The Credit Score Tax: In Many States, Poor Credit Costs More Than a DUI

In most of the country, your auto insurance premium is shaped by a score that reflects your financial history, not your driving record. Credit-based insurance scoring is legal in 46 states and Washington, D.C., and the penalty it imposes on drivers with poor credit is far larger than most people realize.

MoneyGeek analyzed auto insurance premium data from Quadrant Information Services across all 50 states and Washington, D.C., comparing rates for drivers with excellent credit against those with poor credit. The national average credit-score penalty is $2,102 per year. In 30 states, the penalty exceeds $2,000. In many of those states, the credit surcharge is larger than the typical surcharge for a DUI conviction, a penalty that requires an actual dangerous act to trigger.

Four states have banned the practice entirely. Six more introduced legislation in 2026 to restrict or eliminate it. And one state’s data, California’s, reveals the clearest evidence that the penalty is a policy choice, not a market necessity.

Picture two Texas drivers. Both have clean records: no accidents, no tickets, no DUIs. They drive the same car, carry the same full-coverage policy and live at the same address. The only difference between them is a three-digit number that has nothing to do with how they drive.

The driver with poor credit pays $4,197 per year. The driver with excellent credit pays $1,485.

The difference is $2,712. That is $226 per month, every month, indefinitely, for a driver who has never filed a claim and never been cited for a traffic violation. According to Experian’s State of the Automotive Finance Market (Q4 2024), the average monthly payment on a new car loan was $742 in the fourth quarter of 2024. The Texas credit penalty is nearly a third of an extra car payment, month after month, for behavior that occurs entirely outside the vehicle.

And Texas isn’t the worst case. In Wyoming, where the penalty is largest in the country, the spread is $2,876 per year, or $240 per month. Wyoming’s excellent-credit premium is $1,516; its poor-credit premium is $4,392. A driver with a perfect record and poor credit pays almost three times what a driver with excellent credit pays for the same policy.

This is what consumer advocates and an increasing number of state legislators call the credit-score tax on auto insurance: an annual surcharge that is invisible at the point of sale, rarely disclosed on renewal notices and largely unknown to the drivers who pay it.

The penalty varies by state, but the pattern is consistent: drivers with poor credit pay more, often much more, for identical coverage. In many states, the surcharge exceeds what a driver would pay after a DUI conviction. A clean driving record doesn’t protect you from a surcharge larger than the one the system imposes for drunk driving.

And one state’s data reveals exactly how large that penalty would be if the law changed.

Credit-based insurance scoring is not the same as the credit score a lender pulls when you apply for a mortgage. Insurers use a proprietary variant that weighs factors such as payment history, outstanding debt, length of credit history and types of credit used. The score is designed to predict the likelihood that a policyholder will file a claim, not the likelihood that they will repay a loan. Credit is one of several factors that affect car insurance cost, but unlike driving record or vehicle type, it has no direct connection to driving behavior.

The actuarial case for the practice is not invented. Studies conducted by the Federal Trade Commission and state insurance regulators have found that credit-based insurance scores do correlate with claims frequency and severity at a population level. Insurers argue, with some supporting evidence, that the score captures financial stress that leads to deferred maintenance, higher-risk driving behavior and a higher probability of filing claims.

The insurance industry’s strongest practical argument comes from a real-world experiment: Washington state.

In 2021, Washington Insurance Commissioner Mike Kreidler issued emergency rules banning credit-based scoring for auto, homeowners and renters insurance. A Thurston County Superior Court overturned the emergency rule later that year. Kreidler adopted a new permanent rule in 2022; a court struck that down too, finding it exceeded the agency’s statutory authority. When the Washington ban was briefly in effect, industry data showed that more than 60% of drivers saw their premiums increase, because risk was redistributed across the rate pool. Drivers who had benefited from excellent-credit discounts absorbed those costs.

The industry cites Washington as proof that eliminating credit scoring simply shifts costs rather than reducing them.

That argument deserves honest engagement. It is true that credit-based scoring allows insurers to price more granularly: good-credit drivers benefit from lower rates that, in the absence of the scoring system, would be spread across the pool. It is also true that four states with full bans enacted those bans before credit scoring became standard industry practice in the 1990s and early 2000s, so they represent a different market equilibrium rather than a clean “before and after” experiment.

But the actuarial defense runs into a different problem: what credit scores actually measure, at scale, in America.

Median credit scores are lower for Black and Hispanic households than for white households, reflecting documented patterns of wealth inequality, discriminatory lending practices and unequal access to credit over generations. Low-income households carry lower scores than high-income households across all racial groups. A pricing factor that correlates with race and income, even if it also correlates with claims, produces outcomes that a growing number of state legislators find troubling. MoneyGeek’s own research on auto insurance pricing in communities of color has documented these disparities.

The question the policy debate has turned on is not whether credit scores predict claims. They do, in aggregate. The question is whether that predictive power justifies the distributional consequences: charging systematically more to people who are disproportionately poor, disproportionately Black and disproportionately Hispanic, for a risk factor they cannot change quickly, that has nothing to do with how they drive.

