Understanding and Analyzing
Construction Defect Claims
Fin 431 Property-Liability Insurance
University of Illinois
April 22, 2008
Ronald T. Kozlowski, FCAS, MAAA
© 2008 Towers Perrin
Towers Perrin
 Towers Perrin is a professional services firm that helps organizations
improve performance through effective people, risk and financial
management.
 Human Capital Consulting
 Risk & Financial Services Consulting
 Retirement
 Life Actuarial Consulting
 Health & Welfare
 Property & Casualty Actuarial Consulting
 Executive Compensation
 Enterprise Risk Management
 Workforce Effectiveness
 Financial Modeling
 HR Effectiveness
 Reinsurance Placement and Analysis
 Total Rewards Effectiveness
 Mergers, Acquisition & Restructuring
 Research & Surveys
 Employee Communication
 Change Management
 Mergers, Acquisition &
Restructuring
© 2008 Towers Perrin
2
Ronald T. Kozlowski  Biography
 1982-86
Graduated University of Illinois with Bachelors of Science in
Actuarial Science in 1986
 1986-92
Aetna Life & Casualty (Hartford, CT)
— Personal Lines Pricing
— Workers Compensation Ratemaking, Reserving & Planning
— Commercial Auto Ratemaking and Reserving
— Personal Auto Reserving
— Corporate Actuarial
 1992-08
Towers Perrin (“Tillinghast”)
‘92-95  Consulting Actuary (Hartford, CT)
’95-98  Consulting Actuary (Atlanta, GA)
’98-08  Principal & Consulting Actuary (San Francisco, CA)
© 2008 Towers Perrin
3
Ronald T. Kozlowski  Biography
 1992-08 Towers Perrin (“Tillinghast”)
— Insurance companies (e.g., ACE, Allstate, AXA, CNA, Chubb,
Countrywide, Fireman’s Fund, Liberty Mutual, Safeco, XL)
— Other clients (e.g., Allergan, Apple, Bank of America, Duty
Free Stores, Kaiser, KBHome, Ford, Levi Strauss & Co.,
Nordstrom, Novartis, Pella Windows, State of Oregon –
University system)
— Some of my most interesting assignments
– Catastrophe models (earthquake, hurricane, European
straight-line winds, wildfires, winter storms, tornadoes)
– Acquisition evaluation of Malaysian insurance company
for European insurance company
– Evaluation of movie completion bond insurer
– Analyzing construction defect claims for national
homebuilders
© 2008 Towers Perrin
4
California Population Growth and Housing Supply Shortage
 In the late 70s through early 90s, California experienced unprecedented
population and housing growth
 CA population growth was twice the US population growth rate during
parts of period
 Demand for housing exceeded supply
 Construction of multi-family units (condos, townhomes) increased
significantly
 Builders stepped up production
 Unskilled construction labor
 “Cut corners” - cheaper materials and
shorter construction time
 Less supervision
© 2008 Towers Perrin
5
Litigation Ensues
 Aggressive plaintiffs bar in California
 Success in early suits funded additional suits
 Unfavorable legal decisions (Montrose, Stonewall)
 Construction of multi-family units (condos, townhomes) encourages large
cases
 multi-family units four times more likely to sue
 Homeowners associations
 sold on idea by aggressive lawyers
 potential suits against condo board for failure to take action
 Spreads into other states
 lawyers move east
 states use different “theories of liability” and definition of “occurrence”
© 2008 Towers Perrin
6
Litigation Ensues (cont’d.)
© 2008 Towers Perrin
7
Definitions
 Premises & Operations Coverage - Part of a business liability policy that
covers an Insured for bodily injury or property damage liability to members
of the public while they are on business premises.
 Competed Operations Coverage - This form of liability insurance provides
coverage for bodily injury and property damage rising from completed or
abandoned operations…
 An commercial general liability insurance policy…
 is not intended to be a warranty policy
 covers claims arising from work performed by subcontractors
 A warranty policy covers the insured’s own work.
© 2008 Towers Perrin
8
Types of Defects
 Many courts have recognized two primary categories of defects for which
damages are recoverable:
 Defects in design, workmanship and materials
 Faulty drainage
 Improper landscaping
 Inadequate environmental
controls
 Faulty electrical wiring
 Insufficient insulation
 Defective plumbing
 Landslide and earth settlement problems
and irrigation
 Improper materials
 Structural failure or collapse
 Expansive soils
 Land sliding
 Underground water or streams
 Surface failure
 Vertical settlement
 Improper compaction
 Horizontal movement
 Inadequate grading and drainage
 Patent defects vs. latent defects
 CGL covers damage resulting from defects
 No coverage from “your own work”; different from warranty
© 2008 Towers Perrin
9
Type of Defects (cont'd.)
