Auto Insurance Question

Non-investing personal finance issues including insurance, credit, real estate, taxes, employment and legal issues such as trusts and wills
Post Reply
Xpe
Posts: 129
Joined: Wed Sep 17, 2014 7:24 am

Auto Insurance Question

Post by Xpe » Sat May 20, 2017 12:48 pm

Saw another car insurance thread, reminded me of my lack of understanding, but I didn't want to derail their thread so thought I'd start a new one.

I'd love to hear from someone who understands the auto insurance industry.

I feel like I'm missing something, I have a fairly cheap car, got it for $18k 1 year old with 3k miles, now its 5 years old with 15k miles. That should also tell you something about how few miles I drive. I've had one claim 6 years ago, and it was documentedly the other persons fault (they ran a red light and hit me as I went through the intersection).

So with:
fairly cheap car
extremely low miles per year
decent driving record

So I guess for a general question: How is it that I pay $600 per 6mo? I got quotes from a number of companies and they were all in the same range. Also I moved 15 minutes away and my insurance went up 20%. I talked to the company (progressive) and they said it was because the new area has a higher incident rate.

And then for a more actionable question: If I sold my car and bought a car for like $8k, would my insurance go down proportionally? Or do they figure that an older cheaper car is even more likely to have issues?

Also - if I hadn't updated my address, and I got into an accident, would they be able to reject my claim on due to the incorrect address?

anoop
Posts: 427
Joined: Tue Mar 04, 2014 1:33 am

Re: Auto Insurance Question

Post by anoop » Sat May 20, 2017 1:00 pm

They work off of statistics -- age, the car make/model/model year, marital status, ZIP code -- all come into play.

Many years ago I moved from a HCOL area to a LCOL area and my insurance went up.

A couple of years ago, I exchanged a 12 month old car (lemon) for a new one which was about $5K more expensive and my insurance stayed exactly the same.

Some things to consider
- Certain ZIP codes may have higher instances of car vandalism/theft. (Has nothing to do with how safe they may be otherwise.)
- Your car make/model/year may be in high demand for its used parts (hence higher chance of being stolen).
- Your ZIP might have higher % of accidents.
- You don't mention your marital status and whether or not you have kids.

Insurance costs are going up because cars are getting more expensive to fix. Many bumpers today cannot be repaired, so even a minor fender bender requires a new bumper. There is little difference between replacing the bumper of a 5 year old car and that of a new car of the same model even though the two cars have a huge difference in cost. The only time the difference in cost comes into play is in a major accident (because of the payout based on cost), but those are rare anyway, hence the price of the car tends not to affect insurance rates as much.

anoop
Posts: 427
Joined: Tue Mar 04, 2014 1:33 am

Re: Auto Insurance Question

Post by anoop » Sat May 20, 2017 1:03 pm

BTW, Progressive is known to specialize in high-risk policies. Their rates are typically lower than other traditional insurance companies (Amica, AAA, Farmer's, etc.) if you have one or more active at-fault accidents (last 3 years). If not, you will typically get a lower rate moving to a more traditional insurer, the kind would typically refuse to sell you a policy after two at-fault accidents or charge you an obscene amount.

User avatar
Shackleton
Posts: 224
Joined: Mon Dec 29, 2014 6:20 pm

Re: Auto Insurance Question

Post by Shackleton » Sat May 20, 2017 2:59 pm

anoop wrote:BTW, Progressive is known to specialize in high-risk policies. Their rates are typically lower than other traditional insurance companies (Amica, AAA, Farmer's, etc.) if you have one or more active at-fault accidents (last 3 years). If not, you will typically get a lower rate moving to a more traditional insurer, the kind would typically refuse to sell you a policy after two at-fault accidents or charge you an obscene amount.


That is how Progressive started but is no longer true and hasn't been true for at least a decade.
“Superhuman effort isn't worth a damn unless it achieves results.” ~Ernest Shackleton

anoop
Posts: 427
Joined: Tue Mar 04, 2014 1:33 am

Re: Auto Insurance Question

Post by anoop » Sat May 20, 2017 3:06 pm

Shackleton wrote:
anoop wrote:BTW, Progressive is known to specialize in high-risk policies. Their rates are typically lower than other traditional insurance companies (Amica, AAA, Farmer's, etc.) if you have one or more active at-fault accidents (last 3 years). If not, you will typically get a lower rate moving to a more traditional insurer, the kind would typically refuse to sell you a policy after two at-fault accidents or charge you an obscene amount.


That is how Progressive started but is no longer true and hasn't been true for at least a decade.


Ok. My last experience with them was 12 years ago. I had been with them for 3 years because of accidents and they flat out told me they don't try to compete with others for drivers with a good record. It is interesting they changed their business model.

