CORRESP 1 filename1.htm

 

FOIA Confidential Treatment Requested by Citigroup Inc.

 

GRAPHIC

 

John C. Gerspach

Chief Financial Officer

Citigroup Inc.

399 Park Avenue

New York, NY 10022

 

November 18, 2011

 

VIA EDGAR CORRESPONDENCE

Stephanie Hunsaker

Securities and Exchange Commission

100 F Street, NE

Mail Stop 4561

Washington, DC 20549-7553

 

Re:

 

Citigroup Inc.

 

 

Form 10-K for Fiscal Year Ended December 31, 2010

 

 

Filed February 25, 2011

 

 

Form 10-Q for Fiscal Quarter Ended March 31, 2011

 

 

Filed May 5, 2011

 

 

Form 10-Q for Fiscal Quarter Ended June 30, 2011

 

 

Filed August 5, 2011

 

 

File No. 001-09924

 

Dear Ms. Hunsaker:

 

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” or “Commission”) 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 October 26, 2011.

 

1



 

FOIA Confidential Treatment Requested by Citigroup Inc.

 

For your convenience, the text of the Staff’s comments is set forth in bold below, followed in each case by our response.

 

General

 

1.              Please file a legality opinion for your Form S-8 (File No. 333-173683) that does not contain a statement regarding the General Counsel — Human Resources not being admitted to practice law in Delaware.

 

Citi notes for the Staff that on November 7, 2011, a legality opinion was filed with Post-Effective Amendment No. 1 to our Form S-8 (File No. 333-173683) that does not contain a statement regarding the General Counsel Human Resources not being admitted to practice law in Delaware.

 

Form 10-Q for the Quarterly Period ended March 31, 2011

Note 23. Contingencies, page 171

 

2.              We note your response to prior comment 13. Please confirm that in future filings that discuss the Consent Orders you will include an assessment of the impact that the undertakings set forth in the Consent Orders will have on Citi.  If at the time of such future filings you continue to believe that the annual cost of compliance with the Consent Orders will not have a material impact on the results of operations or financial condition of your consumer mortgage business or Citigroup, please disclose such belief.

 

Citi confirms for the Staff that, to the extent we discuss the Consent Orders in our future filings, we will include an assessment of the impact of the undertakings set forth in the Consent Orders on Citi; specifically, whether or not the annual cost of compliance with the Consent Orders will have a material impact on the results of operations or financial condition of Citi or its consumer mortgage business.

 

Form 10-Q for the Quarterly Period ended June 30, 2011

General

 

3.              News articles reported that in May, Citigroup was the target of a data breach that affected over 360,000 credit card accounts.  We note that your quarterly report for

 

2



 

FOIA Confidential Treatment Requested by Citigroup Inc.

 

the period ended June 30, 2011 did not address the data breach or its impact on the company.  Please provide us with an analysis supporting your apparent conclusion that no additional disclosure was needed in your SEC filings.  Please address in your analysis the materiality of the data breach, both from a financial standpoint and from a reputational risk standpoint, and whether you intend to provide disclosure in future filings.

 

As noted by the Staff, on May 10, 2011, Citi identified a high volume of attempts to access credit card accounts in its “Account Online” system (this is a system that enables customers to manage their credit card accounts, e.g., pay bills, view statements and check due dates).  It was determined that a data breach had occurred which resulted in unauthorized access to 360,083 Citi credit card account numbers.

 

Consistent with Citi’s policies and procedures in such instances, systemic alerts were placed on all compromised accounts.  These alerts monitor suspicious activity and queue the account for a fraud risk review if suspicious authorizations are detected.  Customers whose accounts were a part of the breach were notified and, after a risk analysis, the majority of the affected accounts were closed and new cards were reissued; 89% of the reissued cards were activated.

 

By the close of the second quarter 2011, all compromised accounts were identified and substantially all fraud losses were determined.  The 360,083 compromised accounts represented 0.33% of the total 107.5 million North America consumer credit card accounts (both Citi-branded and retail partner cards) at April 30, 2011.  Unauthorized charges occurred on approximately 1.5% of the 360,083 accounts, resulting in an estimated $4.4 million in fraud losses related to this incident.  Such amount represented approximately [***] of Citi’s estimated 2011 annual North America consumer credit card fraud expense.  Operational remediation costs (such as customer notification, service calls and providing identity theft protection) were less than $1 million.