Most Common Types of Construction Defects
21%
Plumbing, Draining,
Other Leaks
Building Structure
21%
Infrastructure
10%
19%
12%
Roof Leaks and
Defects
Internal Systems
Other
17%
Source: California Department of Real Estate/IRMI Conference
© 2008 Towers Perrin
10
Premises & operations and completed operations (construction
defects) have significantly different reporting patterns
 The majority of non construction defect losses are reported as of
four years, whereas construction defect claims have a
significantly slower reporting pattern.
Incremental Reporting Patterns
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Years
Construction defect
Non construction defect
Based on California reporting patterns
© 2008 Towers Perrin
11
Products and Completed Operations Loss Costs for Specific
Contractors Classes 2000-2006 - Oregon
Class
Siding Installation
Ceiling or Wall Installation - Metal
Painting - Exterior Bldgs. - > 3 stories
Sign Painting or Lettering - Inside Bldgs.
Plastering or Stucco Work
Insulation Work - Plastic
Painting - Steel Structures or Bridges
Dry Wall or Wallboard Installation
Plumbing - Commercial and Industrial
Roofing - Residential
Concrete Construction
Plumbing - Residential or Domestic
Masonry
Electrical Work - Within Bldgs.
Roofing - Commercial
Driveway, Parking Area, or Sidewalk
Floor Covering Installation
Fence Erection Contractors
Heating and Air Conditioning Systems
Tile, Stone, Marble, Mosaic, or Terrazzo Work
Sheet Metal Work - Shop and Outside
Electrical Apparatus - Installation, Service, & Repair
© 2008 Towers Perrin
2000-2006
Loss Cost
Change
411.9%
356.0%
347.8%
335.4%
315.9%
300.0%
267.4%
266.1%
243.1%
222.7%
203.5%
186.6%
155.5%
136.5%
98.7%
76.3%
65.1%
62.3%
55.5%
51.6%
26.0%
12.9%
Annual
Loss Cost
Change
31.3%
28.8%
28.4%
27.8%
26.8%
26.0%
24.2%
24.1%
22.8%
21.6%
20.3%
19.2%
16.9%
15.4%
12.1%
9.9%
8.7%
8.4%
7.6%
7.2%
3.9%
2.0%
12
Contractors general liability
 Insureds: homebuilders, general contractors, artisans, suppliers
 Types: practice, project, wraps
 Residential vs. commercial
 Single family vs. multi-family
 Primary or excess
 Policies: policy forms, coverages, exclusions, additional insured
endorsements
 Coverage: premises & operations, completed operations (construction
defect), warranty
 Exposure period: occurrence, claims-made,
sunset provisions
 Exposures: homes, sales/revenue, budgets, payroll
 Decreasing housing prices, less construction
 Competition, new entrants, decreasing rates
© 2008 Towers Perrin
13
Why is analyzing CD claims so complicated?
 Definition
 Construction defect vs. warranty
 Reporting lag/statute of limitation
 Multiple claimants/defendants/insurance companies
 Effects of court decisions
 Changes in policy form and introduction of exclusions
 Additional insured endorsements
 Coding to an accident year
 Specific states developing into problematic states
 “Notice & opportunity to repair laws”
 Case reserving practices
 Insolvencies/possible impact of RRGs
 Implementation of claims-made and sunset provisions
© 2008 Towers Perrin
14
Why is analyzing CD claims so complicated? (cont’d.)
 Binding arbitration clauses
 Increased use of SIRs/deductibles
 Wrap policies
 Quality of construction
 Reallocation of prior cost allocations
 Continuously changing environment
 Overall concern that past may not be predictive
of the future
© 2008 Towers Perrin
15
Definition of a Construction Defect
 There is no industry standard definition of what constitutes a construction
defect
 According to California State Jury Instructions, a construction defect is the
Failure of a building or any building component
a) to be erected in a reasonably workmanlike manner
or b) to perform in a manner intended by the manufacturer or
reasonably expected by the buyer which causes proximate
damage to the structure
© 2008 Towers Perrin
16
Definition of a Construction Defect (cont’d.)
