SecProbe.io

CORRESP
1
filename1.htm

John C. Gerspach

Chief Financial Officer

Citigroup Inc.

399 Park Avenue

New York, NY 10022

January 14, 2011

VIA EDGAR CORRESPONDENCE

Kevin W. Vaughn

Securities and Exchange Commission

100 F Street, NE

Mail Stop 4561

Washington, DC 20549-7553

Re:                             Re: Citigroup Inc.

Form 10-K for Fiscal Year Ended December 31, 2009

Form 10-Q for the Fiscal Period Ended September 30, 2010

File No. 001-09924

Dear Mr. Vaughn:

Enhancement of Citigroup Inc.’s (“Citigroup” or “Citi”) disclosures is an objective that we share with the staff (the “Staff”) of the Securities and Exchange Commission (the “SEC”) and one that we consider in all our filings.  This letter sets forth the responses of Citigroup to the comments of the Staff contained in the Staff’s letter dated December 27, 2010.

For your convenience, the text of the Staff’s comments is set forth in bold below, followed in each case by our response. Please note that all references in this letter to our prior responses or our responses to the Staff’s prior comments refer to Citigroup’s responses to the SEC Staff’s comment letter dated November 4, 2010, which were provided in our response letter dated November 24, 2010 (“Prior Response Letter”).

1

General

1.              Based on previous discussions you had with other members of the commission staff, we understand that it is necessary for you to make certain manual adjustments in order to arrive at more accurate estimates of certain counterparty exposures than the raw outputs of your automated credit risk systems.  It is our understanding that such manual adjustments are not made on a daily or real-time basis.  To the extent that the absolute value of such manual adjustments is significant on a gross basis, these adjustments could be an indication of a limitation in your ability to obtain accurate quantification of credit exposures to certain counterparties within a quick time frame.  Such limitations could negatively affect your ability to have sufficient information to appropriately manage your credit risk with these counterparties in a timely manner. Please consider revising the appropriate section of your future Exchange Act filings to address the following:

a.               Disclose your process for adjusting the outputs of your automated credit risk systems in order to properly estimate your counterparty credit exposure at period end.

b.               Identify how you determine which outputs must be manually adjusted.

c.               Quantify the absolute value of such manual adjustments on a gross basis as of the balance sheet dates presented.

d.               Discuss how the adjusted and unadjusted outputs of your automated credit risk systems are used in managing your credit risk and your exposure to counterparties.

e.               Discuss the extent to which you make similar adjustments on a daily basis.

f.                 To the extent you do not make similar adjustments on a daily basis, disclose why not and discuss limitations on your ability to properly manage your credit exposure to certain counterparties on a timely basis.

g.              Discuss any steps you take to mitigate the lack of accurate outputs from your credit risk systems.

h.              It is our understanding that you believe these adjustments are appropriately made at period end such that the estimates used in your interim and annual financial statement reports are properly presented.  If true, please confirm that understanding in your response as well as your related disclosure.

Citigroup’s credit risk management processes do include making adjustments from time to time to certain outputs from its risk systems; however, these adjustments are primarily to aggregate exposures to clients across multiple products that reside in other systems.  Adjustments may also be required to, for example, reflect recent amendments to collateral, netting and haircut

2

agreements, and input errors.  While Citi continues to streamline its ability to aggregate exposure for and across products through technology enhancements, we do not envision a credit risk practice that would rely solely on automated outputs from our credit systems; rather, such risk measures should always be subject to the continuous review and potential revision by a risk manager.

With respect to the particular discussions referenced in the Staff’s comment above, we note such discussions were in the context of, and focused on, the month-end pre-settlement exposure (PSE) adjustments related to Financial Institutions Risk Management (FIRM) clients within Citi’s broker-dealer entity. Specifically, the discussions focused on May 2010 adjustments, almost all of which related to securities financing transactions (SFT) that are margined.  Because these transactions are fully secured, the overall related risk to Citi is generally small and arises from regular haircuts to market prices. The principal means of managing SFT exposures is through margin calls performed daily by the margin operations department and, therefore, credit systems are not utilized as our primary source of day-to-day risk measurements for this product.  In addition, to the extent there are any breaches of approved counterparty limits, they are escalated to designated credit managers for immediate action and, therefore, are effectively risk managed throughout the month, and not just at month-end.

