April 20, 2006
Ms. Joyce A. Sweeney
Accounting Branch Chief
Mail Stop 4561
Division of Corporation Finance
United States Securities and Exchange Commission
Washington, DC 20549
| Re: | First Data Corporation |
Form 10-K for Fiscal Year Ended December 31, 2005
Form 10-K for Fiscal Year Ended December 31, 2004
File No. 001-11073
Dear Ms. Sweeney:
Included below are the Company’s responses to the comments in your letter dated April 3, 2006. For your convenience, the comments have been incorporated verbatim below and the Company’s responses thereto have been set forth immediately after each comment.
Form 10-K for the Fiscal Year Ended December 31, 2005
Consolidated Financial Statements
Note 8 –Nonderivative and Derivative Financial Instruments, page 112
| 1. | We note your response to comments 2 and 3 of our letter dated January 30, 2006 and the sample hedge documentation that you provided in your letter dated March 7, 2006. In your responses you appear to acknowledge that the difference in the timing of payments of the hedged item (commission payment obligations) and the payments or receipts on the hedging instrument (interest rate swap or swaps) results in economic ineffectiveness. You state that you believe that SFAS 133 does not require you to include the ineffectiveness related to the timing difference of payment terms (i.e. the effect of discounting) in your assessment of hedge effectiveness or measurement of hedge ineffectiveness, and that you believe paragraphs 65 and 134 of SFAS 133 support your position. Although SFAS 133 has specific guidance that allows such a timing difference to be ignored in a hedge that meets the criteria in paragraph 68 of SFAS 133 (a shortcut method hedge), we do not believe that this guidance can be applied to non-short cut hedges. We also do not believe paragraph 65 provides guidance that ineffectiveness due to the difference in the timing of payments can be ignored in an interest rate hedge. Although applying method 1 as described in DIG Issue G7 will result in no ineffectiveness being recognized for certain non-shortcut cash flow hedges, we believe that the application of method 1 to these hedges will result in ineffectiveness. Method 1 requires a comparison of the present value of the cumulative change in the expected future cash flows on the variable leg of the swap with the present value of the cumulative change in the expected future interest cash flows (or in this case commission cash flows) on the floating–rate liability, with both present value calculations using the discount rate used to fair value the swap. Because this method is a comparison of present values, the timing of payments will impact the determination of the amount of ineffectiveness. We believe that for these hedges the timing difference in the payment terms of the hedging instrument and the hedging instrument causes these hedges to not qualify for the effectiveness assessment and measurement method described in DIG Issue G9. As a result we believe that the criteria for hedge accounting were not met for these hedges since your method of assessing effectiveness did not meet the requirements of 30(b) of SFAS 133. Please tell us how you propose to address the impact on your financial statements. |
1
Response:
We recognize your view that the difference in the timing of payments can only be ignored in a hedge that meets the criteria in paragraph 68 of Statement of Financial Accounting Standards No. 133 (“SFAS 133”) (a short cut method hedge); however, based upon the references cited below, the Financial Accounting Standards Board (“FASB”) does not consider the coincidence of cash flow dates to be a necessary requirement for perfect effectiveness in either the critical terms method or the shortcut method. In the non short cut references cited below, the FASB repeatedly emphasizes reset dates must match but there is no requirement for the payment dates to match.
