CREDIT STRESS TESTING FRAMEWORK USES & CHALLENGES TO BANK - - PowerPoint PPT Presentation
CREDIT STRESS TESTING FRAMEWORK USES & CHALLENGES TO BANK - - PowerPoint PPT Presentation
CREDIT STRESS TESTING FRAMEWORK USES & CHALLENGES TO BANK MANAGEMENT USES & CHALLENGES TO BANK MANAGEMENT Jrgen Wienes Richard Vasicek Group Risk Management Enterprise Risk Solutions UniCredit Group UniCredit Group Moody's
Maintainance Of Sustainability & Licence For An Institution: Multiple Factors Demand Sophistacted Usage Of Stress Testing
Regulatory Compliance? Systemic Contagion? g y p Solvency Systemic Contagion? Source and Victim
Stress Test Analytics Consistent Analytics
Resilience To Downturn? L Ab ti C it Investor Confidence? Loss Absorption Capacity Cost of Capital and Funding
2
Variety and Pace of New Regulations for Banking System Many are directed at bank capitalization with need to stress test.
3 Source: McKinsey, BCBS; Dodd-Frank Act; CRD2/3/4; EMIR
Key Challenges for Stress Testing To Maximize Its Utilization
- 1. Managing complexity at Senior Management level deriving from different frameworks, purposes and methodologies.
Therefore design, interpretation of results and managerial actions need to reflect such choices.
- 2. Firm-Wide Stress Testing: As “Silo”-approach may lead to substantial under-estimation of stress impacts, Enterprise
Risk Management captures Risk Types, Balance Sheet and Liquidity together.
- 3. Support the Bank’s planning process, Stress Testing is considered an important tools to assess the robustness.
- 4. Applicability of Platform: To become useful, key drivers and model parameters have to be understood and
pp y y p
- incorporated. Allow Reverse Stress Testing to capture systemic and idiosyncratic vulnerabilities undermining the
Business model viability
- 5. Contingency Plans: correctly identifying triggers and relevant actions for facing the potential consequences
T STRESS TEST K IS A MUST TEGRATED S RAMEWORK 4 INT F
- 1. Managing Complexity
Regulatory Requests For Stress Testing Has Been Growing In Importance and Numbers Over The Few Years p
A third of the top 1000 banks now have to comply with three or more stress The regulatory pressure has increased substantially since 2009 with the emergence
2nd EBA
Global and local stress testing specific regulations
y testing requirements per year*
Regulatory requirements have increased, both in
y g
- f global and local regulations
- 14
CEBS Guidelines
- First EBA
stress test 2 EBA stress test
number and complexity
»
More requests: EBA, IMF, Basel, Local
»
More stringent: Bank-wide and bank-specific, Reverse
rk Key To lexIty
8 14 11 8 10 12
requirements + Basel II Pillar 2 CEBS BIS Principles i.e. Germany
»
More frequent: quarterly to annually
»
More transparent: market confidence
- bjective/disclosure
Framewo ve" Comp
1 3 5 8 4 6 8
- Complexity of
CEBS Guidelines i.e. UK, Australia, Sweden Stress testing & Liquidity Risk Germany
EBA is a major annual exercise in Europe
»
Top 91 banks, covering 65% of the European banking assets and at least 50% of all banking assets in each EU country
ntegrated "Surviv
Source: Market research and own analysis using a sample of 50 countries including G20, EU and BCBS members
2 2005 2006 2007 2008 2009 2010 2011 2012
assets in each EU country
»
Expanded to additional banks by local regulators
»
Comprehensive and complex exercise
In
5
Number of country-specific regulations or guidelines issued for the given year * 3+ Stress testing requirements include: IMF, EBA, local regulation, CEBS guidelines and Basel 3 guidelines Source: Moody's Analytics own estimate based on EBA data, regulatory websites and market research.
- 1. Managing Complexity
Different types and usages of stress testing require a clear strategy for communicating results to Senior Management g g
Types of Stress Testing Usage of Stress Testing Output Recipients of Stress Testing and actions deployment Major condition for successfully
- Sensitivity Analysis
- Scenario Analysis
- Regulatory:
- ICAAP
- European Stress Test
- Capital
Adequacy Assessment Supervisory Authority successfully integrating Stress Test in Bank’s management
- Reverse Stress
Testing
- SIFI (Systemically
Important Financial Institutions)
- Other national
- Robustness of
Business Strategies Top Management Need to integrate Other national regulators’ requests
- Managerial:
- Risk Appetite
Strategies
- Contingency
Plans
- Managerial
different results to form a unique view
- n a Bank’s
capital and
- Risk Appetite
- On-going Stress Test
Focus on specific sub-portfolios’ Actions
- Contingency
Plans Deployment business strategy soundness
Stress Testing is performed according to different design along intended usages (Regulatory and
Managerial). All results need to be successfully and clearly communicated to a Senior Management.
vulnerabilities Deployment 6
To improve the communication, all the relevant stakeholder (Top Management and Business) need to
be involved in the process, starting from the definition of the scenarios down to the sharing of the
- utput and final decision process.
