Front Quant Dev.

Negative interest rates

Initial situation

The reduction in interest rates started to cause problems with lognormal models as rates approached zero, as the lognormal distribution does not allow rates to become negative. A quick solution was required for the bank to be able to value its IRD options contracts

Tasks

The chosen solution was to implement pricing using the normal distribution, which does allow negative underlying prices/rates

Actions

To achieve this, normal volatility surfaces were calculated from market prices and injected into the front-office risk management system, FlexING

Result

The bank was quickly able to use valuations calculated in FlexING, as an alternative to the Summit system which would have required much more effort to convert

OIS discounting

Initial situation

Following the credit crisis and the failure of Lehman, the spread between 3M Libor and the overnight interest rate widened and became more volatile to take account of the risk of counterparty default. It became clear that the previously widespread practice of deriving discount curves based on 3M rates was no longer appropriate as a risk free yield curve

Tasks

For collateralized trades, banks needed to move towards discount curves based on OIS instruments .Given that collateral is posted daily for derivative contracts against counterparty with CSA agreement, OIS discounting is seen as a better approximation to a risk-free interest rate

Actions

New discount curves were configured using OIS rates, and mapped to trades based on the CSA with the counterparty

Result

The bank’s pricing algorithms better reflect their counterparty risk

Quantitative Impact Studies (QIS) – European Banking Authority regulations

Initial situation

The European Banking Authority (EBA) has been monitoring a sample of EU banks since June 2011 in order to assess the impact of the Basel III rules, inviting them to participate in a semi-annual Quantitative Impact Study.

Tasks

As part of the study, Market Risk are required to generate Expected Shortfall (average loss in worst 5% of scenarios) figures for market scenarios which can change each quarter.

Actions

To help Market Risk respond to the requests from the EBA, we worked flexibly to generate the required scenarios and calculate the ES figures.

Result

As a result, Market Risk were able to provide the data on time, without a significant impact on their day-to-day work.