Full Version: Is auto insurance pricing racist?
Front page banner article in the December 20th edition of the San Jose Mercury News: CARS COSTLIER TO INSURE IN BLACK, LATINO AREAS
The article cites a study which found that a hypothetical good driver would pay different auto insurance rates depending on the zip code where they live.
It goes on to quote an advocate of the group conducting the study who makes the extension that this shows auto insurance companies are participating in discriminatory and racist pricing:
The insurance companies responded by saying that they based pricing primarily on age and driving record, followed by zip code, but that race plays no factor in rate determination.
In response:
Questions for debate and my thoughts:
1. Does a correlation between rates and zip codes indicate racist pricing by the insurance companies?
No it doesn't. Statiticians are always careful to point out that correlation is not an indicator of causality. The fact that black and latino zip codes have higher insurance rates is not because they are black and latino neighborhoods. The insurance company statisticians have discovered that those zip codes have a higher preponderance of claims then other neighborhoods, which makes living in them a good predictive indicator of how much will have to be paid out in claims. Perhaps there is more crime in those neighborhoods, perhaps there are some poorly maintained roads, perhaps there is bad traffic, etc. There could be a number of factors contributing to the high rate of claims, but nothing to indicate racism.
2. Should California law be changed to make it illegal to base insurance rates on zip code?
No, since as I described above basing rates on zip codes makes mathematical sense. It charges people who live in high risk neighborhoods more, thus making them accountable for the decision of where they chose to live.
3. The quote from the CA insurance commissioner spokesperson indicates that they think the current system is unfair. Do you agree?
No, the current system bases prices on the likelyhood of loss. This is the only fair way to price insurance.
4. Was it responsible for the newspaper to print this article with an inflammatory headline? (for those not in the Bay Area, this is all over the news here)
Yes. All the study indicate was that zip codes and rates are correlated. To suggest more in the headline is inaccurate and bad journalism. It was the conclusion of the study authors (who appear to have a bone to pick based on their quotes in the article and the petition they've send to the state) that racism is the root cause, but there is no evidence to suggest a causal link between racism and rates. Absent such evidence, the paper should not have printed their statements except as a qualified aside. (i.e. the study authors feel that the results indicate racism in pricing, but the study does not include any evidence this is the case.) It should have simply printed the results of the study with a headline such as: "Zip code major factor in determining auto insurance rates".
Unresponsible journalism such as this only seeks to inflame racial tensions in this country. It preys on those without a knowledge of statistics and plays to people's natural fear that they are somehow being taken advantage of.
The article cites a study which found that a hypothetical good driver would pay different auto insurance rates depending on the zip code where they live.
QUOTE
Rates were 12 percent higher for people with excellent driving records covered by the state's largest auto insurer in 2004 if they lived in a mostly Latino ZIP code, the consumer group said in its analysis. Good drivers in African-American neighborhoods were charged 37 percent more.
One South Bay example: A hypothetical 22-year-old woman with a perfect driving record would have paid $1,277.10 in 2004 if she lived in Morgan Hill's 95037 ZIP code, which is 61.5 percent white, the report found. However, she would have paid $1,402.92 if she lived in Gilroy's nearby 95020 ZIP code, which is 51.1 percent Hispanic, according to the consumer group's analysis.
One South Bay example: A hypothetical 22-year-old woman with a perfect driving record would have paid $1,277.10 in 2004 if she lived in Morgan Hill's 95037 ZIP code, which is 61.5 percent white, the report found. However, she would have paid $1,402.92 if she lived in Gilroy's nearby 95020 ZIP code, which is 51.1 percent Hispanic, according to the consumer group's analysis.
It goes on to quote an advocate of the group conducting the study who makes the extension that this shows auto insurance companies are participating in discriminatory and racist pricing:
QUOTE
These disparities are due to discriminatory insurance regulations that allow insurers to base their rates primarily on where drivers live, instead of how well they drive,'' said Mark Savage, senior attorney with Consumers Union's West Coast office in San Francisco.
The insurance companies responded by saying that they based pricing primarily on age and driving record, followed by zip code, but that race plays no factor in rate determination.
