Page 356 - IC38 GENERAL INSURANCE

A. Underwriting basics

In the previous chapters we have seen that the concept of insurance involves
managing risk through pooling. Insurers create a pool consisting of premiums
that are made by several individuals / commercial / industrial firms /
organizations.

The amount of premium to be paid by each depends on a rate, which is
determined by two factors;

 The probability of loss due to a loss event (caused by an insured peril)
and

 The estimated amount of loss that may arise due to the loss event

Example

Assume the average amount of loss as a result of a fire was Rs 100000 [which we
denote as L]

The average or mean probability of the loss [denoted by P] was 1 out of 100 [or
0.01].

The mean or average expected loss would then be given by: L x P = 0.01 x
100000 = 1000

How can the insurer ensure that the pool is sufficient to compensate for the
losses that are actually incurred?

As we have seen earlier, the whole mechanism of insurance involves pooling of
a large numbers of statistically similar risks so that the law of large numbers
would operate and the probability of number of losses (frequency) as well as
the extent of loss (severity) becomes predictable.

The problem is that all exposures are not alike. A pool of exactly similar [or
‗identical‘] risks may be quite small.

For instance, how many houses would you find that are exactly similar and
located in exactly the same external environment? Not many.

As the pool size increases, it is likely to include non similar risks, which are
exposed to same or similar perils. The insurer faces a dilemma here.

How to create a pool which is large enough so that the risk becomes more
predictable while at the same time ensuring that the pool is sufficiently
homogenous and contains similar risks?.

Insurers have found a solution to the problem.

350