Credit Score & Car Insurance?

[Anonymous]

Anyone know what factors car insurers use when evaluating credit worthiness? I live in a state where Consumer Reports says credit score makes a pretty big difference for car insurance rates, hundreds of dollars a year been good credit and bad. So now I am wondering what exactly the insurers look at -- is it the same as FICO?

I saw a comment on another thread that insurers look at "average credit limit" -- which would seem to imply that closing accounts with low limits would make a difference. For example I have a First Tech visa with a  $5000 limit and I never use it. My CITI/Discover/BoA cards are in the $10k range any my Amex is $25k. So presumably that $5000 card is pulling down my average limit. For FICO it wouldn't hurt me much to close it -- $5000 is only about 5% of my total limit, and the card is not my oldest.

So could I save money by closing it? 

What else do they look at? If car insurance makes that big a difference it's right up there at the #2 most important thing a credit score is used for, right after #1 mortgage rate, in terms of financial impact. At least for me.

[Anonymous]

Hello Canadian!  Haven't seen you on this board in a while.  Hope you had a good Thanksgiving.

The insurance industry does indeed have a credit score that they use and it does differ in some ways from FICO's models.  Like your FICO score, the insurance score is based purely on what is in your credit report.  An insurance company uses that score as one determinant for setting your premium: the others being the obvious ones of driver age, driver's record of previous accidents, zip code you live in, make and model of the car, etc.

The first two ways to improve one's insurance score are (1) to make sure you are doing well on the basic FICO scoring factors (low CC utilization, no lates or collections or similar derogs, decent AAoA, etc.) and (2) sign up for Credit Karma.  In addition to its Vantage scores, Karma also offers a free "insurance" score.  It will give you some idea of what an insurer might see.  (There are two insurance models that are used and Karma has access to one of them.)  You should look at the Karma auto insurance score and compare it to a TU FICO 8 done on the same date.  This will give you a sense of how well you are doing on the non-FICO factors that the insurance model uses.

Average Credit Limit (ACL) is one such factor.  Closing a low CL card does not help you (which was your particular question) since the insurance model (at least the Lexis Nexis one) still counts that closed account in its average, though it treats all closed accounts as having a CL of $0.  Thus closing a card will harm your ACL in the short term, unless you can also convince the creditor to delete the closed account from your report.

Trying to get one's ACL above 11k is not a silly concern, though it's a bad tactic (in my opinion) if it means reducing one's total number of cards to under 6.  Vantage Score also includes ACL as a scoring factor.  If you have a card that could easily get a higher CL by making soft-pull CLI requests, that may be smarter.  That said, if a card is worthless to you and it is not especially old and its CL is < 20% of your current ACL, then there may be a case for closing it even if it takes ten years to get off your reports.

Much better lines of attack (if one is concerned about non-FICO factors in the insurance model) is seeing if you can sharply reduce or eliminate altogether all of your store cards and any auto accounts (Pep Boys, Auto Zone, etc.).  Ideally you'd want to see if you can also get the creditors to delete the accounts once closed from your reports.  The presence of store cards and auto accounts are considered bad in the insurance models.

Another way the insurance models appear to differ from FICO is the vastly greater emphasis they appear to place on the presence of new accounts and the attempt to obtain new accounts.  FICO cares about that but not nearly as much as the insurance models care.

Note that some states prevent auto insurers from using credit scores in their decisions.  I believe those three states are California, Massachusetts and Hawaii.

[newhis]

My DW insurance was due for renew.

Her car insurance score is low, I guess this is because a) new credit history (less than 3 years), b) ACL around 6k, c) new account 11 months ago and d) no previous car insurance record in USA.

I called to lower the payment. I asked about the credit history saying that  now is much better than a year ago, almost 60 points FICO 8, not that much on car insurance from CK.

They agreed to lower the payment and they issue a letter stating that her score is about 50% of those with the company. So  even when the car insurance score didn't go up much, she got a lower rate.

[Anonymous]

@"Hello Canadian!  Haven't seen you on this board in a while.  Hope you had a good Thanksgiving."

