6-K 1 a200325bs-6k.htm 6-K 6-K
UNITED STATES
SECURITIES AND EXCHANGE COMMISSION
Washington, D.C. 20549

Form 6-K
REPORT OF FOREIGN PRIVATE ISSUER PURSUANT TO RULE 13a-16 OR 15d-16
UNDER THE SECURITIES EXCHANGE ACT OF 1934
March 25, 2020
Commission File Number 001-15244
CREDIT SUISSE GROUP AG
(Translation of registrant’s name into English)
Paradeplatz 8, CH 8001 Zurich, Switzerland
(Address of principal executive office)

Indicate by check mark whether the registrant files or will file annual reports under cover of Form 20-F or
Form 40-F.
   Form 20-F      Form 40-F   
Indicate by check mark if the registrant is submitting the Form 6-K in paper as permitted by Regulation S-T Rule 101(b)(1):
Note: Regulation S-T Rule 101(b)(1) only permits the submission in paper of a Form 6-K if submitted solely to provide an attached annual report to security holders.
Indicate by check mark if the registrant is submitting the Form 6-K in paper as permitted by Regulation S-T Rule 101(b)(7):
Note: Regulation S-T Rule 101(b)(7) only permits the submission in paper of a Form 6-K if submitted to furnish a report or other document that the registrant foreign private issuer must furnish and make public under the laws of the jurisdiction in which the registrant is incorporated, domiciled or legally organized (the registrant’s “home country”), or under the rules of the home country exchange on which the registrant’s securities are traded, as long as the report or other document is not a press release, is not required to be and has not been distributed to the registrant’s security holders, and, if discussing a material event, has already been the subject of a Form 6-K submission or other Commission filing on EDGAR.
Indicate by check mark whether the registrant by furnishing the information contained in this Form is also thereby furnishing the information to the Commission pursuant to Rule 12g3-2(b) under the Securities Exchange Act of 1934.
   Yes      No   
If “Yes” is marked, indicate below the file number assigned to the registrant in connection with Rule 12g3-2(b): 82-.






Pursuant to the requirements of the Securities Exchange Act of 1934, the registrant has duly caused this report to be signed on its behalf by the undersigned, thereunto duly authorized.
CREDIT SUISSE GROUP AG
 (Registrant)
Date: March 25, 2020
By:
/s/ Lara J. Warner
Lara J. Warner
Chief Risk Officer
By:
/s/ David R. Mathers
David R. Mathers
Chief Financial Officer












For purposes of this report, unless the context otherwise requires, the terms “Credit Suisse,” the “Group,” “we,” “us” and “our” mean Credit Suisse Group AG and its consolidated subsidiaries. The business of Credit Suisse AG, the direct bank subsidiary of the Group, is substantially similar to the Group, and we use these terms to refer to both when the subject is the same or substantially similar. We use the term the “Bank” when we are only referring to Credit Suisse AG and its consolidated subsidiaries.
Abbreviations are explained in the List of abbreviations in the back of this report.
Publications referenced in this report, whether via website links or otherwise, are not incorporated into this report.
In various tables, use of “–” indicates not meaningful or not applicable.


Pillar 3 and regulatory disclosures 4Q19
Credit Suisse Group AG

Introduction
Swiss capital requirements
Overview of risk management
Risk-weighted assets
Linkages between financial statements and regulatory exposures
Credit risk
Counterparty credit risk
Securitization
Market risk
Interest rate risk in the banking book
Additional regulatory disclosures
List of abbreviations
Cautionary statement regarding forward-looking information






Introduction
General
This Group report as of December 31, 2019 is based on the revised Circular 2016/1 “Disclosure – banks” (FINMA circular) issued by the Swiss Financial Market Supervisory Authority FINMA (FINMA) on October 31, 2019. The revised FINMA circular includes the implementation of the revised Pillar 3 disclosure requirements issued by the Basel Committee on Banking Supervision (BCBS) in March 2017.
This report is produced and published quarterly, in accordance with FINMA requirements. The reporting frequency for each disclosure requirement is either annual, semi-annual or quarterly. This document should be read in conjunction with the Pillar 3 and regulatory disclosures – Credit Suisse Group AG 2Q19 and 3Q19 and the Credit Suisse Annual Report 2019, which includes important information on regulatory capital, risk management (specific references have been made herein to these documents) and regulatory developments and proposals.
The highest consolidated entity in the Group to which the FINMA circular applies is Credit Suisse Group.
These disclosures were verified and approved internally in line with our board-approved policy on disclosure controls and procedures. The level of internal control processes for these disclosures is similar to those applied to the Group’s quarterly and annual financial reports. This report has not been audited by the Group’s external auditors.
For certain prescribed table formats where line items have zero balances, such line items have not been presented.
This report reflects certain updates and corrections to prior period metrics which have been noted in the relevant tabular disclosures, where applicable.
Other regulatory disclosures
In connection with the implementation of Basel III, certain regulatory disclosures for the Group and certain of its subsidiaries are required. The Group’s Pillar 3 disclosure, regulatory disclosures, additional information on capital instruments, including the main features of regulatory capital instruments and total loss-absorbing capacity (TLAC)-eligible instruments that form part of the eligible capital base and TLAC resources, Global systemically important bank (G-SIB) financial indicators, reconciliation requirements, leverage ratios and certain liquidity disclosures as well as regulatory disclosures for subsidiaries can be found on our website.
> Refer to credit-suisse.com/regulatorydisclosures for additional information.
Regulatory developments
In June 2019, the BCBS released a revised framework regarding the treatment of client-cleared derivatives for purposes of the leverage ratio and a revision of the leverage ratio disclosure requirements as part of the Pillar 3 framework. The revision regarding the treatment of client-cleared derivatives aims to align the leverage ratio measurement of client-cleared derivatives with the standardized approach to measuring counterparty credit risk exposures as applied for risk-based capital requirements. Additionally, the revised leverage ratio disclosure requirements set out additional obligations for banks to disclose their leverage ratios based on quarter-end and on daily average values of securities financing transactions. Both revisions will be applicable to the version of the leverage ratio standard that will enter into effect on January 1, 2022.
> Refer to “Regulatory developments” (pages 119 to 120) in III – Treasury, Risk, Balance sheet and Off-balance sheet – Capital management in the Credit Suisse Annual Report 2019 for further information on regulatory developments.
Location of disclosure
This report provides the Pillar 3 and regulatory disclosures required by the FINMA circular for the Group to the extent that these disclosures are not included in the Credit Suisse Annual Report 2019 or in the regulatory disclosures on our website.
> Refer to “Annual Report” under credit-suisse.com/ar for disclosures included in the Credit Suisse Annual Report 2019.
2

Location of disclosures   
FINMA disclosure requirements Location Page number
Overview of risk management, key prudential metrics and risk-weighted assets         
Key prudential metrics [Table KM1] / [Table KM2] Qualitative disclosures: "Treasury, Risk, Balance sheet and Off-balance sheet" 110 - 129
Risk management approach [Table OVA] "Risk management oversight"
"Risk appetite framework"
"Risk coverage and management"
136 - 140
140 - 142
143 - 159
Overview of risk-weighted assets [Table OV1] Qualitative disclosures: "Risk-weighted assets" 124 - 126
Linkages between financial statements and regulatory exposures         
Valuation process [Table LIA] "Fair valuations"
"Critical accounting estimates - Fair value"
"Note 35 - Financial instruments"
65
101
347 - 352
Composition of capital and TLAC         
Differences in basis of consolidation [Table CC2] List of significant subsidiaries and associated entities:
"Note 40 - Significant subsidiaries and equity method investments"
Changes in scope of consolidation:
"Note 3 - Business developments, significant shareholders and subsequent events"
 
387 - 390
 
279
Main features of regulatory capital instruments and TLAC-eligible instruments [Table CCA] Refer to "Capital instruments" under credit-suisse.com/regulatorydisclosures 1
Macroprudential supervisor measures         
Disclosure of G-SIBs indicators [Table GSIB1] Refer to "G-SIB Indicators" under credit-suisse.com/regulatorydisclosures 1
Credit risk         
General qualitative information [Table CRA] "Credit risk" 146 - 149
Additional disclosure related to credit quality
of assets [Table CRB a), b), c) and d)]
"Note 1 - Summary of significant accounting policies"
"Note 19 - Loans, allowance for loan losses and credit quality"
271 - 273
289 - 296
Qualitative disclosure requirements related to credit
risk mitigation techniques [Table CRC a)]: Netting
"Derivative instruments"
"Note 1 - Summary of significant accounting policies"
"Note 27 - Offsetting of financial assets and financial liabilities"
166 - 167
269 - 270
306 - 309
Counterparty credit risk         
Qualitative disclosure requirements [Table CCRA] Transaction rating, credit limits and provisioning: "Credit risk"
Effect of a credit rating downgrade: "Credit ratings"
146 - 149
114 - 115
Securitization         
Qualitative disclosure requirements [Table SECA] "Note 34 - Transfers of financial assets and variable interest entities" 339 - 347
Market risk         
Qualitative disclosure requirements [Table MRA] "Market risk"
"Note 1 - Summary of significant accounting policies"
"Note 32 - Derivatives and hedging activities"
149 - 153
269 - 270
329 - 334
Leverage metrics         
Qualitative disclosures [Table LR2] "Leverage metrics"
"Swiss metrics"
127
128 - 129
Liquidity coverage ratio         
Liquidity risk management [Table LIQA] "Liquidity and funding management" 108 - 115
Liquidity Coverage Ratio [Table LIQ1] Qualitative disclosures: "Liquidity metrics" 110 - 111
Corporate Governance         
Corporate Governance [Appendix 5] "Corporate Governance" 177 - 222
Remuneration         
Remuneration policy [Table REMA] "Compensation" 223 - 256
Remuneration awarded during the financial
year [table REM1] / Special payments [table REM2] /
Deferred remuneration [table REM3]
Senior management: "Executive Board compensation"

Other material risk takers: "Group compensation"
228 - 235

236 - 244
Operational risk         
Qualitative disclosures [Table ORA] "Non-financial risk regulatory capital measurement" 155
Special duties of disclosure for systemically important financial institutions and stand-alone banks         
List and qualification of alleviations granted [Appendix 4] "FINMA Decrees" 118
1
The disclosure will be available by the end of April 2020.
3