Paragraph 30(b) addresses the measurement of hedge ineffectiveness for highly effective relationships. This paragraph does not address whether or not the criteria for qualifying for special hedge accounting have been met. Because paragraph 30(b) addresses measurement of hedge ineffectiveness, the consideration of DIG Issue G7 (“G7”) is appropriate in our case. In your letter dated April 3, 2006 you cited the second paragraph in G7 under Method 1: Change in Variable Cash Flows Method as being applicable to the Company. However, we applied the third paragraph of this section of G7, which provides the underlying principle for the determination of no ineffectiveness, stating:
“The change in variable cash flows methodwill result in no ineffectivenessbeing recognized in earnings if the following conditions are met: (1) the floating-rate leg of the swap and the hedged variable cash flows of the asset or liability are based on the same interest rate index, (2) the interest ratereset dates applicable to the floating-rate leg of the swap and to the hedged variable cash flows of the asset or liability are the same, (3) the hedging relationship does not contain any other basis differences (for example, ineffectiveness could be created if the variable leg of the swap contains a cap and the floating-rate asset or liability does not), and (4) the likelihood of the obligor not defaulting is assessed as being probable. However, ineffectiveness would be expected to result from a difference in the indices used to determine cash flows on the variable leg of the swap (for example, the three-month Treasury rate) and the hedged variable cash flows of the asset or liability (for example, three-month LIBOR) or amismatch between the interest rate reset dates applicable to the variable leg of the swap and the hedged asset or liability.”
Our cash flow hedges of commission payments meet all of the criteria listed in (1) through (4) of paragraph 3 of G7 under the heading Method 1, which indicates that there is no ineffectiveness to record. In undiscounted dollar terms, the cash flows from the derivative provide an exact offset to the variability of our hedged cash flows in the same quarterly accounting period. Therefore, paragraph 3 of G7 under the heading Method 1 results in no ineffectiveness to record.
The FASB does not consider insignificant cash flow timing differences to be a source of ineffectiveness. While we note your observation that paragraph 134 is part of a short cut method illustration, Statement 133 addresses the effect of discounting in other paragraphs that do not qualify for using the short cut method and maintains that the focus of a hedger in assessing whether or not a cash flow hedge is highly effective should be based on whether or not theinterest rate reset dates match.
Paragraph 64 of SFAS 133 states, in part, “In assessing the effectiveness of a cash flow hedge, an entity generally will need to consider the time value of moneyif significant,in the circumstances.” The time value of money with respect to our benchmark interest rate swaps is insignificant and thus does not need to be considered in our assessment of ineffectiveness due to its insignificance and immateriality. The average difference in timing of payments on the interest rate swap and the commission payment is approximately 33 days. Other GAAP literature notes that a timing difference of 90 days or less in most cases cannot be construed as “significant” and further note that GAAP does not require discounting of most items with maturity dates of less than a year. Accordingly, the timing difference in the payment terms does not rise to the level of significance that paragraph 64 would require us to assess it for purposes of qualifying for hedge accounting in our Consolidated Statements of Income pursuant to SFAS 133.
2
The overall concept in paragraph 134 is that reset dates of the swap floating leg and the floating aspect of the hedged cash flows must match, but that the actual payment dates need not match. As noted in your April 3, 2006 letter, this is a short cut method paragraph. However, this concept is repeated throughout SFAS 133 in numerous non-short cut method examples, confirming that the third paragraph of Method 1 of Issue G7 (also a non-short cut method illustration) is the correct application as the FASB intended.
In paragraphs 98 and 99 of SFAS 133, the example addresses certain terms of a cash flow hedge with a basis swap. In such example, the FASB points out that the hedge will be ineffective to the extent that LIBOR has changed between the first of the month (the reset date for the swap) and the tenth of the month (the reset date for the liability). The purpose of this non-short cut example is to emphasize the importance of the reset dates matching.
In paragraphs 153 through 155 of SFAS 133, the FASB provides an example of a non-short cut cash flow hedge of forecasted interest payments. In paragraph 155, the FASB points out that since the critical terms match (notional amount, reset dates, and same benchmark interest rate), the entity may conclude that there will be no ineffectiveness in the hedging relationship (absent a default by the swap counterparty). Our cash flow hedges of commission payments meet both criteria (critical terms match and creditworthiness of counterparty is such that the counterparty can perform pursuant to the terms of the contract). Although the payment dates in the example in paragraph 154 appear to match, paragraph 155 only requires that the reset dates match but does not include a requirement for the payment dates to match in order for the entity to conclude that there will be no ineffectiveness.
