Numerix Proposes an Efficient FVA Calculation Helping Institutions Better Manage the True Cost of OTC Derivatives Funding & Trade Profitability
Numerix (www.numerix.com), the leading provider of cross-asset analytics for derivatives valuations and risk management, today announces a new methodology for calculating Funding Value Adjustment (FVA) for vanilla and exotic deals at both the trade, and portfolio level. Available within the market leading Numerix CrossAsset pricing and risk analytics suite, the universal framework proposes an efficient, practical implementation of the FVA calculation.
The Evolution of Funding Dynamics – Going Beyond OIS Discounting
Post-2008 bank funding costs diverged significantly from risk-free discounting rates resulting in widely disparate derivatives prices. While OIS has emerged as the proxy for the risk-free rate and continues to emerge as the new standard for collateral discounting, a multitude of funding rates are utilized across a portfolio dictated by collateral agreements both to secure funding and manage risk. Though the debate around FVA and how it should be incorporated into the valuation process continues to be discussed, it is clear that collateral is inextricably linked to funding; representing a real charge passed on to traders.
“Collateral was first introduced to mitigate Counterparty Credit Risk, but institutions have realized that if collateral affects the funding costs of a trade, it also affects its profitability, raising questions over how an institution can ultimately pass this cost on to counterparties or build it in to the mark-to-market price of the transaction,” said Satyam Kancharla, Chief Strategy Officer and Senior Vice President at Numerix. “However, with no market consensus on how to report FVA or which funding rate to use, managing the expense of funding is a difficult task. Nevertheless, it’s absolutely imperative that banks understand FVA, how it is calculated and the associated costs so they can make the best trading decisions for their business in terms of profitability.”
A Proposed Methodology for FVA Calculation
The methodology proposed by Numerix for FVA expands upon Vladimir Piterbarg’s framework for valuation in the presence of real collateral. [1] The Numerix FVA implementation is one of the first that allows the computation of the FVA calculation for arbitrary instrument types; even the most complex and bespoke instruments with optionality, callability and triggers. It also allows for an efficient deal-by-deal computation, which is attractive for parallelization. Built on the same foundational analytics and algorithmic approach for calculating Counterparty exposure[2], an American Monte Carlo approach is employed to determine the distribution of prices under future dates resulting in accurate, real-time FVA computations for both exotics and vanilla instruments.[3]
“The Numerix approach for FVA3 simplifies Piterbarg’s complex formula, where we have derived and numerically justified a fast approximated methodology – demonstrating that this computation agrees exactly with the approaches in literature for simple instruments, and achieves a high degree of accuracy for instruments with optionality,” said Serguei Issakov, Global Head of Quantitative Research & Senior Vice President. “We provide a concrete example of FVA calculation for a Bermudan swaption with partial collateralization, proving that the quality of the numerical approximation is excellent. To our knowledge, we’re the first to implement a general approach of Piterbarg’s formula, and such a comparison has not been published previously due to numerical difficulties with the exact multi-rate price of the Bermudan option.”
“As banks come to terms with the true cost of funding collateral and the significant role it plays in overall profitability, the Numerix approach gives users the tools and flexibility to apply FVA in a way that suits their specific needs and methodologies. With Numerix, institutions can not only measure and account for funding cost, but more effectively manage how it could change over time due to the bank’s own credit quality; and ultimately capture PnL impact,” said Steven R. O’Hanlon, Chief Executive Officer & President of Numerix. “The goal of Numerix quantitative research is to increase model transparency and understanding for today’s OTC derivatives market. We’re extremely proud of what Serguei’s team has accomplished and look forward to the efficiencies and innovation it will bring to the marketplace.”
