increase in the number of entities regulated would result in coverage of less than 1% of additional notional activity and swap transactions, and only 4% of additional unique counterparties.  Interestingly, as reflected in the table set forth above, an increase of the de minimis threshold to $15 billion would yield similar results: 34 fewer entities having to register, but reduced coverage of less than 1% of additional notional activity, swap activity and unique counterparties.

Moreover, the data analyzed indicated that a substantial majority of swaps (99% of IRS, 99% CDS and 89% Non-Financial Commodity swaps) involved a registered swap dealer during the final review period. In conclusion, the CFTC stated that it may want to consider whether to set the de minimis threshold to its current $8 billion threshold, allow the threshold to decline to $3 billion, as scheduled, or delay the reduction of the threshold while it continues its efforts to improve data quality.

Separately, the Final Report indicated that the comments received generally expressed support for excluding from an entity’s de minimis calculations swaps entered into on a SEF or DCM and/or centrally cleared, but that the CFTC had not had sufficient time to evaluate several factors that could impact the implementation of such an exclusion.[8]

In addition, the Final Report stated that the CFTC may want to consider: (i) maintaining a single de minimis threshold based on notional amount (instead of a threshold based on multiple factors); and (ii) maintaining the single gross notional de minimis exception (instead of adopting a class-specific approach) or consider adopting a class-specific approach in the future as data quality improves.[9]

Finally, the Final Report indicated that the CFTC may want to consider whether the conditions to the current exclusion to the swap dealer definition for insured depository institutions are overly-restrictive.


[1] Order Establishing De Minimis Threshold Phase – In Termination Date, 81 Fed. Reg. 71, 605 (October 18, 2016).

[2] Swap Dealer De Minimis Exception Final Report, August 15, 2016 (available at http://www.cftc.gov/idc/groups/public/@swaps/documents/file/dfreport_sddeminis081516.pdf).

[3] Swap Dealer De Minimis Exception Preliminary Report, November 18, 2015 (available at: http://www.cftc.gov/idc/groups/public/@swaps/documents/file/dfreport_sddeminis_1115.pdf).  For a summary of this report, click here.

[4] Final Report, at 5.

[5] The CFTC focused on IRS and CDS data because reliable notional data was not available for the other asset classes. Final Report, at 20.  The Final Report highlighted that many of the same limitations noted in the Preliminary Report for Non-Financial Commodity swaps persisted, but that the CFTC nevertheless performed an analysis using counterparty and transaction counts for this asset class. Id. at 19-20.

[6] Id. at 20.

[7] Id. at 21 (Table 1).

[8] Id. at 25.

[9] Id. at 26.

 

CFTC Expands Swap Clearing Requirement

 

On September 28, 2016, the Commodity Futures Trading Commission (the “CFTC”) unanimously approved the expansion of currencies of interest rate swaps subject to mandatory clearing under the U.S. Commodity Exchange Act (the “Act”).[1]  Subjecting standardized swaps to central clearing is intended to decrease risk in the financial system and has been a primary goal of global regulators for several years.

Section 2(h) of the Act makes it unlawful for any person to engage in a swap that is required to be centrally cleared unless that swap is submitted to a derivatives clearing organization (a “DCO”) that is either registered under the Act or exempt from registration under the Act.[2]  This same section of the Act sets forth the process through which the CFTC is to make determinations of whether a swap, or group, category, type or class of swaps should be subject to mandatory clearing.[3] READ MORE

Novo Banco and CDS – A Post-Mortem

 

In 2014, the International Swaps and Derivatives Association, Inc. (“ISDA”), published the 2014 ISDA Credit Derivatives Definitions (the “Definitions”), which updated the 2003 ISDA Credit Derivatives Definitions.[1]

One of the most significant changes in the Definitions was the inclusion of a new credit event for “Governmental Intervention.”[2]  This credit event was intended to address concerns expressed in the market that the credit event for “Restructuring” may not cover certain measures actually taken by governments to support struggling entities, especially banks.  Governmental Intervention[3] is generally defined to include actions or announcements by a “Governmental Authority”[4] that result in, inter alia, the reduction in the rate or amount of interest payable by a reference entity, an expropriation or other event which mandatorily changes the beneficial holder of the relevant obligation, or a mandatory cancellation, conversion or exchange.  This event is similar to Restructuring in certain respects (for example, the reduction in the rate or amount of interest of an obligation may trigger both events).  However, unlike Restructuring, deterioration in creditworthiness is not required to trigger a Governmental Intervention.

