Price optimization and loyalty pricing in car insurance

Price optimization is the practice of adjusting a car insurance price using a prediction of how much you will tolerate, such as how likely you are to shop or leave, instead of how likely you are to cost the insurer money. Regulators in many states said in 2014 and 2015 that this kind of adjustment is unfairly discriminatory, and the NAIC's 2015 white paper listed propensity to shop among practices inconsistent with that standard. A renewal price that exceeds a new customer's price is not proof of the practice, because several legitimate things can cause it, but you can test your own case in an afternoon.

Pricing for risk versus pricing for willingness to pay

A filed insurance rate starts with a cost estimate: expected claims for a group of similar drivers, plus expenses and a margin. Your price is that rate multiplied by rating factors such as age, location, vehicle and driving record. Anyone with the same factors, in theory, pays the same.

Price optimization adds a second model on top. Instead of asking "what will this driver cost us?", it asks "what is the highest price this customer will accept before leaving?" The NAIC's Casualty Actuarial and Statistical Task Force noted in its November 2015 white paper that there is no single accepted definition, and it quoted several. The Casualty Actuarial Society described it as supplementing loss-cost models with quantitative customer demand models. The American Academy of Actuaries described a technique that goes from cost-based rates to final prices by integrating expected costs with expected consumer demand behavior, to meet objectives that might include profit, volume or retention.

The distinction matters because the two questions have different answers for the same person. Two drivers with identical risk can have different willingness to pay. A cost-based price treats them the same. A demand-based price charges the one who is less likely to leave more. Insurers have always used some judgment between the indicated rate and the final one, and the white paper says so, noting that insurers have long weighed how close they can get to the indicated premium without hurting retention. What changed was that data-driven models made it individual, systematic and hard to see.

What the NAIC concluded in 2015

The task force adopted its white paper on November 19, 2015, and the NAIC's Property and Casualty Insurance (C) Committee adopted it on November 21, 2015. The paper says its authors were not expressing an opinion on state policy decisions, but its recommendations were direct. Three are worth knowing:

• Same risk, same premium. The task force recommended that two customers with the same risk profile be charged the same premium for the same coverage, allowing temporary differences between new and renewing customers that come from capping or transition rules.
• Named practices. It listed four rating practices that, at a minimum, it considered inconsistent with laws against unfair discrimination: price elasticity of demand, propensity to shop for insurance, retention adjustment at an individual level, and a policyholder's propensity to ask questions or file complaints.
• Disclosure. It suggested that states consider requiring insurers to disclose whether price optimization is used, differences between proposed prices for existing and new customers, and which rating factors it affects.

It also reported something that explains why this was hard to police: state regulators said they received few rate filings that identified price optimization, either because insurers did not disclose it or because it operated outside the filed rating plan. A 2013 survey of 73 large insurers by the software vendor Earnix, cited in the paper, found that 55% considered customer price elasticity, and that 45% of the largest (more than $1 billion in written premium) already used some form of price optimization. Those figures are 13 years old and come from a vendor survey, so they show that the practice was not marginal in 2013, not what it looks like today.

This article found no newer NAIC white paper replacing this one on the topic. If you need the current position of your state, ask its regulator; see state department of insurance.

The states that moved first

The white paper's Appendix A lists state actions through late 2015. The actions were bulletins, notices and memoranda to insurers rather than new statutes, and most of them rest on existing laws barring "unfairly discriminatory" rates. A summary:

The paper lists nine more jurisdictions that issued bulletins or communicated policies between July and November 2015: Virginia, Indiana, Pennsylvania, Maine, the District of Columbia, Montana, Rhode Island, Delaware and Minnesota.

Two cautions. First, a bulletin is an interpretation of existing law, not a new statute, and it covers rates filed after its date. Second, the paper also says some regulators believed existing law was enough and issued nothing, and many states had received no filing that mentioned the practice. That does not mean the practice is legal there. It means the answer depends on how the state's "not unfairly discriminatory" standard is read. This article describes the position as of the 2015 paper; each bulletin's current status in 2026 was not confirmed, and your state regulator is the authority.

Our companion article on price optimization: why regulators barred charging what you'll pay covers the regulatory reasoning in more depth. This article is about what shows up on your renewal notice.

Renewal price versus new-customer price

An existing customer and a new customer with identical details do not always get the same number, and many of the reasons are legitimate. Here are the main ones, along with what the filing would show.