 Types of defects
 Patent defects
— apparent with reasonable inspection
— statute of limitations requires claim to be submitted within 2 to 3
years of discovery
 Latent defects
— defect is not apparent by reasonable inspection
— more time is allowed to submit a claim, in some cases 10 years
after completion (CA). For comparison purposes, AZ is 8 years,
and WA is 6 years (confirmed by WA supreme court in
September 2001)
 Distinguishing construction defect from warranty
© 2008 Towers Perrin
17
Statutes of Repose and Limitation
 Statutes of Repose - Statutes of repose terminate a manufacturer's
liability for defective products after a statutorily specified number of
years. A person injured after the cut-off date has no recourse to hold the
manufacturer of the defective product accountable.
 Statute of Limitation – A law which sets the maximum period which one
can wait before filing a lawsuit, depending on the type of case or claim.
The periods vary by state. If the lawsuit or claim is not filed before the
statutory deadline, the right to sue or make a claim is forever barred.
© 2008 Towers Perrin
18
Statutes of Repose
NH
WA
VT
MT
ND
ME
MN
OR
MA
ID
WY
WI
SD
NV
UT
RI
MI
PA
IA
NE
CA
NY
IL
CO
KS
OH
IN
DC
WV
MO
NC
TN
AZ
OK
NM
AR
DE
MD
SC
MS
TX
VA
KY
CT
NJ
AL
GA
Years
LA
AK
FL
4-5
6-9
10
> 10
HI
© 2008 Towers Perrin
19
Important Legal Cases (California)
 I - Montrose Chemical Corp v. Superior Court (Canadian Universal Insurance Co) –
1993
 an insurer must defend an insured in a case involving the discharge of hazardous
substances
 II - Montrose Chemical Corp v. Admiral Insurance – 1995
 continuous trigger
 Stonewall Insurance Co. v. City of Palos Verdes Estates – 1996
 Montrose applied to construction defects
 Aas v. William Lyon Company – 2000
 defect without resultant damage is not sufficient for a liability claim
 Presley Homes v. American States Ins. Co. – 2001
 duty to defend the entire action applies if there is a mere potential for coverage
 Lantzy v. Centex Homes – 2003
 addressed “tolling of the statutes”
 L-J v. Bituminous Fire and Marine Ins. Co. – 2004
 no coverage provided to your own work (“your work” exclusion)
© 2008 Towers Perrin
20
Changes in Policies
 Some policy changes/endorsements shift coverage whereas others
eliminate coverage
 Montrose endorsement
 Known loss provisions
 Prior work exclusions
 “Damage to Your Work” exclusion
 Additional insured endorsements
 EIFS exclusion
 Mold exclusions
 Earth movement exclusions
 Residential construction exclusions
 Need to consider how other insurer’s policies
are changing
 Most companies only estimate benefits
© 2008 Towers Perrin
21
Additional Insured Endorsement
 Issues*
 Does coverage specify that liability must “arise out of the named
insured’s ‘act or omissions’ ”?
 Does language limit coverage to the AI’s vicarious liability for acts of the
named insured? Vicarious liability can be defined as “the responsibility
of the superior for the acts of their subordinate”.
 Does coverage terminate for the additional insured when the named
insured’s work is completed?
 Coverage for “ongoing operations” vs. “completed operations”?
— Is policy silent, “includes” or “excludes”
 Cost sharing arrangements
 Should be decided early in the suit
 Pro-rata, tiered, percentage
 Are you tracking use of AI endorsements?
 Can you estimate impacts?
*Source: American Re Construction Defect: Resource Guide, 2005
© 2008 Towers Perrin
22
States where insurance companies
have concerns over construction defects
 1st tier – California
 2nd tier – AK, AZ, CO, FL, HI, MN, NJ, NV, NM, NC, OR, SC, TX, WA
 3rd tier – all other states
 Katrina, Rita, and Wilma affected states - LA, MS, AL, FL
 What to watch out for
 Rapid growth in population
 Rapid growth in construction
 Unskilled labor
 Legal environments
 Theories of liability
 Judicial environment
© 2008 Towers Perrin
23
“Notice and Opportunity to Repair” Legislation
 Generally provide builder with written notice and description of alleged
defects - 90 days before filing lawsuit
 Intent
 decrease frivolous law suits
 reduce legal costs
 California - Calderon Act - 1997
 homeowners association must provide notice of a claim to the developer
and to the members of its association before filing a lawsuit
 California - Steinberg Mandatory Negotiation Bill (7/2002)
 builders, subcontractors, insurers and suing homeowners will be
required to negotiate a solution to specific alleged defects in a timely
manner before a lawsuit can be filed
 California Senate Bill 800 (“Fix It” Law) – 2003
 established building standards to govern
claims against builders
 mandatory pre-lawsuit process
© 2008 Towers Perrin
24
“Notice and Opportunity to Repair” Legislation (cont'd.)
