Price Optimization

Insurance Information Institute

A Three-part series of articles on Allstate’s Complimentary Group Rating

October 10, 2016, Published by Carrier Management
Executive Summary
Opinion: Allstate’s rating process made price optimization the biggest controversy in P/C insurance. But it might not be what it seems, according to James Lynch, chief actuary of the Insurance Information Institute.

In this three-part series of articles, Lynch explains what he calls the ultimate irony—that “the Allstate plan that kicked off the brouhaha might not even be price optimization.” In Part 1, he explains what price optimization is—value-based pricing—and how that differs from the cost-based pricing methods that actuaries use to estimate how much rates should change on an entire book of business. In Part 2, he explains classification plans that introduce individual surcharges and discounts to overall pricing indications for different groups of customers and how Allstate introduced an enhancement to this step—designed to solve “the reversal problem.” In Part 3, he explains that what is being maximized is retention, not price, as he brings together concepts explained in Part 1 and Part 2 of this article series.

Lynch was asked by Allstate officials to review and write about the plan, but he independently chose Carrier Management to publish his article. The article is based on his review of a report by another actuary, information from Allstate and knowledge he has gained from his study of price optimization over a two-year period.

Click on each title to read the entire article. Click here for a printable PDF of the entire series (11 pages).

About Those Rates: Why Allstate’s Rating Program May Not Be What It Seems

October 10, 2016 by James Lynch, Part 1 of 3, Excerpt
Actuaries occupy a quiet, arcane world, their debates obscure: Loss development factors: weighted vs. straight? Premium impact: parallelogram vs. extension of exposures? Reserve methods: chain ladder vs. Bornhuetter-Ferguson.

So Allstate’s Complementary Group Rating process—CGR for short—is one for the books.

The early filings of the company’s CGR plan launched the price optimization controversy in the public eye. The plan, according to its detractors, let Allstate gouge consumers on price, charging the maximum amount each customer would be willing to pay.“This is a watershed moment in the history of insurance consumer protection,” said Robert Hunter in a press release

in late 2014. Hunter is director of insurance at the Consumer Federation of America (CFA), a consumer advocacy group that often finds fault with insurer practices.
“If regulators don’t block this scheme immediately, American consumers will pay a huge price. While we are forced by law to buy these companies’ insurance products in order to drive, there seems to be nothing stopping them from targeting millions of unsuspecting customers with unnecessary and unjustified price hikes.”

The media noticed, particularly after the CFA started calling the practice a “loyalty penalty.” NPR’s headline said, “Being a Loyal Auto Insurance Customer Can Cost You.” Consumer Reports called price optimization a “schmo tax,” as if to say you are a sucker if you don’t shop for insurance and your insurer will jack up your rates unless you do.

Insurers, the reasoning seemed to be, had invented a money milking machine, one they hooked every customer to. The machine, precisely calibrated, could drain just as much money as any customer could possibly stand but not one cent more. Then, at the last possible moment, insurers could close the spigot, leaving each customer drained—not completely dry but given a perfect milking. Consumers wouldn’t notice how much had been sucked from them. That way they would return, replenished and refreshed, at renewal, ready to be suckled again.
As we shall see, this was not the case, but the consumer advocates cried that Allstate’s machine was not the only one. CFA cited surveys by the consulting firm Earnix, which found that 45 percent of large auto insurers owned one.

The ultimate irony, perhaps: The Allstate plan that kicked off the brouhaha might not even be price optimization, at least if you follow the reasoning of one of the better-known actuaries of the past half century.

An Accident of the Rating Process: Allstate’s Cure Explained

October 10, 2016 by James Lynch, part 2 of 3, Excerpt
But pricing actuaries do much more than estimate how much rates should change on an entire book of business. They must estimate the various discounts and surcharges that different groups of customers must pay.

A company might charge $300 on average for collision coverage in a state. But hardly anyone would pay exactly that amount. Male drivers would pay more in most states. Middle-aged drivers would usually enjoy a discount.

All of those adjustments that take the $300 average and turn it into the, say, $278 that you would pay are called rating variables. The system that blends all the variables together—up for male drivers, down for the middle aged, down for those with excellent credit, up for those with a DUI conviction—is known as a classification plan, or class plan for short.
The first auto class plans go back at least six decades, according to a presentation actuary Jeff Kucera made at the Midwest Actuarial Forum in April. The first plans had a handful of variables, maybe a half-dozen.

Actuaries over time have found more and more variables that credibly predict how likely a customer is to file a claim. Today there are hundreds. Married people file claims less frequently than single people. Middle-aged people file less often than young people. Homeowners file less often than renters. (Not all insurers use every variable, and some variables are prohibited in some states.)

It is the actuary’s model that teases out all these differences, and it is the actuary’s job to prove to regulators that the size and direction of change is appropriate—that young drivers should pay a surcharge of, say, 40 percent, not 140 percent, or that persons with excellent driving records should get a credit of 10 percent, not 15 (all of those are numbers I just made up).

This tapestry of factors isn’t perfect. It has one notable flaw. Bass calls it the reversal.

Reversals happen when all those factors collectively say a person’s premium should go down but their premium actually rises. A driver for whom the actuarial indication is a decrease of, say, 4 percent, actually sees their premium rise.