User avatar
Shackleton
Posts: 224
Joined: Mon Dec 29, 2014 6:20 pm

Re: Auto Insurance Question

Post by Shackleton » Sat May 20, 2017 3:09 pm

anoop wrote:
Shackleton wrote:
anoop wrote:BTW, Progressive is known to specialize in high-risk policies. Their rates are typically lower than other traditional insurance companies (Amica, AAA, Farmer's, etc.) if you have one or more active at-fault accidents (last 3 years). If not, you will typically get a lower rate moving to a more traditional insurer, the kind would typically refuse to sell you a policy after two at-fault accidents or charge you an obscene amount.


That is how Progressive started but is no longer true and hasn't been true for at least a decade.


Ok. My last experience with them was 12 years ago. I had been with them for 3 years because of accidents and they flat out told me they don't try to compete with others for drivers with a good record. It is interesting they changed their business model.

It's why they are now the #4 auto insurer and likely to be #3 by the end of the year.
“Superhuman effort isn't worth a damn unless it achieves results.” ~Ernest Shackleton

NorCalDad
Posts: 696
Joined: Sun Mar 25, 2012 11:14 pm
Location: Northern California

Re: Auto Insurance Question

Post by NorCalDad » Sat May 20, 2017 3:13 pm

Have you played with the variables? Higher deductibles, no comprehensive/collision, no add-ons like towing, etc.? I'm guessing you will still want comprehensive/collision on a car with 15,000 miles, but maybe not.

Given how little you drive, have you done the math on going without a car? Are there Uber/Zipcar options available where you live?

User avatar
dm200
Posts: 12930
Joined: Mon Feb 26, 2007 2:21 pm
Location: Washington DC area

Re: Auto Insurance Question

Post by dm200 » Sat May 20, 2017 3:15 pm

Xpe wrote:Saw another car insurance thread, reminded me of my lack of understanding, but I didn't want to derail their thread so thought I'd start a new one.
I'd love to hear from someone who understands the auto insurance industry.
I feel like I'm missing something, I have a fairly cheap car, got it for $18k 1 year old with 3k miles, now its 5 years old with 15k miles. That should also tell you something about how few miles I drive. I've had one claim 6 years ago, and it was documentedly the other persons fault (they ran a red light and hit me as I went through the intersection).
So with:
fairly cheap car
extremely low miles per year
decent driving record
So I guess for a general question: How is it that I pay $600 per 6mo? I got quotes from a number of companies and they were all in the same range. Also I moved 15 minutes away and my insurance went up 20%. I talked to the company (progressive) and they said it was because the new area has a higher incident rate.
And then for a more actionable question: If I sold my car and bought a car for like $8k, would my insurance go down proportionally? Or do they figure that an older cheaper car is even more likely to have issues?
Also - if I hadn't updated my address, and I got into an accident, would they be able to reject my claim on due to the incorrect address?


What are the specifics for the premiums for each component of the insurance. That is important to make a fair comparison.

Jack FFR1846
Posts: 5460
Joined: Tue Dec 31, 2013 7:05 am

Re: Auto Insurance Question

Post by Jack FFR1846 » Sat May 20, 2017 3:53 pm

If you are going through an independent insurance agent, they can go through the various providers and find the best fit for you.

But insurance depends on the things that pertain to you. I had full coverage and very high liability on my Lotus Elise. This is a car where a 1 mph hit will literally total the car because there are no actual bumpers and the body panels at the front and back of the car are single-piece parts that cost $15k to replace (Lotus says you MUST replace and that they cannot be repaired) and take 6 months to ship from England.

Even with all that, I paid $700 a year. I live in a town rated the second lowest risk in my state.
Bogle: Smart Beta is stupid

dwickenh
Posts: 688
Joined: Sun Jan 04, 2015 9:45 pm
Location: Illinois

Re: Auto Insurance Question

Post by dwickenh » Sat May 20, 2017 4:53 pm

Most insurance companies are now using your credit score as part of the rating system. They have found that people with good scores also have less claims. The opposite is true of poor credit leading to increase in losses per their statistics. I am not sure if this is affecting your rates, and there are many other factors.
The market is the most efficient mechanism anywhere in the world for transferring wealth from impatient people to patient people.” | — Warren Buffett

User avatar
AAA
Posts: 806
Joined: Sat Jan 12, 2008 8:56 am

Re: Auto Insurance Question

Post by AAA » Sat May 20, 2017 8:03 pm

I had a similar confusion when my auto insurance renewed recently. For the first time none of our kids lived at home and so were only very occasional drivers of our three cars; the cars were a year older from the previous renewal; no claims during the past year and as we are retired and don't travel much, the annual mileages went down. As a result...our premium went up ! Was told by the insurance co. that it was due to a general rate increase in my area.