Implementation Issue H16 addresses the non-short cut cash flow example of a cross currency interest rate swap used to hedge variable-rate foreign currency-denominated debt. Example 2 illustrates a difference in reset dates as an example of a swap that is not perfectly effective, but still highly effective. Once again, the FASB’s repeated emphasis for cash flow hedge purposes is on assessing the importance of differences in reset dates, not cash flow dates.
In both Implementation Issues G12 and G21, the FASB also addresses non-short cut cash flow hedges. In Issue G12, the FASB states that the short cut method cannot be applied to a cash flow hedge of a variable-rate operating lease, but then advises to apply the guidance in Issue G7, which would include the third paragraph of Method 1 of Issue G7 as cited above. In Issue G21, the FASB addresses a non-short cut method cash flow hedge of floating-rate debt that is prepayable at par. The reader is told that the same perfectly effective hypothetical derivative from Issue G7 that would have otherwise qualified for the short cut method could nevertheless be the same as the actual derivative in use, despite the prepayable nature of the floating-rate debt. Issue G21 notes that even non-short cut eligible relationships can result in no ineffectiveness recorded in earnings by applying the principles in Issue G7. We cite Issues G12 and G21 as examples of cash flow hedge guidance where the FASB has applied the same principles in paragraph 134 to non-short cut hedge relationships. Accordingly, this is why we believe that the paragraph 134 guidance that payment dates need not match is not unique to short-cut hedge relationships, as the FASB repeatedly expands the use of Issue G7 throughout the cash flow hedge literature in numerous different examples.
In no circumstances does the FASB state that the effect of discounting, in either a short cut method or non-short cut method cash flow hedge creates ineffectiveness that would cause the hedging relationship to not be highly effective. Method 1 of Issue G7, paragraph 3, suggests that the effect of discounting can be ignored in the measurement of ineffectiveness of qualifying “highly effective” relationships. Accordingly, we have met the criteria for hedge accounting.
We hope the above expanded details help clarify our application of SFAS 133. If you have further questions, we would appreciate the opportunity to discuss this with you either in person or telephonically at your convenience.
3
| 2. | We note your response to comments 3 and 4 of our letter dated January 30, 2006 and the sample hedge documentation that you provided in your letter dated March 7, 2006. It is not clear to us from the documentation you have provided how you determine the specific forecasted transaction that each layered on swap hedges. For example, it is not clear whether the addition of the $38 million notional swap in 2004 changed the existing hedges in place, such that this swap was designed as part of the same hedge as the existing swaps. If this were the case, we would assume that the addition of the 2004 swap would result in the need to re-designate the existing swaps when the hedged transaction is changed to reflect the additional notional of the new swap. Based on the documentation you provided it appears possible that with each swap you are hedging a different amount of forecasted transactions, separate from the forecasted transactions hedged by the pre-existing swaps. If this were the case, it is not clear how your documentation allows you to know what specific $38 million of forecasted outstanding float this swap hedges. For example, if a change in your forecast of probable outstanding float caused you to determine that for 2010 you were over hedged by $20 million, it is not clear how you would determine which, if any, of the swaps that were hedging forecasted float in 2010 was no longer effective. Additionally, if you decided to de-designate the $38 million notional swap, it is not clear how you would know the specific forecasted notional for each year that was no longer hedged as a result of the de-designation. Please tell us how the documentation you put in place at the inception of each of these cash flow hedges provided sufficient specificity about the particular hedged commission payments being hedged such that if you determined that certain forecasted commission payments were determined to be no longer probable you could identify which of the swaps were hedging those particular forecasted commission payments. Please refer to paragraph 28(a) of SFAS 133 and DIG Issue G13. |