QUOTE
Insurance companies don't use race as part of their rating criteria,'' said Sam Sorich, president of the Association of California Insurance Companies, a trade group. ''The Consumers Union study is a distraction from the fundamental point that insurance companies should be using data that predicts the likelihood of losses. Territory is a significant predictor.
In response:
QUOTE
California Insurance Commissioner John Garamendi accepted a petition...to end ZIP code-based auto-insurance rates. *snip* Norman Williams, a spokesman in Garamendi's office, said Monday that the commissioner is ''looking at how to change the system and make it more fair.
Questions for debate and my thoughts:
1. Does a correlation between rates and zip codes indicate racist pricing by the insurance companies?
No it doesn't. Statiticians are always careful to point out that correlation is not an indicator of causality. The fact that black and latino zip codes have higher insurance rates is not because they are black and latino neighborhoods. The insurance company statisticians have discovered that those zip codes have a higher preponderance of claims then other neighborhoods, which makes living in them a good predictive indicator of how much will have to be paid out in claims. Perhaps there is more crime in those neighborhoods, perhaps there are some poorly maintained roads, perhaps there is bad traffic, etc. There could be a number of factors contributing to the high rate of claims, but nothing to indicate racism.
2. Should California law be changed to make it illegal to base insurance rates on zip code?
No, since as I described above basing rates on zip codes makes mathematical sense. It charges people who live in high risk neighborhoods more, thus making them accountable for the decision of where they chose to live.
3. The quote from the CA insurance commissioner spokesperson indicates that they think the current system is unfair. Do you agree?
No, the current system bases prices on the likelyhood of loss. This is the only fair way to price insurance.
4. Was it responsible for the newspaper to print this article with an inflammatory headline? (for those not in the Bay Area, this is all over the news here)
Yes. All the study indicate was that zip codes and rates are correlated. To suggest more in the headline is inaccurate and bad journalism. It was the conclusion of the study authors (who appear to have a bone to pick based on their quotes in the article and the petition they've send to the state) that racism is the root cause, but there is no evidence to suggest a causal link between racism and rates. Absent such evidence, the paper should not have printed their statements except as a qualified aside. (i.e. the study authors feel that the results indicate racism in pricing, but the study does not include any evidence this is the case.) It should have simply printed the results of the study with a headline such as: "Zip code major factor in determining auto insurance rates".
Unresponsible journalism such as this only seeks to inflame racial tensions in this country. It preys on those without a knowledge of statistics and plays to people's natural fear that they are somehow being taken advantage of.
1. Does a correlation between rates and zip codes indicate racist pricing by the insurance companies? No, unless one makes the case that the Insurance Commission is also racist. All rates are approved by the state.
2. Should California law be changed to make it illegal to base insurance rates on zip code? No, doing so will simply drive up the cost of insurance as the companies will have to invest in additional technology to accurately calculate the "geographic" aspect of their pricing.
3. The quote from the CA insurance commissioner spokesperson indicates that they think the current system is unfair. Do you agree? That the current system is unfair? Maybe, but only because of the requirement to carry insurance.
4. Was it responsible for the newspaper to print this article with an inflammatory headline? (for those not in the Bay Area, this is all over the news here) No, the use of an inflammatory headline is irresponsible, and par for the course.
2. Should California law be changed to make it illegal to base insurance rates on zip code? No, doing so will simply drive up the cost of insurance as the companies will have to invest in additional technology to accurately calculate the "geographic" aspect of their pricing.
3. The quote from the CA insurance commissioner spokesperson indicates that they think the current system is unfair. Do you agree? That the current system is unfair? Maybe, but only because of the requirement to carry insurance.
4. Was it responsible for the newspaper to print this article with an inflammatory headline? (for those not in the Bay Area, this is all over the news here) No, the use of an inflammatory headline is irresponsible, and par for the course.
1. Does a correlation between rates and zip codes indicate racist pricing by the insurance companies?NO. Insurance companies use statistics. So regardless of how careful a driver you are if you live in an area where (statistically) you are more likely to get into an accident (your fault or not) or have your car stolen – your rates are higher.
And note many states (like MA) IMO unjustly spread that risk by forcing all residents of the state to pay more because the city of Boston has a much higher “risk”.