Capping and transition rules. When an insurer introduces a rate change that would move some policies a lot, it may cap how much any one renewal changes. The white paper describes capping as limiting the premium change at renewal on a policy-by-policy basis, usually for a short period; its example is a rule that charges the full approved premium after no more than three renewal cycles. Capping can work in either direction. It keeps a large increase from hitting you at once, and it keeps a decrease from reaching you at once. The paper noted that regulators should weigh the length of the capping period and the effect on others' rates.

Different rating plans or programs. A new-business quote may come from a newer rating plan, a different company within the same group, or a program with new-customer features. The renewing policy may still sit in an older plan. An insurer can have more than one filed plan, but each has to be filed with the state.

Rating factors that changed. Your own record moves. A ticket or at-fault claim ages in or out of the lookback window, your credit-based insurance score changes where the state allows its use, your mileage, vehicle, address or household changes. A new-customer quote uses today's data, and your renewal uses data refreshed at renewal, so differences can go either way.

Base-rate increases. If the insurer has raised its base rate, everyone renewing sees it. A new-customer quote includes the same increase, so a gap here points elsewhere.

Tenure discounts that run out. Some insurers give discounts for years with the company or for continuous coverage that taper or are replaced by other discounts. These are rating features that should appear in the filing. The Consumer Reports analysis described below found that such loyalty discounts ranged from large to nothing.

The white paper's recommended disclosure list includes differences between the proposed prices for an insurer's existing and new customers. That item is where a filing would show an unexplained gap. Not every state requires insurers to file it, so a consumer usually sees the outcome (the renewal notice) rather than the model.

How credit and claims history interact with this

Credit and claims history do two jobs. They are legitimate cost-based factors where your state allows them, and they are also the factors most often blamed for an unexplained renewal increase. To tell the difference:

• A claim or violation. A recent at-fault accident can raise your renewal even when the claim cost nothing out of your pocket. See How a claim reaches your CLUE report and your renewal price for how a claim reaches your claims history and what that does to the renewal price.
• Credit. Where credit is permitted, an updated score can move your renewal. Some states ban credit use or limit how it affects renewals, so the same factor that raised a neighbor's price may be unavailable to your insurer. Our guide to how car insurance rates are calculated summarizes the major state differences.
• A point of confusion. Insurers may say a price changed because of "your risk" when the change was a base-rate increase. Ask which category applies: base rate, factor, or capping.

Price optimization is a different thing from all of these because it is not about you at all; it is about a model of how people like you react to price changes. It also does not need a changed factor to produce a gap. That is why the useful test is to compare against alternatives, not to read the premium breakdown line by line.

Retention modeling, in plain terms

The white paper cites Towers Watson (a consulting firm) as describing these demand models as fit to historical customer-level records of whether each customer renewed or accepted a quote; in its telling they predict how likely a customer is to renew or accept, not who shops more. In plain terms, an insurer feeds records of past customers into a model, and the model predicts who renews at which price. If the output is allowed to set your price, a customer predicted to renew at almost any price gets a higher one than a customer predicted to leave.

A related and legitimate use is for the insurer to set its overall rate level, forecast how many customers it will keep, and plan its budget. The line regulators drew is at the individual level: the task force specifically listed retention adjustment at an individual level and propensity to shop as inconsistent with a prohibition on unfair discrimination, while the broader forecasting that goes with running a business was not the target.

What the evidence says about loyalty penalties

The evidence is mixed, older than most readers would like, and partly from advocates, so read it with those limits in mind.

• Consumer Reports, July 2015. In an analysis of premiums for long-tenured customers, Consumer Reports found that loyalty discounts varied widely: some insurers gave a sizable discount, others a small one, and others none, and in some cases a longtime customer paid more. The magazine later corrected its text to say that several insurers it had first described as imposing penalties did not. It also reported that 53% of its subscribers had stayed with the same insurer for 15 years or more.
• Consumer Federation of America claims. CFA's Bob Hunter told Michigan Public in April 2015 that some customers were charged nearly a third more than less loyal customers with the same risk profile. The insurers he named did not respond to the reporter's questions, and the figure could not be verified independently. Treat it as an advocate's claim.
• Shopping behavior. The NAIC white paper reports that the CFA, citing a Deloitte survey, said only 18% of drivers shop for auto insurance every year and 58% rarely or never do, while an Insurance Information Institute poll found that 68% of people with incomes under $35,000 had compared prices at their last purchase, the highest share of any income group, against 61% for people above $100,000. The Insurance Research Council reported shopping rates between 21% and 26% across income groups within a 12-month window. These sources disagree about who shops, which is the central question for anyone arguing about who gets targeted.

