Posts

/by

c. notes – Evaluating Risk/Return Trade-offs When Margins Are Razor Thin

, ,

It is no secret that decisions today are more complex and far-reaching than ever before, and margins are razor thin. Traditional and non-traditional competitors on the battlefield keep multiplying and plotting to get more of consumers’ business, all while credit unions have to throw resources toward protecting their flank from attacks such as the CFPB, CECL, NCUA’s NEV test, and RBC.

This c. notes outlines advanced approaches to evaluating risk/return trade-offs so that decision-makers can have actionable business intelligence at their fingertips.

To continue reading, please click here.

/by

The Importance of Isolating Variables within Stress Tests

,

Stress testing is an important function of long-term interest rate risk modeling and risk management processes. As with any long-term interest rate modeling, stress testing requires both skill as well as an ability to fully understand and determine which assumptions influence the outputs and which assumptions are driving the results of the stress test.

During a recent model validation we performed, the importance of isolating variables within stress testing was punctuated once again. The objective of this particular stress test was to understand the impact widening credit spreads would have on asset valuations within the net economic value (NEV) simulation. In looking at the impact of widening the credit spreads, we observed that the overall NEV ratio was not as adversely impacted as one would have expected and as the asset devaluation would have implied.

Given that the overall results were not adding up, a deeper dive into the inputs was necessary. In analyzing the inputs, we discovered the model setting that widened the credit spreads on the assets had also increased the rates on the borrowing yield curve, which were being used to value non-maturity deposits (NMDs). The increase in the borrowing yield curve rates was thereby offsetting the asset devaluation caused by widening the credit spreads.

It is perhaps both reasonable and defendable to say that an increase in asset credit spreads could also be strongly correlated with an increase in borrowing rates. Indeed, both the science and math behind this stress test could be wholly appropriate. That said, however, increasing the rates on the borrowing yield curve unnecessarily detracted from both the spirit and objective of the stress test, which was to understand the impact widening the credit spreads would have on asset valuations.

The importance, therefore, of isolating variables when running stress tests and doing sensitivity analysis cannot be underestimated. It is both a science and an art, and interpreting whether the results of the stress test accomplish the objective and make sense should not be lost within the analysis.