Reversals are an accident of the rating process. They are rare, but they are almost inevitable.

Reversals had been an unintentional, unavoidable consequence of the rating system.

One way insurers handle unusual situations is to cap premium changes. To prevent rate shock, a company would file a premium cap of, say, 25 percent. If some confluence of rating factors hiked premium by, say, 75 percent, the increase would be capped at 25 percent.

That is fairly easy to program. The company loads its new rating factors into its systems. The computer compares the new premium with the old. If the premium rises too much, a cap settles in.

It’s a crude response, but it sledgehammers away some of the imperfections of the old system.

It is much harder to find and correct reversals. Not only do you need the old rating factors and the new rating factors; the latest actuarial analysis has to be poured into the computer as well.

This is where Allstate’s CGR plan comes in. It checks for premium reversals as part of the actuarial analysis and sets up a way to eliminate them. It is an additional factor that knocks the indicated rate into place, to be consistent with the actuarial analysis.

Not All ‘Optimization’ Is Price Optimization; Allstate’s CGR Proves It

October 11, 2016 by James Lynch, part 3 of 3, Excerpt

“The purpose of price optimization is to extract as much profit as possible from policyholders, who are often required to purchase insurance policies,” according to a February 2015 press release from the Consumer Federation of America.

Is that what Allstate is doing with its Complementary Group Rating?

First, the CGR plan does not maximize profits. It maximizes retention. Remember all those paragraphs ago (in Part 1) when I asked you to remember that the standard actuarial model assumes that all risks renew no matter how much their rates go up? And if all risks renewed, the model indicated that the company would achieve a certain profit level. And that profit level was the return on equity approved by the regulator.

Allstate’s rating method does something more sophisticated. It understands that some risks will not renew after a rate change. It looks at several sets of rating plans, all of which promulgate fair rates. It tries to figure out which set will give the highest level of retention. In other words, it determines which plan is most like the 100 percent retention assumed by the actuarial model.

The overall profit that results is disclosed in the rate filing. If that profit is too high, the regulator will reject the filing. The only way the insurer can make excessive profits is if the regulator approves the filing. Saying that insurers are maximizing profits or price-gouging, in this case, is akin to saying regulators aren’t doing their jobs. There is no evidence that is happening.

Second, the plan doesn’t price-gouge individuals. If the actuarial models say a policyholder requires a 10 percent increase for the insurer to make a reasonable profit, then that policyholder will see an increase of, at most, 10 percent. That just isn’t price-gouging.

When the model calls for an increase, the premium the policyholder pays is never more than what the actuarial model says that customer should pay. It is hard to price-gouge someone when they are being charged less than what the actuarial model suggests they should pay. (A person for whom the model calls for a decrease sees the opposite situation.)
Allstate’s plan actually seems to improve on the standard ratemaking methods, which can inadvertently result in reversals. Remember in the standard methods, a customer who should see lower rates might see them rise, an inevitable consequence of what has been the state of the actuarial art.

Third, it doesn’t charge each customer the most he or she is willing to pay. It charges a rate that maximizes the retention across the entire portfolio. One individual might be willing to pay more than Allstate wants to charge, but that rate would be part of an overall plan that resulted in fewer customers. So that customer is actually charged less than what he or she might tolerate.

Actuary Bass summed up the argument, writing that Allstate’s plan “reflects business judgments that have always been a part of accepted, sound actuarial ratemaking; but…does [them] better by making sure there are no rate reversals that result from judgmental adjustments to the rating factors.”

Bass’ main point, I think, is that there should be no professional or regulatory qualms about letting a computer replace the old “gut-feel” ratemaking adjustment with a sophisticated, quantitative technique —particularly when the latter results in fairer rates for everyone.

More formally, she concludes that Allstate’s plan:
• Is cost-based, not value-based, which makes it consistent with actuarial ratemaking principles.
• Complies with applicable requirements of regulators and the actuarial profession.
• Is not consistent with the key principles of price optimization, because it doesn’t maximize prices or profits.

Though every state has its own approach to regulating any issue, there are some signs that regulators’ stance has softened as understanding of the practice has grown. Contrast what Florida proscribed—seemingly any adjustment to an indicated rate—with Missouri’s action eight months later. Missouri’s bulletin seems more measured. It said optimizing techniques could “potentially” violate insurance laws. It decried adjustments based on the “propensity to shop for insurance…ask questions or file complaints.”

Nonetheless, it seems unfortunate that the price optimization debate focused on a matter settled long ago—adjusting actuarial indications into the real world—and scorning what appears to be an improvement on the standard method.

The lesson here seems to be that everyone—consumer advocates, regulators, the news media—should avoid a headlong rush to judge such a complex and delicate process as insurance rating. The public would seem to be best served when all parties proceed judiciously, taking time to understand a process thoroughly and isolating what parts might be objectionable and why, before condemning the whole.

National Association of Insurance Commissioners

Price Optimization Activity:
December 2013:
The issue of price optimization was brought before NAIC’s Auto Insurance (C/D) Study Group.

November 2014:
The study group referred the issue to the Casualty Actuarial and Statistical (C) Task Force because “the topic of price optimization goes beyond auto insurance and requires a great deal of actuarial expertise.”

