200 Renaissance Center, Mail Code 482-B12-D21
Detroit, Michigan 48265
(313) 665-0370
Detroit, Michigan 48265
(313) 665-0370
March 12, 2007
Mr. Donald A. Walker, Jr.
Senior Assistant Chief Accountant
United States Securities and Exchange Commission
100 F Street Northeast
Washington, D.C. 20549
Senior Assistant Chief Accountant
United States Securities and Exchange Commission
100 F Street Northeast
Washington, D.C. 20549
Dear Mr. Walker:
This communication is provided in response to your letter dated February 23, 2007, regarding your review of GMAC’s Annual Report on Form 10-K for the Fiscal Year Ended December 31, 2005 filed with the SEC on March 28, 2006, our Quarterly Reports on Form 10-Q for the Fiscal Quarters Ended September 30, 2006, June 30, 2006 and March 31, 2006 and our Form 8-K filed with the SEC on February 16, 2007 (File No. 1-03754). Our responses are presented herein, following each of the respective comments communicated in your letter.
10-K for the Fiscal Year Ended December 31, 2005
Note 16 – Derivative Instruments and Hedging Activities, page 97
| 1. | In light of the proposed restatement of your financial statements for adjustments related to your accounting for hedging activities under SFAS No. 133, please address the following for each type of hedging relationship you have entered into: |
| • | Please tell us how you reassessed or reevaluated your effectiveness testing methodologies for each of your hedging relationships to which you apply hedge accounting for compliance with the requirements of SFAS No. 133; | ||
| • | Please provide an example for each type of hedging relationship showing us your effectiveness testing methodology both prior to and subsequent to your conclusion to restate; | ||
| • | Please tell us if you apply any hurdle, trigger, or two-part effectiveness testing methodologies to any of your hedging relationships; and | ||
| • | Please tell us how you determined that your effectiveness testing methodologies now comply with paragraphs 20 and 28 of SFAS No. 133 for each type of hedging relationship. |
United States Securities and Exchange Commission
Mr. Donald Walker
March 12, 2007
Page 2
Mr. Donald Walker
March 12, 2007
Page 2
Response
As described in our August 17, 2006 response to your July 27, 2006 letter, GMAC had entered into the following primary hedging relationships during the years 2003-2005:
| • | fair value hedges of fixed rate debt (both long-haul and shortcut) at GMAC Global Automotive Finance (GAF) and Residential Capital Corporation (ResCap), mortgage servicing rights (long-haul) at ResCap and mortgage loans and mortgage loans held for sale (long-haul) at ResCap and Commercial Mortgage (CM), as well as | ||
| • | cash flow hedges of floating rate debt (long-haul and shortcut) at GAF, ResCap and CM, foreign currency denominated financial assets and liabilities (matched terms) at GAF and CM, and future issuances of debt (long-haul) at ResCap. |
The restatement described in our Form 8-K of February 16, 2007 was primarily driven by our conclusion, subsequent to the filing of the above referred August 17, 2006 response, that fair value hedges of callable fixed rate debt at GAF should not have been afforded hedge accounting treatment. This conclusion was primarily based on our determination that the hedge effectiveness assessment methodology, in place since GMAC’s adoption of SFAS 133 in 2001, was essentially a trigger-based methodology, which does not satisfy the requirements of SFAS 133.
In addition, the new management team of our previously wholly owned subsidiary CM has concluded that the hedge effectiveness assessment documentation in place for CM’s fair value hedges of fixed rate certificates of deposit and cash flow hedges of variable rate debt did not satisfy the requirements of SFAS 133. Further, ResCap management has determined that their cash flow hedges of anticipated issuances of floating rate debt failed to satisfy the requirements of SFAS 133. Although the impacts of these reconsiderations were not material to our historical financial statements, we have now changed the historical hedge accounting treatment of these hedging relationships.
Please note that we sold approximately 78% of our equity in CM on March 23, 2006. Subsequent to the sale, CM is no longer consolidated in our financial statements and our remaining interest is accounted for under the equity method of accounting.