31 States with NOR Laws
NH
WA
MT
ME
VT
ND
MN
OR
MA
ID
NY
WI
SD
WY
MI
RI
PA
IA
NE
CT
OH
NV
UT
CA
IL
NJ
DC
IN
CO
WV
DE
VA
KS
MO
KY
MD
NC
TN
AZ
OK
NM
AR
SC
GA
MS
TX
AL
LA
AK
FL
Year Enacted
2002&Prior
2003
2004
2005
2006
2007
No law
HI
PR
VI
*Source: National Association of Home Builders (NAHB); if more than one NOR – chart shows year of latest reform
© 2008 Towers Perrin
25
Changes in Case Reserving Practices
 Actuaries look at the past to predict the future
 Significant changes in case reserving by insurers
 Movement from “generalist” claims staff to specialized units
 Less loss development due to higher initial case reserves
 Attention to reducing legal costs
— vendors (e.g., lawyers, expert witnesses)
– insurance companies want more control
– fixed fee arrangements
— Arbitration/mediation
 Why don’t some home builders estimate case reserves?
 Gross estimates versus net estimates
© 2008 Towers Perrin
26
Insolvencies
 Reliance, Legion, Frontier (United Capital)
 California Insurance Guarantee Association
 denies indemnity coverage if other insurance is available; may
provide some defense
 remaining solvent companies to share loss
 Residual impact on loss development and severities
 Future concerns
 Insurance companies and risk retention groups with low retentions
 Deductibles/self-insured retentions
 Claims-made policies and sunset provisions
© 2008 Towers Perrin
27
Why is analyzing CD claims so complicated?
 Definition
 Construction defect vs. warranty
 Reporting lag/statute of limitation
 Multiple claimants/defendants/insurance companies
 Effects of court decisions
 Changes in policy form and introduction of exclusions
 Additional insured endorsements
 Coding to an accident year
 Specific states developing into problematic states
 “Notice & opportunity to repair laws”
 Case reserving practices
 Insolvencies/possible impact of RRGs
 Implementation of claims-made and sunset provisions
© 2008 Towers Perrin
28
Why is analyzing CD claims so complicated? (cont’d.)
 Binding arbitration clauses
 Increased use of SIRs/deductibles
 Wrap policies
 Quality of construction
 Reallocation of prior cost allocations
 Continuously changing environment
 Overall concern that past may not be predictive
of the future
© 2008 Towers Perrin
29
Cost Allocation Matrix - illustrative
 Due to Montrose, a continuous trigger applies
 triggers any policy between the date of project completion or the date of third-
party damage and the date of remediation
Maturity
12
24
36
48
60
72
84
1998
2
Accident Date
1999
2
2000
2
2001
2
2002
2
2003
2
2004
2005
96
2
2
 Most companies code claim counts only to oldest
© 2008 Towers Perrin
30
Cost Allocation Matrix – illustrative (cont.)
 Greater use of additional insured endorsements spreads losses to more
companies
Maturity
12
24
36
48
60
72
84
1998
4
Accident Date
1999
4
2000
4
2001
4
2002
4
2003
4
2004
2005
96
4
4
 Losses move from general contractors to artisans
© 2008 Towers Perrin
31
Cost Allocation Matrix – illustrative (cont.)
 Montrose exclusion or “known loss” stops policies from responding once
an incidence is known
Maturity
12
24
36
48
60
72
84
1998
4
Accident Date
1999
2000
96
4
4
2001
2002
2003
2004
2005
 Fewer exposed policies mean higher severity
© 2008 Towers Perrin
32
Cost Allocation Matrix – illustrative (cont.)