March 2015:
The task force issued its first draft of a white paper on price optimization. Numerous comments were received from regulators, consumer groups and industry trade groups.

May 2015:
An updated draft of the white paper was issued and opened for comment through July 3, 2015.

November 2015:
NAIC adopts Price Optimization White Paper.

State Commissioner Bulletins:

Maryland, October 31, 2014
Ohio, January 29, 2015 – File not available
California, February 18, 2015 – File not available
New York, March 18, 2015
Florida, May 14, 2015
Vermont, June 24, 2015 – File not available
Washington, July 10, 2015 – File not available
Indiana, July 20, 2015
Pennsylvania, August 22, 2015
Maine, August 24, 2015 – File not available
District of Columbia, August 25, 2015
Rhode Island, September 18, 2015 – File not available
Montana, September 18, 2015
Delaware, October 1, 2015 – File not available
Colorado, October 29, 2015
Minnesota, November 16, 2015
Connecticut, December 4, 2015 – File not available
Alaska, December 8, 2015 – File not available
Missouri, January 12, 2016
Virginia, April 16, 2016 – File not available

Find additional links to related items at: http://www.naic.org/committees_c_catf.htm

State Regulation of Price Optimization

February 18, 2015, the California Insurance Commissioner issued a notice to more than 750 property and casualty insurers, announcing that “any use of price optimization in the ratemaking/pricing process or in a rating plan is unfairly discriminatory in violation of California law.”

March 18, 2015, the New York Department of Financial Services (DFS) issued a Section 308 inquiry letter requesting information about the use of price optimization by insurers.

May 14, 2015, the Florida Office of Insurance Regulation (OIR) issued Informational Memorandum OIR-15-04M, which bans certain price optimization practices.

July 20, 2015 – Indiana Insurance Commissioner Stephen Robertson issued an official Bulletin 219, alerting insurers to the Departments position that the use of price optimization in establishing insurance rates is not permitted. Indiana is the seventh state to explicitly ban price optimization, following Washington, Florida, Maryland, Ohio, Vermont and California.

August 22, 2015 – The Pennsylvania Insurance Department issued Price Optimization; Notice 2015-06 to remind insurers about the Department’s longstanding prohibition against the use of price optimization techniques in property and casualty insurance rates.

August 25, 2015 – The District of Columbia’s Department of Insurance, Securities and Banking (DISB) issued Bulletin 15-IB-06-8/15 notifying insurers doing business in the District that price optimization is discriminatory and that it violates the District’s anti-discrimination insurance laws. The bulletin directs any insurer using price optimization to rate insurance policies in the District to cease such practice.

September 18, 2015 – Rhode Island issued Insurance Bulletin #2015-8 reminding insurers that “base rates and rating classes must be based on factors specifically related to an insurer’s expected losses and expenses.” Insurers that use price optimization must submit revised filing that remove such factors by November 18, 2015.

September 18, 2015 – Montana issued an Advisory Memorandum concluding that “the use of price optimization constitutes an illegal, unfairly discriminatory practice.” Insurers currently using price optimization must file new rating plans no later than February 1, 2016.

October 29, 2015 – Colorado Commissioner Marguerite Salazar issued Bulletin B-5.36.

November 16, 2015 – Minnesota Commerce Commissioner Mike Rothman issued Administrative Bulletin No 2015-3, which ordered Property & Casualty Insurers to cease using “price optimization” in conjunction with the personal lines policies in the State of Minnesota.

January 12, 2016 – Missouri Department of Insurance issued bulletin 16.02, banning price optimization.

April 16, 2016 – Virginia Insurance Commissioner issued administrative letter 2013-03 banning price optimization.

Media Coverage

Please click the title to view the full article. Articles appear oldest to most recent.

As Data About Drivers Proliferates, Auto Insurers Look to Adjust Rates
April 18, 2014, NY Times [Excerpts]
It’s no surprise that car accidents, speeding tickets and where you live all affect how much you pay for automobile insurance. But consumer groups say that other, apparently unrelated, factors are unfairly being used to set rates.

Two organizations, the Consumer Federation of America and the New York Public Interest Research Group, have released separate reports contending that the factors insurance companies are using to determine rates — like educational level and occupation — are detrimental to consumers, especially to low-income customers.
More data available online about all of us, he said, along with increasingly sophisticated software to interpret those details, gives insurance companies better information than ever in targeting insurance rates.

California, Massachusetts and Hawaii do not allow credit scores to be taken into consideration by insurance companies when determining rates. California also prohibits using educational level or occupation.

The fact that California does not allow the use of education level, occupation or credit score shows “it’s possible to have a rating system without these discriminatory factors,” Mr. Morrison said.

Mr. Hunter of the consumer federation said he grew concerned last year when he saw reports by Earnix, a company that sells software to some of the nation’s largest auto insurers, discussing the growing prevalence of something called “price optimization.”

Price optimization incorporates data about a company’s operating costs, consumer behavior and the competitive environment to achieve the optimal desired outcome — like maximizing profits or expanding a business.

Meryl Golden, North American general manager of Earnix, said there was nothing underhanded about it. “It just replaces broader judgments with more fine-tuned analytics,” she said.