2. Should California law be changed to make it illegal to base insurance rates on zip code?
NO
3. NO
4. Was it responsible for the newspaper to print this article with an inflammatory headline? (for those not in the Bay Area, this is all over the news here)
Of course. This is the kind of trash that sells newspapers.
And note many states (like MA) IMO unjustly spread that risk by forcing all residents of the state to pay more because the city of Boston has a much higher “risk”.
2. Should California law be changed to make it illegal to base insurance rates on zip code?
NO
3. NO
4. Was it responsible for the newspaper to print this article with an inflammatory headline? (for those not in the Bay Area, this is all over the news here)
Of course. This is the kind of trash that sells newspapers.
1. Does a correlation between rates and zip codes indicate racist pricing by the insurance companies?
2. Should California law be changed to make it illegal to base insurance rates on zip code?
I’m going to lump these together. I don’t believe the zip code ranking is necessarily racist, as many have pointed out. What it IS is unfair to good drivers who live in not-so-good areas. Not every resident in every high-priced zip code is a minority. Let me give you just one example:
I used to live in a rental house in zip code 90807 (a very nice area of Long Beach, CA called “Bixby Knolls” with the attendant cachet), but the house was actually over the “border” in unincorporated Signal Hill. The poohbahs of Signal Hill got it into their collective pointy heads to lobby for their own zip code. They won, and SH got its very own zip code (though not its own post office, go figure). Now, Signal Hill encompasses some very dicey areas. OK, so I call my insurance company to update my address…and am told my insurance will go up nearly $700 a year! I impress upon them that I am the same person living in the very same house, on the very same street, with the very same vehicle in the very same circumstances as the day before. My house did not get picked up by a tornado and plopped somewhere else. I’m told it doesn’t matter; the zip code is different. (I’d be curious to see how it could be justified based on statistics, since the zip code didn’t even EXIST up to that point!)
And I have since moved to an even “less desirable” zip code, but since the move also corresponded with a change of vehicle, I can’t say how much rate difference there was from Signal Hill…but I know for a fact that it is big difference from 90807.
The only statistic that I can see that would have any validity for jacking rates up based on zip code, is auto theft. Lindsay Lohan’s driving record is for crap, and I’m sure she has a Brentwood zip code.
Where you l live doesn’t matter a tinker’s damn as to how you drive. Should it matter that you happen to live in the same zip code as a notoriously accident-prone freeway intersection? In my opinion, heck no.
In conclusion, yes, California should change their method of calculating insurance to something NOT exclusively based on zip code.
3. The quote from the CA insurance commissioner spokesperson indicates that they think the current system is unfair. Do you agree?
Yes.
4. Was it responsible for the newspaper to print this article with an inflammatory headline? (for those not in the Bay Area, this is all over the news here)
Hey, inflammatory headlines are a newspaper’s bread and butter. Nothing new there, move along now...
2. Should California law be changed to make it illegal to base insurance rates on zip code?
I’m going to lump these together. I don’t believe the zip code ranking is necessarily racist, as many have pointed out. What it IS is unfair to good drivers who live in not-so-good areas. Not every resident in every high-priced zip code is a minority. Let me give you just one example:
I used to live in a rental house in zip code 90807 (a very nice area of Long Beach, CA called “Bixby Knolls” with the attendant cachet), but the house was actually over the “border” in unincorporated Signal Hill. The poohbahs of Signal Hill got it into their collective pointy heads to lobby for their own zip code. They won, and SH got its very own zip code (though not its own post office, go figure). Now, Signal Hill encompasses some very dicey areas. OK, so I call my insurance company to update my address…and am told my insurance will go up nearly $700 a year! I impress upon them that I am the same person living in the very same house, on the very same street, with the very same vehicle in the very same circumstances as the day before. My house did not get picked up by a tornado and plopped somewhere else. I’m told it doesn’t matter; the zip code is different. (I’d be curious to see how it could be justified based on statistics, since the zip code didn’t even EXIST up to that point!)
And I have since moved to an even “less desirable” zip code, but since the move also corresponded with a change of vehicle, I can’t say how much rate difference there was from Signal Hill…but I know for a fact that it is big difference from 90807.
