Insurance pricing anomoly
Insurance pricing anomoly
Author
Discussion

boyse7en

Original Poster:

8,249 posts

196 months

Thursday 30th March 2017
quotequote all
Maybe this is common knowledge, but it was a new one on me.

Got a quote from Bennetts for my BMW F800.

Fully comp £85 with an excess of £200 + £250 voluntary excess.

I reduced the voluntary excess to zero, and my price reduced to £78! Don't get how that works, but I'm not complaining.

catso

16,587 posts

298 months

Thursday 30th March 2017
quotequote all
We'll never understand how insurance companies work.

One of my bikes has no provision to carry a pillion; no seat, no footrests and no means of fitting them but my cover includes pillion.

Whilst speaking with the insurance co. at renewal time I told them this and asked if I would save anything by not having pillion cover - they came back wanting around £80 more for solo only...? confused

sc0tt

18,315 posts

232 months

Thursday 30th March 2017
quotequote all
catso said:
We'll never understand how insurance companies work.

One of my bikes has no provision to carry a pillion; no seat, no footrests and no means of fitting them but my cover includes pillion.

Whilst speaking with the insurance co. at renewal time I told them this and asked if I would save anything by not having pillion cover - they came back wanting around £80 more for solo only...? confused
Fat chicks = Slower Bike

??

Birky_41

4,668 posts

215 months

Thursday 30th March 2017
quotequote all
Black magic...

Mr2Mike

20,143 posts

286 months

Thursday 30th March 2017
quotequote all
That will likely be because their ultra secret, carefully compiled stats* show that anyone that can afford to pay a higher excess is at greater risk of crashing and wiping out a bus queue of pensioners. You know, the same stats that show if someone randomly crashes into you then you are suddenly at greater risk of another accident.



* Random number bullst generator

LuS1fer

43,570 posts

276 months

Thursday 30th March 2017
quotequote all
Yes, I found that increasing the excess had no effect on the premium so I reduced it and it was either the same or lower.
However, some insist on a compulsory excess you can't get out of.

Rubin215

4,201 posts

187 months

Thursday 30th March 2017
quotequote all
I always go for zero voluntary excess anyway.
fked if I'm paying any more if I ever do have a claim!

Gnits

1,156 posts

232 months

Thursday 30th March 2017
quotequote all
Funny how nobody gets to see the 'statistics'!
Whole system needs a proper overhaul before it goes the way of the banks.

catso

16,587 posts

298 months

Thursday 30th March 2017
quotequote all
Getting back to the pillion business, why is it that insurance companies have developed this charging more (or not) for pillion cover?

I understand the extra risks etc. but when I insure my car they don't ask me if I take passengers.

The same with mileages, they never used to ask.

There was a time when insurance was simple, now it just seems to get more complicated, more conditional and of course more expensive. confused

black-k1

12,817 posts

260 months

Friday 31st March 2017
quotequote all
Oh how I miss Loons justifications of these "features" in insurance. I'm sure tomorrow morning would be the perfect time for him to give the answers!!!

There is no real logic in insurance, particularly low volume insurance like that for motorcycles. Much of it is based entirely on what percentage of market share each company wants, what are the competition doing/offering and what have been the big loss makers/claims over the last couple of years.

Prof Prolapse

16,163 posts

221 months

Friday 31st March 2017
quotequote all
The problem, as I understand it, is insurance companies actuaries have extremely strict, and I believe international guidelines, they must by law work to. The "pools" they use to generate these quotes, can be very shallow, and not processed in a manner which is governed by reasoned hypothesis. So you get very odd results, but the insurance company cannot deviate from their agreed process.

It's pure speculation on my part, but a solution would seem to be have an independent third party, presumably a University, aggregate the and process all of the claim data from all UK. The data set would be much larger, so much more reliable even within their strange system. Quotes would no longer see huge variation, as they could only vary by as much as the fees, leading to them cutting their own throats for your business. Many jobs would also be lost, but no one would care.

But as I understand it there's another kicker here, by law insurance companies, I believe, must be independent from the state. So there is a huge question mark over whether it would be possible, also, turkey's don't vote for Christmas, who will convince all car insurers in the country to make themselves redundant?

There also the inconvenient truth, that vehicle insurance, especially motorcycle isn't that profitable, so clumsy government overhauls could mean it quickly ceased to be a viable business. In which case we'd all be fked. This is the biggest threat I see to any change.

So there's not a great deal that can be done within the current legislative framework, but they are working to a strict framework, not shafting you, is my point. It also does have a logic, even if the calculations used don't always mesh with reality. They also are highly likely to have advantages we lay people are simply not aware of. It has likely evolved this way for a reason, we just can't see it not being trained actuaries.

Finally, any truly probability based insurance quotation system, would see the end of the advantageous anomalies as well as the questionable ones. Most car savvy enthusiasts, including myself, have exploited these at some point.

Just a thought.



Edited by Prof Prolapse on Friday 31st March 09:14

sc0tt

18,315 posts

232 months

Friday 31st March 2017
quotequote all
We use risk modelling which will incorporate past claims PP.

