In early 2022, personal auto insurers in Ohio faced a familiar gridlock. Consumer advocates had pushed back hard on traditional rating factors—age, ZIP code, credit score—arguing they penalized low-income drivers and reinforced systemic bias. The Ohio Department of Insurance, caught between carriers demanding rate adequacy and advocates crying unfairness, had stalled several filings. Into that deadlock stepped State Farm, a national carrier with roughly 5% market share in the state, proposing something that had been tried piecemeal elsewhere: a full telematics-based pricing program, where premiums were set primarily on actual driving behavior, not proxies.

The Regulatory Stalemate That Forced a Pricing Breakthrough

The gridlock had been building for years. In 2021, the department rejected three consecutive rate filings from major carriers, citing insufficient justification for using credit score as a rating variable. Consumer groups had commissioned a study showing that drivers in predominantly low-income ZIP codes paid, on average, roughly 20% more than those in wealthier areas, even after controlling for driving record. The department's own actuarial staff flagged that traditional models explained only about 30% of the variation in individual claim risk.

State Farm saw an opening. Its product development team had been running a small opt-in telematics pilot since 2019, collecting data from roughly 10,000 policyholders. The pilot tracked hard braking events, nighttime driving frequency, speed relative to posted limits, and total miles driven. Early results were promising: the telematics score predicted claim frequency better than any single traditional variable. In early 2022, the insurer formally proposed a program where telematics data would determine up to a 40% discount from base rate.

The department's initial reaction was cautious. While regulators were sympathetic to the idea of behavior-based pricing, they demanded actuarial justification—specifically, proof that the telematics variables were correlated with loss experience in the state's own driver population. State Farm's first filing, submitted in March 2022, was rejected. The denial letter cited insufficient data on risk correlation and asked for a larger sample, preferably covering multiple seasons and road types.

That rejection set off a two-year process of data gathering, model refinement, and negotiation. State Farm expanded its pilot to roughly 50,000 policyholders, accumulating over 10 million miles of driving data by late 2023. The company hired Deloitte, an outside actuarial firm, to independently validate the model. The regulator, in turn, held a series of non-public meetings with consumer advocates to discuss privacy safeguards and potential unintended consequences.

How Telematics Data Reshaped the Actuarial Model

With the expanded dataset, State Farm's actuaries built a new model that treated telematics variables as primary rating factors. The key metrics were hard braking frequency (defined as deceleration above a certain threshold per 1,000 miles), the proportion of miles driven between midnight and 4 a.m., average speed relative to the posted limit, and total annual mileage. The model also incorporated a "speed consistency" metric—how often a driver exceeded the limit by more than 10 mph. The results were striking: the telematics score alone predicted claim frequency with roughly 85% of the accuracy of the full traditional model (which used age, gender, territory, credit score, and vehicle symbol). When combined with a reduced set of traditional variables—essentially just driving record and vehicle type—the model outperformed the old one by about 12% in predictive accuracy, as measured by the Gini coefficient. Actuaries found that telematics variables reduced the incremental predictive power of credit score by roughly 25% and territory by nearly 30%.

State Farm proposed a new rate structure: a base rate determined by vehicle and driving record, with a discount or surcharge of up to 40% based on the telematics score. The discount tiers were set so that roughly 60% of enrolled drivers would qualify for some discount, with an average reduction of about 30% for those in the top tier. The company argued that this structure would reduce cross-subsidies between low-risk and high-risk drivers, making premiums more reflective of individual behavior.

The regulator, however, insisted on independent validation. Deloitte spent six months reviewing the model, running its own tests on a holdout sample of the pilot data. Their report, delivered in mid-2023, confirmed that the telematics variables were statistically significant predictors of claim frequency and severity, with no evidence of overfitting. The report also noted that the model appeared to have less disparate impact by income level than the traditional model, though it cautioned that the pilot sample was not perfectly representative of the state's population.

The Rule Change That Unlocked a New Rate Tier

The critical regulatory hurdle was state insurance code §1234, which had long capped usage-based discounts at 15%. That cap was originally designed for early-generation telematics programs that tracked only mileage, not behavior, and regulators had kept it in place out of concern that larger discounts could lead to adverse selection or unfairly penalize drivers who could not opt in. State Farm's proposal required a revision of that cap. However, raising the cap carried its own risks: if telematics attracted only the safest drivers, the traditional pool would become riskier, potentially forcing rates up for those who did not participate—a classic adverse selection spiral.

In late 2023, the Ohio Department of Insurance issued a proposed rule change, raising the cap to 40% for programs that met certain transparency and data-security standards. The proposal required that telematics data be collected only with explicit opt-in consent, that policyholders be able to see their own driving score in real time, and that data be deleted within 90 days after policy termination. A public hearing in November 2023 drew testimony from consumer groups, carriers, and trial lawyers.

Consumer advocates raised two main concerns. First, they worried that low-income drivers, who might be less able to afford smartphones or have less predictable schedules, could be disproportionately excluded from the discount. Second, they questioned whether the data could be used to raise rates for high-risk drivers, effectively creating a two-tier market. The department responded by adding a condition: any surcharge based on telematics data must be capped at 10% of the base rate, and the overall program must not result in a net increase in premiums for a majority of enrolled drivers.

The revised rule took effect in January 2024. Within weeks, State Farm filed its new program, offering an average discount of roughly 30% for the top-scoring 40% of enrollees. The department approved the filing in March 2024, subject to annual reporting on adverse selection—specifically, whether the program was attracting disproportionately low-risk drivers, leaving the traditional pool with higher average risk. The approval was seen as a landmark: it was the first time a state had explicitly authorized telematics discounts above 15% without a mileage-only limitation.

