CORRESP 1 filename1.htm 11-7-2012 SEC Comment Letter (Redaction)
Confidential Treatment
Requested by JPMorgan Chase & Co.
Page 1


December 3, 2012


Ms. Suzanne Hayes
Assistant Director
Division of Corporation Finance
United States Securities and Exchange Commission
100 F Street, N.E.
Washington, D.C. 20549

Re:     JPMorgan Chase & Co.
Form 10-K for Fiscal Year Ended December 31, 2011
Filed February 29, 2012
Form 10-Q for Fiscal Quarter Ended March 31, 2012
Filed May 10, 2012
Forms 8-Ks Filed July 13, 2012
Form 10-Q for Fiscal Quarter Ended June 30, 2012
Filed August 9, 2012
File No. 001-05805

Dear Ms. Hayes:

We are in receipt of the letter dated November 7, 2012 to Douglas L. Braunstein, Executive Vice President and Chief Financial Officer of JPMorgan Chase & Co. (“JPMorgan Chase” or the “Firm”), from the staff (the “Staff”) of the Securities and Exchange Commission (the “Commission”), regarding the above-referenced filings.

Certain confidential portions of this letter were omitted by means of redacting a portion of the text. The word “[Redacted]” has been inserted in place of the portions so omitted. Pursuant to the provisions of 17 C.F.R. §200.83, we have separately submitted a copy of this letter containing the redacted portions of this letter with the Staff and have requested confidential treatment for the redacted portions of this letter.

To assist in your review of our responses to the comments set forth in the Staff's letter, we have set forth below in full the comments contained in the letter, together with our responses.


Form 10-K for the Fiscal Year Ended December 31, 2011
Item 9A: Controls and Procedures, page 20
1.We note your response to prior comment two of our letter dated July 25, 2012 where you indicated that there was a material difference in the size and characteristics of the synthetic credit portfolio during the first quarter of 2012. As part of your analysis of the overall effectiveness of the design of the controls, including your evaluation of the severity of the control deficiencies identified, please tell us whether you considered both the potential size and complexity


Confidential Treatment
Requested by JPMorgan Chase & Co.
Page 2

and the related volume of activity that reasonably could have occurred in 2011 and the activity that was expected in future periods. For example, refer to the section of the Commission's Guidance Regarding Management's Report on Internal Control Over Financial Reporting (http://www.sec.gov/rules/interp/2007/33-8810.pdf) related to evaluation of control deficiencies that indicates that you should consider the volume of activity in the class of transactions exposed to the deficiency that has occurred in the period, or that is expected in future periods. In this regard, we note that it would appear that this volume of activity, and the likelihood that it could have occurred as of December 31, 2011, or was expected to occur in future periods, should be considered in evaluating whether the design of your controls was effective as of December 31, 2011.

The Firm assesses the effectiveness of its internal controls over financial reporting (“ICFRs”) based upon the COSO framework. The Firm's process for assessing its ICFR is a “top-down” risk-based approach, which is designed to identify key financial reporting controls that, if not designed appropriately or if not operating effectively, may result in a material misstatement of the Firm's financial statements.
As part of the Firm's assessment of the overall effectiveness of the design of its internal controls over financial reporting, the Firm considered the potential size, complexity and volume of activity that reasonably could have occurred in CIO at year-end 2011 and that was expected to occur in future periods. The Firm concluded that the design of the CIO valuation control process was appropriate at December 31, 2011 and March 31, 2012, but that a material weakness existed in the operating effectiveness of the VCG process as of March 31, 2012. Further, the Firm concluded that the significance and nature of the changes in CIO's synthetic credit portfolio during the first quarter of 2012 could not have been reasonably expected in December 2011.
[Redacted]


2.     Additionally, please tell us whether you have identified any deficiencies as of
December 31, 2011, March 31, 2012, or June 30, 2012 in the risk assessment component of your controls related to timeliness of identifying and addressing new or changed financial reporting risks. Please also describe how you evaluated the severity of any such deficiencies.

