SecProbe.io

CORRESP
1
filename1.htm

corresp

    J. David Williamson

Senior Executive Vice President and Chief Financial Officer

e-mail: david.williamson@cibc.com

Direct : 416-956-3364

Assistant : 416-956-3387

April 04, 2008

VIA EDGAR

Mark Webb

Branch Chief, Financial Services Group

United States Securities and Exchange Commission

Washington, D.C. 20549

    Re:

     Canadian Imperial Bank of Commerce

Form 40-F for the Fiscal Year Ended October 31, 2007

File Number 001-14678

Dear Mr. Webb:

     Thank you for your letter dated March 19, 2008, addressed to Michael G. Capatides of Canadian
Imperial Bank of Commerce (“CIBC” or “the Company”), setting forth comments of the staff of the
Division of Corporation Finance (the “Staff”) on CIBC’s Form 40-F for the fiscal year ended October
31, 2007.

     We appreciate the effort that went into the Staff’s comments. We have provided our responses
in the order the comments appeared in the Staff’s comment letter, which we have reproduced in bold
face text. Our responses follow each comment.

Form 40-F

Exhibit B.3(c), Management’s Discussion and Analysis

U.S. Subprime Residential Mortgage Market, page 40

    1.

    In the disclosure of your exposure to the U.S. subprime residential mortgage market through
derivative contracts hedged with investment-grade counterparties, we note that as of October
31, 2007, 35% of the notional amount of derivatives was with an “A” rated U.S. financial
guarantor. This counterparty was subsequently confirmed to be ACA Financial Guaranty Corp. in
your Form 6-K filed December 19, 2007, in which you disclosed that there was a reasonably high
probability of incurring a large charge following ACA’s credit rating being downgraded from
“A” to “CCC” by Standard & Poors. You further stated that the mark of the hedge protection
from ACA was $1.71 billion as of October 31, 2007, and $2.0 billion as of November 30, 2007.

    We note that in your Form 6-K filed January 15, 2008, you confirmed that you had recorded a
$2 billion fair value adjustment to the counterparty credit protection receivable from ACA
as a result of the credit downgrade. Taking into account your discussion of risk
measurement processes and risk rating methodologies presented on page 63, please tell us how
you considered ACA Capital Holdings, Inc.’s press release dated November 9, 2007, which
discloses Standard & Poor’s decision to place them on CreditWatch with negative
implications, in assigning an internal risk rating to the ACA exposure and determining that
a fair value adjustment was not required as of October 31, 2007.

    Our approach in fair valuing the hedges with ACA as of October 31, 2007, took into account
all available market and counterparty credit information as of that date. The derivative
contracts were valued on a gross mark to market basis less a valuation adjustment for credit
risk to ACA. The mark to market values were based on external quotes. Our credit valuation
adjustment was calculated on a formulaic basis based on the gross mark to market, an
estimate of potential future exposure and the internal credit risk rating of ACA. The
internal credit risk rating was determined based on a number of established parameters that
were both quantitative and qualitative in nature. Rating agency information, views and
ratings outlook (e.g. credit watch with negative implications) were considered in the
context of the full analysis of the counterparty credit risk.

    Our credit analysis of ACA took into account all available credit information as of October 31, 2007,
including Standard & Poor’s (S&P) review on October 29, 2007 which confirmed ACA’s rating at A
with a stable outlook. In the case of financial guarantors, rating agency risk ratings have recently become a more
significant factor in the determination of our internal risk ratings, as we believe that they have a significant impact on
the public’s perception of the companies’ viability. Based on our analysis, we assigned an internal credit risk
rating equivalent to BBB minus for ACA as at October 31, 2007.

2

    As noted on November 9, 2007, S&P revised its rating from A (stable) to A (negative watch).
While representing a negative development, given the A rating was not changed as per S&P’s
definition of CreditWatch below, we viewed our internal credit risk rating (equivalent to
BBB minus) as still appropriate pending further developments.

CreditWatch highlights the potential direction of a short- or long-term rating. It
focuses on identifiable events and short-term trends that cause ratings to be placed
under special surveillance by Standard & Poor’s analytical staff. These may include
mergers, recapitalizations, voter referendums, regulatory action, or anticipated
operating developments. Ratings appear on CreditWatch when such an event or a
deviation from an expected trend occurs and additional information is necessary to
evaluate the current rating. A listing, however, does not mean a rating change is
inevitable, and whenever possible, a range of alternative ratings will be shown.
CreditWatch is not intended to include all ratings under review, and rating changes
may occur without the ratings having first appeared on CreditWatch. The “positive”
designation means that a rating may be raised; “negative” means a rating may be
lowered; and “developing” means that a rating may be raised, lowered, or affirmed.