 “Prior acts” or prior work” exclusion restricts policies to the year of
construction
 “prior” exclusions usually refer to work done while insured by insurer
 Wraps can have similar affects
Maturity
12
1998
24
36
48
60
72
84
96
4
Accident Date
1999
2000
2001
2002
2003
2004
2005
© 2008 Towers Perrin
33
Actuarial Analysis – Exposures/Underwriting
 Homebuilders, general contractors, artisans, suppliers
 Practice, project, wraps
 Residential vs. commercial
 Residential : single family vs. multi-family
 Primary or excess
 States
 Policy forms, coverages, exclusions, additional insured endorsements
 Premises & operations, completed operations (construction defect),
warranty
 Occurrence, claims-made, sunset provisions
 Homes, sales/revenue, budget, payroll
© 2008 Towers Perrin
34
Actuarial Analysis - Issues to Address (cont'd.)
 Coding/availability of loss data
 California and other states
 Residential vs. commercial
 Single family vs. multi-family
 Missing case reserves
 Insurers – ALAE
 Homebuilders – loss or ALAE
 Claim counts
 Reported, closed
 Developer/contractor vs. subs/artisans  CWP, CWNP
 Definition & mix of CD/non CD claims
 CWIP, CWEP
 Report year triangles
 Limits/reinsurance/wrap
 Accident/COE triangles
 ALAE - inside or outside of limits
 Individual claim listing
 Sub-classes (EIFS, mold, AI)
© 2008 Towers Perrin
35
The mix of construction defect to
non-construction defect claims is important
 The mix of construction defect losses and non construction defect losses
is important to understanding the potential reporting pattern.
Cumulative Loss Development Patterns
1.000
% Reported
0.800
0.600
0.400
0.200
0.000
1
2
3
4
5
6
7
8
9 10 11 12 13 14 15 16 17
Years
0% CD/ 100% Non-CD
25% CD/ 75% Non-CD
50% CD/ 50% Non-CD
75% CD/ 25% Non-CD
Based on California reporting patterns
© 2008 Towers Perrin
36
Traditional actuarial methods may not be good predictor
 Loss development methodologies assume that past is a good predictor of
the future
 Construction losses have been affected by a constantly changing
environment
 Better to use frequency and severity methodologies
 Management can understand results
 Monitoring of results is easier
© 2008 Towers Perrin
37
RECOMMENDATIONS
Recommended Construction Defect Reserving Methodology
 Combination of
 Report Year Loss Development
 Frequency/Severity Method for “Pure” IBNR
Ultimate
Loss &
ALAE
Where
© 2008 Towers Perrin
=
(
Reported
Loss &
ALAE
IBNR =
Reserve
)(
+
IBNR
Claim
Counts
RY
Supplemental
Reserve
)(
+
( ) ( )(
x
CWP
Ratio
x
)
IBNR
Reserve
CWP
Severity
)
38
Reported CD Claim Count Development - illustrative
Reported Claim Counts
Accident
Year
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
12
0
0
0
114
162
334
340
465
379
350
182
179
72
42
47
77
© 2008 Towers Perrin
24
0
0
83
396
565
753
830
1,280
898
851
582
475
250
201
243
36
0
0
261
701
914
1,166
1,879
1,734
1,414
1,357
1,088
727
458