The Firm did not identify deficiencies in the risk assessment component of the Firm's control framework regarding financial reporting risks at December 31, 2011, March 31, 2012 or June 30, 2012.
As noted in the Firm's response to the Staff's comment 1, although the Firm concluded that the design of the CIO valuation control process was effective, including as of March 31, 2012, the Firm did identify a material weakness in the operating effectiveness of the CIO valuation control process at March 31, 2012. The Firm does not view the material weakness in the operating effectiveness as a risk assessment issue. Rather, the Firm considers the finance supervision and engagement deficiency it noted to be an integral part and a key component of the CIO valuation control activity itself.
[Redacted]
   




Confidential Treatment
Requested by JPMorgan Chase & Co.
Page 3

Form 10-Q for Fiscal Quarter Ended March 31, 2012
Market Risk Management, page 73
Value-at-Risk, page 73

3.     We note your response to the sixth bullet point of prior comment 10 of our letter dated July 25, 2012 regarding the need for disclosure of certain model, assumptions, and parameter changes to your value-at-risk (VaR) models. Please clearly explain to us how you considered your disclosure obligations pursuant to Item 305(a)(4) of Regulation S-K for the VaR model change that occurred within CIO during the first quarter of 2012, and how you evaluated materiality for purposes of your conclusions in this instance. For example, please specifically discuss any quantitative or qualitative metrics you consider in assessing VaR model changes, and describe whether these factors change depending on the type of VaR model change being made. Also, as part of your response, please tell us whether you would consider any mathematical model change related to the calculation of market or credit risk that is required to be submitted to the Division of Trading and Markets for approval pursuant to Rule 15c3-1e(a)(8) as material for disclosure under Item 305(a)(4), if approval is ultimately granted. If not, please tell us the types of circumstances where you believe disclosure under Item 305(a)(4) would not be required.

Item 305(a)(4) of Regulation S-K requires disclosure in a registrant's Form 10-K or 10-Q regarding changes in a Registrant's VaR model if there have been changes in “key model characteristics, assumptions and parameters used in providing quantitative information about market risk” and “if the effects of any such change is material [emphasis added]".

The Firm monitors changes in its VaR models on a monthly basis for the Investment Bank and on a quarterly basis for its others lines of business (LOB) to assess whether any such changes would require disclosure under Item 305(a)(4). With respect to the first quarter of 2012, the VaR models that were used to report VaR in the Firm's Quarterly Report on Form 10-Q for the quarter ended March 31, 2012 (“2012 First Quarter 10-Q”) were the same models as those used to report VaR in prior quarters, including in the Firm's 2011 Annual Report on Form 10-K and the Quarterly Report on Form 10-Q for the quarter ended March 31, 2011. As was noted in footnote (c) on p. 73 of the 2012 First Quarter Form 10-Q, the methodology used to calculate CIO's VaR in the first quarter 2012 was consistent with the methodology used to calculate CIO's VaR in 2011 (as reported in the relevant quarterly and annual reports).1 Accordingly, although there was an interim change in the VaR model during the first quarter, that change had been reversed by the time the 2012 First Quarter 10-Q was filed.  As a result, the Firm believes there was no model change within the meaning of Item 305(a)(4), which applies to VaR as reported in annual and quarterly reports, and, in any event, there could not have been a “material” change in the context of the VaR as reported in the 2012 First Quarter 10-Q.

[Redacted]


Note 3 – Fair value measurement, page 91
4.     We note your response to comments 11 and 12 of our letter dated July 25, 2012 and have the following questions:



Confidential Treatment
Requested by JPMorgan Chase & Co.
Page 4

We understand from your disclosure in your third-quarter Form 10-Q that the CIO valuation control group (VCG) typically establishes price testing thresholds. It is not clear how the acceptable range is defined, especially for instruments with a wide bid-offer range. Further, tell us how the VCG ensures that the range is sensitive enough that it is actually providing independent price verification. As part of your response, please clarify the process for reviewing the thresholds, including to whom the results of the analysis are reported, the factors that drive the thresholds, and how often the thresholds have changed over time.

With the exception of the most liquid exchange traded instruments, there is generally not one single estimate of fair value for a particular instrument to which all market participants would unanimously agree. This is because market transactions occur in the normal course of business throughout a range (often the bid-ask spread); this market activity creates a range of reasonable fair values for any particular instrument. The Firm must apply judgment within such a range to identify an appropriate specific estimation of fair value. The range of reasonable values is narrow (but not non-existent) for the most liquid instruments and generally increases as liquidity decreases. The objective of pricing thresholds is to approximate the reasonable range of fair values for a particular instrument so that VCG can identify when a variance between a front office mark and external pricing information may not be appropriate because it is outside the range of reasonable fair values.

[Redacted]

In your response to both comments 11 and 12, you have noted that the VCG will make valuation adjustments for various factors including the size of the net open risk position. Tell us which specific portfolio or positions you factor in the net open risk position.

