The$WC$Loss$DevelopmentTail$ - Confex
Transcript of The$WC$Loss$DevelopmentTail$ - Confex
![Page 2: The$WC$Loss$DevelopmentTail$ - Confex](https://reader030.fdocuments.in/reader030/viewer/2022012513/618bfd22736af804695ccaaa/html5/thumbnails/2.jpg)
A Typical WC Reserve Analysis: 15 DYs
AY Years of Development
Triangle of Historical Development Data
Development Factors Helpful
Tail
Region
![Page 3: The$WC$Loss$DevelopmentTail$ - Confex](https://reader030.fdocuments.in/reader030/viewer/2022012513/618bfd22736af804695ccaaa/html5/thumbnails/3.jpg)
Es?ma?ng the WC Tail @ DY 15
You have 15 year triangle & the latest large claim lis?ng. Your op?ons?
• Accept the case? • Extrapolate ILDFs and PLDFs? How? • Use external data? Different state? • Adjust the case? Use a primi?ve predic?ve model?
• Use more complex predic?ve models?
![Page 4: The$WC$Loss$DevelopmentTail$ - Confex](https://reader030.fdocuments.in/reader030/viewer/2022012513/618bfd22736af804695ccaaa/html5/thumbnails/4.jpg)
Accept the Case Reserve?
• What rate of medical infla?on was assumed, if any?
• Stair-‐stepping is quite common. • Medical condi?on oTen evolves with aging. • Reopened claims poten?al. • Expected value of future payments is typically 25%-‐60% higher than the sum of projected payments un?l age at death.
![Page 5: The$WC$Loss$DevelopmentTail$ - Confex](https://reader030.fdocuments.in/reader030/viewer/2022012513/618bfd22736af804695ccaaa/html5/thumbnails/5.jpg)
Adjus?ng the Case Reserve Using A Large Claim Lis?ng
Data in the large claim lis?ng (AY, DY, Age at Injury or Current Age, Paid to Date, Case Reserve, Injury Descrip?on, Gender)
A Typical Approach: • Split reserve into medical and indemnity. • What rate of future medical cost escala?on was assumed by
the claims adjuster? • Remove adjuster’s medical cost escala?on adjustment, using
the claimant’s life expectancy. • Assume constant on-‐level incremental paids un?l claimant
dies or claim is closed, and inflate future medical payments at your chosen rate of medical cost escala?on.
![Page 6: The$WC$Loss$DevelopmentTail$ - Confex](https://reader030.fdocuments.in/reader030/viewer/2022012513/618bfd22736af804695ccaaa/html5/thumbnails/6.jpg)
Offse_ng Factors?
• Permanent disability claims may close even if the claimant keeps living.
• Claimant’s condi?on may change, especially if they become elderly.
• In some states, at age 65 medicare picks up WC medical. Not true in Oregon & Washington.
• Can test by comparing actual PLDFs with PLDFs expected if claims close only due to death and on-‐level medical costs remain constant for future years—for DYs 30+.
![Page 7: The$WC$Loss$DevelopmentTail$ - Confex](https://reader030.fdocuments.in/reader030/viewer/2022012513/618bfd22736af804695ccaaa/html5/thumbnails/7.jpg)
SAIF’s Actual PLDFs – 1.0 SAIF PLDFs Less 1.0
0.0000
0.0100
0.0200
0.0300
0.0400
0.0500
0.0600
0.0700
6 7 8 9 10 11 12 13 14
Year of Development
PLDF Le
ss 1.0
![Page 8: The$WC$Loss$DevelopmentTail$ - Confex](https://reader030.fdocuments.in/reader030/viewer/2022012513/618bfd22736af804695ccaaa/html5/thumbnails/8.jpg)
Payout Patterns--Lifetime v. Short Term MPD Payments for a Single Accident Year
0.00
5.00
10.00
15.00
20.00
25.00
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49
Development Year (DY)
$ M
illio
ns
Lifetime
ShortTerm
![Page 9: The$WC$Loss$DevelopmentTail$ - Confex](https://reader030.fdocuments.in/reader030/viewer/2022012513/618bfd22736af804695ccaaa/html5/thumbnails/9.jpg)
Death Rate v. Inflation by DY
0.0%
5.0%
10.0%
15.0%
20.0%
25.0%
22 26 30 34 38 42 46 50 54 58 62 66 70
Development Year (DY)
9% In
flatio
n v.
Dea
th R
ate
Death9% Infl
![Page 10: The$WC$Loss$DevelopmentTail$ - Confex](https://reader030.fdocuments.in/reader030/viewer/2022012513/618bfd22736af804695ccaaa/html5/thumbnails/10.jpg)
We assume a brontosaurus tail.