I shopped around and also talked with an independent insurance agent and as it turned out the premium was fairly representative, we renewed. It wasn't worth it to us to go with a co. I never heard of just for a relatively small annual saving.

Katietsu
Posts: 728
Joined: Sun Sep 22, 2013 1:48 am

Re: Auto Insurance Question

Post by Katietsu » Sat May 20, 2017 8:36 pm

Look at the breakdown for your premium. My premium is high compared to what I see people posting here. But only 20% of my premium is for comprehensive/collision. We have no accidents and are demographically a low risk. But my state has a high incidence of lawsuits, courts favorable to the plaintiff, and underinsured/uninsured drivers. Therefore, the only way to make a noticeable reduction would be to move or reduce my liability coverage (currently $300k).

miamivice
Posts: 577
Joined: Tue Jun 11, 2013 11:46 am

Re: Auto Insurance Question

Post by miamivice » Sat May 20, 2017 9:52 pm

Honestly, the mechanics of why an insurance quote is a particular amount is not really an issue. All that matters is how many quotes you get, who is the lowest, and your comfort factor with those who provided quotes.

User avatar
dm200
Posts: 12930
Joined: Mon Feb 26, 2007 2:21 pm
Location: Washington DC area

Re: Auto Insurance Question

Post by dm200 » Sun May 21, 2017 6:52 am

dwickenh wrote:Most insurance companies are now using your credit score as part of the rating system. They have found that people with good scores also have less claims. The opposite is true of poor credit leading to increase in losses per their statistics. I am not sure if this is affecting your rates, and there are many other factors.


Other than making more money for the insurance company, is there any statistical or actuarial or scientific basis justifying using a credit report for setting premiums? Anecdotally, I know a lot of folks with great credit who are terrible drivers and those with less than perfect credit who are good drivers.

Even if there is, then seems to me double counting to use BOTH driving record and credit score to set premiums.

User avatar
mrc
Posts: 937
Joined: Sun Jan 10, 2016 6:39 am
Location: right here

Re: Auto Insurance Question

Post by mrc » Sun May 21, 2017 7:17 am

dm200 wrote:Other than making more money for the insurance company, is there any statistical or actuarial or scientific basis justifying using a credit report for setting premiums? Anecdotally, I know a lot of folks with great credit who are terrible drivers and those with less than perfect credit who are good drivers.

Even if there is, then seems to me double counting to use BOTH driving record and credit score to set premiums.


I've wondered about this ever since I heard about the use of credit scores to set rates. How exactly does (an imperfect method for estimating) credit worthiness correlate with an insured's risk? It seems likely to be profoundly discriminatory.
A great challenge of life: Knowing enough to think you're doing it right, but not enough to know you're doing it wrong. — Neil deGrasse Tyson

User avatar
Shackleton
Posts: 224
Joined: Mon Dec 29, 2014 6:20 pm

Re: Auto Insurance Question

Post by Shackleton » Sun May 21, 2017 7:28 am

mrc wrote:
dm200 wrote:Other than making more money for the insurance company, is there any statistical or actuarial or scientific basis justifying using a credit report for setting premiums? Anecdotally, I know a lot of folks with great credit who are terrible drivers and those with less than perfect credit who are good drivers.

Even if there is, then seems to me double counting to use BOTH driving record and credit score to set premiums.


I've wondered about this ever since I heard about the use of credit scores to set rates. How exactly does (an imperfect method for estimating) credit worthiness correlate with an insured's risk? It seems likely to be profoundly discriminatory.

Yes, there is a definite correlation between credit score and claims when viewed in aggregate with a bunch of other factors. I guarantee that insurance companies would not spend all that money on credit reports if they were not able to satisfy their c-suite execs that it added value since auto insurance is actually a fairly low margin industry. Anecdotal evidence is useless when doing the sort of statistical analysis used by insurance companies.
“Superhuman effort isn't worth a damn unless it achieves results.” ~Ernest Shackleton

dwickenh
Posts: 688
Joined: Sun Jan 04, 2015 9:45 pm
Location: Illinois

Re: Auto Insurance Question

Post by dwickenh » Sun May 21, 2017 8:04 am

Ability to pay all of your bills, save for retirement, spend less than you make are all correlated with certain personality traits that may be appealing to an insurance company taking a risk on someone they have never met. If you think of all the people you know, which ones would you consider to be the most risky? This is not true in all cases, but is true often enough to become statistically important for profits.

Dan
The market is the most efficient mechanism anywhere in the world for transferring wealth from impatient people to patient people.” | — Warren Buffett

Xpe
Posts: 129
Joined: Wed Sep 17, 2014 7:24 am

Re: Auto Insurance Question

Post by Xpe » Sun May 21, 2017 9:39 am

Thanks for all the responses. My credit score is ~790 and I'm married.

