Insurance Class Actions Insider -

Auto insurance policies typically provide for insureds to be reimbursed for expenses they incur in assisting in the defense of a lawsuit against the insured that is being defended by the insurer.  Some putative class actions have been brought on the theory that insurers fail to proactively determine whether these expenses are incurred and pay them even where insureds have not  requested reimbursement.  (See my October 11, 2011 post for more on this.)  The Pennsylvania Superior Court (that state’s intermediate appellate court) recently held, in a 2-1 decision, that under the language of Erie Insurance Exchange’s policy, the plaintiff could not recover without making a request for reimbursement. 

In Albert v. Erie Insurance Exchange, 2013 PA Super 59, 2013 Pa. Super. LEXIS 146 (Pa. Super. Ct. Mar. 20, 2013), the auto policy provided that the insurer would pay for “reasonable expenses [of] anyone we protect may incur at our request to help us investigate or defend a claim or suit.  This includes up to $100 a day for actual loss of earnings.”  Id. at *2.  The trial court dismissed the complaint (on the equivalent of a motion to dismiss) based on the plaintiff’s failure to allege that she made any request for reimbursement under this provision.  It appears that the plaintiff did not want to make that allegation because it would be detrimental to pursuit of a class action (perhaps on the assumption that many insureds do not request reimbursement under this coverage, and those who do request it likely receive it).  The Pennsylvania Superior Court majority agreed with the trial court’s reasoning that the insurance policy did not require Erie to advise the plaintiff of the terms of the policy, which the plaintiff had read and signed.  The court also relied upon a policy provision requiring that the insured notify the insurer or insurance agent “when a loss happens,” concluding that this provision required an insured to make a claim for the reimbursement of litigation expenses.  Id. at *8.  The court cited with approval the Ohio Supreme Court’s decision in a similar case, Kincaid v. Erie Ins. Co., 944 N.E.2d 207 (Ohio 2010).  The court also rejected arguments that an obligation to advise the insured about the availability of this coverage was created when Erie hired counsel to defend the insured.  The court found no basis for bad faith liability where there was no duty to inform the insured about the policy provision, and no duty to provide reimbursement without a request.  Id. at *17. 

Judge Colville dissented with respect to the breach of contract and declaratory judgment claims. The dissent would have held that the policy was silent with respect to whether a request for reimbursement is required.  To the extent that the policy required the insured to provide notice to the insurer “[w]hen there is an accident or loss,”  Judge Colville concluded that a claim for reimbursement of litigation expenses would not be an “accident” or a “loss,” and thus this policy provision was inapplicable.

Although appellate courts in both Ohio and Pennsylvania have now ruled in favor of insurers on this issue, insurers may wish to continue to monitor this issue and review their own policy language that governs this type of claim.  There is of course no assurance that a court in some other jurisdiction would not agree with the dissent in this case.  It also might be good business to pay this coverage proactively – it may create goodwill and not cost very much to pay $100 to those insureds who actually testify at a deposition or trial.

Last week, in my August 7, 2012 post, I reported on a new class action filed against AIG/Chartis involving the subrogation “made whole” doctrine.  A recent Florida federal district court decision in a putative class action on this issue is significant because it struck the class allegations on the pleadings.

In Vandenbrink v. State Farm Mutual Automobile Insurance Company, 2012 U.S. Dist. LEXIS 108696 (M.D. Fla. Aug. 3, 2012), the plaintiffs allege that on auto insurance claims, State Farm improperly attempts to recover its PIP and/or Med Pay payments out of insureds’ settlements with third-party tortfeasors, without first determining whether the insured has been made whole.  Judge James S. Moody, Jr. of the Middle District of Florida dismissed the complaint under Iqbal and Twombly because, among other grounds, the plaintiffs had failed to “at least make some showing as to what amount would be required to make Plaintiffs whole and what amount was actually received.”  Id. at *6.  More significantly, Judge Moody granted State Farm’s motion to strike the class allegations on the pleadings because individual issues regarding each particular settlement clearly would predominate:

