Rivera’s job title at the Latino Action Network Foundation, an advocacy organization in New Jersey, is “programs coordinator,” so she figured she wasn’t getting dinged for her career. And her credit score—another factor that car insurers often consider—is in the high 700s.

Then it struck her: She hadn’t gone to college. Rivera didn’t know whether Liberty was charging her more as a result, but it seemed smart to compare prices. And she soon found a car insurance policy with somewhat less coverage that would save her almost $1,800 per year. It was offered by Citizens United Reciprocal Exchange (CURE), a regional, not-for-profit insurance company that doesn’t factor education or occupation into its premiums.

A Liberty Mutual spokesperson declined to comment on Rivera’s experience. But a new Consumer Reports investigation found that Liberty Mutual is one of several large auto insurers that ask prospective customers about their education level or job title—and quote some drivers different premiums based on the results.

It had never occurred to Rivera that her old car insurer might use education in its pricing formulas, a practice she considers deeply unfair. “I was completely blown away,” she says. “I just don’t understand how they get away with it.”

Pricing Based on Your Bio

What you pay for auto insurance has a lot to do with your behavior on the road, including past collisions and speeding tickets, along with factors such as what car you own and how many miles you drive each year.

But when they set rates, insurers can also lean on biographical details that you don’t control, such as your age or gender. They may also look at financial factors that seem to have little to do with how safely you drive, including your credit history and whether you rent or own your home. In 2015, Consumer Reports found that socioeconomic factors sometimes outweigh driving records and habits in the premiums that insurers set.

Some insurers also consider people’s education and occupation when they decide what to charge, arguing that there’s a correlation between those factors and the financial risk of insuring a customer.

Even if that’s true on average, it’s unfair to individuals, says Chuck Bell, a policy advocate and insurance expert at Consumer Reports. “No one should have to pay a penny more for auto insurance just because they haven’t graduated from college or have a working-class job,” he says.

What’s more, this controversial practice disproportionately affects Black and Latino drivers, who on average are less likely to have college degrees and white-collar jobs. Insurers aren’t allowed to charge people differently for car insurance because of their race, but pricing based on education and job title can have a similar effect, Bell says. “These factors unfairly penalize drivers of color and working-class people.”

To understand how insurers are using education and occupation to set premiums, Consumer Reports requested almost 900 online car insurance quotes from nine insurers. We used 21 ZIP codes that covered Washington, D.C., and six states, including New Jersey and Oregon, where lawmakers are debating bills to ban the practice.

To gather the quote requests, we created a hypothetical 30-year-old woman named Sara Lynn shopping for her state’s minimum required coverage. She owns a 2016 Toyota Camry LE and has a clean driving record. The only details we varied among requests were her education level and occupation.

Consumer Reports found evidence that three major insurance companies—Geico, Liberty Mutual, and Progressive—use education when calculating car insurance quotes. All three quoted higher premiums, on average, to consumers who had completed less education. Geico and Progressive also factor in occupation, the study indicated, quoting higher prices to consumers with service jobs than to managers and executives.

Consumer Reports found disparities in pricing based on education. The chart above shows that the average Geico quote for someone with an advanced degree was $700, while the average quote for an otherwise identical person without a high school diploma was $815.

For instance, in Brooklyn Center, Minn.—a suburb of Minneapolis—the version of Sara Lynn who is a cashier without a high school diploma got a quote from Progressive for an annual premium $150 higher than the rate quoted to the Sara Lynn who was an executive with a graduate degree.

And in Hoboken, N.J., our hypothetical cashier who didn’t finish high school got a quote from Geico for an annual premium that was $455 above the price quoted to an otherwise identical woman with an executive job title and a graduate degree.

Our findings square with earlier research, such as a 2014 study from the New York Public Interest Research Group that found that Geico and Progressive were among the insurers charging more to nonprofessional workers and people with less education in New York state.

Consumer Reports found disparities in pricing based on job title as well. The chart shows that the average Geico quote for an executive was $718, while the average quote for an otherwise identical cashier was $814. (Numbers don’t add up to $815 because of rounding.)