We have confirmed that none of the remaining hedging relationships utilized any hurdle, trigger, or two-part effectiveness testing methodologies1. We have reviewed and re-confirmed our belief that the hedge effectiveness assessment methodologies and documentation in place for our remaining hedging relationships satisfy the requirements of SFAS 133. The balance of this response will present the examples of effectiveness assessment methodologies utilized by GMAC in each hedging relationship.
Effective January 1, 2006, ResCap elected the fair value measurement method for subsequent measurements of separately recognized servicing assets in accordance with SFAS No. 156,Accounting for Servicing of Financial Assets. As the existing hedge relationships were de-designated as of that date, we have excluded the discussion of the MSR hedging relationships from the remainder of this response.
1 In preparing this response, we have interpreted the phrase “hurdle, trigger, or two-part effectiveness testing methodologies” to mean methodologies in which a limited effectiveness test is performed initially, followed by a second, more robust effectiveness test depending on the results of the initial test.
United States Securities and Exchange Commission
Mr. Donald Walker
March 12, 2007
Page 3
Mr. Donald Walker
March 12, 2007
Page 3
The effectiveness testing methodologies utilized for our remaining hedging relationships are the same — both prior to and following our decision to restate. We refer you to our August 17, 2006 letter for a more complete description of these hedging relationships and responses to earlier staff comments.
Fair Value Hedges of Fixed Rate Debt and Cash Flow Hedges of Floating Rate Debt (which utilize the shortcut method for assuming no ineffectiveness)
All shortcut hedging relationships are evaluated at hedge inception to ensure that the relationship complies with all the required criteria of paragraph 68 of SFAS 133 and that our documentation supports that conclusion. All critical terms in our shortcut hedges are exact mirrors of the corresponding terms in the hedged debt. Any change in key terms of the hedging relationship’s debt or swap (e.g., buy back of debt or partial termination of swap) including a material deterioration in the swap counterparty’s credit risk (as described in the response to your second question which follows) would result in the hedging relationship being de-designated, thus terminating hedge accounting treatment.
All shortcut hedging relationships are evaluated at hedge inception to ensure that the relationship complies with all the required criteria of paragraph 68 of SFAS 133 and that our documentation supports that conclusion. All critical terms in our shortcut hedges are exact mirrors of the corresponding terms in the hedged debt. Any change in key terms of the hedging relationship’s debt or swap (e.g., buy back of debt or partial termination of swap) including a material deterioration in the swap counterparty’s credit risk (as described in the response to your second question which follows) would result in the hedging relationship being de-designated, thus terminating hedge accounting treatment.
Cash Flow Hedges of Foreign Currency Denominated Financial Assets and Liabilities (which utilize the critical matched terms method for assuming no ineffectiveness)
All critical matched terms hedging relationships are evaluated at hedge inception to ensure that all critical terms of the hedged item and the hedging instrument are mirror images of each other and that our documentation supports that conclusion in accordance with the guidance provided by paragraph 65 of SFAS 133 and DIG Issue No. G9,Cash Flow Hedges: Assuming No Ineffectiveness Critical Terms of the Hedging Instrument and the Hedged Transaction Match in a Cash Flow Hedge (DIG G9). Any change in key terms of the hedging relationship (changes in either the hedged item or the hedging instrument and/or a material deterioration in the counterparty’s credit risk as described in the response to your second question which follows) would result in the hedging relationship being de-designated, thus terminating hedge accounting treatment. As we were finalizing our decision to restate for the fair value hedges of the callable fixed rate debt, we reaffirmed that all critical terms of these relationships are matched.
All critical matched terms hedging relationships are evaluated at hedge inception to ensure that all critical terms of the hedged item and the hedging instrument are mirror images of each other and that our documentation supports that conclusion in accordance with the guidance provided by paragraph 65 of SFAS 133 and DIG Issue No. G9,Cash Flow Hedges: Assuming No Ineffectiveness Critical Terms of the Hedging Instrument and the Hedged Transaction Match in a Cash Flow Hedge (DIG G9). Any change in key terms of the hedging relationship (changes in either the hedged item or the hedging instrument and/or a material deterioration in the counterparty’s credit risk as described in the response to your second question which follows) would result in the hedging relationship being de-designated, thus terminating hedge accounting treatment. As we were finalizing our decision to restate for the fair value hedges of the callable fixed rate debt, we reaffirmed that all critical terms of these relationships are matched.