Response:
Our documentation of the cash flow hedges of commission payments meets the requirements of SFAS 133 with respect to “sufficient specificity” based on the following:
| 1. | More than 80% of our commission payment obligations are paid on the same day each month to selling agents indicating that the majority of the payments are concurrent and the payments are homogeneous. Each layered on swap hedges an additional amount of the concurrent and homogeneous payments. Accordingly, the “first payments received/made” technique addressed in DIG Issue G13 and DIG Issue G25 is not applicable to our interest rate swaps of commission payments. DIG Issue G25 states that “payments made or received concurrently may not be arbitrarily sorted to minimize or achieve a desired amount of ineffectiveness in the hedging relationship” supporting our position that G25 can not be applied to our interest rate swaps of commission payments. |
| 2. | We are projected to be significantly under hedged by approximately $5.8 billion to $7 billion in each of the years from 2006 through 2018 and historically have always been significantly under hedged. It is the Company’s policy to be under hedged by index rate and on a cumulative basis. Based on the predictability of the underlying transaction, we believe that it is extremely remote that the portfolio of forecasted transactions to the extent of the hedges would no longer be probable of occurring. |
| 3. | Our hedge documentation defines our hedged item as “basket of commission based on T-Bill index rate.” Since the majority of our commission payments are made on the same day, and we are significantly under hedged, our hedge documentation is sufficiently specific with respect to the hedged commission payments based on the T-Bill index rate. |
SFAS 133 indicates that there is no need for our hedge documentation to include further specificity with respect to specific individual commission payments as long as our aggregate hedges related to such basket are less than the aggregate hedged cash flows. However, in the event the aggregate forecasted transactions become less than the aggregate hedges we would be faced with determining the impact on each of our hedging instruments. If this were to occur (which would be in violation of our current hedge program), we could not rely on a pre-determined “ordering” of our derivatives (e.g. FIFO or LIFO) as to which derivative is impacted first. Accordingly, we believe SFAS 133 would require that the effect of our over hedged condition be recorded in our Consolidated Statements of Income, should this ever occur. If the over hedging condition grew to such a severity that we could no longer satisfy an appropriate dollar offset (e.g. 80 %), then we would be forced to discontinue hedge accounting at such date for all the cash flow hedges of forecasted commission payments due to our inability to isolate the impact of over hedging to a single derivative. As we mentioned earlier, we are notably under hedged so the likelihood of this scenario is remote.
4
| 3. | Aside from your cash flow hedges of your commission payment obligation, for each type of hedging relationship, please tell us how you determined that they met the criteria for hedge accounting pursuant to paragraphs 20, 21, 28 and 29 of SFAS 133. Specifically address the following for each type of hedging relationship: |
| • | the nature and terms of the hedged item or transaction; |
| • | the nature and terms of the derivative instruments; |
| • | the specific documented risk being hedged; |
| • | the type of SFAS 133 hedge (fair value, cash flow, etc.); and |
| • | the quantitative measures you used to assess effectiveness of each hedge both at inception and on an ongoing basis. |
Response:
Our response to questions number 7 and 8 of our letter to you dated January 6, 2006 address your questions with respect to the fair value hedges related to certain long-term portfolio investments. Accordingly, our responses to questions 3 and 4 will address the remaining hedging programs consisting of, hedges of net investment in a foreign operation, foreign currency hedges, hedges of certain equity investments and interest rate swaps on fixed rate debt.
