PDA

View Full Version : WC loss constanst, violation of SOP?


Gareth Keenan
04-05-2011, 11:10 PM
Actuarial Statement of principle on rate making # 3 indicates that a rate must cover for all of the costs associated with an individual risk transfer. item # 4 brings up that a rate must be reasonable, and not excessive, inadequate, nor unfairly discriminatory blah blah blah.

Chapter 11 of Werner introduces the idea of a loss constant, so you can charge to the same loss ratio for your large and small risks. Does this violate the actuarial statement of principles?

Cheers,

Gareth Keenan

MightySchoop
04-06-2011, 09:14 AM
No, it does not. If you're spreading your risk fairly, all policies should have the same expected loss ratio (well, if you don't count some weird stuff that happens with expenses in the Work comp line).

Werner mentions several reasons why small risks in Work Comp have worse loss experience than large risks. The experience rating used in Work Comp does not correct this fully, for reasons you'll understand when you read the exam 8 material. Without making some adjustment for this discrepancy, your large risks will be subsidizing your small risks, which would be a violation of the statement of principles. The loss constant is an attempt to eliminate that subsidy.

JasonScandopolous
04-06-2011, 10:55 AM
No, it does not. If you're spreading your risk fairly, all policies should have the same expected loss ratio (well, if you don't count some weird stuff that happens with expenses in the Work comp line).


The last line of the segment in Werner is almost obviously true; if company size is a significant variable for differentiating loss ratios (whether part of a "sophisticated multivariate technique" or not), why wouldn't you price using it? And why wouldn't you use a fixed expense constant for fixed expenses? etc.

I do have a minor comment on MightySchoop's above statement: "all policies should have the same expected loss ratio". This can be false for other reasons, in a company with ERM and/or sophisticated treatment of investment income. This is because of the increased cost of the capital that needs to be held on riskier exposures, given equal expected losses. Or more simply, "the risk load". Not sure if this is on 5A syllabus (I havent actually finished reading it yet), but, could be good to make note of.

Vorian Atreides
04-06-2011, 11:00 AM
The last line of the segment in Werner is almost obviously true; if company size is a significant variable for differentiating loss ratios (whether part of a "sophisticated multivariate technique" or not), why wouldn't you price using it? And why wouldn't you use a fixed expense constant for fixed expenses? etc.

I do have a minor comment on MightySchoop's above statement: "all policies should have the same expected loss ratio". This can be false for other reasons, in a company with ERM and/or sophisticated treatment of investment income. This is because of the increased cost of the capital that needs to be held on riskier exposures, given equal expected losses. Or more simply, "the risk load". Not sure if this is on 5A syllabus (I havent actually finished reading it yet), but, could be good to make note of.
The bolded, while true, is outside the scope of Exam 5. The base line is that all risks should have the same loss ratio. Deviations from that are exceptions that require documentation (why are those other risks "riskier"?) and, as such, are not part of the Exam 5 syllabus.