JPMorgan Chase has elected to apply the portfolio exception to its market making derivative portfolios and related cash instruments within the IB line of business.

You have noted in this response as well in response to our comment 21 that the VCG is independent of the trading function. However, you have also noted that the respective VCG groups are located within both the CIO and Investment Bank (IB). Please discuss and disclose in future filings the structure of the VCG groups within the various business lines, including its reporting structure in each line.

The Firm historically maintained a VCG in both the IB and CIO. In both cases VCG was part of the respective IB/CIO Finance functions and reported directly to the respective IB/CIO Controller. The Controllers report to the respective IB/CIO CFOs, who each have dual reporting to the CFO of the Firm as well as to the IB Chief Executive Officer and the Firm's Chief Investment Officer, respectively.

In October 2012 the Firm combined the existing VCGs in CIO and IB (including activities covering Mortgage Banking) into a single VCG. The combined VCG continues to be part of the Firm's Finance function and is managed and evaluated accordingly. In addition, the Firm established a firm-wide Valuation Governance Forum (VGF) comprising senior finance and risk executives to oversee the management of risks arising from valuation activities across the Firm. The firm-wide VGF is chaired by the firm-wide head of VCG, and also includes sub-forums for the Investment Bank, Mortgage Bank, and Corporate functions, including CIO.



Confidential Treatment
Requested by JPMorgan Chase & Co.
Page 5

The Firm will provide an expanded description of reporting lines of the combined VCG function in its 2012 Form 10-K.

Separately for each CIO and IB, please explain the process when the front office's mark is outside of the VCG's acceptable range and identify the parties responsible for making the final determination of the mark to be used. Please clarify how and whether this information is communicated to the trader that posted the marks.

[Redacted]

The results of the independent price testing process are reported to individual traders on a monthly basis, either in person or via email. This reporting includes all variances between front office marks and VCG independent marks and any resulting adjustments made by VCG.


5.     We note your response to comment 13 of our letter dated 25, 2012, and have the following questions:

Your response indicates that you have applied liquidity adjustments to certain financial instruments. Please tell us amounts for your entire portfolio where you have made adjustments that are greater than 10% of the price or spread yet are still in Level 2.

[Redacted]

Please reconcile the statement that there was enough market activity for these instruments to be considered Level 2 with your other statements about holding positions in indices that were not as liquid. Specifically address the extent to which you consider your own positions and your own transactions when assessing the level of market activity in a particular index.

When determining the classification of instruments within the fair value measurement hierarchy the Firm evaluates the type and source of market information available as well as the level of market activity. The classification within the fair value hierarchy is made by considering available market information for a particular instrument without regard to entity-specific information, such as the size of position held by the Firm.

[Redacted]


Note 24 – Business Segments, page 163
6.     We note your response to comment 19 from our letter dated July 25, 2012 regarding how the carrying values of your reporting units are determined for purposes of performing your goodwill impairment analysis. We are continuing to evaluate your response, and we may have further comment in the future.
 




Confidential Treatment
Requested by JPMorgan Chase & Co.
Page 6

Form 8-K filed July 13, 2012
Item 4.02(a) Non-Reliance on Previously Issued Financial Statements or a Related Audit Report or Completed Interim Review, page 2
 
7.     We note your response to comments 21 and 23 from our letter dated July 25, 2012 and have the following comments:

We understand that based on your review and analysis of emails, voice tapes, and interviews suggest trader intent to mark positions advantageously. It is not clear from your response exactly how this advantageous marking was being done. Please further describe how these positions were being marked advantageously.

[Redacted]

Tell us the extent to which the traders were marking long positions and short positions incorrectly, but still within the bid-offer spread and within the VCG established thresholds.

When the Firm made its decision to restate its first quarter financials, the Firm determined to restate the positions in the entire synthetic credit portfolio to an objectively determined benchmark “mid” valuation, notwithstanding that the prior valuations were within the VCG established thresholds. As noted in the Firm's response to the Staff's comment 4, the Firm does not believe that there is one single estimate of fair value for the positions in the synthetic credit portfolio to which all market participants would unanimously agree. However, the evidence uncovered during the management Task Force review raised questions about whether trader's marks represented unbiased assessments of exit prices for the positions in the CIO synthetic credit portfolio and therefore the Firm reached a determination to restate its financial statements to reflect an objectively determined benchmark “mid” valuation for all the positions in the portfolio. Given the inherent difficulty of (i) isolating the traders' bias to specific individual positions, and (ii) conducting a position-by-position re-evaluation of the front office marks without hindsight bias, the Firm determined that remarking all of the positions in the synthetic credit portfolio to an objectively determined "mid" valuation was reasonable under the circumstances. For the entire synthetic credit portfolio there was an aggregate $660 million (pre-tax) difference between the traders' marks on the portfolio as of March 31, 2012, taking into account the previously included March 31, 2012 threshold adjustments, and the benchmark “mid” price recalculated by the Firm.
As part of the VCG's verification process, tell us whether they consider recent transactions at period end to establish their thresholds, or verify if the trader marks within the bid-offer spread represent a reasonable basis for an exit price.