Swiss capital requirements
FINMA requires the Group to fully comply with the special requirements for systemically important financial institutions operating internationally. The following tables present the Swiss capital and leverage requirements and metrics as required by FINMA.
> Refer to “Swiss requirements” (pages 117 to 119) in III – Treasury, Risk, Balance sheet and Off-balance sheet – Capital management – Regulatory framework and “Swiss metrics” (pages 128 to 129) in III – Treasury, Risk, Balance sheet and Off-balance sheet – Capital management in the Credit Suisse Annual Report 2019 for further information on general Swiss requirements and the related metrics.
Swiss capital requirements and metrics
   Phase-in Look-through

end of 4Q19

CHF million
in %
of RWA

CHF million
in %
of RWA
Swiss risk-weighted assets                           
Swiss risk-weighted assets 291,282 291,282
Risk-based capital requirements (going-concern) based on Swiss capital ratios                           
Total 40,401 13.87 42,499 14.59
   of which CET1: minimum  14,273 4.9 13,108 4.5
   of which CET1: buffer  13,923 4.78 16,021 5.5
   of which CET1: countercyclical buffers  845 0.29 845 0.29
   of which additional tier 1: minimum  9,030 3.1 10,195 3.5
   of which additional tier 1: buffer  2,330 0.8 2,330 0.8
Swiss eligible capital (going-concern)                           
Swiss CET1 capital and additional tier 1 capital 1 52,691 18.1 49,757 17.1
   of which CET1 capital 2 36,740 12.6 36,740 12.6
   of which additional tier 1 high-trigger capital instruments  8,310 2.9 8,310 2.9
   of which additional tier 1 low-trigger capital instruments 3 4,707 1.6 4,707 1.6
   of which tier 2 low-trigger capital instruments 4 2,934 1.0
Risk-based requirement for additional total loss-absorbing capacity (gone-concern) based on Swiss capital ratios                           
Total according to size and market share 33,789 5 11.6 5 41,653 14.3
Reductions due to rebates in accordance with article 133 of the CAO (5,406) (1.856) (6,665) (2.288)
Reductions due to the holding of additional instruments in the form of convertible capital in accordance with Art. 132 para 4 CAO (1,467) (0.504)
Total, net 28,383 9.744 33,522 11.508
Eligible additional total loss-absorbing capacity (gone-concern)                           
Total 38,576 13.2 41,138 14.1
   of which bail-in instruments  37,172 12.8 37,172 12.8
   of which tier 2 low-trigger capital instruments  1,032 0.4 3,966 1.4
   of which non-Basel III-compliant tier 2 capital  372 6 0.1
Rounding differences may occur.
1
Excludes tier 1 capital, which is used to fulfill gone-concern requirements.
2
Excludes CET1 capital, which is used to fulfill gone-concern requirements.
3
If issued before July 1, 2016, such capital instruments qualify as additional tier 1 high-trigger capital instruments until their first call date according to the transitional Swiss "Too Big to Fail" rules.
4
If issued before July 1, 2016, such capital instruments qualify as additional tier 1 high-trigger capital instruments no later than December 31, 2019 according to the transitional Swiss "Too Big to Fail" rules.
5
Consists of a base requirement of 10.52%, or CHF 30,643 million, and a surcharge of 1.08%, or CHF 3,146 million.
6
Non-Basel III-compliant tier 2 capital instruments are subject to phase-out requirements. The amount includes the amortization component of CHF 58 million and the unamortized component of CHF 314 million.
4

Swiss leverage requirements and metrics
   Phase-in Look-through

end of 4Q19

CHF million
in %
of LRD

CHF million
in %
of LRD
Leverage exposure                           
Leverage ratio denominator 909,994 909,994
Unweighted capital requirements (going-concern) based on Swiss leverage ratio                           
Total 40,950 4.5 45,500 5.0
   of which CET1: minimum  15,470 1.7 13,650 1.5
   of which CET1: buffer  13,650 1.5 18,200 2.0
   of which additional tier 1: minimum  11,830 1.3 13,650 1.5
Swiss eligible capital (going-concern)                           
Swiss CET1 capital and additional tier 1 capital 1 52,691 5.8 49,757 5.5
   of which CET1 capital 2 36,740 4.0 36,740 4.0
   of which additional tier 1 high-trigger capital instruments  8,310 0.9 8,310 0.9
   of which additional tier 1 low-trigger capital instruments 3 4,707 0.5 4,707 0.5
   of which tier 2 low-trigger capital instruments 4 2,934 0.3
Unweighted requirements for additional total loss-absorbing capacity (gone-concern) based on Swiss leverage ratio                           
Total according to size and market share 36,400 5 4.0 5 45,500 5.0
Reductions due to rebates in accordance with article 133 of the CAO (5,824) (0.64) (7,280) (0.8)
Reductions due to the holding of additional instruments in the form of convertible capital in accordance with Art. 132 para 4 CAO (1,467) (0.161)
Total, net 30,576 3.36 36,753 4.039
Eligible additional total loss-absorbing capacity (gone-concern)                           
Total 38,576 4.2 41,138 4.5
   of which bail-in instruments  37,172 4.1 37,172 4.1
   of which tier 2 low-trigger capital instruments  1,032 0.1 3,966 0.4
   of which non-Basel III-compliant tier 2 capital  372 6 0.0
Rounding differences may occur.
1
Excludes tier 1 capital, which is used to fulfill gone-concern requirements.
2
Excludes CET1 capital, which is used to fulfill gone-concern requirements.
3
If issued before July 1, 2016, such capital instruments qualify as additional tier 1 high-trigger capital instruments until their first call date according to the transitional Swiss "Too Big to Fail" rules.
4
If issued before July 1, 2016, such capital instruments qualify as additional tier 1 high-trigger capital instruments no later than December 31, 2019 according to the transitional Swiss "Too Big to Fail" rules.
5
Consists of a base requirement of 3.625%, or CHF 32,987 million, and a surcharge of 0.375%, or CHF 3,413 million.
6
Non-Basel III-compliant tier 2 capital instruments are subject to phase-out requirements. The amount includes the amortization component of CHF 58 million and the unamortized component of CHF 314 million.
5

Overview of risk management
General
Fundamental to our business is the prudent taking of risk in line with our strategic priorities. The primary objectives of risk management are to protect our financial strength and reputation, while ensuring that capital is well deployed to support business activities. Our risk management framework is based on transparency, management accountability and independent oversight. Risk management is an integral part of our business planning process with strong involvement of senior management and the Board of Directors. Risk measurement models are reviewed by the Model Risk Management team, an independent validation function, and regularly presented to and approved by the relevant oversight committee.
> Refer to “Risk management oversight” (pages 136 to 140), “Risk appetite framework” (pages 140 to 142) and “Risk coverage and management” (pages 143 to 159) in III – Treasury, Risk, Balance sheet and Off-balance sheet – Risk management in the Credit Suisse Annual Report 2019 for information on risk management oversight including risk culture, risk governance, risk organization, risk types, risk appetite, risk limits, stress testing and strategies/processes to manage, hedge and mitigate risks.
Risk reporting
Risk reporting is performed regularly and there are numerous internal control procedures in place, in particular the standard operating procedures, risk and control assessment and independent report review. These ensure the reporting and measurement systems are up to date and are working as intended. They cover: validation and authorization of risk measurement data, status summary reports, data reconciliation, independent checks/validation and error reports to capture any failings. Senior management and the Board of Directors are informed about key risk metrics, including Value-at-Risk (VaR), Economic Risk Capital (ERC), key risks and top exposures with the monthly Group Risk Report.
Key risks
The Group is exposed to several key banking risks such as:
Credit risk (refer to section “Credit risk” on pages 12 to 43);
Counterparty credit risk (refer to section “Counterparty credit risk” on pages 44 to 53);
Securitization risk (refer to section “Securitization risk” on pages 54 to 61);
Market risk (refer to section “Market risk” on pages 62 to 66);
Interest rate risk in the banking book (refer to section “Interest rate risk in the banking book” on pages 67 to 70); and
Operational risk.
> Refer to “Non-financial risk regulatory capital measurement” (page 155) in III – Treasury, Risk, Balance sheet and Off-balance sheet – Risk management – Risk coverage and management in the Credit Suisse Annual Report 2019 for information on operational risk.
The Basel framework proposes various approaches for determining capital requirements which banks have to abide by in order maintain regulatory compliance. Credit Suisse has adopted a modelled approach with respect to most of its risk types, both for regulatory and internal requirements, in order to ensure our capital resources are appropriate to our risk profile.
6

Risk-weighted assets
With the adoption of the revised FINMA circular, risk-weighted assets (RWA) presented in this report, including prior period comparisons, are based on the Swiss capital requirements.
> Refer to “Swiss requirements” (pages 117 to 119) in III – Treasury, Risk, Balance sheet and Off-balance sheet – Capital management in the Credit Suisse Annual Report 2019 for further information on Swiss capital requirements.
The following table presents an overview of total Swiss RWA forming the denominator of the risk-based capital requirements. Further breakdowns of RWA are presented in subsequent sections of this report.
RWA of CHF 291.3 billion as of the end of 4Q19 decreased 4% compared to the end of 3Q19, mainly resulting from a decrease relating to movements in risk levels in credit risk and market risk and a negative foreign exchange impact, partially offset by an increase relating to external model and parameter updates in credit risk.
RWA flow statements for credit risk, counterparty credit risk (CCR) and market risk are presented in subsequent parts of this report.
> Refer to “Risk-weighted assets” (pages 124 to 126) in III – Treasury, Risk, Balance sheet and Off-balance sheet – Capital management in the Credit Suisse Annual Report 2019 for further information on risk-weighted assets movements in 2019.
OV1 – Overview of Swiss risk-weighted assets and capital requirements 
     
Risk-weighted assets
Capital
requirement
1
end of 4Q19 3Q19 4Q18 4Q19
CHF million   
Credit risk (excluding counterparty credit risk) 144,984 146,413 139,867 11,599
   of which standardized approach (SA)  25,518 24,935 13,190 2,042
   of which supervisory slotting approach  4,212 3,509 2,403 337
   of which advanced internal ratings-based (A-IRB) approach  115,254 117,969 124,274 9,220
Counterparty credit risk 20,365 23,044 17,613 1,629
   of which standardized approach for counterparty credit risk (SA-CCR) 2 1,830 2,964 2,559 146
   of which internal model method (IMM)  17,486 19,060 14,086 1,399
   of which other counterparty credit risk 3 1,049 1,020 968 84
Credit valuation adjustments (CVA) 6,892 8,402 5,743 551
Equity positions in the banking book under the simple risk weight approach 10,202 10,410 8,378 816
Settlement risk 219 148 259 18
Securitization exposures in the banking book 13,333 14,393 12,541 1,067
   of which securitization internal ratings-based approach (SEC-IRBA)  7,751 8,222 6,915 620
   of which securitization external ratings-based approach (SEC-ERBA), including internal assessment approach (IAA)  1,555 1,622 1,727 125
   of which securitization standardized approach (SEC-SA)  4,027 4,549 3,899 322
Market risk 15,192 18,376 18,643 1,215
   of which standardized approach (SA)  1,981 2,031 2,393 158
   of which internal model approach (IMA)  13,211 16,345 16,250 1,057
Operational risk (AMA) 68,318 70,475 71,040 5,466
Amounts below the thresholds for deduction (subject to 250% risk weight) 11,777 11,249 11,109 942
Total  291,282 302,910 285,193 23,303
1
Calculated as 8% of Swiss risk-weighted assets, based on total capital minimum requirements, excluding capital conservation buffer and G-SIB buffer requirements.
2
Calculated under the current exposure method.
3
Includes RWA for contributions to the default fund of a central counterparty and loans hedged by centrally cleared CDS.
7