- 2. Firm-Wide Stress Testing:
From Risk-wide Stress Test to Enterprise Risk Management Understanding and Capturing the Linkages g p g g
- Loans and Receivables with
Banks
- Loans and Receivables with
ASSETS
Ri k Wid
- Credit Risk
- Market Risk
Risk Map
LIABILITIES
- Deposit From Banks
- Deposits From
Customers And Debt Securities In Issue
- Loans and Receivables with
Customers
- Financial Assets held for
trading Cash and Cash Balances
- Financial Investments
Risk-Wide Integrated Stress Test
Market Risk
- Operational Risk
- Financial Investment Risk
- Real Estate Risk
- Business Risk
Securities In Issue
- Financial Liabilities
Held For Trading
- Financial Liabilities
Designated At Fair Value
- Hedging Instruments
P i i F Ri k
- Financial Investments
- Hedging Instruments
- Property, Plant and
Equipment
- Goodwill
Oth I t ibl A t
- Business Risk
- Provisions For Risk
And Charges
- Tax Liabilities
- Liabilities Included In
Disposal Group Classified As Held For Sale
- Many banks performed risk wide S.T. including credit
- Other Intangible Assets
- Tax Assets
- Non-Current Assets and
Disposal Groups Classified as Held For Sale O h A Sale
- Other Liabilities
- Shareholders’ Equity:
- Capital and Reserves
- Available-For-Sale
Assets Fair Value Reserve and Cash-Flow
risk, market risk, operational risk, but ...
- …. risk wide S.T. underestimates the true risk because
it does not take into consideration other elements of balance sheet coming from liabilities (e.g. deposits,
- Other Assets
Hedging Reserve
- Net Profit (Loss)
Enterprise Risk Management
- wn bonds,…) and more in general all the liquidity risk
- elements. (Firm wide S.T.)
- Banks have to improve S.T. methodology in order to
really capture and manage all kind of risk. Vice versa a
- Bank Balance Sheet Structure
- Business Strategies vs. change of market trends
Enterprise Risk Management stress scenario coming from an increase of reputational risk instead of a sovereign default or recessive outlook could affect more on the liquidity
- side. (The 2009 crisis provides impact both from credit
and liquidity side) 7 Firm-Wide Stress Test and liquidity side).
- 3. Planning Process:
Stress Testing integral part the Bank’s planning process to show robustness of Strategy and Budget-Assumptions
Business Strategy
(Target volumes
P&L Forecasting “Book Capital”
Model common equity
gy g p
Key performance indicators
(Target volumes, growth, risk appetite, target rating, dividend, policy, cost strategy, M&A,…)
Balance Sheet Forecasting
q y (raises/buybacks) Dividends/Retained earnings Minority interests
Forecasting Inputs to stress (risk factors)
Minority interests Sub debt maturing/issuance Provisions/EL/Deductions
Systemic risk Macroeconomic factors)
- Credit drivers (PD, LGD,
EAD)
- Losses (€)
RegCap
RWA = f (EAD PD LGD)
Scenarios
(Regulatory/ Business)
Losses (€)
- Liquidity drivers (funding
risk, concentration risk…)
- Market Drivers (Interest
RWA = f (EAD, PD, LGD)
ECap
( rates, FX, housing prices…)
ECap
EC = f (EAD, PD, LGD, correlations)
Source: Moody's Analytics
8
With all that properly defined as objective, where is an institution in its capabilities and what does the technical platform need to provide differently to the past?
