ESTABLISHING RISK BOUNDARIES. Michel Rochette, MBA, FSA Caribbean Actuarial Association Annual Meeting Trinidad & Tobago December 4th 2008

Transcription

1 ESTABLISHING RISK BOUNDARIES Michel Rochette, MBA, FSA Caribbean Actuarial Association Annual Meeting Trinidad & Tobago December 4th 2008

2 TOPICS Context from 2006 to 2008 Risk appetite and ERM Definition and its evolution Value of articulating risk appetite Stakeholders influence on risk appetite statement Components of risk appetite and responsibilities Ex. of a risk appetite statement: ING Summary of methods to determine risk appetite Success factors

3 Risk Appetite: 2006 UK FSA Most firms have documented their approach for risk management through risk policies/procedures/risk appetite. However, «risk appetite» is not well understood throughout many firms to a level of clarity that provides a reference point for all material decision making. A big step exists between defining and applying risk appetite. UK FSA Insurance Sector Briefing, Risk Management in Insurance, 2006

4 Risk Appetite: 2008 UK FSA For insurers demonstrating a strong integration of risk and capital management: Clearly articulated and quantified risk appetite, tolerances, and trigger points for each risk. Processes are set to assess on a continuous basis the level of risk appetite. Coherent and well articulated processes to actively manage risk exposures that exceed risk appetite: risk monitoring. UK FSA Insurance Sector Briefing, 2008

5 Risk Appetite and ERM Risk Appetite Strategy ERM Framework

6 Strategy Objectives: markets, products and services, distribution channels, stakeholders Financial goals: Capital goals in relation to solvency issues. ROA and ROE without considering risk taking. RAROC if integrating risk taking into the strategic framework. Value creation goals if objective is to maximize shareholder s: Embedded Value. Non Financial goals: customer satisfication, corporate social responsibility objectives.

7 ERM Framework Enterprise risk Policy: All existing key risks: financial/operational/business/strategic Emerging risks: «known and unknown risks» Champion of Risk: CRO who can initiate a discussion of risk appetite at the Board/top management level, supported by a centralized risk unit. Risk technology: control of risk taking through risk limits, risk reporting through a dashboard. Businesses: risk management at the unit level. Audit/Compliance: independent oversight of the risk framework.

8 Risk Appetite: Evolution Turnbull Report: Risk appetite reflected indirectly by «those risks which are acceptable» UK COSO I: Focused on internal controls only COSO II ERM: Give management reasonable assurance that strategic objectives will be met within risk appetite CAS ERM Definition: Process to manage risks to create value. Risk appetite not explicit but indirect. Solvency II: Risk tolerance limits and business strategy must be defined. UK FSA Prudential Regulations: Risk appetite defined. ISO 31000: Risk appetite is defined indirectly in relation to value creation and risk acceptability.

9 Risk Appetite: Definitions COSO II ERM: Amount of risk that an entity is willing to accept in pursuit of value. Would add: «in line with the firm s strategic objectives taking into the capability of its ERM framework. Similar to a mission statement but focused on risk: Impact that risk can have on the capacity of the firm to attain its strategic objectives. Defines boundaries of what is «too much» or «too little» and what is «acceptable» or «non accpetable» in relation to the firm s strategic objectives.

10 Risk Appetite: Another View Risk Universe Strategy ERM Framework Retained Risk

11 Value of Articulating Risk Appetite Allow a FI to: Clarify desired risks: retained and non retained. Set the tone from the Top. Preferable to a bottom-up approach which tends to overemphasize exisiting risks. Estimate/Assess their impact, both financial and non financial ex. social responsibility Evaluation of risks, not a valuation of risks! Establish clearly the risk preferences of the company: Are we risk averse, risk takers in light of potential returns?

12 Value of Articulating Risk Appetite Set a consistent communication - transparency - from management to : Business units/product lines External parties: Shareholders: can diversify away if they don t like it! Regulators: Part of Pillar II and III of SolvencyII/Basel II. Other stakeholders: Employees may not want to be part of your organization. Ex. Army! Customers as well. Recent example: AIG only mentions the word risk appetite without ever elaborating about it in their official published documents.

13 Value of Articulating Risk Appetite Top down approch is preferable because: Stakeholders requirements are discussed explicitely among board members. Allows a more balanced view of risks instead of just focusing on one group: credit agencies, financial analysts, employees, shareholders, regulators, customers, society at large! More forward looking: Introduces forward thinking in terms of desired risk profile, not just existing risk profile! Can link risk appetite with strategic goals and required capital to support growth and risks. Board members/management are on the same page on risk appetite. Management can then react/take action if the risk profile exceeds/is below its desired/target risk appetite.