In this instance, the threshold inquiry is whether the Plaintiffs were made whole through their settlements. If this case were to proceed, the most important issues to settle will be individual in nature. The issues will include the damages incurred by an individual plaintiff, the amount of the settlement, and the portion of the settlement that actually was for medical payments. Since the individual factual inquiry will predominate this litigation, making any sort of class litigation highly impractical, the class allegations will be stricken. Since a class action is inappropriate, only one of the named plaintiffs may proceed in this action. The second, as decided upon by plaintiffs, must file a separate action.

Id. at *9.

This decision should be quite helpful to insurers defending against class actions on this issue.  The increased willingness that some federal and state courts have shown towards motions to strike class allegations saves the parties substantial time and expense in discovery in cases that, based on the pleadings, would not be appropriate for class certification.

Last week there was significant media attention given to a report issued by the Consumer Federation of America regarding the use of software by insurers to adjust bodily injury claims, such as "Colossus," typically under auto insurance policies.  The report explains:

Over the past ten to fifteen years, the payment of bodily injury claims covered by automobile or home and property insurance has evolved from a system based primarily upon the experience and knowledge of claims’ adjusters to a computer-based assessment that has the potential to be easily and broadly manipulated by insurers. This technology has enabled many insurers to increase profits by reducing the amount paid to consumers who file bodily injury liability claims, including uninsured and underinsured motorist claims.  . . . The authors’ primary objective in writing this report is to inform regulators about the technical complexity of this topic and the need to exercise better oversight regarding how these systems can be manipulated to the detriment of consumers.

The report was the subject of articles by Leslie Scism and Erik Holm in the Wall Street Journal, by Denise Johnson in Insurance Journal, by Patricia Sabatini in the Pittsburgh Post-Gazette, by Grant Gross in Computerworld, and others.

Denise Johnson’s article quotes a statement from the National Association of Insurance Commissioners (NAIC) regarding the Consumer Federation report as follows:

“Computerized claims systems are generally geared for use in the claims settlement portion that involves pain and suffering, which can be a very subjective process. It is the expectation of state insurance regulators that these systems are just one factor in helping a claim representative reach a settlement conclusion,” the NAIC said in a statement.

In my view, the Consumer Federation is digging up an old issue that has been litigated fairly extensively for a long time (as they say, this has been going on for ten to fifteen years) and it has largely been resolved because class action cases were dismissed, had certification denied or they settled.  This is certainly nothing new and it’s unclear why it’s being highlighted now.  As I’ve described in posts on this blog over the last year, this is the type of issue on which the Illinois Appellate Court reversed class certification last year, although the Oregon Supreme Court upheld a large verdict after a class action trial.

One thing that seems to be missing from the Consumer Federation report and the articles about it is that, whatever use is made of software in evaluating bodily injury claims, the end result is simply a settlement offer that the claimant or insured are free to accept or reject.  The claimant or insured usually have an attorney if the injury is any kind of serious injury (or even a lot of times when they have a soft-tissue injury of questionable significance).  The software is nothing more than a tool that can be used in negotiations.  There is never any obligation for a claimant or insured to accept any settlement offer.  If insurers make offers that are too low the cases will not settle and juries or judges will decide how much to award.  Insurers also might risk bad faith penalties in some circumstances in some jurisdictions if they fail to make a “fair” offer before trial. 

Litigants who are trying to settle cases never want to pay more than they have to. That is why mediations predictably often start with relatively extreme positions on both sides until finally, at the end of the day, the parties might come close enough to try to reach a settlement.  Even before the advent of this kind of software, insurers could train adjusters to be tough negotiators in settling cases.  Especially when insureds and claimants have aggressive lawyers on their side, I’m not sure what is wrong with an insurer also having a strong negotiator, perhaps aided by computer data regarding prior settlements, on its side.  If insurers enter into settlements that are too generous, it will only lead to higher premiums for everyone.  Maybe I’m missing something, but if similar cases are really settling for less than they used to settle for, the only reason for that seems to be because plaintiffs or their lawyers are becoming weaker negotiators.  That seems hard to believe.