Geico, Liberty Mutual, and Progressive are among the largest auto insurers in the country. The other six insurers we tested didn’t appear to consider education or occupation when they offered quote prices, at least in the 21 ZIP codes we studied. That includes some big players, like Allstate and State Farm, along with regional companies such as NJM Insurance and Plymouth Rock Assurance.

This means that people can find plenty of quotes that don’t take education or occupation into account. But insurers are very secretive about their rating formulas, which makes it difficult for the average insurance shopper to learn which companies ignore those factors. That means it’s hard to know where to go to look for the best rate.

‘Multiple Gut Punches’

Insurance companies are generally barred from considering a customer’s race or income, along with religion and some other factors, when setting insurance rates. But consumer advocates argue that placing people in broad categories based on education and job title effectively does the same thing, disproportionately hurting non-white customers and anyone with a low income.

In 2019, 40 percent of white people 25 and older had a bachelor’s degree or more education, compared with 26 percent of Black people and 19 percent of Hispanic people, according to the U.S. Census. Meanwhile, 41 percent of working white people had management or professional jobs in 2019—the highest-paying category of jobs, according to the Bureau of Labor Statistics—compared with just 32 percent of Black people and 23 percent of Hispanic or Latino people.

People who complete less education are more likely to then work in service-industry or blue-collar jobs, so pricing that considers both education and career can hit the same person twice for essentially the same circumstances. Then, if they earn less money and have a harder time paying their bills, they may have lower credit scores—another factor that can make for higher premiums. Add it all together, and a person may get penalized over and over again for basically the same reason.

“The way some of these companies price insurance, if you’re working-class and don’t have a college degree, it’s not just that you get one gut punch,” says Doug Heller, an insurance expert at the Consumer Federation of America. “You get multiple gut punches because each of these factors is another penalty on your premium.”

When premiums are unaffordable, low-income families who have to drive to get to work or school must decide whether to cut back on some other critical bill or drive uninsured, which can lead to enormous fines—or catastrophic lawsuits in the event of a crash.

These pricing practices feel wrong to most people. A 2016 survey commissioned by the CFA found that 69 percent of Americans said it’s unfair to use education to price car insurance policies, and 63 percent said it’s unfair to use occupation.

However, insurance companies and their lobbyists argue that there’s a perfectly valid reason to use these factors to set rates: They accurately predict how financially risky a customer will be to insure.

“Actuarial science justifies the use of education and occupation,” says Alex Hageli of the American Property Casualty Insurance Association (APCIA), which lobbies on behalf of the industry. The better an auto insurer can predict how much financial risk a particular driver brings, the more accurately it can charge that driver, he says. Otherwise, everyone else would have to shoulder some of the burden of insuring that riskier driver.

The industry points to reports from a handful of state regulators that analyzed industry-provided data and concluded that education and occupation can be good predictors of financial losses to the companies.

“Insurance companies work very hard to ensure every customer pays the right rate,” says Michael Barry, head of public affairs at the Insurance Information Institute, an industry-funded research center. “Insurers do an excellent job, generally, of charging low-risk customers low rates and high-risk customers high rates.”

Deciding What’s Fair

Insurance companies don’t give independent researchers access to their data, so there’s no way to verify the strength of the correlations among education, job title, and the financial rewards and risks for insurers. Assuming those relationships are real, it’s not clear what causes them.

But according to consumer advocates, those correlations don’t matter.

The tug-of-war over education and occupation—as well as other contentious rating factors, including credit scores and gender—essentially comes down to two competing definitions of fairness.

To an insurer, a fair rate is one that closely matches a person’s premiums to the company’s financial risk, no matter how much control the consumer has over the factors that go into the pricing formula. That way, every driver pays their fair share, they argue.

But to consumer advocates—and a growing cadre of lawmakers around the country—fairness means something broader. They argue that because Black and Latino people have diminished access to higher education and employment opportunities, using those factors to price their car insurance policies just adds another burden to a skewed economy that already makes things harder for members of those groups. The racial and ethnic disparity doesn’t have to be intentional to be real: It happens automatically because of disparities in schools and economic opportunities.

“This is piling on a type of discrimination on top of other types of discrimination that have together held families back from being economically successful,” says Dena Mottola Jaborska, associate director of New Jersey Citizen Action, an advocacy group lobbying for a bill that would ban the use of education, occupation, and credit score for setting premiums in the state.