VCG utilizes various pricing sources including recent transactions, dealer quotes, and consensus pricing to verify front office marks. The specific data used in the price verification process varies based on the judgment of VCG.

As discussed above in the Firm's response to the Staff's comment 4, the OTC derivatives comprising the synthetic credit portfolio are transacted, unlike exchange-traded derivatives, on a counterparty to counterparty basis, and therefore, there is no single “price” for an instrument to which all market participants would unanimously agree, since market transactions occur in the normal course of business within a range, and each dealer will individually quote what it determines to be a fair market price at


Confidential Treatment
Requested by JPMorgan Chase & Co.
Page 7

which it is willing to transact. The thresholds established by VCG are intended to represent a reasonable range of an exit price for a position, and are determined using market data that is reflective of recent market activity. Accordingly, VCG would generally, absent evidence to the contrary (such as the voice recordings and electronic correspondence uncovered by the Firm's management Task Force review) accept a front office mark if such front office mark is within the established threshold determined by VCG.

You note in your response that VCG makes two types of adjustments to the front office markets [sic]. Please tell us if these adjustments were made to the positions in question (where the traders had marked the positions advantageously), and if so, whether the VCG group noted any issues with the marks. Please specifically explain whether, and if so, how the VCG made adjustments to these positions without recognizing that they were being marked differently or more advantageously.

[Redacted]

Please tell us whether VCG or anyone in management inquired further about any adjustments, and whether there were any controls in place that would have required them to do so.

[Redacted]

Please quantify the positions that were marked advantageously by the traders and any VCG adjustments that were made to these positions both prior to and after the restatement (separate from the correction of the error).
 
As noted above, in light of the evidence uncovered during the management Task Force review, which raised questions as to whether the trader's marks represented unbiased assessments of exit prices for the positions in the CIO synthetic credit portfolio, the Firm reached a determination to restate its financial statements to reflect an objectively determined benchmark "mid" valuation for all positions in the portfolio (rather than isolating specific positions).

[Redacted]

We understand that CIO's valuation control environment was not appropriately enhanced in light of the significant changes to the size and characteristics of the portfolio and also that you note in Item 4 of your June 30, 2012 Form 10-Q that you have instituted more formal controls around establishment and monitoring of your price testing thresholds. However, in your response you have indicated that you believe the controls over valuation were designed appropriately. Please tell us how you considered that it appears that the price testing thresholds were not sensitive enough to ensure that the trader's marks were appropriate, which would appear to be a weakness in the design of the control.

[Redacted]

Further, tell us how you considered that it appears that the controls were not sensitive enough to detect that the portfolio had grown in both size and complexity during the quarter ended March 31, 2012, such that additional measures were warranted.


Confidential Treatment
Requested by JPMorgan Chase & Co.
Page 8


[Redacted]


8.     We note your response to comment 22 from our letter dated July 25, 2012 and have the following questions:

You have told us in response to comment 11 that VCG verifies front office marks by evaluating independently derived market data. In response to comment 13, you indicated that the fair values are verified using readily available prices derived from independent third party data providers and quoted market prices from dealers, which represent observable inputs. You further indicated that based on this information you concluded these positions represented Level 2 inputs. This conclusion appears to be inconsistent with your disclosure on page 3 of Item 4.02(a) of your Form 8-K that you did not have independent third party data to run your VaR model for certain positions and therefore had to rely on trader marks. Please explain.

[Redacted]

Please quantify the positions where you used trader marks to calculate your VaR, and further explain the reasons for the lack of availability of information (i.e. illiquidity of positions etc.).

[Redacted]

Please disclose in future filings the fact that the data used to calculate your VaR is not the same as what is used in your financial statements, and explain the reasons for the difference.
    
In future filings, the Firm will provide an enhanced description of the data sources used in its VaR calculations and why such information may differ from valuations used for financial statement purposes.


