Banks tout break clauses as capital mitigant

Typically, there are two different counterparty risk charges where break clauses could have an impact – the basic default risk capital charge and the CVA charge. In both calculations, the capital requirement depends on the size of the counterparty exposure, so reducing the exposure – which is what a break clause does, banks argue – should mean the required capital is also cut.

But there is little scope to achieve reductions in the default risk charge, as banks are required to calculate exposure over a one-year horizon, with the effective maturity of the trade capped at five years. As a result, break clauses – which are often inserted at intervals of five years – would not shorten the period over which counterparty exposure is calculated. The relief that could be obtained on the CVA calculation is more meaningful, because the exposure has no maturity cap: if the trade has a maturity of 30 years, the counterparty exposure for CVA purposes is calculated over a 30-year period.

Dealers argue there are two ways to approach the issue. One is to regard the break clause as a ceiling on the swap’s maturity – so, if a bank had a 30-year swap with a break clause at 10 years, it would model the exposure up to the 10-year point. The other approach is to model the exposure for the full life of the trade, but to include scenarios in which the counterparty’s credit deteriorates and the clause is exercised, and ones in which it is not. The result would be a kind of average maturity that reflects the probability of a break being exercised. Either approach would result in the exposure – and the resulting capital charge – being reduced.

Citi argues for the first of the two approaches. If the bank determines it is highly likely to exercise the break clause, the duration of the trade should be the same as the point at which the break is inserted. Up to that point, however, the swap should still be treated in risk and capital terms as a longer-maturity trade, it believes.

“If we determined that we would exercise the break clause in a 30-year swap at 10 years, up to the 10-year point the swap would still display the dynamics and economics of a 30-year swap. But at the 10-year point we would simply stop recording any contribution to the netting set from that trade. So the trade moves with the sensitivity of a 30-year swap, but at 10 years we essentially stop the exposure profile,” says Citi’s Wayne.

Others have questioned the approach. “In that situation, you need to be absolutely sure you exercise the break when you say you are going to, and you have to have a bulletproof governance policy in place to ensure that happens. The big question is what happens if you are out-of-the-money on the trade at that point? Do you still pull the trigger?” asks the head of interest rates swaps trading at one European bank.

Citi insists its governance policy tackles those issues, but requests for further detail had not been answered by the time Risk went to press.

Other banks favour the more refined approach in which probabilities are assigned to the exercise of a break clause based on a related metric, and the capital is then reduced by an appropriate amount – some sources suggest basing exercise probabilities on the credit rating or credit default swap spread of the counterparty, or an economic metric such as the extent to which the bank is in-the-money.

In a simple example, if the bank estimated a 25% chance of the counterparty’s credit spread widening to 500 basis points at the point of the break, which would cause the bank to exercise, the exposure after the point of the break would now retain a weight of 75%. But that would be one scenario among hundreds when calculating the exposure.

Banks could not expect a regulator to accept that kind of judgement without a clear, reliable governance process, warns Jon Gregory, a counterparty credit risk consultant at Solum Financial and former global head of credit quantitative analytics at Barclays Capital. “You need to be able to establish a clear link between the relief you are claiming, how you intend to model the exposures, and what actually happens in real life. If you decided to model exercise probability using credit spreads and the extent to which a trade is in-the-money, that needs to be reflected in the bank’s governance policy. The bank would have to state that if the counterparty’s credit spread widens beyond a threshold level – while the bank is in-the-money – then it intends to exercise the break. This link will need to be proven to regulators,” he says.

But exercise decisions are not based solely on the level of counterparty risk and the value of the swap, says one rates structurer at a European bank in London. “You also have to take into account the portfolio effects. For example, if a trade has a break clause, and the counterparty’s credit spread blows out, what happens if – by exercising that option – the bank would break a netting set? These are considerations that will also have to be factored in. It’s not a simple process.”