Cash Flow Hedges of Floating Rate Debt (which utilize method 1 — change in variable cash flows — of DIG Issue No. G7 for assuming no ineffectiveness)
Similar to Company A’s hedging strategy in DIG Issue No. G13,Cash Flow Hedges: Hedging the Variable Interest Payments on a Group of Floating-Rate Interest Bearing Loans, GMAC has defined the hedged item as a specified first dollar amount of the expected probable future interest payments based upon a forecasted aggregate dollar principal balance of variable rate bond obligations and projects future bond balances using expected prepayment activity and expected debt issuance. This hedging relationship does not meet the criteria for use of the short-cut method because the debt is prepayable via a clean-up call on the related securitized assets. Although this hedging relationship does not qualify for the shortcut method, GMAC’s designation of the hedged risk as the “first dollar” of interest payments, coupled with GMAC’s ability to project (based upon projected prepayment and debt issuance activity) that it is probable that such cash flows will always occur, GMAC has been able to assess that these hedging relationships will result in the measurement of zero ineffectiveness via application of Method 1 – Change in Variable Cash Flows Method of DIG Issue No. G7,Cash Flow Hedges: Measuring the Ineffectiveness of a Cash Flow Hedge under Paragraph 30(b) When the Shortcut Method is Not Applied (DIG G7). Specifically, GMAC evaluates the conditions listed below in our initial assessment and documentation of the hedging relationship and reassesses effectiveness each quarterly
Similar to Company A’s hedging strategy in DIG Issue No. G13,Cash Flow Hedges: Hedging the Variable Interest Payments on a Group of Floating-Rate Interest Bearing Loans, GMAC has defined the hedged item as a specified first dollar amount of the expected probable future interest payments based upon a forecasted aggregate dollar principal balance of variable rate bond obligations and projects future bond balances using expected prepayment activity and expected debt issuance. This hedging relationship does not meet the criteria for use of the short-cut method because the debt is prepayable via a clean-up call on the related securitized assets. Although this hedging relationship does not qualify for the shortcut method, GMAC’s designation of the hedged risk as the “first dollar” of interest payments, coupled with GMAC’s ability to project (based upon projected prepayment and debt issuance activity) that it is probable that such cash flows will always occur, GMAC has been able to assess that these hedging relationships will result in the measurement of zero ineffectiveness via application of Method 1 – Change in Variable Cash Flows Method of DIG Issue No. G7,Cash Flow Hedges: Measuring the Ineffectiveness of a Cash Flow Hedge under Paragraph 30(b) When the Shortcut Method is Not Applied (DIG G7). Specifically, GMAC evaluates the conditions listed below in our initial assessment and documentation of the hedging relationship and reassesses effectiveness each quarterly
United States Securities and Exchange Commission
Mr. Donald Walker
March 12, 2007
Page 4
Mr. Donald Walker
March 12, 2007
Page 4
reporting period by verifying that no changes in these terms have occurred. Under this measurement methodology, there is no ineffectiveness recognized when (as provided by DIG G7):
| 1. | the floating rate leg of the swap and the hedged variable interest payments of the bonds are based on the same interest rate index, | ||
| 2. | the interest rate reset dates applicable to the floating rate leg of the swaps and to the hedged variable interest payments of the bonds are the same, | ||
| 3. | the hedging relationship does not contain any other basis differences, | ||
| 4. | the likelihood of the counterparty not defaulting is assessed as being probable, | ||
| 5. | the initial fair value of the swap at the inception of the hedging relationship is zero or somewhat near zero, and | ||
| 6. | the notional amount of the swap matches the principal amount of the bonds being hedged. |
On an ongoing basis, as long as no changes in these conditions have occurred, GMAC is able to recognize no ineffectiveness in earnings.