- 4. Applicability:
Application level varies according to regulatory requirements and bank overall risk management sophistication g p
Main driver for stress testing (% of respondents) Leaders
Input into business strategy s driven 44% 22% 34% Regulation driven Business driven
Practitioners Leaders
» Leaders in practice
» Comprehensive process and governance
» Leaning towards business usage Business Both
d Followers
» Dedicated resources » Models and systems » Input into the business strategy and part of ERM » Both regulatory and business » Comprehensive process and governance » Mostly quantitative analysis » Some automation » Dedicated resources Both regulatory and
» Responding to regulation
» Process in place
ven
Laggards
business » All risks are stressed » Both bank-wide and specific tests » Both regulatory and business » All risks are stressed » Both bank-wide and specific test » Process in place » Manual / no automated calculations » Some quantitative analysis » Regulation driven » Several risks stressed
» Early stage
» Lack of process » Manual/ no automated calculations » Expert judgment
Regulation driv
(credit, market risk) » Bank-wide test » No dedicated resources » Specific stress testing
- nly
» Only market risk stressed
Sophistication Methodologies governance, tools & dedicated Embryonic process and tools, no dedicated
9
20% of banks interviewed 34% of banks interviewed 32% of banks interviewed 15% of banks interviewed
g g resources resources
Source: Moody's Analytics; Credit Stress-Test Survey 2011
- 4. Applicability
Recent Trends in Stress Testing in order to overcome Past Shortfalls……
1.Perform portfolio analysis at Group in addition to Business Unit / Asset Class levels in order to
understand cross-correlation effects.
2.Rather than performing analysis in silo by business function, seek to establish linkages between Risk
Management, Portfolio Management, Origination, and Treasury
- Use Stress Testing to understand balance sheet behavior through sensitivities analysis
Use Stress Testing to understand balance sheet behavior through sensitivities analysis
- Improve robustness of Fund Transfer Pricing and Liquidity Management through understanding of
spread decomposition (base rate, credit, optionality, margins) and sensitivities to market changes .
3.Rather than analysis performed on a top-down basis only, use Solutions that allows for bottom-up
analysis that will provide opportunities to understand Stress Test Impacts so to:
- Challenge Business Unit Planning by comparing risk / return metrics under varying potential
g g y p g y g p economic conditions in order to ensure resulting impacts are still appropriate to expectations
- Force Senior Management to maintain Oversight of Performance relative to Risk Tolerance
4.Improve upon Model Parameterization that support Stress Testing Programs
- Seek to stress key risk metrics cohesively rather than using an independent approach
- Adopt quantitative in addition to qualitative analysis to derive model risk parameterization
10
Adopt quantitative in addition to qualitative analysis to derive model risk parameterization
- Trend : Tie Model Parameterization to Macro-Economic Indicators to facilitate production of
Plausible Scenarios
- 4. Applicability
Sensitivity Analysis - Understand How Resilient Your Balance Sheet Is Under Different Scenarios; Identify Vulnerabilities ; y
1.What impact will an adverse scenario have on your RoA, RoE, Earnings, Capital? 2 Financial performance metrics and risk appetite are linked to planning
- 25.0%
Deviation from Target Value (%) – Annual Forecast
2.Financial performance metrics and risk appetite are linked to planning
- 10.0%
8 0%
- 15.0%
- 18.0%
- 12.0%
8 0%
- 11.0%
- 20.0%
- 15.0%
- 10 0%
Capital RoE RoA
- 1.0%
- 1.5%
- 7.0%
- 2.0%
- 4.0%
- 5.0%
- 7.0%
- 8.0%
- 4.0%
- 6.0%
- 8.0%
10.0%
- 5.0%
0.0% Earnings NPLs RAROC 4.0% 6.0% 5.0% 8.0% 7.0% 5.0% 7.5% 5.0% 10.0% 15.0% 23.0% 20.0% 25.0% 30.0% Recovery - 1/10 Mild - 1/25 Severe - 1/50 Extreme - 1/80
11
- 4. Applicability
Stress Scenario Analysis: Credit Value-at-Risk (VaR)
1.Via simulation, portfolio credit loss distributions are constructed and yield estimates of Expected Loss
(EL), Unexpected Loss (UL), and Economic Capital. (EC).
2.For Stress Testing, one can specify the estimation of the portfolio loss distribution conditional on the
realization of a given market scenario (economic state).. Facilitates comparative analysis to base case.
3 E ti
ti l ti d th k i k t l ti t ifi i i i
3.Estimating correlations and other key risk parameters relative to specific economic scenarios in a
consistent approach is critical to estimating the conditional loss distribution
12
- 4. Applicability
Forward looking Parameterization for Stress Testing VaR Analysis
1.Organizations tend to perform Origination & Portfolio analysis using Through-the-Cycle inputs 2.Stress Testing can assist institutions to understand vulnerability to experiencing dramatic shifts in 2.Stress Testing can assist institutions to understand vulnerability to experiencing dramatic shifts in
business performance during periods of market cyclicality
3.To measure impacts of Cyclicality, key to use Point-in-Time Measures for:
- Migration Rates: credit migrations can explain up to 50% of portfolio volatility and 35% of
Economic Capital (EC). Matrices estimated on data at peaks/troughs produce very different risk results, 30-35% increase in transition states.