    As at October 31, 2007, our valuation of derivative contracts with ACA was net of a credit
valuation adjustment of US$17 million.

Liquidity Facilities to ABCP Conduits, page 40

    2.

    We note that during the second half of 2007 you purchased $3.1 billion asset-backed
commercial paper issued by your sponsored conduits as a result of liquidity tightening that
had affected the asset-backed securities market. We also note disclosures in your Critical
Accounting Policies and Estimates section on page 76 relating to consolidation of variable
interest entities (VIEs) to which you provide liquidity facilities, and note that as of
October 31, 2007, you had not consolidated any of these entities. Please provide us with a
comprehensive analysis clearly stating the authoritative guidance you relied upon to support
your determination that you were not the primary beneficiary and therefore not the
consolidator of the sponsored conduits to which you provided liquidity facilities.

    As at October 31, 2007, we concluded that we were not the primary beneficiary of any of the
CIBC sponsored conduits (the “Conduits”) under AcG-15 Consolidation of Variable Interest
Entities as our holdings of Asset Backed Commercial Paper (“ABCP”) issued by the Conduits,
together with our other variable interests in the Conduits, did not result in an expectation
that CIBC would absorb the majority of expected losses.

    In making this determination, we considered 1) the nature of the risks in the Conduits and
2) the variability the Conduits were designed to create and pass along to their variable
interest holders.

    The risks we identified included credit risk associated with the potential default of the
assets held by the Conduits and certain interest rate spread risks. These risks are
reflected in the variable interests that we identified and assessed below.

3

    The variable interests that we identified included the ABCP held by CIBC and other parties,
the Global Style Liquidity facility (“GSL”) provided by CIBC to the Conduits, and certain
fees earned by CIBC from the Conduits that are deemed to be variable interests. We
determined that the credit enhancements provided by the sellers to the Conduits are not
considered variable interests under AcG-15.

    We assessed that the variable interests flowing to CIBC as a result of providing the GSL
facility are insignificant to the primary beneficiary analysis. The pricing for the
liquidity line upon a draw is generally expected to be higher to the Conduits than the cost
of raising ABCP. The GSL facilities are used by the Conduits to manage liquidity risk (i.e.
the inability of the Conduits to roll their short-term debt), but not credit risk associated
with the potential default of the underlying assets as amounts that can be drawn under the
liquidity facility are reduced with respect to assets that are in default. Furthermore, the
exposure of credit risk to CIBC is further reduced given the high quality of underlying
assets and substantial credit enhancements provided by the sellers to the Conduits. Since
the pricing of the liquidity would be more expensive than the cost of issuing ABCP and the
proceeds of a draw from the liquidity line can only be used to repay maturing ABCP, the
Conduits are expected to fund themselves through the roll of ABCP rather than through the
liquidity line. Therefore, we expect that the liquidity line would only be used to manage
extreme liquidity distress as a last resort; as a result, we have assessed that the
probability of a draw is extremely low. Consequently, the variable interest as a result of
providing the liquidity facility is insignificant to the primary beneficiary analysis.

    We have also assessed that the variability of certain fees earned by CIBC (deemed to be
variable interests) was very low.

    We concluded that the credit risk associated with the underlying assets held by the Conduits
and interest rate spread risks are the only material sources of variability and that these
risks substantially accrue to the ABCP holders as a group on a pro-rata basis. Therefore,
we concluded that we are not the primary beneficiary of the Conduits to the extent that our
ABCP holdings are sufficiently below 50% to take into account the small amount of
variability we absorb as the provider of the GSL facility and as an earner of certain fees.

    As at October 31, 2007, we did not consolidate any of the Conduits as we owned significantly
less than 50% of the ABCP issued by each of the Conduits and therefore did not consider
ourselves the primary beneficial interest holder. However, we would note that during the
period from September 10, 2007 to October 17, 2007 (inclusive), we were the primary
beneficiary of one of the Conduits, Sound Trust, due to the level of our ABCP holdings in
that entity during the period. As a result, we made appropriate consolidation adjustments.
The level of our holdings in Sound Trust dropped below the consolidation threshold by
October 31, 2007 and as such it was not consolidated at that date.

