Introduction to Casualty Actuarial Science

Ken Fikes, FCAS, MAAA Director of Property & Casualty

Email: [email protected]

Ken Fikes, FCAS, MAAA

Casualty Actuarial Science

Two major areas are measuring 1. Written Premium Risk
Pricing

2. Earned Premium Risk

Reserving

Ken Fikes, FCAS, MAAA

Definitions

What is a Loss Reserve?

Amount necessary to settle unpaid claims

Why are Loss Reserves Important?

Accurate evaluation of financial condition & underwriting income

Ken Fikes, FCAS, MAAA

Definitions

Accounting Aspects of Loss Reserves

Balance Sheet

Assets

Liabilities Surplus

Ken Fikes, FCAS, MAAA

Definitions

Case Reserves
Claim reported but not yet paid Assigned a value by a claims adjuster or by formula

Bulk + IBNR reserves include:

Reserves for claims not yet reported (pure IBNR) Claims in transit Development on known claims Reserves for reopened claims

Ken Fikes, FCAS, MAAA

Life Cycle of a Claim Reserve

7/11/01 Accident reported 4/2/01 Accident occurs Pure IBNR Claims in Transit 8/1/01 Accident entered into records as $1,000 Formula Reserve

8/18/02 Settlement agreed $30,000 Case Reserve

1/1/02 Estimate revised $25,000 Case Reserve

10/5/01 Individual reserve established $10,000 Case Reserve

8/25/02 Payment sent $30,000 Case Reserve

9/2/02 Claim draft clears Closed

Ken Fikes, FCAS, MAAA

Other Considerations

Factors Affecting Loss Reserves

External or Environmental
Society Regulation Judiciary Seasonality Residual Market Inflation Economy

Ken Fikes, FCAS, MAAA

Basic Reserving Techniques

Expected Loss Ratio Method Loss Development Method Bornhuetter/Ferguson Method

Ken Fikes, FCAS, MAAA

Expected Loss Ratio Method

EXPECTED LOSS RATIO (ELR) The anticipated ratio of projected ultimate losses to earned premiums. Sources: Pricing assumptions Industry data

Ken Fikes, FCAS, MAAA

Expected Loss Ratio Method

EXAMPLE OF ELR USING PRICING ASSUMPTIONS

Commissions Taxes General Expenses Profit

20% 5% 12% (2%)

Total 35% Amount to pay for loss & loss expense ---- 65% of premium
Ken Fikes, FCAS, MAAA 10

Expected Loss Ratio Method

Estimating Reserves Based on ELR - Example
Earned Premium Expected Loss Ratio Paid Losses = $100,000 = 0.65 = $10,000

Total

= ($100,000 x 0.65) - $10,000

Reserve = $65,000 - $10,000 = $55,000

Ken Fikes, FCAS, MAAA

11

Expected Loss Ratio Method

Estimating Reserves Based on ELR
Use when you have no history such as:
New product lines Radical changes in product lines Immature accident years for long tailed lines

Can generate negative reserves if Ultimate Losses < Paid Losses

Ken Fikes, FCAS, MAAA

12

Basic Reserving Techniques: Definitions

Loss Development
The financial activity on claims from the time they occur to the time they are eventually settled and paid.

Triangles
Compiled to measure the changes in cumulative claim activity over time in order to estimate patterns of future activity.

Loss Development Factor

The ratio of losses at successive evaluations for a defined group of claims (e.g. accident year).

Ken Fikes, FCAS, MAAA

13

Basic Reserving Techniques:

Compilation of Paid Loss Triangle

The losses are sorted by the year in which the accident occurred. The losses are summed at the end of each year. Losses paid to date are shown on the most recent diagonal. The data is organized in this way to highlight historical patterns.
14

Ken Fikes, FCAS, MAAA

Basic Reserving Techniques:

Compilation of Paid Loss Triangle
The goal is to estimate the total amount that will ultimately be

paid

Accident Year 1996 1997 1998 1999 2000 2001

12 3,780 4,212 4,901 5,708 6,093 6,962

Cumulative Paid Losses ($000 Omitted) Development Stage in Months 24 36 48 60 6,671 7,541 8,864 10,268 11,172 8,156 9,351 10,987 12,699 9,205 10,639 12,458 9,990 11,536

72 10,508

Final Total Cost ??? ??? ??? ??? ??? ???