- Correlation Rates: Estimates show firm levels of systematic risk sensitivity to market factors can
increase 20-30% during periods of market stress. E.g. Financials RSQs of 65% to 80-90%.
- Market Sharpe Ratio: Risk Tolerance input
1.200
- Market Sharpe Ratio: Risk Tolerance input
relative to Return – used for Cash Flow discounting in risk neutral model framework. Can be used to stress Market Liquidity During
0 700 0.800 0.900 1.000 1.100
Can be used to stress Market Liquidity. During period of peak Financial crisis in 2008 Sharpe Ratio doubled when compared to historical average
0.300 0.400 0.500 0.600 0.700 NA ‐ IG NA ‐ IG Avg EU ‐ IG EU ‐ IG Avg
13
average..
‐ 0.100 0.200
- 4. Applicability
A framework for VaR Portfolio Analysis driven by Macroeconomic indicators to derive Scenario Loss distributions indicators to derive Scenario Loss distributions
Macroeconomic Indicators Credit Risk Drivers (PD, Portfolio Risk – Capital
A B C
Macroeconomic Indicators ( , LGD, Correlations) p Requirements
A B C Macro- economic Indicators
Common Sources
- f Risk
(Country/Industry)
Risk Drivers- PD, LGD, Correlations Portfolio Loss Distribution
1.MA is constructing for Unicredit Group, correlation models based on regional / sector / product type
factors consistent to Unicredit’s portfolio composition. Such model also embeds macro-economic p p
- indicators. Model to be used for both base case and stress testing analysis.
2.Will enable stress test analysis model parameterization of Unicredit’s internal Credit Portfolio Model to
be driven through changes in economic scenarios facilitating intuitive causal analysis be driven through changes in economic scenarios facilitating intuitive causal analysis.
3.Such stress scenario results will provide insight as to associated changes to Portfolio Volatility and
Expected Shortfall, which are relevant for understanding:
14
- Changes in Risk Profile across Portfolio Segmentation allowing for Risk Mitigation
- Dynamic adjustments to Strategic Business Plans
- 4. Applicability
Stress Test Cases – Impact on Portfolio Risk
15
- 4. Applicability
Stress Testing vs. Reverse Stress Testing: Reconciliation
16
- 4. Applicability
Reverse Stress Testing is powerful tool to identify Bank’s vulnerabilities both from a systemic and an idiosyncratic standpoint
- 1. Using VaR Trial-by-Trial
capability –given a target loss bl id if d
y y p
- 1. Portfolio Tail Risk and Macro Analysis
– we are able to identify and quantify in Value terms different shock impacts in the Tail Region for each macro
2000 4000 6000
Capital Millions
Tail Risk Analysis
scenario
500 1000 1500 2000
Tail Exposure
- 2. Scenario and Contingency Analysis
- 2. Moody’s Analytics Reverse
Stress Testing Reports help senior management to put in g p place capital contingency plans and to develop the firm’s risk appetite, business strategy and risk limits
17
- 4. Capabilities
Reverse stress testing analysis g y
Analysis offers a unique opportunity for Banks to better understand their business and focus management’s attention on the areas where weakness could turn out to be focus management s attention on the areas where weakness could turn out to be potentially harmful to the entire organization
1.Consistent and cohesive portfolio reverse stress testing metric 2.Flexibility, Transparency, and Usability 3.Allows for Reconciliation of Stress Testing and Reverse Stress Testing 4.Helps Institutions to Meet Regulatory Requirements
18
With goals of an Organizations risk culture and stress testing framework defined, how does one develop Contingency Planning relative to Business Strategy?
- 5. Contingency Plan:
A Bank must complement its Business strategy with contingency plans in order for it to timely identify and address critical situations p y y
Level A
Triggers
Level B Level C
LOSS LOSS CORE TIER 1 RATIO
Level B actions Level C actions Level A actions Postponement of capital expenditure initiatives (capital spending) securitizations and recourse to risk mitigation Resolution Plan Review of credit limits changes in the overall strategy and business plan: reduction of exposures to specific
Actions
techniques (asset hedging or selling); addressing capital structure via reduction of share buy-back, reduction
- r halting of dividend
exposures to specific sectors, countries, regions, products or portfolios tighten underwriting policies and
19
- r halting of dividend
payments, capital raising, sale of assets policies and processes (i.e.. no new lending with revenue below expected loss);