It did not take long for this new credit event to be probed and tested by a set of straightforward, but unusual, facts that apparently were not specifically considered by the drafters. In August 2014, Banco de Portugal, the central bank of Portugal, applied resolution measures to Banco Espírito Santo, S.A. (“BES”), a bank organized in Portugal that was experiencing distress.  These measures included a €4.9 billion rescue package for BES, and the transfer of numerous assets, liabilities and deposit-taking operations from BES to a new “good bank,” Novo Banco, S.A. (“Novo Banco”).  However, on December 29, 2015, Banco de Portugal announced the re-transfer of five senior Euro-denominated bonds (having almost €2 billion in principal) from Novo Banco back to BES.[5]  This re-transfer resulted in significant losses to bondholders, up to 90% in secondary market trading.

It was clear that Banco de Portugal constituted a “Governmental Authority” and that it had taken a binding action pursuant to its resolution law. However, irrespective of the undeniable losses suffered by creditors,[6] there was an issue as to whether the re-transfer of the bonds at issue from BES to Novo Banco affected creditors’ rights in one of the ways specified in Section 4.8(a) of the Definitions.

In accordance with the standard practice in the credit default swap (“CDS”) market, purchasers of protection asked the relevant 15-member Determinations Committee (the “Committee”) to determine whether the central bank’s actions amounted to a Governmental Intervention, hence triggering protection payments under CDS contracts.  Under the relevant rules, an 80% supermajority (12 of 15 members) is required for a Committee to decide whether or not a credit event has occurred.[7]   The Committee fell just short of the required percentage, with 11 members voting that there was no Governmental Intervention.  Although the Governmental Intervention event was generally intended to protect investors from governmental actions negatively affecting the value of obligations, the majority of the Committee concluded that the transfer of debt to another institution did not constitute a “mandatory cancellation, conversion or exchange” and did not have “an analogous effect” to the events specifically enumerated in the definition of Governmental Intervention.

Nevertheless, without the required supermajority, pursuant to the relevant rules governing Committees, the matter was referred for “external review.”[8]  External reviews of Committee decisions are quite rare and entail at least three “experts” nominated by Committee members deciding the issue.  A unanimous decision is needed to override the Committee’s original “No” vote where, as here, over 60% of Committee members voted against the occurrence of a credit event.  On February 15, 2016, the external review panel released a unanimous decision to uphold the negative determination.  In short, the panel decided that the central bank’s transfer did not constitute a mandatory cancellation, conversion or exchange of the obligations, and was not analogous to those types of event.[9]  On the latter point, the panel stated that taking a broader review of the word “analogous” would result in this clause “dominat[ing] the whole of the definition, which is inconsistent with [the] careful and detailed drafting” of the definition.[10]

In addition to seeking a determination that a Governmental Intervention had occurred, protection buyers pursued one more potential avenue for payout under the Definitions: seeking a determination that a “successor” event had occurred with respect to CDS naming Novo Banco as reference entity. A successor event occurs under the Definitions if, generally, more than 25% of the relevant obligations of a reference entity are transferred to another entity and more than 25% of the relevant obligations of the reference entity remain with the reference entity.[11]  The Committee sought additional information from Banco de Portugal and Novo Banco to determine whether this threshold had been exceeded.  After receiving this additional information, the Committee unanimously decided on March 3, 2016 that a successor event did not occur.[12]

Many industry participants believed that the Governmental Intervention was designed to protect against precisely the type of result that precipitated from Banco de Portugal’s actions.  Nevertheless, the Committee concluded, and the external review panel agreed, that the actions of Banco de Portugal did not align with the requirements of the Definitions.  Thus, one lesson from this situation is that Determinations Committees may take a formalistic view of requirements under the Definitions, regardless of the scope of losses incurred by creditors or the spirit or purpose of the language in the Definitions.  CDS purchasers should take note and not overlook the precise words of the Definitions.


[1] The implementation date for the updated Definitions was September 22, 2014.

[2] For a more complete description of the changes to ISDA Credit Derivatives Definitions, click here.

[3] “Governmental Intervention” is defined in full, as follows:

(a) “Governmental Intervention” means that, with respect to one or more Obligations and in relation to an aggregate amount of not less than the Default Requirement, any one or more of the following events occurs as a result of an action taken or an announcement made by a Governmental Authority pursuant to, or by means of, a restructuring and resolution law or regulation (or any other similar law or regulation), in each case, applicable to the Reference Entity in a form which is binding, irrespective of whether such event is expressly provided for under the terms of such Obligation:

(i) any event which would affect creditors’ rights so as to cause:

(A) a reduction in the rate or amount of interest payable or the amount of scheduled interest accruals (including by way of redenomination);

(B) a reduction in the amount of principal or premium payable at redemption (including by way of redenomination);

(C) a postponement or other deferral of a date or dates for either (I) the payment or accrual of interest, or (II) the payment of principal or premium; or

(D) a change in the ranking in priority of payment of any Obligation, causing the Subordination of such Obligation to any other Obligation;

(ii) an expropriation, transfer or other event which mandatorily changes the beneficial holder of the Obligation;

(iii) a mandatory cancellation, conversion or exchange; or

(iv) any event which has an analogous effect to any of the events specified in Sections 4.8(a)(i) to (iii).

