Q&A with Emily Hollis

Emily Hollis
August 24, 2009 | COMMENTS

Question: The Net Interest Income Simulation (NII) shows a gain in net income should rates rise; however the Net Economic Value (NEV) analysis shows a negative value in the “moderate” interest rate risk range. Are the reports erroneous? Also, if you do the NII analysis, is the NEV analysis not important?

Differing results with the NII simulation versus the NEV analysis are becoming more and more common in this low interest rate environment.

The NII simulation replicates earnings over a period of time. It does not capture cash flows beyond the initial horizon period of the simulation. Therefore, a key point to remember is the market value volatility of longer-term cash flows is not taken into consideration, and subsequently, its impact is not shown. Maximum rates that are embedded in loans and investments such as adjustable-rate mortgages (ARM) might not affect income projections if the rates are not moved high enough in the simulation.

Let's look at an example: 5/1 hybrid ARMs that were issued two years ago at a rate of 5.50 percent might be capped at 7.50 percent should rates move up substantially. However, below this level, ARM rates would move in tandem with the market and show an increase in earnings as the mortgages were reset.

On the other hand, the NEV analysis will show a fair value loss on the market values and will capture the entire 30-year's worth of cash flows, therefore modeling interest rate paths that could potentially impact interest rate resets.

To help clarify, an example is outlined below. This financial institution has 85 percent of its assets in real estate loans, which at first glance is a red flag for high risk. These real estate loans are fairly interest rate sensitive because the rates of the ARMs and LOCs will reset as interest rates change. When rates move up, the variable rate sectors of the balance sheet will reset up to their defined cap the first year, which lessens the risk associated with the fixed-rate, 30-year loans that will not reset. 

Now let's examine the liabilities: 33 percent of the liabilities and equities are in money market accounts, which reset quickly with rising interest rates. However, the risk of these accounts is mitigated by the checking and regular share accounts, which barely move in a rising rate environment. As you can see in the table, the asset liability spread increases by 15 basis points and net interest income by 29 basis points in the up 300 basis-point scenario. When you analyze the entire balance sheet, from an income perspective the credit union makes money when rates rise over a one-year period.

Alternatively, the NEV analysis shows a 1.04 percent increase in net economic value to capital in the base-case scenario and a decrease from that in fair value of 4.45 percentage points in the up 300 basis-point scenario. Reason being that the market devaluation of the 35 percent of assets, which are in 30-year real estate loans, is not supported by liability market increases on the shorter term liabilities. The NEV takes into consideration all cash flows instead of those for just one year. The institution is approaching a “high” amount of interest rate risk with a 45 percent negative change of NEV from the base-case scenario.

In conclusion, one test shows minimal interest rate risk and the other shows high interest rate risk. Intuitively this seems flawed; however, it is not. The example outlined here highlights the limitations of the NII simulation. While we perform both, we put more credence in the NEV analysis; without it, long-term cash flows and embedded options (i.e., caps) are not adequately tested and measured.

Emily Hollis is president of ALM First Financial Advisors in Dallas, Texas. Contact Hollis at 800-752-4628 or ehollis@almfirst.com.