For purposes of avoiding class action exposure on this kind of issue, as I see it the key for an insurer is not to have hard-and-fast immutable rules about how software is used and what kinds of offers are made by claim professionals, but rather to use software as a potential aid where appropriate, together with the claim professional’s experience, education, discretion and guidance from a supervisor, and keeping mind applicable law in the jurisdiction, in order to try to reach a settlement.

After Wal-Mart v. Dukes, plaintiffs’ lawyers tend to file more narrowly-tailored, single state class actions, often focusing on insurers’ compliance with state statutes or regulations.  Recent filings against GEICO and Progressive, two of the country’s largest auto insurers, are good examples of this trend: 

• Davis v GEICO Casualty Company.pdf, Case No. 2012CA005024 (Florida Circuit Court, 15^th Judicial Circuit in and for Palm Beach County, filed Mar. 22, 2012):  This case focuses on a Florida statute requiring insurers to make disclosures regarding UM/UIM coverage, and obtain a signed written consent if an insured chooses not to buy UM/UIM coverage or chooses limits lower than the bodily injury coverage limits.  See Fla. Stat. § 627.727.  After a few entertaining paragraphs about the cavemen, the gecko and “15 minutes can save 15%,” the complaint alleges that GEICO fails to comply with this Florida statute by failing to provide Florida consumers who buy their policies by phone or over the Internet in 15 minutes with the required disclosures, and failing to obtain their signatures. The complaint vaguely pleads that GEICO’s electronic signature process for Internet sales fails to comply with Florida law with respect to electronic signatures.  The complaint seeks, among other relief, reformation of the policies of putative class members such that they provide UM/UIM limits equal to the bodily injury coverage limits.  This is not unique to Florida – some other states have similar requirements for declining UM/UIM coverage or buying lower limits for such coverage. 
• Beavers v Progressive Casualty Company.pdf, No. CV 12 779206 (Ohio Court of Common Pleas, Cuyahoga County, filed Mar. 28, 2012):  This case focuses on Ohio regulations requiring that, when an insurer pays a claim for a total loss of a vehicle, it must provide notice to the claimant that sales tax will be reimbursed if a replacement vehicle is purchased within 30 days, and the insurer must provide such reimbursement if a claim for sales tax is timely submitted.  See Ohio Admin. Code § 3901-1-54(H)(7).  The complaint alleges that Progressive provides a notice that states only that “we will include applicable sales taxes and fees when required by law.”  (Complaint, ¶ 10.)  The plaintiff claims that this notice is insufficient “because it fails to provide the claimant with any information regarding when taxes and fees are ‘required by law.’”  (Id., ¶ 11.) This seems to go against the basic legal principle that people are presumed to know the law, although in reality few average citizens take the time to explore the intricacies of insurance department regulations.

Other insurers may want to pay attention to these filings because it’s common for plaintiffs’ lawyers to file class actions first against the insurers with the largest market shares and then, if they have some success, follow with suits against carriers with smaller market shares.  (Although sometimes this happens the other way around, perhaps because plaintiffs’ lawyers think the smaller insurers will have less familiarity with class actions and hire less qualified defense counsel, and will try to obtain favorable rulings they can then use against the larger carriers.)  

Insurers seeking to avoid class action exposure in the post-Wal-Mart era would be well-served to devote resources to beefing up their compliance department.  Careful compliance with state statutes and regulations should help avoid being sued in some of these types of lawsuits.