Fair Value Hedges of HFS Loans (which utilize regression analysis)
ResCap — Loans held for sale are aggregated into similar asset classifications based upon their projected price sensitivity to interest rate shocks. Each asset group within a similar asset classification must have a similar weighted-average price change within an 80% to 120% range within all shock points. Due to changes in the underlying pools of hedged loans, these hedging relationships are de-designated and re-designated on a daily basis. Scenario analyses used in these assessments are updated, and portfolios are rebalanced daily. Hedge effectiveness of each daily hedge period is assessed based upon a historical regression analysis, which is applied consistently to its designated loans held for sale portfolio. To be considered highly effective, three statistical thresholds must be met (1) R2 between 0.80 and 1.00, (2) slope between - -0.8 and -1.25 and (3) F-statistic in excess of the critical value based upon the number of data points used in the regression (using a 95% confidence interval).
CM — Hedged mortgage loans are aggregated into groups of similar assets with similar risk attributes. These hedging relationships are designated and de-designated on a daily basis. Daily, each new loan is evaluated and designated to a hedge pool which shares the same exit strategy, risk composition and similar duration. Each portfolio is modeled on a daily basis and contains the duration, dollar value of a one basis point movement in interest rates, weighted average life, and gain/loss for each similar asset within the pool. In order to substantiate that each individual product type continues to meet the designation of a similar asset, shock analyses are performed on a monthly basis for each “product” group/strata. In order for the loans to be grouped together and qualify as a portfolio of similar assets, each asset group must have a similar weighted-average price change within an 80% to 120% range across all shock points. The effectiveness of these hedge relationships is assessed both on a prospective and retrospective basis utilizing regression analysis. The data points used in the regression analysis are the change in the fair value of the hedging instrument and the change in the total fair value of the hedged item. Each day effectiveness is assessed by calculating R2 (must have a value between .80 and 1.00 to be considered highly effective otherwise
United States Securities and Exchange Commission
Mr. Donald Walker
March 12, 2007
Page 5
Mr. Donald Walker
March 12, 2007
Page 5
hedge accounting is not applied). At the end of each month, we again assess the effectiveness of these hedging relationships using regression analysis and to be considered highly effective, three statistical thresholds must be met (1) R2 between 0.80 and 1.00, (2) slope between -0.8 and -1.25, and (3) F-statistic in excess of the critical value based upon the number of data points used in the regression (using a 95% confidence interval) otherwise hedge accounting is not applied for the entire month.
| 2. | For each of your hedging relationships, specifically tell us how you considered the differential between your credit rating and the counterparty’s credit rating in identifying and in assessing hedge effectiveness, both initially and on an ongoing basis. |
Response
By way of background, GMAC manages counterparty credit risk through periodic monitoring and approval of financially sound counterparties and through limiting the potential credit exposures to individual counterparties to predetermined exposure limits. In fact, GMAC has never incurred a loss due to the failure of a derivative counterparty to perform according to the terms of the derivative contract. At December 31, 2006 and 2005, the market value of derivative financial instruments in an asset or receivable position (from GMAC’s perspective) was $2.5 billion and $3.0 billion, including accrued interest of $0.6 billion and $0.7 billion, respectively. As of December 31, 2006, more than 74% of our exposure is with counterparties with a Fitch rating of A+ or higher (or an equivalent rating from another rating agency if a counterparty is not rated by Fitch), as compared with more than 84% as of December 31, 2005.
Each of GMAC’s reporting segments involved in derivative transactions (GAF, ResCap and CM) monitors and manages counterparty credit risk at the reporting segment level, within aggregate limits set by GMAC. Each reporting segment has a derivatives risk management policy which, amongst other things, calls for setting aggregate exposure limits for each counterparty based on the counterparty’s size and credit rating, measuring counterparty credit risk for derivative transactions according to mark to market as well as “Maximum Potential Exposure” calculations and reporting of outstanding derivatives portfolio at least quarterly. In general, theses policies are designed and implemented to ensure that GMAC is not at risk of counterparty default.