 

Thus, from a financial standpoint, the data breach did not have a material impact on Citi’s North America credit card businesses.  With regard to reputational risk, Citi believes that the high activation rate of the reissued credit cards clearly indicates that no meaningful reputational damage or loss of client confidence occurred, and there were no investor questions relating to the incident.  Accordingly, for these reasons, Citi did not deem the incident to warrant disclosure in its second quarter 2011 Form 10-Q and Citi does not intend to provide disclosure on this particular matter in future filings.

 


[***] The following information has been redacted in accordance with Citigroup’s request for confidential treatment.

 

3



 

FOIA Confidential Treatment Requested by Citigroup Inc.

 

As with other operational and financial risks, Citi will continue to monitor data breach and other cybersecurity risks and incidents and evaluate the need for disclosure based on the respective quantitative and qualitative factors, including those set forth in the Staff’s guidance on cybersecurity matters issued on October 13, 2011.

 

North America Consumer Mortgage Lending, page 50

 

4.              We note your disclosure on page 52 related to foreclosures in process and the impact these have had on your lending portfolios and operations.  Specifically, we note that the lengthening of the foreclosure process has inflated the amount of 180+ day delinquencies in mortgage statistics and 90+ day delinquencies in consumer non-accrual loans statistics.  We also note that the lengthening of the foreclosure process has caused a “build” in nonaccrual assets on the balance sheet.  To the extent reasonably possible, please tell us and revise future filings to provide more quantitative information about the impact of foreclosures in process on your lending statistics and non-accrual loans and non-accrual assets.  Please also consider including a footnote to your related delinquencies and nonaccrual loans and non-accrual assets tabular disclosure that addresses the impact of foreclosures in process to each relevant statistic and/or balance, as appropriate and discuss whether you have observed any geographic trends.

 

We note for the Staff that, on page 54 of our third quarter 2011 Form 10-Q, similar to the several prior quarters, we continued to include the following disclosure on the implications of the lengthening of the foreclosure process, including the fact that the lengthening of the foreclosure process is particularly pronounced in the judicial states:

 

“In addition, as previously disclosed, while Citi’s active foreclosures in process continued to decrease during the third quarter of 2011, foreclosures in process for two years or more continued to increase, particularly in judicial states (i.e., those states that require foreclosures to be processed via court approval).  Combined with continued pressure on home prices, high unemployment rates and continued uncertainty surrounding potential governmental actions or requirements with respect to mortgages, the lengthening of the foreclosure process:

 

4



 

FOIA Confidential Treatment Requested by Citigroup Inc.

 

(1)                               subjects Citi to increased “severity” risk, or the loss on the amount ultimately realized for property subject to foreclosure given the continued pressure on home prices in particular markets, thus potentially increasing Citi’s net credit losses;

(2)                               inflates the amount of 180+ day delinquencies in Citigroup’s mortgage statistics and Consumer non-accrual loans (90+ day delinquencies);

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

(4)                               causes additional costs to be incurred given the longer process.”

 

For so long as material and relevant, we currently intend to include disclosure similar to the above in our future filings.  The purpose of this disclosure has been, and continues to be, to emphasize to investors the various impacts that the lengthening of the foreclosure process, particularly in the judicial states, has on various aspects of Citi’s businesses: increased severity risk; loans remain in our delinquency statistics for longer periods; creates a dampening effect on net interest margin (which is already under pressure given the low interest rate environment, as we have also disclosed for several quarters); and, overall, increases the costs of foreclosure (e.g., maintenance of property, etc.).