Linkages between financial statements and regulatory exposures
This section shows the various sources of differences between the carrying values presented in the Group’s financial statements prepared in accordance with accounting principles generally accepted in the US (US GAAP) and the exposure amounts used for regulatory purposes. The identification, classification and presentation of these sources of differences requires a significant amount of management judgement and is based on the information available at the time. As such, reclassifications have been made compared to the prior year. Management believes that the estimates and assumptions used in the preparation of these disclosures are prudent, reasonable and consistently applied.
The following table shows the differences between the scope of accounting consolidation and the scope of regulatory consolidation, broken down by how the amounts reported in the Group’s financial statements correspond to regulatory risk categories. The column about the securitization framework includes securitizations in the banking book, whereas securitizations in the trading book are included in the column about market risk. Foreign exchange risk in the banking book is captured by the Internal Model Approach (IMA) in market risk. Positions with foreign exchange risk in the banking book are not included in the column about market risk. Cash collateral is excluded from market risk. However, the cash leg of securities financing transactions (SFT) in the trading book is included in the column about market risk.
LI1 - Differences between accounting and regulatory scopes of consolidation and mapping of financial statements with regulatory risk categories
   Carrying values Carrying values of items subject to:

end of 4Q19




Published
financial
statements




Regulatory
scope of
consolidation



Credit
risk
frame-
work

Counter-
party
credit
risk
frame-
work



Securiti-
zation
frame-
work



Market
risk
frame-
work
Not subject
to capital
require-
ments or
subject to
deduction
from capital
Assets (CHF million)   
Cash and due from banks 101,879 101,487 99,956 10 0 0 1,521
Interest-bearing deposits with banks 741 1,167 1,167 0 0 0 0
Central bank funds sold, securities purchased under resale agreements and securities borrowing transactions 106,997 106,997 25 106,972 0 98,244 0
Securities received as collateral, at fair value 40,219 40,219 0 40,219 0 40,190 0
Trading assets, at fair value 1 153,797 147,302 8,883 64,394 2 1,718 150,080 0
Investment securities 1,006 1,006 990 0 16 0 0
Other investments 5,666 5,848 3,086 0 382 464 1,916
Net loans 296,779 297,095 267,989 208 27,868 1,586 (309)
Goodwill 4,663 4,668 0 0 0 0 4,668
Other intangible assets 291 291 0 0 0 0 291
Brokerage receivables 35,648 35,648 2,245 28,159 0 0 5,249
Other assets 39,609 38,917 18,502 5,137 1,551 6,386 7,380
Total assets  787,295 780,645 402,843 245,099 31,535 296,950 20,716
Liabilities (CHF million)   
Due to banks 16,744 17,139 0 0 0 0 17,139
Customer deposits 383,783 383,793 0 0 0 0 383,793
Central bank funds purchased, securities sold under repurchase agreements and securities lending transactions 27,533 32,597 0 32,573 0 20,988 24
Obligation to return securities received as collateral, at fair value 40,219 40,219 0 40,219 0 40,190 0
Trading liabilities, at fair value 1 38,186 38,252 12 14,577 0 56,746 478
Short-term borrowings 28,385 23,370 0 0 0 5,628 17,742
Long-term debt 152,005 150,364 0 0 0 50,966 99,398
Brokerage payables 25,683 25,683 0 20,413 0 0 5,270
Other liabilities 31,043 25,402 418 8,562 0 639 15,810
Total liabilities  743,581 736,819 430 116,344 0 175,157 539,654
There are items in the table which attract capital charges according to more than one risk category framework. As an example, derivatives assets/liabilities held in the regulatory trading book are shown in the column about market risk and in the column about counterparty credit risk.
1
Trading assets/liabilities on the balance sheet reflect the balance after considering netting benefit of cash collateral hence reflect a lower balance than disclosed in the market risk column as cash collateral is not part of the market risk framework.
2
Includes assets pledged as collateral since collateral posted is subject to counterparty credit risk.
8

LI1 - Differences between accounting and regulatory scopes of consolidation and mapping of financial statements with regulatory risk categories (continued)
   Carrying values Carrying values of items subject to:

end of 4Q18




Published
financial
statements




Regulatory
scope of
consolidation



Credit
risk
frame-
work

Counter-
party
credit
risk
frame-
work



Securiti-
zation
frame-
work



Market
risk
frame-
work
Not subject
to capital
require-
ments or
subject to
deduction
from capital
Assets (CHF million)   1
Cash and due from banks 100,047 99,827 98,677 262 0 0 899
Interest-bearing deposits with banks 1,142 1,461 1,461 0 0 0 0
Central bank funds sold, securities purchased under resale agreements and securities borrowing transactions 117,095 117,095 42 117,053 0 107,831 0
Securities received as collateral, at fair value 41,696 41,696 0 41,696 0 41,696 0
Trading assets, at fair value 2 132,203 126,936 9,722 51,359 3 1,360 129,595 0
Investment securities 2,911 1,479 1,471 0 8 0 0
Other investments 4,890 4,971 2,114 0 1,068 499 1,290
Net loans 287,581 288,215 267,012 122 20,204 1,213 (192)
Goodwill 4,766 4,770 0 0 0 0 4,770
Other intangible assets 219 219 0 0 0 0 219
Brokerage receivables 38,907 38,907 2,446 28,498 0 0 7,971
Other assets 37,459 36,747 16,819 8,606 1,196 3,617 7,259
Total assets  768,916 762,323 399,764 247,596 23,836 284,451 22,216
Liabilities (CHF million)   1
Due to banks 15,220 16,032 0 0 0 0 16,032
Customer deposits 363,925 363,828 0 0 0 0 363,828
Central bank funds purchased, securities sold under repurchase agreements and securities lending transactions 24,623 30,277 0 30,188 0 23,496 89
Obligation to return securities received as collateral, at fair value 41,696 41,696 0 41,696 0 41,696 0
Trading liabilities, at fair value 2 42,169 42,212 0 16,311 0 59,746 627
Short-term borrowings 21,926 16,536 0 0 0 3,466 13,070
Long-term debt 154,308 152,058 0 0 0 46,685 105,373
Brokerage payables 30,923 30,923 0 23,097 0 0 7,826
Other liabilities 30,107 24,635 394 7,787 0 1,369 15,096
Total liabilities  724,897 718,197 394 119,079 0 176,458 521,941
There are items in the table which attract capital charges according to more than one risk category framework. As an example, derivatives assets/liabilities held in the regulatory trading book are shown in the column about market risk and in the column about counterparty credit risk.
1
Prior period has been corrected.
2
Trading assets/liabilities on the balance sheet reflect the balance after considering netting benefit of cash collateral hence reflect a lower balance than disclosed in the market risk column as cash collateral is not part of the market risk framework.
3
Includes assets pledged as collateral since collateral posted is subject to counterparty credit risk.
For financial reporting purposes, our consolidation principles comply with US GAAP. For capital adequacy reporting purposes, however, entities that are not active in banking and finance are not subject to consolidation (i.e. insurance, commercial and certain real estate companies). Also, FINMA does not require consolidating private equity and other fund type vehicles for capital adequacy reporting. Further differences in consolidation principles between US GAAP and capital adequacy reporting relate to special purpose entities (SPEs) that are consolidated under a control-based approach for US GAAP but are assessed under a risk-based approach for capital adequacy reporting. In addition, FINMA requires us to consolidate companies which form an economic unit with Credit Suisse or if Credit Suisse is obliged to provide compulsory financial support to a company. The investments into such entities, which are not material to the Group, are treated in accordance with the regulatory rules and are either subject to a risk-weighted capital requirement or a deduction from regulatory capital.
All significant equity method investments represent investments in the capital of banking, financial and insurance entities and are subject to a threshold calculation in accordance with the Basel framework and the Swiss Capital Adequacy Ordinance.
> Refer to “Note 40 – Significant subsidiaries and equity method investments” (pages 387 to 390) in VI – Consolidated financial statements – Credit Suisse Group in the Credit Suisse Annual Report 2019 for a list of significant subsidiaries and associated entities.
9

In addition to the differences between accounting and regulatory scopes of consolidation as shown in table LI1 there are further main sources of differences between the financial statements’ carrying value amounts and the exposure amounts used for regulatory purposes.
LI2 - Main sources of differences between regulatory exposure amounts and carrying values in financial statements
   Items subject to:

end of


Credit
risk
frame-
work
Counter-
party
credit
risk
frame-
work
1

Securiti-
zation
frame-
work


Market
risk
frame-
work
4Q19 (CHF million)   
Asset carrying value amount under regulatory scope of consolidation 402,843 245,099 31,535 296,950
Liabilities carrying value amount under regulatory scope of consolidation 430 116,344 0 175,157
Total net amount under regulatory scope of consolidation 402,413 128,755 31,535 121,793
Off-balance sheet amounts 67,994 0 28,901 0
Differences due to application of potential future exposures (SA-CCR) 2 0 4,290 0 0
Differences due to the application of internal models for derivatives (IMM) and SFTs (VaR) 0 (50,941) 0 0
Other differences not classified above 1,056 2,112 (2,914) 0
Exposure amounts considered for regulatory purposes  471,463 84,216 57,522 3
4Q18 (CHF million)   4
Asset carrying value amount under regulatory scope of consolidation 399,764 247,596 23,836 284,451
Liabilities carrying value amount under regulatory scope of consolidation 394 119,079 0 176,458
Total net amount under regulatory scope of consolidation 399,370 128,517 23,836 107,993
Off-balance sheet amounts 67,244 0 26,736 0
Differences due to application of potential future exposures (SA-CCR) 2 0 4,222 0 0
Differences due to the application of internal models for derivatives (IMM) and SFTs (VaR) 0 (51,187) 0 0
Other differences not classified above (297) 976 (2,405) 0
Exposure amounts considered for regulatory purposes  466,317 82,528 48,167 3
The funded portion of the default funds for clearing houses are recorded as a brokerage receivable in accounting. For these positions there is no exposure amount considered for regulatory purposes.
1
Counterparty credit risk includes client cleared exposures, whereas such agency exposures are not reported in the financial statements. Additionally, the column counterparty credit risk and the column market risk take into account the impact of collateral pledges received in SFTs.
2
Calculated under the current exposure method.
3
The concept of “exposure amounts considered for regulatory purposes” is not applicable for market risk as for example for the VaR model.
4
Prior period has been corrected.
> Refer to “Comparison of the standardized and internal model approaches” (pages 19 to 23) in Credit risk – Credit risk under the standardized approach for further information on the origins of differences between carrying values and amounts considered for regulatory purposes shown in the table above.
10

