1. A Statistical Analysis of The Auto BI Settlement Process and Structure of Negotiated Payments in The Presence of Fraud and Buildup Richard A. Derrig, President, OPAL Consulting LLC Visiting Scholar, Wharton School President, OPAL Consulting LLC Visiting Scholar, Wharton School University of Pennsylvania University of Pennsylvania Grezgorz A. Rempala Associate Professor, Statistics University of Louisville 2005 WRIEC Salt Lake City, Utah, USA August 8-11, 2005

4. NEGOTIATION Liability claims are negotiated not “paid by the insurer”. Liability claims are negotiated not “paid by the insurer”. First party claims have payment regulations both good (Cooperation) and bad (Time Frames for Payment) re fraud. First party claims have payment regulations both good (Cooperation) and bad (Time Frames for Payment) re fraud. Negotiation subject only to bad faith and unfair claim practice regulations Negotiation subject only to bad faith and unfair claim practice regulations Two-person game: Adjusters and Claimant/Attorneys, but not suitable for game theory model. Two-person game: Adjusters and Claimant/Attorneys, but not suitable for game theory model. Example in paper is Auto Bodily Injury Liability – Mass Data Example in paper is Auto Bodily Injury Liability – Mass Data

5. Table 1 BI Negotiation Leverage Points Adjuster Advantages Adjuster has ability to go to trial Company has the settlement funds Attorney, provider, or claimant needs money Adjuster knows history of prior settlements Adjuster can delay settlement by investigation Settlement authorization process in company Initial Determination of Liability

6. Table 2 BI Negotiation Leverage Points Attorney/Claimant Advantages Attorney/Claimant can build-up specials Asymmetric information (Accident, Injury, Treatment) Attorney/Claimant can fail to cooperate Attorney has experience with company Investigation costs the company money Attorney can allege unfair claim practices (93A) Adjuster under pressure to close files

7. NEGOTIATION Claim Payment Components Claim Payment Components Demands and Offers Demands and Offers Time Frames for Rounds Time Frames for Rounds Anchoring and Adjusting Anchoring and Adjusting Offer/Demand Ratios Offer/Demand Ratios Settlements Settlements Mass BI Data for 1996 AY Mass BI Data for 1996 AY Statistical Modeling Statistical Modeling

8. Table 3 - 1 Final Negotiation Model Negotiation Variables 1996 data 1 (422 Claims in Data Set) Disability Unknown0.43110.70.0010 1st Demand Ratio0.0139.80.0017 BI IME No Show-0.3846.30.0118 BI IME Not Requested-0.1526.10.0135 BI IME Performed with positive outcome-0.1685.10.0245 Suspicion-0.0277.00.0080 1 1996 data set - includes "Unknown Disability" claims and claims with a 1st Demand Amount 2 1996 data set - excludes 64 "Unknown Disability" claims

9. NEGOTIATION Claim Payment Components Claim Payment Components Demands and Offers Demands and Offers Time Frames for Rounds Time Frames for Rounds Anchoring and Adjusting Anchoring and Adjusting Offer/Demand Ratios Offer/Demand Ratios Settlements Settlements Mass BI Data for 1996 AY Mass BI Data for 1996 AY Statistical Modeling Statistical Modeling

10. Table 4 Negotiation - Steps (dollars) Claim Demand 1 Offer 1 Demand 2 Offer 2 Demand 3 Offer 3BI Example 110,0004,1007,5004,6006,0004,7505,500 Example 535,0006,00020,00011,000na 15,000 AVG 4 Rds28,0834,67915,3855,97811,4527,5368,536 STD 4 Rds.41,3403,11015,1073,67010,3445,1576,265 AVG 3 Rds.21,4885,55211,2667,211na 7,926 STD 3 Rds.19,0264,07412,3855,094na 5,608

11. NEGOTIATION Claim Payment Components Claim Payment Components Demands and Offers Demands and Offers Time Frames for Rounds Time Frames for Rounds Anchoring and Adjusting Anchoring and Adjusting Offer/Demand Ratios Offer/Demand Ratios Settlements Settlements Mass BI Data for 1996 AY Mass BI Data for 1996 AY Statistical Modeling Statistical Modeling

12. Table 6 Negotiation – Offer/Demand Ratios by Round 4 ROUNDS (100 claims) O 1 /D 1 O 2 /D 2 O 3 /D 3 BI/D 3 Average0.2460.4760.7240.798 Std. Dev.0.1530.2130.2110.191 3 ROUNDS (119 claims) O 1 /D 1 O 2 /D 2 BI/D 2 Average0.3930.7080.766 Std. Dev.0.6100.2120.191

13. Offer Demand Ratios (Sorted by Descending Losses) – Figure 1

14. Offer Demand Ratios (Sorted by Descending 1st Demands) – Figure 2

15. Table 7 Offer/Demand Ratio Dependence on Demand RatioRoundsIntercept S.E. Demand (000) Coefficient O 1 /D 1 20.550.08-0.0074 O 2 /D 2 20.780.02-0.0061 BI/D 2 20.840.02-0.0057 O 1 /D 1 30.280.02-0.0013 O 2 /D 2 30.560.03-0.0055 O 3 /D 3 30.790.03-0.0058 BI/D 3 30.840.02-0.0040 All intercept and demand coefficients significant at 1%