 

We further supplementally advise the Staff that, as the Staff would likely expect, in the ordinary course of business, while Citi may begin the foreclosure process when loans are 90+ days delinquent (become non-accrual), and thus all such loans have the potential to become “foreclosures in process,” not all such loans are in fact “foreclosures in process” at a particular point in time.  For example, for loans that become 90+ days delinquent, Citi may not refer the loan to the foreclosure process as we continue to work with the borrower as part of our loss mitigation activities.  Moreover, a previously instituted foreclosure in process could later be placed on “hold” temporarily for similar reasons.  Depending on the outcome of these loss mitigation efforts, we may begin or re-institute, as applicable, the foreclosure process on these loans.

 

Accordingly, we do not believe that the Staff’s suggested quantitative information on foreclosures in process would be helpful or relevant to our disclosure of the overall impacts associated with lengthening foreclosure timelines as such information would represent only a snapshot in time and would not necessarily accurately convey the impact of our foreclosures in process on our delinquency and other statistics.  Rather, we believe the more meaningful information is the qualitative disclosure we have provided regarding the overall increase in the

 

5



 

FOIA Confidential Treatment Requested by Citigroup Inc.

 

length of time it takes to complete foreclosures, as illustrated by the increase in foreclosures in process for two years or more, especially in judicial states, that impacts Citi’s lending statistics.  We agree, however, that this disclosure could be revised to make the above points more clear, and intend to consider this as part of the disclosure in our 2011 Annual Report on Form 10-K.

 

Value at Risk for Trading Portfolios, page 75

 

5.              We note your response to prior comment 11 regarding your VAR methodology. In addition to your proposed future disclosure related to the number of VAR backtesting exceptions, please include the supplemental discussion regarding your expectations as to why the number of backtesting exceptions is lower than what would be statistically expected and the scenarios where the number of backtesting exceptions may be higher than you expect (i.e. in periods of unstable market conditions).  Additionally, please expand your disclosure in future filings to discuss the rationale of diversification benefits and how that benefit is calculated.

 

In response to the Staff’s comment, Citigroup included the following disclosure on page 73 of its third quarter 2011 Form 10-Q regarding why the number of backtesting exceptions is lower than what would be statistically expected (as well as the number of back testing exceptions):

 

“Citigroup uses full Monte Carlo simulation, which Citi believes is conservatively calibrated to incorporate the greater of short-term (most recent month) and long-term (up to three years) market volatility.  The Monte Carlo simulation involves approximately 300,000 market factors, making use of 180,000 time series, with market factors updated daily and model parameters updated weekly.

 

The conservative features of the VAR calibration contributes approximately a 20% add-on to what would be a VAR estimated under the assumption of stable and perfectly normally distributed markets.  Under normal and stable market conditions, Citi would thus expect the number of days where trading losses exceed its VAR to be less than 2 or 3 exceptions per year; periods of unstable market conditions could increase the number of exceptions.  During the last four quarters, there was one back testing exception where trading losses exceeded the VAR estimate at the Citigroup level.”

 

6



 

FOIA Confidential Treatment Requested by Citigroup Inc.

 

In addition, we included the following disclosure related to the changes to diversification benefits during the third quarter of 2011:

 

“The main driver of the impact due to position changes was a change in the risk profile of equity markets within S&B and the corollary loss of diversification benefits across the portfolio.”

 

In addition, Citigroup intends to provide disclosure similar to the following in its 2011 Annual Report on Form 10-K as a footnote to the disclosed VAR tables to discuss the rationale of diversification benefit:

 

“ ‘Diversification benefit’ equals the difference between the total VAR and the sum of the VARs tied to each individual risk type.  The benefit reflects the fact that the risks within each and across risk types are not perfectly correlated and, consequently, the total VAR on a given day will be lower than the sum of the VARs relating to each individual risk type.  The determination of the primary drivers of changes to the diversification benefit is made by an examination of the impact of both model parameter and position changes.”

 

We intend to continue to include these disclosures, updated or revised as necessary, in our future filings so long as appropriate.

