- DB Dashboard
- Financials
- Filings
-
Holdings
- Transcripts
- ETFs
-
Insider
- Institutional
- Shorts
-
CORRESP Filing
Deutsche Bank (DB) CORRESPCorrespondence with SEC
Filed: 20 Dec 13, 12:00am
DEUTSCHE BANK AKTIENGESELLSCHAFT
Securities and Exchange Commission
Division of Corporation Finance
100 F Street, NE
Washington, D.C. 20549
Attention: | Stephanie J. Ciboroski, Senior Assistant Chief Accountant |
Division of Corporation Finance |
December 20, 2013
Re: | Deutsche Bank Aktiengesellschaft |
Form 20-F for Fiscal Year Ended December 31, 2012 |
Filed April 15, 2013 |
Form 6-K Filed October 29, 2013 |
File No. 001-15242 |
Ladies and Gentlemen:
We, Deutsche Bank Aktiengesellschaft (sometimes referred to as “Deutsche Bank”, the “Bank” or the “Group”), have received your letter, dated December 6, 2013 (the “Comment Letter”), providing comments on our Annual Report on Form 20-F for the Fiscal Year Ended December 31, 2012 (the “Annual Report” or the “2012 Form 20-F”), our Form 6-K filed October 29, 2013 and our letter dated November 1, 2013 responding to your last letter dated September 30, 2013. In response to the Comment Letter, please note the following.
Form 20-F for Fiscal Year Ended December 31, 2012
01 – Management Report
Results of Operations, page 14
Provision for Credit Losses, page 17
• | Refer to your response to prior comment 2. You state that in accordance with IAS 39, an uplift in the value of loans recorded at fair value on initial consolidation is reflected in interest income. Please clarify the following: |
1
• | What you mean by “an uplift in the value”, by stating whether this is a reference to a change in expected cash flows subsequent to acquisition. |
The term ‘uplift in value’ refers to subsequent increases in cash flows following the date of acquisition. Such increases relate primarily to those due to accelerated repayments (in the case of loans not acquired with a deep discount) or due to subsequent improvements in the credit quality of loans (in the case of loans acquired with a deep discount).
• | To the extent that it references a change in expected cash flows, clarify what types of loans experience such as increase in expected cashflows subsequent to acquisition, and specifically, whether you record an increase in expected cash flows subsequent to acquisition in interest income for loans that were deemed tonot have an incurred loss at acquisition (e.g.,not acquired at a deep discount as discussed in IAS 39.AG5) |
If loans are not impaired at the acquisition date, then the cash flows used to compute the effective interest rate (EIR) are those based on the contractual terms of the loans, considering prepayment and call options, but excluding future credit losses (both incurred and expected). If the loans are impaired at the acquisition date, then the cash flows used to compute the EIR are determined on the basis of the expected cash receipts, after reduction for incurred credit losses only in line with IAS 39.AG5.
• | To the extent that you record an increase in interest income related to an increase in the expected cash flows of loans that were deemed to not have an incurred loss at acquisition, tell us what guidance you relied upon for your accounting, and address how you considered the definition of the effective interest method at IAS 39.9, which indicates that an entity shall not consider future credit losses when calculating the effective interest rate. |
At the date of change of control, we recorded loans with a fair value of €129.9 billion from Postbank (refer to page 287 of the IFRS Financial Statements 2012), which were comprised of €126.9 billion ‘not acquired at deep discount’ and €3 billion of loans ‘acquired at deep discount’ respectively.
With respect to the €126.9 billion of ‘not acquired at deep discount’ loans, the Bank has subsequently recognised interest income on the loans for ‘uplifts in value’ (increases in expected cash flows) as the loans repaid more quickly than was originally anticipated at the acquisition date. The interest income recognised in this respect related primarily to portfolios of mortgage loans in Germany which were originated during periods of high interest rates (between 2000 to 2006). Due to lowering of market interest rates after the acquisition date, adjustments to the carrying amount and interest income were recorded as prescribed by IAS 39.AG8.
2
The remainder of the portfolio of €3 billion of loans which were considered impaired at the date of acquisition comprise primarily loans backed by commercial properties. The Bank has not observed significant increases in expected cash flows due to improvements in credit quality of such loans since the acquisition.
• | Discuss how you calculate and accrete fair value discounts for loans acquired with an incurred loss at acquisition and without an incurred loss at acquisition, and specifically, address how you consider incurred credit losses, expected credit losses, and interest rate marks for each type of loan. To the extent you believe it would be useful to the analysis, provide a representative example of calculating the discount and accreting the discount for each type of loan. |
The following illustrates how we calculate and accrete fair value discounts for loans in both categories.
