Denny Yu, Product Manager Risk for Numerix, explores one proposed methodology for the funding value adjustment (FVA) and its use in regulatory and accounting areas.
Introduction
As a result of the financial turmoil in capital markets in 2008, banks and derivative participants were forced to introduce a plethora of new accounting and regulatory adjustments to their derivative valuations and risk management processes. The price of a vanilla swap was no longer just the net discounted values of each leg but included adjustments for the counterparty credit quality (CVA), the firm’s own credit quality (DVA), liquidity (LVA) and costs of funding (FVA). While much has been written in literature on the calculation of CVA and DVA, the industry debate is now turning to the calculation and relevancy of a firm’s funding value adjustment (FVA). This article will focus on one proposed methodology for the FVA adjustment and its use in accounting and regulatory arenas.
Prior to the financial crisis, when calculating the trading P&L banks took into account fluctuations in the portfolio value due to changes in market risk factors. For OTC derivatives that were collateralized, banks assumed there would be liquidity in the marketplace and did not explicitly consider financing costs in the portfolio P&L. At this point, LIBOR was often the benchmark banks used to charge for financing as well as the benchmark for crediting interest on posted collateral. Thus, collateral financing costs were largely offset by the interest received on the collateral they posted and the final result was that the portfolio P&L typically did not include cost of funding.
However, as the crisis deepened, banks were less willing to lend at terms longer than overnight. As a result, the lending rate which was typically benchmarked to the LIBOR rate changed to the overnight indexed swap rate for that market. In the US, the Fed Funds rate was the overnight rate used while in Europe, EONIA and SONIA became the standard. This change to fund collateral at the overnight rate resulted in a fundamental change to valuation of collateralized trades. Furthermore, banks were charged funding spreads on top of overnight rates with different spreads based on the credit quality of the borrowing bank. Banks now found that the cost of funding collateral played a significant part in the P&L of its trading portfolio. Therefore, banks began looking for a way to measure and account for this funding cost that could change over time due to the bank’s own credit quality and the notion of a funding value adjustment emerged.
Funding Value Adjustment – A Proposed Methodology Defined
Now, profit and loss included not only changes due to market factors, but also non-trade related factors such as counterparty credit quality, collateral financing costs and interest received on posted collateral. Banks had to face the new trade P&L reality which included the market P&L, and a host of additional adjustments.
Before, delving into the FVA calculation, let’s review the financing operation within a bank that is active in derivative transactions. Typically, financing for collateral posting is facilitated by a funding desk inside the bank. The funding desk borrows from the external markets and pays a spread that reflects its own credit quality. In turn, the funding desk charges the trading desks for financing of collateral and can charge the external funding spread (internal transfer pricing agreement) or add an additional spread on top. The external funding spread for the bank reflects the credit quality of the bank and can be a significant amount for lower-rated institutions.
Though the methodology is still evolving, one school of thought we focus on in this paper discusses FVA as the sum of a funding charge adjustment (FCA) and a funding benefit adjustment (FBA). It is measured at the portfolio level where FVA is calculated for the trades that are part of a specific credit support annex(CSA).
Let’s begin with a suggested definition of the funding charge adjustment. The FCA is a calculation that can be done by multiplying the collateral posted according to a CSA by the funding spread charged to the trading desk. The funding charge adjustment is always a negative charge as it reflects cash outflows from the bank to the marketplace. The more collateral that is required by a trading desk, the higher the FCA.
To calculate the Funding Benefit Adjustment we first look at a two-way CSA agreement. In a typical two-way CSA, negotiated parameters such as margin frequency, collateral thresholds, minimum transfer amounts and rounding lead to uncollateralized exposure. CSA parameters introduce friction into the collateral posting and receiving process because there will always be some credit exposure under these parameters. This amount of unsecured derivative value provides a benefit to the counterparty with negative exposure because it essentially represents collateral that is not posted and is a financing savings. This is known as the funding benefit adjustment (FBA) and can be calculated as the uncollateralized negative exposure multiplied by the funding spread. A perfect CSA, one with no friction, assumes that there is no credit exposure due to negotiated CSA parameters mentioned earlier.
In a CSA with no friction, collateral is immediately posted when it is called since the friction in the way of CSA parameters does not exist.
FVA = – Funding Charge Adjustment + Funding Benefit Adjustment
FCA is the funding spread charged to trading desk * actual collateral financed
FBA is the funding spread charged to trading desk * additional collateral that would have been financed under a perfect CSA
In the case of a one-way CSA agreement, typical when trading with sovereigns, supranational and agencies (SS&A), the firm that is posting collateral has only FCA and the SS&A firm receiving the collateral has only FBA.
As Dodd-Frank Act (Dodd-Frank) in the US and European Market Infrastructure Regulation (EMIR) in Europe push derivative trades to central clearing, FVA can serve as another potential criteria for banks deciding on which trades to move to central clearing. Most banks consider default by central counterparties (CCPs) minimal and often do not calculate a CVA for trades that are cleared. On the other hand, these CCPs often provide default protection by imposing overcollateralization of centrally cleared trades.
For banks deciding to trade bilaterally or centrally clear trades, the economic impact can be very different for either choice. Suppose we have an AA-rated bank that is deciding to enter a transaction with an A-rated bank or clear the same trade through a clearinghouse. In the case of trading with the A-rated bank, the collateral required will likely be less than the full exposure in the case where the AA-rated bank has negative exposure.
Conversely, the same trade with negative exposure will require the posting of an initial margin with the clearinghouse that is usually for the full exposure amount to cover clearing member defaults. In addition, the AA-rated bank may also have to post additional collateral daily in the name of variation margin to cover exposure fluctuation due to market volatility.
The choice of central clearing now involves reducing counterparty credit risk at the cost of material additional financing costs to the bank. The cost of central clearing for the trade can be measured by calculating the FVA of the trade where the bank only has a funding charge adjustment as a result of having to post initial margin. In this example, the CVA charge of doing business with the A-rated bank can be compared to the FVA of centrally clearing the trade.
Impact on Derivatives Valuation
As a result of the adjustments made on the derivatives portfolio due to credit of the counterparty and the self, the portfolio profit and loss for the bank is no longer the change in value of the trade itself but includes a CVA and FVA adjustment.
Trading desk Profit & Loss = (P&L of the trade) + CVA + FVA
There is an interesting idea emerging on a bank’s use of FVA to manage its trading portfolio. Since the funding desk can adjust the funding charge it applies to the trading desks, it can apply a smaller charge for trades the bank wants the trading desk to take on and apply a higher charge for trades the bank does not want the trading desk to enter. This may be done to drive trading with certain counterparties, sectors, ratings or other categories of trades or trading partners.
Conclusions
There is no clear consensus in the industry on the FVA calculation and as various methodologies continue to be debated many banks argue its compliance under IFRS International Financial Reporting Standards (IFRS) and Generally Accepted Accounting Principles (US GAAP) accounting rules governing the fair value measurement of OTC derivatives. While most banks do not include an accounting entry for FVA in their financial statements, more and more banks believe that funding is becoming as important as credit risk in derivatives valuation and expect FVA to be included in their accounting reports in the near future.