Docket ID OCC-2011-0008 Docket No. R-1415 RIN 7100 AD74 RIN 3064-AD79

Download as pdf or txt
Download as pdf or txt
You are on page 1of 6

July 11, 2011

Office of the Comptroller of Currency (OCC) 250 E Street, SW Washington, DC 20219 Docket ID OCC-2011-0008 Jennifer J. Johnson Board of Governors or the Federal Reserve System 20th Street and Constitution, NW Washington, DC 20551 Docket No. R-1415; RIN 7100 AD74 Gary K. Van Meter Acting Director Office of Regulatory Policy Farm Credit Administration 1501 Farm Credit Drive McLean, VA 22102-5090 RIN 3052-AC69 Robert E. Feldman Executive Secretary Attention: Comments Federal Deposit Insurance Corporation 550 17th Street, NW Washington, DC 20429 RIN 3064-AD79

Alfred M. Pollard General Counsel Attention: Comments/RIN 2590-AA45 Federal Housing Finance Agency Fourth Floor 1700 G Street, NW Washington, DC 20552 RIN 2590-AA45

Re: Margin and Capital Requirements for Covered Swap Entities Ladies and Gentlemen: Markit 1 is pleased to submit the following comments to the Office of the Comptroller of the Currency (the OCC), the Board of Governors of the Federal Reserve System (the Board of Governors), the Federal Deposit Insurance Corporation (the FDIC), the Farm Credit Administration, (the FCA), and the Federal Housing Finance Agency (the FHFA and, together with the other agencies, the Agencies) relating to the Agencies proposed rule regarding margin and capital requirements for covered swap entities (the Proposed Rule). 2 Introduction Markit is a service provider to the global derivatives markets, offering independent data, valuations, risk analytics, and related services for swaps and security-based swaps (SB swaps) across many regions and asset classes in order to reduce risk, increase transparency, and improve operational efficiency in these markets. Markit supports the objectives of the Dodd-Frank Wall Street Reform and Consumer Protection Act (the DFA) 3 of increasing transparency and efficiency in the OTC derivatives markets, of reducing both systemic and counterparty risk, and of detecting market manipulation and abuse. Executive Summary The requirement for Covered Swap Entities (CSEs) to compute and collect both initial margin (IM) and variation margin (VM) for uncleared swaps across asset classes, products and counterparties will pose significant operational challenges to the marketplace. Further, for IM and VM collection to result in the desired reduction of systemic risk, calculation needs to be performed in a reliable, unbiased, timely and consistent fashion.

Markit is a financial information services company with over 2,300 employees in North America, Europe and Asia Pacific. The company provides independent data and valuations for financial products across all asset classes in order to reduce risk and improve operational efficiency. Please see www.markit.com for additional information. 2 Margin and Capital Requirements for Covered Swap Entities, 76 Fed. Reg. 27564 (published May 11, 2011). 3 Dodd-Frank Wall Street Reform and Consumer Protection Act, Pub. L. 111-203, 124 Stat. 1376 (2010).

July 11, 2011 Page 2

We believe that the proposed rules can be improved in a number of aspects, and that independent third party providers (ITPPs) can provide services related to IM and VM that will help achieve the above goals and deliver benefits to all stakeholders. ITPPs can ensure that valuations, margin models and data inputs are reliable, comparable, and unbiased and, as a result, ensure that the collection of IM and VM will result in the desired reduction of systemic risk. As further discussed below, we believe that: (1) counterparties may not be able to agree upon a valuation method as part of their trading documentation that is as detailed as required by the Proposed Rule, so counterparties should be able to reference inputs, methodologies, or valuations provided by appropriately qualified ITPPs; (2) the inputs and assumptions used in internal IM models should be independent or should be independently verified, and all of the Agencies should explicitly permit CSEs to delegate IM calculation to ITPPs; (3) DCO models are generally not appropriate to calculate IM for uncleared swaps; and (4) the rules should reflect that expected liquidation horizons vary between swaps based on a number of factors. Comments 1. Trading Documentation That Forms the Basis For VM Calculation Should be Practicable and Allow Delegation Similar to the comparable rule proposed by the Commodity Futures Trading Commission (the CFTC), the Proposed Rule would require CSEs 4 to execute trading documentation that specifies: (i) the methods, procedures, rules, and inputs for valuing swaps for the purpose of calculating variation margin requirements, and (ii) dispute resolution procedures concerning the valuation of swaps or assets collected as collateral for the same purpose. 5 We believe that, in principle, requiring counterparties to a swap transaction to agree on valuation methodologies prior to or contemporaneously with execution could be useful to reduce the potential for disagreement about the valuation of swaps during their lifetime. However, given the complexities of valuing swaps, we are concerned that the provisions as proposed could require agreement on a nearly impossible degree of detail. We believe that this concern could be addressed through the following changes to the proposed rules. a. Agreement on Valuation Methodologies and Inputs Should Only Be Required To a Practicable Level of Detail

