Insurance Rate Monitoring - Part 2

YEARLY VARIATIONS

1. If we consider a base underwriting year where nothing major happened, then a policy renewing in January the following year on ‘as before’ terms can be treated in the same way as if it renewed ‘as before’ in August (base year). 

2. However, if the base underwriting year had a large systemic event in July, then a policy renewing ‘as before’ in January would be on considerably weaker terms than if it renewed ‘as before’ in August. This is allowed by tracking the IELR at a policy level.

3. Below is an example to explain the importance of tracking the IELR between a policy and class level.

EXAMPLE

2007	EPI	Estimated Claims	IELR
Policy 1	1mil	0.3mil	30%
Policy 2	1mil	0.6 mil	60%
TOTAL	2mil	0.9mil	45%

1. Total 2007 IELR is 45% ($0.3m of claims plus $0.6m of claims against $2m of premium). 

2. In 2008, only the second risk renews with a 5% rate increase. 

2008	EPI	Estimated Claims	IELR
Policy 1	Not renewed
Policy 2	1.05 mil	0.6 mil	57.1%
TOTAL	1.05 mil	0.6 mil	57.1%

1. The 2008 IELR becomes 57.1% (being 0.6/1.05).

2. If we observe the rate movements of renewals then the class was up 5% (1mil to 1.05mil)

3. If we observe the rate movements of the class at business level, rate movement would  decrease 21% (45%/57.1 -1)

4. As policies does not have the same IELR, the premium-weighted average does not work as mentioned previous post..


CONCLUSION
1.We can see from the example that simply looking at rate changes on renewal business would imply different results as compared to when we take into account the relative size and profitability of the renewal business to the lapsed business.

2. The differences are miniscule but as year-on-year data builds up these small differences can become magnified into much more material differences.

3. Much time and effort can be spent on developing rate monitoring tools but we need to ensure that the system is measuring what we actually need.

4. We need to build up policy level information on Estimated Premium Income (EPI), IELRs and rate movements to obtain accurate information.