Decision trees, scenario analysis, monte carlo simulation and scenario planning JUC – Netværk i Compliance & Risk Management
13 likes3,113 views
The document discusses using scenario planning to address the risk of routing bribery payments through third-party vendors. A workshop is proposed to generate scenarios on this issue over the next 5 years. Participants identify trends, strengths/weaknesses, and unknown factors. Scenarios are created on the impact of bribery cases, including a scenario where a large fine is imposed. Strategies are formulated to monitor for "early signals" of scenarios and centralize compliance functions. The process outlines interviews, workshops, and follow up needed to conduct the scenario planning exercise.
Decision trees, scenario analysis, monte carlo simulation and scenario planning JUC – Netværk i Compliance & Risk Management
- 1. Hernan Huwyler 4th meeting Governance, Risk and Compliance Management Scenario Planning
- 3. Words of estimative probability We know it is misleading since 1964! Sherman Kent study
- 4. More problems with adjectives risk management Enterprise Operational Strategic Project IT Portfolio Legal Reputational Business continuity Financial Fraud
- 5. The real RM problems Large actual vs. budget differences – missing targets, growth in particular – uninsured losses – complex renegotiations – the “journalist” approach Biased decisions – “recalibrations” on past experiences – bonus schemes – poor articulation of strategies impacted on risk identification None of them addressed by colourful heat maps RM not changing any decision and not detecting unknown issues
- 6. The intern test Discover how managing… insulated RM policy library of risks ad-hoc workshops KRIs reporting emerging risks …. takes 2 months to master
- 7. Decision-making in budgeting – current operations in long-term planning – development and bidding plans Quantify risk – analytical tools to model risks – tips to better predictions (calibration techniques) for mid-managers Investment analysis – approving investments Holistic ISO 31000 – still need to report risks Goals to achieve?
- 8. Sales volumen Sales price Labor / fuel costs Inflation FX rates Interest rate Fines / claims Construction time Approach Cash flow EBITA Capital budget Tender price Tariff pass-through request Budgets for current business units Development plans Investment requests Financing alternatives Bidding Predictive models for one scenario to simulations Assumptions Model Decisions
- 9. Model risks in the current planning activities – excellent for investment decisions Assumptions in demand, costs and time – internal and external risk factors (e.g. competitors) – alternatives when decisions are made (what if) Payoff tables with net present values NPVs – maximum expected cash flow – most likely cash flow – PERT estimate: (best case + worse case + av. case * 4 ) / 6 – alternative: 10% and 90% most likely cases Decision trees
- 10. Decision trees Decision Event - Probability 10 yr contract NPV Expected Individual subsidiary No price renegotiation - 90% Good negotiation - 10% 1,000 2,000 900 200 1,100 As part of a consortium No price renegotiation - 80% Good negotiation - 20% 500 3,000 400 600 1,000 NPV of revenues and costs Each event affects many budget assumptions Combine analytical risk identification Transfer this model to mgmt accounting
- 11. Scenario analysis Scope: physical assets, data, people, reputation Assess current controls on assets → vulnerabilities Early indicators of occurring scenarios Identify key assumptions Identify impacts from external sources → “out of trends” Craft plausible futures → “stories” or “Alternative worlds” Time horizon → tip: 3/5 ys Alternative outcomes → primary and secondary losses Simulation Based on internal and external data Threats Assets Losses
- 12. Calculate alternative possible outcomes of a risk – understand how risk affects the budget assumptions • What if there is a new regulation? Bad weather? – project consequences for all assumptions – tip: limit to assumptions with the higher impact • Several possible scenarios – macro economics, demand, cost prices – optimistic, baseline and adverse → stress testing – assign probabilities to each scenario – tip: involve mgmt, finance, hr, mkt and ops experts Objectives – identify top risks – ensure consistency in assumptions • bias: estimate a good economy for sales and bad for costing – exercise flexibility • develop contingency and prevention plans • identify early warnings if an scenario occurs Scenario analysis
- 13. Scenario analysis Assumptions Best Base Worst Growth in demand 7% 5% 2% Tariff increase – inflation (in real terms) 2% 0% -7% Days of outstanding sales 50 days 70 days 180 days Impact on cash flow Growth in demand 2% 0% -5% Tariff increase – inflation (in real terms) 8% 0% -15% Days of outstanding sales 5% 0% -10% Scenario analysis Cashflow % Occurrence Best €50M 20% €30MBase €40M 50% Worst €0M 30% Cash flow w/o scenario analysis Cash flow w/ scenario analysis