Valuation process
The Basel capital adequacy framework and the Swiss regulation provide guidance for systems and controls, valuation methodologies and valuation adjustments and reserves to provide prudent and reliable valuation estimates.
Financial instruments in the trading book are carried at fair value. The fair value of the majority of these financial instruments is marked to market based on quoted prices in active markets or observable inputs. Additionally, the Group holds financial instruments which are marked to models where the determination of fair values requires subjective assessment and varying degrees of judgment depending on liquidity, concentration, pricing assumptions and the risks affecting the specific instrument.
Control processes are applied to ensure that the reported fair values of the financial instruments, including those derived from pricing models, are appropriate and determined on a reasonable basis. These control processes include approval of new instruments, timely review of profit and loss, risk monitoring, price verification procedures and validation of models used to estimate the fair value. These functions are managed by senior management and personnel with relevant expertise, independent of the trading and investment functions.
In particular, the price verification function is performed by Product Control, independent from the trading and investment functions, reporting directly to the Chief Financial Officer (CFO), a member of the Executive Board.
The valuation process is governed by separate policies and procedures. To arrive at fair values, the following type of valuation adjustments are typically considered and regularly assessed for appropriateness: model, parameter, credit and exit-risk-related adjustments.
Management believes it complies with the relevant valuation guidance and that the estimates and assumptions used in valuation of financial instruments are prudent, reasonable and consistently applied.
> Refer to “Fair valuations” (page 65) in II – Operating and financial review – Credit Suisse – Other information, to “Fair value” (page 101) in II – Operating and financial review – Critical accounting estimates and to “Note 35 – Financial instruments” (pages 347 to 352) in VI – Consolidated financial statements – Credit Suisse Group in the Credit Suisse Annual Report 2019 for further information on fair value.
11

Credit risk
General
This section covers credit risk as defined by the Basel framework. CCR, including those that are in the banking book for regulatory purposes, and all positions subject to the securitization framework are presented in separate sections.
> Refer to “Counterparty credit risk” (pages 44 to 53) for further information on the capital requirements relating to counterparty credit risk.
> Refer to “Securitization” (pages 54 to 61) for further information on the securitization framework.
The Basel framework permits banks to choose between two broad methodologies in calculating their capital requirements for credit risk: the standardized approach or the internal ratings-based (IRB) approach. Off-balance-sheet items are converted into credit exposure equivalents through the use of credit conversion factors (CCF).
The reported credit risk arises from the execution of the Group’s business strategy through the divisions and is predominantly driven by cash and balances with central banks, loans and commitments provided to corporate and institutional clients, loans to private clients including residential mortgages and lending against financial collateral.
Risk management objectives and policies for credit risk
> Refer to “Credit risk” (pages 146 to 149) in III – Treasury, Risk, Balance sheet and Off-balance sheet – Risk management – Risk coverage and management in the Credit Suisse Annual Report 2019 for information on risk management objectives and policies for credit risk, including our credit risk profile, the setting of credit risk limits, the structure and organization of credit risk management.
Credit risk reporting
Credit risk is subject to daily monitoring and reporting, and is governed by internal policies & procedures and a framework of limits and controls. The groups credit risk exposure is subject to formal monthly reporting through the Group Risk Report which provides summary information in relation to the credit risk portfolio composition, rating profile, and the largest single name loans and commitments. The Group Risk Report also provides qualitative commentary on key credit risk matters and developments, and is discussed at Board of Directors Risk Committee and distributed to the Board of Directors and Executive Board members.
Credit quality of assets
The amounts shown in the following tables are the US GAAP carrying values according to the regulatory scope of consolidation that are subject to the credit risk framework.
The following tables present a breakdown of exposures by geographical areas, industry and residual maturity.
CRB - Geographic concentration of gross credit exposures

end of

Switzerland

Americas
Asia
Pacific

EMEA

Total
4Q19 (CHF million)   
Loans and debt securities 207,888 56,330 45,228 86,644 396,090
Off-balance sheet exposures 1 17,842 55,521 5,191 26,024 104,578
Total  225,730 111,851 50,419 112,668 500,668
4Q18 (CHF million)   2
Loans and debt securities 208,570 55,450 40,365 86,844 391,229
Off-balance sheet exposures 1 16,661 54,177 5,584 25,757 102,179
Total  225,231 109,627 45,949 112,601 493,408
The geographic distribution is based on the domicile of the counterparty, shown pre-substitution.
1
Revocable loan commitments, which are excluded from the disclosed exposures, can attract risk-weighted assets.
2
Prior period has been corrected.
12

CRB - Industry concentration of gross credit exposures

end of
Financial
institutions
1
Commercial

Consumer
Public
authorities

Total
4Q19 (CHF million)   
Loans and debt securities 164,034 86,141 140,687 5,228 396,090
Off-balance sheet exposures 2 31,064 72,445 888 181 104,578
Total  195,098 158,586 141,575 5,409 500,668
4Q18 (CHF million)   3
Loans and debt securities 153,666 87,562 137,143 12,858 391,229
Off-balance sheet exposures 2 24,488 76,594 848 249 102,179
Total  178,154 164,156 137,991 13,107 493,408
Exposures are shown pre-substitution.
1
Includes exposures to central banks of CHF 89.1 billion and CHF 81.5 billion as of the end of 4Q19 and 4Q18, respectively.
2
Revocable loan commitments, which are excluded from the disclosed exposures, can attract risk-weighted assets.
3
Prior period has been corrected.
CRB - Remaining contractual maturity of gross credit exposures

end of
within
1 year
1 within
1-5 years

Thereafter

Total
4Q19 (CHF million)   
Loans and debt securities 170,769 169,680 55,641 396,090
Off-balance sheet exposures 2 41,778 56,880 5,920 104,578
Total  212,547 226,560 61,561 500,668
4Q18 (CHF million)   3
Loans and debt securities 166,363 173,674 51,192 391,229
Off-balance sheet exposures 2 41,853 55,641 4,685 102,179
Total  208,216 229,315 55,877 493,408
1
Includes positions without agreed residual contractual maturity.
2
Revocable loan commitments, which are excluded from the disclosed exposures, can attract risk-weighted assets.
3
Prior period has been corrected.
13

The following tables show the amounts of impaired exposures and related allowances and write-offs, broken down by geographical areas and industry.
CRB - Geographic concentration of allowances, impaired loans and write-offs

end of
Allowances
individually
evaluated
for
impairment
Allowances
collectively
evaluated
for
impairment



Total
allowances

Impaired
loans with
specific
allowances
Impaired
loans
without
specific
allowances


Total
impaired
loans


Gross
write-
offs
4Q19 (CHF million)   
Switzerland 511 182 693 1,301 335 1,636 152
EMEA 26 29 55 177 68 245 60
Americas 57 83 140 150 13 163 20
Asia Pacific 14 49 63 87 0 87 75
Total  608 343 951 1,715 416 2,131 307
4Q18 (CHF million)   
Switzerland 475 180 655 1,046 710 1,756 221
EMEA 70 26 96 179 120 299 3
Americas 19 61 80 30 15 45 24
Asia Pacific 44 33 77 98 0 98 32
Total  608 300 908 1,353 845 2,198 280
CRB - Industry concentration of allowances, impaired loans and write-offs

end of
Allowances
individually
evaluated
for
impairment
Allowances
collectively
evaluated
for
impairment



Total
allowances

Impaired
loans with
specific
allowances
Impaired
loans
without
specific
allowances


Total
impaired
loans


Gross
write-
offs
4Q19 (CHF million)   
Financial institutions 37 25 62 48 0 48 0
Commercial 426 272 698 1,059 335 1,394 213
Consumer 145 46 191 608 81 689 94
Total  608 343 951 1,715 416 2,131 307
4Q18 (CHF million)   
Financial institutions 50 29 79 86 0 86 0
Commercial 412 224 636 736 693 1,429 184
Consumer 146 47 193 531 152 683 96
Total  608 300 908 1,353 845 2,198 280
14

The following table presents a comprehensive picture of the credit quality of the Group’s on and off-balance sheet assets.
CR1 – Credit quality of assets

end of

Defaulted
exposures
Non-
defaulted
exposures

Gross
exposures

Allowances/
impairments

Net
exposures
4Q19 (CHF million)   
Loans 1 2,924 381,588 384,512 (951) 383,561
Debt securities 90 11,488 11,578 0 11,578
Off-balance sheet exposures 2 110 104,468 104,578 (191) 104,387
Total  3,124 497,544 500,668 (1,142) 499,526
2Q19 (CHF million)   
Loans 1 2,671 364,340 367,011 (888) 366,123
Debt securities 103 13,046 13,149 0 13,149
Off-balance sheet exposures 2 143 105,517 105,660 (174) 105,486
Total  2,917 482,903 485,820 (1,062) 484,758
1
Loans include all on-balance sheet exposures that give rise to a credit risk charge and exclude debt securities, derivatives, securities financing transactions and off-balance sheet exposures.
2
Revocable loan commitments, which are excluded from the disclosed exposures, can attract risk-weighted assets.
The definitions of “past due” and “impaired” are aligned between accounting and regulatory purposes. However, there are some exemptions for impaired positions related to troubled debt restructurings where the default definition is different for accounting and regulatory purposes.
> Refer to “Note 1 – Summary of significant accounting policies” (pages 271 to 273) and “Note 19 – Loans, allowance for loan losses and credit quality” (pages 289 to 296) in VI – Consolidated financial statements – Credit Suisse Group in the Credit Suisse Annual Report 2019 for further information on the credit quality of loans including past due and impaired loans.
The following table presents the changes in the Group’s defaulted loans, debt securities and off-balance sheet exposures, the flows between non-defaulted and defaulted exposure categories and reductions in the defaulted exposures due to write-offs.
CR2 – Changes in defaulted exposures
2H19
CHF million   
Defaulted exposures at beginning of period  2,917
Exposures that have defaulted since the last reporting period 743
Returned to non-defaulted status (137)
Amounts written-off (131)
Other changes (268)
Defaulted exposures at end of period  3,124
15

