It's common today to discount the importance of that old dependable asset liability standby, the gap report. Yet, the gap report still provides an important window on interest rate risk. Let's review basic gap analysis, and then look at two banks and their gap results.
Gap is the difference between the amount of assets and liabilities on which interest rates are reset during any particular bucket of time. If a bank has both $5 million in assets and $5 million in liabilities that reprice in any given time window, changes in interest rates should not change the bank's net interest margin. This is known as a balanced gap position.
If instead, $10 million in assets reprice with only $5 million in liabilities repricing, the bank is in an asset sensitive position. An asset sensitive bank will enjoy a net interest margin increase if interest rates increase. Of course, as we've seen over the past few years, the asset sensitive bank will have net interest margin compression if rates fall.
The converse situation, with $5 million in assets repricing during the same period that $10 million in liabilities reprice is known as a liability sensitive position. Here, if interest rates increase net interest margin will decline. Similarly, if interest rates fall the liability sensitive bank will anticipate a wider net interest margin.
One of the key criticisms of gap analysis is that it fails to account for optionality in assets and liabilities. That is, if rates fall and your assets prepay faster, or if rates rise and the average life of your assets extend, this information is typically not given by a simple static gap report. Another criticism concerns repricing assumptions on non-maturity deposits. This is a critical assumption and a recent blog post shows our perspective.
In order to deal with these criticisms, when we create a gap analysis we don't stop with the static information from the ledger. Instead, we examine each line item and every time bucket of assets and liabilities for potential prepayment activity. Some line items such as U.S. Treasury securities, are typically non-prepayable bullets and no adjustment is needed. Loans however, typically amortize, as well as prepay at different speeds for different interest rate environments. By adjusting our base case gap analysis for these, and other, likely behavior traits, we begin the process of converting static accounting data into a dynamic asset liability format. We further extend the analysis by examining the change in the bucketed cash flows for each of the rate shock environments.
So now that we have literally pages of detailed cash flow information, how do we evaluate it? It's here that a picture truly is worth 1000 words. Let's take a look at two examples. The chart shows cumulative gap for each of the time buckets and how it changes in each interest rate shock scenario.
The first bank shows a cumulative gap that is a negative at the one year point, indicating a slight liability sensitive position. Note however that all of the lines representing the various interest rate scenarios are tightly bound. This indicates that regardless of the movement in interest rates the bank's gap position is not expected to change significantly.
Now let's look at the second bank. Here, the cumulative gap position is positive, indicating asset sensitivity, in the one year bucket. Note however the significantly wider variability in gap given changes in interest rates. The message is clear. As rates change this bank's gap position, especially in the one to three year bucket, will vary substantially. If you were to look at the underlying cash flow detail, he would find this bank holds significant volumes of fixed rate securities and loans that reprice in the one to three year window. By examining the area of widest variability in gap, you have a head start on identifying the time periods with the most inherent interest rate risk.
In fact, if we were to look at the other measurements of interest rate risk for the same banks we would find that earnings at risk, and economic value of equity show behavior consistent with gap. That is, the first bank shows a very flat to neutral earnings at risk, and economic value of equity, with a slight bias toward long term liability sensitivity. The second bank, with wider variability in gap results, also shows substantially higher variability in both earnings at risk and EVE.
For most community banks, without a concentration of complex instruments on the balance sheet, the three basic measurements of interest rate risk should generally show consistent results. Don't shy away from using gap to identify where your interest rate risk lies simply because it's an older, time-tested tool. As long as you understand the limitations of gap, it remains a highly effective measure of interest rate risk.
Photo provided by Clemson.