Data mining is now used to set insurance rates; critics cry foul
April 16, 2014, by Herb Weisbaum | @TheConsumerman, ETCNBC.com [Excerpt]
Insurance is a different kind of product. You n eed it to get a mortgage and you’re supposed to have it to drive a car. That’s why the industry is highly regulated.

You might not like the price you pay for your home or auto insurance policy, but by law those premiums must be based on actuarial risk–the expected cost to provide you that coverage.

Some insurance companies now use sophisticated software to help them set their rates. The industry says the process, called “price optimization,” is simply a way to be more efficient. Consumer advocates believe it is being used to get around risk-based pricing.

Actuaries, Algorithms and Drivewise
Exclusivefocus Summer 2014, NAPAA
Unbeknown to many in the P&C business, a new rate making methodology has been slowly infiltrating the automobile insurance industry over the past two years. With so much information available via the web, and the accompanying explosion of companies scouring the web for this data, someone was bound to find a way to turn their collection of information into a “pot of gold.” That pot of gold has come to the insurance industry in the form of mathematical algorithms.

Good hands, great data
October 30, 2014, By Jennifer Alsever, Fortune.com (Excerpt)
A hard look at how data could be better used internally has helped insurance company Allstate tighten its operations and reduce its risk. Wall Street seems to agree.
In July 2011, Floyd Yager, then senior vice president of analytics at Allstate Insurance, huddled in a conference room with Suren Gupta, the company’s chief information officer. Yager had an idea. He had heard about open source software called Hadoop that could process millions of datasets in amazing speed on a just a dozen computers. Yager had a small team that could mine this information but he needed Gupta’s buy in: the computers, the storage, the network and the capacity to go forth. Could the company’s information technology department work with him to get it done?

Maryland Insurers Using ‘Price Optimization’ Ordered to File Corrective Action Plan
November 3, 2014, InsuranceJournal.com [Excerpt]
The Maryland Insurance Administration (MIA) issued a bulletin alerting insurers that the use of “price optimization” in Maryland is in violation of §27-212(e)(1) of the state’s Insurance Article.

MIA said it is requiring insurers that currently utilize price optimization to rate insurance policies in Maryland to file a corrective action plan with MIA by no later than Jan. 1, 2015.

According to the bulletin, price optimization refers to the practice of varying rates based on factors other than the risk of loss — such as the likelihood that policyholders will renew their policies and the willingness of certain policyholders to pay higher premiums than other policyholders.

“The MIA has determined that the use of price optimization results in rates that are unfairly discriminatory in violation of the §27-212(e)(1) of the Insurance Article,” the bulletin stated. “As a result, insurers may not use price optimization to rate policies in Maryland.”

CFA Rips Allstate’s Auto Insurance Pricing
December 16th, 2014, by Editor, Corporate Crime Reporter [Excerpt]
An Allstate document discovered by the Consumer Federation of America (CFA) provides the first clear evidence that the insurance giant is basing customers’ auto insurance premiums on a new factor, called “marketplace considerations,” that has nothing to do with the risk that a driver will cause an accident or file any claims.

This and similar schemes, often referred to as “price optimization,” have been developed by insurance companies and consulting firms to increase profits by raising premiums on individuals who are unlikely to shop around to find a better price.

Price Optimization Allegations Challenged, NAIC Investigating Practice
December 18, 2014, by Don Jergler, InsuranceJournal.com
Reaction to industry-wide price optimization allegations from a consumer group known to call out insurers has been somewhat critical yet guarded – however it appears regulators may already be starting to look into the practice.

Consumer Federation of America this week said it has “clear evidence” that insurance giant Allstate is basing auto insurance premiums on a factor called “marketplace considerations,” which CFA argues have nothing to do with a driver’s risk.

Allstate, state regulator challenged on auto rates
January 13, 2015, By Richard Piersol, Lincoln Journal Star (Excerpt)
The Consumer Federation of America challenged Nebraska Insurance Director Bruce Ramge on an Allstate auto insurance pricing system that CFA said unfairly raises premiums for thousands of Nebraskans.

The group urged Ramge to disapprove and prohibit Allstate from using the rating plan, if it has already taken effect, and to bar other insurers from using anything like it. As of 2013, Allstate had the ninth largest share of auto insurance in Nebraska.

Ramge said the department is in the initial stages of reviewing the matter. “It is too early to comment,” Ramge said in an email Tuesday morning.

“Price optimization” puts the screws to loyal car insurance customers
January 26, 2015, by Insure.com, TheStreet.com
Price optimization is a little-known technique that penalizes loyal auto insurance policyholders by charging them higher rates because their insurer believes they are unlikely to shop around for a better price. By mining data on customers and testing incremental price increases, some insurers try to predict who is less likely to switch companies, and then charge those customers more in order to squeeze out higher profits.

The practice is under review by some state insurance regulators, thanks to the spotlight shined on it by the Consumer Federation of America (CFA).

Ohio Insurance Director Warns Insurers Against Use of Price Optimization
February 10, 2015, InsuranceJournal.com
Lieutenant Governor and Department of Insurance Director Mary Taylor has warned Ohio insurers against the use of price optimization that can result in unfair discrimination.
Taylor recently issued Bulletin 2015-01, dated Jan. 29, to Ohio insurance companies noting that the use of price optimization violates Ohio law.