The following table shows the aging analysis of accounting past-due exposures.
CRB - Aging analysis of accounting past-due exposures 
   Current Past due

end of

Up to
30 days
31–60
days
61–90
days
More than
90 days

Total

Total
4Q19 (CHF million)   
Financial institutions 15,315 88 1 3 47 139 15,454
Commercial 108,805 642 74 73 728 1,517 110,322
Consumer 157,676 504 83 57 493 1,137 158,813
Public authorities 1,208 26 0 0 0 26 1,234
Gross loans held at amortized cost  283,004 1,260 158 133 1,268 2,819 285,823
Gross loans held at fair value 12,662
Gross loans  298,485
4Q18 (CHF million)   
Financial institutions 12,871 107 19 3 45 174 13,045
Commercial 104,361 461 101 83 861 1,506 105,867
Consumer 153,107 528 65 45 519 1,157 154,264
Public authorities 1,173 13 0 0 0 13 1,186
Gross loans held at amortized cost  271,512 1,109 185 131 1,425 2,850 274,362
Gross loans held at fair value 14,873
Gross loans  289,235
Loans that are modified in a troubled debt restructuring are reported as restructured loans. Generally, restructured loans would have been considered impaired and an associated allowance for loan losses would have been established prior to the restructuring. As of December 31, 2019, CHF 168 million were reported as restructured loans.
> Refer to “Note 19 – Loans, allowance for loan losses and credit quality” (page 296) in VI – Consolidated financial statements – Credit Suisse Group in the Credit Suisse Annual Report 2019 for further information on restructured exposure.
Credit risk mitigation
Credit Suisse actively mitigates credit exposure through the use of legal netting agreements, security over supporting financial and non-financial collateral or financial guarantees and through the use of credit hedging techniques, primarily credit default swaps (CDS). The recognition of credit risk mitigation (CRM) against exposures is governed by a robust set of policies and processes that ensure enforceability and effectiveness.
Netting
> Refer to “Derivative instruments” (pages 166 to 167) in III – Treasury, Risk, Balance sheet and Off-balance sheet – Risk management – Risk portfolio analysis and to “Note 1 – Summary of significant accounting policies” (pages 269 to 270) in VI – Consolidated financial statements – Credit Suisse Group in the Credit Suisse Annual Report 2019 for information on policies and procedures for on- and off-balance sheet netting.
> Refer to “Note 27 – Offsetting of financial assets and financial liabilities” (pages 306 to 309) in VI – Consolidated financial statements – Credit Suisse Group in the Credit Suisse Annual Report 2019 for further information on the offsetting of derivatives, reverse repurchase and repurchase agreements, and securities lending and borrowing transactions.
Collateral valuation and management
The policies and processes for collateral valuation and management are driven by:
a legal document framework that is bilaterally agreed with our clients;
a collateral management risk framework enforcing transparency through self-assessment and management reporting; and
any prevailing regulatory terms which must be complied with.
For exposures collateralized by financial collateral (e.g. marketable securities), collateral valuations are performed on a daily basis and any requirement for additional collateral (e.g. frequency and process for margin calls) is governed by the legal documentation. The market prices used for daily collateral valuation are a combination of internal pricing sources, as well as market prices sourced from trading platforms and external service providers where appropriate.
For exposures collateralized by non-financial collateral (e.g. real estate, ships, aircraft), valuations are performed at the time of credit approval and periodically thereafter depending on the type of collateral and the loan-to-value (LTV) ratio in accordance with documented internal policies and controls. Valuations are based on a combination of internal and external reference price sources.
16

Primary types of collateral
The primary types of collateral are described below.
Collateral securing foreign exchange transactions and over-the-counter (OTC) trading activities primarily includes:
Cash and US Treasury instruments;
G-10 government securities; and
Other assets that are eligible as per the uncleared margin rules (including supranationals and equities).
Collateral securing loan transactions primarily includes:
Financial collateral pledged against loans collateralized by securities of clients of the private, corporate and institutional banking businesses (primarily cash and marketable securities);
Real estate property for mortgages, mainly residential, but also multi-family buildings, offices and commercial properties; and
Other types of lending collateral, such as accounts receivable, inventory, plant and equipment.
Concentrations within risk mitigation
Credit Suisse, primarily through its Global Markets division, is an active participant in the credit derivatives market and trades with a variety of market participants, principally commercial and investment banks. Credit derivatives are primarily used to mitigate investment grade credit exposures. Where required or practicable, these trades are cleared through central counterparties (CCP), reducing the potential risk against individual CRM providers.
As a result of a strong domestic franchise, Credit Suisse has a significant volume of residential mortgage lending in Switzerland and a resultant concentration of residential real estate collateral. Credit Suisse has clear underwriting standards with regard to mortgage lending and ensures that the composition of the real estate portfolio is subject to ongoing monitoring, periodic revaluation, and assessment of the geographical and borrower composition of the portfolio.
Credit Suisse provides loan facilities to private clients against financial collateral such as cash and marketable securities (e.g. equities, bonds, or funds). The financial collateral portfolio within risk mitigation is generally diversified and the portfolio is subject to ongoing monitoring and reporting to identify any concentrations, which may result in lower LTV ratios or other mitigating actions.
> Refer to “Credit risk” (pages 161 to 167) in III – Treasury, Risk, Balance sheet and Off-balance sheet – Risk management – Risk portfolio analysis in the Credit Suisse Annual Report 2019 for further information on credit derivatives, including a breakdown by rating class.
CRM techniques – overview
The following table presents the use of CRM techniques. Credit Suisse recognizes the CRM effect of eligible collateral either as a reduction from the exposure at default (EAD) value of the secured instrument or as an adjustment to the probability of default (PD) or loss given default (LGD) associated with the exposure. All exposures that are secured through eligible collateral are disclosed as “Net exposures partially or fully secured”. Eligible collateral amounts, regardless of which CRM technique has been applied, are disclosed as “Exposures secured by collateral”. Exposures secured by credit derivatives do not include certain immaterial positions, where the credit derivative is recognized with an adjustment to the LGD.
CR3 – CRM techniques
   Net exposures Exposures secured by

end of


Unsecured
Partially
or fully
secured


Total


Collateral

Financial
guarantees

Credit
derivatives
4Q19 (CHF million)      
Loans 1 145,288 238,273 383,561 196,864 7,243 2
Debt securities 11,119 459 11,578 282 0 0
Total  156,407 238,732 395,139 197,146 7,243 2
   of which defaulted  609 1,797 2,406 1,246 175 0
2Q19 (CHF million)   
Loans 1 120,720 245,403 366,123 208,561 5,582 95
Debt securities 12,804 345 13,149 311 0 0
Total  133,524 245,748 379,272 208,872 5,582 95
   of which defaulted  397 1,865 2,262 1,280 158 0
1
Loans include all on-balance sheet exposures that give rise to a credit risk charge and exclude debt securities, derivatives, securities financing transactions and off-balance sheet exposures.
17

Credit risk under the standardized approach
General
Under the standardized approach, risk weights are determined either according to credit ratings provided by recognized external credit assessment institutions (ECAI) or, for unrated exposures, by using the applicable regulatory risk weights.
Credit risk exposure and CRM effects
The following table presents the effect of CRM (comprehensive and simple approach) on the standardized approach capital requirements’ calculations. RWA density provides a synthetic metric on the riskiness of each portfolio.
CR4 – Credit risk exposure and CRM effects
   Exposures pre-CCF and CRM Exposures post-CCF and CRM

end of
On-balance
sheet
Off-balance
sheet

Total
On-balance
sheet
Off-balance
sheet

Total

RWA
RWA
density
4Q19 (CHF million)   
Sovereigns 72,456 24 72,480 72,344 12 72,356 433 1%
Institutions - Banks and securities dealer 1,552 1,492 3,044 1,549 396 1,945 558 29%
Institutions - Other institutions 268 0 268 268 0 268 268 100%
Corporates 7,721 7,615 15,336 7,112 2,558 9,670 7,818 81%
Retail 1,006 139 1,145 1,006 139 1,145 1,021 89%
Other exposures 17,346 2,140 19,486 17,346 1,954 19,300 15,420 80%
   of which non-counterparty related assets  7,942 0 7,942 7,942 0 7,942 7,942 100%
Total  100,349 11,410 111,759 99,625 5,059 104,684 25,518 24%
2Q19 (CHF million)   
Sovereigns 67,825 0 67,825 68,044 5 68,049 806 1%
Institutions - Banks and securities dealer 1,167 618 1,785 1,146 300 1,446 307 21%
Institutions - Other institutions 76 0 76 76 0 76 76 100%
Corporates 7,719 7,196 14,915 7,018 2,228 9,246 7,518 81%
Retail 1,088 152 1,240 1,088 152 1,240 1,071 86%
Other exposures 17,485 1,940 19,425 16,216 1,750 17,966 14,099 78%
   of which non-counterparty related assets  7,815 0 7,815 7,815 0 7,815 7,815 100%
Total  95,360 9,906 105,266 93,588 4,435 98,023 23,877 24%
Exposures by asset class and risk weight
The following table presents the breakdown of credit exposures by asset class and risk weight, which correspond to the riskiness attributed to the exposure according to the standardized approach.
18

CR5 – Exposures by asset class and risk weight
   Risk weight

end of


0%


20%


50%


75%


100%


150%


Others
Exposures
post-CCF
and CRM
4Q19 (CHF million)   
Sovereigns 71,825 26 274 0 112 119 0 72,356
Institutions - Banks and securities dealer 0 1,539 317 0 85 4 0 1,945
Institutions - Other institutions 0 0 0 0 268 0 0 268
Corporates 0 1,222 1,997 0 6,201 250 0 9,670
Retail 0 0 0 494 651 0 0 1,145
Other exposures 3,918 0 0 0 15,370 0 12 19,300
   of which non-counterparty related assets  0 0 0 0 7,942 0 0 7,942
Total  75,743 2,787 2,588 494 22,687 373 12 104,684
   of which past due  0 0 0 0 102 185 0 287
2Q19 (CHF million)   
Sovereigns 66,764 605 304 0 62 314 0 68,049
Institutions - Banks and securities dealer 0 1,395 46 0 5 0 0 1,446
Institutions - Other institutions 0 0 0 0 76 0 0 76
Corporates 0 1,076 1,975 0 5,957 238 0 9,246
Retail 0 0 0 675 565 0 0 1,240
Other exposures 3,890 1 0 0 14,068 0 7 17,966
   of which non-counterparty related assets  0 0 0 0 7,815 0 0 7,815
Total  70,654 3,077 2,325 675 20,733 552 7 98,023
   of which past due  0 0 0 0 48 406 0 454
Comparison of the standardized and internal model approaches
Background
We have regulatory approval to use a number of internal models for calculating our Pillar 1 capital charge for credit risk (default risk). These include the advanced-internal ratings-based (A-IRB) approach for risk weights, Internal Models Method (IMM) for derivatives credit exposure, and repo VaR for securities financing transactions (SFT). These modelled based approaches are used for the vast majority of credit risk exposures, with the standardized approaches used for only a relatively small proportion of credit exposures.
Regulators and investors are increasingly interested in the differences between capital requirements under modelled and standardized approaches. This is due, in part, to ongoing and future regulatory changes by the BCBS, such as the new standardized approaches for counterparty credit risk (SA-CCR) and credit risk as well as the restrictions on the use of internal models for certain portfolios in 2022. As such, FINMA requires us to disclose further information on differences between credit risk RWA computed under internal modelled approaches, and current standardized approaches. FINMA also requires us to disclose the differences between the EAD based on internal modelled approaches and the EAD used in the leverage ratio.
Key methodological differences
The differences between credit risk RWA calculated under the internal modelled approaches and the standardized approaches are driven by the risk weights applied to counterparties and the calculations used for measuring EAD.
Risk weights: Under the A-IRB approach, the maturity of a transaction, and internal estimates of the PD and downturn LGD are used as inputs to the Basel risk-weight formula for calculating RWA. In the standardized approach, risk weights are less granular and are driven by ratings provided by ECAI.
EAD calculations: Under the IMM and repo VaR methods, counterparty exposure is computed using monte-carlo simulation models or VaR models. These models allow for the recognition of netting impacts at exposure and collateral levels for each counterparty portfolio. The standardized approach is based on market values at the balance sheet date plus conservative add-ons to account for potential market movements. This approach gives very limited recognition to netting benefits and portfolio effects.
19