There have been an unusually large number of significant decisions in insurance class actions over the last couple of weeks.  I will not be able to discuss all of them in detail but thought you would like to have shorter summaries of them.  Here is the first installment regarding recent P&C class action decisions: 

• Folks v. State Farm Mut. Auto. Ins. Co., 2012 U.S. Dist. LEXIS 43294 (D. Colo. Mar. 29, 2012):  This case involves personal injury protection (PIP) coverage and a requirement in Colorado’s No Fault Act that insurers must offer enhanced PIP coverage that includes coverage for pedestrians.  State Farm failed to offer the enhanced benefits and, after losing some prior court decisions, modified its policy language.  State Farm also provided voluntary relief, re-adjusting a number of claims.  The court denied class certification, finding that numerosity was not satisfied because, excluding the people who were given voluntary relief, the plaintiff could only speculate that there were about 50 class members that might obtain relief in the suit.  The court also concluded that a reformation claim was not suitable for certification under Rule 23(b)(2) because reformation is an equitable remedy that requires an individualized analysis of whether the class member relied on State Farm’s past practices and the degree of injustice or hardship.  To the extent certification was sought under Rule 23(b)(3), predominance also was not satisfied because the prior litigation, together with State Farm’s voluntary relief program, had resolved the common legal and liability issues, and what remained were only individualized determinations regarding reformation of policies and the amounts claimed by individual class members.  This case is a good example of how a voluntary relief program can potentially defeat a putative class action (for more on that, see my Sept. 16, 2011 blog post). 
• Cox v Allstate Insurance Company.pdf, No. CIV-07-1449-F, slip op. (W.D. Okla. Mar. 28, 2012):  This is a class certification decision in a first-party property insurance case involving claims for wildfire damage.  The two proposed Oklahoma statewide classes were: (1) insureds who had recoverable depreciation held back from their claim and did not receive a letter from Allstate explaining the amount of depreciation and how to recover it; and (2) insureds whose policy limits were increased automatically by Allstate’s “Property Insurance Adjustment” feature.  The claim on behalf of the first class was that the failure to send a letter regarding the holdback was by itself bad faith and resulted in unjust enrichment to Allstate.  The claim on behalf of the second class was that Allstate’s method of increasing policy limits resulted in overinsurance and unduly high premiums, also resulting in unjust enrichment.  The court held that commonality was not satisfied under Wal-Mart v. Dukes as to proposed class (1) because Allstate’s general policy was to send the letter, and, as the plaintiffs’ expert admitted, some people who did not receive the letter recovered all depreciation they were owed.  Determining whether people failed to receive the letter and sustained an injury would require a file-by-file individualized analysis.  The court reached a similar conclusion with respect to the class of insureds claimed to be overinsured – a file-by-file analysis would be required to determine if there was overinsurance.  The court also found that predominance was not satisfied because, with respect to class (1), a bad faith claim could not be pursued without a breach of contract, and, under Wal-Mart, Allstate would be entitled to present its individualized defenses to breach of contract claims.  The court rejected an argument that Allstate was somehow estopped from asserting defenses in litigation that it had not asserted during the claim process.  With respect to class (2), a determination of whether Allstate inappropriately raised the policy limit would require comparing every individual policy limit with the property’s fair market value, an individualized analysis that defeated certification.  This case is a good example of how courts are applying Wal-Mart v. Dukes in insurance class actions.

Do insurance companies charge premiums for coverage that can never be triggered?  That is the essential allegation in Keeling v. Esurance Ins. Co., 2012 U.S. Dist. LEXIS 26998 (S.D. Ill. Mar. 1, 2012).  In my October 4, 2011 blog post, I wrote about a Seventh Circuit decision finding federal jurisdiction in this case, based on the possibility of punitive damages pushing the amount in controversy over $5 million.  After jurisdiction was established, Esurance challenged the complaint in a motion to dismiss.  The motion to dismiss was denied (except for dismissal of a fraudulent misrepresentation claim). 

The plaintiff claimed that underinsured motorist (UIM) coverage of $20,000/$40,000 was illusory under Esurance’s policies because it would never be paid.  The plaintiffs focused on the following provision in Esurance’s policies:

“Underinsured motor vehicle” means a land motor vehicle or trailer of any type to which a bodily injury liability bond or policy applies at the time of the accident but its limit for bodily injury liability is less than the limit of liability for this coverage.