- 15. Estimated performance Year Estimatedcashflow 1 2 3 4 5 6 7 Project net cash flows Best case (5%) Worse case (95%) Risk of being short of cash after overoptimistic assumptions
- 16. Simulate outputs for more +1k scenarios (runs) – Output: 1 cash flow – 2 project NPV – Not only the most likely scenario Lower and upper limits in assumptions – from previous experience or subjectively – tip: previous sensibility analysis to identify key assumptions, focus on high to moderate risks – random distribution to simulate an aggregated result Discuss aggregated scenarios with stakeholders – Do we have capabilities (EUR, FTEs, know-how) to deal with extreme scenarios? Monte carlo
- 17. Monte carlo in SAP
- 18. Trade-off Choose between capital investments Allocate available capital Identify competitive advantages Risk and return Risk exposure Net present value at risk as % investment Internalrateofreturn Low risk and return High risk and return A B
- 19. Recommendation Hubbard, D. (2009), The failure of risk management (here) Key Improvement to Risk Management “Use calibrated probabilities to express your uncertainties and use those uncertainties in Monte Carlo models of your organization as a system.”
- 20. Thomas, P. (2013) The Risk of Using Risk Matrices (here) Cox, L.A. (2008) What's wrong with risk matrices Flage , R. (2012) A reflection on some practices in the use of risk matrices Elmontsri, M. (2014) Review of the strengths and weaknesses of risk matrices (here) References
- 21. Hernan Huwyler 4th meeting Governance, Risk and Compliance Management Workshop Scenario Planning
- 22. Discussion case: bribery by 3Ps
- 23. Scenario planning? Scenarios are stories about the future Participative learning exercise Scenario analysis applied to strategic planning –1st step: scenario generation future plausible and rigorously constructed events affecting performance objectives critical questions faced by decision-makers –2nd step: scenario planning reactions for scenarios with great uncertainty
- 24. “What if” in scenario planning
- 25. What are the issues to be explored? – macro scenarios, market or investment decisions or learning – Tip: link them to core problems and growth plan assumptions How long should the time horizon be? – I.e.: 5 years long term planning / 20 years for concessions Which business units should be included? Workshop How we can mitigate the risk of routing bribery transactions through third-party vendors in the next 5 years? 1 Frame the issue
- 26. Which are the managers dealing with the issue to explore? Who has the experience to produce better inputs? Are other people representing interests outside the organization? Workshop Risk owner: compliance SMEs: internal audit, legal, contracting, finance External dialog: investor relations, CEO? 2 Identify participants
- 27. To the participants: Which are the trends and key control vulnerabilities in our organization and sector? Which data supports them? – Tip: look for blind spots / weak signals Workshop (possible answers) More 3rd party advisors to obtain licenses Increasing number of invoices submitted for services never performed detected by audits Misrepresentation issues in booking services by the shared service center Background checks disqualified many vendors Increasing impact on the reputation after corruption cases 3 Identify the known
- 28. Strengths Ethical leadership Coordination Intellectual capital Healthy bonus incentives Strong due diligence / background check Weaknesses Strong need to use external advisors to get licenses Controls by shared services center Decentralized compliance program Weak financial controls Opportunities Leverage skills across compliance, finance and audit Implement a fraud/corruption detection solution Collaborative efforts on GRC Threats Impact on the reputation Political changes Increased scrutiny by stakeholders Criminal liability Emerging competition 3 SWOT analysis (illustrative)
- 29. 3 impact/predictability grid Impact Predictability Low High HighLow FCPA UK ABL AML Enforcement in developing countries Increased reputational loss Populist movements
- 30. To the participants: What are the external unknown factors? Workshop (possible answers) Extraterritorial anti-bribery laws enforced in current markets Tougher licenses to obtain by authorities Improper payments detected by AML ERP solutions to prevent corruptions Criminal liability for corporations going global 4 Identify the unknown
- 31. If you could speak with an oracle, what things would you like to know about the corruption cases in this organization in 2022? If the organization were to collapse after a corruption scandal by 2022, what might have caused the collapse and why? 4 Brainstorming questions
