In a recent DerivSource webinar, Arthur Rabatin, CIO, derivatives counterparty and funding risk at Deutsche Bank, offered a brief introduction to the new BCBS standardized approach for measuring counterparty risk (SA-CCR) and explored some of the implications of adopting this new methodology. Read on for insight on how SA-CCR impacts data management and understanding of CSAs and your trades is key to successful implementation.
SA-CCR is a very important and long overdue change in terms of how the industry looks at counterparty risk—and it is impact will extend well beyond counterparty risk. Firms that are not directly impacted at the moment need to keep an eye on developments as they may be affected in the future.
SA-CCR replaces the current exposure method (CEM) and the standardized approach (SM). It will have a big impact on collateralization, requiring risk-sensitive treatment of collateral and the recognition of initial margin (IM). Marginings for both the variation as well as IM impacts both the replacement cost as well as the potential future exposure (PFE) add-on calculation.
Hedging sets are being introduced in addition to netting sets. Hedging sets are a categorization of asset classes, within which netting is allowed. There’s quite a strict definition and there will be no diversification allowed across different asset classes.
The methodology of SA-CCR is very prescriptive, which was one of the regulators’ objectives. It serves as a tool—as do all the methodologies with the standardized approach (SA) prefix—for regulators to not just better control individual firms, but to get a better view of the industry and how systemic risk works. By allowing less discretion for local authorities, regulators can have a global view to compare banks across different regulatory regimes.
There will be new supervisory factors, which will change the risk factors—a direct result of the crisis. It’s a reflection of post-crisis volatility, and the fact that negative mark-to-market (MTM) for derivatives has an impact on credit risk is also taken into account.
Key Definitions for SA-CCR Calculations
Most firms are familiar with CEM calculations—there will still be a component of replacement cost plus a PFE component. There’s an additional refactor or an alpha factor of 1.4 that we multiply it with, and that is consistent with the alpha factor that’s applied to the internal model.
A major change is that replacement cost now includes margining—variation margin (VM) that reduces MTM, as well as the IM part of the replacement cost calculations. IM behaves quite differently from VM. In a VM world, collateral can be reposted. IM—certainly how the regulations currently define it—is segregated and cannot be reused or rehypothecated. That makes it a very robust separate collateralization. The threshold amount and minimum transfer amount are key features of the credit support annex and of collateralization. The higher the threshold amount, or the higher the minimum transfer amount, the greater the exposure to a counterparty.
With regards to the PFE, the key calculation is the so-called aggregate add-on, which is adjusted by a multiplier. The multiplier is another factor that allows exposure to be reduced through the recognition of initial margin. IM is by definition over-collateralization, because it is calculated at the inception of the trade irrespective of the P&L on any given date. The multiplier will never be higher than 1 because it will always reduce the exposure. But it will never be reduced to zero, so the regulators have introduced a flaw of 5% on the value.
The add-on aggregate is simply the sum of the add-on’s asset class. The PAC add-on is a way to express a potential future exposure through risk sensitivities of a trade.
The advantage of SA-CCR is it’s a very prescriptive process for following the Bank for International Settlements (BIS) guidelines. Worked examples are available so firms can use either spreadsheets or build quick tools to cross-check the calculations they are building into their larger systems.
Data Challenges and Requirements
Transactions need to be broken down by netting set, which effectively provides an individual netting set with the counterparty. Primary risk drivers are assigned per asset class, before transactions are grouped into hedging sets.
Not every trade will fully belong in one asset class, so BIS has said there will be a primary risk driver. If firms have a really complex book, they will need to look at the split with local regulators. Once the netting set and asset classes have been defined, there follows the initial grouping of trades. Trades are standardized in terms of the duration of the trade, which is the maturity, and the trade notional. This is very prescriptive, very formulaic, and is carried out by asset class.
“SACCR is very prescriptive, with a strong focus on margining and on the definition of CSAs. The advantage is there are examples to use and it is easy to test during implementation.”
Next comes the standardization of maturities—one of the areas where margining is very important. For an uncollateralized trade, the maturity is the time maturity of the trade. If the trade is highly collateralized, or is collateralized, then the maturity becomes a margin period of risk calculation. There is a very big difference between collateralized and uncollateralized. The supervisory deltas, which are adjustments particularly for options trades, will be relevant.
Once the trades are standardized, firms can re-aggregate the standardized trades into buckets, which can then yield the add-on for each asset class. They then apply the supervisory factors defined in the BIS document for each hedging set, and add the different add-ons for each hedging set. There is no diversification effect allowed across different asset classes. That will then yield a PFE add-on for the entire netting set and counterparty.
SACCR is very prescriptive, with a strong focus on margining and on the definition of CSAs. The advantage is there are examples to use and it is easy to test during implementation.
Understanding Trades and CSAs is Key
The actual risk calculations, and analytics around the calculations, are actually straightforward, and there are worked-out examples to follow. However, in order to do these calculations, firms have to have a very good understanding of their trades and client service agreements (CSAs), especially as some CSAs pre-date the financial crisis. There are a lot of documents and they can be very complex. Understanding trades, classifying them correctly, identifying the trade types is key. Once that is done, the calculation itself should not be a challenge.
When it comes to theoretical challenges, there is hopefully not a methodological issue, but there could be challenges in defining what constitutes the notional of a trade or what asset class a trade is part of, which is not normally straightforward. Although the regulators have worked very hard to eliminate all of these interpretations, there are some left, and depending on the nature of the trading book, these challenges could grow.
For example, identifying the notional of a trade is not straightforward if it involves roller coaster-type notional swaps, or hybrid books, where there are different asset classes in one trade. In particular, structured hedging activities for real clients is an area where there is a lot of complexity of the trades. However, the data challenge, understanding the trades and CSAs is probably the biggest hurdle for most organizations.
Overlap with FRTB
The whole space is very much evolving, especially the regulator view in terms of the capitalization of market risk. Traditionally, CVA and credit risk from an risk-weighted- asset (RWA) point of view were two different things. CVA is considered a market risk rather than a credit risk measure. The key to that is to adjust the price, by calculating a fair value of the derivative’s price.
Over the last year, thanks to regulatory pressure, there has been increased collateralization and pressure to move to margining. More recently, FRTB has moved to what’s called a standardized approach for CVA, which is taking the sensitivities out of the accounting CVA world, to calculate a capital CVA number on the back of that. The SA prefix is cropping up more and more. Firms need to observe that and will likely receive more clarity from regulators over the coming year or two. The two areas are evolving, and especially with regards to margining, there will likely be much more alignment.
Which department owns SA-CCR?
A lot of credit or capital calculation used to be carried out not in the front office, but in a middle office or in a market research-controlled environment. What has changed, or is currently changing, is that there has to be much more ownership of certain calibration within the front-office functions—with trade processing, trade booking, with client-facing functions—because understanding exactly how trades are booked is absolutely critical to getting the SA-CCR calculation right.
Firms should think of SA-CCR as an opportunity rather than just a cost or a tax on their business, because everything they need to do in terms of understanding trades and legal agreements should be crucial to running an efficient operation. If any of that is a challenge, then probably addressing that challenge is overdue whatever the regulatory demands.
To listen to the full webinar on the implementation challenges and best practices for SA-CCR, click here.