It's honestly just really frustrating that 'miles driven' doesn't seem to have much weight, when to me it seems like for the same customer, driving the same car, in the same area, if you cut your mileage in half, that should relate to a 30-50% reduction in premium. But it seems more like 5-10%.

I would LOVE to go without a car, but I'm on daycare pick-up duty, and right now the daycare is like 5-10 miles away. I'd be up for biking, but I crash too much(lol), which is fine when it's just me but cant be having that with a kid on the back. Plus 10 miles would be a long ride for a toddler anyways. Our daycare has a location < 1 mile from the house, but that location doesn't take kids as young as ours, so in a year or so it may be a possibility.

Was looking at the best cars to insure, and even those cars are like $1000 per year to insure anyways, so I don't think my premium is that crazy, it's just annoying paying like $.30 per mile for insurance, on top of probably another $.30 per mile for depreciation.

I'll probably look at AAA insurance, I didn't realize they even did insurance. I've been a member for years and their roadside assistance has been great.

User avatar
dm200
Posts: 12930
Joined: Mon Feb 26, 2007 2:21 pm
Location: Washington DC area

Re: Auto Insurance Question

Post by dm200 » Sun May 21, 2017 10:44 am

mrc wrote:
dm200 wrote:Other than making more money for the insurance company, is there any statistical or actuarial or scientific basis justifying using a credit report for setting premiums? Anecdotally, I know a lot of folks with great credit who are terrible drivers and those with less than perfect credit who are good drivers.

Even if there is, then seems to me double counting to use BOTH driving record and credit score to set premiums.


I've wondered about this ever since I heard about the use of credit scores to set rates. How exactly does (an imperfect method for estimating) credit worthiness correlate with an insured's risk? It seems likely to be profoundly discriminatory.


"Discriminatory" is not always bad or wrong. We all "discriminate" all the time, from what we eat for breakfast to who we date and marry to the kind of car we drive. The "key" is whether "discriminatory" is unfair, or illegal or improper. I suppose (perhaps?) those with higher credit scores may be less likely to be late paying the premiums - and late payers or those that have to be reminded all the time or get the cancellation notices - and then are rescinded when they eventually do pay - drive up the costs.

We cannot get into the weeds of the issue (board rules), but I see and have strong opinions about financial (unjustified) exploitation of the poor or those with certain kinds of credit history or those of limited education or sophistication. Much of what I learned 50+ years ago (and still I think being taught) in Economics 101 about product/service pricing is just not true for many types of products and services. Competition (for such consumers) does not drive costs down the way I learned in Economics 101

talzara
Posts: 271
Joined: Thu Feb 12, 2009 7:40 pm

Re: Auto Insurance Question

Post by talzara » Sun May 21, 2017 12:20 pm

dm200 wrote:Other than making more money for the insurance company, is there any statistical or actuarial or scientific basis justifying using a credit report for setting premiums? Anecdotally, I know a lot of folks with great credit who are terrible drivers and those with less than perfect credit who are good drivers.

Even if there is, then seems to me double counting to use BOTH driving record and credit score to set premiums.


Yes. Credit history is one of the best predictors of claims frequency and claims amount. Here is an FTC report from 2007: Credit-Based Insurance Scores

Pages 20-23 of the report cite four different studies that found credit history was correlated with auto insurance claims risk. The studies are from MetLife, an insurance industry trade association, the University of Texas, and the Texas Department of Insurance.

The FTC also ran its own statistical analysis. On pages 98-99 of the PDF, they show the risk by credit history decile. The worst credit risks are more than twice as risky for auto insurers as the best credit risks:

  • 1.9 times the property damage liability risk
  • 2.4 times the body injury liability risk
  • 2.6 times the collision risk
  • 2.5 times the comprehensive risk

In fact, credit history correlates better to risk than whether you've been in an at-fault accident. The average driver gets into an at-fault accident about once every 15 years, but the CLUE database removes claims after 7 years. A lot of bad drivers show up with 0 claims, and a lot of good drivers show up with 1 claim. The data is too sparse.

Driving history is also a weak predictor of risk. The data is also too sparse. A lot of bad drivers have no tickets because they got lucky, and a lot of good drivers have tickets because they got unlucky.

