What you pay for car insurance depends on more than what you drive and how you drive it.
It depends on where you live, your credit score, and sometimes your occupation and education level.
High school dropouts may pay more than people with master’s degrees — even when the well-educated have worse driving records. Janitors may pay more than business executives. People who rent may pay more than home owners. People who live in poor, crime-ridden parts of town will pay much more than others.
Of course, your driving record affects this too, as does the car you drive.
Insurers say it’s all an effort to match rates to the risk presented by a driver. Critics, however, say insurers are discriminating against the poor.
The rate differences can be dramatic.
If you live in ZIP code 63147 — which stretches along north Broadway into the city’s Baden neighborhood — count on paying the highest auto insurance rates in the St. Louis area.
Residents of other poor city ZIPs face also face high rates, according to a rate analysis by Carinsurance.com, an online auto insurance market.
A good driver might pay $1,974 to cover a 2012 Honda Accord in the 63147 ZIP, according to the survey, which averaged the rates of six big insurance companies.
Move the same car and driver to the southern or western suburbs — such as Crestwood or Chesterfield — and its owner would save about $700. Move it to the far suburbs of Metro East — such as O’Fallon or Columbia — and the savings are about $1,000.
Insurance companies say claims statistics explain much of the difference. Payouts are higher in areas plagued by theft and vandalism. Areas of dense population also produce more accidents, they say.
All the other factors used in rating — be it late payments on credit cards or the absence of a high school diploma — can be directly tied to a person’s chances of filing an auto claim, says Steven Weisbart, chief economist at the Insurance Information Institute, an industry organization.
The object of rating is to group people of similar characteristics, then charge them by the claims record of that group. “To most people, that seems fair,” says Weisbart.
BETTER DRIVERS?
Consumer advocates say something sneakier is going on. They say insurers are using factors designed to screen people by wealth, as well as risk, figuring that they can sell more financial products to people with money.
Critics complain less about ZIP code ratings than they do about the use of factors such as education, job titles, marital status, credit scores and home ownership.
Companies often charge a poor but safe driver more than a well-off but dangerous one, according to the Consumer Federation of America.
The group last month released a survey of auto rates in 12 cities, including St. Louis. It compared rates offered two 30-year-old women living on the same street in a middle-income ZIP code looking for equal auto coverage.
One was single, a high school grad who rents her home and work as a receptionist. She had a clean driving record but had been without insurance for 45 days.
The other driver was a married executive with a master’s degree who owns a home. She was at fault for an accident that caused $800 in damage.
“In two-thirds of the 60 cases studied, large auto insurers quoted higher premiums to safe drivers than to those responsible for an accident,” the federation reported. In most cases, the difference was more than 25 percent.
“State insurance regulators should require auto insurers to explain why they believe factors such as education and income are better predictors of losses than at-fault accidents,” said Robert Hunter, the federation’s insurance director, and a former Texas state insurance commissioner.
In St. Louis, Progressive offered a higher rate to the receptionist. Farmers Insurance didn’t offer coverage to the receptionist at all, although it would cover the executive. Allstate and State Farm offered better rates to the accident-free receptionist, according to the survey.
State Farm, which has nearly a quarter of Missouri’s car insurance market, charged the good-driver receptionist less in all 12 cities.
Weisbart says the study is flawed, in part because the receptionist had a break in insurance coverage. That break can justify higher rates even though it has nothing to do with accidents.
It costs a company money to issue a policy, he notes, and someone who let their last policy lapse may not keep up the payments on a new one. Some people sign up for insurance in order to register their cars, planning on dropping the policy the next month.
But other comparisons also show higher rates for people with less education and lower-paying jobs.
OR BETTER CUSTOMERS?
Doug Heller, former director of Consumer Watchdog, gives a presentation featuring the St. Louis insurance rates offered by Geico. It’s designed to show how things with little obvious connection to driving can affect rates.
He used Geico’s rate-quote website to get rates for University City drivers identical except in education and occupation.
A driver with a master’s degree and an executive job was quoted $481. With only a high school diploma, the executive’s premium went to $540. Make the high school grad into a waiter and the premium went to $694.
Using Geico’s website, I compared quotes for two fictional 35-year-old bachelors with flawless driving records living around the corner from each other in a middle-class part of Richmond Heights.
The manager with a master’s degree was quoted $343 for a six-month premium. The high school dropout working as janitor would pay $436 for the same coverage.
Geico did not respond to a request for comment.
Insurance rating systems differ widely by company, and consumers can profit by shopping around. In the 63147 ZIP on the north side, rates for the Accord varied from $1,227 to $3,562 in the Carinsurance.com study. (The excessive rate may indicate an insurer that doesn’t want to do business in that area.)
Can things like income, occupation and payment record really predict accidents? “I don’t believe so, and I’m an actuary,” said Hunter.
Weisbart differs. He points to a 2007 study by the Federal Trade Commission of credit scores in insurance rating.
“Credit-based insurance scores are effective predictors of risk under automobile policies,” the FTC concluded. “They are predictive of the number of claims consumers file and the total cost of those claims.”
Weisbart feels the same can be said of education, job title and the other factors, though he didn’t offer any studies that proved his point.
Hunter says it’s not enough to show that people with poor credit or low education have more accidents, he says. Actuarial science demands a reason to believe that one causes the other, i.e. correlation is not the same as causation.
“If I were to ask about hair color, they would provide me with undeniable proof that blondes cause more accidents, or maybe they cause less. That doesn’t mean it’s meaningful,” says Hunter.
In fact, the FTC said it couldn’t tell why people with low credit scores had higher insurance claims.
Critics think insurers should drop such factors and give more weight to a person’s actual driving record.
However, Weisbart thinks using multiple factors can better assess a driver’s risk. “They wouldn’t be doing this if they didn’t have the data,” he says.
Insurers may have another reason. Heller, of Consumer Watchdog, believes companies use low rates to lure in well-off customers, then shift the cost of claims to lower-income customers.
“The rich executive who is married probably has a more expensive car, and a second car and a home he wants to insure and money that he could put in your affiliated bank,” he says.
Insurance departments in Missouri and Illinois largely stay out of this debate. Insurers don’t need state approval for rates in the two states. The states’ basic requirement is that rates not be discriminatory. Missouri law says rates can’t be excessive, and that credit scores can’t be the only factor in setting rates.
