Introduction to Property Exposure Rating

1
Introduction to Property Exposure Rating

Thomas Cosenza, FCAS, MAAA August 8, 2007
2
Data needed for exposure rating

• To select the severity curve, the hazard
  characteristics for subject business is needed.
• Prospective non-cat gross loss and ALAE ratio for
  subject business
• Prospective Subject Premium
• Current limits profile with SIR/attachment point
  by premium

3
Understand the limits Profile

• Business interruption and/or contents included?
• Policy limit or location limit.
• Key locations only or all locations
• Are locations properly valued (ITV)
• Are there added or excluded coverages
• Are there excess policies/large deductibles, if
  so need attachment point
• Subscription policies
• Gross or net of facultative purchases
• Homeowners Coverage A or all coverages

4
Property Loss Curves

• Lloyds
• Salzmann (1960 INA Homeowners data)
• Reinsurer Curves (Swiss Re, Munich Re, etc)
• Ludwig (1984-1988 Homeowners and Small Commercial
  data)
• ISOs PSOLD (Recent Commercial data)
• ISOs PSOLD (Recent Homeowners data)

5
Property Loss Curves Advantages/(Disadvantages)

• Lloyds Curves
• (Very old data)
• (Does not vary by amount of insurance or
  occupancy class)
• (Underlying data is largely unknown (marine
  losses? WWII Fires?))
• Salzmann (Personal Property)
• Based on actual Homeowners data
• Varies by Construction/Protection Class
• (Very old data from 1960)
• (Does not vary by amount of insurance)
• (Building losses only and Fire losses only)
• (Salzmann recommends not using them, only meant
  as an example)
• Reinsurer Curves (Swiss Re, Munich, Skandia, etc)
• Documented study (some curves) on personal
  commercial reinsurance business
• (Old data)
• (No publicly available documentation)
• (Does not vary by occupancy class)

6
Property Loss Curves Advantages/(Disadvantages)

• Ludwig Curves (Personal and Commercial)
• Based on actual Homeowners and Commercial data,
  (but uses Hartford small commercial property book
  may not be good for large national accts)
• Varies by Construction/Protection Class for HO
  and Occupancy Class for Commercial
• Includes all property coverages and perils
• (Old data 1984 - 1988)
• ISOs PSOLD
• Recent Data updated every 2 years
• Varies by amount of insurance, occupancy class,
  state, coverage, and peril
• Continuous Distribution (no need for
  Interpolation)
• Based on 2,000,000 occurrences
• 4 Perils BG1(Fire), BG2(Wind), Special(All
  Other) and All Perils
• Special Update will be available in September
  2007
• (Based on ISO data only)
• (ISO data limited for large accounts)
• (Mixed exponential curves allow unlimited loss)
• (Huge number of curves- How well were they fit?)
• 3 class groups times 60 size of risk bands
• Additional curves by subclass and state

7
PSOLD Endurance Approach

• Loss-to-value curves based on PSOLD parameters
• Use 5 size of risk groups(0-1m,1m-5m,5m-10m,10m-50
  m, 50m )
• 3 Classes(Light, Medium, Heavy)
• Ignore state differences (data not credible)
• Cap loss at 125 of total limits and 150 of
  building content limits.

8
(No Transcript)
9
PSOLD Endurance Classes
10
Property Loss to Layer Calculation

• Expected Loss to the Layer (Premium)(Expected
  Ground Up Loss Ratio)( Exposed)

• FLS (First Loss Scale) (Limited Average
  Severity)/(Unlimited Expected Severity)
• FLS(X) LAS(X)/E(X)

11
Property Exposure Rating Example Primary Policy
12
Property Exposure Rating Example Excess Policy
13
Property Exposure Rating Example

• As deductibles (attachment point) as a percentage
  of TIV increase
• Subject premium tends to decrease
• The dollar amount of loss to the layer may not
  change but the burn as a of subject premium
  does.
• The closer to pro-rata premium will be needed for
  excess of loss contracts

14
Excess Policies

• Excess on Excess is extremely difficult to write.
• Limits and SIRs/Attachment Points are less
  stable than a primary book.
• Experience may be less credible
• Difficult to calculate underlying rate changes
• Policy may cover multiple locations with single
  limit
• May be difficult to allocate premium by location
• SIRs/Attachment Points are extremely important.
• To properly exposure rate either a 1) policy
  listing or 2) limits profile matrix showing
  limit and SIRs/Attachment Points by premium is
  needed.
• Using an Average SIR/Attachment Point may lead to
  inaccurate calculations.