Prof Prolapse

16,163 posts

221 months

Friday 31st March 2017
quotequote all
sc0tt said:
We use risk modelling which will incorporate past claims PP.
Assuming you mean you share data from other companies regarding past claims to help guide risk assessment, I get that. But my understanding is that, correct me if I'm wrong, this simply feeds into your individual assessments. You don't, like in my pie in the sky suggestion, have a single independent statistical body making all of your actuarial needs as an industry. Larger sample sizes make for more accurate data.

Again my understanding , is you don't, as an industry, share data on the scale I refer to, and never will. It was suggested, (I think correctly by Loon, that such sharing of intellectual property would likely violate competition laws. That sounds perfectly feasible.

I'm just shooting the breeze, like everyone else here, I don't know enough. I had considered learning more about the actuarial process but then I remembered I would rather remove my own eyes out with rusty spoons.








sc0tt

18,315 posts

232 months

Friday 31st March 2017
quotequote all
Prof Prolapse said:
Assuming you mean you share data from other companies regarding past claims to help guide risk assessment, I get that.
Nope, in laymans terms we use a super computer to analyse risks and calculate possible outcomes.

Doing what you say would create a monopoly and as such a fixed price across the board.

Again, just shooting the breeze.

Prof Prolapse

16,163 posts

221 months

Friday 31st March 2017
quotequote all
sc0tt said:
Nope, in laymans terms we use a super computer to analyse risks and calculate possible outcomes.

Doing what you say would create a monopoly and as such a fixed price across the board.

Again, just shooting the breeze.
I don't understand how that can be true. You pool risk based on multiple data fields. Just because you have an individuals claims history does not mean you do not have a unique interpretation of risk in related fields, postcode, vehicle type, these rely on the datasets you hold as a company. They're your company's property, not shared.

I'm open to there being good reasons why my idea is st, but I don't see how it is a monopoly. By definition to be one, "an individual would need to be in the sole possession of a commodity", there is no possession, and there is no commodity. This is just using a shared tool to make the product, in this case insurance. Companies would be free to operate and compete independently, the reasons for doing so would just be limited to how they operate as a business, their overheads, etc. Probability based on reality, would become the domain of the statisticians, not privately owned individuals.

It's also a bit of specious reasoning to think that, if risk is based on reality, that some how one companies reality can be different from someone else's. The risk, and therefore core pricing, must be identical if it is accurate.








Berz

406 posts

223 months

Friday 31st March 2017
quotequote all
sc0tt said:
Prof Prolapse said:
Assuming you mean you share data from other companies regarding past claims to help guide risk assessment, I get that.
Nope, in laymans terms we use a super computer to analyse risks and calculate possible outcomes.

Doing what you say would create a monopoly and as such a fixed price across the board.

Again, just shooting the breeze.
Sounds like there are, or could be, several layers to this:

1. Raw data/claims history/sample pool/whatever.
2. Super computer to analyse the data and produce a model which says person A on bike B is more or less risky than person X on bike Y.
3. Algorithms/actuaries weight the risk per the preferred insurance company profile (deter high risk yoofs, specialise in insurance for women, etc.).

I don't see how having a larger pool of data to pull from creates a monopoly. It should allow you to be more accurate with your insurance models and allow you to weed out high risk customers you might not want, or encourage high risk customers but with an appropriate premium. Different companies will (guess) have different appetites, some might want to specialise in imports, for example, and that difference is what stops it being a monopoly. Although, would you lose competition in that vertical maybe? Say every insurance company knows exactly how risky boy racers are would they have their little race to the bottom and then one company for that profile survives?

This is all thinking out loud of course, I don't knows the ins and outs just find it interesting smile

sc0tt

18,315 posts

232 months

Friday 31st March 2017
quotequote all
Prof Prolapse said:
I don't understand how that can be true. You pool risk based on multiple data fields. Just because you have an individuals claims history does not mean you do not have a unique interpretation of risk in related fields, postcode, vehicle type, these rely on the datasets you hold as a company. They're your company's property, not shared.

I'm open to there being good reasons why my idea is st, but I don't see how it is a monopoly. By definition to be one, "an individual would need to be in the sole possession of a commodity", there is no possession, and there is no commodity. This is just using a shared tool to make the product, in this case insurance. Companies would be free to operate and compete independently, the reasons for doing so would just be limited to how they operate as a business, their overheads, etc. Probability based on reality, would become the domain of the statisticians, not privately owned individuals.

It's also a bit of specious reasoning to think that, if risk is based on reality, that some how one companies reality can be different from someone else's. The risk, and therefore core pricing, must be identical if it is accurate.
For once, I agree.


Gavia

7,627 posts

122 months

Friday 31st March 2017
quotequote all
I've just renewed and happy with my figure of £256 for the year. I'd prefer there to be a massive spread of prices, otherwise they'll all just move to the middle price and that will cost me more.

djt100

1,744 posts

216 months

Friday 31st March 2017
quotequote all
I Been getting some quotes this morning on go compare, wanted to see what bikes were going to sting me with insurance.
some odd results
2004 Fireblade £348
2006 ZX10R £244 (ZX6R £279)

in fact getting quotes on daytona 675, GSXR6/750/1000 £50 between then but all over £300

Prices are all over the price and seem to have nothing to do with cc of the bike.

MrGman

1,674 posts

237 months

Friday 31st March 2017
quotequote all
As i always say, they use a random number generator!

Much like how fully comp is usually cheaper than TPFT but i imagine this is as it shows you value your bike/car and are therefore more likely to take less chances/risks?!