Basic Statistical and Financial Tools: What Can They Reveal About Your Risks? (RIF001) Speakers: Laura Langone, Senior Director, Global Risk & Insurance, PayPal, Inc. Michael Elliott, Senior Director of Knowledge Resources, The Institutes
Basic Statistical and Financial Tools: What Can They Reveal About Your Risks? Introduction How do you evaluate insurance program options? Do you find yourself asking Should my company change deductibles from $250,000 to $500,000 or $1 million? What criteria do you use to evaluate these questions? There a various tools and analyses that may be used: Providing a process for companies to determine their risk tolerance and risk appetite Estimating a company s exposure to loss (expected loss, worst case scenarios) Comparing the company s insurance program options to other company capital investment options
Learning Objectives At the end of this session, you will: Explore basic statistical and financial tools available for risk management. Use these tools to analyze organizational risks and select appropriate risk financing solutions. Communicate to top management statistical and financial justifications for chosen risk financing techniques.
Overview Financial Tools Available Loss triangles Loss forecasting Retention analysis Understanding key financial concepts Total cost of risk Cash flow analysis Comparing financial options Risk-adjusted return on capital Engaging your broker and obtaining buy-in from management
Scenario Workers compensation line of business Current retention = $250,000 per accident Trying to decide whether to increase the retention to $500,000, or even $1,000,000 per accident Develop losses Look at premiums and other costs at alternative retention levels Analyze cash flow
Reported Claims Paid Amounts Plus Case Reserves (Incurred Losses) (Losses limited to $250,000 per accident) Evaluation (months after beginning of accident year) Accident Year 12 24 36 48 60 2012 $1,400,000 $1,960,000 $2,380,000 $2,660,000 $2,800,000 2013 $1,650,000 $2,310,000 $2,805,000 $3,135,000 2014 $1,115,000 $1,561,000 $1,895,500 2015 $1,400,000 $1,960,000 2016 $1,600,000
$3,000,000 2012 Accident Year IBNR $2,500,000 $2,000,000 $1,500,000 $1,000,000 $500,000 $0 2012 2013 2014 2015 2016 IBNR Reserves $1,400,000 $840,000 $420,000 $140,000 $0 Case Reserves $1,120,000 $1,260,000 $1,120,000 $840,000 $420,000 Paid Losses $280,000 $700,000 $1,260,000 $1,820,000 $2,380,000
$3,500,000 2016 Accident Year Projections $3,000,000 $2,500,000 $2,000,000 $1,500,000 $1,000,000 $500,000 $0 2016 2017 2018 2019 2020 IBNR Reserves $1,600,000 $960,000 $480,000 $160,000 $0 Case Reserves $1,280,000 $1,440,000 $1,280,000 $960,000 $480,000 Paid Losses $320,000 $800,000 $1,440,000 $2,080,000 $2,720,000
2016 Per Accident Retentions $250,000 Retention $500,000 Retention $3,000,000 $3,200,000 $3,400,000 $3,200,000 $3,500,000 $3,800,000 $1,000,000 Retention $3,350,000 $3,600,000 $3,950,000
2016 Estimated Total Cost of Risk (TCOR) Per Accident Retention $250,000 $500,000 $1,000,000 Retained Claims (expected) $3,200,000 $3,500,000 $3,600,000 Risk transfer expense 900,000 500,000 250,000 Loss control expense 50,000 50,000 50,000 RM administrative expense 40,000 40,000 40,000 Total cost of risk (expected) 4,190,000 4,090,000 3,940,000 Total cost of risk (high) 4,390,000 4,390,000 4,290,000 Total cost of risk (low) 3,990,000 3,790,000 3,690,000 Add one full-retention loss: $4,440,000 $4,590,000 $4,940,000
Cash Flow Analysis of 2016 Expected Claim Amounts Total cost of risk - $250,000 retention - Expected Amount 2016 2017 2018 2019 2020 2021 Estimated Percentage Paid Claims 10 15 20 20 20 15 Claim Payments 320,000 480,000 640,000 640,000 640,000 480,000 3,200,000 Risk transfer expense 900,000 900,000 Loss control expense 50,000 50,000 RM administrative expense 40,000 40,000 Total cash outflow 1,310,000 480,000 640,000 640,000 640,000 480,000 4,190,000 PV Factor (6.9% per annum) 0.9672 0.9048 0.8464 0.7917 0.7406 0.6928 Net Present Value TCOR $1,267,017 $434,285 $541,671 $506,708 $474,002 $332,555 $3,556,239
Another Approach The typical approach ignores the capital charge and capital at risk Retention level options can be evaluated using three metrics: Economic Cost of Risk Risk Adjusted Return Risk Adjusted Return on Capital Economic Cost of Risk = L+P+C where L = Discounted Retained Losses P = Discounted Premiums C = Capital Charge What is the capital charge? It is the charge related to holding capital to support the insurance program What is the company s cost of setting aside additional capital to support the insurance program Higher retentions have a higher capital charge and lower retentions have a lower capital charge
Economic Cost of Risk Optimality Defined as Minimum Economic Cost $40,000,000 $35,000,000 ECOR = L+P+C Economic Cost of Risk $30,000,000 $25,000,000 $20,000,000 $15,000,000 where L = Disc. Ret. Losses P = Premiums C = Capital Charge $10,000,000 $5,000,000 $0 <null> 2,500,000 $5,000,000 $10,000,000 $25,000,000 $35,000,000 $50,000,000 $100,000,000 Per-Claim Retention The capital charge can be interpreted in various ways whether or not funds are actually segregated from working capital. This approach typically identifies the optimal retention option.
