CUNA Comment Letter

NCUA’s Request for Comments on the Valuation of Credit Union Non-Maturity Shares

April 24, 2002

Ms. Becky Baker
Secretary of the Board
National Credit Union Administration
1775 Duke Street
Alexandria, Virginia 22314

RE: NCUA’s Request for Comments on the Valuation of Credit Union Non-Maturity Shares

Dear Ms. Baker:

The Credit Union National Association is pleased to comment on the agency’s Request for Comments on the Valuation of Non-Maturity Shares. CUNA’s comments, which were developed by CUNA’s Vice President of Economics, Michael Schenk, are attached. CUNA represents more than 90% of our nation’s over 10,300 state and federal credit unions.

Thank you for the opportunity to express our views on this issue.

Sincerely,


Mary Mitchell Dunn
Associate General Counsel

CUNA’S COMMENTS ON THE VALUATION OF NON-MATURITY SHARES

In general, CUNA believes that the specific methods used to justify a credit union’s non-maturity share assumptions should be left to each credit union and should be judged on a case-by-case basis. Imposition of a standard calculation methodology is undesirable since, as explained in the n/e/r/a research report, each has unique weaknesses. While standardization may help NCUA conduct examinations more efficiently, the goal should be accurate risk measurement.

We agree in principle with the idea of establishing "safe harbors" for non-maturity share assumptions (such as a maturity of 1.0 years for money market shares, 2.5 years for regular shares, and 3.0 years for share drafts). However, we suspect the safe harbors outlined in the n/e/r/a paper are too restrictive since these were developed outside the credit union movement. In this regard, we find it disturbingly inconsistent for the n/e/r/a report to deem OAS models "not appropriate" for credit unions because credit union deposits "may behave in materially different fashion" from those of banks and thrifts, yet use bank and thrift research to recommend appropriate safe harbors.

The n/e/r/a suggests three ways to assess the accuracy of the effective maturities recommended in their report. We believe that additional statistical study of credit union deposits conducted using currently available call report data is appealing. However, data limitations would severely limit the use of study results.

CUNA does not agree that the call report should be modified to capture data that would allow the NCUA to calculate median deposit retention rates. Requiring credit unions to track and report the aggregate balances in new accounts would be unnecessarily time-consuming, burdensome, and operationally unworkable for some credit unions. Notwithstanding this objection, any additions to the call report should only apply to credit unions that desire to use assumptions outside the safe harbor.

CUNA supports the idea of obtaining data from a sample of credit unions. As a first step, we believe that NCUA should survey ALM software providers to see if any of these vendors have centrally-collected information. The results (and conclusions) of that analysis should then be communicated to the movement with additional opportunity to comment. We further believe that this study should be conducted before the Board reaches any other conclusions or takes any other action on this topic.

CUNA does not agree that a Treasury rate should be treated as the (exclusive) discount rate for valuing non- maturity deposits. Credit unions should be given the flexibility to use other discount rates as well. The use of treasuries as suggested by the agency seems to imply that deposits be evaluated from the viewpoint of the depositor. Instead, we believe that these deposits should be viewed from the standpoint of an "acquirer" (or "seller"). In this regard, the appropriate discount rate for many credit unions would be a wholesale rate like jumbo certificates. Jumbo rates have merit, but since this market does not have maturities corresponding to the assumed lives of many non-maturity deposits, other alternatives also are desirable. For some credit unions, FHLB advances may be appropriate. The FHLB market is well-understood and provides a "complete", longer-term yield curve with readily observable rates.

CUNA agrees that the characteristics of a non-maturity account (not its labeling) are important determinates of value and are currently evaluating the documentation appropriate to use "safe harbor" assumptions.