![Page 11: The$WC$Loss$DevelopmentTail$ - Confex](https://reader030.fdocuments.in/reader030/viewer/2022012513/618bfd22736af804695ccaaa/html5/thumbnails/11.jpg)
Model v. Actual SAIF PLDFs Less 1.0
0.00
0.01
0.02
0.03
0.04
0.05
0.06
0.07
6 10 14 18 22 26 30 34 38 42 46 50 54 58
Year of Development
PLDF
Les
s 1.0
ModelSAIF
S
![Page 12: The$WC$Loss$DevelopmentTail$ - Confex](https://reader030.fdocuments.in/reader030/viewer/2022012513/618bfd22736af804695ccaaa/html5/thumbnails/12.jpg)
Mortality Model vs. SAIF’s Actual
• 9% rate of future medical cost escala?on assumed.
• Mortality rates of general popula?on assumed.
• Model fit well out to development year (DY) 40.
• Model no?ceably underes?mated actual development beyond DY 40.
![Page 13: The$WC$Loss$DevelopmentTail$ - Confex](https://reader030.fdocuments.in/reader030/viewer/2022012513/618bfd22736af804695ccaaa/html5/thumbnails/13.jpg)
MPD payments: Stegosaurus tail.
![Page 14: The$WC$Loss$DevelopmentTail$ - Confex](https://reader030.fdocuments.in/reader030/viewer/2022012513/618bfd22736af804695ccaaa/html5/thumbnails/14.jpg)
Washington State Fund PLDFs – 1.0
0.0%
0.1%
0.2%
0.3%
0.4%
0.5%
0.6%
10.5 12.8 15 17.3 19.5 21.8 24 26.3 28.5 30.8 33 35.3 37.5 39.8 42 44.3 46.5 48.8 51 53.3 55.5 57.8 60
Washington State FundMedical Tail
Incr
emen
tal P
ayou
t
Maturity (Years)
![Page 15: The$WC$Loss$DevelopmentTail$ - Confex](https://reader030.fdocuments.in/reader030/viewer/2022012513/618bfd22736af804695ccaaa/html5/thumbnails/15.jpg)
Washington State FundMedical Tail
0.00
0.01
0.02
0.03
0.04
0.056 10 14 18 22 26 30 34 38 42 46 50 54 58
Year of Development
PLD
F L
ess 1
.0
Washington State FundMedical Tail
0.00
0.01
0.02
0.03
0.04
0.056 10 14 18 22 26 30 34 38 42 46 50 54 58
Year of Development
PLD
F L
ess 1
.0
![Page 16: The$WC$Loss$DevelopmentTail$ - Confex](https://reader030.fdocuments.in/reader030/viewer/2022012513/618bfd22736af804695ccaaa/html5/thumbnails/16.jpg)
Model v. Actual SAIF PLDFs Less 1.0
0.00
0.01
0.02
0.03
0.04
0.05
6 10 14 18 22 26 30 34 38 42 46 50 54 58
Year of Development
PLD
F L
ess 1
.0
ModelSAIF
S
![Page 17: The$WC$Loss$DevelopmentTail$ - Confex](https://reader030.fdocuments.in/reader030/viewer/2022012513/618bfd22736af804695ccaaa/html5/thumbnails/17.jpg)
SAIF PLDFs Less 1.0Future Cost v. On-Level
0.00
0.01
0.02
0.03
0.046 10 14 18 22 26 30 34 38 42 46 50 54 58
Year of Development
PLD
F L
ess 1
.0
Future CostOn-Level
SA
![Page 18: The$WC$Loss$DevelopmentTail$ - Confex](https://reader030.fdocuments.in/reader030/viewer/2022012513/618bfd22736af804695ccaaa/html5/thumbnails/18.jpg)
Model v. Empirical SAIF Closure Rates
0.00
0.05
0.10
0.15
0.20
0.25
0.30
20 24 28 32 36 40 44 48 52 56 60
Year of Development
Ann
ual C
losu
re R
ate
SAIF 7-YrSAIF 14-YrModel
SAI
![Page 19: The$WC$Loss$DevelopmentTail$ - Confex](https://reader030.fdocuments.in/reader030/viewer/2022012513/618bfd22736af804695ccaaa/html5/thumbnails/19.jpg)
On-Level Incremental Paid per Claim
02,0004,0006,0008,00010,00012,00014,000
16 21 26 31 36 41 46 51 56
Development Year (DY)
($)
![Page 20: The$WC$Loss$DevelopmentTail$ - Confex](https://reader030.fdocuments.in/reader030/viewer/2022012513/618bfd22736af804695ccaaa/html5/thumbnails/20.jpg)
Washington State Fund On-‐Level Incremental Severi?es by DY
WALI Medical Excluding Hearing Loss Trended Incremental Paid Per Active Claim
1000
2000
3000
4000
5000
6000
7000
8000
5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59 61 63
DY Center of Average
Incr
emen
tal S
ever
ity
9 Yr Weighted Avg
![Page 21: The$WC$Loss$DevelopmentTail$ - Confex](https://reader030.fdocuments.in/reader030/viewer/2022012513/618bfd22736af804695ccaaa/html5/thumbnails/21.jpg)
Is There an Elder Bulge?