The only statistic that I can see that would have any validity for jacking rates up based on zip code, is auto theft. Lindsay Lohan’s driving record is for crap, and I’m sure she has a Brentwood zip code.
Where you l live doesn’t matter a tinker’s damn as to how you drive. Should it matter that you happen to live in the same zip code as a notoriously accident-prone freeway intersection? In my opinion, heck no.In conclusion, yes, California should change their method of calculating insurance to something NOT exclusively based on zip code.
3. The quote from the CA insurance commissioner spokesperson indicates that they think the current system is unfair. Do you agree?
Yes.
4. Was it responsible for the newspaper to print this article with an inflammatory headline? (for those not in the Bay Area, this is all over the news here)
Hey, inflammatory headlines are a newspaper’s bread and butter. Nothing new there, move along now...
QUOTE
Daffygrl
Where you l live doesn’t matter a tinker’s damn as to how you drive. Should it matter that you happen to live in the same zip code as a notoriously accident-prone freeway intersection? In my opinion, heck no.
Where you l live doesn’t matter a tinker’s damn as to how you drive. Should it matter that you happen to live in the same zip code as a notoriously accident-prone freeway intersection? In my opinion, heck no.
Well the statistics seem to indicate that it’s not just how you drive but where you drive that has an effect or the likelihood of your having an accident. Obviously where you live has a great impact on the likelihood of you car being stolen.
Many states (I live in MA) go beyond just the area in setting rates. They require the insurance companies to give drivers with no accidents and no tickets better rates. Not sure if CA does this.
QUOTE(Ted @ Jan 5 2006, 08:25 AM)
QUOTE
Daffygrl
Where you l live doesn’t matter a tinker’s damn as to how you drive. Should it matter that you happen to live in the same zip code as a notoriously accident-prone freeway intersection? In my opinion, heck no.
Where you l live doesn’t matter a tinker’s damn as to how you drive. Should it matter that you happen to live in the same zip code as a notoriously accident-prone freeway intersection? In my opinion, heck no.
Well the statistics seem to indicate that it’s not just how you drive but where you drive that has an effect or the likelihood of your having an accident. Obviously where you live has a great impact on the likelihood of you car being stolen.
Many states (I live in MA) go beyond just the area in setting rates. They require the insurance companies to give drivers with no accidents and no tickets better rates. Not sure if CA does this.
California doesn't do this, and therein lies the rub. Zip code is still the major deciding factor. I'm considered a "good driver" but all that gets me is a miniscule discount. Ironically, the proposition to change this was PASSED in 1988, but it has taken this long to force the insurance companies to comply.
But yes, where you drive should also be a deciding factor. If it was, my insurance rates would probably be half what they are now with the zip code system (since I work in an area that has much lower rates than where I live). I also drive less than 12,000 miles a year. Not much fairness in CA's system.
DaffyGrl,
If, as you admit, "where you drive" is a legitimate rating factor, how do you propose that insurance companies incorporate that? Claims information is collected and correlated on the basis of zip codes. Most accidents happen within 5 miles of home, and where you live is indicative of choices that you've made, hence, of your risk profile.
I understand your dislike of the current system, but what practical system would you use to replace it?
If, as you admit, "where you drive" is a legitimate rating factor, how do you propose that insurance companies incorporate that? Claims information is collected and correlated on the basis of zip codes. Most accidents happen within 5 miles of home, and where you live is indicative of choices that you've made, hence, of your risk profile.
I understand your dislike of the current system, but what practical system would you use to replace it?
QUOTE(Bikerdad @ Jan 5 2006, 09:28 AM)
DaffyGrl,
If, as you admit, "where you drive" is a legitimate rating factor, how do you propose that insurance companies incorporate that? Claims information is collected and correlated on the basis of zip codes. Most accidents happen within 5 miles of home, and where you live is indicative of choices that you've made, hence, of your risk profile.
I understand your dislike of the current system, but what practical system would you use to replace it?
If, as you admit, "where you drive" is a legitimate rating factor, how do you propose that insurance companies incorporate that? Claims information is collected and correlated on the basis of zip codes. Most accidents happen within 5 miles of home, and where you live is indicative of choices that you've made, hence, of your risk profile.