However, “underinsured motor vehicle” does not include any vehicle or equipment:

1. To which a bodily injury liability bond or policy applies at the time of the accident but its limit for bodily injury liability is less than the minimum limit for bodily injury liability specified by the financial responsibility law of Illinois.

Id. at 7-8 (emphasis added).  The class was defined as insureds who bought UIM coverage from Esurance with limits of $20,000/$40,000 in Illinois.  The claim was that this coverage was worthless to insureds because the minimum required coverage under Illinois law is $20,000/$40,000.  Thus, the plaintiff asserted that if an Esurance policyholder bought UIM coverage of $20,000/$40,000, the only way coverage would apply  under the definition of “underinsured motor vehicle” would be if the underinsured motorist had coverage less than Illinois law required, which was unlikely unless the driver was from out of state.  But if the underinsured motorist had lower limits than what Illinois law requires ($20,000/$40,000), then the exclusion (paragraph “1” above) would bar coverage, and therefore there would never be coverage. 

Esurance argued that coverage should be based on whether the underinsured motorist or his or her insurer pays less than the limits, rather than what the policy limits are, but the court rejected that position, finding it inconsistent with the terms of the policy and applicable Illinois law.  The court also rejected Esurance’s argument based on the filed rate doctrine because the case was not a challenge to premium rates but rather a challenge to the illusory nature of the coverage.

One problem insurance companies sometimes experience, which can lead to these kinds of class actions, is that when policy forms are written or revised that is done by the underwriting department without involvement of the claims department and without involvement of any lawyers who are familiar with the kinds of issues raised in coverage litigation and class actions.  The issue presented by this case seems like precisely the type of issue that could be flagged by a company that is proactive in attempting to identify problems that might lead to class actions.  (See my blog post about Rob Herrington’s book, “Verdict for the Defense,” for more on that.)

In my view, in defending a case raising the kind of issue that Keeling does, insurers should not file motions to dismiss reflexively.  Too often these motions are filed because that is the standard playbook, or to avoid burdensome discovery if the judge will stay discovery while the motion is decided, without thinking about the consequences of losing the motion.  You do not want a judge to rule against you early, as a matter of law, on an issue of contract interpretation that is at the very heart of the case.  (A plaintiff cannot seek such a ruling before class certification because of the one-way intervention rule that applies in class actions.)  You may be better off defending against class certification and not risking an early adverse ruling on a contract interpretation issue that is a close call or, even worse, on which you can identify substantial weaknesses in your own position.  Another common strategy for filing a motion to dismiss is to whittle down some of the causes of action before discovery and class certification.  But unless the discovery can be separated by causes of action, keeping some additional causes of action in the case until class certification also can sometimes help the defense case by broadening the issues on which individualized adjudication is necessary, where narrowing them with a motion to dismiss might make the case easier to certify.  All of this should be considered before the trigger is pulled on a motion to dismiss.

Numerous class actions have been brought against auto insurers on the theory that they should be required to pay under collision coverage not only for the cost of repairing damage to a vehicle but also for the diminished value that a vehicle might sustain because it was in an accident.  This claim for diminished value might be because of damage that is unrepairable (typically weakened metal in the framing) or based on the “stigma” that used car buyers might place on a vehicle that was in a serious accident (think of those “Carfax” advertisements), or both.  The issue of diminished value has also arisen prominently in property insurance recently – the Eleventh Circuit certified to the Georgia Supreme Court the question of whether a property policy provides coverage for alleged diminished value of a building after property damage has been repaired.  See Royal Capital Development, LLC v. Maryland Casualty Company, 659 F.3d 1050 (11th Cir. 2011).  The insured in that case is making a novel argument trying to extend the Georgia Supreme Court precedent on diminished value in auto insurance to the property insurance context.  If the Georgia Supreme Court rules in favor of the insured in that case, we might see a rash of new lawsuits, potentially including class actions, on that issue.  This is an issue that the insurance industry should continue to pay careful attention to.

