CORRESP 1 filename1.htm

 

 

April 26, 2006

 

VIA EDGAR

 

Securities and Exchange Commission

100 F Street, NE

Washington DC, 20002

Attention:                      Ms. Joyce Sweeney
Accounting Branch Chief

 

RE:         Response to SEC Staff Comments on

UnionBanCal Corporation Annual Report on

Form 10-K for the fiscal year ended December 31, 2005

File No. 1-15081

 

Dear Ms. Sweeney,

 

We refer to the comments of the SEC staff in your letter dated April 13, 2006 on the UnionBanCal Corporation filing listed above.

 

Below are the comments of the Staff transmitted in your letter dated April 13, 2006 together with our responses to those comments.

 

Form 10-K for Fiscal Year Ended December 31, 2005

 

Consolidated Financial Statements

 

Note 19 - Derivative Instruments, page F-90

 

Hedging Strategies for Variable Loans and Certificates of Deposit, page F-91

 

1.             We have reviewed your response to comment 3 of our letter dated March 28, 2006. Please provide us with an example describing how you would perform the

 

400 CALIFORNIA STREET, SAN FRANCISCO, CALIFORNIA 94104

 

P. O. BOX 45000, SAN FRANCISCO, CALIFORNIA

 

415 765 0400

 


 

following for a cash flow hedge of the forecasted issuance and rollover of fixed-rate certificates of deposits:

 

•                  how you forecast the issuance and rollover;

 

•                  the terms of the hedge instrument; and

 

•                  how you determined that the hedging relationship met each of the criteria in paragraphs 28-29 of SFAS 133, as applicable.

 

Response to Comment 1.

 

On a quarterly basis, all deposit taking business units produce a 12-month rolling forecast of their expected monthly certificate of deposit (CD) balances. This information is used as the basis for any decisions and planning with respect to entering into or terminating any hedges. On a monthly basis, actual 1-month, 3-month and 6-month CD balances are downloaded from bank source systems and compared to actual hedge balances. In the unlikely event that more hedges were on the books than forecasted balances of  CDs being hedged, these transactions would no longer qualify for hedge accounting. However, this has never occurred since we hedge only a portion of the forecasted CDs and only those CDs with maturities of 6 months or less (short-term CDs). We believe that based on our historical information issuances and rollovers of CDs beyond the forecasted 12-month period are probable of occurring.

 

The hedge instruments are either purchased caps or cap corridors with reset tenors between 1 and 6-months and with a maturity date of less than 3 years. The cap corridors may be entered into as a single instrument or as two instruments (a purchased cap and a written cap) designated in combination and entered into contemporaneously. In addition, the cap corridors must have the following characteristics:  (1) the strike rate of the purchased cap is lower than that of the written cap; (2) both the purchased and written cap are based on the same notional amount, start date, reset/payment dates, index, day count convention, maturity date and other relevant critical terms; and (3) the net premium for entering into the transaction is either zero, or is a net payment paid by us.

 

At inception, hedges are entered into with reset frequencies intended to match the tenors of our CD portfolios (e.g., 1-month LIBOR swaps are matched with 1-month maturity CDs, etc.). In forecasting our hedged CD balances, an assumption is made that the forecasted rollovers will generally have the same original maturity as the current portfolio. We also ensure that the notional balances of the hedge instruments will not exceed the current or expected amount of the hedged CD balances. Our designation of these hedging instruments follows a broad designation approach, similar to the “first payments” approach discussed in DIG Issue G13. That is, the hedging relationship is not narrowly limited to interest payments on specifically identified individual CDs, but instead relates to a specified amount of CDs and forecasted CDs that are deemed probable of being outstanding. As such, the hedging relationship is deemed to be intact as long as the existing and forecasted levels of short-term CDs in amounts equal to or greater than the notional amounts of the hedging instruments are deemed probable over the term of the hedge periods. Changes in the make-up of individual CDs within the portfolio do not change the designated hedging relationship. Our hedge strategy documentation acknowledges that changes in the CD portfolio can (and will) occur, such that not all of our hedging instruments will continue to have reset frequencies that match the maturities of the hedged CDs. When this

 


 

occurs, this is not deemed to be a discontinuance of the hedging relationship, but instead an event that impacts the assessment of effectiveness and measurement of ineffectiveness, as discussed below.

 

Hedging the rollover of short-term fixed rate CDs is an approved strategy, formally documented within our SFAS 133 policy and specifically allowed by DIG Issue G19 as a cash flow hedge of benchmark interest rates. At inception and on an ongoing quarterly basis, we determine the effectiveness of the hedging relationships using regression analysis to demonstrate the degree of effectiveness that exists. For the majority of our hedging instruments, the derivative interest reset frequency matches the tenor of the forecasted CD balances. In these relationships, the only source of ineffectiveness is the degree to which the timing of resets does not match the timing of CD maturity and rollover. Therefore the primary assessment test consists of a regression analysis that examines the differences in reset dates that may exist between the hedging instruments and the hedged items. The key regression outputs that are analyzed in assessing effectiveness include R squared, slope, t and F statistics. We update this regression analysis quarterly for both retrospective and prospective assessment purposes in order to justify the continuing use of hedge accounting on these kinds of hedges and would discontinue hedge accounting in the event that the test failed. The assessment performed at the end of a reporting period also serves as our prospective effectiveness assessment for any new hedges entered into during subsequent reporting periods. Ineffectiveness is measured using the hypothetical derivative method, which incorporates the effect of differences in reset dates between our hedging instruments and the hedged items. For hedging relationships involving options, we apply the guidance in DIG Issue G20. Therefore, we base our assessment of effectiveness on the terminal value of such instruments, and accordingly include total changes in fair value (time and intrinsic) in the assessment of hedge effectiveness.