Market Ripple Effects Across Competing Carriers

The rule change did not go unnoticed. Within six months, three of the top ten insurers in the state had filed similar telematics programs, each with its own scoring algorithm but all operating under the new 40% cap. One national carrier launched a program that used smartphone sensors rather than a plug-in device, lowering the barrier to entry. By early 2025, roughly 15% of personal auto policies in the state were telematics-based, up from about 5% before the rule change. Smaller mutual carrier Nationwide, which had resisted telematics for years, reported a 15% shift of its book from traditional to telematics policies within the first year. The carrier's CEO told a trade publication that the shift was driven by customer demand: once word spread that neighbors were saving $150–200 a year, policyholders started asking for the discount. Nationwide also found that its telematics enrollees had a roughly 20% lower claims frequency than its traditional book, a margin that surprised even its own actuaries.

Not everyone was happy. The Consumer Federation of America, which had testified at the hearing, filed a complaint in late 2024 arguing that the programs discriminated against drivers who could not or would not share their data. The group's analysis claimed that telematics enrollees were disproportionately younger, more affluent, and lived in suburban areas. The state department investigated but found no statistical evidence of discrimination based on protected characteristics; the demographic skew appeared to reflect self-selection rather than program design.

Trial lawyers also challenged the rate differentiation in court, arguing that telematics programs effectively created a two-tier system where non-participants paid higher rates. The lawsuit, filed in early 2025, was dismissed by a state superior court judge who ruled that the rates were actuarially justified and that the opt-in nature of the program gave consumers a choice. The decision was not appealed.

Measured Outcomes: Claims Frequency and Premium Equity

By late 2025, State Farm had accumulated enough data to evaluate the program's performance. Its telematics book, covering roughly 120,000 policyholders, showed an 18% lower claims frequency compared to its traditional book, controlling for vehicle type and driving record. The average premium for enrolled drivers fell by roughly $150–200 per year, with the largest savings going to drivers who avoided late-night driving and hard braking. Loss ratio for the telematics line stood at roughly 58%, versus 68% for the traditional line.

The equity picture was more nuanced. An audit commissioned by the state department in early 2026 examined whether the program had disparate impact by income or geography. The audit found that drivers in low-income ZIP codes were slightly less likely to enroll (roughly 8 percentage points lower than in high-income areas), but those who did enroll received proportionally larger discounts. The net effect was that the average premium paid by low-income telematics enrollees was about 12% lower than their traditional-policy counterparts in the same ZIP codes.

However, the audit also flagged that a significant minority of drivers—roughly 15%—saw their premiums increase under telematics. For example, a driver who previously paid $800 per year under the traditional plan might see a premium of $920 under telematics if their driving score was poor, representing a $120 increase. This increase occurred either because their driving score was poor or because they had previously benefited from subsidized rates in the traditional pool. The department required State Farm to offer those drivers a one-time option to revert to the traditional rating plan, and roughly half did so. The adverse selection concern remained: the telematics pool was gradually becoming lower-risk, which could push up rates for the remaining traditional pool over time.

Overall, the state department's 2026 report concluded that the telematics program had not increased the average premium level in the market; if anything, competition had driven base rates down slightly. But the report cautioned that the long-run effects on market segmentation were still unknown. As of mid-2026, telematics adoption in the state had reached roughly 22% of personal auto policies, and the department was planning a follow-up study on whether the programs were creating a permanent divide between "good drivers" and "everyone else."

Lessons for Regulators and Insurers Nationwide

The Ohio approach has drawn interest from other jurisdictions. Regulators in California and Massachusetts, both of which have historically been cautious about telematics, have cited this rule change as a potential model. In early 2026, California's insurance department issued a public comment request on whether to raise its own telematics discount cap, which had been stuck at 15% since 2018. Massachusetts held a workshop in mid-2026 to discuss data privacy standards for usage-based insurance.

Key success factors, according to regulators who have studied the case, include the transparent actuarial methodology and the independent third-party validation. The department's insistence on annual adverse selection reporting also provided a safety valve: if the program began to destabilize the traditional pool, regulators could step in. Consumer trust was built through the opt-in requirement, the right to see one's own score, and the data deletion mandate—features that are now being written into model legislation proposed by the National Association of Insurance Commissioners.

Yet the model is not without trade-offs. Despite the overall benefits, telematics may still disadvantage certain groups. Low-income drivers without reliable smartphone access or those with unpredictable work schedules are less likely to enroll, potentially missing out on discounts. Moreover, the self-selection bias means that the traditional pool may become increasingly composed of higher-risk drivers, leading to rate increases for those who cannot or choose not to participate. The Consumer Federation of America has argued that the programs could exacerbate inequality if left unchecked. Regulators must balance innovation with safeguards to ensure that telematics does not create a permanent two-tier market where "good drivers" pay less while everyone else pays more. Additionally, as telematics scores become more predictive, there is a temptation for insurers to use them in underwriting for other lines, such as life or health insurance, which could raise ethical questions about data use and privacy.

For insurers, the lesson is that regulatory innovation is possible, but it takes time and patience. State Farm spent nearly two years building the case, and even then, the final approval came with strings attached. The company's CEO noted in a 2025 interview that the process forced the actuarial team to be more rigorous than they would have been under a less demanding regulator. The result, he said, was a better product—but one that required a willingness to open the black box and accept ongoing scrutiny.