Hedge of a Net Investment in a Foreign Operation
| 1. | The nature of the hedged item is our net investment in a specific foreign investment (either 100% ownership or majority ownership) the terms of which are denominated in a currency other than FDC’s functional currency. |
| 2. | The nature of the derivative instrument is a fixed to fixed cross-currency swap. The terms of the swap are as follows: 1) the notional amount is less than the notional amount of the investment in the entity, 2) the derivative instrument has a 10 year term, 3) the derivative instrument includes an initial and final settlement of the currency in which our investment was made, and 4) the swap includes semi-annual interest payments in which FDC pays a fixed rate of interest in the currency in which the investment was made and receives a fixed rate of interest in US dollars. |
| 3. | The specific documented risk that we are hedging is the change in book value of the investment resulting from changes in foreign currency rates (the change in foreign currency rates between the currency in which the investment was made, i.e. Euro in certain of our recent European and our functional currency which is the US dollar). |
| 4. | The hedge is designated as a hedge of a net investment pursuant to SFAS 133. |
| 5. | SFAS 133 requires that the gain or loss on a hedging derivative instrument that is designated as, and is effective as, an economic hedge of the net investment in a foreign operation shall be reported in the same manner as a translation adjustment of the net investment to the extent it is effective as a hedge. The phrase “to the extent it is effective as a hedge” determines the amount of ineffectiveness that is recognized in earnings. The FASB in DIG Issue H8 addresses the issue of how to determine the amount of hedge ineffectiveness in a net investment hedge and provides implementation guidance. Such guidance does not require a quantitative measure to be performed at the inception of the hedge and on an ongoing basis if certain criteria are met. As discussed in response to question number 4, below, we use the forward rate method to measure hedge ineffectiveness and believe that the terms of our net investment hedges meet the criteria in DIG Issue H8 such that no hedge ineffectiveness exists. |
5
Cash Flow Hedges
Foreign Currency Hedges
| 1. | We are hedging the forecasted fee revenue earned in our money transfer business when the fee is paid in a currency other than our functional currency (US dollars). |
| 2. | The nature of the derivative instrument is a foreign currency forward contract. The term of the derivative instrument is to hedge a specific amount of the forecasted fee revenue in the period in which such fee revenue will occur. |
| 3. | The specific documented risk being hedged is the change in cash flows of the fee revenue due to the changes in the foreign currency rate between the currency in which the fee revenue is earned and the US dollar (FDC’s functional currency). |
| 4. | The hedge is a cash flow hedge pursuant to SFAS 133. |
| 5. | Paragraph 65 of SFAS 133 permits us to conclude that changes in cash flows attributable to the foreign currency risk being hedged are expected to completely offset at inception and on an ongoing basis since the critical terms of the forward contract match that of the underlying. Accordingly, SFAS 133 requires that we verify that the critical terms continue to match on an ongoing basis and assess the creditworthiness of the counterparty to ensure that they can perform under the terms of the contract. |
Hedge of an Equity Investment
As disclosed in footnote 8 in our 2005 Form 10-K, during 2005 we sold our investment in CheckFree corporation’s common stock which we hedged using costless collars. Upon the sale of our investment in CheckFree Corporation, the costless collars were terminated and all amounts recorded in OCI were recognized in the “investment gains and losses” line item of the Consolidated Statements of Income.
| 1. | We were hedging the change in the forecasted sale price of the equity investment. |
| 2. | The nature of the derivative instrument was a costless collar. |
| 3. | The specific documented risk being hedged was the volatility in cash flows due to the change in the equity share price. |
| 4. | The hedge was a cash flow hedge pursuant to SFAS 133. |
| 5. | Since (a) the hedging instrument was a costless collar, (b) the exposure being hedged was the variability in expected future cash flows attributed to the price of the equity, and (c) the assessment of effectiveness was based on total changes in the costless collar’s cash flows, DIG Issue G20 permited us to focus on the hedging instrument’s terminal value in determining under paragraph 28(b) whether the hedging relationship was expected to be highly effective in achieving offsetting cash flows attributable to the hedged risk during the term of the hedge. Furthermore, since 1) the critical terms matched, 2) the strike prices of the costless collar matched the specified levels beyond which the exposure is being hedged, 3) the costless collar inflows or outflows at maturity were expected to completely offset the change in the hedged transaction’s cash flows, and 4) the costless collars could only be exercised on their contractual maturity date; we presumed that the hedging relationship was perfectly effective pursuant to DIG G20. Accordingly, we did not have to perform a quantitative assessment of effectiveness. |
Fair Value Hedge
Interest Rate Swap
| 1. | We are hedging the change in interest rates on our fixed rate debt. |
| 2. | The nature of the derivative instrument is an interest rate swap whereby we are exchanging a fixed rate of interest for a variable rate of interest (3 Month LIBOR). |
| 3. | The specific documented risk that we are hedging is the change in interest rates. |
| 4. | The hedge is a fair value hedge pursuant to SFAS 133. |
| 5. | The interest rate swaps meet all the criteria in paragraph 68 of SFAS 133. Accordingly, the hedge will be perfectly effective at inception and on an ongoing basis. |
| 4. | For items discussed in response to comment 5 above, please tell us whether you use the short-cut method or matched terms for assuming no ineffectiveness for any of your hedging relationships that qualify for hedge accounting treatment under SFAS 133. If so, please tell us how you determine that the hedging relationship meets each of the conditions in paragraph 68 or 65 of SFAS 133. |
6
Response:
Hedge of a Net Investment in a Foreign Operation
We determine the amount of ineffectiveness to be recorded in earnings by applying the changes in forward rates method as described in DIG Issue H8. Since the critical terms match in each of our net investment hedges and we have elected to perform our ineffectiveness calculation based on forward rates, no hedge ineffectiveness exists.