• Permanently disabled claimant becomes elderly.
• Spouse can’t con?nue to provide as much home care.
• WC carrier picks up cost of home care. • Effects of exis?ng disabili?es from work tend to more seriously affect quality of life as the claimant becomes elderly.
![Page 22: The$WC$Loss$DevelopmentTail$ - Confex](https://reader030.fdocuments.in/reader030/viewer/2022012513/618bfd22736af804695ccaaa/html5/thumbnails/22.jpg)
Another reason for the rise in on-‐level incremental severi?es for late DYs.
• New research: Examined on-‐level incremental severi?es by age-‐at-‐injury, gender, claim type & DY.
• As the DY increases, the composi?on of surviving claimants by age-‐at-‐injury shiTs drama?cally to lower ages, where on-‐level severi?es are much higher.
• Younger workers are given the hazardous jobs.
![Page 23: The$WC$Loss$DevelopmentTail$ - Confex](https://reader030.fdocuments.in/reader030/viewer/2022012513/618bfd22736af804695ccaaa/html5/thumbnails/23.jpg)
Average On-‐Level Incremental Paid PTD/PPD, Male/Female
Avg Age DYs DYs DYs DYs
at Injury 16-25 26-40 41+ 16+
15-35 5,957 8,579 16,094 7,482
36-45 5,495 6,707 5,952
46+ 2,647 5,132 3,509
All 4,630 7,126 11,749
![Page 24: The$WC$Loss$DevelopmentTail$ - Confex](https://reader030.fdocuments.in/reader030/viewer/2022012513/618bfd22736af804695ccaaa/html5/thumbnails/24.jpg)
Mul?ple Regression
• Dependent Variable:
On Level Incremental Severity
• Independent Variables:
Age-‐at-‐Injury, DY
![Page 25: The$WC$Loss$DevelopmentTail$ - Confex](https://reader030.fdocuments.in/reader030/viewer/2022012513/618bfd22736af804695ccaaa/html5/thumbnails/25.jpg)
% Young at Injury by DY
DY
20
30
40
50
60
% Young
At Injury
46%
57%
71%
87%
98%
![Page 26: The$WC$Loss$DevelopmentTail$ - Confex](https://reader030.fdocuments.in/reader030/viewer/2022012513/618bfd22736af804695ccaaa/html5/thumbnails/26.jpg)
$7,000 Young & $3,500 Older
DY 20
DY 30
DY 40
DY 50
DY 60
% Injured Young
46%
57%
71%
87%
98%
Wtd. Severity
5,100
5,495
5,985
6,545
6,930
![Page 27: The$WC$Loss$DevelopmentTail$ - Confex](https://reader030.fdocuments.in/reader030/viewer/2022012513/618bfd22736af804695ccaaa/html5/thumbnails/27.jpg)
Average On-‐Level Incremental Paid PTD/PPD, Male/Female
Avg Age DYs DYs DYs DYs
at Injury 16-25 26-40 41+ 16+
15-35 5,957 8,579 16,094 7,482
36-45 5,495 6,707 5,952
46+ 2,647 5,132 3,509
All 4,630 7,126 11,749
![Page 28: The$WC$Loss$DevelopmentTail$ - Confex](https://reader030.fdocuments.in/reader030/viewer/2022012513/618bfd22736af804695ccaaa/html5/thumbnails/28.jpg)
Es?ma?ng the WC Tail, PCAS 2005
Expected value of future payments is typically 25%-‐60% higher than the sum of projected payments un?l age at death.
See Sec?on 8 of Paper
![Page 29: The$WC$Loss$DevelopmentTail$ - Confex](https://reader030.fdocuments.in/reader030/viewer/2022012513/618bfd22736af804695ccaaa/html5/thumbnails/29.jpg)
A Very Simple PPD Claim • Jeremy’s right leg amputated in 2004 because of work injury. He is 55.
• Ar?ficial leg costs $1,000. • Leg must be replaced every 15 years, at double the prior cost.
• Jeremy is expected to live un?l age 78, so adjuster sets up a case reserve to cover one replacement leg when Jeremy is 70.