The following table provides a summary of the key conceptual differences between the internal models approach and the current standardized approach.
Key differences between the standardized approach and the internal model approach
Standardized approach Internal model approach Key impact
EAD for
derivatives   
Current Exposure Method is simplistic
(market value and add-on):
replaced with SA-CCR in 2020.
Internal Models Method (IMM)
allows Monte-Carlo simulation to
estimate exposure.
For large diversified derivatives portfolios,
standardized EAD is higher than model EAD.
No differentiation between margined and
unmargined transactions.
Ability to net and offset risk factors within the
portfolio (i.e. diversification).
Impact applies across all asset classes.
Differentiates add-ons by five exposure
types and three maturity buckets only.
Application of multiplier on IMM exposure
estimate.

Limited ability to net.
Variability in holding period applied to collateralized
transactions, reflecting liquidity risks.

Risk
weighting   
Reliance on ECAIs: where no rating is
available a 100% risk weight is applied (i.e. for
most small and medium-size enterprises and funds).
Reliance on internal ratings where each
counterparty/transaction receives a rating.
Model approach produces lower RWA
for high-quality short-term transactions.
Crude risk weight differentiation with 4 key weights:
20%, 50%, 100%, 150% (and 0% for AAA
sovereigns; 35%, 75% or 100% for mortgages;
75% or 100% for retail).
Granular risk sensitive risk weights differentiation
via individual PDs and LGDs.

Standardized approach produces lower RWA
for non-investment grade and long-term
transactions.
No differentiation for transaction features.
LGD captures transaction quality features
incl. collateralization.
Impact relevant across all asset classes.
Application of a 1.06 scaling factor.
Risk
mitigation   
Limited recognition of risk mitigation.

Risk mitigation recognized via
risk sensitive LGD or EAD.
Standardized approach RWA
higher than model approach RWA
for most collaterals.
Restricted list of eligible collateral.
Wider variety of collateral types eligible.
Impact particularly relevant for lombard
lending and SFTs.
Conservative and crude regulatory haircuts.


Repo VaR allows use of VaR models to
estimate exposure and collateral for SFTs.
Approach permits full diversification
and netting across all collateral types.



Maturity
in risk
weight   
No differentiation for maturity of transactions,
except for interbank exposures in a coarse
manner.
No internal modelling of maturity.

Model approach produces lower RWA
for high-quality short-term transactions.



Regulatory RWA function considers
maturity: the longer the maturity
the higher the risk weight
(see chart "Risk weight by maturity").



The following chart shows standardized risk weights, and model based (A-IRB) risk weights for loans of varying maturity. The graphs are plotted for a AA-rated corporate senior unsecured loan with a LGD of 45% (consistent with Foundation-IRB, F-IRB), and a AA-rated corporate senior secured loan with a LGD of 36%. The graphs show that standardized risk weights are not sensitive to maturity, whereas A-IRB risk weights are sensitive to maturity. In particular, under A-IRB, lower maturity loans receive lower risk weights reflecting an increased likelihood of repayment for loans with a shorter maturity.
20

Key methodological differences between internally modelled EAD and EAD used in leverage ratio
The exposure measure used in the leverage ratio also differs from the exposure measure used in the internal modelled approach. The main methodological difference is that leverage ratio exposure estimates do not take into account physical or financial collateral, guarantees or other CRM techniques to reduce the credit risk. Leverage ratio exposures also do not fully reflect netting and portfolio diversification. As a result, leverage ratio exposures are typically larger than model based exposures.
The following table shows the internal model-based EAD, along with average risk weight, compared to an estimate of the exposure measure used in the leverage ratio calculation. Estimates are provided at Basel asset class level. As expected, leverage exposure measures exceed internal model-based EAD for banks and corporates where the impacts of netting, diversification and CRM are large.
Leverage exposure estimate
   Internal model approach

EAD
Risk
weight
Leverage
exposures
1
Basel asset class (CHF billion, except where indicated)   
Corporates 173 53% 320
Banks 27 29% 59
Sovereigns 26 7% 26
Retail 201 16% 200
1
The leverage exposure estimates only consider those exposures which are comparable to the credit risk RWA calculation under internal model approach and hence excludes exposures such as trading book, securitization and non-credit exposures. Asset class leverage ratio based exposures are approximate and provided on a best efforts basis.
It should be noted that credit risk capital requirements based on the internal model based approach are not directly comparable to capital requirements under the leverage ratio. The reason for this is that the 3% leverage ratio capital requirement can be met with total tier 1 capital, including capital for market risk and operational risk.
Risk-weighted assets under the standardized and internal model approaches
Credit risk RWA computed under the standardized approach are higher than those based on the internal models for which we have received regulatory approval. Higher risk-weights under the standardized approach rules are a material driver of the higher RWA for all Basel asset classes. The standardized exposure calculations also lead to some higher RWA, with the corporate and bank asset classes being most significantly affected.
Corporate asset class
The table “Leverage exposure estimate” shows that the EAD for corporates computed under the internal model approach is CHF 173 billion. The EAD for corporates under the standardized approach is significantly higher. This difference is driven mainly by the standardized exposure calculations for OTC derivatives and secured financing transactions. For these products, exposures calculated under the standardized approach are higher than the model based exposures because the standardized approach does not fully recognize the benefits of netting, portfolio diversification and collateral. The exposure calculated under the leverage ratio is higher than the EAD computed using internal models. This is because CRM, netting and portfolio diversification are not reflected in the leverage ratio exposure calculation.
Another significant driver of the increase in credit risk RWA under the standardized approach is higher risk weights. The exposure weighted-average risk weight under the internal model approach is 53%. This is significantly lower than the risk weights assigned to corporates under the standardized approach.
The following graph shows the risk weights assigned to counterparties under the A-IRB approach and the standardized approach. For the IRB risk weight curve, an LGD value of 45% and a maturity adjustment of 2.5 years are chosen, as these are the Basel Foundation IRB parameters. For counterparties in the AAA to BB+ range (based on external ratings), higher risk weights (20%, 50% and 100%) are assigned under the standardized approach than under the A-IRB approach. For the corporate asset class, approximately three-quarters of the Group’s exposures are in this range (based on internal ratings), and this is a key driver for the higher RWA under the standardized approach. The different treatments of loan maturity in the model based approach and standardized approach are not a material cause of RWA differences.
The Group’s exposure weighted-average maturity of its corporate portfolio is lower than the foundation IRB value of 2.5 years, and lower maturities would result in a lower model-based risk weight curve than shown in the graph. In addition, the PD for each rating shown in the graph are consistent with the Group’s PD masterscale.
21

An additional driver of higher risk weights within the corporate asset class are counterparties without an external rating. Under the standardized approach, counterparties without an external rating receive a fixed risk weight of 100%. This applies to a large proportion of the Group’s exposures, among them non-banking financial institutions and specialized lending. This fixed standardized risk weight is typically higher than the model based risk weight with for example, the average model based risk weight of specialized lending being approximately 40%.
> Refer to “CR6 – Credit exposures by portfolio and PD range” (pages 28 to 35) for further information on EAD and risk weights for each credit rating for the corporate asset class.
Bank asset class
The table “Leverage exposure estimate” shows that the EAD for banks under the internal model approach is CHF 27 billion. The EAD for banks calculated under the standardized approach is significantly higher. This is driven predominantly by the exposure calculations for both OTC derivatives and secured financing transactions and, to a lesser extent, the exposure calculations for listed and centrally cleared derivatives. For these products, exposures calculated under the standardized approach are much higher than the model based exposures because the standardized approach does not fully recognize the benefits of netting, portfolio diversification and collateral. The exposures calculated under the leverage ratio are significantly higher than the EAD computed using internal models. This is because CRM, netting and portfolio diversification are not reflected in the leverage ratio exposure calculation.
In addition, there is a significant increase in credit risk RWA under the standardized approach due to higher credit risk-weights. The exposure weighted-average risk-weight under the internal model approach is 29%. This is significantly lower than the risk weights assigned to banks under the standardized approach where a significant amount of the Group’s exposures would attract a risk weight of 50%.
The following graph shows the risk weights assigned to counterparties under the A-IRB approach and the standardized approach. For the IRB risk weight curve, an LGD value of 45% and a maturity adjustment of 2.5 years are chosen, as these are the Basel Foundation IRB parameters. The graph shows that counterparties in the AAA to BBB+ range (based on external ratings) attract higher risk weights (20% and 50%) under the standardized approach than under the A-IRB approach. In excess of three-quarters of the Group’s exposures fall in this range (based on internal ratings) and this leads to higher RWA under the standardized approach for these counterparties. The different treatments of loan maturity in the model based approach and standardized approach are not a material cause of RWA differences.
> Refer to “CR6 – Credit exposures by portfolio and PD range” (pages 28 to 35) for further information on EAD and risk weights for each credit rating for the bank asset class.
The Group’s exposure weighted-average maturity of its bank portfolio is lower than the foundation IRB value of 2.5 years, and lower maturities would result in a lower model based risk weight curve than shown in the graph. In addition, the PD for each rating shown in the graph are consistent with the Group’s PD masterscale.
Sovereign asset class
The table “Leverage exposure estimate” shows that the EAD for sovereigns under the internal model approach is CHF 26 billion. This is comparable to the EAD calculated under the standardized approach and the leverage ratio exposure. This is because the majority of the sovereign exposure is in the form of uncollateralized loans, i.e. there are no material differences in the exposure calculation.
The impact of employing standardized credit risk weights to the sovereign portfolio is an overall increase in credit risk RWA. The exposure weighted-average risk weight under the internal model approach is less than 7%. This is lower than the risk weights assigned to counterparties under the standardized approach.
The following graph shows the risk weights assigned to counterparties under the A-IRB approach and the standardized approach. For the IRB risk weight curve, an LGD value of 45% and a maturity adjustment of 2.5 years are chosen, as these are the Basel Foundation IRB parameters. The graph shows that counterparties in the AAA to A range (based on external ratings) would attract lower risk weights (0% and 20%) under the standardized approach than under the A-IRB approach. The majority of the Group’s exposures have extremely low risk-weights under the A-IRB approach and would attract risk weights of 0% under the standardized approach. The remaining exposures would receive higher risk weights under the standardized approach (20%, 50% or 100%) than under the A-IRB approach. Overall, this would lead to higher RWA under the standardized approach. The different treatments of loan maturity in the model based approach and standardized approach are not a material cause of RWA differences.
> Refer to “CR6 – Credit exposures by portfolio and PD range” (pages 28 to 35) for further information on EAD and risk weights for each credit rating for the sovereign asset class.
22

