Asset/Liability Systems Strategy

Townsend Walker, a senior vice president at Bank of America, explains how to choose the right asset/liability management system.

For banks, interest rate exposure is central to the business; managing it successfully is how banks make money. The first step in the management process is measuring the exposure. That is why it is useful to review the different approaches to measurement and establish some criteria for which method to use.

For banks, it is not simply a matter of fixed versus floating rate exposures. Most banks have a dozen or so different rate indices attached to their assets and liabilities, such as prime, Libor, CD or Treasury notes. This is a reflection of the different sources from which they draw their funds and the customization of lending rates to attract borrowers.

It is the nature of banking that depositors and borrowers have different appetites and so create mismatches and risk for a bank. There are mismatches in quantity, maturity and in rate indices. Some indices are averages of past rates (lagging indices) and some assets and liabilities contain embedded options. These dimensions should be included in the methods used to measure exposure.

There are several ways to measure the interest rate risk of banks: static gap, dynamic gap, duration gap, funds allocation, simulation, market value and the Office of the Controller of the Currency's method for risk-based capital.

Static gap

Gap analysis reviews all of the assets, liabilities and equity accounts and identifies the items with interest rates. These items are put in a "maturity basket" according to when the rate on the item will change. Assets that reprice in a given bucket are compared to liabilities that reprice in the same bucket. The difference is called a "gap."

Dynamic gap

Dynamic gap analysis takes into account a bank's plans, the evolution of its business and capital structure objectives. A static gap analysis is prepared for various times in the future and then they are linked to observe the exposure pattern.

Duration gap

One of the motivations for the duration gap measure is that the maturity buckets rarely show a consistent pattern of asset or liability sensitivity. The three-to-six-month bucket shows the bank will gain if rates go up, the six-to-12-month bucket shows it will lose if rates go up, the one-to-two-year bucket shows it will gain if rates go up, and so on. It is not easy to hedge that kind of exposure. On the other hand, it may not be necessary to hedge every exposure; it may be enough to ensure that, on average, the magnitude and timing of the liability cash flows match the magnitude and timing of the asset cash flows. A way to arrive at this average is to calculate the duration of the bank's assets and the duration of its liabilities. The net of the two numbers is your bank's exposure to interest rate movements.

Funds allocation

The funds allocation approach ranks assets, liabilities and equity by the frequency with which their rate changes. At the least frequent end of the liability/equity spectrum is common equity; at the most frequent end is Fed funds purchased. Assets on the least frequent end of the spectrum are the bank building and furniture; on the most frequent end, Fed funds sold. The allocation procedure is to fund the fixed assets first, then 30-year loans, then five-year loans, then one-year loans, and so on, starting with equity, then long-term liabilities, medium-term liabilities, and so on. For example, the funding for a prime rate commercial loan portfolio may be six-month CDs; when all of the six-month CDs have been used, three-month CDs are used, and then Eurodollar deposits. The result is a clear picture, on the basis of the frequency of rate change, of the sectors of the balance sheet where the bank does, or does not, have interest rate exposure.

Simulation models

Simulation models make assumptions about the future volume and mix of loans and the family of future interest rates (Fed funds, three-month Libor, prime, five-year Treasuries, etc.). The heart of simulation models are equations, generally derived from historical information, relating external variables to the behavior of the balance sheet and income statement. Simulation models are not easy to use correctly; their virtue is that they allow your bank to examine systematically what may happen in the future.

Market value

The market value approach is based on the premise that the bank with the highest market value will have the highest future earnings potential. Market value is the risk-adjusted value of future income streams. First, the future cash flows, without option characteristics, are discounted at current market rates. Those asset and liability items that have embedded options are adjusted to account for the option behavior in different interest rate scenarios. Then to test the interest rate sensitivity of your bank, the individual discount rates are moved up and down. The result is the change in the current market value of your bank.

OCC

The OCC is approaching interest rate risk on the basis of what a change in interest rates does to the economic value of the bank. Economic value is defined as the net present value of the assets less the net present value of the liabilities, and the net present value of the off-balance-sheet items. In concept, it is the same as the market value approach noted above. The difference is that because the OCC approach is designed to collect data in a uniform fashion, it has specified many of the behavioral assumptions about retail deposits and mortgages. Each balance sheet category is assigned risk factors on the basis of interest rates moving up and down 2 percent. Mortgages and mortgage products receive special treatment designed to capture their option features.

Criteria for choosing an approach

Asset/liability models serve a number of purposes besides identifying interest rate exposure. From an interest rate exposure point of view, a model should:

1. clearly define the current macro sensitivity of your bank to changes in rates

2. identify basis risk

3. identify option risk

4. be able to incorporate future plans and events

5. reveal all of the choices for remedying an undesirable exposure

6. be easy to use and understand

The dynamic gap approach satisfies criteria 1, 4, 5 and 6.

The duration gap approach satisfies criteria 1, 5 and 6.

The funds allocation approach satisfies criteria 1, 2, 5 and 6.

The simulation model satisfies criteria 2, 3, 4 and 5.

The market value approach satisfies criteria 1, 2, 3 and 5.

As you can see, the answer rests in using a combination of approaches. The ones chosen will depend on your bank's characteristics.

Townsend Walker is the author of Managing Risk with Derivatives: A Guide to Bankers and Their Customers, just published by the American Bankers Association.