 

Country and Cross-Border Risk, page 84

 

6.              We note your response to prior comment 15 and your enhanced disclosure on page 84 of your Form 10-Q. We also note slide 27 of your third quarter earnings presentation which is provided on your website and illustrates your country risk exposure as of September 30, 2011.  We continue to believe that disclosure of your gross exposure by country (i.e. Greece, Ireland, Italy, Portugal and Spain, aka GIIPS) would be beneficial to investors and is consistent with industry practice by other financial institutions.  Accordingly, as previously requested, please revise your disclosure in future filings to present the gross exposure you have to each of these countries separately broken out by sovereigns, financial institutions, corporations and trading/AFS exposure.  Additionally, please respond to the following and clarify future disclosures as appropriate:

 

7



 

FOIA Confidential Treatment Requested by Citigroup Inc.

 

In response to the Staff’s comment, we continue to believe that Citi’s disclosure of its gross and net exposure to the GIIPS, in the aggregate, (i) is appropriate in light of the comparative size of the exposures to Citi’s overall balance sheet, (ii) strikes the appropriate balance with the confidentiality considerations of Citi’s long-standing relationships in these countries and (iii) perhaps most importantly, is responsive to the questions and concerns of our investors.

 

As we noted in our prior response letter dated August 23, 2011 (“Prior Response Letter”), it was in response to investor requests that we disclosed our gross exposure to the GIIPS in our second quarter 2011 Form 10-Q.  Other information we provided in the third quarter 2011 Form 10-Q that was responsive to investor questions or concerns included (i) the amount of Greece sovereign entity or Greek financial institution exposure included within our GIIPS net current funded exposure (page 83), and (ii) the percentage of purchased credit protection on sovereign reference names (also page 83).  While the general feedback we have received from investors is that Citi has been one of the more forthright and detailed disclosure providers on this topic, we will of course continue to review and revise our disclosures in this area to reflect investor concerns.

 

We also note for the Staff that, while disclosure of this information on a country-by-country basis, in some form, may be appropriate for certain other financial institutions, it is our view and understanding that disclosure decisions are to be based on materiality, our investors and the individual facts and circumstances of the institution.  In other words, “industry practice” is not, and we believe should not be, the rationale for disclosure of this information on a country-by-country basis.

 

Finally, and more generally, we also note for the Staff that it is unclear to us when disclosure of exposures to foreign countries would be required in addition to what is currently required under Industry Guide 3, Cross-Border Risk disclosures.  If the SEC wants such disclosure, then we believe a framework needs to be developed.  Moreover, providing a standard definition of exposure would help to ensure uniformity of application.

 

To expand on certain of the points above:

 

·                  Citi’s aggregate net current exposure to the GIIPS was $7.2 billion at September 30, 2011, which was less than 0.4% of Citigroup’s $1.9 trillion balance sheet as of such date.  Citi’s aggregate gross exposure to the GIIPS was $20.6 billion, or approximately 1.0% of Citi’s balance sheet.  We fail to see how a further country-by-country breakdown of these

 

8



 

FOIA Confidential Treatment Requested by Citigroup Inc.

 

amounts, which would represent no more than a subset of the gross and net aggregate exposures, would be beneficial or provide material incremental information to investors, particularly given the client relationship and confidentiality considerations, discussed in more detail below.

 

·                  The sovereign entities of the GIIPS are long-standing, important clients of Citigroup.  We believe that public disclosure of the individual sovereign’s banking activities with Citi is not appropriate and does not consider client confidentiality concerns.  We also believe that these long-standing relationships differentiate Citi from those financial institutions, which may disclose these exposures on a country-by-country basis as they do not generally have the global client base and the accompanying considerations that Citi has.  Moreover, aggregation of the GIIPS exposures permits Citi to maintain some level of flexibility with respect to continuing to serve important clients in this region, without sending unnecessary or unintended signals to the markets that modest increases or decreases in a given country from period to period reflect Citi’s outlook for that specific country.  As we noted in the Prior Response Letter, period-to-period disclosures, on a country-by-country basis, could easily be interpreted as Citi’s withdrawal of support for its clients within one particular country or increase the perception of the country as more risky than others.  We do not believe this result would be helpful to the already significant market volatility and instability in the region.

 

As such, we do not intend to disclose our gross and net exposure to the GIIPS on a country-by-country basis as we do not believe such disclosure would add any material information for our investors (nor have they generally asked for it), but would seriously jeopardize Citi’s client relationships and further exacerbate instability within the region.  As referenced above, Citi will continue to re-assess its GIIPS disclosures based on investor questions and concerns, market conditions and the other factors discussed in this response.