As a provider of valuations for a variety of financial instruments including swaps and security-based swaps across various asset classes and regions, 6 we understand that valuing these products entails a number of complexities. If the Proposed Rule requires counterparties to agree upon valuation methodologies in advance that account for all minute details of these methodologies, complying with these requirements could become overly expensive and time consuming, and even impractical. For example, documentation designed to accurately assess the value of swaps on any given day until maturity of the swap would have to specify ex ante, among other things: (i) the exact mechanics of the pricing model used, (ii) the method to calibrate the model to account for market data, (iii) the parameters used for simulating future reiterations of the relevant pricing variables, (iv) the exact methods of interpolation and extrapolation, and (v) the approach for choosing comparable market data when market data required by the pricing model is not observable.
4 5 6

CSEs are swap entities subject to regulation by the Agencies. See Proposed Rule, 76 Fed. Reg. at 27589 (to be codified at 12 C.F.R. __.5(b).

Markit Portfolio Valuations provides valuation services for swaps and other financial products to investment managers, fund administrators and custody banks that represent thousands of end-users. During an average month, Markit Portfolio Valuations produces approximately 1,800,000 valuations across many products and asset classes.

July 11, 2011 Page 3

These inputs will be difficult if not impossible to agree upon at the time of execution. We therefore believe the Agencies should clarify that the Proposed Rule does not require counterparties to agree upon all of the specific details of methods and inputs that will be used for valuing a swap. Instead, the Proposed Rule could achieve the desired objective by requiring counterparties agreement only on a more general set of inputs and models for valuation purposes. b. Counterparties Should be Allowed to Delegate Determinations Regarding Swap Valuation to ITPPs

The Agencies could ensure that counterparties execute appropriate trading documentation without requiring them to agree on an impractical level of detail in advance by allowing them to reference the inputs, models, fallbacks, or the actual swap valuation provided by ITPPs for VM calculation purposes. This would ensure that the basis for VM calculation is flexible enough to react to future market events or changes in valuation methodologies because ITPPs can adjust their pricing models and inputs on an ongoing basis when appropriate. Market participants are already accustomed to delegating such tasks to ITPPs today. For example, many mutual funds in the U.S. solicit daily valuations from one or more ITPPs for use in calculating a funds net asset value. We therefore believe that the Agencies should permit counterparties to mutually agree in their trading documentation to reference the inputs, methods, procedures, and rules, or the actual swap valuation provided by an ITPP as the basis for their VM calculation. 7 2. The Proposed Standardized Lookup Table Disregards Important Factors The Proposed Rule would allow CSEs to calculate IM by consulting a standardized lookup table or by using an internal model that meets several enumerated requirements. 8 We believe that certain problems could arise from the use of the proposed standardized lookup table. An IM requirement that is solely based on the asset class, maturity, and notional amount of a swap or SB swap 9 is unlikely to accurately reflect the risk of a portfolio of swaps. CPSS/IOSCO, for example, recently acknowledged that margin requirements should be established to account for, among other factors, the liquidity of different products as well as the characteristics of individual swaps. 10 Other important characteristics to consider in this respect include the credit quality of the counterparty and the delta adjustment for option positions, as was recognized in some recently proposed rules. 11 Failing to account for the delta adjustment for option positions could result in an IM that significantly overstates the risk related to low delta positions. Any grid based IM should therefore at least allow delta-weighting for option positions, and should not exceed the premium that was paid upfront.
For example, trading documentation could state that the counterparties to the swap mutually agree to utilize swap valuations as provided by ITPP X, and, having performed sufficient due diligence on Xs methods, procedures, rules, inputs and alternative methods, the parties mutually agree that these are acceptable for the purpose of valuing this swap. If a higher degree of detail was required, the ITPP could provide the counterparties with documented methods, procedures, rules and/or inputs. Such document could be referenced as an addendum to the trading document. 8 See Proposed Rule, 76 Fed. Reg. at 27590 (to be codified at 12 C.F.R. __.8). 9 See id. at 27592. 10 See CPSS/IOSCO Guidance on the application of the 2004 CPSS-IOSCO Recommendations for Central Counterparties to OTC Derivatives CCPs at 15 (May 2010) (The appropriate liquidation time horizon may vary among products.); id. at 14 (In setting margin requirements, a CCP should use models and parameters that capture the risk characteristics of the products cleared (including historical price volatility, market volatility, and whether the products exhibit non-linear price characteristics).). 11 See Reporting by Investment Advisers to Private Funds and Certain Commodity Pool Operators and Commodity Trading Advisors on Form PF, 76 Fed. Reg. 8068, 8079 n.113 (published Feb. 11, 2011) (The form would require the adviser to report the total gross notional value of its funds derivative positions, except that options would be reported using their delta adjusted notional value.).
7

