Posts

/by

Some Examiner Requests Conflict With Written Guidance – Make Sure to Get Clarity of Reasoning

We are curious as to why some examiners are requesting that credit unions establish risk limits based on 12-month net income simulations.

This request is interesting in light of what was outlined in the Interagency Advisory on Interest Rate Risk Management.  The guidance states:  “When using earnings simulation models, IRR exposures are best projected over at least a two-year period.”  The guidance goes on to say:  “However, to fully assess the impacts of certain products with embedded options, longer time horizons of five to seven years are typically needed.”

If your examiner requests that you establish risk limits based on a 12-month simulation, consider asking:  “How will managing risks with such a short-term view protect our credit union and the insurance fund, especially in light of historically low interest rates?”

Your examiner may counter by saying they use NEV to look long-term.  However, remember that NEV is defined as the fair value of assets minus the fair value of liabilities and tells you nothing about your earnings today or how they can change as rates change.

/by

Does Your NEV Analysis Really Capture Fair Value Of Assets?

, ,

If your credit union runs a Net Economic Value (NEV) analysis as part of the A/LM process, how do you determine the credit union’s loan discount rate assumptions?

Sometimes we hear credit unions say that their current loan offering rates are used as the discount rates in their NEV models.  We do not believe this is the best method.

While current offering rates reflect what a specific credit union is willing to accept in terms of yield on new volumes, they do not necessarily reflect what an investor looking to purchase an existing portfolio may demand.  For example, if the credit quality of a credit union’s auto loan portfolio has materially degraded since the loans were made—a situation many credit unions are experiencing—then an investor is going to require a return that exceeds today’s low rates to compensate for the risk.

Additionally, there may be times when a credit union is intentionally pricing a portfolio above or below market to affect growth in that portfolio.  Many credit union executives would say that their loan rates are better than banks—in other words, they are “below market rates.”

As far as the credit risk component is concerned, lately we have heard comments suggesting that credit risk should not be factored into an NEV analysis.  This is simply not true.

In Letter to Credit Unions 99-CU-12, the NCUA says “NEV equals the fair value of assets minus the fair value of liabilities.”  What is the definition of fair value?  According to 12CFR NCUA, Section 704.2, fair value is defined as “the amount at which an instrument could be exchanged in a current, arms-length transaction between willing parties.”  The definition further states, “Valuation techniques should incorporate assumptions that market participants would use in their estimates of values, future revenues, and future expenses, including assumptions about interest rates, default, prepayment, and volatility.”

If you have gone through the process of having a third party value your portfolio, this can be helpful.  Note, in many cases such as mortgages, the price will only be for a portion of the loans, not for 100% of the portfolio.  Often, a material portion of the portfolio will not receive a price since there is not an efficient market for mortgages made several years ago that now have high LTVs.

If you do not have market values then for each portfolio, as you set assumptions, consider if you would have to take a loss selling all of the parts (the good and the bad).  Not only should your unique credit experience play a role in this answer, but so should the geographic region in which the loans were made.  A credit union selling loans from any of the sand states will typically take larger losses.

Once you establish an assumed price, you can use that price to calculate the assumed discount rate.  Once the base simulation is done, run alternate sets of assumptions to calculate the sensitivity to your results.