Price optimization is an insurance company’s practice of varying premiums based upon factors such as whether a consumer has complained about a policy or the amount or percentage change of the consumer’s premium over prior years.

See ODI bulletin: https://insurance.ohio.gov/Legal/Bulletins/Documents/2015-01.pdf

Insurance commissioner bans ‘price optimization’ in California
Insurers put on notice to stop illegal pricing activities.
February 23, 2015, By Patricia L. Harman, PropertyCasualty360.com
Property and casualty insurers in California have been notified by insurance commissioner Dave Jones that “price optimization” in setting rates is discriminatory and violates state law.

According to the commissioner, the practice in essence bases what a group of individuals will pay for insurance on their “willingness to pay a higher premium relative to other individuals or classes.” Customers less likely to compare prices for policies in a particular area may find themselves paying more for the same coverage offered at a lower price in a different jurisdiction because those residents are savvier when it comes to shopping for rates.

“It is illegal for an insurer to charge people different rates based on their sensitivity to price increases or the likelihood that they will comparison shop,” said Jones in a statement. “Price optimization represents a fundamental threat to fairness in rating. As insurance commissioner, I remain committed to ensuring California maintains a healthy and vibrant insurance marketplace, while making sure consumers remain protected.”

California Joins Maryland and Ohio in Addressing “Price Optimization”
February 24, 2015, by Thomas Curvin, Tracey Ledbetter, Phillip Stano, Mary Jane Wilson-Bilik | Sutherland Asbill & Brennan LLP, JDSupra.com
The California Insurance Commissioner issued a notice on February 18, 2015 to more than 750 property and casualty insurers doing business in California, announcing that “any use of Price Optimization in the ratemaking/pricing process or in a rating plan is unfairly discriminatory in violation of California law.”

The notice defines price optimization as “any method of taking into account an individual’s or class’s willingness to pay a higher premium relative to other individuals or classes.” The California Insurance Commissioner contends that price optimization “does not use actuarially sound methods to estimate the risk of loss” and “represents a fundamental threat to fairness in rating.”

In taking this action, California joins insurance regulators in Maryland and Ohio, both of whom also recently issued insurance bulletins addressing price optimization.
The Maryland Insurance Administration issued Bulletin 14-23 on October 31, 2014. The Maryland Bulletin defined price optimization as “varying rates based on factors other than risk of loss, including, but not limited to: (a) the likelihood that a policyholder will engage in activities that result in policy turnover [defined to include shopping with other carriers for a lower premium, cancelling a policy before the expiration of the policy term, or failing to renew a policy at the renewal of the policy term]; and (b) the willingness of a policyholder to pay a higher premium compared to other policyholders.” The Bulletin stated that price optimization, “by its nature,” involves discriminating against policyholders of the same class based on factors other than actuarial risk. The Maryland Insurance Administration required every insurer that then used price optimization in Maryland to file a corrective action plan by January 1, 2015.

The Ohio Department of Insurance followed Maryland’s lead by issuing Bulletin 2015-01 on January 29, 2015. The Ohio Department defined price optimization as the practice of “varying premiums based upon factors that are unrelated to risk of loss in order to charge each insured the highest price that the market will bear.” The Bulletin took the position that price optimization allowed insurers to set premiums based on factors that were unrelated to the risk of loss or expense. It required insurers that use price optimization to rate insurance policies in Ohio to submit a SERFF filing compliant with the Bulletin by March 31, 2015, with proposed effective dates no later than May 31, 2015 for new business and June 30, 2015 for renewal business.

The NAIC Casualty Actuarial and Statistical (C) Task Force is conducting research on the use of price optimization, including the regulatory implications, and is expected to issue a white paper documenting the relevant issues sometime this year. We will continue to monitor and report on price optimization developments in the states and at the NAIC.
Full article at: http://www.jdsupra.com/legalnews/california-joins-maryland-and-ohio-in-ad-76681

Investor wonders if Allstate’s use of big data violates civil rights
March 13, 2015, By Joe Cahill, ChicagoBusiness.com (Excerpt)
Companies are embracing “big data” as a tool for creating personalized products, pitches and pricing for a wide range of customers.
But is it also a primrose path to civil rights violations? An Allstate shareholder wants the giant auto insurer to find out.

Calvert Investments, a money manager known for advancing social causes, is seeking a report on possible civil rights liabilities arising from Allstate’s use of big data in underwriting and pricing insurance policies. The $13 billion fund has asked Allstate to include in this year’s proxy statement, a shareholder resolution calling for such a report.
The proposed resolution would “request that the board prepare a public report, at a reasonable cost and omitting proprietary information, by Oct. 31, 2015, describing how the board and company management identify, oversee, and analyze civil rights risks related to Allstate’s use of big data, how they mitigate these risks, and how they incorporate assessment results into company policies and decision-making.”

Allstate, not surprisingly, isn’t crazy about the idea. In a 13-page letter to the U.S. Securities and Exchange Commission, company lawyers gave several reasons why Allstate shouldn’t have to include Calvert’s resolution in the proxy statement for its annual shareholders meeting May 19.

TOO COMPLEX
They argue the proposal improperly meddles in day-to-day company affairs, that it deals with matters too complex for shareholders to address, and that Allstate already has “substantially implemented” the aims of the resolution through its existing policies and procedures for risk management and legal compliance. For good measure, Allstate calls the proposal too “vague and indefinite” to implement.