When a counterparty’s credit rating decreases below the A- level (Fitch equivalent) established by the policy, GMAC management evaluates the circumstances surrounding that counterparty including exposure, number of trades, alternative counterparties in the market place, etc., to determine whether to terminate GMAC’s existing derivative transactions with the counterparty or, in instances when management does not expect the counterparty to default, to seek a written affirmative approval from the respective management committee to maintain the transactions. Management has demonstrated a practice of actively managing counterparty exposures and terminating derivative transactions with large receivable positions in order to reduce exposures to individual counterparties. Additionally, we reduce credit risk on the majority of our derivative financial instruments by entering into legally enforceable agreements that permit the closeout and netting of transactions with the same counterparty upon occurrence of certain events. To further mitigate the risk of counterparty default, GMAC maintains collateral agreements with certain counterparties. As applicable. the agreements require
United States Securities and Exchange Commission
Mr. Donald Walker
March 12, 2007
Page 6
Mr. Donald Walker
March 12, 2007
Page 6
both parties to maintain cash deposits in the event the fair values of the derivative financial instruments meet established thresholds.
The remainder of this response addresses your question as applied in our effectiveness assessment methodologies. We have limited the response to the four methodologies not involved in the restatement -
| • | shortcut, | ||
| • | critical matched terms, | ||
| • | change in variable cash flows, and | ||
| • | regression. |
Fair Value Hedges of Fixed Rate Debt and Cash Flow Hedges of Floating Rate Debt (which utilize the shortcut method for assuming no ineffectiveness)
GMAC only enters into interest rate swaps with counterparties whose credit ratings meet GMAC’s pre-set limits. At hedge inception, GMAC fully expects the counterparty to meet its contractual obligations under the contractual provisions of the derivative. In accordance with paragraph 70 of SFAS 133, the difference between GMAC’s credit rating and the counterparty’s credit rating does not preclude GMAC from assuming no ineffectiveness in a hedge of interest rate risk.
GMAC only enters into interest rate swaps with counterparties whose credit ratings meet GMAC’s pre-set limits. At hedge inception, GMAC fully expects the counterparty to meet its contractual obligations under the contractual provisions of the derivative. In accordance with paragraph 70 of SFAS 133, the difference between GMAC’s credit rating and the counterparty’s credit rating does not preclude GMAC from assuming no ineffectiveness in a hedge of interest rate risk.
Excerpt from paragraph 70 of SFAS 133:
Comparable credit risk at inception is not a condition for assuming no ineffectiveness even though actually achieving perfect offset would require that the same discount rate be used to determine the fair value of the swap and of the hedged item or hedged transaction. To justify using the same discount rate, the credit risk related to both parties to the swap as well as to the debtor on the hedged interest-bearing asset (in a fair value hedge) or the variable-rate asset on which the interest payments are hedged (in a cash flow hedge) would have to be the same. However, because that complication is caused by the interaction of interest rate risk and credit risk, which are not easily separable, comparable creditworthiness is not considered a necessary condition to assume no ineffectiveness in a hedge of interest rate risk.
In fact as described in the general comments section of the response to Derivative Implementation Group Issue No. E4,Hedging—General: Application of the Shortcut Method(DIG E4), although changes in the credit rating of the counterparty could create ineffectiveness, the shortcut method allows companies not to recognize that ineffectiveness in current earnings.
