1. 1 Dirty Data on Both Sides of the Pond: Results of the GIRO Working Party on Data Quality 2008 CAS Ratemaking Seminar Boston, Ma.

2. 2 Data Quality Working Party Members Robert Campbell Louise Francis (chair) Virginia R. Prevosto Mark Rothwell Simon Sheaf

3. 3 Agenda Literature Review Horror Stories Survey Experiment Actions Conclusions

4. 4 Literature Review Data quality is maintained and improved by good data management practices. While the vast majority of the literature is directed towards the I.T. industry, the paper highlights the following more actuary- or insurance-specific information: Actuarial Standard of Practice #23: Data Quality Casualty Actuarial Society White Paper on Data Quality Insurance Data Management Association (IDMA) Data Management Educational Materials Working Party

5. 5 Actuarial Standard of Practice #23 Provides descriptive standards for: selecting data, relying on data supplied by others, reviewing and using data, and making disclosures about data quality http://www.actuarialstandardsboard.org/pdf/asops/asop023_097.pdf

6. 6 Insurance Data Management Association The IDMA is an American organization which promotes professionalism in the Data Management discipline through education, certification and discussion forums The IDMA web site: Suggests publications on data quality, Describes a data certification model, and Contains Data Management Value Propositions which document the value to various insurance industry stakeholders of investing in data quality http://www.idma.org

7. 7 Cost of Poor Data Olson – 15% - 20% of operating profits IDMA - costs the U.S. economy $600 billion a year. The IDMA believes that the true cost is higher than these figures reflect, as they do not depict “opportunity costs of wasteful use of corporate assets.” (IDMA Value Proposition – General Information).

8. 8 CAS Data Management Educational Materials Working Party Reviewed a shortlist of texts recommended by the IDMA for actuaries (9 in total) Publishing a review of each text in the CAS Actuarial Review (starting with the current (August) issue) Paper published in the Winter 2007 CAS Forum combines and compares reviews “Actuarial IQ (Information Quality)” published in the Winter 2008 CAS Forum

9. 9 Agenda Literature Review Horror Stories Survey Experiment Actions Conclusions

10. 10 Horror Stories – Non-Insurance Heart-and-Lung Transplant – wrong blood type Surgery on wrong side very frequent, but preventable Bombing of Chinese Embassy in Belgrade Mars Orbiter – confusion between imperial and metric units Fidelity Mutual Fund – withdrawal of dividend Porter County, Illinois – Tax Bill and Budget Shortfall

11. 11 Horror Stories – Rating/Pricing Exposure recorded in units of $10,000 instead of $1,000 Large insurer reporting personal auto data as miscellaneous and hence missed from ratemaking calculations One company reporting all its Florida property losses as fire (including hurricane years) Mismatched coding for policy and claims data

12. 12 Horror Stories - Reserving NAIC concerns over non-US country data Canadian federal regulator uncovered: Inaccurate accident year allocation Double-counted IBNR Claims notified but not properly recorded

13. 13 Horror Stories - Reserving June 2001, the Independent in liquidation, became the U.K.’s largest general insurance failure. A year earlier, its market valuation had reached £1B. Independent’s collapse came after an attempt to raise £180M in fresh cash by issuing new shares failed because of revelations that the company faced unquantifiable losses. The insurer had received claims from its customers that had not been entered into its accounting system, which contributed to the difficulty in estimating the company’s liabilities.

14. 14 Horror Stories - Katrina US Weather models underestimated costs for Katrina by approx. 50% (Westfall, 2005) 2004 RMS study highlighted exposure data that was: Out-of-date Incomplete Mis-coded Many flood victims had no flood insurance after being told by agents that they were not in flood risk areas.

15. 15 Agenda Literature Review Horror Stories Survey Experiment Actions Conclusions

16. 16 Data Quality Survey of Actuaries Purpose: Assess the impact of data quality issues on the work of general insurance actuaries 2 questions: percentage of time spent on data quality issues proportion of projects adversely affected by such issues

17. 17 Results of Survey

18. 18 Survey Conclusions Data quality issues have a significant impact on the work of general insurance actuaries about a quarter of their time is spent on such issues about a third of projects are adversely affected The impact varies widely between different actuaries, even those working in similar organizations Limited evidence to suggest that the impact is more significant for consultants

19. 19 Agenda Literature Review Horror Stories Survey Experiment Actions Conclusions

20. 20 Hypothesis Uncertainty of actuarial estimates of ultimate incurred losses based on poor quality data is significantly greater than those based on better quality data

21. 21 Data Quality Experiment Examine the impact of incomplete and/or erroneous data on the actuarial estimate of ultimate losses and the loss reserves Use real data with simulated limitations and/or errors and observe the potential error in the actuarial estimates

22. 22 Data Used in Experiment Real data for primary private passenger bodily injury liability business for a single no- fault state Eighteen (18) accident years of fully developed data; thus, true ultimate losses are known