More broadly speaking, adjustments for any reason are generally recorded at month-end in a centralized reporting system.  The risk analytics department approves all simulation-based exposure adjustments and/or exposure tables provided for exposure calculations for any adjustments, which are further submitted to risk managers for review and approval.  In addition, a month-end portfolio review and mandated sign-off procedure is conducted by all industry risk managers for all outstandings and facilities.  Risk managers receive a “Month-end Adjustment Analysis and Metrics” report which includes all exposure adjustments by amount (pre and post), reason for adjustment, organization and analyst.

In addition, Citi’s overall credit management practices include numerous additional controls designed to monitor counterparty exposures on a more frequent basis than month-end, examples of which include:

·                  Daily exposure reports for the primary regulatory agencies that include exposures for each of our top 20 banks and brokers, non-bank finance companies and corporate names (therefore, a total of 60 counterparties).

3

·                  Risk management analyst teams performing daily monitoring of exception reports to identify any counterparty exposures that exceed credit-approved limits.

·                  As noted above, the margin operations team monitoring daily margin status and reporting disputes and calls not  met to the risk management teams for immediate action. This occurs not just for the SFTs noted above, but for derivatives as well.

·                  Upon request by any risk manager, daily focus positions, which aggregate exposure information across Citi, can be produced for any counterparty not otherwise included in Citi’s top exposure daily reports.

Based on all of the factors and timely controls described above, Citi confirms for the Staff that we believe the adjustments required for aggregation or other purposes are appropriately made at period end (including both month-end and quarter-end) or, if required, more frequently (including on an immediate basis), such that the estimates used in our interim and annual financial statements reports, including our balance sheet, are properly presented.  We do not view these adjustments as a limitation in our ability to obtain accurate quantification of credit exposures to counterparties, and thus we do not believe these processes require additional disclosure in our Exchange Act reports.

Form 10-Q for the period ended September 30, 2010

U.S. Consumer Mortgage Lending, page 52

2.              We note your disclosure that “the growing amount of foreclosures in process” could inflate the amount of 180+ day delinquencies, could result  in an increase in non-accrual loans, and/or could have a dampening effect on net interest margin as non-accrual assets build on the balance sheet.  Please tell us and revise your future filings to also disclose the potential effects of foreclosure delays on servicing advances, including the collectability of such advances and the value of MSRs, as well as the potential effects on the timing of sales of loans, other real estate owned and mortgage backed securities and any related effects on the valuation and impairment of these assets.

As disclosed on page 52 of our third quarter Form 10-Q (“3Q Form 10-Q”), as was further supplemented by the disclosure included in our filing on Form 8-K, filed with the SEC on November 17, 2010, due to the increased attention to the foreclosure process on an industry-wide basis, Citi intensified the ongoing review of its foreclosure processes with respect to its U.S. mortgage portfolios.  As the Staff is aware and as we have disclosed, to date, Citi has not

4

identified systemic deficiencies in its existing foreclosure processes.  However, as we disclose on page 52 of the 3Q Form 10-Q, “…Citi’s review of its existing and historical processes continues and, depending on the results of that review, or if any industry-wide adverse regulatory or judicial actions are taken in respect of foreclosures, Citi’s ability to continue to carry out its current foreclosure processes, and its financial results of operations and financial condition, could be adversely affected.”

Because Citi’s review of its own processes continues, and because there continues to be extensive judicial, regulatory and media focus on foreclosure process issues, in response to the Staff’s comment, Citi intends to enhance the disclosure referenced by the Staff, beginning with our 2010 Annual Report on Form 10-K (“2010 Form 10-K”) as follows:

“Any increase or backlog in the number of foreclosures in process, whether related to Citi’s foreclosure process issues or industry-wide efforts to prevent or forestall foreclosure, has broader implications for Citigroup’s U.S. consumer mortgage portfolios.  Specifically, because Citigroup is unable to take possession of the underlying assets and sell the properties on a timely basis, growth in the number of foreclosures in process:

·                  inflates the amount of 180+ day delinquencies in Citigroup’s mortgage statistics;

·                  increases consumer non-accrual loans (90+ day delinquencies);

·                  creates a dampening effect on Citi’s net interest margin as non-accrual assets build on the balance sheet;

·                  negatively impacts the amounts ultimately realized for property subject to foreclosure; and

·                  causes additional costs to be incurred in connection with legislative or regulatory investigations.

