View Full Version : What is the point of "Off Balance Factor"

04-27-2005, 10:34 AM
Why do we even need this ? It is created by the way we tabulate all these numbers. To calculate the Adjusted On-level EP, can not I just use the following ? (following All 10 Page 8)

Sate wide indication 1.1014

Class On-Level EP Loss
1 14370 11003
2 9438 6541
3 8002 5618
Total 31810 23162

So State wide expected premium is 31810 * 1.1014 = 35035.53
then distribute this premium proportionally to the loss amount of each class

Class 1 Adjusted on-level EP = 11003/23162 * 35035.53 = 16643.47
Class 2 Adjusted on-level EP = 6541/23162 * 35035.53 = 9894
Class 3 Adjusted on-level EP = 5618/23162 * 35035.53 = 8497.95

The numbers match exactly.
If I use this way on the exam, will I get points ?

04-27-2005, 12:04 PM
Off-balance factors are discussed in detail in Exam 9. For now, just deal with it.

04-27-2005, 12:10 PM
The off balance is used so that your overall indication doesn't change due to different relativites underlying your new indicated relativies vs the ones currently underlying your current rates.

The way you are doing it looks arithmetically identical to me. Your method does it in one less step. The question, however, might want you to give indicated non balanced rates first and then also balance it back so doing this shortcut might not help.

Banquet of Chestnuts
04-27-2005, 05:34 PM
The ratemaking part of this whole exam seems overly complicated. They seem to have decided that ratemaking must be done in tabular form with over-aggregated data and have molded the theories around it.

Methinks "For now, just deal with it." is sound advice.

04-27-2005, 06:27 PM
To the extent that you have varying exposure levels by class, you'll need to consider that in your off-balance factor. I suggest doing it the way that it is shown in Finger. Having said that, the off-balance factor is completely meaningless if you are balancing back to a base class since it divides out.

04-28-2005, 07:54 AM
The purpose of the off-balance factor, much like the CPCU readings, is to piss me off.

04-28-2005, 11:15 AM
Having said that, the off-balance factor is completely meaningless if you are balancing back to a base class since it divides out.No, it isn't. You can have an off-balance factor with either type of relativity--relative to a base class or to a group average.

Two recent exam questions have asked the reason for the off-balance factor, and they didn't give the same answer. According to 2004 Question 14, the adjustment is made "to retain the same overall rate level". This strikes me as wrong because most of the time you have a nonzero overall indication and a relativity change, and you're trying to hit the indication, not retain the same rate level. But the other answers were even more wrong.

According to 2003 Question 37(b), The off-balance factor is necessary because the average relativity under indicated rates is different from the average relativity underlying the current rates.I think that's a better answer.

04-28-2005, 06:00 PM
yes, it is. I agree with everything else you said though. Mathematically, though, when you are doing the Finger method, the off-balance factor cancels out since you are taking the credibility weighted indicated adjustment/off balance factor to find the adjustment necessary to balance back to an overall revenue neutral change. Take that value for class C and divide by that value for the base class, and the off balance factor cancels out. Now, if you are balancing to statewide, that's not true, and I think that's how most actual rating plans work, so that might be what you're thinking of.

None of that really matters. What matters is memorizing off-balance factor calculation, and why it's important, and regurgitating that information in a little over a week.

04-29-2005, 12:21 PM
I think it's easier to understand if you forget about the average relativity and simply consider the total premium collected under each set of relativities. For example, under the old set of relativities, your total premium might be $1,000,000. Using your new relativities, your total premium might come out to be only $970,000. So you need to make up that $30,000 somewhere (unless you're filing for a 3% rate decrease), and that's what the off-balance factor does--it increases the $970,000 back up to $1,000,000. In this example, the off-balance factor (if you're multiplying) = 1,000,000 / 970,000 = 1.031. You can also look at it as getting only 970,000 / 1,000,000 = 97% of the original premium, so you can divide by .97 and get the same figure. I would think that the CAS would accept either method since they are mathematically equivalent.