Thanks ) Yes I have been quiet lately. I've been in the garden since April so I poke my head in here less frequently. Planning to stay in the garden until this coming April when I may go looking for a mortgage. At that time my last inq will roll off, my AAoA will go over 2yrs, and I will get my annual bonus which will complete my saving for a down payment. It's not a coincidence those things all happen at the same time  

@"Average Credit Limit (ACL) is one such factor.  Closing a low CL card does not help you (which was your particular question) since the insurance model (at least the Lexis Nexis one) still counts that closed account in its average, though it treats all closed accounts as having a CL of $0.  Thus closing a card will harm your ACL in the short term, unless you can also convince the creditor to delete the closed account from your report."

Hmm. Can't see how I'm going to get anything deleted. All my cards were used at one point and all are in good standing, never late. I closed two accounts because they had an AF. One was Amex PRG, I'm not sure if charge cards are treated differently. The other was a LOC that I had which I wasn't using and I closed because it had a $25 AF. 

@"Trying to get one's ACL above 11k is not a silly concern, though it's a bad tactic (in my opinion) if it means reducing one's total number of cards to under 6."

The ACL of my eight open accounts is $11,312 right now so I guess I'm good there. If you add the two I closed as $0 limit it's $9050. So I guess what I need to do is work on CLI's until I get another $950 ACL, which would require a total of $9500 in CLI's on my existing accounts (divide by 10 accts incl 2 closed).

That seems like something to strive for  

@"seeing if you can sharply reduce or eliminate altogether all of your store cards and any auto accounts (Pep Boys, Auto Zone, etc.)."

The only store card I have is the Amazon store card from Synchrony Bank, I don't know if it counts as a store card in a bad way. I guess it probably does. But I guess I also like the 5% back at Amazon so I'll keep it.

@"Note that some states prevent auto insurers from using credit scores in their decisions.  I believe those three states are California, Massachusetts and Hawaii."

I'm in WA and here they use it. Which probably hurt me last year when I got set up with car insurance. Hopefully when it renews my much improved credit score gives me a break!

[Anonymous]

Since you have plenty of credit cards now, my guess is that your overwhelming driver is preparing for your mortgage, and that part of your plan is not applying for any new accounts -- except for the mortgage.

The good news as far as the auto insurance score goes is that the longer you can refrain from opening a new account, the better.  The auto insurance models seem extremely sensitive to inquiries and the opening of new accounts.  You have got a lot else positive going for you: almost no store cards, no auto accounts, a high ACL, etc.

I think that Tom Thumb mentioned that auto insurance companies rarely pull your insurance score again by default, though if you ask them to they will.  If your next premium is timed conveniently so that your score is much better, you may want to ask them to pull it again.

LexisNexis defines a new account as one opened in the last 24 months.   When your policy renews next year, how many new accounts will you have at that point?

[Anonymous]

OK! So my average credit limit is over 11k now, including closed accounts as zero CL. Mostly achieved by CLI on Discover and Amex. My Amex is at $40k now, maybe tempting a financial review (bring it, I've been honest).

All without a HP! Just a regular SP CLI every few months.

So now it's going to take a month likely before these new limits reflect on my report.

What then? Call the car insurer and ask to revaluate me?

Some questions: was that $11k average CL from something official? An industry standard? Or just a guess? Should I keep trying to raise ACL or am I there now? I mean it doesn't hurt. But I don't want to call the insurer a bunch of times would like to get this done just once.

Not sure how much more I can increase it. CITI will likely give me another CLI soon but probably only another $1500 which only will raise my ACL by $150. Need to wait another 60 days for Discover, at least, and I don't know if I want to push Amex any further. Maybe I can get my ACL to $12k but much higher than that seems unlikely. I need another $40k CLI to get over $15k ACL and I don't see that happening.

[CreditDunce]

LN has published a list of reason codes:

https://consumer-solutions.custhelp.com/app/answers/detail/a_id/2523/~/how-do-i-learn-more-about-sco...

Code 3038 is the one for Average bank credit line

https://consumer-solutions.custhelp.com/ci/fattach/get/2930352/0/filename/Attract+Auto+3+0+Expanded+...

Good luck in trying to unravel the codes.  When I tried, I kept finding if I tried maximizing one reason code, I would hurt another code.  You can also purchase your LN auto or home score if you like.  I don't believe the insurance risk score increases your premiums much unless it is really low.    I am assuming your insurance company uses LN.  You can tell by looking at your credit reports after requesting a quote.  You will see a SP via LN.  Not all insurance companies use insurance risk scores.

ETA: The best auto LN codes are for ACL above $10,532.  The best home LN codes are for ACL above $13,001.  Note, IIRC high limits on store cards hurts your score.

