Some say that extremely self-aware people have a higher incidence of deja vous. Technically, it is thought that that the experience of having done something or been somewhere before is a result of information hitting the brain at different times (we are talking millionths of a second). We are not sure what causes the phenomenon, but today's discussion may give you the feeling.
A few weeks ago we introduced the 3 pillars in managing banks' spread gap: 1) Funds Transfer Pricing, 2) Expense Allocation, and 3) Economic Capital. We left that story at the introductory phase so today we cover funds transfer pricing in more detail.
Funds transfer pricing (FTP) refers to the attribution of "spread gap" between liabilities and assets of a bank. At community banks 30% to 60% of the "spread gap" is generated on the liability side of the balance sheet.
It is important for a bank to understand the contribution of those engaged in lending and also those engaged in deposit gathering. Within a bank, treasury charges FTP rates for each lending transaction and credits an FTP rate for each deposit. While FTP is strictly an internal performance measure, the process is considered a "best practice," as it helps allocate resources, risk and return to sections of the bank (to aid in management).
The simplest method is to calculate FTP as a single pool. Here, Treasury determines one rate for all assets and liabilities. An example of this method is calculating a bank's current cost of funding and comparing gross yield on a loan to that current cost of funds. One problem with this approach is that a point-in-time cost of funds is applied to loans that have different maturities than the deposits (mismatch in term). Another problem with this method is that deposits or assets are not matched for options such as prepayments, floors or caps (mismatch in option). The task of modeling indeterminate deposits is another tricky issue that leaves single pool methodology lacking.
Another method that some banks use is called specific matching FTP. Under this structure, a specific liability is matched to every identifiable asset of an equal amount (taking into account maturity, amortization and options). An example of this method occurs when Treasury allocates a 5Y fixed-rate loan against 5Y brokered deposits and a 10Y fixed-rate loan against equity. There are numerous problems with this method. First, it is impossible to find perfectly matched assets and tie them to perfectly matched liabilities. Second, while some traditional accounting practitioners view equity as "free," economically this is not the case. In an economic capital framework, equity has a charge (related to the risk of the bank) and that charge is perfectly (one-for-one) and positively attributable to interest rates. This is to say that the cost of equity goes up, as interest rates rise, all else being equal.
A third and more sophisticated method to calculate FTP is called matched maturity. This approach uses the term, amortization of principal and explicit and implicit optionality of loans and deposits to convert all assets and liabilities to a spread over a yield curve. A yield curve (in most cases it is the LIBOR swap curve) is used to create a term structure for FTP. All assets and deposits are then priced off this yield curve, after adjusting for term, amortization and optionality. All rates are converted to monthly coupons, since most accounting systems at banks run typically on a monthly cycle.
While the matched maturity method may seem daunting to some, the inherent concepts are already built into many banks' internal models. At BIG, all of our products (hedging, loan pricing, loan valuation, relationship profitability and liability management) make use of the matched maturity method.
If you think that you have read this before, it's only deja vous.