Calculated decision - JSE MAGAZINE

Calculated decision

Implementation of the latest JSE Clear value-at-risk approach has resulted in a notable amount of initial margin being returned to investors

Calculated decision

The JSE’s new methodology for enumerating margin requirements for its interest rate derivatives market has seen approximately R4 billion in initial margin   returned to participants. Crucially, this was achieved without increasing JSE Clear’s (the derivative market central counterparty) risk exposure, or diluting its level of risk management.

JSE Clear’s value-at-risk (VaR) methodology – which replaces the former conservative JSPAN framework – considers the market risk profile of a participant’s entire cleared portfolio. Previously, it would simply aggregate the results after considering the profile of each individual position, without acknowledging the fact that a participant’s exposure was potentially only to a small section of the yield curve.

‘In essence, the new methodology maps a participant’s entire portfolio to a single exposure across the yield curve – the curve depicting the relationship between interest rates and time to maturity,’ says Paul du Preez, Quantitative Risk Manager at the JSE.

He offers an example of a participant with positions in the R186 and R209 bonds. These are two of the most liquid on the SA market and mature in 2026 and 2036 respectively. ‘A participant with a position in the R186 will thus be affected by a change in the 0 to 10-year region of the curve, while a participant with a position in the R209 will be affected by a change in the 0 to 20-year region of the curve. However, in the futures market, participants can have both long and short positions – with a long position making money when interest rates decrease, and a short position making money when interest rates increase.’

Du Preez adds: ‘A participant with opposite positions – one long and one short – in the R186 and R209 is thus only exposed to changes in the 10 to 20-year region of the curve, since the opposite exposures in the 0 to 10-year region offset each other. This fact is now incorporated into the initial margin framework.’

In this example – and under the new methodology – the above spread position would require approximately 85% less initial margin than was required under the old methodology. ‘In the interest rate derivatives market, participants often make use of various instruments to express a view on a very specific portion of the yield curve. The above example shows how the R186 and R209 could be used to express a view on the 10 to 20-year region of the curve, but this is but one example in a virtually infinite range of possibilities,’ he says.

‘Without the ability to map a participant’s exposure to single exposure across the curve, portfolios are essentially margined at a gross level and not a net level, which is not representative of the true risk to which participants are exposed.’

Addressing the shortcomings of JSPAN means that there is no longer a situation where the amount of initial margin being called from market participants is excessive. The excessive margin levels that were associated with JSPAN unnecessarily drained liquidity from the system and actually prevented certain exposures from being cleared – potentially preventing participants from effectively hedging their interest rate risk.

Price-making banks were the primary beneficiaries of the return of billions of rands in initial margin on the implementation date of VaR. ‘These banks now face less cost with regard to funding their initial margin requirements, which creates potential for better pricing and hence more liquidity,’ he says.

For now, this new methodology is only in place on interest rate derivatives markets. According to Du Preez, the JSE intends to move all derivatives markets onto a similar margining methodology, which the implementation of its integrated trading and clearing initiative enables.

By Hilton Tarrant
Image: Gallo/GettyImages