1. Individual Insureds differ in potential risk and amount of insurance coverage.
2. We can ensure each group pays its share of losses and avoid anti-selection while ensuring fair discrimination.
ANTI-SELECTION
1. A situation where sellers have information that buyers do not, or vice versa, about some aspect of product quality. In the case of insurance, adverse selection is the tendency of those in dangerous jobs or high-risk lifestyles to get life insurance.
2. To fight adverse selection, insurance companies try to reduce exposure to large claims by limiting coverage or raising premiums.
OTHER CONSIDERATIONS IN RATING DISTINCTIONS
Operational
|
Social
|
Legal
|
Actuarial
|
·
Clear objective definition of which demographic is in group
·
Administrative expense
|
·
Privacy
·
Causality
·
Affordability
|
·
Constitutional
·
Statutory
·
Regulatory
|
·
Homogeneity
·
Reliability
·
Credibility
|
BASIC METHOD FOR RATE RELATIVITY
1. Loss ratio relativity method - Produces changes in relativity.
Class
|
Premium
$
|
Losses
$
|
Loss
Ratio
|
Loss
Ratio Relativity
|
Current
Relativity
|
New
Relativity
|
1
|
1000000
|
600000
|
0.60
|
1.00
|
1
|
1
|
2
|
3000000
|
700000
|
0.23
|
0.39
|
1.17
|
0.45
|
2. Pure premium relativity method - Produces indicated relativity.
Class
|
Exposures
|
Losses $
|
Pure Premium
|
Pure Premium Relativity
|
1
|
6,000
|
900,000
|
150
|
1
|
2
|
7,000
|
1,300,000
|
186
|
1.24
|
CREDIBILITY
1. Weightage assigned to a given body of data
2. Designated by Z. Below are common properties.
3. 0 < Z < 1 (data is given full weight and fully credible at Z = 1)
4. Diff Z/ Diff E > 0 (credibility increases as experience increases)
5. Diff (Z/E) / Diff E <0 (Percentage change in credibility decrease as volume of experience increases)
ESTIMATING CREDIBILITY
1. Bayesian
Z = E/ (E+K)
Note: E = Exposure, K = expected variance within classes or variance between classes
2. Classical or Limited Fluctuation
Z = (n/k)^0.5
Note: n = observed number of claims, k = full credibility standard
EXAMPLE OF APPLYING CREDIBILITY
1. Calculating new relativity from loss ratio
Class
|
Loss Ratio
|
Credibility (Z)
|
Credibility Weighted Loss Ratio
(LR)
|
Loss Ratio Relativity
|
Current Relativity
|
New Relativity
|
1
|
0.7
|
0.6
|
0.66
|
1
|
1
|
1
|
2
|
0.4
|
0.8
|
0.44
|
0.67
|
2
|
1.33
|
Total LR
|
0.6
|
Note: Credibility weighted Loss Ratio is calculated as follow:
LR= (Z)LRclass i + (1-Z) LRstate
2. Off-Balance Adjustment
Class
|
Premium
|
Current Relativity
|
Premium @ Base Class Rates
|
Proposed Relativity
|
Proposed Premium
|
1
|
2,000,000
|
1
|
2,000,000
|
1
|
2,000,000
|
2
|
3,000,000
|
2
|
1,500,000
|
1.33
|
1,995,000
|
Total
|
5,000,000
|
3,995,000
|
