Decay rates on non-maturity deposits (including MMA, Savings, NOW accounts, etc.) relate to cash-flow and liquidity projections made by the Bank, and, when combined with "Beta" factors, the re-pricing impacts of such deposits in IRR modeling. Increasingly, regulators are requesting that banks develop their own customized decay rates for non-maturity deposits rather than use the industry averages or “rules of thumb”. Regulators believe that the use of customized rates may better reflect each bank’s own customer base and the local competition for deposits. Customized rates could then improve cash flow projections and IRR modeling performed by banks. Under current regulatory guidelines, banks are responsible for understanding and testing the assumptions they use in their IRR models. Thus, decay rate studies may be deemed to be necessary by regulators in developing model parameters or in validating them.
The OTS used to compile statistics or averages which had been widely used by community banks. With the OTS merger into OCC, such data may not remain current. Also, there is another set of average decay rates in common used, which are generally known as the "FDICIA 305" decay rates which have been in use for several years. Many banks have relied one or the other of these sources for decay rate assumptions rather than developing their own. This is usually because estimation of decay rates (as well as customized “beta” factors and prepayment speeds) can often be data and labor intensive to prepare. However, considering the regulatory interest in customized rates, it now seems that banks should consider adding a process to critically review deposit decay assumptions and, perhaps, develop and maintain customized decay rates. There is literature available on the Internet which describes acceptable methodologies for doing such decay studies. (e.g., see page 16 by clicking here).
Another related activity which banks may wish to consider in addition to customized decay studies is the use of "sensitivity" analysis for the assumptions used in IRR models. Such sensitivity testing is also mentioned in published regulatory guidance on IRR modeling. The objective of performing such sensitivity testing of the model (i.e., running the model using alternate decay rates such as "up by 50%" and "down by 50%", but holding all else constant) is to determine how sensitive the model results are to each assumptions or parameter chosen by the bank. Sensitivity tests can also be performed on other assumptions/model parameters (e.g., beta factors, prepay speeds, etc.), to let the Bank know which parameters or variables have the largest impacts on results. Those with relatively large impacts should be more carefully analyzed, while those with very low or even minimal impact may not need as much analysis and estimation effort.
An additional area concern in IRR modeling these days should be the common assumption that CDs will roll-off based on scheduled maturities. This is normally a fairly good assumption, due to the fact that depositors do not want to lose the interest penalty by cashing in a CD before its maturity. However, given the very low rates currently in place, the penalty is quite small. A major upward movement in rates might make CD-holders very willing to pay the penalty in order to get higher rates immediately. Banks may wish to consider forecasting some roll-off of low rate CDs with longer maturity dates in conjunction with higher (e.g., +300 and +400) rate shock scenarios. Given the Fed’s current stance and public statements, this may not be expected over the next year or more, but rates will eventually change, and anticipation of the impacts of a sudden upward shock could add to the predictive ability of the bank’s IRR model.
Financial Senior Audit Manager
Jim Cole, Financial Risk Manager, has a diverse and comprehensive background covering many different aspects of the financial area, such as budgeting and strategic planning, asset liability management, profitability and liquidity analysis, financial systems and processes, public offerings, risk management, accounting and financial reporting, investment performance monitoring and financial hedging.