Page 42 - Monocle Quarterly Journal Vol 1 Issue 1 Q4

BANKING
“One of the challenges faced in modelling future cash ows, is the fact that bank rates are often determined by the bank itself, rather than by the market.”
ere are also other behavioural aspects that impact the assets and liabilities side of the balance sheet. In particular, banks tend to lend on an internal or ‘prime’ rate, which is adjusted non-linearly to prevailing market rates. And, much liquidity risk modelling has excluded the impact of business growth and budgeting and forecasting targets. What we need to do, then, is include interest rate sensitivity models and asset/liability portfolio growth models with a holistic consideration of liquidity risk behavioural models.
Interest Rate Sensitivity
Cash ow amounts are usually dependent on the amount of interest charged on/accrued to the asset/liability. One of the challenges faced in modelling future cash ows, is the fact that bank rates are often deter- mined by the bank itself, rather than by the market. is is problematic when predicting the bank’s response to changes in market interest rates.
Our methods would estimate the relationship between a change in market rates, and the resulting change in internal lending and deposit rates. is is then input to a simulation to determine the level of interest charged on individual and corporate accounts, and therefore the value of projected cash ows.
For products with a reference rate based on an internal lending or deposit rate, rather than a market rate, Monocle has a methodology to predict future levels of internal lending and deposit rates, given the levels and changes in market rates.
Historically, there’s been a strong relationship between internal rates and market rates, but this may not always be the case. As market rates decrease, banks generally apply these ‘savings’ to their customers by reducing internal lending rates, to remain competitive. Similarly, as market rates rise, the banks pass these costs onto their customers by raising internal lending rates.
By looking at 5 to 10 years of market rates of all tenors, we can estimate the relationship between these rates and internal bank rates over the same period. is relationship can then be embedded within a statistical model which translates movements in market rates into a probability of change for internal rates. When the probability of change reaches a key value, the reference or prime rate is assumed to increase or decrease, depending on recent movements in market rates.
40