436
48
0
74
495
980
1,229
1,858
2,355
2,266
2,057
1,974
1,690
1,047
769
60
122
193
777
1,214
1,841
2,303
2,986
3,012
2,741
2,513
2,253
1,469
72
189
354
1,034
1,682
2,285
2,857
3,787
3,810
3,401
3,132
2,855
Evaluation age in Months
84
96
108
120
276
364
518
718
505
758 1,101 1,438
1,490 1,920 2,452 3,087
2,169 2,602 3,154 3,746
2,739 3,339 3,978 4,515
3,570 4,308 4,979 5,566
4,659 5,482 6,337 7,146
4,528 5,297 6,207
4,185 5,200
3,766
132
919
1,808
3,783
4,202
4,898
6,212
144
1,137
2,114
4,118
4,484
5,280
156
1,230
2,300
4,318
4,727
168
1,286
2,384
4,444
180
1,289
2,466
192
1,301
39
Reported CD Claim Count Development Factors - illustrative
Reported Claim Count Loss Development Factors
Accident
Year
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
1-2
3.49
3.48
2.25
2.44
2.75
2.37
2.43
3.19
2.65
3.48
4.74
5.15
© 2008 Towers Perrin
2-3
3.13
1.77
1.62
1.55
2.26
1.36
1.57
1.59
1.87
1.53
1.83
2.17
3-4
1.90
1.40
1.35
1.59
1.25
1.31
1.46
1.45
1.55
1.44
1.68
4-5
2.62
1.57
1.24
1.50
1.24
1.27
1.33
1.33
1.27
1.33
1.40
5-6
1.55
1.83
1.33
1.39
1.24
1.24
1.27
1.26
1.24
1.25
1.27
6-7
1.46
1.43
1.44
1.29
1.20
1.25
1.23
1.19
1.23
1.20
Evaluation age in Months
7-8
8-9
9-10 10-11 11-12 12-13 13-14 14-15 15-16
1.32
1.42
1.39
1.28
1.24
1.08
1.05
1.00
1.01
1.50
1.45
1.31
1.26
1.17
1.09
1.04
1.03
1.29
1.28
1.26
1.23
1.09
1.05
1.03
1.20
1.21
1.19
1.12
1.07
1.05
1.22
1.19
1.13
1.08
1.08
1.21
1.16
1.12
1.12
1.18
1.16
1.13
1.17
1.17
1.24
40
Reported CD Claim Count Development Factors - illustrative
Reported Claim Count Loss Development Factors
Accident
Year
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
1-2
3.49
3.48
2.25
2.44
2.75
2.37
2.43
3.19
2.65
3.48
4.74
5.15
4.46
© 2008 Towers Perrin
2-3
3.13
1.77
1.62
1.55
2.26
1.36
1.57
1.59
1.87
1.53
1.83
2.17
1.84
1.84
3-4
1.90
1.40
1.35
1.59
1.25
1.31
1.46
1.45
1.55
1.44
1.68
1.56
1.56
1.56
4-5
2.62
1.57
1.24
1.50
1.24
1.27
1.33
1.33
1.27
1.33
1.40
1.34
1.34
1.34
1.34
5-6
1.55
1.83
1.33
1.39
1.24
1.24
1.27
1.26
1.24
1.25
1.27
1.25
1.25
1.25
1.25
1.25
6-7
1.46
1.43
1.44
1.29
1.20
1.25
1.23
1.19
1.23
1.20
1.21
1.21
1.21
1.21
1.21
1.21
Evaluation age in Months
7-8
8-9
9-10 10-11 11-12 12-13 13-14 14-15 15-16
1.32
1.42
1.39
1.28
1.24
1.08
1.05
1.00
1.01
1.50
1.45
1.31
1.26
1.17
1.09
1.04
1.03
1.01
1.29
1.28
1.26
1.23
1.09
1.05
1.03
1.02
1.01
1.20
1.21
1.19
1.12
1.07
1.05
1.04
1.02
1.01
1.22
1.19
1.13
1.08
1.08
1.06
1.04
1.02
1.01
1.21
1.16
1.12
1.12
1.08
1.06
1.04
1.02
1.01
1.18
1.16
1.13
1.11
1.08
1.06
1.04
1.02
1.01
1.17
1.17
1.13
1.11
1.08
1.06
1.04
1.02
1.01
1.24
1.16
1.13
1.11
1.08
1.06
1.04
1.02
1.01
1.20
1.16
1.13
1.11
1.08
1.06
1.04
1.02
1.01
1.20
1.16
1.13
1.11
1.08
1.06
1.04
1.02
1.01
1.20
1.16
1.13
1.11
1.08
1.06
1.04
1.02
1.01
1.20
1.16
1.13
1.11
1.08
1.06
1.04
1.02
1.01
1.20
1.16
1.13
1.11
1.08
1.06
1.04
1.02
1.01
1.20
1.16
1.13
1.11
1.08
1.06
1.04
1.02
1.01
1.20
1.16
1.13
1.11
1.08
1.06
1.04
1.02
1.01