14 Stakeholders Influence: Board Risk preferences of individual board members/management: Risk averse vs risk takers. Risk Averse Type Board: Focuses on «value preservation». Reduces earnings volatility. Low impact of extreme events! «Keep us out of trouble» We don t want surprises! Concerns about legal fines, external scrutinity if they take too much risk. Wants to keep their desired ratings. Usually found in mutuals. Wants to preserve capital. Less concerned about capital efficiency. Incurring losses is perceived to be negative. Don t consider the gains realized before losses occured.

15 Stakeholders Influence: Board Risk Taker Type Board: Focused on «Value Enhancement». Considers risk vs opportunity relationship. Focuses on higher returns and risks. Anticipates «newer» risks, capitalizes on them, optimizes the risk/return relationship. Concept of efficient frontier! Optimizes use of capital. Capital management and risk management are done proactively. Usually found in public companies. When risks materialize, board shouldn t panic if within target risk appetite! Risk and losses are not viewed as negative!

16 Stakeholders Influence: Regulators Risk preferences of the local/global regulators: Asian: stricter, more rules based. European: more principle based. US: more rules based Stricter on Admitted assets, Single regulator - OSFI/UK FSA vs a diversified group of regulators US SEC, NAIC, OCC, OTS, FED, FDIC My prediction: US will tend towards a «single regulator model» common view, not one organization! Internationally: Moving towards «college»

17 Stakeholders Influence: Rating Ag. Risk preferences of rating agencies: Impact on agencies rating: Financial Strength or Claims Paying ability. If risk appetite is expressed solely as «desired AA rating», constraints immediately risk appetite to a certain overal probability of default/ruin. SP s ERM evaluation method: Risk Appetite is part of their Governance evaluation: «Clearly articulated risk tolerance is a key factor.»

18 Stakeholders Influence: Others Risk preferences of : Employees/customers/clients/policyholders: Risk of loosing key employees if taking too much risk! Will customers buy our products if the firm may not longer be there to service them in the future? Ex. GM/Ford In a pension plan, ratio of projected active/retired employees would certainly affect your desired risk appetite. Shareh0lders: If long-term/passive investors, may be willing to tolerate more risks. Political groups/media/advocacy groups.

19 «Risk Appetite»: Components Risk Capacity: Maximum amount of risk that an enterprise is able to accept in line with its mission/values/strategic goals. Risk appetite per se: Overall statement about the amount and type of risk that an enterprise is willing to accept in line with its strategic goals. Risk Target: Optimal level of risk desired. Risk Tolerance: Max/Min amount of risk for each class/subclass of risk. Risk Limits/Budgets: Thresholds not to exceed/min to accept. Not all firms have all these components!

20 Components: Risk Capacity Influenced by the quality of its risk management framework and processes: Overall ERM effectiveness: Sources could be an external view as assessed by a rating agency, external governance score. Management of past losses, especially unexpected and risk transfer options. Influenced by the amount and quality of its capital structure or Value of the business: Amount: measured by RBC, rating agencies required capital, economic view. Quality: Tier 1 versus Tiers 2 & 3 capital. Liquidity of capital: sources and availability particularly in times of stress. Access to central banks liquidity facilities: US recent history with AIG for ex. Systemic view by governments/markets: Too big too fail! Too big to rescue! Think of how Iceland was affected by the combined effect of risk appetite of its banks on the country itself. Value: Value of the business model to generate economic value.

21 Components: Risk Appetite Lower than Risk Capacity and if focused on downside risk: Defined as acceptable/non acceptable volatility of capital - quantitative component/metric over a certain horizon for certain risks deemed to be acceptable/non acceptable. qualitative component Quatitative metric: prob of ruin/ certain target rating/ minimum regulatory capital ratio Golden ruleon acceptable/non acceptable risks: Would our stakeholders be surprised if we annonced losses due to this risk? Think of AIG with credit derivatives! Focused on existing balance sheet risks/preservation of capital. Capital centric statement. Ex. «Level of risk that results in no more than a 0,1% chance of failure over a one-year horizon, where failure is defined as loosing 100% of capital, measure by US GAAP.»

22 Components: Risk Appetite If focused on downside/upside risk: Defined as an acceptable/non acceptable volatility of value - quantitative component/metric over a certain horizon for certain risks deemed to be acceptable/non acceptable. qualitative component Value metric: could be economic value/embedded value based on discounted earnings/cash flows at WACC. Focused not only on existing balance sheet risks but also takes into account emerging risks in line with strategy. Value centric statement, but not necessarily optimizing risk/return relationship as it expresses risk preferences. Tends towards a portfolio view of risks.

