Many bank chief executives are frustrated by the inability of their institutions to develop a consistent approach for interest rate risk taking. Notes one CEO: "Our strategies are driven by our underlying philosophy, which seems to change with our outlook on interest rates."
First Manhattan Consulting Group believes that many financial institutions should improve the consistency of their interest rate risk philosophy by making it independent of transient rate fluctuations. Only by so doing can institutions raise the return on the economic capital earmarked for rate positioning.
We estimate that if the rate risks undertaken by the average institution were appropriately capitalized, they would absorb some 15% of total bank capital, more than any other individual bank business. Systematizing the approach to rate positioning will raise risk-adjusted returns, primarily by lowering the amount of needed capital. Hence, the return on equity on interest rate risk taking, which today averages between 8% and 10%, could conceivably reach into the teens.
Part of the reason for today's low risk-adjusted returns is that many banks either slight certain forms of rate risk or take them counterproductively. When bankers discuss rate risk, they often refer only to "core" risk - that caused by the impact of general rate changes on a mismatched asset-liability position. But many are learning that there are at least three other types of exposure - basis, option, and yield curve twist - and at a goodly number of institutions the absolute amount of dollars at risk to each of these three types can exceed that at risk to general rate movements. Small wonder that the Office of the Comptroller of the Currency is expecting bank management to study these risks more closely.
In subsequent columns, we will address the questions of how best to manage basis, embedded options, and yield-curve-twist risks. For the rest of this piece, we will propose a philosophy and a strategy for coping with core interest rate risk.
First Manhattan Consulting Group feels that a strategy of day-to-day or even quarter-to-quarter position tinkering tends to increase earnings volatility with little improvement in average earnings. For such an "activist" risk-taking philosophy to succeed, all of the following conditions must prevail:
1. Rate levels must be forecastable.
2. The bank must be able to forecast rates better than can the general marketplace.
3. The investor must regard earnings achieved through forecasting as sustainable profits deserving of a high multiple.
Most banks in open-market economies cannot regularly improve on the interest rate judgments of the general marketplace, as expressed in the shape of the yield curve. They simply do not have access to superior information on financial flows, which is prerequisite to better-than- consensus forecasting.
Nevertheless, it does not follow that because a bank cannot outguess the marketplace, it should take zero interest rate risk. A zero risk position is, in any event, impossible to achieve.
To be sure, it is often argued that a riskless position is achievable by choosing assets and liabilities with identical cash-flow characteristics. Then rate changes would have equal impact on each side of the balance sheet.
But in a solvent bank, assets exceed liabilities. So even if one matches each liability cash flow with that of an asset, there remains the question of what is to be done with the equity. It, too, must be invested. But in what?
If the equity is invested in, say, very short-term instruments, there is, by definition, little or no risk to its principal value. But since the rate on such instruments fluctuates widely, there is considerable risk to earnings and the growth rate of equity. If rates rise, income, and thus the additions to equity values, will increase sharply. If rates decline, income, and thus the growth in equity values, will be much more modest.
Therefore by investing its equity in an instrument that reprices frequently or daily (often referred to as a zero duration of equity), the bank achieves market-value stability at the price of income instability.
Conversely, a bank that cash-flow-matches each liability with an asset and then invests its equity in 20-year fixed-rate bonds (a long-duration equity) stabilizes its income for the life of that bond. However, the market value of the equity becomes highly unstable, rising when rates fall and vice versa.
In First Manhattan Consulting Group's opinion, neither of these polar examples - which are representatives of the extremes actually found in banks - will command the enthusiasm of shareholders. Investors should not be especially interested in owning a bank with a stable initial equity value but highly uncertain increments to that value. Nor would they be attracted to an institution whose earnings growth was protected from rate fluctuations but whose equity value could drop dramatically when rates rise.
If a bank's equity should not be invested in overnight instruments like fed funds or in long-term bonds, where should it be put? Given that banks are not superior rate forecasters, the sensible answer is the proverbial middle way: Invest the equity in such a way as to balance the tradeoff between market-value and income instability.
Placing the equity in, say, five-year, fixed-rate assets (with earnings on the equity systematically reinvested in new five-year assets) simultaneously reduces the market-value risk inherent in the 20-year fixed investment and the income risk inherent in the short-term investment.
Indeed, doing so reduces the market-value risk by two to three times that of a 20-year investment, while the earnings variability is half that of adopting a zero-duration stance.
Another reason for choosing the middle-of-the-road risk-taking strategy is the desire to stabilize the bank's leverage, expressed in market-value terms - i.e., the ratio of the market value of assets to that of equity. Regulators, rating agencies, and analysts are increasingly viewing a stable leverage ratio - in market as well as book terms - as an important gauge of financial strength.
Hence it is advisable for banks to invest their equity in assets with weighted maturities comparable to those of the total asset base. This ensures that a given change in rates will have virtually the same proportional impact on both the numerator and the denominator of the leverage ratio.
Based on the foregoing, banks should recognize the difficulties of arguing that active risk positioning produces higher returns rather than a higher variability in returns. Having done so, they should craft specific and systematically applied strategies to:
1. Minimize the mismatching of those assets funded by liabilities (i.e., cash-flow-match this part of the balance sheet).
2. Invest the equity in a ladder of assets with an aggregate maturity somewhere between two and five years, picking a given maturity point and then sticking with it.
3. Use plain-vanilla interest rate swaps and other hedges where the same effect can be created at a lower transaction cost than in the cash market.
4. Measure actual results to ensure that they are in line with expectations and that underlying maturity assumptions for assets and liabilities with indeterminate maturities are robust.
Mr. Toevs and Mr. Zizka are managing vice presidents of First Manhattan Consulting Group, New York