I understand your dislike of the current system, but what practical system would you use to replace it?
So, when I drive to Redondo Beach (good area, low rates) to work , but I live in Long Beach (not-so-good area, high rates), according to your statement above I should receive low rates because I'm driving to a "good" area. How does that square with the insurance companies using the zip code where I live as the deciding factor?
Also, if I live in the not-so-good area for 10+ years, have an excellent driving record with no tickets or accidents, should I still be penalized for living in an "undesirable" zip code? Makes absolutely no sense to me whatsoever. I want what was voted on in 1988: calculate rates based on a person's driving habits and history. My driver's license does not identify me as "Zip Code 90805". Tough luck for the statisticians; I'm sure they'll find a way.
QUOTE
Daffygrl
California doesn't do this, and therein lies the rub. Zip code is still the major deciding factor. I'm considered a "good driver" but all that gets me is a miniscule discount. Ironically, the proposition to change this was PASSED in 1988, but it has taken this long to force the insurance companies to comply.
California doesn't do this, and therein lies the rub. Zip code is still the major deciding factor. I'm considered a "good driver" but all that gets me is a miniscule discount. Ironically, the proposition to change this was PASSED in 1988, but it has taken this long to force the insurance companies to comply.
I find it hard to believe that it took that long to “get the insurance companies to comply”. The reality is that it took that long for the legislators in CA to pass the change in the regulation.
As far as where you drive to work some states ask for this info and I believe take this into consideration. But remember if you live in Long Beach you still must drive daily into, out of and through Long Beach to get home and this is one reason you pay more. Most accidents do take place near home.
Are your rates competitively set in CA. Some states allow this (MA does not) and this allows you to pick a company that may give you a better rate because you have a better driving record.
Another practice used in many states (like MA) is rate leveling. In this scheme the rates are lowered in the cities (like Boston) where the majority of the accidents happen and raised in the suburbs. Not sure if CA does this. I hate it because you end up paying $1,000 or more to insure virtually any car no matter how old, in the state.
Another rip off used in MA is that drivers pay a multi year penalty if they get a traffic ticket. Their insurance rates go up – the catch is that the STATE gets the money!
QUOTE(DaffyGrl @ Jan 5 2006, 12:53 PM)
QUOTE(Bikerdad @ Jan 5 2006, 09:28 AM)
DaffyGrl,
If, as you admit, "where you drive" is a legitimate rating factor, how do you propose that insurance companies incorporate that? Claims information is collected and correlated on the basis of zip codes. Most accidents happen within 5 miles of home, and where you live is indicative of choices that you've made, hence, of your risk profile.
I understand your dislike of the current system, but what practical system would you use to replace it?
If, as you admit, "where you drive" is a legitimate rating factor, how do you propose that insurance companies incorporate that? Claims information is collected and correlated on the basis of zip codes. Most accidents happen within 5 miles of home, and where you live is indicative of choices that you've made, hence, of your risk profile.
I understand your dislike of the current system, but what practical system would you use to replace it?
So, when I drive to Redondo Beach (good area, low rates) to work , but I live in Long Beach (not-so-good area, high rates), according to your statement above I should receive low rates because I'm driving to a "good" area. How does that square with the insurance companies using the zip code where I live as the deciding factor?
Because most driving takes place within 5 miles of the home address.
QUOTE
Also, if I live in the not-so-good area for 10+ years, have an excellent driving record with no tickets or accidents, should I still be penalized for living in an "undesirable" zip code?
Tell me, have the rest of the folk living in your "undesirale" zip code gone 10+ years without an accident? I doubt it. Which means that even though you may be a flawless driver, you have chosen to live in an area that is, statistically speaking, inhabited by CrashBarones. Which greatly increases your chances of being in the wrong place, at the wrong time. If, as is quite likely, the area also has a higher ratio of uninsured drivers, thats another bump up in the risk profile. QUOTE
Makes absolutely no sense to me whatsoever. I want what was voted on in 1988: calculate rates based on a person's driving habits and history. My driver's license does not identify me as "Zip Code 90805". Tough luck for the statisticians; I'm sure they'll find a way.