Definitions, §4.8(a) (emphasis added).

[4] The term “Governmental Authority” includes, inter alia, any de facto or de jure government (or agency, instrumentality, ministry or department thereof), any court, tribunal administrative or other governmental, inter-governmental or supranational body and any authority or other entity designated as a resolution authority or charged with the regulation or supervision of the financial markets of the “reference entity” or some or all of its obligations.  Definitions, §4.9(b).

[5] It appears from Banco de Portugal’s announcement of the measure that it specifically selected these bonds because they were “intended for institutional investors.” See Banco de Portugal, Banco de Portugal approves decisions that complete the resolution measure applied to BES (December 29, 2015) (available at: https://www.bportugal.pt/en-US/OBancoeoEurosistema/ComunicadoseNotasdeInformacao/Pages/combp20151229-2.aspx). Among other things, the central bank stated that “[t]he selection of the above-mentioned bonds was based on public interest and aimed to safeguard financial stability and ensure compliance with the purposes of the resolution measure applied to Banco Espírito Santo, S.A. This measure protects all depositors of Novo Banco, the creditors for services provided and other categories of unsecured creditors.” Id.

[6] Some US$430 million in net protection payments would be triggered if it was determined that a credit event had occurred.

[7] ISDA supported a decision of a Governmental Intervention having occurred, arguing that the re-transfer of the bonds at issue constituted a “conversion or exchange” or had “an analogous effect” to an exchange, and that a negative determination “would lead to an illogical outcome.” See ISDA Determinations Committee Request, Has a Governmental Intervention Credit Event occurred with respect to Novo Banco, S.A.? (December 30, 2015) (available at http://dc.isda.org/documents/2016/01/novo-banco-pai-3.pdf). Specifically, ISDA argued as follows:

A decision that a Governmental Intervention has not occurred would lead to an illogical outcome where holders of the Transferred Bonds suffer a near complete economic loss for the arbitrary reason that they are institutional investors, but this loss is not mitigated by their CDS protection despite paying a higher premium for protections against the risk of governmental interventions.  This outcome will further diminish CDS’s efficacy as a tool to hedge credit risk and further erode confidence in the product. . . .  This action is taken under the “bail in” regime that the European banking regulators seek to implement, and is precisely the type of governmental intervention risk that the Definitions were designed to cover.

[8] Specifically, Section 4.1(a) of the 2014 ISDA Credit Derivatives Determinations Committees Rules (available at http://dc.isda.org/wp-content/files_mf/1414437004ICM2319997111v10DC_Rules_2014.pdf) provides that “[a]ny DC Question relating to DC Resolutions to be made by Supermajority under . . . 3.1(c) (Credit Event Resolution) . . . shall be referred to the external review process described in this Section 4 (External Review) . . . if a Convened DC . . . holds a binding vote on, but is unable to Resolve by a Supermajority, such DC Question.”

[9] Among other things, the panel pointed to the use of the word “transfer” elsewhere in the Definitions and concluded that “it would seem . . . that the draftsman has deliberately decided not to include the expression of ‘transfer’ as an event within the definition of a [Governmental Intervention].”  Novo Banco External Review; Decision and Analysis of the External Review Panel of the ISDA EMEA Determinations Committee with respect to DC issue Number 2015123002 pursuant to Section 4 of the 2016 ISDA Credit Derivatives Determination Committees Rules, at 4 (February 15, 2016) (available at http://dc.isda.org/documents/2016/02/nb-er-decision.pdf).

[10] Id. at 6.

[11] Definitions, at §2.2(a). More specifically, in this case, if the Committee concluded that BES succeeded to more than 25% of Novo Banco’s relevant obligations but more than 25% of relevant obligations remained with Novo Banco, then the CDS would be divided accordingly into new CDS transactions with each of BES and Novo Banco, respectively, as the reference entity. See id. at §2.2(a)(iv) and 2.2(n).

[12] See ISDA EMEA Credit Derivatives Determinations Committee, Statement (March 3, 2016) (available at: http://dc.isda.org/documents/2016/03/emea-dc-meeting-statement-3-march-2016.pdf). It appears that, in addition to Novo Banco’s €7.35 billion of outstanding senior debt, the European Central Bank had provided Novo Banco with some €7 billion in loans. Therefore, €2 billion of re-transferred bonds constituted less than 25% of Novo Banco’s aggregate obligations at the time.