![Page 30: The$WC$Loss$DevelopmentTail$ - Confex](https://reader030.fdocuments.in/reader030/viewer/2022012513/618bfd22736af804695ccaaa/html5/thumbnails/30.jpg)
Three Scenarios
Scenario (Age at Death)
Number of Legs
Cost of New
Leg
Total Future Payments
< 70
0
$0
$0
70 - 84
1
$2,000
$2,000
85 +
2
$4,000
$6,000
![Page 31: The$WC$Loss$DevelopmentTail$ - Confex](https://reader030.fdocuments.in/reader030/viewer/2022012513/618bfd22736af804695ccaaa/html5/thumbnails/31.jpg)
Expected Value – 4.7% Med. Infl. Age at Death Future
Payments
Probability Fut. Pay x Probability
< 70
$0
25 %
$0
70 - 84
$2,000
50 %
$1,000
85 +
$6,000
25 %
$1,500
Expected Value of Future Payments
$2,500
![Page 32: The$WC$Loss$DevelopmentTail$ - Confex](https://reader030.fdocuments.in/reader030/viewer/2022012513/618bfd22736af804695ccaaa/html5/thumbnails/32.jpg)
Expected Value – 9.7% Med. Infl. Age at Death Future
Payments
Probability Fut. Pay x Probability
< 70
$0
25 %
$0
70 - 84
$4,000
50 %
$2,000
85 +
$20,000
25 %
$5,000
Expected Value of Future Payments
$7,000
![Page 33: The$WC$Loss$DevelopmentTail$ - Confex](https://reader030.fdocuments.in/reader030/viewer/2022012513/618bfd22736af804695ccaaa/html5/thumbnails/33.jpg)
Wrestling with Low Credibility and a Few Large Incremental Paids
AY DY 46 DY 47 DY 48 DY 49 DY 50
1960 20 50 40
1961 50 225 55
1962 45 20 60
![Page 34: The$WC$Loss$DevelopmentTail$ - Confex](https://reader030.fdocuments.in/reader030/viewer/2022012513/618bfd22736af804695ccaaa/html5/thumbnails/34.jpg)
Wrestling with Low Credibility and a Few Large Incremental Paids
• Timing of the large payments within any given AY tends to be random among the latest 8 calendar years. • Cap individual incremental paids at 1 standard devia?on above the mean and re-‐distribute the excess equally among all the latest 8 diagonals, separately by AY.
![Page 35: The$WC$Loss$DevelopmentTail$ - Confex](https://reader030.fdocuments.in/reader030/viewer/2022012513/618bfd22736af804695ccaaa/html5/thumbnails/35.jpg)
Factors Offset One Another?
Tail FaEeners: 1. Elder Bulge 2. Age-‐at-‐Injury Effects 3. Expected >> Pay Un?l Exp. Year of Death
Tail Thinners: • Permanent Disability Claims Close for Reasons
Other than Death. • Medicare Assumes Responsibility (Some States) • Increasing %-‐age of Females
![Page 36: The$WC$Loss$DevelopmentTail$ - Confex](https://reader030.fdocuments.in/reader030/viewer/2022012513/618bfd22736af804695ccaaa/html5/thumbnails/36.jpg)
CAS Paper Indica?ons for SAIF
Model v. Actual SAIF PLDFs Less 1.0
0.00
0.01
0.02
0.03
0.04
0.05
0.06
0.07
6 10 14 18 22 26 30 34 38 42 46 50 54 58
Year of Development
PLDF
Les
s 1.0
ModelSAIF
S
Model v. Actual SAIF PLDFs Less 1.0
0.00
0.01
0.02
0.03
0.04
0.05
0.06
0.07
6 10 14 18 22 26 30 34 38 42 46 50 54 58
Year of Development
PLDF
Les
s 1.0
ModelSAIF
S
![Page 37: The$WC$Loss$DevelopmentTail$ - Confex](https://reader030.fdocuments.in/reader030/viewer/2022012513/618bfd22736af804695ccaaa/html5/thumbnails/37.jpg)
Latest Indica?ons for SAIF
Model v. Actual SAIF PLDFs Less 1.0
0.00
0.01
0.02
0.03
0.04
0.05
6 10 14 18 22 26 30 34 38 42 46 50 54 58
Year of Development
PLD
F L
ess 1
.0
ModelSAIF
S
![Page 38: The$WC$Loss$DevelopmentTail$ - Confex](https://reader030.fdocuments.in/reader030/viewer/2022012513/618bfd22736af804695ccaaa/html5/thumbnails/38.jpg)
Washington State Fund PLDFs – 1.0
0.0%
0.1%
0.2%
0.3%
0.4%
0.5%
0.6%
10.5 12.8 15 17.3 19.5 21.8 24 26.3 28.5 30.8 33 35.3 37.5 39.8 42 44.3 46.5 48.8 51 53.3 55.5 57.8 60
Washington State FundMedical Tail
Incr
emen
tal P
ayou
t
Maturity (Years)