Excerpt from DIG E4:
Statement 133 acknowledges in paragraph 70 that a hedging relationship that meets all of the applicable conditions in paragraph 68 may nevertheless involve some ineffectiveness (notwithstanding the supposed “assumption of no ineffectiveness”). Yet Statement 133 permits application of the shortcut method, which does not recognize such ineffectiveness currently in earnings. For example, the change in the fair value of an interest rate swap may not offset the change in the fair value of a fixed-rate receivable attributable to the hedged risk (resulting in hedge ineffectiveness) due to a change in the creditworthiness of the counterparty on the swap . Although an expectation of such hedge ineffectiveness potentially could either (a) preclude fair value hedge accounting
United States Securities and Exchange Commission
Mr. Donald Walker
March 12, 2007
Page 7
Mr. Donald Walker
March 12, 2007
Page 7
at inception or (b) trigger current recognition in earnings under regular fair value hedge accounting, the shortcut method masks that ineffectiveness and does not require its current recognition in earnings. In fact, the shortcut method does not even require that the change in the fair value of the hedged fixed-rate receivable attributable to the hedged risk be calculated.
On an ongoing basis, GMAC’s monitoring of counterparty credit ratings ensures that any increase in counterparty default risk is identified and evaluated as to whether GMAC can still rely on the counterparty to fulfill its obligations and conclude that the hedge will remain a highly effective hedge of the designated risk (changes in the fair values or cash flows related changes in the designated benchmark interest rate). This practice is in line with the discussion included in DIG Issue No. G10,Cash Flow Hedges: Need to Consider Possibility of Default by the Counterparty to the Hedging Derivative(DIG G10) which, while focused on cash flows hedges, also addresses the issue for shortcut method hedges of either fair values or cash flows.
Excerpt from DIG G10:
In applying the shortcut method for hedges of interest rate risk with interest rate swaps (as discussed in paragraphs 68, 114, and 132), an entity must similarly consider the likelihood of the counterparty’s compliance with the contractual terms of the hedging derivative that require the counterparty to make payments to the entity. Implicit in the criteria for the shortcut method is the requirement that a basis exists for concluding on an ongoing basis that the hedging relationship is expected to be highly effective in achieving offsetting changes in fair values or cash flows.
As further required for cash flow hedges by the discussion in DIG G10, in order to comply with paragraph 28(b) a company must evaluate the possibility that the counterparty in a cash flow hedge will not default to be at least probable in order to conclude that the hedge will remain highly effective in achieving offsetting cash flows. Management believes that GMAC’s derivative risk management policies allow us to reach that conclusion as the policies implicitly require termination of any derivative transaction which would not meet that criterion.
Cash Flow Hedges of Foreign Currency Denominated Financial Assets and Liabilities (which utilize the critical matched terms method for assuming no ineffectiveness)
Similar to the preceding discussion, through GMAC’s derivative risk management policies, GMAC only enters into derivatives with counterparties whose credit ratings meet GMAC’s pre-set limits and for whom management has concluded that the possibility of non-default is at least probable. Prospective assessments of effectiveness are performed qualitatively in accordance with DIG G9, which requires an entity to assess any adverse changes in counterparty default risk.
Similar to the preceding discussion, through GMAC’s derivative risk management policies, GMAC only enters into derivatives with counterparties whose credit ratings meet GMAC’s pre-set limits and for whom management has concluded that the possibility of non-default is at least probable. Prospective assessments of effectiveness are performed qualitatively in accordance with DIG G9, which requires an entity to assess any adverse changes in counterparty default risk.
Excerpt from DIG G9:
Because the assessment of hedge effectiveness in a cash flow hedge involves assessing the likelihood of the counterparty’s compliance with the contractual terms of the derivative designated as the hedging instrument, the entity must also assess whether there have been adverse developments regarding the risk of counterparty default, particularly if the entity planned to obtain its cash flows by
United States Securities and Exchange Commission
Mr. Donald Walker
March 12, 2007
Page 8
Mr. Donald Walker
March 12, 2007
Page 8
liquidating the derivative at its fair value. If there are no such changes in the critical terms or adverse developments regarding counterparty default, the entity may conclude that there is no ineffectiveness to be recorded. In that case, the change in fair value of the derivative can be viewed as a proxy for the present value of the change in cash flows attributable to the risk being hedged.
Management believes that GMAC’s derivative risk management policies satisfy this requirement as the policies implicitly require termination of any derivative transaction which would not meet that criterion.