Fact pattern
Assume that at 1 January 2013 the Bank purchased a fixed rate loan with a principal of 100. The purchase price is 88, which was the fair value of the loan at the date of acquisition. The Bank determined that the discount of 12 is made up of the following:
• | mark of 6 related to an increase in the market interest rate since the loan’s origination; |
• | liquidity mark of 2; |
• | incurred credit losses of 3; |
• | expected credit losses of 1. |
DB would record the loan on its books at 88 at the acquisition date.
Treatment under IFRS
If the loan was assessed as impaired, in calculating the EIR, the Bank would reduce the contractual cash flows for the incurred credit loss of 3, but exclude the expected credit loss of 1. Note that the marks for interest rates and liquidity are already incorporated into the fair value of the loan (88) and as such would not adjust the contractual cash flows.
Going forward, the interest income recognised for the loan would effectively be reduced for the incurred credit losses (net presented in interest income).
If the loan was not assessed as impaired, then in calculating the EIR, the bank would not reflect the incurred nor expected credit losses in the cash flows. As such going forward there would be a gross presentation of interest income and impairment losses.
• | Finally, quantify the amount of loans acquired with an incurred loss at acquisition and those without an incurred loss at acquisition for each period presented. |
3
As mentioned in our comment letter response dated Novermber 1, 2013 (response to comment 3) we do not segregate pre- and post-consolidation loan portfolios in the Postbank Group. We are therefore unable to quantify the amount of loans acquired with an incurred loss at acquistion and those without an incurred loss at acquistion for periods post-consolidation.
Note 1 – Significant Accounting Policies, page 250
Impairment of Loans and Provision for Off-Balance Sheet Positions, page 262
2. | Refer to your response to prior comment 7. You calculate your loan loss allowance rates for the separately identified portfolios for delinquency periods of 30 days up to the date of charge-off, and your loan loss allowance rates are calculated based upon historical data, including the charge-off rate. Given that the charge-off rate appears to be a significant input into your allowance estimate, and the charge-off period is so lengthy, please clarify the following: |
• | Whether the charge-off rate is the rate, based on historical experience, at which each identified portfolio of loans by delinquency period is expected to charge-off. |
We hereby confirm that the Loan Loss Allowance (LLA) rate determines the level of allowance required for a particular delinquency stage and, as described in our prior response letter, is determined based on our historic economic loss experience, factoring in charge-offs and subsequent recoveries for those assets already charged-off.
In general, we charge-off at two dates:
1. | For our uncollateralized loan positions (‘going concern’ groups) we conduct a partial write off at 180/270 days past due retaining only the expected recoverable amount, being the unprovisioned loan portion at that date. At the time of the partial write off the charge-off rate equals the LLA rate. Hence the loan portion ‘charged-off’ equates to 100% of the LLA provision at that date. LLAs for the remaining ‘recoverable’ portion are calculated over the remaining life of the loan based on prior collection history for the various portfolios. For our collateralized loans (‘loss to liquidation’ groups) at the charge off date (840 days past due) the loan is written down to the collateral value. Collateral value is calculated and reviewed on an individual basis post charge-off. |
2. | In the second step the remaining loan position is written-off when we deem the recovery process to be finalized or when we deem the exposure to be unrecoverable. At the latest, the full charge-off for uncollateralized loans is performed at 1980 or 2070 days past due for the ‘going concern’ loans. The date of final charge off for collateralized loans depends on the liquidation periods in the relevant jurisdiction. Liquidation periods vary dependant on country specific regulations. |
4
The 180 and 270 days past due charge-off dates are applied based on our collection experience. The uncollaterized loans are assessed as portfolios and will include loans of mixed recoverability. On average the LLA rate of the uncollaterized portfolios at primary charge-off is 80%.
• | If the foregoing is true, identify the delinquency period to which you apply a charge-off rate that is very close to or at 100%. |
As the uncollaterised loans are assessed on a portfolio basis both pre- and post-primary charge-off, the LLA rate/charge-off rate varies between portfolios and may never reach 100%. It is dependant on the loans the portfolios consist of. As detailed above we do have a final cut off date (1980/2070 days past due) after which all loan positions are fully charged-off and we assume a 100% unrecoverability for all uncollateralized loans.
• | How you recalibrated your loan loss allowance rates when you changed your charge-off policy to extend charge-off timelines. Specifically, address the risks that it is taking more time (up to 840 days in some cases) to observe actual losses and recoveries through your charge-off rates now that your charge-off timeline is longer, and that your charge-off rates may not be reflective of the current credit environment on a timely basis. Discuss how you mitigate those risks in your loan loss allowance rates. |
We hereby confirm that neither our LLA rate methodology nor our charge-off timelines have been adjusted since 2002, though we have performed on-going reviews to ensure their continued applicability.
