Search
RSS
Our Attorneys
Twitter
Twitter
« Maine financial institutions...In The News | Main | FinCEN Assesses the Maximum Civil Money Penalty for Illegal Disclosure of a SAR »
Monday
Jan162012

12 Steps to Managing Interest Rate Risk: Feds issue new FAQs

On January 12th, the federal banking agencies issued Supervisory Guidance and a Financial Institution Letter regarding interest rate risk (IRR) management.  The Guidance contained answers to 12 “Frequenty Asked Questions” intended to clarify the 2010 interagency Advisory on Interest Rate Risk Management.  Here is a summary of the new guidance. 

  1. Choosing a model.  Institutions should choose IRR vendor models after considering cost, risk profile, transparency, level of automation, amount of vendor support, and the model’s ability to reasonably capture current and future risks.  To manage the risk of vendor termination, institutions should also maintain in-house knowledge and contingency plans.
  2. Modeling new strategies.  Prior to adopting a new product or strategy, institutions should consider the ability to measure IRR from the new strategy.
  3. Types of IRR measurements.  Institutions should measure IRR on both earnings and the economic value of capital.  Regulators note that there are now many financial products with embedded options which have significant prepayment or extension risk.  Institutions are expected to use longer simulation time horizons (i.e. longer than one year), and more complex institutions should consider 5-7 year simulations.  Complex or structured securities should be modeled individually.  Homogenous whole-loan portfolios should be aggregated by product type, coupon band, maturity, and prepayment volatility.
  4. Institutions with non-complex balance sheets.  Regulators understand that a less-sophisticated IRR measurement program may be appropriate.
  5. Modeling rate shocks.  The 2010 Advisory recommends modeling interest rate changes of 300 or 400 basis points.  Regulators further recommend that modeling should consider the current level of rates relative to a “normal rate cycle.”  Thus, in low-rate environments, institutions should focus more on positive-rate shocks.  When rates are extremely low, institutions should focus more on larger rate shock scenarios.
  6. Consideration of four components of IRR.  At least annually, institutions should consider all elements of IRR:  repricing mismatch, basis risk, yield curve risk, and options risk.  Analyses should also be run whenever an institution’s risk profile changes (e.g. merger, new products).  If an institution’s risk profile shows particular sensitivity to a specific scenario, it should be monitored monthly or quarterly.
  7. Board-approved thresholds.  Management should establish limits, triggers, or thresholds for stress scenarios and compare to board-established risk tolerance.  Results should be periodically reported to appropriate management committee and board.  Institutions are also encouraged to conduct  and report the results of “nonstandard or less frequently run stress tests.”
  8. Modeling “no growth” scenarios.  Because growth can mask the effects of IRR, management should include no growth scenarios when modeling risk.
  9. Verifying accuracy of model.  Institutions should periodically test whether their IRR measurement model is working through back-testing and independent reviews meeting three criteria:  (1) evaluation of conceptual soundness; (2) ongoing modeling to ensure proper functioning; and (3) outcomes analysis to evaluate model performance.
  10. Effective back-testing practices.  Effective back-testing should include isolation of key drivers of IRR, including actual interest rates, prepayment speeds, other runoff, and new volumes.  Periodically comparing offering rates with modeled behavior helps ensure that the model reflects an institution’s ongoing business practices.
  11. Use of industry estimates.  Internal data sets are preferred because industry estimates and vendor defaults may fail to measure important variables like geography or customer types.  Defaults should only be a starting point for difficult to measure variables.
  12. Deposit Decay Rate Assumptions.  Management should consider both deposit and NMD decay-rate assumptions, particularly when customer behaviors change during stress periods or due to external factors.  As an example:  customers will gravitate to insured deposits during times of economic stress, which influences NMD decay rates.

PrintView Printer Friendly Version

EmailEmail Article to Friend