[Anonymous]

Like CreditD, I also observed a certain amount of "robbing Peter to pay Paul" with the LN codes.  I.e. the more you try to optimize your profile with respect to code X the more you start to hurt your profile with code Y.

That said, I also observed many areas where its very straightforward -- the obvious way eliminating all penalty with respect to code Z does not create a problem for any other codes.  That there are a lot of examples of that.

If I remember right, ACL is a straightforward win.  You won't hurt some other code by increasing your ACL.

Eliminating all the traditional derogs is another example: lates, collections, chargeoffs.  There's no part of the LN algorithm that likes those.

Eliminating all store cards and all auto accounts (pep boys, autozone, etc.).  Again, no part of the LN algorithm likes those.  (Obviously if your profile consisted solely of store cards, then eliminating all store cards would ultimately hurt your LN score, but here I am assuming you have at least three major CCs and that you can still keep your util very low without the store cards.)

Keeping your CC utilization extremely low is a simple win.  (If I remember right, LN cares about actual dollar value of debt as well.) 

Increasing your AAoA is a simple win.  (Curiously, if your Age of Oldest Account is much older than your AAoA, then that's bad in LN.  I think they use that as one of their markers for people who open lots of accounts.)

Assuming you have several accounts already, then avoiding any further inquiries or new accounts is a simple win.  LN is very sensitive to new credit (meaning it is bad) and I don't remember seeing any codes that implied they'd ding you if you went for ten years (say) without an inquiry or a new account.

[Anonymous]

Thanks! So for my car insurance question, it sounds like I'm there with >$11k in ACL, at least in terms of ACL. That includes only one store card, my Amazon store card, which I regularly use and I like earning the 5% so I won't ditch that. The rest are bank credit cards plus a PLOC.

I'll ping them in April. By then a few things will have happened: These new limits will have posted, my AAoA will have reached 24m, and my last inquiry will be >1yr old which may help from a FICO perspective. No derogs. I'll make sure my balances post small that month as well (and for several reasons, may go mortgage shopping around then).

It's interesting that having >13k ACL is beneficial for some purpose. I don't see it happening anytime really soon but it seems achievable. I need to rack up another $18k in CLI's on my existing cards to get there since I currently have 118k/10 = 11.8k ACL (including closed accounts as 0). Seems achievable in a year or so, my last Discover CLI was $6k so I just need 3 more like that in a year and I can get to $13k ACL for whatever (little) it's worth. Mostly I've been chasing CLI's to make my overall utilization lower.

[Anonymous]

Sounds like a good plan, C-I-S.  A few quick thoughts:

*  I seem to remember some reason codes that implied LN was especially sensitive to CC utilication and to even the raw dollar value of debt.  Therefore, if I were you, I would lower your util to a really low number.  E.g. all cards zero except one (AZEO) with $10 or so on the remaining card.  My point here is that you should not assume that a 5% utilization (in the typical FICO way) would be safe.  For you, 5% would be almost $6000 of debt -- way too much to be optimal for LN.

*  A fairly secret part of FICO's algorithm is that many models drop a card from its computation of total credit limit if that card's CL is big enough.  A group of folks have been trying to figure out where those break points might be.  For old FICO models they were not very high, like 29.9k or 34.9k, if I remember right.  LN might also do this.  Thus, it's possible that your 40k card(s) might be actually lowering your total CL, and therefore your ACL.  E.g. while they were 24, 26, 28k they helped a lot, but then they cross some breakpoint and the card starts being ignored.  I do not know that LN does this, I am just saying that FICO does it, so it's possible.

40k is probably fine.  (Again, who knows?)  We appear to have some recent confirmation that FICO 8 does not drop 50k cards from total CL, though apparently some earlier FICO models do.  But if I were you, I wouldn't go out of my way to increase a card's CL any higher than what you have now, in pursuit of some phantom ACL benefit. 

The likelihood that you would get some homeowner insurance premium benefit for increaseing your ACL from 11 to 13 strikes me as unlikely anyway.  11k is a very strong ACL.

*  You mention that you might want to shop for a mortgage at the same time that you ask your auto insurance company to re-pull your LN scores.  Be sure to delay all the mortgage inquiries till AFTER the LN pull and resulting new premium.  LN is extremely sensitive to inquiries and new credit.