A bank's focal point for market risk: The transfer pricing mismatch unit

Journal of Bank Cost & Management Accounting, The, 2000 by Chittenden, John

2. To what extent can the Mismatch Unit's net earnings be controlled by the ALM department?

The honest answer to this question is the unsatisfying "it depends Of course, generally speaking, the more precise the transfer pricing methods, the more meaningful the Mismatch Unit results. It is often helpful to analyze the quality of Mismatch Unit results by making broad distinctions as follows.

There are areas where Treasury has substantive control over the Mismatch Unit's earnings:

* Treasury (ALM) performs the net interest forecasts. When based on good data, these forecasts will reveal major anticipated balance sheet and market rate movements which affect the Bank's rate risk position. This is a solid basis for net interest margin hedging within the timing and materiality constraints imposed by the availability of this data.

* All market risk hedges are proposed and executed by Treasury (ALM). Their results are reported in the Mismatch Unit earnings.

There are areas where Treasury has approximate control over the Mismatch Unit's earnings. A few among many examples:

* Non-contractual maturity deposits have both balance and rate characteristics that are not directly related to money market rates, are not precisely predictable, and are therefore not fully hedgeable.

* Prime basis risk can only be partially hedged. The Bank's credit standing is probably not sufficient for it to hedge all of the Bank's Prime loans on a cost efficient basis in the Prime/Libor swap market.

*Prepayment behavior of fixed rate mortgage loans is, at best, "covered" by an option charge, but these charges cannot be expected to correspond exactly with the real gains/losses on the prepayment options in the loan products.

There are areas where Treasury has little control over the Mismatch Unit's earnings:

* Most earnings-based profitability analyses focus on month-to-month changes. Monthly trends in the Mismatch Unit P&L are not meaningful indicators of the effectiveness of interest rate risk management. They are subject to days-in-month accrual choppiness and the happenstance timing of accounting entries. Longer term P&L measurements must be used. Perhaps even quarter to quarter results are not long enough to be meaningful.

* Exact measurements of risk gradations in the Mismatch Unit's position are probably not possible. Therefore, when the P&L is used together with equity allocation to produce an ROE, this risk/return ratio must be interpreted with a wide latitude. For example, a movement from a 13% ROE to a 15% ROE from quarter to quarter may not have significance. This is discussed in greater detail below.

* Specific exceptions must be given for ALCO-directed trades which have an immediate effect on the interest rate risk position, but whose effect cannot be immediately hedged.

3. What are the legitimate sources of profit for the Mismatch Unit? Should all profits from interest rate risk be in a Trading book rather than the Mismatch Unit?

Although there are only a few U.S. Banks which have significant trading-- for-profit strategies, those that do clearly segregate trading from Treasury operations. Profits belonging to the Trading book(s) are clearly defined and broadly understood. They are basically short-term market rate bets in a defined, balanced book of business. This operation perhaps entails trading long term assets/liabilities but with gains/losses based on short-term realization and mark-to-market accounting. Currency trading might, of course, be another trading book. Such trading books are usually found only in large, money center banks. From this perspective,"trading for profit"does not include opportunistic Fed Funds or Repo trading as part of the Bank funding operations.