41
Tempered Reported CD Claim Count Development Factors –
illustrative
Reported Claim Count Loss Development Factors
Accident
Year
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
1-2
3.49
3.48
2.25
2.44
2.75
2.37
2.43
3.19
2.65
3.48
4.74
5.15
5.75
© 2008 Towers Perrin
2-3
3.13
1.77
1.62
1.55
2.26
1.36
1.57
1.59
1.87
1.53
1.83
2.17
2.28
2.39
3-4
1.90
1.40
1.35
1.59
1.25
1.31
1.46
1.45
1.55
1.44
1.68
1.72
1.77
1.81
4-5
2.62
1.57
1.24
1.50
1.24
1.27
1.33
1.33
1.27
1.33
1.40
1.42
1.43
1.45
1.46
5-6
1.55
1.83
1.33
1.39
1.24
1.24
1.27
1.26
1.24
1.25
1.27
1.27
1.26
1.26
1.26
1.26
6-7
1.46
1.43
1.44
1.29
1.20
1.25
1.23
1.19
1.23
1.20
1.19
1.18
1.17
1.17
1.16
1.15
Evaluation age in Months
7-8
8-9
9-10 10-11 11-12 12-13 13-14 14-15 15-16
1.32
1.42
1.39
1.28
1.24
1.08
1.05
1.00
1.01
1.50
1.45
1.31
1.26
1.17
1.09
1.04
1.03
1.01
1.29
1.28
1.26
1.23
1.09
1.05
1.03
1.03
1.01
1.20
1.21
1.19
1.12
1.07
1.05
1.02
1.02
1.01
1.22
1.19
1.13
1.08
1.08
1.04
1.01
1.01
1.01
1.21
1.16
1.12
1.12
1.07
1.03
1.01
1.01
1.01
1.18
1.16
1.13
1.11
1.07
1.03
1.01
1.01
1.01
1.17
1.17
1.12
1.11
1.06
1.03
1.01
1.01
1.01
1.24
1.17
1.12
1.11
1.06
1.03
1.01
1.01
1.00
1.24
1.16
1.12
1.10
1.05
1.03
1.01
1.01
1.00
1.24
1.15
1.11
1.10
1.05
1.03
1.01
1.01
1.00
1.25
1.15
1.11
1.10
1.04
1.03
1.01
1.01
1.00
1.25
1.14
1.11
1.10
1.04
1.03
1.01
1.01
1.00
1.25
1.14
1.10
1.09
1.03
1.03
1.01
1.01
1.00
1.25
1.13
1.10
1.09
1.02
1.03
1.01
1.01
1.00
1.25
1.13
1.10
1.09
1.02
1.03
1.01
1.01
1.00
42
Homebuilder Example – Reported Claim Counts
(1) Claim Count as of December 31, 2005
COE Year
1997
1998
1999
2000
2001
2002
2003
2004
2005
© 2008 Towers Perrin
1
-
2
1
1
1
1
1
1
1
1
1
2
3
-
1
1
2
2
1
2
3
Report Year Lag
4
5
6
1
1
1
1
2
2
3
2
2
1
2
7
2
1
3
2
8
2
2
1
9
3
2
1
Total
10
11
11
9
5
5
5
2
2
43
Homebuilder Example – Reported Claim Count Frequency
(1) Claim Count as of December 31, 2005
COE Year
1997
1998
1999
2000
2001
2002
2003
2004
2005
1
-
2
1
1
1
1
1
1
1
1
1
2
3
-
1
1
2
2
1
2
3
Report Year Lag
4
5
6
1
1
1
1
2
2
3
2
2
1
2
7
2
1
3
2
8
9
2
2
1
3
2
1
7
1.43
1.18
0.43
8
2.14
1.18
9
0.71
Total
10
11
11
9
5
5
5
2
2
(2) Frequency Per 1,000 Deliveries
COE Year
1997
1998
1999
2000
2001
2002
2003
2004
2005
© 2008 Towers Perrin
1
0.59
0.42
0.40
0.33
0.31
0.50
2
0.59
0.43
0.42
0.38
0.33
0.28
3
1.18
0.87
0.38
0.80
1.00
Report Year Lag
4
5
6
0.71
0.71
1.43
0.59
0.59
0.59
0.87
0.87
1.30
1.25
0.83
0.83
0.77
0.38
0.80
Total
7.14
6.47
4.78
3.75
1.92
2.00
1.67
0.59
0.50
44
Homebuilder Example – Estimate Ultimate Frequency
(2) Frequency Per 1,000 Deliveries
COE Year
1997
1998
1999
2000
2001
2002
2003
2004
2005
Average
© 2008 Towers Perrin
1
0.59
0.42
0.40
0.33
0.31
0.50
2
0.59
0.43
0.42
0.38
0.33
0.28
3
1.18
0.87
0.38
0.80
1.00
0.28
0.30
0.60
Report Year Lag
4
5
6
0.71
0.71
1.43
0.59
0.59
0.59
0.87
0.87
1.30
1.25
0.83
0.83
0.77
0.38
0.80
0.83
0.68
1.04
7
1.43
1.18
0.43
8
2.14
1.18
9
0.71
Total
7.14
6.47
4.78