The Group’s exposure weighted-average maturity of its sovereign portfolio is lower than the foundation IRB value of 2.5 years, and lower maturities would result in a lower model-based risk weight curve than shown in the following graph. In addition, the PD for each rating shown in the graph are consistent with the Group’s PD masterscale.
Retail asset class
The EAD of the retail asset class under the internal model approach is CHF 201 billion, which is comparable to the EAD calculated under the standardized approach and the leverage ratio. This is because the majority of retail exposure is on-balance sheet exposure.
The application of the standardized approach would lead to higher credit risk RWA. The exposure weighted-average risk weight is 16% using internal model approach. This is lower than the risk weights assigned to counterparties under the standardized approach. The maturity of the loan has no impact on the modelled risk weights in the retail asset class.
The retail portfolio consists mainly of residential mortgage loans, lombard lending and other retail exposures, and further analysis for each of these portfolios is provided below:
Residential mortgages: Under the standardized approach, fixed risk weights are applied depending on the LTV, i.e. risk weight of 100% for LTV > 80%, risk weight of 75% for 80% > LTV > 67% and risk weight of 35% for LTV < 67%. The internal model-based approach however takes into account borrowers’ ability to service debt more accurately, including mortgage affordability and calibration to large amounts of historic data. The Group’s residential mortgage portfolio is focused on the Swiss market and the Group has robust review processes over borrowers’ ability to repay. This results in the Group’s residential mortgage portfolio having a low average LTV and results in an average risk weight of 18% under the A-IRB approach.
Lombard lending: For lombard lending, the average risk weight using internal models is 11%. RWA under the standardized approach would be higher for these exposures.
Other retail exposures: Other retail exposures are risk-weighted at 75% or 100% under the standardized approach. This yields higher RWA compared to the A-IRB approach where the average risk-weight is 39%.
Conclusion
Overall, the Group’s credit risk RWA would be significantly higher under the standardized approach than under the internal model based approach. For most Basel asset classes, this is due to standardized risk weights being much higher than the IRB risk weights for high quality investment grade lending, which is where the majority of the Group’s exposures are. For certain asset classes, standardized exposure calculations also lead to significantly higher RWA. This is where the standardized exposure methods give limited recognition to economic offsetting and diversification for derivatives and SFTs at a portfolio level.
The credit risk RWA under the standardized approaches described above is not reflective of the capital charges under the new standardized approach for credit risk on which the BCBS published new rules in December 2017. This new standardized approach for credit risk is more risk sensitive and employs a different approach for incorporating external ratings. In addition, there is a new standardized approach for counterparty credit risk (SA-CCR), which prescribes a standardized calculation of EAD for derivative transactions. SA-CCR, which was implemented in January 2020, will more accurately recognize the risk mitigating effect of collateral and the benefits from legal and economic offsetting. These regulatory changes could potentially lead to very different results to the ones described above.
The credit risk RWA computed under the internal model-based approach provide a more risk-sensitive indication of the credit risk capital requirements and are more reflective of the economic risk of the Group. The use of models produces a strong link between capital requirements and business drivers, and promotes a proactive risk culture at the origination of a transaction and strong capital consciousness within the organization. A rigorous monitoring and control framework also ensures compliance with internal as well as regulatory standards.
23

Credit risk under internal ratings-based approaches
General
Under the IRB approach, risk weights are determined by using internal risk parameters and applying an asset value correlation multiplier uplift where exposures are to financial institutions meeting regulatory defined criteria. We have received approval from FINMA to use, and have fully implemented, the A-IRB approach whereby we provide our own estimates for PD, LGD and EAD.
PD parameters capture the risk of a counterparty defaulting over a one-year time horizon. PD estimates are mainly derived from models tailored to the specific business of the respective obligor. The models are calibrated to the long run average of annual internal or external default rates where applicable. For portfolios with a small number of empirical defaults, low default portfolio techniques are used.
LGD parameters consider seniority, collateral, counterparty industry and in certain cases fair value markdowns. LGD estimates are mainly based on an empirical analysis of historical loss rates. To reflect time value of money, recovered amounts on defaulted obligations are discounted to the time of default and to account for potential adverse outcomes in a downturn environment, final parameters are chosen such as they reflect periods where economic downturns have been observed and/or where increased losses manifested. For portfolios with low amount of statistical values available conservative values are chosen based on proxy analysis and expert judgement. For much of the private, corporate and institutional banking businesses loan portfolio, the LGD is primarily dependent upon the type and amount of collateral pledged. The credit approval and collateral monitoring process are based on LTV limits. For mortgages (residential or commercial), recovery rates are differentiated by type of property.
EAD is either derived from balance sheet values or by using models. EAD for a non-defaulted facility is an estimate of the expected exposure upon default of the obligor. Estimates are derived based on a CCF approach using default-weighted averages of historical realized conversion factors on defaulted loans by facility type. Estimates are calibrated to capture negative operating environment effects. To comply with regulatory guidance in deriving individual observed CCF values as basis for the estimation are floored at zero, i.e. it is assumed that drawn exposure can never become lower in the run to default.
> Refer to “Credit risk” (pages 146 to 149) in III – Treasury, Risk, Balance sheet and Off-balance sheet – Risk management – Risk coverage and management in the Credit Suisse Annual Report 2019 for further information on PD and LGD.
Risk weights are calculated using either the PD/LGD approach or the supervisory risk weights approach for certain types of specialized lending.
Reporting related to credit risk models
> Refer to “Model validation” (pages 25 to 26), “Use of internal ratings” (page 27) and “Credit Risk Review” (page 27) for further information on the scope and main content of the reporting related to credit risk models.
Rating models
The majority of the credit rating models used in Credit Suisse are developed internally by Core Credit Models, a specialized unit within the Quantitative Analysis and Technology area in the risk organization. These models are independently validated by Model Risk Management team prior to use in the Basel III regulatory capital calculation, and thereafter on a regular basis. Credit Suisse also uses models purchased from recognized data and model providers (e.g. credit rating agencies). These models are owned by Core Credit Models and are validated internally following the same governance process as models developed internally.
All new or material changes to rating models are subject to a robust governance process. Post development and validation of a rating model or model change, the model is taken through a number of committees where model developers, validators and users of the models discuss the technical and regulatory aspects of the model. The relevant committees opine on the information provided and decide to either approve or reject the model or model change. The ultimate decision making committee is the Risk Processes & Standards Committee (RPSC). The responsible Executive Board Member for the RPSC is the Chief Risk Officer (CRO). The RPSC sub-group responsible for credit risk models is the Credit Methodology Steering Committee (CMSC). RPSC or CMSC also review and monitor the continued use of existing models on an annual basis.
The following table provides an overview of the main PD and LGD models used by Credit Suisse. It reflects the portfolio segmentation from a credit risk model point of view, showing the RWA, type and number of the most significant models, and the loss period available for model development by portfolio. As the table follows an internal risk segmentation and captures the most significant models only, these figures do not match regulatory asset class or other A-IRB based segmentation.
Some of the portfolios shown in the table sum up multiple rating models. The distinction criteria determining which model applies, differs from portfolio to portfolio. Corporates, banks and non-banking financial institutions are split by turnover and geography. For funds, the distinction criteria is the different form of funds e.g. mutual-, hedge-funds etc., whereas for income producing real estate (IPRE), it is corporate vs. private counterparties. The distinction criteria for Sovereign is global governments vs. Swiss Canton vs. local governments (e.g. cities).
24

CRE - Main PD and LGD models used by Credit Suisse
   PD    LGD   

Portfolio


Asset class

RWA (in
CHF billion)
Number
of years
loss data

No. of
models


Model comment

No. of
models


Model comment
Statistical and hybrid models using e.g. industry and counterparty segmentation, collateral types and amounts, seniority and other transaction specific factors with granularity enhancements by public research and expert judgement
Corporates Corporates, retail 60 >15 years 2 Statistical scorecards using e.g. balance sheet, P&L data and qualitative factors 3
Banks and other financial institutions Banks, corporates 10 >30 years 5 Statistical scorecard and constrained expert judgement using e.g. balance sheet, P&L data and qualitative factors
Funds Corporates

11

>10 years

4

Statistical scorecards using e.g. net
asset value, volatility of returns and
qualitative factors


Statistical model using e.g. counterparty segmentation, collateral types and amounts
Residential mortgages Retail 13 >10 years 1 Statistical scorecard using e.g. LTV, affordability, assets and qualitative factors 1
Income producing real estate Specialized lending, retail 20 >10 years 2 Statistical scorecards using e.g. LTV, debt service coverage and qualitative factors
Commodity
traders
Corporates,
specialized lending
2

>10 years

1

Statistical scorecard using e.g.
volume, liquidity and duration of
financed commodity transactions


Sovereign Sovereign,
corporates

2


>10 years


1


Statistical scorecards using e.g.
GDP, financials and qualitative
factors
1


Statistical models using e.g. industry
and counterparty segmentation,
seniority and other transaction
specific factors
Ship
finance
Specialized
lending

3


>10 years


1


Statistical scorecard using e.g.
freight rates, ship market values,
operational expenses and group
information
1


Statistical model using e.g. LTV
and counterparty attributes

Lombard,
Securities
Borrowing &
Lending
Retail


14


>10 years


1


Merton type model using e.g.
LTV, collateral volatility and
counterparty attributes
1