Rates are calculated based on a person's driving habits and history. You have a history of driving repeatedly through a high risk area. You have a history of garaging your car in an area with a higher crime rate. You recieve a discount for having a no accidents. Now, you may disagree with the method that they use for slicing and dicing the geography, but are you willing to pay the additional costs that using some other method of slicing and dicing the geography will incur? And what would that method be?I agree, the statisticians can find a way. How much will it cost?
QUOTE(bikerdad)
Because most driving takes place within 5 miles of the home address.
That makes no sense to me. My vehicle spends 10-11 hours OUTSIDE my zip code 5 days out of 7.
QUOTE
Tell me, have the rest of the folk living in your "undesirale" zip code gone 10+ years without an accident? I doubt it. Which means that even though you may be a flawless driver, you have chosen to live in an area that is, statistically speaking, inhabited by CrashBarones. Which greatly increases your chances of being in the wrong place, at the wrong time. If, as is quite likely, the area also has a higher ratio of uninsured drivers, thats another bump up in the risk profile.
That’s most likely true…which circles you right back around to the headline stating that the method is racist. Most uninsured drivers are poor, most poor are minorities, thus the policy has racist elements.
Some interesting statistics from the NHTSA/DOT:
QUOTE
Fatal crashes decreased slightly (0.6 percent) from 2002 to 2003, and the fatality rate dropped to a new historic low of 1.48 fatalities per 100 million vehicle miles of travel in 2003.
The nonmotorist fatality rate per 100,000 population has declined by 52.4 percent from 1975 to 2003.
The nonmotorist injury rate per 100,000 population has declined by 45.6 percent from 1988 to 2003.
The percent of alcohol-related fatalities has declined from 60 percent in 1982 to 40 percent in 2003. NHTSA
The nonmotorist fatality rate per 100,000 population has declined by 52.4 percent from 1975 to 2003.
The nonmotorist injury rate per 100,000 population has declined by 45.6 percent from 1988 to 2003.
The percent of alcohol-related fatalities has declined from 60 percent in 1982 to 40 percent in 2003. NHTSA
Statistics by CA county: NHTSA/DOT
So, since crashes, injuries, drunk driving are all down, insurance rates should also go down, right?
QUOTE
Rates are calculated based on a person's driving habits and history. You have a history of driving repeatedly through a high risk area. You have a history of garaging your car in an area with a higher crime rate. You recieve a discount for having a no accidents. Now, you may disagree with the method that they use for slicing and dicing the geography, but are you willing to pay the additional costs that using some other method of slicing and dicing the geography will incur? And what would that method be?
I agree, the statisticians can find a way. How much will it cost?
I agree, the statisticians can find a way. How much will it cost?
You don’t state where you live, but I bet it isn’t in CA. The demographics of LA County are rat's nest, with expensive, so-called "desirable" neighborhoods within spitting distance of areas you wouldn't want to be caught in late at night. I don’t live in a ghetto or the barrio, but because of my ZIP CODE, I am considered right along with areas that could be considered such (right back to that whole racist thing). Ironically enough, my immediate neighborhood consists of all working, middle class folks, with a Caucasian majority.
You sincerely don't have any issue with my original scenario where a brand new zip code was priced HIGHER when there were NO statistics to back it up???
Sorry, we'll have to agree to disagree on this one.
QUOTE
That’s most likely true…which circles you right back around to the headline stating that the method is racist. Most uninsured drivers are poor, most poor are minorities, thus the policy has racist elements.
Not so. Racist implies that a decision is based on race alone. Many uninsured drivers are not minorities and I would bet the statisticians have no idea what race the people are in any particular area. And even if they do the rate setting process is not based on race – thus minorities living in areas with low accident and loss stats get the same low rates as non minorities.
QUOTE(DaffyGrl @ Jan 5 2006, 02:48 PM)
QUOTE(bikerdad)
Because most driving takes place within 5 miles of the home address.
That makes no sense to me. My vehicle spends 10-11 hours OUTSIDE my zip code 5 days out of 7.
Which means that it spends the majority of time inside your zipcode.