3.75
1.92
2.00
1.67
0.59
0.50
1.01
1.66
0.71
7.13
45
Homebuilder Example – Claim Count Reporting Pattern
(2) Frequency Per 1,000 Deliveries
COE Year
1997
1998
1999
2000
2001
2002
2003
2004
2005
Report Year Lag
4
5
6
0.71
0.71
1.43
0.59
0.59
0.59
0.87
0.87
1.30
1.25
0.83
0.83
0.77
0.38
0.80
1
0.59
0.42
0.40
0.33
0.31
0.50
2
0.59
0.43
0.42
0.38
0.33
0.28
3
1.18
0.87
0.38
0.80
1.00
Average
0.28
0.30
0.60
0.83
0.68
Selected Pattern
0.05
0.05
0.08
0.11
0.11
© 2008 Towers Perrin
7
1.43
1.18
0.43
8
2.14
1.18
9
0.71
Total
7.14
6.47
4.78
3.75
1.92
2.00
1.67
0.59
0.50
1.04
1.01
1.66
0.71
7.13
0.14
0.14
0.22
0.10
1.00
46
Homebuilder Example – Projecting Ultimate Claim Counts
(1) Claim Count as of December 31, 2005
COE Year
1997
1998
1999
2000
2001
2002
2003
2004
2005
1
-
2
1
1
1
1
1
1
1
1
1
2
3
-
1
1
2
2
1
2
3
Report Year Lag
4
5
6
1
1
1
1
2
2
3
2
2
1
2
7
2
1
3
2
8
9
2
2
1
3
2
1
7
1.43
1.18
0.43
8
2.14
1.18
9
0.71
Total Ultimate
10
10
11
12
11
16
9
17
5
16
5
18
5
23
2
25
2
29
(2) Frequency Per 1,000 Deliveries
COE Year
1997
1998
1999
2000
2001
2002
2003
2004
2005
Report Year Lag
4
5
6
0.71
0.71
1.43
0.59
0.59
0.59
0.87
0.87
1.30
1.25
0.83
0.83
0.77
0.38
0.80
1
0.59
0.42
0.40
0.33
0.31
0.50
2
0.59
0.43
0.42
0.38
0.33
0.28
3
1.18
0.87
0.38
0.80
1.00
Average
0.28
0.30
0.60
0.83
0.68
1.04
1.01
1.66
0.71
7.13
Selected Pattern
0.05
0.05
0.08
0.11
0.11
0.14
0.14
0.22
0.10
1.00
© 2008 Towers Perrin
Total Ultimate
7.14
7.14
6.47
7.18
4.78
7.06
3.75
7.03
1.92
6.20
2.00
7.06
1.67
7.51
0.59
7.01
0.50
7.27
47
Loss Projection Calculations
 Insurance Companies
 loss development on reported





claims
claim reporting pattern
future # of claims
closed w/payment statistics
severity per claim
pure premium per
payroll/revenue
 National Home Builders
 loss development on reported







© 2008 Towers Perrin
claims
claim reporting pattern
future # of claims
claim frequency per 1000
deliveries
# of homes per claim
# of homes involved in a claim
per 1000 deliveries
severity per claim
pure premium per home
48
Homebuilder statistics
 NERA/NAHB study “Construction Defect Disputes: Getting to Yes
without Going to Court”
 Statistics from 261 home builders surveyed as part of NAHB
monthly and quarterly Housing Market Index – 2004 survey
— Estimated per unit cost of homebuilder general liability and
umbrella/excess insurance costs - $2,700
– Some homebuilders as high as $15,000
— AZ, CA, and NV costs are 87% higher than states with low
liability costs (GA, IL, MN, MO, PA, SC)
— Average cost of insurance per $1,000 of revenue for single-family
builders
– in litigation intensive states - $9.32
– in states with low liability costs - $3.76
– Average difference in costs is equivalent to $3,000 per unit
© 2008 Towers Perrin
49
Ronald T. Kozlowski, FCAS, MAAA
Towers Perrin
525 Market St., Suite 2900
San Francisco, CA 94105-2708
Phone: (415) 836-1025
Email: ron.kozlowski@towersperrin.com