23 Components: Risk Target Specifies the optimal level of risk that an organization desires taking into account its risk capacity, risk appetite and desired returns. Efficient frontier concept: for a given level of capital capital centric approach or returns value centric approach - where do I want to be in terms of risk given my strategic goals? Target risk profile vs actual risk profile? Set risk objectives so that if risk is outside target monitoring of risk profile then actions are taken to reduce/enhance/increase risk taking. Could be done overall and by type of major risk class. Not all firms have risk targets.

24 Components: Risk Tolerance Sinceestimating risk capacity/appetite/target is not a perfect exercise, tolerance sets bands around which company is tolerating fluctuations of its risk appetite/target. Similar to the statistical concept of estimating a mean from a sample: Real mean = sample mean +/- Variability/Noize Set so that the aggregation of total risk is within the overall organization s risk appetite/target. Certain risks like SOX/Fraud/Legal Compliance: Zero Tolerance Financial risks: Tolerance expressed as a +/- yearly IRR duration mismatch, % of ALM, Greeks, GAP, Unexpected losses, yearly expected losses above a certain threshold, % economic capital depleted, volatility of embedded value Non financial risks: min customer satisfaction rates, employee retention rates, % of clients funds retained

25 Components: Risk Limits/Budgets Max not to exceed/min to accept. Practical/day-to-day constraints on business activities with some risk tolerances. Limits/risk budgets can be set up for: Business units, product lines, country, types of risks, concentration, market limit of securities held, existing, future derivatives -. Ex. ABCP recent problems in Canada. CDP Capital held 1/3 of market too much..didn t have a market limit Devising an overall limit system should be done so that it akes into account all acceptable/non acceptable risks, correlation, aggregation of risks, & risk tolerances in order to tend towards the firm s desired risk target/risk appetite. Risk limits should also be explained/negotitated with business units and embedded into compensation schemes. Limits should be established in the same units: Capital/Value As much an art as science here!

26 Risk Appetite: Responsibilities Board: Approves, discusses & challenges the Risk Appetite Statement. Reviews it annually & authorizes exception. Communicates it to stakeholders. Management: Reviews/discusses the risk capacity exercise. Proposes the risk appetite to the Board along with its components: target/tolerances/risk limits. Negotiates/explains the limits with the business units. Reports risk appetite to the Board. Frequency: quaterly. ERM Group: Performs the risk capacity/appetite/target/tolerance/limits exercise. Monitors the overall risk appetite/limit system. Updates analysis with changes in external environment, strategy

27 Examples of Risk Appetite: ING Risk appetite measured along 3 dimensions: Earnings at Risk, Capital At Risk, Economic Capital Earnings at Risk (EaR) is a measure of the potential reduction in IFRS earnings from expectations, assuming no mitigating management actions, during a moderate (i.e. 1 in 10 ) stressscenario. Capital at Risk (CaR) is the potential reduction of the current net asset value (based on fair values) of the balance sheet over the next year relative to the expected value during a moderate (i.e. 1 in 10 ) stress scenario, and assuming no mitigating management action. Economic Capital (EC) is the amount of capital required to absorb unexpected losses in times of severe stress given ING s AA target rating, 99,95%, (i.e. 1 in 2000 ). Integrates shareholder s point of view: EaR & CaR Integrates rating agencies/debtholders point of view: EC Integrates their banking and insurance operations/all risks Risk appetite appears 34 times in their 2007 Financial Statements compared to 1 time in AIG s 2007 Statements!

28 Risk Appetite: Methods Simple like KPI/KRIs combined in a scorecard indicator. Easy to set up and monitor. Concept of the Green/Amber/Red zones. Heat Map Approach: Evaluate Likelihood and Impact. Risk Appetite is the boundary line. Efficient frontier Approach: Investment Perspective. «Sophisticated Approach» : EC/Enterprise/Embedded Value Modelling. Recommend: Combination of methods if sophisticated modelling fails!

29 Risk Appetite: Success Factors Integrate both internal and external stakeholders different risk tolerances into the process from value protection to value creation. Integrate process within the overall strategy, culture and risk capabilities. Consider past historical decision making, reactions to events to assess risk appetite/tolerance. If CRO is fired all the time, maybe risk appetite is lower than said! ING CRO is leaving? Integrate non financial and financial risks: portfolio view of risks. Create a few measures that are practical and that represent the most critical aspects of the business. Communicate it through the firm! From top-down to bottom-up feedback.