Ken Fikes, FCAS, MAAA

15

Basic Reserving Techniques:

Paid Loss Development Factors
Evaluation Interval in Months Accident Year 1996 1997 1998 1999 2000 2001 12-24 1.765 1.790 1.809 1.799 1.834 24-36 1.223 1.240 1.240 1.237 36-48 1.129 1.138 1.134 48-60 1.085 1.084 60-72 1.052 72 to Ultimate ???

Sample Calculation for Accident Year 1997: 12-to-24 Months 1.790 = 7,541 / 4,212

From the end of the accident year (at 12 months) to the end of the following year (at 24 months), paid losses for 1997 grew 79%. During the next year (from 24 to 36 months), paid losses experienced an additional 24% growth (or development) and so forth. Loss Development Factors (LDFs) are also known as: Age-to-Age factors Link Ratios
Ken Fikes, FCAS, MAAA 16

Basic Reserving Techniques:

Paid Loss Development Factors
Evaluation Interval in Months Accident Year 12-24 24-36 1996 1.765 1.223 1997 1.790 1.240 1998 1.809 1.240 1999 1.799 1.237 2000 1.834 2001 Simple Average - All Years 1.799 1.235 Simple Average - Latest 3 Years 1.814 1.239 36-48 1.129 1.138 1.134 48-60 1.085 1.084 60-72 1.052 72 to Ultimate ???

1.134

1.085

1.052

1.134

XXX

XXX

Simple Average - Excluding High & Low 1.799 1.239 1.134 Weighted Average - All Years 1.803 1.235

XXX

XXX

1.134

1.085

1.052

Selected Loss Development Factors 1.800 1.235 1.134

1.085

1.052

1.070
17

Ken Fikes, FCAS, MAAA

Basic Reserving Techniques:

Application of Paid LDM
Evaluation Interval in Months 12-24 LDFs 1.800 24-36 1.235 36-48 1.134 48-60 1.085 60-72 1.052 72 to Ultimate 1.070 Final Total Cost 11,244 12,985 15,215 17,588 19,109 21,435

Accident Year 1996 1997 1998 1999 2000 2001

12 3,780 4,212 4,901 5,708 6,093 6,962

Cumulative Paid Losses ($000 Omitted) Development Stage in Months 24 36 48 60 6,671 8,156 9,205 9,990 7,541 9,351 10,639 11,536 8,864 10,987 12,458 13,517 10,268 12,699 14,401 15,625 11,172 13,797 15,646 16,976 12,532 15,477 17,550 19,042

72 10,508 12,136 14,220 16,437 17,859 20,032

Sample Calculations for Accident Year 2001: At 24 Months: At 36 Months: 12,532 = 6,962 x 1.800 15,477 = 12,532 x 1.235 or 15,477 = 6,962 x 1.800 x 1.235

12 to Ult 3.079
Ken Fikes, FCAS, MAAA

Cumulative Development Factors 24 to Ult 36 to Ult 48 to Ult 60 to Ult 1.710 1.385 1.221 1.126

72 to Ult 1.070
18

Basic Reserving Techniques:

Paid LDM Projections & Reserves
Reserve Estimate @ 12/31/01 = $32.241 million
Actual Paid Losses @ 12/31/01 (2) 10,508 11,536 12,458 12,699 11,172 6,962 65,335 Cumulative Development Factors to Ultimate (4) 1.070 1.126 1.221 1.385 1.710 3.079 Estimated Ultimate Losses [(2) x (4)] (5) 11,244 12,985 15,215 17,588 19,109 21,435 97,576 Actual Paid Losses @ 12/31/01 (6) 10,508 11,536 12,458 12,699 11,172 6,962 65,335 Estimated Loss Reserves {(5) - (6)} (7) 736 1,449 2,757 4,889 7,937 14,473 32,241

Loss

Accident Year (1) 1996 1997 1998 1999 2000 2001 Total

Selected LDFs (3) 1.070 1.052 1.085 1.134 1.235 1.800

Ken Fikes, FCAS, MAAA

19

BORNHUETTER-FERGUSON APPROACH APPLIED TO A NON-INSURANCE EXAMPLE

Given the following, how many home runs will Barry Bonds hit this year? You initially expected he would hit 40 home runs this year He has hit 20 home runs through 40 games There are 160 games in a season Three pieces of information are need to perform a Bornhuetter-Ferguson (B-F) projection: Expected Ultimate Value Cumulative Loss Development Factor Amount Incurred To Date