SEC reviewers have yet to rule on Allstate’s request for assurance that the agency won’t object if the company excludes Calvert’s proposal from the proxy statement.
Whatever the SEC says, questions about the civil rights implications of big data aren’t likely to go away.

A report last month by President Barack Obama’s Council of Economic Advisers warned that “big data may facilitate discriminatory pricing strategies that target consumers based on factors outside their own control, or run afoul of antidiscrimination provisions in existing laws such as the Fair Credit Reporting Act or Civil Rights Act.”

Insurance is emerging as a flashpoint in the debate. Few industries have been as aggressive in deploying big data and related technologies. Insurers collect reams of information on customers, using it to evaluate risks and price policies.

Allstate has been at the forefront with technologies such as telematics, which tracks a customer’s driving habits. Customers who let Allstate put telematics devices in their cars can get discounts. Effective use of these technologies has helped limit claims and bolster profit margins at Allstate in recent years.

ENABLES PROFILING
But Calvert and others see potential for discrimination in the use of “behavioral” data to determine risk and price policies.

“While allocating risk and setting higher fees for higher-risk customers is central to the insurance business model, this use of big data also enables customer profiling, with potentially problematic civil rights implications,” Calvert wrote in support of its proposal.

Calvert points out that assessing risk levels and setting prices based on driving behaviors can result in higher rates for some safe drivers. For example, insurers that identify a link between late-night driving and highway accidents might raise rates on customers who regularly drive at night. This would penalize late-shift workers with safe driving records, many of whom are members of minority groups protected by civil rights laws.

The SEC may not object if Allstate excludes Calvert’s proposal. But that won’t insulate Allstate from claims that big data pricing practices unfairly discriminate. At some point, the company will have to address the issue directly.

Click here to read correspondence related to the Calvert proposal: The Allstate Corporation; Rule 14a-8 no-action letter – SEC

N.Y. Regulator Studying How Car, Other Insurance Rates Are Set
Financial watchdog Benjamin Lawsky sends letter to insurers inquiring about mathematical pricing models
March 19, 2015 , By Leslie Scism, The Wall Street Journal
New York’s top financial watchdog warned insurers this week about charging higher prices to customers least likely to shop around, the latest state to raise questions about how the industry uses data when setting rates.

Department of Financial Services Superintendent Benjamin Lawsky sent a letter to hundreds of car and property insurers asking for any details on math-driven models that seek to predict how customers react to various price levels, according to a copy reviewed by The Wall Street Journal. The practice, according to the letter, could violate a state law that prohibits “unfairly discriminatory” rates.

New York DFS Opens Inquiry Into Price Optimization
March 20, 2015 InsuranceJournal.com
The New York Department of Financial Services (DFS) on March 18 sent out letters to hundreds of property/casualty insurers operating in the state, seeking information concerning price optimization.

New York joins a growing number of states around the country that are examining the practice of price optimization. Maryland, Ohio and California have prohibited the use of price optimization in their states in recent months, and the National Association of Insurance Commissioners’ (NAIC) Casualty Actuarial and Statistical Task Force is currently looking at the practice and plans to issue a report this year.

Insurers that currently use price optimization models or considerations are directed to respond to more detailed questions such as describing in detail all such price optimization models and considerations used, as well as how the company specifically incorporates the results of price optimization models and considerations into its pricing decisions. Insurers are directed to respond by no later than April 15, 2015.

The following is a copy of the letter from the New York Department of Financial Services: http://www.sutherland.com/portalresource/NY-308-Letter.pdf

N.Y. DFS Eyes Insurers’ Cybersecurity, Price Optimization
April 6, 2015, InsuranceJournal.com
New York state regulators recently sent a letter to insurers operating in the state, seeking detailed information on their cybersecurity measures.
Benjamin Lawsky, superintendent of the New York Department of Financial Services (DFS), asked insurers in a March 26 letter to provide responses to some 16 wide-ranging cybersecurity questions.

These questions include: the curriculum vitae and job description of the current chief information security officer; descriptions of how the insurer maintains information security procedures to address the goals of confidentiality, integrity, and availability; any significant changes to the insurer’s information technology (IT) portfolio over the last two years; and protections used to safeguard data that is accessible to third-party service providers.

Additionally, the DFS on March 18 asked insurers operating in the state to provide information concerning so-called “price optimization” techniques.
New York joins a growing number of states around the country that are examining the practice of price optimization. Maryland, Ohio and California have prohibited the use of price optimization in their states in recent months.

[Georgia] Allstate looking to adopt controversial pricing model
April 17, 2015, By Caitlin Bronson, IBAMag.com (Excerpt)
Documents recently obtained by an Atlanta news team suggest one of the nation’s leading property/casualty insurers is seeking to change the way it prices policies – something that may bolster arguments in favor of an independent agent’s value in a highly commoditized space.

In forms submitted to the Georgia insurance commissioner’s office, Allstate proposes to adopt a controversial method of pricing auto insurance policies called “price optimization.” Unlike other auto insurers, Allstate has not previously engaged in this method of pricing, which critics say is designed to boost insurer prices while downplaying the actual risk of the insured driver by charging loyal customers more for auto insurance.