    3.

    Given that you have identified consolidation and variable interest entities as a critical
accounting policy, please consider discussing in your future filings:

    •

    The scenarios where you would have to consolidate any of these entities for which
you provide liquidity facilities, and your expectation of the likelihood of such
consolidation; and

4

    •

    The frequency of which you reconsider, and the typical triggers which require you to
reconsider, whether you are the primary beneficiary and therefore would be required to
consolidate these entities.

We will disclose in our future Form 40-F filings that we will reconsider our primary
beneficiary analysis in the Conduits whenever our level of interest in the ABCP issued by
the Conduits significantly changes, or in the less frequent event that the liquidity
protection we provide to the Conduits is drawn or amended. We will also disclose that
consolidation of a particular Conduit would become more likely to the extent that our ABCP
holdings in that Conduit approaches 45% as a result of the relatively small amount of
variability stemming from our other variable interests in the Conduit. Finally, we will
disclose that a significant increase in our holdings of ABCP issued by the Conduits would
become more likely in a scenario in which the market for bank sponsored ABCP suffered a
significant deterioration such that the Conduits were unable to roll their ABCP.

Visa Gain, Page 40

    4.

    We note your disclosure that you recognized a $456 million gain with respect to the worldwide
restructuring of Visa based on the fair value of shares of Visa, Inc. that you received on
October 2, 2007. Please address the following:

    •

    Tell us how you previously accounted for your membership interests in Visa Canada
and Visa International prior to the restructuring; and

We accounted for our membership interests in Visa Canada and Visa International at their
nominal cost prior to the restructuring in October 2007.

    •

    Tell us how you determined the fair value of shares of Visa, Inc. received in the
absence of an active market for these shares. Please provide us with a comprehensive
response which includes the key assumptions made and valuation models employed.

    A valuation report prepared by an independent valuation firm on behalf of the major Canadian
banks that participated in the Visa restructuring was used as the basis for the
determination of the estimated fair value of the Visa shares in October 2007. We translated
the mid point of the range of US dollar denominated per share fair values (which were
determined as described in the paragraph below) set out in the valuation report into Canadian dollars, and multiplied the result by the number of Visa shares we expected to own at the time of Visa’s initial public offering (IPO). The estimate of the number of Visa shares we expected to own at the time of the IPO was calculated based on the number of
shares we received upon the restructuring in October 2007, adjusted downwards by our estimate of the true-up that we expected would apply to our ownership interests shortly
before the IPO in accordance with the terms of the worldwide restructuring. That estimated true-up was based on an interim notice provided by Visa based on the relative weighting of
Visa’s revenue and expense numbers from the different Visa regions across the world.

    The US dollar denominated range of the estimated per share fair value of the Visa shares set
out in the valuation report was based only on publicly available information at the time.
The valuation methodology involved the estimation of what the fully distributed

5

    trading value of the Visa shares would be after the IPO and the various discounts that
should apply to that value for the purpose of estimating the fair value of the Visa shares
received by the banks in October 2007. The fully distributed trading value was the estimate
of what the market capitalization of Visa would be after the IPO after the elimination of
the “IPO discount” and the assumption of normal trading volumes.

    The estimation of the post-IPO fully distributed trading value of the Visa shares was based
on estimating Visa’s fiscal 2007 and 2008 earnings, and then applying a market observed
price to earnings multiples for similar companies, in particular MasterCard. The estimation
of Visa’s fiscal 2007 and 2008 earnings was based on Visa’s pro forma historical results
disclosed in its public filings, adjusted by various items that the valuator believed should
be taken into account in arriving at a forward-looking earnings estimate.

    The discounts that were subtracted from the fully distributed trading value in respect of
the 51% of the Visa shares expected to be subject to mandatory redemption shortly after the
IPO (at a price based on the IPO price) included factors for the estimated “IPO discount”,
underwriters fees that were expected to be factored into the redemption price, and the risk
that the IPO might not occur. The “IPO discount” refers to the observation that an IPO
typically prices at a discount from what the stock trades at shortly after the IPO.

    The discounts that were subtracted from the fully distributed trading value in respect of
the 49% of the Visa shares expected to be subject to a three year restriction period after
the IPO included factors for a “restricted stock discount” and the risk that the IPO might
not occur. The “restricted s