 

There can be instances in which the tenors of the hedging instruments no longer match the tenor of the hedged item, due to changes in the make-up of the CD portfolio. For instances, such changes may require that 1-month LIBOR caps be allocated to CDs with 3-month maturities. Thus, ineffectiveness can potentially exist due to this mismatch in tenors, causing differences in the frequency of interest rate resets of the hedging instrument versus that of the hedged item. Such mismatches do not always exist because we generally attempt to first match hedging instruments and hedged items of like tenor, as noted above. However, they can and do occur. In order to support the assumption that despite the differences in tenor/reset frequency, the hedging instrument can be expected to be an effective hedge of the underlying cash flows, we perform an additional regression test that evaluates the relationship between LIBOR rates of differing tenors (i.e., 1, 3 and 6-months). Since that analysis is performed quarterly and so long as that analysis continues to demonstrate high effectiveness for the differing rate tenors, any hedging relationships for which mismatched tenors occur will be deemed to be effective retrospectively and prospectively. Any mismatch in tenors is also taken into account in the construction of our hypothetical derivatives used in the measurement of ineffectiveness.

 

We believe that these hedging relationships meet all of the applicable criteria of paragraphs 28 and 29 of SFAS 133, as follows:

 

28. a. We prepare formal documentation that is in place at the inception of each hedge, which consists of the following:  a hedge strategy documentation manual, which describes each of our hedging relationships, including our risk management objective and strategy, the type of instrument used, the hedged transaction, the nature of the risk being hedged, how hedge effectiveness will be assessed and how ineffectiveness will be measured. The information

 


 

contained in our hedge strategy documentation manual is referenced on each trade ticket at the time a new hedge instrument is entered into.

 

28. b. Described above.

 

28. c. and d. are not applicable.

 

29. a. Described in our response letter of April 5, 2006.

 

29. b. Described above.

 

29. c. All of our forecasted transactions are with customers who are external parties.

 

29. d. The forecasted transactions do not relate to items that will be subsequently remeasured at fair value.

 

29. e. through g. are not applicable.

 

29. h. Described above.

 

Hedging Strategy for “MarketPath” Certificates of Deposit, page F-93

 

2.             We have reviewed your response to comment 3 of our letter dated March 28, 2006. We note that you account for derivatives embedded in your MarketPath CDs and total return swap at fair value with changes recognized directly in earnings. Please tell us the following regarding your fair value hedges of MarketPath CDs:

 

•                  the nature of the bifurcated embedded derivative(s);

 

•                  the terms of the total return swap;

 

•                  the specific documented risk being hedged;

 

•                  how you determined that the hedging relationship met each of the criteria in paragraphs 20-21 of SFAS 133 to qualify for hedge accounting; and

 

•                  how you measure ineffectiveness and assess effectiveness of these hedges both at inception and on an ongoing basis.

 

Response to Comment 2. As we mentioned in our letter of April 5, 2006 and in our telephone conversation on April 14, 2006, MarketPath CDs pay interest at maturity based upon the change in the value of the S&P 500 index. On the date of issuance of a MarketPath CD, we would enter into an S&P 500 indexed option contract. Each option’s premium consisted of an upfront payment based upon a fixed, market-rate of interest over the life of the CD, which locked in a fixed market rate on the MarketPath CD. The upfront premium is then amortized over the life of the CD as interest expense. The interest payable to the CD customer is equal to the change in the fair value of the intrinsic value of the S&P 500 index option since the customer receives exactly the same amount as that which is paid on the option contract. At maturity there could be a difference of less than $10 on each CD to compensate for grace period days. That amount is reflected in interest expense at the maturity date.

 


 

As we indicated in our previous letter, this product was originally offered by Jackson Federal Bank. In order to minimize the number of option contracts (in the range of 300 to 400 contracts) and the fact that this product was no longer going to be offered after the acquisition was completed, the option contracts were “closed out” and replaced with a total return swap contract. The swap contract’s cash flows mirror all of the cash flows from the individual option contracts. Therefore, there are no differences between the cash flows from the options themselves and the swap contract.

 

As we indicated in our telephone conversation, this hedging strategy is more accurately described as an economic hedge rather than a “fair value” hedge under SFAS 133. We will change our disclosure in our March 31, 2006 Form 10-Q to indicate that this strategy is an economic hedging strategy rather than describing the strategy as a “fair value” hedge. This clarification will have no effect on our financial statements for any period.

 

*     *    *

 

In connection with responding to the comments of the Staff, we acknowledge that:

 

•                            we are responsible for the adequacy and accuracy of the disclosure in the filing;

 

•                            staff comments or changes to disclosure in response to staff comments do not foreclose the Commission from taking any action with respect to the filing; and

 

•                            we may not assert staff comments as a defense in any proceeding initiated by the Commission or any person under the federal securities laws of the United States.

 

Please call me at 415-765-2595 or Jacqueline Bean at 415-765-2742 if you have any questions or comments regarding the foregoing.

 

 

Sincerely yours,

 

 

 

 

 

 

David I. Matson

 

 

Vice Chairman and

 

 

Chief Financial Officer

 

 

cc:

 

Sharon M. Blume

 

 

U.S. Securities and Exchange Commission

 

 

Takashi Morimura

 

 

Chief Executive Officer, UnionBanCal Corporation