 

·                  Your disclosure on page 84 indicates that your “net funded exposure” has not been reduced by SFAS 5 or SFAS 114 credit loss reserves associated with those exposures.  Please clarify whether it incorporates any fair value adjustments that may be required if the assets are available-for-sale or trading securities; and

 

Citi notes for the Staff that our disclosure on page 84 of the second quarter 2011 Form 10-Q and page 83 of the third quarter 2011 Form 10-Q clearly indicates that the securities held in our trading and AFS portfolios which are included in our disclosed net current

 

9



 

FOIA Confidential Treatment Requested by Citigroup Inc.

 

funded exposure for the respective periods are “marked to market” daily.  Thus, we do not believe any additional clarification is necessary.

 

·                  Your disclosure on page 84 indicates that you have reduced your “net funded exposure” by $7.0 billion of hedge positions, consisting of purchased credit protection from non-GIIPS financial institutions.  We note per your response to the last bullet of prior comment 15, that the $7.0 billion represents the notional amount of credit derivatives related to these countries.  Please disclose this fact in future filings, and provide clarifying disclosure indicating that credit risk of the counterparty is excluded from this amount.  Additionally, disclose the fact that losses could still result in your positions even if there is purchased credit protection because the purchased credit protection contracts only pay out under certain scenarios and thus not all losses may be covered by the credit protection.

 

In response to the Staff’s comment, we included the following disclosure on page 82 of our third quarter 2011 Form 10-Q (footnote 4 to the table):

 

“Credit protection purchased from financial institutions predominately outside of GIIPS, France and Belgium.  As the counterparties on this purchased credit protection are predominantly from financial institutions outside of GIIPS, France and Belgium, the related counterparty credit exposure is not included in the amounts set forth in the table. In addition, such protection generally pays out only upon the occurrence of certain credit events with respect to the country or borrower covered by the protection, as determined by a committee composed of dealers and other market participants.  The credit events do not fully cover all situations that may adversely affect the value of Citi’s exposure and, accordingly, Citi could still experience losses despite the existence of the credit protection.”

 

Regarding the Staff’s comment to disclose the fact that the $7.0 billion represents the notional amount, we note for the Staff our disclosure on page 86 of the second quarter 2011 Form 10-Q as well as page 85 of the third quarter 2011 Form 10-Q that credit derivatives, for country risk purposes, are reported based on net notional amounts.

 

7.              We note your disclosure here that you have additional, locally-funded exposure in the GIIPS to retail customers and small businesses, the vast majority of which is in Spain and Greece.  Please quantify for us your total gross locally-funded exposure,

 

10



 

FOIA Confidential Treatment Requested by Citigroup Inc.

 

by country and lending category (e.g. retail customers, small business, etc.).  Please also consider disclosing this quantified information in future filings.

 

In response to the Staff’s comment, we included the following disclosure on page 83 of our third quarter 2011 Form 10-Q for the GIIPS and Belgium and France, respectively:

 

“In addition, at September 30, 2011, Citi had approximately $8.2 billion of locally-funded exposure in the GIIPS, generally to retail customers and small businesses as part of its local lending activities.  The vast majority of this exposure is in Citi Holdings (Spain and Greece).”

 

“In addition, at September 30, 2011, Citi had approximately $3.3 billion of locally-funded exposure in Belgium and France to retail customers and small businesses as part of its local lending activities.  The vast majority of this exposure is in Citi Holdings (Belgium).”

 

We do not believe that further breakdown of this disclosure is necessary or material to investors.

 

*********

 

In connection with responding to the Staff’s comments, Citigroup acknowledges that it is responsible for the adequacy and accuracy of the disclosures in its filings; that the Staff’s comments or changes to disclosure in response to Staff comments do not foreclose the Commission from taking any action with respect to the filing; and that Citigroup 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.

 

 

 

Sincerely,

 

 

 

 

 

/s/ John C. Gerspach

 

 

 

John C. Gerspach

 

Chief Financial Officer

 

11