The only adjustment has been a move from full charge-off at 180/270 days past due to a partial write-off to the level of the LLA provision, as discussed above. This change in the charge-off process was made in 2009. Prior to this date we experienced high levels of recovery post charge-off and we therefore believe a partial rather than full charge-off at these dates is appropriate.
As detailed in our prior response our LLA rates are revised on a quarterly basis based on the current loan portfolio data and will therefore reflect any changes in the credit environment, e.g. each and every LLA rate per delinquency bucket takes into account all future expected write-offs and recoveries. We therefore believe the application of provisioning based on our current LLA methodology provides comfort that loans are adequately provided for up to and between the charge-off dates.
Note 14 – Amendments to IAS 39 and IFRS 7, “Reclassification of Financial Assets,” page 309
3. | Refer to your response to prior comment 11. Please revise your disclosure in future filings within Operating and Financial Review and Prospects or elsewhere by expanding your discussion of the underlying factors that contributed to the difference between the fair value and carrying value as of December 31, 2012, the significant amount of additional provisions and losses upon sale of the reclassified assets during the nine months ended September 30, 2013, and why you believe no further impairment is necessary by addressing the following: |
• | The credit quality of your securitized assets classified as covered bonds, CDOs/CLOs, Student Loan ABS, and commercial mortgage ABS, as well as your loans classified as residential mortgages. As an example, the credit quality discussion could address subordination within a security or other credit enhancement available to your position or tranche of a security. |
5
• | Quantify the amount of each of these securitized asset and loan classes where you have observed delinquencies in payments of principal or interest, increased probabilities that a borrower will enter bankruptcy or financial reorganization, or counterparties facing significant financial difficulties. Discuss how these factors have changed over time, including whether factors changed in 2013 such that you recorded significantly higher provisions on these reclassified assets during 2013 relative to 2012. |
• | For the securitized asset and loan classes you have quantified in response to the preceding bullet point, tell us what evidence you have observed that supports the lack of impairment on these instruments as of December 31, 2012 and whether that changed during 2013. |
We will enhance our disclosure on the carrying value/fair value differential of the reclassified financial assets by providing a more granular breakdown of these amounts by asset type for the Securitized Assets and Loan trading assets reclassified to loans. We plan to provide this in the relevant footnote of the Financial Statements. Additionally we will provide commentary on the asset classes with significant carrying value to fair value differences for each period presented addressing the causes of key movements in the differentials. A table similar to that shown below will be included in the footnote to the Financial Statements.
€m | Dec 31, 2013 | Dec 31, 2012 | ||||||||||
Trading Assets reclassified as loans: | Carrying value | Fair value | FV-CV Difference | Carrying value | Fair value | FV-CV Difference | ||||||
Covered Bond | ||||||||||||
CDO/CLO | ||||||||||||
Student Loans | ||||||||||||
Commercial Mortgages | ||||||||||||
Senior Secured | ||||||||||||
Residential Mortgages | ||||||||||||
Other | ||||||||||||
|
|
|
|
|
| |||||||
Securitized assets Total | ||||||||||||
|
|
|
|
|
| |||||||
Commercial Mortgages | ||||||||||||
Residential Mortgages | ||||||||||||
Municipal Loans | ||||||||||||
Other | ||||||||||||
|
|
|
|
|
| |||||||
Loan Total | ||||||||||||
|
|
|
|
|
| |||||||
Grand Total | ||||||||||||
|
|
|
|
|
|
Additionally in the financial statement footnotes we will expand our commentary on loan loss provisions incurred on reclassified assets during the year including commentary on trigger events occurring in the period. Also we will provide commentary on the derisking losses incurred during the period.
6
We will provide additional information regarding the credit quality of the reclassified assets by showing the contribution of the reclassified assets to existing key credit risk disclosures in the Risk Report within our Management Report. An example of the detail to be provided is detailed below. We will provide comparative year detail.