July 11, 2011 Page 4

However, we believe that any standardized table, by its very nature, will often result in IM requirements that are either too high or too low, and, as a consequence, create systemic risk or economically damaging results. These effects would likely alternate as market conditions change. Therefore, we question the utility of the standardized lookup table that is provided in Appendix A for IM calculation. In any event, given that the IM requirements generated by the lookup table will frequently exceed IM requirements that are based on pure risk considerations, we expect CSEs to generally prefer the use of an internal model. 3. The Use Of Internal IM Models Should Be Based On Sources That Are Independent or Independently Validated While Delegation Of IM Calculation To ITPPs Should Be Explicitly Permitted While we are generally supportive of the use of internal models for IM calculation, we note that certain problems could originate from their use. First, not all CSEs will have sufficient resources to develop and receive approval for their own internal model, which would force them to use the lookup table calculation. This, in turn, could result in an un-level playing field between larger and smaller CSEs that compete for business in the marketplace. Second, the Proposed Rule would only require that CSEs establish an internal process to validate their IM model. 12 While we support the requirement for IM models to be validated, we believe that requiring IM models to be validated only internally while not requiring validation of the data inputs or assumptions creates the potential for conflicts of interest. Third, market participants will need to accurately predict pre-trade the IM that will be demanded from various potential counterparties for a new transaction. However, this would be very difficult to achieve if most CSEs calculate IM based on their own proprietary models. We believe that these issues could be addressed if CSEs use independent inputs and models or if those inputs and models are independently validated, and by permitting CSEs to delegate the actual IM computation to qualified ITPPs. a. The Independence of IM Models and Inputs Is Important, as Is Independent Validation of Portfolio Offsets, Back Tests, and Stress Tests

Consistent with other proposed rules, 13 we believe that the Agencies should ensure that the inputs used for IM models are objective by requiring CSEs to use independent sources or by requiring ITPPs to validate the data used. We believe that using independent inputs and/or performing an independent assessment of input data, assumptions, stress tests, or portfolio offsets would provide all stakeholders with increased transparency and ultimately secure greater systemic risk reduction. Any validations should be performed on a recurring basis to take changing market conditions and advancements in methodologies into account. Notably, if an ITPP were to determine that reliable and independent data was not available, it could highlight the shortfalls of the data used and perform an analysis of the risk. Further, ITPPs could validate the appropriateness of portfolio offsets, back tests, and stress tests. 14 Calculating portfolio offsets requires one to consider complex correlation assumptions. Correlation

See Proposed Rule, 76 Fed. Reg. at 27590 (to be codified at 12 C.F.R. __.8(f)(2)). See Swap Trading Relationship Documentation Requirements for Swap Dealers and Major Swap Participants, 76 Fed. Reg. 6715 (proposed Feb. 8, 2011). 14 The Agencies requested comment on whether IM models should take some degree of portfolio offset into consideration and whether certain measures would adequately account for offsetting exposures, diversification, and other hedging benefits. See
13

12

July 11, 2011 Page 5

assumptions are subjective and are difficult to predict based on historical observations or theoretical considerations. Further, the correlated nature of the relevant risks will be interpreted differently by the various CSEs. ITPPs with the relevant quantitative expertise could provide an independent view into whether portfolio offsets are justified, thereby reducing the subjectivity inherent in such determinations. b. CSEs Should be Permitted to Delegate Margin Calculation