15
Excess Policies Average SIR vs Actual
16
Subscription Policy - Example

• Actual reinsurance layer is 20 of 5m xs 5m.
• Losses above 10m are not relevant.

17
Insurance to Value (ITV) Commercial Insurance

• Ranges from 20 to 40 Undervalued
• Reasons
• Underwriters are accepting reported values as
  10-20 under as "close-enough" on new business
  and when they do spot checks at renewal
• Values may not be updated for years
• Renewal values are not being kept current.
  Average inflation in construction has been
  running close to 10 over the past few years and
  renewal updates (if any) has not been adequate.
• Underwriters are only spot-checking and some are
  not including any ITV analysis in their workflow
  at all .
• Blanket limit can allow individual locations to
  be underinsured
• Margin Clauses can somewhat mitigate that issue

18
Insurance to Value (ITV) Homeowners

• Per MSB 2006 Study 57 of the homes are
  undervalued by 21
• Reasons
• Replacement cost coverage on homeowners decreases
  incentive for policyholders to insure to value
• Renewals are undervalued
• Inspections may not be done for years
• Policy characteristics change every year
• Remodeling a 233B industry, accounts for 40 of
  all residential construction and improvements.
• Households living in their homes more than 2
  years accounted for 86 of total remodeling
  dollars.
• Mystery Wings, missing 465 sq ft on 7-10 of
  records
• Average cost of Kitchen in top 35 Market? 43,213
• Average cost of a Room Addition? 27,028

19
Insurance to Value Key Questions

• What models and versions (if any) are being used
  for ITV analyses?
• How often is the ITV vendors model/cost guide
  updated?
• How long does it take the carrier to implement
  the updates once received from the vendor?
• How often has the carrier been making updates,
  quarterly or annually?When was the last update?
• Are policy values based on Reconstruction or New
  Construction basis?
• Has the carrier been using vendors recommended
  settings or have custom factors been applied?
  Have any changes been made to these factors
  recently?
• Who is preparing ITV analyses agents,
  engineers, or underwriters?
• How is ITV analysis integrated into the
  underwriters workflow?
• On what size buildings do guidelines recommend
  values be checked?
• What method is used to update values at renewal?

20
Insurance to Value Key Questions

• How far in advance of the renewal date are values
  updated?
• What values are used to run the CAT model?
• Are portfolio values updated before the CAT
  models are run?
• What audit procedures have been used to make sure
  values are kept current on individual risk or on
  the portfolio?
• What projections have been included in CAT
  analyses for inflation and demand surge?
• Are values used for analyses 100 values or 80
  /90 values(co-insurance) or limits exposed?
• What tools and procedures in place for evaluating
  contents values?
• What tools are in place for evaluating business
  interruption values?
• How is data quality tied to incentive
  compensation?
• How close is close enough?

21
Exposure Rating - Homeowners

• Loss curves are based on Coverage A (Buildings)
• Coverage A can range from 40 to 60 of total
  limit on standard policies. For high value homes
  Coverage A can be a low as 20 due to high
  contents coverage.

22
Other Issues Cat Loads

• Cat Models(RMS, AIR, EQECAT) can be used
  determine loss to reinsurance layers.
• Per Risk covers are the most difficult to cat
  model
• RMS has issues capping at the occurrence limit
• Alternative Method Exposure rating using the
  Cat Loss Ratio (Gross CAT Loss/Subject Premium)
  instead of the Non-Cat Loss Ratio.
• Good for a reasonability check
• May overstate cat load due to peril specific sub
  limits.

23
Other Issues Relativity Approach

• Used for higher layers with little or no
  experience
• Lower more credible experience layers are
  compared to exposure rating the relativity is
  applied to exposure rating of higher layer
• Good check for fit of exposure curve

24
Example - Experience vs Exposure Rating
25
Advantages of exposure rating

• Relatively easy to do
• Any excess layer can be priced
• Use of current limits profile enables up to
  date view of excess layer pricing.
• Shifts in limits and attachment points over time,
  which may make experience rating difficult, are
  irrelevant here.
• Addresses free cover issues. Free cover
  exists when the top of your reinsurance layer
  exceeds the largest trended loss in your data.

26
Disadvantages of exposure rating

• Selected severity curve may not properly reflect
  clients subject business
• Selected gross loss and ALAE ratio may not
  appropriately reflect exposed risks
• Exposure rating does not consider clients actual
  loss experience in excess layers