Cash Flow Hedges of Floating Rate Debt (which utilize method 1 — change in variable cash flows — of DIG Issue No. G7 for assuming no ineffectiveness)
Similar to the preceding discussions, through GMAC’s derivative risk management policies, GMAC only enters into derivatives with counterparties whose credit ratings meet GMAC’s pre-set limits and for whom management has concluded that the possibility of non-default is at least probable. For all non-shortcut cash flow hedges of interest rate risk, differences between the credit spread of the hedged item and that included in the discount rate used to value the hedging instrument will have no impact on the hedge effectiveness.
Similar to the preceding discussions, through GMAC’s derivative risk management policies, GMAC only enters into derivatives with counterparties whose credit ratings meet GMAC’s pre-set limits and for whom management has concluded that the possibility of non-default is at least probable. For all non-shortcut cash flow hedges of interest rate risk, differences between the credit spread of the hedged item and that included in the discount rate used to value the hedging instrument will have no impact on the hedge effectiveness.
As discussed in the response section of DIG G7, the discount rates applicable to determining the fair value of the swap designated as the hedging instrument should also be applied to the computation of the present value of the cumulative changes in the hedged cash flows because the sole source of variability on the cash flows of the floating leg of the swap and the cash flows of the floating rate debt is the same floating rate index. Thus, no ineffectiveness results for either initial differences in the credit spread or subsequent changes in the credit spread between the hedged item and the hedging instrument. Further, for these hedging relationships, GMAC uses the change in variable cash flows method described in DIG G7, which allows an entity to assume no ineffectiveness in certain situations when specific criteria are met. The DIG G7 criteria include a requirement that an entity assess the derivative counterparty’s likelihood of not defaulting as at least probable. Management believes that GMAC’s derivative risk management policies allow us to reach that conclusion as the policies implicitly require termination of any derivative transaction which would not meet that criterion.
Fair Value Hedges of Held For Sale Loans (which utilize regression analysis)
GMAC management believes differentials in, as well as changes in, counterparty and GMAC credit ratings have the potential to create ineffectiveness for this fair value hedging relationship. The potential ineffectiveness is driven by the differing impact of credit or performance risk on the fair value of the hedging instrument. The degree of impact depends on, among other things, the hedge period, the hedging instrument, and the level of collateral or other supporting elements to the relationship.
GMAC management believes differentials in, as well as changes in, counterparty and GMAC credit ratings have the potential to create ineffectiveness for this fair value hedging relationship. The potential ineffectiveness is driven by the differing impact of credit or performance risk on the fair value of the hedging instrument. The degree of impact depends on, among other things, the hedge period, the hedging instrument, and the level of collateral or other supporting elements to the relationship.
Hedging instruments used by CM in this hedging relationship are subject to a number of clauses that reduce credit risk. These clauses include: credit support annexes (CSAs), and netting.
| • | CSAs: CSAs generally require daily mark-to-market (MTM) of the derivative and include the use of MTM thresholds of $5 million. Collateral is posted in either direction if the MTM value exceeds the contract’s relevant threshold. |
United States Securities and Exchange Commission
Mr. Donald Walker
March 12, 2007
Page 9
Mr. Donald Walker
March 12, 2007
Page 9
| • | Netting: Since these hedges are typically entered under a standardized master agreement (e.g., International Swaps and Derivative Association (ISDA) agreements) with the hedge counterparty, if a Company defaults on one contract with a counterparty then it must default on all outstanding contracts with the counterparty. |
Supplementing these specific transactional enhancements, counterparty credit risk is further mitigated by GMAC’s previously discussed derivative risk management policies, resulting in minimal impact to the effectiveness of the hedging relationship.
While the hedging instruments used by CM are covered by Master ISDA agreements and CSAs, the primary hedging instruments in ResCap’s program are liquid Fannie Mae Mandatory Forward sale commitments. These contracts generally are not written on an ISDA contract and generally do not require posting of collateral. A forward commitment represents an agreement to sell mortgage loans or mortgage-backed securities at a specified date and price. At the time the contract is entered, the delivery price is set so that the contract has a value of zero for both parties. Forward commitments are used to lock in a price today for loans either already warehoused or still in the pipeline that will be sold at a future time. These commitments are frequently referred to as To-Be-Announced (TBA) securities. The duration of a TBA security can vary; generally ranging between one to four months following trade date, with an average duration at any given point of approximately 45 days.