Further, we believe that all stakeholders would benefit if the Agencies explicitly permit CSEs to delegate the actual margin calculation to qualified independent third parties in addition to providing the choice of using a lookup table or internal model. ITPPs could increase transparency and comparability in the IM calculation based on the nature of their independent data inputs. Further, because ITPPs have no financial interest in the IM requirements that they calculate, delegation would alleviate conflict of interest concerns and the potential for disputes between the counterparties. Finally, IM calculation by independent third parties could provide counterparties with the much-needed ability to estimate and compare the expected IM requirement for various potential counterparties before deciding to enter into a new swap transaction with one of them. We note that both the FHFA and FCA have considered explicitly permitting their regulated entities to delegate the IM calculation to independent third parties. 15 The other Agencies, however, do not address this in the same manner but do not provide any particular rationale. For the reasons discussed above, we believe that all of the Agencies should permit CSEs to delegate their IM calculation to appropriately qualified ITPPs. 4. DCO Models Cannot Accurately Reflect the Risks Associated with Uncleared Swaps The Agencies requested comment on whether CSEs should be allowed to use models of derivatives clearing organizations (DCOs) for IM calculation of uncleared swaps or SB swaps. 16 While DCOs are well suited to provide valuations and IM for the more liquid and standardized swaps, we believe that several problems could arise from applying a DCO IM model to swaps that are not cleared by any DCO. Many swaps remain uncleared because they do not conform to the characteristics required by DCOs to clear them. This might be due to their insufficient liquidity, their high degree of customization, or the lack of relevant historical or current data. Since uncleared swaps pose greater risks to counterparties than cleared swaps, we believe that their IM cannot simply be determined by using models that were designed for cleared swaps. Even if those DCO models were altered, such modifications as well as the mapping and transformation between cleared and uncleared swaps would introduce significant uncertainty and risk to the IM calculation. Similarly, we are concerned with the Agencies proposal to require CSEs to solely benchmark any IM model against DCO models by ensuring that the initial margin required [by the CSE] is not less than what a derivatives clearing organization or a clearing agency would require for similar transactions. 17 We are skeptical that the sole use of DCO models in this context could provide constructive comparisons for uncleared swaps. Instead, we believe that a recent IM calculation for a given uncleared swap or SB swap performed by a qualified ITPP, in addition to other valuation methods, would represent a more meaningful benchmark.
Proposed Rule, 76 Fed. Reg. at 27573 (Questions 16-22). We believe that portfolio offsets should be taken into account for IM calculation, but that they must be calculated accurately in order to ensure that IM is neither too high nor too low in relation to risk. 15 See Proposed Rule, 76 Fed. Reg. at 27595, 27596 (to be codified at 12 C.FR. 624.11(a)(1)(ii); 1221.11(a)(1)(ii)). 16 See id. at 27573. 17 Id. at 27580.

July 11, 2011 Page 6

5. Expected Liquidation Periods For IM Calculation Should Vary to Account for Actual Risk The Agencies propose requiring a standard minimum time horizon for IM calculation of 10 business days for uncleared swaps or SB swaps, regardless of the product or asset class, 18 and request comment on whether this time horizon should vary. 19 We believe that the minimum liquidation horizon assumed for a swap or SB swap should indeed vary not only between asset classes but also between products, depending on their liquidity. As a provider of Liquidity Metrics and Scores for a variety of products, we have considerable experience making these types of determinations. We have found that the traditional approach of using only observed trading volumes to gauge liquidity fails to produce meaningful results because most individual swap instruments trade only infrequently, and that one should attempt to measure their prospective liquidity instead. The measurement of the expected liquidation period and the liquidity for swaps must involve the compilation of multiple inputs and will often vary significantly between products, and even maturities, within the same asset class. However, reasonably accurate gauges can be derived from a combination of observable factors such as trade frequency, standard market size and average transaction size. Accuracy can be further increased through the inclusion of factors such as bid/offer spreads, the agreement on the price, the number of market makers, and others. We also believe that the liquidity of a specific product at any time is only a snapshot, and that minimum observation periods must be used to ensure that the ongoing liquidity (and its variance over time) of an asset is taken into account. We therefore urge the Agencies to allow for the use of a variety of expected liquidation periods, and to take the above factors into consideration when deciding on what inputs and methods to accept for the determination of liquidation periods as element of IM calculation. * * * * *

We appreciate the opportunity to provide these comments on these proposed regulations. We thank the Agencies for considering our comments. In the event you may have any questions, please do not hesitate to contact the undersigned or Marcus Schler at marcus.schueler@markit.com. Sincerely,

Kevin Gould President Markit North America, Inc.

18 19

See id. at 27590 (to be codified at 12 C.F.R. __.8(d)(1)). See id. at 27580.

You might also like