The affordability burden data makes this concrete. In West Virginia, the credit-score penalty ($2,391) consumes 4.27% of median household income. In Louisiana, the penalty ($2,439) consumes 4.19%. In Wyoming, the surcharge ($2,876) consumes 3.97% of income. These are not abstract percentages. They represent real money extracted, every year, from drivers who are already among the least financially resilient in their states and who have the fewest low-income car insurance options available to offset the cost.

The legislative activity of 2026 suggests that, in a growing number of statehouses, the answer to the distributional question is: no, predictive accuracy alone is not enough to justify this.

Six states introduced or advanced legislation in the current legislative session to restrict or ban credit-based insurance scoring in auto insurance. A federal bill was reintroduced in 2025. Two are the most consequential.

Illinois: The Bill That Creates Regulatory Infrastructure. Senate Bill 1486 passed the Illinois House 66-40 on March 19, 2026, and returned to the Senate for a concurrence vote on House amendments. The bill doesn’t directly ban credit scores but creates a rate-review framework: it requires the Department of Insurance to review and approve rate filings, prohibits rates that are “excessive, inadequate, or unfairly discriminatory” and requires 60 days’ notice before premium increases of 10% or more. Illinois is currently one of only two states (along with Wyoming) that doesn’t require prior approval of insurance rate increases. Illinois drivers with poor credit pay a $2,296 annual penalty, 180% of the excellent-credit rate. The Illinois measure may be the most consequential bill in the current wave. It creates regulatory infrastructure that could effectively challenge credit-based rates as “unfairly discriminatory” without a direct ban, potentially serving as a model for states where outright prohibition meets stronger industry resistance.

New York: The Low-Penalty State That Acted Anyway. Assembly Bill A10524, introduced March 6, 2026, by Assemblymember Jen Lunsford, would require the Department of Financial Services to prohibit the use of credit scores, small-area ZIP codes and income as auto insurance rating factors. New York currently has the lowest credit-score penalty among non-ban states in this dataset: $1,442 per year (162%). The bill is pending in the Assembly Insurance Committee. New York’s case is notable for advocates: even the smallest non-ban penalty in the country, translated to a political bill, has enough support to advance.

Four additional states and the federal government have active measures:

West Virginia carries both active legislation and the highest affordability burden in the dataset: the $2,391 annual credit penalty consumes 4.27% of median household income in a state where median income is $55,948, the lowest in this study. The federal PAID Act, sponsored by Representatives Rashida Tlaib of Michigan, Bonnie Watson Coleman of New Jersey and Mark Takano of California, would impose civil penalties of at least $2,500 per violation through FTC enforcement. The current Republican-controlled Congress is unlikely to advance it, but its reintroduction signals sustained federal-level pressure.

The tables below covers all 50 states and Washington, D.C. Section 1 lists the 46 non-ban states and D.C., ranked by dollar difference from highest to lowest. Section 2 lists the four ban states with modeled rates that consumers in those states don’t pay.

The pattern visible in this data is not random. The states with the highest home and auto insurance affordability burdens, West Virginia (4.27%), Louisiana (4.19%), Wyoming (3.97%), Texas (3.58%), Montana (3.39%), are states where insurance costs are already high relative to income, and where a $2,000 to $3,000 annual credit penalty falls hardest on households already stretched thin.

California’s $3,553 modeled penalty is the organizing data point of this study. It is simultaneously the largest surcharge in the dataset and the surcharge that no California consumer pays. The only mechanism producing that outcome is a 38-year-old state law. Forty-six other states don’t have that law. Their residents pay the penalty, or some version of it, every year.

The credit-scoring industry and insurance carriers respond that the alternative, eliminating risk segmentation by credit, means higher premiums for the 65% to 70% of Americans with good or excellent credit, who currently benefit from the system. This is a real cost, not a hypothetical one. Washington’s experience showed it in practice. The question is whether that cost-shifting is a reason to maintain the status quo, or an obstacle to be managed in the design of a fairer transition.

The legislative trajectory of 2026 suggests the balance is shifting. Six states introduced bills this year; one passed a chamber. A federal bill is circulating for the fourth time. The four ban states have demonstrated, over more than 30 years in some cases, that insurance markets function without credit-based scoring. California’s functioning market, and its modeled penalty, is the clearest evidence available that the surcharge is a choice, not a market necessity.

What the data cannot resolve, but what the legislative activity reflects, is a growing consensus that a pricing factor should be evaluated not only on its predictive accuracy but on its distributional consequences. Credit scores predict claims. They also, predictably, charge the most to the people who can afford it the least.

MoneyGeek analyzed the auto insurance premium difference between drivers with excellent credit and poor credit across all 50 states and Washington, D.C. Premium data was sourced from Quadrant Information Services for MoneyGeek’s standard base driver profile, varying only by credit tier. The dollar difference was calculated as (poor credit annual premium minus excellent credit annual premium).