Further, any increase in the time to complete foreclosure sales may result in an increase in servicing advances and may negatively impact the value of Citigroup’s MSRs and mortgage-backed securities, in each case due to an adverse change in the expected timing and amount of cash flows to be received.”

We supplementally advise the Staff that Citi does not currently believe an increase in the amount of foreclosures in process would impact the collectability of servicing advances due to the relatively low delinquency rates in our securitizations and the delinquency and cross-

5

collateralization provisions of our agreements, which protect us with regard to advances made to investors.  In addition, Citi does not believe there would be an impact on the timing of sales of loans, because substantially all our loan sales are from new originations. Citi appropriately considers any potential impacts of foreclosure delays on the valuation and impairment of loans and other real estate owned in its periodic loan loss reserve and valuation processes.

Representations and Warranties, page 69

3.              We note your response to comment 5 from our letter dated November 4, 2010, as well as your disclosure on page 71 that you are not able to estimate a range of reasonably possible loss as defined by ASC 450 related to indemnifications for breaches of representations and warranties, and we remain unclear why you believe you are in full compliance with the above referenced guidance. Given the large realized loss and provision for repurchase losses recorded in the third quarter of 2010 and in previous periods, it is apparent that these indemnifications matters are significant and that there could be exposure to loss in excess of amounts accrued.  Thus, we continue to believe a more transparent discussion of the reasonably possible loss or range of possible loss is required so that an investor may understand your liability exposure.  We believe that where a range of reasonably possible loss is estimable, and the top of the range is in excess of the amount accrued, which appears to be the case given the addition of significant probable incurred losses each period relating to loans sold in historical periods, the range should be disclosed pursuant to ASC 450-20-50-3.  Please tell us and revise your disclosure in future filings to provide the range of reasonably possibly losses for all counterparties for which this disclosure is possible.  These disclosures should be updated quarterly as additional information is obtained.

As noted in our Prior Response Letter, with respect to Citi’s representation and warranty exposure as it relates to GSEs and the prior sales of whole loans to private investors in our U.S. consumer mortgage lending business, Citi intends to provide a sensitivity analysis, beginning in our 2010 Form 10-K, to illustrate the sensitivity that the repurchase reserve has to changes in the key assumptions utilized in determining the reserve.  For Citi’s representation and warranty exposure relating to our private label securitizations, see our response to Staff comment #5 below.

4.              We note your disclosure on page 70 that a significant assumption into the repurchase reserve is the “claims appeal success rate,” which is your expected success at rescinding an investor claim by satisfying the investor demand for more information, disputing the

6

claim validity, etc.  Please tell us in more detail how the appeals process with the investor typically works, and how it may vary by type of investor.  For example, tell us the types of information you would typically provide in rejecting the original request for repurchase, how long the investor has to review and appeal your original rejection, and the typical results in any appeal of your original rejection.  Please tell us whether any changes in the original rejection conclusion are built into your “claims appeal success rate.”  Please also tell us how your original rejection impacts the timing and amount of the repurchase reserve, such as whether the reserve is immediately reduced after your original rejection, or whether the repurchase reserve is adjusted after any appeals process by the investor is complete.

Citi supplementally advises the Staff that the claims appeals process is substantially the same for each type of investor (e.g., GSE, private investor, etc.). Upon the receipt of a repurchase claim from any such investor, a repurchase coordinator within Citi’s U.S. consumer mortgage business will review the nature of the claim and work with various other applicable units within Citi to resolve the claim.  The types of information provided by Citi in rejecting the repurchase request will necessarily vary based on the particular violation asserted.  For example, if the investor is asserting that adequate title is not in the loan package, Citi will attempt to obtain such title and submit