Advocates say that governments have a special responsibility to set up guardrails to ensure fairness on this issue because the vast majority of drivers in the U.S. are required by law to buy car insurance.

“It’s entirely appropriate for regulators to say, ‘Even if these factors are predictive, we’re still not going to let you use them. You have lots of other data points you can use,’” says Amy Bach, executive director of United Policyholders, an insurance-focused advocacy group. “‘Use the legitimate ones—the ones that are not controversial—and set a good example.’”

Some insurers already do just that. CURE, the not-for-profit insurance company that now counts Rivera as a customer, decided not to include those factors in its own pricing because of the harm it would do, says chief operating officer Eric Poe, despite a business case for using them. “When you employ these income proxies, you dramatically increase the number of people who can’t afford car insurance,” Poe says. “Can you think of two criteria that are more correlated to income?”

Insurers and trade groups we asked say they haven’t studied how using these factors affects drivers of color and low-income people. “We don’t collect the data that would be necessary to answer that question,” says David Snyder, APCIA’s vice president for international policy, when asked whether the two factors can lead to discrimination. Both Geico and Liberty Mutual, two of the companies that quoted different prices by education level or occupation, said their rates are fair and competitive. Geico also said the company would not have achieved its towering market position if the rating factors it uses were discriminatory.

Progressive, the third company that appeared to use education or occupation to price its policies, did not respond to several requests for comment.

And Farmers—which asked drivers about their job but did not appear to use that information to set prices in the ZIP codes we analyzed—said the ability to use rating factors that help predict risk is important for setting accurate prices.

Pushing for Reform

A growing number of measures around the country are restricting the kinds of personal information insurers can use when they set rates.

A New Jersey bill being considered right now would stop insurers in the state from factoring in education and job title when setting rates, as would a wider-ranging proposal in Oregon. Starting in 2020, Michigan has barred insurers from using occupation, education, and certain other biographical data to set prices. California and Massachusetts already prevent insurers from considering education and occupation.

In 2017, the New York Department of Financial Services adopted a rule requiring insurers to prove that using occupation and education is not unfairly discriminatory, or else to drop the factors altogether. Allstate, Geico, Liberty Mutual, and Progressive all agreed to stop using both factors in New York.

And more changes could be on the way. In Congress, a proposal from Sen. Cory Booker, D–N.J., and Reps. Bonnie Watson Coleman, D-N.J., and Rashida Tlaib, D-Mich., would extend the bans across the country—an unusual step since the industry has long been regulated at the state level.

“Proxies like where you work or whether you have a college degree don’t weed out bad drivers; they just create a two-tier system,” Watson Coleman tells CR. “That second tier, where payments are higher for those who can afford them the least, is functionally racist—it’s the kind of systemic racism that we talk about being pervasive and recalcitrant. But it’s not just a question of racial bias. These practices are the same burden for any working American family.”

A number of the current proposals take aim at additional rating factors alongside education and occupation. Chief among the targets is credit score, a factor that currently weighs heavily in consumers’ premiums. CR has found that a low credit score can add anywhere between $500 and $2,000-plus to annual premiums compared with what drivers who have excellent credit pay.

In 2020, the National Association of Insurance Commissioners, a guiding body for state regulators, began studying the potentially discriminatory effects of certain insurance underwriting and rate-setting practices. The organization plans to release a report in the next couple of months, a spokesperson says.

Snyder, the APCIA vice president, says that taking rating factors off the table can create “unfair subsidies” for risky drivers. But consumer groups argue that enacting them would benefit not just drivers but also everyone else on the road.

Insurers should rely on information about how a person drives, says Chuck Bell, the insurance expert at Consumer Reports, which supports the New Jersey bill and other reform efforts. The more your driving record affects your rates, the greater your incentive to drive safely. That’s good for everyone.

People have much less control over biographical details. “It’s not practical to tell people, ‘Why don’t you get a doctorate or become an executive at a big company?’” Bell says. “We think it’s much better to give consumers feedback based on their ability to operate a car. That would make auto insurance more equitable for millions of people.”