23. 23 Actuarial Methods Used Paid chain ladder models Bornhuetter-Ferguson Berquist-Sherman Closing rate adjustment Incurred chain ladder models Inverse power curve for tail factors No judgment used in applying methods

24. 24 Completeness of Data Experiments Vary size of the sample; that is, 1)All years 2)Use only 6 accident years 3)Use only last 3 diagonals

25. 25 Data Error Experiments Simulated data quality issues: 1.Misclassification of losses by accident year 2.Late processing of financial information 3.Overstatements followed by corrections in following period 4.Definition of reported claims changed 5.Early years unavailable

26. 26 Measure Impact of Data Quality Compare Estimated to Actual Ultimates Use Bootstrapping to evaluate effect of different random samples on results

27. 27 Estimated Ultimates based on Paid Losses

28. 28 Est. Ults. based on Adjusted Paid

29. 29 Est. Ults. based on Incurred Losses

30. 30 Results of Adjusting Experience Period The adjusted paid and the incurred methods produce reasonable estimates for all but the most immature points. However these points contribute the most dollars to the reserve estimate. The paid chain ladder method, which is based on less information (no case reserves, claim data or exposure information), produces worse estimates than the methods based on the incurred data or the adjusted paid data. Methods requiring more information, such as Bornhuetter- Ferguson and Berquist-Sherman, performed better It is not clear from this analysis that datasets with more historical years of experience produce better estimates than datasets with fewer years of experience.

31. 31 Experiment Part 2 Next, we introduced three changes to simulate errors and test how they affected estimates: 1.Losses from accident years 1983 and 1984 have been misclassified as 1982 and 1983 respectively. 2.Approximately half of the financial movements from 1987 were processed late in 1988. 3.Incremental paid losses for accident year 1988 development period 12-24 overstated by a multiple of 10. This was corrected in the following development period. An outstanding reserve for a claim in accident year 1985 at the end of development month 60 was overstated by a multiple of 100 and was corrected in the following period.

32. 32 Comparison of Actual and Estimated Ultimate Losses Based on Error-Modified Incurred Loss Data

33. 33 Standard Errors for Adjusted Paid Loss Data Modified to Incorporate Errors

34. 34 Results of Introducing Errors The error-modified data reflecting all changes results in estimates having higher standard errors than those based on the clean data For incurred ultimate losses, the clean data has the lowest bias and lowest standard error Error-modified data produced: More bias in estimates More volatile estimates

35. 35 Distribution of Reserve Errors

36. 36 Results of Bootstrapping Less dispersion in results for error free data Standard deviation of estimated ultimate losses greater for the modified data (data with errors) Confirms original hypothesis: Errors increase the uncertainty of estimates

37. 37 Conclusions Resulting from Experiment Generally greater accuracy and less variability in actuarial estimates when: Quality data used Greater number of accident years used Data quality issues can erode or even reverse the gains of increased volumes of data: If errors are significant, more data may worsen estimates due to the propagation of errors for certain projection methods Significant uncertainty in results when: Data is incomplete Data has errors

38. 38 Agenda Literature Review Horror Stories Survey Experiment Actions Conclusions

39. 39 Actions – What can we do? Data Quality Advocacy Data Screening or Exploratory Data Analysis (EDA)

40. 40 Data Quality Advocacy Data quality – measurement Data quality – management issues

41. 41 DQ - Measurement Redman – advocates a sampling approach to measurement & depends on how advanced data quality at company currently is Other – automated validation/accuracy techniques

42. 42 DQ - Management Issues Manage Information Chain (from Data Quality, the Field Guide by Redman) establish management responsibilities describe information chart understand customer needs establish measurement system establish control and check performance identify improvement opportunities make improvements

43. 43 Data Screening Visual Histograms Box and Whisker Plots Stem and Leaf Plots Statistical Descriptive statistics Multivariate screening Software – Excel, R, SAS, etc.

44. 44 Example Data Texas Workers Compensation Closed Claims Some variables are: Incurred Losses Paid Losses Attorney Involvement Injury Type Cause Type County

45. 45 Box Plot

46. 46 Box and Whisker Plot

47. 47 Histogram/Frequency Polygon

48. 48 Categorical Data – Bar Plots

49. 49 Descriptive Statistics Year of Date Licensed has minimum and maximums that are impossible

50. 50 Agenda Literature Review Horror Stories Survey Experiment Actions Conclusions

51. 51 Conclusions Anecdotal horror stories illustrate possible dramatic impact of data quality problems Data quality survey suggests data quality issues have a significant cost on actuarial work Data quality experiment shows data quality issues have significant effect on accuracy of results Data Quality Working Party urges actuaries to become data quality advocates within their organizations Techniques from Exploratory Data Analysis can be used to detect data anomalies

52. 52 Dirty Data on Both Sides of the Pond: Results of the GIRO Working Party on Data Quality Questions?