but with all values converted to British pounds sterling.     The basic concept in the gap report is the net present value (NPV) of the banking book. The PVBP report measures the difference between the market values of assets and liabilities in the banking book. To calculate NPV we require a discount rate, and it represents a mark-to-market of the book. The rates used are always the zero-coupon rates derived from the benchmark government bond yield curve, although some adjustment should be made to this to allow for individual instruments. Gaps may be calculated as differences between outstanding balances at one given date, or as differences of variations of those balances over a time period. A gap number calculated from variations is known as a mar- gin gap. The cumulative margin gaps over a period of time plus the initial difference in assets and liabilities at the beginning of the period are identi- cal to the gaps between assets and liabilities at the end of the period. The interest-rate gap differs from the liquidity gap in a number of detail ways, which include:   ■ whereas for liquidity gap all assets and liabilities must be accounted for, only those that have a fixed rate are used for the interest-rate gap; ■ the interest-rate gap cannot be calculated unless a period has been defined because of the fixed-rate/variable-rate distinction. The interest- rate gap is dependent on a maturity period and an original date.   The primary purpose in compiling the gap report is to determine the sensitivity of the interest margin to changes in interest rates. As we noted earlier, the measurement of the gap is always "behind the curve" as it is an historical snapshot; the actual gap is a dynamic value as the banking book continually changes.       CRITIQUE OF THE TRADITIONAL APPROACH   Traditionally, the main approach of ALM concentrated on interest sensi- tivity and net present value sensitivity of a banks loan/deposit book. The usual interest sensitivity report is the maturity gap report, which we reviewed briefly earlier. The maturity gap report is not perfect however, and can be said to have the following drawbacks:   ■ the re-pricing intervals chosen for gap analysis are ultimately arbi- trary, and there may be significant mismatches within a re-pricing interval. For instance, a common re-pricing interval chosen is the 1- year gap and the 1-3 year gap; there are (albeit extreme) circum-