Credit history avoids these problems because it's a much denser data set. You can look at how someone performed over 10 years, for a dozen different accounts, with bills being due 12 months a year.
Last edited by talzara on Sun May 21, 2017 12:43 pm, edited 1 time in total.

talzara
Posts: 271
Joined: Thu Feb 12, 2009 7:40 pm

Re: Auto Insurance Question

Post by talzara » Sun May 21, 2017 12:42 pm

Xpe wrote:It's honestly just really frustrating that 'miles driven' doesn't seem to have much weight, when to me it seems like for the same customer, driving the same car, in the same area, if you cut your mileage in half, that should relate to a 30-50% reduction in premium. But it seems more like 5-10%.


Yes, mileage is direct related to crash risk. Insurance companies have known about this effect for 100 years. However, people tend to lie about their mileage, so the insurance companies are reluctant to offer bigger discounts.

To get the really big discounts, you would have to allow the insurance company to track your driving. Since you drive only 3,000 miles a year, you are the perfect candidate for pay-per-mile insurance (PPM) or usage-based insurance (UBI).

MetroMile is the only PPM insurer that has a wide footprint in the United States. You would get about a 50% discount for driving 3,000 miles a year. The premium is calculated retrospectively, based on the actual number of miles you drove. They company only offer policies in seven states.

All of the big auto insurers except GEICO offer UBI policies on a prospective basis. You get very little discount immediately, but they track your driving and they offer you a discount at the next renewal. The discounts are much smaller, so you should expect about 20-30% discount for driving 3,000 miles a year. The tracking devices also record how you drive, so the discounts could go away if you drive aggressively.

miamivice
Posts: 577
Joined: Tue Jun 11, 2013 11:46 am

Re: Auto Insurance Question

Post by miamivice » Sun May 21, 2017 12:48 pm

I don't think that a good / excellent / decent driving record results in much less risk.

I'm personally a terrible driver. I have a hard time remembering to make head checks so I occasionally cut people off and they blare their horn at me. Sometimes I don't see pedestrians crossing the street and have nearly run over cyclists crossing the street. I also get confused easily on whose turn it is at a 4 way stop and sometimes cut in line. I'll make a bunch of other mistakes too.

However, I have a great driving record. In the past 5 years I've only received 1 ticket. Considering I drive 20,000 to 22,000 miles per year in difficult traffic situations, I think that is great.

But I don't pretend that my great driving record means I'm a great driver. Rather, I keep focusing on the areas that I need to improve on and try my best to be the best driver I can be.

oslocal
Posts: 9
Joined: Sat May 21, 2016 5:30 pm

Re: Auto Insurance Question

Post by oslocal » Wed Sep 13, 2017 6:15 pm

I went from driving about 12,000 miles per year to something like 3,000. Almost 3 years ago now, I got two speeding tickets within about one month time. Seems like GEICO picked up on it about 9 months later when they sent me a letter they wouldn't insure me. That's from a rate of 800 for 6 months with collision and two cars as well as 300/500 coverage.

I shopped around for alternatives which were mostly in the 3000 range and one in the 1500 range. I chose Metromile and it's been around 1000 for 6 months based on my driving and lately closer to 900.

Every so often I get emails from GEICO asking me to "please come back" and asking why did I leave. Funny... I guess I left before the policy was up. They're still asking for 1500.

User avatar
grabiner
Advisory Board
Posts: 20571
Joined: Tue Feb 20, 2007 11:58 pm
Location: Columbia, MD

Re: Auto Insurance Question

Post by grabiner » Wed Sep 13, 2017 10:27 pm

talzara wrote:
Sun May 21, 2017 12:42 pm
Xpe wrote: It's honestly just really frustrating that 'miles driven' doesn't seem to have much weight, when to me it seems like for the same customer, driving the same car, in the same area, if you cut your mileage in half, that should relate to a 30-50% reduction in premium. But it seems more like 5-10%.
Yes, mileage is direct related to crash risk. Insurance companies have known about this effect for 100 years. However, people tend to lie about their mileage, so the insurance companies are reluctant to offer bigger discounts.
The relationship of mileage to crash risk is reduced because high-mileage drivers have more accidents and tickets per year at the same risk level. If you get into an accident every 50,000 miles, you will have two accidents in the last five years if you drive 20,000 miles per year, or one if you drive 10,000 miles per year. Thus a driver who drives 20,000 miles per year with no accidents is less risky than one who drives 10,000 miles per year with no accidents.