Form 8-K filed July 13, 2012
Exhibit 99.1 – Earnings Release – Second Quarter 2012 Results
Retail Financial Services, page 6

9.     We note your response to comment 24 regarding the $1.4 billion reduction in your Retail Financial Services allowance for loan losses. In your response, you note that you adjust your statistical calculation to take into consideration model imprecision, external factors and current economic events that have occurred but that are not yet reflected in the factors used to derive the statistical calculation, as further discussed on pages 168 and 169 of your 2011 Form 10-K. In order to provide additional transparency in future filings, please address the following for each of your respective portfolio segments:

Given the significant judgment management applies in making the adjustment to the statistical calculation, as discussed on page 169 of your 2011 Form 10-K, please tell us and consider disclosing in future filings the amount of this component of the allowance for loan losses.



Confidential Treatment
Requested by JPMorgan Chase & Co.
Page 9

The Firm reports three portfolio segments in its allowance for loan losses (“Allowance”): (i) wholesale (ii) consumer, excluding card, and (iii) credit card. Also, as addressed on pages 168-169 and 252-253 of the 2011 Form 10-K, the Firm's allowance for loan losses includes three separate and distinct components: (a) an asset-specific component, which addresses loans considered to be impaired, including those modified in troubled debt restructurings ("TDRs"); (b) an allowance for losses on purchased credit-impaired loans; and (c) a formula-based allowance, which addresses all other loans in the retained loan portfolio. The formula-based component of the Allowance consists of a base statistical loss estimate and an adjustment to the base statistical loss estimate (the “Adjustment”).

[Redacted]

Given the factors disclosed on page 169 of your 2011 Form 10-K that drive this component of the allowance, please provide disclosure describing how these factors have directionally impacted this component of the allowance over the past two years.

In future filings, the Firm will enhance its description of the factors that are considered in estimating losses in the portfolio, including factors driving the Adjustment and how the Adjustment is directionally impacted.
[Redacted]

Tell us whether you have made any changes to the historical loss periods used for any of your portfolio segments during the past two years. If so, please describe the changes and discuss how these changes affected the qualitative component of the allowance discussed above.

The Firm does not base its allowance for loan losses upon the levels of losses realized in specific historical periods. Rather, the base statistical loss estimate is based on current loss estimates, which consider probability of default (PD) and loss given default (LGD).

For risk-rated loans, the PD and LGD factors are differentiated by the risk rating and the expected maturity of the underlying loan. The Firm has consistently used PD estimates based on the longest period available of historical default observations (nearly 30 years), using credit rating agency default statistics. LGD estimates are primarily based on the nature and structure of the credit facility, the type and amount of collateral provided by the borrower, and other relevant factors. These estimates are periodically validated using historical default observations to refine the Firm's process of assigning LGDs to specific facilities. The Firm's loss estimate for risk-rated loans is determined by applying the PD and LGD factors to the current risk rating of each loan facility. Accordingly, the risk ratings assigned to each facility is the primary means by which the Firm ensures that loss estimates reflect management's current assessment.

For scored loans, the Firm uses a number of forecasting models (e.g., portfolio roll-rate models, loan-level loss forecasting models) to develop its product-level base statistical loss estimates. These models consider factors such as collateral values (e.g., home prices) and employment data to project future delinquencies and probabilities of default of loans in the portfolio. For collateralized loans, the Firm separately develops product-level loss severity assumptions using the same sources of collateral value information. Based upon these estimated probability of default and loss severity assumptions, the Firm then calculates expected loss factors for individual pools of loans with similar risk characteristics


Confidential Treatment
Requested by JPMorgan Chase & Co.
Page 10

(essentially forecasts of expected charge-offs). The product-level base statistical loss estimate is ultimately computed by multiplying period-end loan balances by the expected loss factors over the loss emergence period. The Firm has defined its loss emergence period to represent the time period between the date at which the loss is estimated to have been incurred and the ultimate realization of that loss (through a charge-off).


Form 10-Q for the Quarter Ended June 30, 2012
Management’s Discussion and Analysis, page 3
Selected European exposure, page 104
10. We note your response to comment six of our letter dated July 25, 2012, as well as your disclosure in your quarterly filing and have the following questions:

We note your disclosure that the “Trading” column includes approximately $9.0 billion of net purchased protection, including single name, index and tranched credit derivatives that are part of the Firm’s market making activities as well as the synthetic credit portfolio held by CIO. The majority of the $9.0 billion relates to the synthetic credit portfolio held by the CIO. Please tell us why these amounts are included in the “Trading” column and not the “Portfolio Hedging” column as these amounts are not part of your market making and appear to serve the same function as the instruments in your “Portfolio Hedging” column.