[New Mexico] Insurance companies explore higher prices for loyal customers
April 18, 2015 By Rosalie Rayburn / Journal Staff Writer, ABQJournal.com (Excerpt)
Albuquerque, NM– Think your good driving record and years of loyalty to your car insurance company will get you better rates?
Maybe not.

Some insurance companies nationwide have a new method of calculating rates known as “price optimization” that uses statistical modeling to predict how likely a customer is to stick with what they’ve always done and keep renewing their policy, even if the price increases.

New Mexico Insurance Superintendent John Franchini this week said he intended to reject insurer Allstate’s recent rate filing with the state, based on an analysis of the filing.
Price optimization is now used by 45 percent of large insurance companies in North America, according to a 2013 survey by customer analytics software company Earnix.
Allstate would not directly respond to the consumer group’s allegations or say whether it uses price optimization in calculating rates.

Legal Alert: NAIC Update – Spring 2015
April 22, 2015, www.Sutherland.com [Excerpt]
The National Association of Insurance Commissioners (NAIC) held its first national meeting for 2015 in Phoenix, Arizona, from March 26 through March 31. Noteworthy new initiatives include cybersecurity, price optimization and the development of a model law for unclaimed life insurance benefits.

The following are notable highlights of the meeting, along with a summary of subsequent actions taken by the NAIC following the meeting:

2. Price Optimization
Use of certain “price optimization” techniques for property and casualty insurance has been taken up by the Casualty Actuarial and Statistical (C) Task Force, which is now working on a white paper on the topic. This undertaking was prompted by a referral from the Auto Insurance (C/D) Study Group, which concluded that the topic of price optimization goes beyond auto insurance and requires actuarial or statistical expertise. The Task Force prepared an initial draft white paper for comment before its March meeting. The stated purpose of the white paper is to provide research on price optimization, identify potential benefits and problems with price optimization, and present options for state regulatory responses.

Being A Loyal Auto Insurance Customer Can Cost You
May 8, 2015, NPR.org
Many companies reward their most loyal customers with incentives, discounts and freebies. But in car insurance, the opposite can actually happen. A driver can be punished with a higher premium just for being loyal to the company.　

It’s called price optimization, and it happens to lots of people all the time. A driver could have no history of accidents but all of a sudden their car insurance goes up.
Among the major insurance companies we contacted, only Progressive and State Farm told us they don’t price optimize.

Allstate initially denied it uses “any form of the price optimization tool” described in this story. But in 2011, Allstate filed a report with the SEC saying that it does indeed use price optimization.

Eventually, after a lot of back-and-forth emails, Allstate spokesman Justin Herndon amended the statement. “We do not engage in price optimization that seeks to charge the highest rate the market will bear,” Herndon said.

In a statement, Allstate said it is “committed to operating with absolute integrity. The Consumer Federation of America’s allegations of illegal pricing methods continue to be wrong and misinformed. Our prices are legal and actuarially sound.”

For now, there are two class-action lawsuits seeking damages for price optimization.

Meanwhile, the National Association of Insurance Commissioners plans to issue recommendations to states on the practice by the fall.

http://www.npr.org/2015/05/08/403598235/being-a-loyal-auto-insurance-customer-can-cost-youu

Webcast: The Price Is Right? Predictive Modeling, Big Data and “Price Optimization”: Regulatory and Litigation Developments
May 12, 2015, www.Sutherland.com
Predictive analytics, fueled by big data and driven by our constant use of smartphones, apps and web browsers, are increasingly used by insurers in marketing and distribution, underwriting, in-force management and claims. Recently, certain uses of predictive modeling in pricing personal lines policies, dubbed “price optimization” by many, have drawn the attention of state regulators, consumer advocates and class action plaintiffs. Regulators in Maryland, Ohio, California and New York have addressed the topic, the National Association of Insurance Commissioners has formed a “price optimization” task force to study the issue, and class actions challenging “price optimization” have been filed in at least two states. Tune in to hear Sutherland lawyers Eric Arnold, Tom Curvin, Mary Jane Wilson-Bilik, Dave Hancock and Tracey Ledbetter discuss potential regulatory and litigation issues associated with “price optimization” and related practices.

Resources:
Video – Play | mp4 | 60 min
Podcast | mp3 | 60 min
Presentation & Materials

Florida Bans Price Optimization; Insurers Question Definition
May 19, 2015, By Amy O’ Connor, InsuranceJournal.com (Excerpt)
A memorandum released last week by Florida Insurance Commissioner Kevin McCarty warning insurers in the state not to use price optimization in rating has gotten cheers from a consumer advocate group and jeers from an insurance industry trade association that says the definition of the practice used is too vague.

The memo, released May 14 by the Florida Office of Insurance Regulation (OIR), was addressed to all property/casualty insurers authorized to do business in Florida to “emphasize the requirements of the Florida Insurance Code in connection with insurers’ use or potential use of price optimization in determining policyholder premiums.”
Price optimization introduces information such as supply and demand or competition into the rating of policyholders. The practice took center stage late last year when the Consumer Federation of America (CFA) charged that insurance giant Allstate was basing auto insurance premiums on the “marketplace considerations” factor.

