Strategic Modeling for High Performance: Asset / Liability Management
The successful performance of a community bank ultimately depends on proper understanding and management of interest rate risk. Sometimes, however, bank management is too complacent about modeling changes in interest rates and the resulting impact on earnings. Many institutions have more complexity embedded in their balance sheets than they realize, andif they rely on simple call-report based ALM reporting systems, they may not be getting the best information on which to base decisions. In the battle for net interest margin, it pays to have a sound decision-making process including a robust asset/liability model.
We know that deposit rates, loan rates, and bond yields rise and fall with market conditions and general interest rate trends. The story doesn’t end there, however. We have to remember is that it is the relationship between short and long rates, or the shape of the yield curve, that ultimately dictatesearnings performance. Borrowing costs are driven by interest rates in the short end of the maturity spectrum, while the bank’s asset yields (rates on loans and bonds) are priced off of the longer end. When the difference between short and long rates is minimal (a flat yield curve), bank margins are compressed. In times when the yield curve is steep, on the other hand, margin widens and earnings improve given the appropriate time lag. The trick is to identify and project what the yield curve is doing to our cost of funds and asset yields today and in the future. Remember too, that the complication of yield curve twists is added to the uncertainty of cash flows from all sorts of optionality. The number of variables is often enough to drive management crazy. It can be done, though, and in order to get it all right, we just need the proper tools including an ALM system that incorporates the essential assumptions that must be modeled.
Any basic model must simulate the impact of re-pricing volumes of assets and liabilities into new reinvestment rates or costs. To do it right, we also must incorporate time lags, and re-pricing sensitivities of the various types of balance sheet items. Further, we must have the ability to model changes in the shape of the yield curve along with growth and sector diversification strategies. The model must have back testing capabilities for users, and the vendor should provide proof of validation by an outside auditor.
The best way to view a bank is as a single system made up of various constituent parts. These component parts include funding sources, loan structures, and investments among others. A good asset / liability model will piece together the mechanics of this system so that the bottom-line effect of interactions among the different parts can be clearly seen and reported.Importantly, the reports themselves must be readable and useable by management, the board of directors, and regulatory officials if the output is to be useful.This requires reports that are specifically designed for various users. The board of directors must have a set of reports that tells them essential information about the bank’s balance sheet risk. Management will need more in-depth reporting on which meaningful decisions can be made in terms of loan or liability pricing, investment management and strategy going forward. The bank should also have the sort of reports requested by examiners including those regarding assumptions and back tests. All of these reports are essential in order to clarify and communicate the bank’s interest rate risk position to management, directors, and regulatory staff. More importantly, they are critical to running a profitable and productive institution.
Jeffrey F. Caughron
Associate Partner
The Baker Group LP