Also, I have had several insurers set rates by commuting distance, which is more reliably estimated, and easier for the insurance company to prove fraud. If you work downtown and live 20 miles from downtown, you should have a 20-mile commute (or be able to prove that you actually have a 5-mile commute to a train station because you have a monthly train pass).
David Grabiner

rec7
Posts: 2146
Joined: Tue Oct 28, 2008 7:22 pm
Location: ST LOUIS

Re: Auto Insurance Question

Post by rec7 » Thu Sep 14, 2017 8:01 am

Your auto insurance can double just by moving to a different area of your city sometimes. My friend saw a 50% increase by moving within the same city.
Disclaimer: You might lose money doing anything I say. Although that was not my intent. | Favorite song: Sometimes He Whispers Jay Parrack

deltaneutral83
Posts: 359
Joined: Tue Mar 07, 2017 4:25 pm

Re: Auto Insurance Question

Post by deltaneutral83 » Thu Sep 14, 2017 8:45 am

Car theft must be popular in San Diego and I know a person whose rates tripled overnight when they moved there from MCOL east coast. In addition, credit score is usually a good indicator of most things involving payments. That seems fairly obvious to me. In the end, best thing to do is price it out to 3 or 4 companies and pick, not much else you can do.

User avatar
Pranav
Posts: 171
Joined: Thu Aug 25, 2016 11:25 am
Location: Austin, Texas, United States

Re: Auto Insurance Question

Post by Pranav » Thu Sep 14, 2017 8:54 am

Shackleton wrote:
Sat May 20, 2017 3:09 pm
anoop wrote:
Shackleton wrote:
anoop wrote:BTW, Progressive is known to specialize in high-risk policies. Their rates are typically lower than other traditional insurance companies (Amica, AAA, Farmer's, etc.) if you have one or more active at-fault accidents (last 3 years). If not, you will typically get a lower rate moving to a more traditional insurer, the kind would typically refuse to sell you a policy after two at-fault accidents or charge you an obscene amount.
That is how Progressive started but is no longer true and hasn't been true for at least a decade.
Ok. My last experience with them was 12 years ago. I had been with them for 3 years because of accidents and they flat out told me they don't try to compete with others for drivers with a good record. It is interesting they changed their business model.
It's why they are now the #4 auto insurer and likely to be #3 by the end of the year.
I am interested in the source for the ranking.
Pranav

User avatar
dm200
Posts: 12930
Joined: Mon Feb 26, 2007 2:21 pm
Location: Washington DC area

Re: Auto Insurance Question

Post by dm200 » Thu Sep 14, 2017 10:22 am

talzara wrote:
Sun May 21, 2017 12:20 pm
dm200 wrote:Other than making more money for the insurance company, is there any statistical or actuarial or scientific basis justifying using a credit report for setting premiums? Anecdotally, I know a lot of folks with great credit who are terrible drivers and those with less than perfect credit who are good drivers.

Even if there is, then seems to me double counting to use BOTH driving record and credit score to set premiums.
Yes. Credit history is one of the best predictors of claims frequency and claims amount. Here is an FTC report from 2007: Credit-Based Insurance Scores
Pages 20-23 of the report cite four different studies that found credit history was correlated with auto insurance claims risk. The studies are from MetLife, an insurance industry trade association, the University of Texas, and the Texas Department of Insurance.
The FTC also ran its own statistical analysis. On pages 98-99 of the PDF, they show the risk by credit history decile. The worst credit risks are more than twice as risky for auto insurers as the best credit risks:
  • 1.9 times the property damage liability risk
  • 2.4 times the body injury liability risk
  • 2.6 times the collision risk
  • 2.5 times the comprehensive risk
In fact, credit history correlates better to risk than whether you've been in an at-fault accident. The average driver gets into an at-fault accident about once every 15 years, but the CLUE database removes claims after 7 years. A lot of bad drivers show up with 0 claims, and a lot of good drivers show up with 1 claim. The data is too sparse.
Driving history is also a weak predictor of risk. The data is also too sparse. A lot of bad drivers have no tickets because they got lucky, and a lot of good drivers have tickets because they got unlucky.
Credit history avoids these problems because it's a much denser data set. You can look at how someone performed over 10 years, for a dozen different accounts, with bills being due 12 months a year.
Interesting - This "association" puzzles me greatly. Of course, "association" does not demonstrate "cause", since bad driving may cause bad credit scores - or some other factor "causes" both bad credit score and bad driving.

talzara
Posts: 271
Joined: Thu Feb 12, 2009 7:40 pm

Re: Auto Insurance Question

Post by talzara » Thu Sep 14, 2017 1:25 pm

grabiner wrote:
Wed Sep 13, 2017 10:27 pm
talzara wrote:
Sun May 21, 2017 12:42 pm
Yes, mileage is direct related to crash risk. Insurance companies have known about this effect for 100 years. However, people tend to lie about their mileage, so the insurance companies are reluctant to offer bigger discounts.
The relationship of mileage to crash risk is reduced because high-mileage drivers have more accidents and tickets per year at the same risk level. If you get into an accident every 50,000 miles, you will have two accidents in the last five years if you drive 20,000 miles per year, or one if you drive 10,000 miles per year. Thus a driver who drives 20,000 miles per year with no accidents is less risky than one who drives 10,000 miles per year with no accidents.
The crash risk vs. mileage curve is indeed concave. However, that's not why the insurance companies act that way.

