Case Study: The Impact of an Interest Rate Hike on Mortgage Default Rates

Alexander Denev

The analysis we present in this chapter examines the application of Bayesian nets to the potential increase in default rates, income and loss given default in a portfolio of UK household mortgages of a bank due to an increase in the Bank of England’s Key Rate. We want to investigate the effects of this policy decision in a time interval of less than 1.5 years. We do not want to commit to a precisely specified time horizon, as the timing of the impacts of some policy decisions cannot be determined exactly.

We shall use a combination of a dynamic Bayesian net (which we shall train on real-world data, to model a time series of macroeconomic variables) and another non-dynamic Bayesian net. Finally, we also show how to perform reverse stress testing along the lines discussed in Chapter 4.

The decision by the BoE of whether to increase the base interest rate, which is 0.5% at the time of writing, depends on many crucial elements: UK’s inflation, and inflation expectations; import costs, eg, commodity prices; the value of sterling; the outlook for the economy; household income and indebtedness. Information about these key variables changes frequently, and hence the probabilities in