- 32. To the participants: What are the best and worst case scenarios for the identified trends on the performance of the scoped BUs? Workshop (possible answers) Fines Penalties Reputation 5 Create possible scenarios
- 33. 5 Scenario matrix UncertaintyA LowHigh Uncertainty B Low High Scenario 1 Scenario 2 Scenario 3 Scenario 4
- 35. 5 Fact sheet Scenario: Considerable fine after a BU used a vendor to channel bribery payments undetected by accounting Top manager of BU avoids a due diligence on new vendor Top manager uses the new vendor to route bribery payments to public officers Public officers facilitate issuing permits for new operations Large fine is set after a whistleblower exposed the scheme to the authorities Description Scenario data Weak supervisory controls Decentralized operation of a compliance program Scheme out of the scope of internal audit Impact – Other costs Legal costs: 1M EUR Forensics advisors: .5M EUR Internal reorganization: 2M EUR Impact - Fine Best: 1M EUR Worst: 20M EUR Improper payments Best: .1M EUR Worst: 1M EUR
- 36. Do the envisioned follow logically from what is known? Are the futures plausible from what we know? Workshop (possible answers) Increased AML regulations are linked to anti- bribery 6 Test of plausibility
- 37. To the participants to create “stories”: How the organization and external stakeholders would react to each scenario? Workshop (possible answers) Competitors would target affected clients Public banks would deny financing 7 Anticipate interactive dynamics
- 38. Corruption impact/predictability matrix Impact Predictability Low High HighLow Review Contingency plans Monitor Develop scenarios
- 39. To the participants: What strategic initiatives should implement now to address plausible scenarios? What are the “early signals” for each scenario? Workshop (possible answers) Crisis protocol for ethics and compliance Centralize the compliance functions for due diligence Monitor “early signals” 8 Formulate strategies
- 40. Initiation On boarding Generation Planning Initial meeting for objective and scope setting Interviews with senior executives and cross sections 3 workshops for scenario generation Scenario presentation 3 workshops for scenario planning Follow up the scenario- based strategies External and internal analysis Data collection and research Data collection and research 6 hours 1 hour per interview (10) 4 hrs per workshop (3) 4 hours 4 hrs per workshop (3) 1 hour per interview (10) Process and schedule
- 41. Supply chain quality of inputs, supply relative to industry demand, price of raw materials Product consumer tastes, market demand, availability of substitutes and complements Competition pricing and other forms of rivalry, new entrants, product and process innovations, rivals’ price Uncertainties list Industry Operations labor relations, availability of inputs, production variability and downtime Liability product liability, emission of pollutants Credit delays with collectibles Ethics fraud, corruption, opportunistic behaviors by managers or employees R&D R&D activities, regulatory approval of new products Company level at corporate or business levels
- 42. Social social unrest, diversity, cultural, shift in social concerns, skills, expectations from stakeholders Technologic disruptive innovations, digital transformation, cyber threats, business intelligence, automatization, mobile technologies, renewable fuel sources Economic exchange rates, interest rates, inflation, terms of trade, China and India, disposable income Environmental climate change, natural disasters, available raw materials, environmental consciousness of consumers Political Brexit, nationalizations, far-right politics, terrorism, war, changes in government, geopolitical stability, liberalization government policies fiscal and monetary policies, trade restrictions, regulations affecting the business sector, tax policy, environmental regulations, changes in collective bargaining law Uncertainties list General (STEEP forces approach)
- 43. Budget accuracy improved – by validating assumptions – RM as a decision making process – budgets disclosed risks (rather than hide them) – Some scenarios became real RM became visible, relevant and ongoing – interest in profitability moved to RM – risk culture linked to strategic plans Better understanding of the relationships between risk events – by quantitative modelling and scenario planning Final thoughts
- 44. Recommendation Chermack, T. (2011), Scenario planning in organizations “Scenario planning is a revolutionary alternative to traditional strategic planning because it recognizes the unpredictable nature of the future”
- 45. Miller, K. (2003) Scenarios, real options and integrated risk management Schwenker, B. (2013) Scenario-based strategic planning - Developing strategies in an uncertain world Kahane, A. (2012) Transformative scenario planning Martelli, A (2014) Models of scenario building and planning References
- 46. Closing
Editor's Notes
- #39: /