Historically, the value of TBA contracts has not varied as a result of changes in GMAC’s credit rating because TBA collateral is agency-backed, the contract term is short and it generally pertains to funded loans.
With respect to the hedged item, while the valuations of loans held for sale depend on the underlying borrowers’ credit risk, GMAC’s credit risk has no impact on most HFS loan pools. In fact, GMAC’s credit risk has only a minimal impact on valuations of certain loan pools covered by agency guarantees, in that GMAC’s credit rating is one consideration amongst many factors (e.g., size relative to competitors, servicing ability, loan type) considered by the agency in setting the guarantee cost, with loan type being the most significant factor.
Conclusion
While credit rating differentials are clearly understood and monitored by GMAC at the inception of a hedging relationship as well as on an ongoing basis for all hedging relationships, for hedging relationships which assume no ineffectiveness, the existing guidance provided by SFAS 133 and DIGs E4, G7, G9 and G10 allow GMAC to focus on ensuring that the likelihood that the counterparty will not default remains at least probable in order to enable management to conclude that these hedging relationships will remain highly effective. Management believes that GMAC’s derivative risk management policies allow us to reach that conclusion as the policies implicitly require termination of any derivative transaction which would not meet that criterion.
While credit rating differentials are clearly understood and monitored by GMAC at the inception of a hedging relationship as well as on an ongoing basis for all hedging relationships, for hedging relationships which assume no ineffectiveness, the existing guidance provided by SFAS 133 and DIGs E4, G7, G9 and G10 allow GMAC to focus on ensuring that the likelihood that the counterparty will not default remains at least probable in order to enable management to conclude that these hedging relationships will remain highly effective. Management believes that GMAC’s derivative risk management policies allow us to reach that conclusion as the policies implicitly require termination of any derivative transaction which would not meet that criterion.
For fair value hedges of loans held for sale, based on the preceding discussion of relevant aspects of the hedges used and hedged items in GMAC’s specific hedging relationships as well as GMAC’s derivative risk management policies, GMAC
United States Securities and Exchange Commission
Mr. Donald Walker
March 12, 2007
Page 10
Mr. Donald Walker
March 12, 2007
Page 10
management believes that the differential in GMAC’s credit rating and the counterparty’s does not significantly impact hedge effectiveness initially or on an ongoing basis. The impact on the hedging instrument of differences in or changes in either party’s credit rating are either (1) handled outside the valuation of the hedge through CSAs and netting or (2) insignificant due to the short term nature of the hedging relationship. With respect to the held for sale loans, GMAC’s credit risk has a de minimus impact on valuations. Finally, GMAC’s derivative risk management policies require the monitoring of credit exposures we have to all derivative counterparties and implicitly require termination of any derivative transaction in which the counterparty’s likelihood of default became probable.
* * * * *
In addition to the above responses, following is the statement that you requested.
The company acknowledges that:
| • | We are responsible for the adequacy and accuracy of the disclosure in the filings; | ||
| • | 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. |
We appreciate your prompt attention and responsiveness to this matter. In the event further information is necessary please feel free to contact us. I can be reached at (313) 665-0370 or you may contact Linda Zukauckas, GMAC’s Controller and Principal Accounting Officer, at (313) 665-4327.
Sincerely,
/s/ Sanjiv Khattri
Sanjiv Khattri
Chief Financial Officer
Chief Financial Officer
cc: Amit Pande
Assistant Chief Accountant, United States Securities and Exchange Commission
Dave Irving
Staff Accountant, United States Securities and Exchange Commission
Eric Feldstein
Chief Executive Officer, GMAC LLC
Linda Zukauckas
Controller and Principal Accounting Officer, GMAC LLC
Jim Young
Chief Accounting Officer, Residential Capital, LLC