Traditional insurance policies give minimal discounts for low-mileage driving. A car that is driven 1 mile a year will be charged 80% as much as a car that is driven 12,000 miles a year.

However, the same insurance companies are willing to offer substantial discounts for verified mileage. For example, in State Farm's UBI program, a car that is driven 1 mile a year will only pay 50% as much as a car that is driven 12,000 miles a year.

The insurance companies simply don't believe self-reported mileage.

They have good reason not to. Here's a study from the California Department of Insurance on underreporting of mileage: http://www.insurance.ca.gov/0400-news/0 ... 2606-2.pdf

Compare Self-Reported Mileage (Chart 1, page 19) to Actual Mileage (Chart 2, page 20). If California drivers are to be believed, they're driving less than the Japanese.
grabiner wrote:
Wed Sep 13, 2017 10:27 pm
Also, I have had several insurers set rates by commuting distance, which is more reliably estimated, and easier for the insurance company to prove fraud. If you work downtown and live 20 miles from downtown, you should have a 20-mile commute (or be able to prove that you actually have a 5-mile commute to a train station because you have a monthly train pass).
It's just as easy to prove that self-reported mileage is fraudulent. Just check the odometer.

The problem is that insurance companies only look for fraud when a claim is filed. They're not checking your odometer every month, and they won't check your train pass every month.

talzara
Posts: 271
Joined: Thu Feb 12, 2009 7:40 pm

Re: Auto Insurance Question

Post by talzara » Thu Sep 14, 2017 2:08 pm

dm200 wrote:
Thu Sep 14, 2017 10:22 am
Interesting - This "association" puzzles me greatly. Of course, "association" does not demonstrate "cause", since bad driving may cause bad credit scores - or some other factor "causes" both bad credit score and bad driving.
Insurance companies only have to demonstrate correlation, not causality. As long as the rate factor is justified by their actual claims history, they can use it. They don't actually want to discover the real cause, because there's a risk that it will prove distasteful .

There are many hypotheses for why credit history works so well in auto insurance.

Some say that credit score is a proxy for race or income. However, studies have shown that the association remains even after controlling for race and income. Drivers with good credit histories are less risky, on average, than drivers of the same race and income who have bad credit histories.

Others say that people with good credit history are simply paying out of pocket. They're still getting into car crashes, but the insurance company is seeing a reduced claims risk.

The insurance companies say that credit history is a proxy for responsibility. Someone who is responsible is likely to pay bills on time and drive carefully. There is also a lower risk of insurance fraud.

User avatar
djpeteski
Posts: 164
Joined: Fri Mar 31, 2017 9:07 am

Re: Auto Insurance Question

Post by djpeteski » Thu Sep 14, 2017 2:24 pm

Keep in mind that part of your insurance coverage is liability. If there is an auto accident, and you are sued, the auto insurance company is responsible. In order to maintain a umbrella policy you need to have a high minimum amount of liability on your auto insurance. Typically that amount is 300K personal and 100K property. The state minimums are typically much lower.

So, if you live in a sue happy area, this may be reflected in your rates.

I'd shop around a bit and see if you can get your premiums lowered. Make sure you do a same coverage comparison. It seems like you are a bit of an outlier and you are being lumped into a group that pays a bit higher than what would be expected.
Last edited by djpeteski on Thu Sep 14, 2017 2:26 pm, edited 1 time in total.

MrsBDG
Posts: 43
Joined: Thu Apr 13, 2017 4:04 pm

Re: Auto Insurance Question

Post by MrsBDG » Thu Sep 14, 2017 2:25 pm

It's possible people with great credit scores have more available cash and avoid making smaller claims, either due to a high deductible or just based on choosing to pay cash for repairs in order to avoid insurance costs increasing.

Jags4186
Posts: 1224
Joined: Wed Jun 18, 2014 7:12 pm

Re: Auto Insurance Question

Post by Jags4186 » Thu Sep 14, 2017 2:34 pm

Without knowing your limits and deductibles this question is impossible to answer. $600 could be an excellent rate or it could be highway robbery.