Maximizing Risk Adjusted Return Value Generation Through Risk Management $ ECOR ECOR + RAR RAR Option 1 2 3 4 <n> 1 2 3 4 <n> 1 2 3 4 <n> Economic savings from <no insurance> (RAR) is interpreted as the value generated for the firm through the purchase of insurance.
Maximizing RAROC 15.0% RAROC = Risk Adjusted Return On Capital Economic Return Economic Capital Risk Adjusted Return on Capital 10.0% 5.0% 0.0% -5.0% -10.0% <null> $2,500,000 $5,000,000 $10,000,000 $25,000,000 $35,000,000 $50,000,000 Per-Occurrence Retention RAROC provides a means to compare treatments of different risks and across lines of business - i.e. a tool to implement true Enterprise Risk Management.
Capacity Does Not Imply Desirability 100 Just because a company may have the capacity to bear substantial risk, does not mean that it s in the firm s best interest to do so. Capacity to Bear Additional Risk (millions) 90 80 70 60 50 40 30 20 10 0 Working Capital Pre-Tax Earnings Cashflow Surplus Cashflow Earnings per Share Stockholder's Equity Net Sales Total Assets tolerable range for maximum unexpected losses A firm s capacity to bear risk can be thought of as a pain threshold above which the decision makers do not wish to experience. Capacity is often found through an analysis of the firm s financials, and application of rules of thumb. Wall Street s expectations often play a significant role. By definition, capacity goes across all sources of risk, not just insurable risks. Also, as expected losses are budgeted for, expected losses do not erode capacity. A large strong organization with many investment opportunities should not retain large amounts of fortuitous risk, as its capital base may be better employed.
Risk-adjusted Return on Capital Example: Worker s Compensation Analysis What retentions are optimal for the economic cost of capital to insure this risk? Inputs: Premium at various retentions Loss forecasts at various confidence levels Understand maximum probable loss worse case scenario Modeling by Oliver Wyman
Risk-adjusted Return on Capital Example: Medical Stop Loss Analysis Should HR self-insure this risk? Inputs: Premium at various retentions Loss forecasts at various confidence levels Understand maximum probable loss worse case scenario Modeling by Oliver Wyman
Summary Objectives Learned: There are various solid financial tools available for risk managers to make better risk-based decisions If used correctly, a risk retention analysis framework will provide risk managers a process for quantifying and optimizing the insurance purchase on an annual basis Next steps: Engage your broker and actuary Build time into your insurance renewal process Enhance decision making with senior management Add value to other internal stakeholders on these approaches
Questions
Developing Losses Collect and organize past data Limit individual losses Apply loss development factors 1 2 3 22
Predictive Modeling Training Data Test Data Production
Instances Predictive Modeling Data WC Claims Fraud Attributes Class Label Name Age Body part previously injured Attorney involvement Witness Fraudulent Claim Anna 35 Y Y N Y Carlos 42 N N Y N David 53 N N N N Jason 27 Y Y N Y Sonia 32 N Y Y N New Instance Gregory 45 Y Y Y?
Information Gain from Various Attributes 1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 Prev. Injured Body Part Age Attorney involvement Days to Report Claim Day of week Witness
Classification Tree WC Claim Fraud No Previously Injured Body Part Yes Age Attorney involvement < 40 => 40 No Yes Number of Medical Visits Day of week < 3 > 3 Monday Other than Monday Days to Report Claim < 1 > 1 No Witness Yes Prob. Fraud =.02 Prob. Fraud =.80
Classification as a Set of Rules If (body part previously injured) AND (an attorney is involved) AND (day of week is Monday) AND (no witness) THEN Class = Fraud Likely Refer for Further Investigation If (body part not previously injured) AND (age less than 40) AND (number of medical visits less than 3) AND (claim reported within 1 day) THEN Class = Fraud Highly Unlikely