Big Data’s Big Guns: Allstate
May 26, 2015, By Nathan Golia, Insurancenetworking.com (Excerpt)
Like some of its peers, Allstate has seen itself as a big data company for a while, according to chief data officer Floyd Yager. But the impetus to do more with the data it had collected over the years has never been greater.

“Allstate has always had very good data, but it does exactly what it was designed to do: make us “transactionally” efficient on a product basis,” Yager says. “Now it’s about, ‘How do I take that operationally efficient data and turn it into a customer/household view and understand all the products attached to a person?'”
Allstate has focused heavily on master data management and data governance over Yager’s tenure, creating party and household IDs for data. The company is also building a team to work across business areas on analytics projects rather than siloing big data projects within certain units.

Allstate wants to sell customer driving data for profit
June 1, 2015, by Caitlin Bronson, Insurance Business America [Excerpt]
Privacy concerns over the use of telematics by insurance companies were heightened last week as one of the nation’s leading auto carriers said it was considering selling consumer driving data.

According to statements made Thursday by Allstate Chairman and CEO Tom Wilson, the property/casualty insurer hopes to profit from the sale of telematics data and then pass on savings to consumers by lowering premiums.

The possibility of insurers selling driving information is one of the reasons roughly half of the driving population views telematics negatively, according to a 2014 survey from Deloitte. Analyst John Lucker actually anticipates concerns over privacy will bifurcate the auto insurance market in the near-term – some carriers will adopt telematics, while others will stick to traditional underwriting factors for those concerned with privacy.

Allstate just patented technology to collect all kinds of information about drivers on the road
June 23, 2016, By Jonathan Fisher, BusinessInsider.com (Excerpt)
Allstate was recently granted a patent for a driving-behavior database that would let the insurance company “evaluate drivers’ physiological data, including heart rate, blood pressure and electrocardiogram signals.”

With this technology, Allstate could calculate how safe a driver is by tapping into sensors in the steering wheel and brakes, or adding cameras inside a car.

New Questions Raised As Companies Like Allstate Consider Monetizing Customer Data
June 19, 2015, By Steve Shoaff, Business2Community.com (Excerpt)
Few industries today remain untouched by disruptive technology. As companies start recognizing that their traditional business models are being challenged in the digital world, many are evaluating new ways to extract value from their existing assets. This includes evaluating new opportunities to maximize the value of their customer data such as using it for direct financial gain.

Vermont Issues Bulletin on Price Optimization in Personal Lines Ratemaking
July 2, 2015, InsuranceJournal.com
Vermont regulators issued a bulletin last week addressing the issue of “price optimization” in personal lines ratemaking.
The Vermont Department of Financial Regulation (DFR) noted in the bulletin that some property/casualty insurers have been relying upon the practice of price optimization to help determine the premiums that they will charge to policyholders.

But DFR warned property/casualty insurers that adjustments to rates may not be based on non-risk-related factors and that going forward, all personal lines rate filings must disclose whether the company uses non-risk-related factors such as “price elasticity of demand.

“The DFR Insurance Bulletin No. 186 was issued on June 24 and is applicable for all property/casuality insurers issuing personal lines policies in Vermont.

Consumer Federation of America Report

Smoking Gun Reveals Allstate’s Illegal Auto Insurance Pricing Scheme [Excerpt]

Document Provides Evidence That American Insurance Companies Are Using So-Called “Price Optimization” To Push Up Premiums Based on a Consumer’s Shopping Habits in Violation of “Unfair Discrimination” Laws.

Washington, D.C. (December 16, 2014) –
An Allstate document discovered by Consumer Federation of America (CFA) provides the first clear evidence that the insurance giant is basing customers’ auto insurance premiums on a new factor, called “marketplace considerations,” that has nothing to do with the risk that a driver will cause an accident or file any claims. This and similar schemes, often referred to as “Price Optimization,” have been developed by insurance companies and consulting firms to increase profits by raising premiums on individuals who are unlikely to shop around to find a better price.

The document – a filing submitted to the Wisconsin Department of Insurance (available here) – includes 1,300 pages of tables listing approximately 100,000 “micro-segments into which each one of Allstate Property and Casualty’s Wisconsin customers is placed. Allstate then assigns each micro-segment to one of one thousand “complementary groups,” which determine policyholders’ premiums and can range from giving a customer a 90% discount off the standard rate to increasing his or her premium by 800%, depending on Allstate’s analysis of the individual policyholder’s “marketplace considerations.”

CFA’s findings have been distributed in a letter (available here) to all of the nation’s State Insurance Commissioners. The revelations are also being presented by CFA to a conference call of the National Association of Insurance Commissioners today.

Read the full article here.
Related:
Media Access to Allstate Documents Blocked
Letter to all State Insurance Commissioners
Wisconsin Auto Rate Filing

Following are links to numerous items from the Consumer Federation of America on the subject of Price Optimization:

Consumer Information

Consumer Alert: New Auto Insurer Pricing Scheme Makes It Important For All Policyholders to Shop Around, 08/27/2014. (PDF)
Consumer Alert: WARNING: Your Insurance Company May Raise Your Auto and Home Insurance Rates if They Think You Won’t Shop Around, 04/07/2014. (PDF)

Press Releases

Testimony and Comments

Allstate Press Releases

View more from the Consumer Federation of America. URL not found