16. Table 8 Negotiation – Steps (Ratio) Step 0 Step 1 Step 2 Step 3 Step 4 Step 5 Step 6 Step 7 Claim 0 / D1O 1 / D 1 O 1 / D 2 O 2 / D 2 O 2 / D 3 O 3 / D 3 O 3 / BI Settle ment Example 100.4100.5470.6130.7670.7920.8641.000 Example 500.1710.3000.550na 0.7331.000 Note: Step 6 for 3 rounds is O 2 /BI AVG 4 Rds0.2460.3770.4760.6080.7240.909 STD 4 Rds.0.1530.1920.2130.2240.2110.148 AVG 3 Rds.0.3930.5790.708na 0.920 STD 3 Rds.0.6100.4510.212na 0.137

17. Table 9 Model for the First Round Ratio (O 1 /D 1 ) VariableCoefficient Standard Errorp-Value Demand 1 (000's)-0.00410.0009<.0001 O 1 / BI Settlement0.85610.0674<.0001 Suspicion-0.00200.00120.1021 BI Total Paid (000's)0.01010.00450.0264 Serious Injury0.21930.09380.0204 Log Disability Wks-0.05950.03360.0781 Three or more claimants-0.12630.06380.0493 Disability Unknown-0.26820.14800.0715 Settlement Lag 1-2 yrs0.20120.06500.0022 R 2 =.66

18. Figure 1: The Massachusetts Negotiation Data Estimated standardized rates of the NHPP of arrival of O/D for 2-, 3- and 4-negotiation rounds.

19. Table 10 Logistic Classifier of Fast and Slow Claims VariableCoefficient Standard Errorp-Value Demand 1 (000's)-0.06780.03270.0385 O 1 / D 1 -5.46602.84400.0546 Report Date – Accident Date (days)-0.02970.01030.0038 Three or more claimants -1.69901.05800.1082 BI IME Not Requested3.13001.09400.0042 BI IME Performed with Positive Outcome2.54601.44900.0789 Intercept 3.0120 1.7000 0.0764

20. Figure 2: Empirical rates for ‘slow’ and ‘fast negotiations’ (solid lines) along with the rates estimated on the basis of the logistic regression classifier (dashed lines) for the subset of 58 negotiations histories from the Massachusetts dataset

21. Figure 3: 95% confidence tunnel for both ‘slow’ and ‘fast’ fitted rates for the subset of 58 negotiations histories from the Massachusetts dataset

22. Offer / Demand Ratios (Sorted by Descending Pre- Settlement Ratio) – Figure 3

23. NEGOTIATION Future Modeling Work Demands and Offers Demands and Offers Role of Time Frames Role of Time Frames Role of Covariates (Injury, etc) Role of Covariates (Injury, etc) Anchoring and Adjusting Anchoring and Adjusting Offer/Demand Ratios Offer/Demand Ratios Settlements Settlements Statistical Models Statistical Models Mass BI Data for 1996 AY Mass BI Data for 1996 AY Another Data Set Needed Another Data Set Needed

24. References Cooter, Robert D. and Daniel L. Rubinfeld, (1989), Economic Analysis of Legal Disputes and Their Resolution, Journal of Economic Literature, 27, 1067- 1097 Cooter, Robert D. and Daniel L. Rubinfeld, (1989), Economic Analysis of Legal Disputes and Their Resolution, Journal of Economic Literature, 27, 1067- 1097 Derrig, Richard, and Herbert I. Weisberg, (2004a), Determinants of Total Compensation for Auto Bodily Injury Liability Under No Fault: Investigation, Negotiation and the Suspicion of Fraud, Insurance and Risk Management, 71:4, 633-662, January. Derrig, Richard, and Herbert I. Weisberg, (2004a), Determinants of Total Compensation for Auto Bodily Injury Liability Under No Fault: Investigation, Negotiation and the Suspicion of Fraud, Insurance and Risk Management, 71:4, 633-662, January. Epley, Nicholas, and Thomas Gilovich, (2001), Putting Adjustment Back in the Anchoring and Adjustment Heuristic: Differential Processing of Self-Generated and Experimenter-Provided Anchors, Psychological Science, 12:5, 391-396. Epley, Nicholas, and Thomas Gilovich, (2001), Putting Adjustment Back in the Anchoring and Adjustment Heuristic: Differential Processing of Self-Generated and Experimenter-Provided Anchors, Psychological Science, 12:5, 391-396. Loughran, David, (2002) Deterring Fraud: The Role of General Damage Awards in Automobile Insurance Settlements, RAND, Working Paper, August. Loughran, David, (2002) Deterring Fraud: The Role of General Damage Awards in Automobile Insurance Settlements, RAND, Working Paper, August. Raiffa, Howard, (1982), The Art and Science of Negotiation, The Belknap Press of Harvard University Press. Raiffa, Howard, (1982), The Art and Science of Negotiation, The Belknap Press of Harvard University Press. Ross, Lawrence, H., (1980), Settled Out of Court, (Chicago, III: Aldine). Ross, Lawrence, H., (1980), Settled Out of Court, (Chicago, III: Aldine). Tversky, A., and D. Kahneman, (1974), Judgment Under Uncertainty: Houristics and Biases, Science, 195, 1124-1130. Tversky, A., and D. Kahneman, (1974), Judgment Under Uncertainty: Houristics and Biases, Science, 195, 1124-1130. Wright, W.F. and U. Anderson, (1989), Effects of Situation Familiarity and Incentives on use of the Anchoring and Adjustment Heuristic for Probability Assessment, Organizational Behavior and Human Decision Processes, 44, 68-82. Wright, W.F. and U. Anderson, (1989), Effects of Situation Familiarity and Incentives on use of the Anchoring and Adjustment Heuristic for Probability Assessment, Organizational Behavior and Human Decision Processes, 44, 68-82.