The "Portfolio Hedging" column represents net protection purchased through the Firm's credit portfolio management activity, which is generally intended to mitigate the credit risk associated with specific lending and derivatives counterparty exposure; this activity generally uses single-name credit derivatives. The synthetic credit portfolio held by CIO was originally intended to protect the Firm more generally in a stressed credit environment, rather than on a name-specific basis. In light of this difference, as well as the substantial changes that occurred in the synthetic credit portfolio in the first quarter of 2012, and the active and ongoing reductions in the synthetic credit portfolio, the Firm believes that presentation of this portfolio separate from "Portfolio Hedging" activities is appropriate and is consistent with internal risk reporting. To provide transparency, the Firm has disclosed this distinction, as well as the aggregate amount of net purchased protection provided by the synthetic credit portfolio in the five named countries.

We also note your disclosure in footnote “d” to your table on page 104 that beginning on March 31, 2012, the Firm’s country risk reporting reflects enhanced measurement of tranched credit derivative products. It is not clear what “enhanced measurement” actually means in this context. Please tell us and disclose in future filings the nature of such enhancements and the reasons for including its effects in the table.

Beginning in the first quarter of 2012, the Firm's country risk reporting was enhanced to reflect all tranched credit derivative products using a full revaluation approach. [Redacted] In future filings, the Firm will enhance its disclosure to more clearly describe the measurement of tranched credit derivative products in its country risk reporting.

We also note in your response that both your trading and portfolio hedge columns include index credit derivatives. You also indicate that the table does not contemplate any potential


Confidential Treatment
Requested by JPMorgan Chase & Co.
Page 11

ineffectiveness of your hedges. In your future filings, in addition to stating that your effectiveness may vary based on a number of factors, please more clearly discuss how the indices that you use as hedges relate to your exposures to Greece, Italy, Ireland, Portugal, and Spain, and describe specific risks (i.e. correlation) that may the cause the hedge to not be effective. For transparency purposes, please consider disclosing the historical correlation between the indices and your European exposure.

In future filings, the Firm will enhance its description of the potential ineffectiveness of the Firm's CDS protection as a hedge of the Firm's exposures. This ineffectiveness may vary based on a number of factors, including the maturity of the Firm's CDS protection (which in some cases may be shorter than the Firm's exposures), the named reference entity (i.e. in a particular country, the Firm may experience losses on specific exposures that are different than the named reference entities in the CDS), and the contractual terms of CDS (which may have a defined credit event that does not align with an actual loss realized by the Firm).

The Firm's country risk disclosures are based on the Firm's risk to an immediate default of all counterparties or obligors in a specific country, with zero recovery. These amounts therefore consider the effect from net purchased or sold protection of credit derivatives that would be realized upon such a simultaneous default. Accordingly, for index credit derivatives (which provides a pro rata amount of protection against the default of each of the named reference entities included in the index), the Firm's country risk reporting considers the amount of protection related to any reference entities (in the relevant country) that are part of an index credit derivative. For example, if the Firm has written protection with $100 notional (and zero current fair value), and the index has 10 names, the credit derivative would provide $10 of protection with respect to each of the named reference entities; if two of such reference entities were located in Italy, the Firm's country risk reporting would report sold protection of $20 in Italy. The Firm's country risk reporting is not intended to measure the Firm's risk to an actual individual counterparty or obligor default (which may or may not be offset by net purchased protection through a single-name, index or tranched credit derivative) or to the default of some but not all obligors in a particular country. Rather, the Firm's country risk reporting is intended to report the Firm's risk to a broad default of all exposures in a particular country. Therefore the correlation between the performance of an index as a whole and any of its underlying exposures individually is not measured in the Firm's country risk reporting framework.



****************


This is to acknowledge that (i) the Firm is responsible for the adequacy and accuracy of the disclosure in its filings; (ii) Staff comments or changes to disclosure in response to Staff comments do not foreclose the Commission from taking any action with respect to the filings; and (iii) the Firm 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.






Confidential Treatment
Requested by JPMorgan Chase & Co.
Page 12

If you have any questions or request any further information, please do not hesitate to contact me.

Very truly yours,


/s/ Douglas L. Braunstein
Douglas L. Braunstein
Chief Financial Officer