Merton type model using e.g.
LTV, collateral volatility and
counterparty attributes
Model development
The techniques to develop models are carefully selected by Core Credit Models to meet industry standards in the banking industry as well as regulatory requirements. The models are developed to exhibit “through-the-cycle” characteristics, reflecting a PD in a 12 month period across the credit cycle.
All models have clearly defined model owners who have primary responsibility for development, enhancement, review, maintenance and documentation. The models have to pass statistical performance tests, where feasible, followed by usability tests by designated Credit Risk Management experts to proceed to formal approval and implementation. The development process of a new model is thoroughly documented and foresees a separate schedule for model updates.
The level of calibration of the models is based on a range of inputs, including internal and external benchmarks where available. Additionally, the calibration process ensures that the estimated calibration level accounts for variations of default rates through the economic cycle and that the underlying data contains a representative mix of economic states. Conservatism is incorporated in the model development process to compensate for any known or suspected limitations and uncertainties.
Model validation
Model validation for risk capital models is performed by the Model Risk Management function. Model governance is subject to clear and objective internal standards as outlined in the Model Risk Management policy and the Model Validation Policy. The governance framework ensures a consistent and meaningful approach for the validation of models in scope across the bank. All models whose outputs fall into the scope of the Basel internal model framework are subject to full independent validation. Externally developed models are subject to the same governance and validation standards as internal models.
The governance process requires each in scope model to be validated and approved before go-live; the same process is followed for material changes to an existing model. Existing models are subject to an ongoing governance process which requires each model to be periodically validated and the performance to be monitored annually. The validation process is a comprehensive quantitative and qualitative assessment with goals that include:
to confirm that the model remains conceptually sound and the model design is suitable for its intended purpose;
to verify that the assumptions are still valid and weaknesses and limitations are known and mitigated;
to determine that the model outputs are accurate compared to realized outcome;
25

to establish whether the model is accepted by the users and used as intended with appropriate data governance;
to check whether a model is implemented correctly;
to ensure that the model is fully transparent and sufficiently documented.
To meet these goals, models are validated against a series of quantitative and qualitative criteria. Quantitative analyses may include a review of model performance (comparison of model output against realized outcome), calibration accuracy against the longest time series available, assessment of a model’s ability to rank order risk and performance against available benchmarks. Qualitative assessment typically includes a review of the appropriateness of the key model assumptions, the identification of the model limitations and their mitigation, and ensuring appropriate model use. The modeling approach is re-assessed in light of developments in the academic literature and industry practice.
Results and conclusions are presented to senior risk management including the RPSC; shortcomings and required improvements identified during validation must be remediated within an agreed deadline. The Model Risk Management function is independent of model developers and users and has the final say on the content of each validation report.
Model governance at Credit Suisse follows the “three lines of defense” principle. Model developers and owners provide the first line of defense, Model Risk Management the second line, and Internal Audit the third line of defense. Organization independence ensures that these functions are able to provide appropriate oversight. For Credit Risk models, the development and validation functions are independent up to the CRO (Executive Board level). Internal Audit has fully independent reporting into the Chair of the Board of Directors Audit Committee.
Stress testing of parameters
The potential biases in PD estimates in unusual market conditions are accounted for by the use of long run average estimates. Credit Suisse additionally uses stress-testing when back-testing PD models. When predefined thresholds are breached during back-testing, a review of the calibration level is undertaken. For LGD/CCF calibration stress testing is applied in defining Downturn LGD/CCF values, reflecting potentially increased losses during stressed periods.
Descriptions of the rating processes
All counterparties that Credit Suisse is exposed to are assigned an internal credit rating. The rating is assigned at the time of initial credit approval and subsequently reviewed and updated regularly. Where available, Credit Risk Management employs rating models relative to the counterparty type that incorporate qualitative and quantitative factors. Expert judgement may further be applied through a well governed model override process in the assignment of a credit rating or PD, which measures the counterparty’s risk of default over a one-year period.
Corporates (excluding corporates managed on the Swiss platform), banks and sovereigns (primarily in the investment banking businesses)
Where used, rating models are an integral part of the rating process. To ensure all relevant information is considered when rating a counterparty, experienced credit officers complement the outputs from the models with other relevant information not otherwise captured via a robust model-override framework. Other relevant information may include, but is not limited to peer analysis, industry comparisons, external ratings and research and the judgment of credit experts. This analysis emphasizes a forward looking approach, concentrating on economic trends and financial fundamentals. Where rating models are not used the assignment of credit ratings is based on a well-established expert judgment based process which captures key factors specific to the type of counterparty.
For structured and asset finance deals, the approach is more quantitative. The focus is on the performance of the underlying assets, which represent the collateral of the deal. The ultimate rating is dependent upon the expected performance of the underlying assets and the level of credit enhancement of the specific transaction. Additionally, a review of the originator and/or servicer is performed. External ratings and research (rating agency and/or fixed income and equity), where available, are incorporated into the rating justification, as is any available market information (e.g., bond spreads, equity performance).
Transaction ratings are based on the analysis and evaluation of both quantitative and qualitative factors. The specific factors analyzed include seniority, industry and collateral.
Corporates managed on the Swiss platform, mortgages and other retail (primarily in the private, corporate and institutional banking businesses)
For corporates managed on the Swiss platform and mortgage lending, the PD is calculated directly by proprietary statistical rating models, which are based on internally compiled data comprising both quantitative factors (primarily LTV ratio and the borrower’s income level for mortgage lending and balance sheet information for corporates) and qualitative factors (e.g., credit histories from credit reporting bureaus, management quality). In this case, an equivalent rating is assigned for reporting purposes, based on the PD band associated with each rating. Collateral loans (margin lending), which form the largest part of “Other retail”, is also following an individual PD and LGD approach. This approach is already rolled out for loans booked on the Swiss platform and for the majority of international locations; the remaining international locations follow a pool PD and pool LGD approach. Both approaches are calibrated to historical loss experience. Most of the collateral loans are loans collateralized by securities.
The internal rating grades are mapped to the Credit Suisse Internal Masterscale. The PDs assigned to each rating grade are reflected in the following table.
26

CRE - Credit Suisse counterparty ratings
Ratings PD bands (%) Definition S&P Fitch Moody's Details
AAA 0.000 - 0.021
Substantially
risk free
AAA
AAA
Aaa
Extremely low risk, very high long-term
stability, still solvent under extreme conditions
AA+
AA
AA-
0.021 - 0.027
0.027 - 0.034
0.034 - 0.044
Minimal risk

AA+
AA
AA-
AA+
AA
AA-
Aa1
Aa2
Aa3
Very low risk, long-term stability, repayment
sources sufficient under lasting adverse
conditions, extremely high medium-term stability
A+
A
A-
0.044 - 0.056
0.056 - 0.068
0.068 - 0.097
Modest risk


A+
A
A-
A+
A
A-
A1
A2
A3
Low risk, short- and mid-term stability, small adverse
developments can be absorbed long term, short- and
mid-term solvency preserved in the event of serious
difficulties
BBB+
BBB
BBB-
0.097 - 0.167
0.167 - 0.285
0.285 - 0.487
Average risk

BBB+
BBB
BBB-
BBB+
BBB
BBB-
Baa1
Baa2
Baa3
Medium to low risk, high short-term stability, adequate
substance for medium-term survival, very stable short
term
BB+
BB
BB-
0.487 - 0.839
0.839 - 1.442
1.442 - 2.478
Acceptable risk


BB+
BB
BB-
BB+
BB
BB-
Ba1
Ba2
Ba3
Medium risk, only short-term stability, only capable of
absorbing minor adverse developments in the medium term,
stable in the short term, no increased credit risks expected
within the year
B+
B
B-
2.478 - 4.259
4.259 - 7.311
7.311 - 12.550
High risk

B+
B
B-
B+
B
B-
B1
B2
B3
Increasing risk, limited capability to absorb
further unexpected negative developments
CCC+
CCC
CCC-
CC
12.550 - 21.543
21.543 - 100.00
21.543 - 100.00
21.543 - 100.00
Very high
risk

CCC+
CCC
CCC-
CC
CCC+
CCC
CCC-
CC
Caa1
Caa2
Caa3
Ca
High risk, very limited capability to absorb
further unexpected negative developments

C
D1
D2
100
Risk of default
has materialized
Imminent or
actual loss

C
D

C
D

C


Substantial credit risk has materialized, i.e. counterparty
is distressed and/or non-performing. Adequate specific
provisions must be made as further adverse developments
will result directly in credit losses.
Transactions rated C are potential problem loans; those rated D1 are non-performing assets and those rated D2 are non-interest earning.
Use of internal ratings
Internal ratings play an essential role in the decision-making and the credit approval processes. The portfolio credit quality is set in terms of the proportion of investment and non-investment grade exposures. Investment/non-investment grade is determined by the internal rating assigned to a counterparty.
Internal counterparty ratings (and associated PDs), transaction ratings (and associated LGDs) and CCF for loan commitments are inputs to RWA and ERC calculations. Model outputs are the basis for risk-adjusted-pricing or assignment of credit competency levels.
The internal ratings are also integrated into the risk management reporting infrastructure and are reviewed in senior risk management committees. These committees include the RPSC and the Capital Allocation & Risk Management Committee (CARMC).
Credit Risk Review
Governance and supervisory checks within credit risk management are supplemented by the credit risk review function. The credit risk review function is independent from credit risk management with a direct functional reporting line to the Risk Committee Chair, administratively reporting to the Group CRO. Credit risk review’s primary responsibility is to provide timely and independent assessments of the Group’s credit exposures and credit risk management processes and practices. Any findings and agreed actions are reported to senior management and, as necessary, to the Risk Committee.
EAD covered by the various approaches
The following table shows the part of EAD covered by the standardized and the A-IRB approach for each of the asset classes. The F-IRB approach is currently not applied.
CRE - EAD covered by the various approaches

end of 4Q19
Standardized
approach
A-IRB
approach
EAD (in %)   
Sovereigns 75 25
Institutions - Banks and securities dealer 13 87
Institutions - Other institutions 17 83
Corporates 7 93
Residential mortgages 0 100
Retail 1 99
Other exposures 100 0
Total  23 77
27

Credit risk exposures by portfolio and PD range
The following table presents the main parameters used for the calculation of capital requirements for IRB models.
CR6 – Credit risk exposures by portfolio and PD range

end of 4Q19
Original
on-balance
sheet gross exposure
Off-balance
sheet exposures
pre CCF

Total
exposures

Average
CCF
EAD post-
CRM and
post-CCF
1
Average
PD
Number of
obligors
(thousands)

Average
LGD
Average
maturity
(years)


RWA
2
RWA
density

Expected
loss


Provisions
Sovereigns (CHF million, except where indicated)   
0.00% to <0.15% 22,619 351 22,970 94% 22,905 0.03% < 0.1 8% 1.3 467 2% 0
0.15% to <0.25% 68 85 153 100% 140 0.22% < 0.1 47% 2.8 77 55% 0
0.25% to <0.50% 71 0 71 71 0.37% < 0.1 53% 1.5 44 61% 0
0.50% to <0.75% 49 0 49 49 0.64% < 0.1 42% 4.9 52 105% 0
0.75% to <2.50% 63 3 66 45% 64 1.83% < 0.1 53% 1.6 78 123% 1
2.50% to <10.00% 1,067 0 1,067 45% 268 6.31% < 0.1 51% 2.6 537 201% 9
10.00% to <100.00% 20 0 20 0% 20 16.44% < 0.1 52% 2.5 56 276% 2
100.00% (Default) 258 0 258 17 100.00% < 0.1 58% 1.2 18 106% 0
Sub-total  24,215 439