Credit Quality of Financial Instruments neither Past Due nor Impaired (both tables for 2013 and 2012)
Dec 31, 2013 | ||||||||||||||||||||||||||||
in € m. 1 | iAAA–iAA | iA | iBBB | iBB | iB | iCCC and below | Total | |||||||||||||||||||||
Due from banks | 0 | 0 | 0 | 0 | 0 | 0 | 0 | |||||||||||||||||||||
Interest-earning deposits with banks | 0 | 0 | 0 | 0 | 0 | 0 | 0 | |||||||||||||||||||||
Central bank funds sold and securities purchased under resale agreements | 0 | 0 | 0 | 0 | 0 | 0 | 0 | |||||||||||||||||||||
Securities borrowed | 0 | 0 | 0 | 0 | 0 | 0 | 0 | |||||||||||||||||||||
Financial assets at fair value through profit or loss 2 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | |||||||||||||||||||||
Financial assets available for sale 2 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | |||||||||||||||||||||
Loans 3 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | |||||||||||||||||||||
Thereof: IAS 39 reclassified loans | 0 | 0 | 0 | 0 | 0 | 0 | 0 | |||||||||||||||||||||
Other assets subject to credit risk | 0 | 0 | 0 | 0 | 0 | 0 | 0 | |||||||||||||||||||||
Financial guarantees and other credit related contingent liabilities 4 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | |||||||||||||||||||||
Irrevocable lending commitments and other credit related commitments 4 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | |||||||||||||||||||||
|
|
|
|
|
|
|
|
|
|
|
|
|
| |||||||||||||||
Total | 0 | 0 | 0 | 0 | 0 | 0 | 0 | |||||||||||||||||||||
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Additionally we will extend our disclosure on impaired reclassified loans in the Risk Report to provide detail, when material, on reclassified loan positions when past due but neither renegotiated nor impaired and for reclassified loans positions that are renegotiated, but not impaired.
In subsequent interim financial statements we will review whether the carrying value versus fair value positions have moved substantially since year end in order to determine whether additional quantitative or qualitative disclosure is required in the interim reports.
Form 6-K Filed October 29, 2013
Exhibit 99.1
Litigation, page 95
4. | We note your disclosure that you currently estimate that as of September 30, 2013, the aggregate future loss of which the possibility is more than remote, but less than probable is approximately € 1.3 billion. We also note your disclosure on page 4 of your third quarter earnings presentation that is available on your website that you describe this same € 1.3 billion as reasonably possible losses above provisions, which are recognized pursuant to accounting standards when the outflow of funds is determined to be more than remote (>10%) but less than probable (<50%) and an estimate of such outflow reliably can be made. Please respond to the following: |
• | Clarify whether this € 1.3 billion estimate represents only matters which are contingent liabilities, that is, those matters for which no provision has been recorded, or whether this amount also includes matters where a provision has been recorded, but there is an additional amount that is in excess of the amount accrued which is more than remote, but less than probable. |
7
• | To the extent that this € 1.3 billion estimate excludes any component related to matters for which you have recorded a provision, please include clarifying disclosure discussing this fact. |
We confirm that this € 1.3 billion estimate represents only claims which are contingent liabilities, that is, those claims for which no provision has been recorded. As stated in Note 29 “Provisions” of our 2012 audited financial statements (which are included in our 2012 Form 20-F), “Where a provision has been taken for a particular claim, no contingent liability is recorded.”
Mortgage Repurchase Demands, page 101
5. | Refer to your response to prior comment 24 and your disclosure on page 101 of your interim report which shows a continued increase in outstanding repurchase requests and provisions related to repurchase demands as of September 30, 2013. Given the uncertainty and complexities related to estimating provisions for repurchase demands, please confirm that you will revise your disclosure in future filings to discuss your lack of information on loans sold as whole loans (roughly 45% of all loans sold/securitized), and for those securitized loans where you do have information, that you have not observed a direct correlation between delinquency status and demand activity as repurchase claims have been received on loans that are current, modified, and paid in full. |
We will revise the disclosure in our Form 20-F for year end 2013 to discuss this issue using text similar to the following:
‘Deutsche Bank anticipates that additional mortgage repurchase demands may be made in respect of the mortgage loans that it has sold, but cannot reliably estimate their timing or amount. Deutsche Bank did not act as servicer for the loans sold to third parties as whole loans (which constitute almost half of all U.S. residential mortgage loans sold from 2005 through 2008) and, once sold, Deutsche Bank generally ceased to have access to information about their performance. While loan performance is publicly available on the mortgage loans that Deutsche Bank securitized, no direct correlation has been observed between their performance and repurchase demands received. Demands have been received on loans that have defaulted, as well as loans that are current and loans that have been repaid in full.’
8
Deutsche Bank Aktiengesellschaft acknowledges that it is responsible for the adequacy and accuracy of the disclosure in its filing, that staff comments or changes to disclosure in response to staff comments do not foreclose the Commission from taking any action with respect to the filing and that it 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.
Should you have any questions or comments, please feel free to contact the undersigned, Karin Dohm (ph: +49(69)910-31183; fax: +49(69)910-39741; e-mail: karin.dohm@db.com) or Mathias Otto (ph: +49-69-910-33962; fax: +49-69-910-38572; e-mail: mathias.otto@db.com).
Deutsche Bank Aktiengesellschaft
Karin Dohm | Mathias Otto | |
Head of Group External Reporting and | Deputy General Counsel Germany, | |
APAG, Chief Accounting Officer | Central and Eastern Europe |
9