Theoretically, the final version of Basel II was published in June 2004.
Practically, nothing is really ‘final’ because both the financial industry and its instruments are in full evolution, making necessary new rules so that regulation is ahead of the curve. This section highlights issues raised by Basel’s consulta- tive document of April 2005 on trading activities.4
The first subject that should hold the reader’s attention in respect to this docu- ment is the creation of a family of securities financing transactions(SFTs), which eventually may become as important as OTC transaction are today. These SFTs include:
● Securities lending
● Securities borrowing
● Securities margin lending, and repurchase, and
● Reverse repurchase agreements.
Indeed, in this April 2005 consultative document, the Basel Committee com- pares and contrasts SFTs to OTC derivatives, making the point that generally SFT transactions are undertaken with a counterparty against which a probability of default can be determined. Also, they:
● Generate a current exposure
● Have a random (read: unknown) future market value, and
● Create an exchange of payments, or of financial instruments against payment.
The implication of the factors listed above is that financial institutions will be well advised to use a methodology similar to tracking position risk in the loans book, as explained in section 6. While some of the mechanics will differ, the gen- eral systems approach remains valid.
A similar statement is valid in connection to another major issue raised by the April 2005 Basel consultative document: counterparty credit risk(CCR). This is the bilateral credit risk of transactions with uncertain exposures, that can vary over time with the movement of underlying market factors – hence the wisdom of tracking them intraday, interactively.
As a term, CCR refers to the likelihood that the counterparty to a transaction could default prior to final settlement of the transaction’s cash flows (Herstatt risk). This
contrasts to the credit institution’s exposure to a loan, because in the latter case credit exposure is unilateral and only the bank faces a risk of loss. By contrast,
● With CCR the risk of loss is bilateral, and
● The market value of the transaction can be positive or negative to either counterparty.
Notice that the treatment of CCR arising from OTC derivatives was first set forth in an amendment to the 1988 Basel Accord. What Basel II does is to update this treatment for transactions booked in either the trading book or the banking book.
It also advances repo-style treatment of CCR and other types of transaction.
For starters, the basis of existing treatment of OTC derivatives, known as the cur- rent exposure method (CEM), is that of reflecting potential future exposure, cal- culated by applying a weighting factor to the notional principal amount.
However, because the risk-sensitivity of this treatment appears limited, particu- larly with regard to the internal ratings-based (IRB) method, supervisors propose to enhance this procedure for OTC derivative transactions.
Enhancement is provided by introducing a new treatment for securities financing transactions (discussed in the opening paragraphs of this section). The consulta- tive document referred to advances three alternative methods for calculating expo- sure at default (EAD) or exposed amount for transactions involving CCR in the banking book or trading book:
● An internal model using expected positive exposure (EPE)
● A standardized method, and
● The existing current exposure approach (CEM).
These alternatives represent points in a continuum of sophistication in risk man- agement, and they aim to provide incentives for banks to improve their handling of CCR by adopting more accurate approaches. Given their dependence on cur- rent information, every one of them, particularly so the most advanced, can ben- efit from virtual balance sheet solutions.
One of the more interesting practical issues connected to CCR, which can bene- fit from interactive computational finance, is the tracking of wrong-way risk. This is a new term introduced by Basel and used for identifying synergies in credit exposure. Two types should be distinguished:
● General wrong-way risk, which arises when the probability of default of counterparties positively correlates with general market risk.
● Specific wrong-way risk, the result of exposure to a particular counterparty, which positively correlates with the counterparty’s probability of default (PD).
Real-time access to databases, experimentation, simulation, and the use of knowledge engineering artifacts are valuable tools for wrong-way risk studies.
The same statement is valid about the calibration of stress probability of default (SPD) and stress loss given default (SLGD).
The challenge of computing unexpected losses (UL) is a key reason why the requirements for internal loss given default (LGD) calculation in the A-IRB method of Basel II had to be rewritten. Theoretically, but only theoretically, the LGD parameter can be seen as an average, default-weighted, loss ratio – which is taken as not being associated to a particular economic scenario. Practically, this is not the case. As far as the LGD parameter is concerned, the down scenario must be included in estimates. Therefore, it has to be entered into the UL risk-weight func- tion. Along this frame of reference, regulators essentially promote three LGDs:
● Mean LGD, taken at lower level
● Downturn LGD, taken at crisis time, and
● Expected LGD, which accounts for the current economic environment.
The downturn LGD can be calculated from LGDs in time periods characterized by large credit losses. Mean LGD and downturn LGD might be identical in the case of credit exposures where the LGD is independent of cyclical movements. The careful reader will appreciate that these notions open up a huge perspective for experi- mentation involving all material positions in the loans portfolio (see also section 4).
With Basel II, downturn LGD is applied to non-defaulted loans, both when deter- mining UL and when determining expected losses (EL). This is a simplification which permits banks to use only a single estimated value of LGD to determine the regulatory capital requirements, calculated for each individual category of:
● Assets, and
● Collateral.
Similar notions also apply to probability of default in market downturns or if sys- temic risk becomes significant, owing to adverse economic or other conditions.
This concept leads to SPD, by converting the input probability of default into a stress PD applying an appropriately adjusted UL risk-weight function. The latter would be prescribed by supervisors.