No recent, independent U.S. study that measures renewal premiums against new-customer premiums across insurers turned up in the research for this article. Without one, no one outside the insurers can say how common a loyalty penalty is today. The practical conclusion is the same as it was in 2015: price differences between carriers are large, and the only way to know whether you are in a good position is to check.

Test your own renewal: five steps

1. Get the facts from the renewal notice. Note the new premium, the change from last term, and any reason the insurer gives. reading a car insurance renewal notice: rate-change notice rules explains what notices must say in many states.
2. Ask the insurer what changed. The question to ask is: "Of this increase, how much is from the company's rate change, how much from my own rating factors, and how much from any capping or transition rule?" You are entitled to ask; the answer you get is useful whether or not it is complete.
3. Get a new-customer quote from the same insurer. Use the same drivers, vehicles, limits, deductibles and discounts, entered the way the renewal has them. If the insurer sells through more than one channel (agent, online, phone), run the quote in at least one different channel. A markedly lower number for a new policy deserves a direct question.
4. Get at least three quotes elsewhere. Use our renewal rate increase checker to see whether your increase is out of line, then compare like for like with the how to compare car insurance quotes apples to apples method.
5. Decide, and file a complaint if the answer is unexplained. If an insurer cannot explain a gap and you suspect a non-risk adjustment, your state insurance department takes complaints and can ask the company to justify its rating.

A worked example with invented numbers

Say your renewal notice shows $2,140 for six months, up from $1,980. You get these figures for the same coverage:

How to read this: the same-insurer gap of $270 is the thing to question. If the insurer explains it as a one-time capping or plan difference, you have an answer and a decision about whether to stay. If it cannot, you can move to a lower quote, or you can ask to be re-rated into the plan the new-customer quote used. Insurers B and D are cheaper without any loyalty story at all, which also tells you that the difference might be a normal price spread between carriers and not a penalty. These numbers are hypothetical, not market data.

What this means for how often to shop

Regulators banned the practice in several states, but the underlying economics have not gone away: an insurer's price for a given customer depends partly on what it expects that customer to do. The cheapest protection is to make yourself a customer who compares. Our guide on how often should you shop for car insurance? suggests when to do that, and why your premium went up at renewal with no claims covers the other reasons a premium rises. A discount for staying put is worth taking, but only after you have confirmed you are not paying more for it than a new customer pays elsewhere.

Common questions

Is price optimization illegal?

In the states whose regulators issued bulletins or notices in 2014 and 2015, the regulator said it conflicts with the state's ban on unfairly discriminatory rates, and insurers there were told to stop and to remove its effects from filings. Elsewhere, the answer depends on how the state reads its own rating law. The NAIC white paper itself took no position on any state's policy. Check your state's insurance department for current guidance.

Does an insurer have to tell me if it uses it?

Not everywhere. Vermont's 2015 bulletin told insurers to disclose in personal-lines filings whether they use non-risk-related factors, and the NAIC paper suggested other states consider similar requirements. Regulators told the task force that few filings mentioned the practice, which suggests that disclosure was not routine. As a consumer you will mostly see the result, not the model.

Is a loyalty discount the opposite of a loyalty penalty?

Yes, in principle, and both exist as concepts. A discount for tenure or continuous coverage is a filed rating feature that lowers your price. The Consumer Reports analysis found tenure discounts ranged from sizable to nothing across insurers and states. A discount that exists on paper can still be smaller than the gap between your renewal and what a new customer would pay elsewhere, so compare totals, not discount labels.

Will shopping hurt my rate or credit?

Our article do car insurance quotes hurt your credit score? explains how quote requests are treated. Switching has its own steps, covered in how to switch car insurance without a gap.

Can I ask for my renewal to be re-rated?

You can always ask. Some insurers will re-run a policy against current plans or discounts, review drivers and mileage, or apply a discount you had not claimed. There is no right to the new-customer price, but the request costs nothing, and the answer tells you something.

Key takeaways

• Price optimization adjusts a price using predicted willingness to pay or shop, not expected claims; the NAIC's 2015 white paper named individual retention adjustments and propensity to shop as inconsistent with unfair-discrimination law.
• Maryland acted first (Oct. 31, 2014), followed by Ohio, California, New York, Florida, Vermont, Washington and at least nine other jurisdictions in 2015, mostly by bulletin under existing law; check your regulator for the current position.
• A renewal price above a new-customer price can come from capping, older rating plans, changed factors or expired discounts, so ask the insurer which one applies.
• Evidence of how common loyalty penalties are today is thin; the main U.S. evidence is from 2015 and partly from advocates.
• The reliable test is practical: a same-insurer new-customer quote plus three outside quotes for identical coverage.