As I stated elsewhere we pay right around $2900/yr for insurance on two compact Japanese sedans. Seems high considering we have no points and one no fault accident that the other drivers insurance covered. But it's the best we can find so we have to deal with it for the coverages we want. I'm unwilling to trade $1000 in premiums to drop collision on two cars which are under 5 years old. As many of you know any body damage these days costs well north of $1000 to fix. A simple "dent" and some "scratches" on my driver's side door cost nearly $3900 to repair.

talzara
Posts: 271
Joined: Thu Feb 12, 2009 7:40 pm

Re: Auto Insurance Question

Post by talzara » Thu Sep 14, 2017 6:51 pm

djpeteski wrote:
Thu Sep 14, 2017 2:24 pm
I'd shop around a bit and see if you can get your premiums lowered. Make sure you do a same coverage comparison. It seems like you are a bit of an outlier and you are being lumped into a group that pays a bit higher than what would be expected.
That's already been done:
Xpe wrote:
Sat May 20, 2017 12:48 pm
So I guess for a general question: How is it that I pay $600 per 6mo? I got quotes from a number of companies and they were all in the same range.
However, you're correct that this case is an outlier.

The OP owns one car and drives it 3,000 miles a year. Only 10% of American cars are driven 3,000 miles or less every year, and that includes second and third cars. The average American car is driven 12,000 miles a year.

Americans drive a lot. If you drive less than 90% of your fellow Americans, then you will be overcharged if you stay with a traditional insurance policy. The only way to avoid subsidizing them is to switch to a UBI or PPM policy that gives steep discounts for low mileage.

Wakefield1
Posts: 548
Joined: Mon Nov 14, 2016 10:10 pm

Re: Auto Insurance Question

Post by Wakefield1 » Thu Sep 14, 2017 7:27 pm

I suspect that mileage (and date) that is recorded on my "Inspection Certification:Virginia State Police" is available to my insurance company. Or at least the mileage on my biannual "Vehicle Emissions Inspection Report".

talzara
Posts: 271
Joined: Thu Feb 12, 2009 7:40 pm

Re: Auto Insurance Question

Post by talzara » Fri Sep 15, 2017 9:52 am

Wakefield1 wrote:
Thu Sep 14, 2017 7:27 pm
I suspect that mileage (and date) that is recorded on my "Inspection Certification:Virginia State Police" is available to my insurance company. Or at least the mileage on my biannual "Vehicle Emissions Inspection Report".
Carfax also collects odometer readings every time you get an oil change.

There are some companies that don't tolerate mileage underreporting. In the California DOI study, chart 9 (page 27) shows the percentage of cars that are underreported by more than 5000 miles. Company 5 is seeing 30% of its covered vehicles underreport by that much, but company 2 is only seeing 10%. Company 2 must be doing something to crack down on mileage underreporting.

However, most insurance companies don't do this. I don't know why. According to the numbers in the California DOI study, the average driver is underreporting mileage to his insurance company by 30%!

neilpilot
Posts: 899
Joined: Fri Dec 04, 2015 1:46 pm
Location: Memphis area

Re: Auto Insurance Question

Post by neilpilot » Fri Sep 15, 2017 9:58 am

talzara wrote:
Fri Sep 15, 2017 9:52 am
Wakefield1 wrote:
Thu Sep 14, 2017 7:27 pm
I suspect that mileage (and date) that is recorded on my "Inspection Certification:Virginia State Police" is available to my insurance company. Or at least the mileage on my biannual "Vehicle Emissions Inspection Report".
Carfax also collects odometer readings every time you get an oil change.
Not for my 2 cars. I never report the mileage to Carfax when I change oil.

talzara
Posts: 271
Joined: Thu Feb 12, 2009 7:40 pm

Re: Auto Insurance Question

Post by talzara » Fri Sep 15, 2017 10:30 am

neilpilot wrote:
Fri Sep 15, 2017 9:58 am
talzara wrote:
Fri Sep 15, 2017 9:52 am
Wakefield1 wrote:
Thu Sep 14, 2017 7:27 pm
I suspect that mileage (and date) that is recorded on my "Inspection Certification:Virginia State Police" is available to my insurance company. Or at least the mileage on my biannual "Vehicle Emissions Inspection Report".
Carfax also collects odometer readings every time you get an oil change.
Not for my 2 cars. I never report the mileage to Carfax when I change oil.
Every time you "get" an oil change. A do-it-yourselfer would "do" an oil change" or "change my oil."

Besides, insurance companies aren't buying your car. They're only covering your car. They don't need to get odometer readings at every oil change. You might even get oil changes at a small shop that doesn't report to Carfax.

They only need to get enough odometer readings to assure themselves that mileage isn't being underreported. State inspections are only done every 2 years, and not every state does inspections. They would need something that's reported more frequently and is available in every state.

If you do all your own maintenance and never show up in Carfax, then they can just put in a rate factor for unverifiable mileage. It's the same thing they do when they can't pull your credit report.

I think there aren't enough low-mileage American drivers for them to address this market. Low-mileage drivers are also entitled to lower premiums, so they would be an even smaller fraction of revenues. However, mileage comes for free with UBI policies, so they might as well use it.

Post Reply