Cash Flow Hedges
Foreign Currency Hedges
The assessment of effectiveness for our foreign currency cash flow hedges, both prospectively and retrospectively, is made by confirmation that the critical terms (notional amount, fair value of forward contract is zero at inception of hedge, the period the forecasted transaction occurs is the same period of the forward contract, and the change in expected cash flows on the forecasted transaction is based on the forward exchange rate) match both at inception and on an ongoing basis. As long as these conditions are validated, the hedges are assumed to be highly effective and qualify to use paragraph 65 of SFAS 133. Because there is a perfect match in the critical terms of the hedge and the hedged transaction, no ineffectiveness results. Changes in the fair value of the hedge will be viewed as a proxy for changes in the present value of the change in the cash flows of the hedge item.
Hedge of an Equity Investment
We apply DIG Issue G20 in assessing effectiveness of the costless collars. Since (a) the hedging instrument is a costless collar, (b) the exposure being hedged is the variability in expected future cash flows attributed to the price of the equity, and (c) the assessment of effectiveness is based on total changes in the costless collar’s cash flows, and we meet the four criteria in G20 as stated in our response to sub question 5 of question number 3, above, we conclude that the hedging relationship is considered to be perfectly effective.
Fair Value Hedge
Interest Rate Swap
We apply the shortcut method criteria in paragraph 68 of SFAS 133 for assuming no ineffectiveness with respect to our interest rate swaps on our fixed rate debt. We determine that the hedging relationship meets each of the conditions in paragraph 68 of SFAS 133 as indicated below.
| 1. | The notional amount of the interest rate swap matches the principal amount of the specified percentage of debt being hedged. |
| 2. | The fair value of the swap at inception of the hedging relationship is zero. |
| 3. | The formula for computing net settlements under the swap is the same for each net settlement. |
| 4. | The debt is not prepayable. |
| 5. | The index on which the variable leg of the swap is based (3 Month LIBOR) matches the benchmark interest rate designated as the interest rate risk being hedged for the hedging relationship. |
| 6. | There are no other terms in the debt or the swap that would invalidate our assumption of no ineffectiveness. |
| 7. | The expiration date of the swap matches the maturity date of the debt. |
| 8. | There is no floor or ceiling on the variable interest rate of the swap. |
| 9. | The interval between repricings (3 months) of the variable leg of the swap is frequent enough to justify an assumption that the variable payment or receipt is at a market rate. |
7
* * * *
The Company acknowledges that:
| • | The Company is responsible for the adequacy and accuracy of the disclosures in the Company’s filings; |
| • | Staff’s comments or changes to disclosure in response to staff comments do not foreclose the Commission from taking any action with respect to the Company’s filings; and |
| • | The Company may not assert staff comments as a defense in any proceedings initiated by the Commission or any person under the federal securities laws of the United States. |
If you have any questions or wish to discuss any of our responses, please contact me at 303-488-8388.
Sincerely,
/s/ Kimberly S. Patmore
Kimberly S. Patmore
Executive Vice President and Chief Financial Officer
8