Ken Fikes, FCAS, MAAA

20

BORNHUETTER-FERGUSON APPROACH APPLIED TO A NON-INSURANCE EXAMPLE

The three pieces of information for our example : Before the season started, how many home runs would we have expected Barry Bonds to hit? Expected Ultimate Value = 40

To project season total from current statistics, multiply the current statistics by 4 since the season is 1/4 completed. Cumulative Loss Development Factor = 4.000 He has already hit 20 home runs. Amount Incurred To Date = 20

Ken Fikes, FCAS, MAAA

21

BORNHUETTER-FERGUSON APPROACH APPLIED TO A NON-INSURANCE EXAMPLE

B-F Projection: Ultimate Value = (Expected Value*IBNR Factor)+(Inc. to Date) IBNR Factor = 1.000 - (1.000/LDF) = 1.000 - (1.000/4.000) = .75 (In Other Words, 75% of the season is left to be played) Ultimate Value = (40 * .75) + 20 = 50 The B-F Method projects that Barry Bonds will hit 50 home runs this year. Games 0-40 Games 41-80 Games 81-120 Games 121-160 20 Home Runs 10 Home Runs 10 Home Runs 10 Home Runs

Ken Fikes, FCAS, MAAA

22

BORNHUETTER-FERGUSON APPROACH APPLIED TO A NON-INSURANCE EXAMPLE

Comparison of B-F with Two Other Methods

Incurred Loss Development Method Ultimate Value = Incurred To Date * Cumulative LDF = 20 * 4.000 = 80 Home Runs Games 0-40 20 Home Runs Games 41-80 20 Home Runs Games 81-120 20 Home Runs Games 121-160 20 Home Runs

Expected Loss Ratio Method Ultimate Value = Expected Value = 40 Home Runs Games 0-40 10 Home Runs Games 41-80 10 Home Runs Games 81-120 10 Home Runs Games 121-160 10 Home Runs

Ken Fikes, FCAS, MAAA

23

Example
EXAMPLE You are given the following losses evaluated at 12/31/2006. Use the paid loss development method to estimate the required reserves by accident year. Assume all losses are fully developed at 60 months. Cumulative Paid Losses ($000 Omitted) Development Stage in Months 24 36 48 6,000 6,400 7,000 7,600 9,000 9,600 10,500 10,800 11,520

Accident Year 2002 2003 2004 2005 2006

12 3,000 3,200 3,500 3,800 5,000

60 11,340

Ken Fikes, FCAS, MAAA

24

Solution
12-24 2.000 Age-to-Age Development Factors 24-36 36-48 48-60 1.500 1.200 1.050 Cumulative Development Factors 24 to Ult 36 to Ult 48 to Ult 1.890 1.260 1.050 (1) Paid Losses @ 12/06 11,340 11,520 10,500 7,600 5,000 (2) Dev Factors to Ult 1.000 1.050 1.260 1.890 3.780 (3)=(1)*(2) Estimated Ultimate Losses 11,340 12,096 13,230 14,364 18,900 60-Ult 1.000

12 to Ult 3.780

60 to Ult 1.000 (4)=(3)-(1) Estimated Loss Reserve 576 2,730 6,764 13,900
25

Accident Year 2002 2003 2004 2005 2006

Ken Fikes, FCAS, MAAA

Further Reading
For additional information on Loss Reserving, see the following references at www.casact.org/admissions/syllabus/exam6.pdf
Wiser, et al., Loss Reserving, Foundations of Casualty Actuarial Science (Fourth Edition), Casualty Actuarial Society, 2001, Chapter 5, pp. 197-285. Bornhuetter, R.L; and Ferguson, R.E., The Actuary and IBNR, PCAS LIX, 1972, pp. 181195. Including discussions of paper: Cooper, W.P., PCAS LX, 1973, pp. 161-164; and White, H.G., PCAS LX 1973, pp. 165-168. Brosius, E., Loss Development Using Credibility, CAS Study Note, March 1993.

Ken Fikes, FCAS, MAAA

26