QUOTE
QUOTE
Tell me, have the rest of the folk living in your "undesirale" zip code gone 10+ years without an accident? I doubt it. Which means that even though you may be a flawless driver, you have chosen to live in an area that is, statistically speaking, inhabited by CrashBarones. Which greatly increases your chances of being in the wrong place, at the wrong time. If, as is quite likely, the area also has a higher ratio of uninsured drivers, thats another bump up in the risk profile.
That’s most likely true…which circles you right back around to the headline stating that the method is racist. Most uninsured drivers are poor, most poor are minorities, thus the policy has racist elements.
Most poor are not minorities. There are more poor whites than any other group. While blacks, Hispanics and American-Indians have higher rates of poverty than whites, Asians and Jews have lower rates. If the contention that the ratings are racist is to hold up, then you would have to compare comparable socio-economic zip codes with completely different racial compositions with different rates, AND no other explanation for the difference. I don't see it happening.
QUOTE
Some interesting statistics from the NHTSA/DOT:
Statistics by CA county: NHTSA/DOT
So, since crashes, injuries, drunk driving are all down, insurance rates should also go down, right?
QUOTE
Fatal crashes decreased slightly (0.6 percent) from 2002 to 2003, and the fatality rate dropped to a new historic low of 1.48 fatalities per 100 million vehicle miles of travel in 2003.
The nonmotorist fatality rate per 100,000 population has declined by 52.4 percent from 1975 to 2003.
The nonmotorist injury rate per 100,000 population has declined by 45.6 percent from 1988 to 2003.
The percent of alcohol-related fatalities has declined from 60 percent in 1982 to 40 percent in 2003. NHTSA
The nonmotorist fatality rate per 100,000 population has declined by 52.4 percent from 1975 to 2003.
The nonmotorist injury rate per 100,000 population has declined by 45.6 percent from 1988 to 2003.
The percent of alcohol-related fatalities has declined from 60 percent in 1982 to 40 percent in 2003. NHTSA
Statistics by CA county: NHTSA/DOT
So, since crashes, injuries, drunk driving are all down, insurance rates should also go down, right?
Actually, it doesn't say that crashes have declined, it says fatal crashes have declined. What you aren't considering is that long-term medical care is more expensive, the lawsuit lottery has driven up costs, vehicle repair is more expensive, and people are driving more. Note that there is a correlation between some of the stats you've provided, and some of the factors I've mentioned. For instance, airbags save lives, but in modest collisions airbags frequently deploy when they aren't actually needed (due to the seatbelt.) Voila, add the cost of replacing the airbag to the repair.
QUOTE
You don’t state where you live, but I bet it isn’t in CA. The demographics of LA County are rat's nest, with expensive, so-called "desirable" neighborhoods within spitting distance of areas you wouldn't want to be caught in late at night. I don’t live in a ghetto or the barrio, but because of my ZIP CODE, I am considered right along with areas that could be considered such (right back to that whole racist thing). Ironically enough, my immediate neighborhood consists of all working, middle class folks, with a Caucasian majority.
No, I don't live in California, although it seems like it sometimes. 
However, I'm quite familiar with its geography. Over the years, I've spent a lot of time in Reseda, Oceanside, Long Beach, Anaheim and the High Desert. My hometown has very similar characteristics, and they do rate by zip code here, with the added bonus of the "newest" car population of any Metropolitan Survey Area. Which means we all pay higher rates because of the swarms of late model cars that are sharing the roads with us, subject to damage. And yet, I choose to live here... Furthermore, if your neighborhood is "within spitting distance" of such bad areas, why shouldn't they consider that in your rating? In your case, you happen to share the zipcode with the bad area.
QUOTE
You sincerely don't have any issue with my original scenario where a brand new zip code was priced HIGHER when there were NO statistics to back it up???
Nope, because I'm fairly confident that there are statistics to back it up. How do you know there weren't? As a "one-off" statistics project, they could easily have analyzed 6 months of claims from the zip code, then reanalyzing them based on the new zip codes. X is the amount of claims from ZipPreSplit, Y came from ZipA, Z came from ZipB. Set rates accordingly.You do know that you could move to a more amenable zip code, right? Or even state?
But you choose not to, because the trade-offs you make to live were you live are worth it to you. Well, this is just another trade off.Grace and peace, BD
This is a simplified version of our main content. To view the full version with more information, formatting and images, please click here.