In Moeller v. Farmers Insurance Company of Washington, 2011 Wash. LEXIS 957 (Wash. Dec. 22, 2011), the Supreme Court of Washington recently held, in a 5-4 decision, that Farmers’ auto insurance policies provided coverage for diminished value.  The majority also affirmed the trial court’s certification of a class.  While this is an outlying decision that expressly rejected the majority view among other courts around the country in these cases, insurers may wish to take another look at their policy language, particularly in Washington but also in other states in which this issue has not yet been conclusively resolved, in order to try to avoid the impact of this decision and a few others like it. 

Coverage

In Moeller, the policy’s grant of coverage for collision provided that “[w]e will pay for loss to your Insured car caused by collision less any applicable deductibles.”  Id. at *2.  The term “loss” was defined as “direct and accidental loss of or damage to your Insured car, including its equipment.”  Id.  The policy also included “Limits of Liability” and “Payment of Loss” clauses providing as follows:

Limits of Liability

Our limits of liability for loss shall not exceed:

1. The amount which it would cost to repair or replace damaged or stolen property with other of like kind and quality, or with new property less an adjustment for physical deterioration and/or depreciation.

. . . .

Payment of Loss

We may pay the loss in money or repair or replace damaged or stolen property.

Id. 

Farmers argued that these two provisions unambiguously barred coverage for diminished value.  It argued that under the policy it had the option to repair the damaged vehicle, and if it chose that option that would fully satisfy its contractual obligation.  The majority rejected this argument, concluding that an average consumer reading the policy would expect to be paid for diminished value in addition to the cost of repairs:

We must read an insurance contract as an average person would read it.  Thus, the lens through which we view this question is from the point of view of the consumer. From this point of view, the bargain of the contract is to return the consumer to his preaccident position with respect to the value of his car. Strictly construing the limiting language of Farmers’ policy, as we must, it does not convey to the average policyholder that the value of coverage may be less if Farmers repairs a vehicle rather than replacing or "totaling" it. Rather, the reasonable expectation is that, following repairs, the insured will be in the same position he or she enjoyed before the accident.

Id. at *13 (citation omitted; emphasis added).

Chief Justice Madsen wrote a strong dissent, joined by three other justices.  She wrote that the majority’s interpretation was “a contrived interpretation of clearly worded provisions,” that the majority’s construction was “untenable under language that gives Farmers the right to repair or replace, at its option,” and that it “creates a new obligation with no basis whatsoever in the contract.”  Id. at *31-32 (Madsen, C.J., dissenting).  The chief justice further explained that the “reasonable expectations” doctrine referenced by the majority had not previously been adopted in Washington, and it does not apply unless there is an ambiguity.  Id. at *34-35. 

When I try to take off my defense lawyer hat and look at this objectively, I really think the dissenters have the better side of the argument here.  If you give this policy language to a nonlawyer and ask them what they think it means, it seems doubtful that people would expect the insurance company to pay not only to repair the physical damage to a vehicle but also for intangible diminution in value.  I think it’s similar to the flood exclusion issue that arose in Hurricane Katrina, and which occupied a lot of my time.  When I explained to nonlawyers that the plaintiffs’ lawyers were arguing that the flood exclusion did not apply to what occurred in New Orleans on the grounds that there were man-made causes (in addition to the hurricane), the reaction universally was that the argument seemed ridiculous.  Some people commented that only lawyers would argue about something like that.  Eventually, after years of litigation, the Louisiana Supreme Court unanimously rejected that argument.

Class Certification

On class certification, the majority in Moeller found certification appropriate.  It started with a principle that is apparently applied in Washington but contrary to the law in the federal courts and most other places – that “the trial court should err in favor of certifying the class.”  Id. at *19.  The central issue in dispute was whether the plaintiff could prove liability and damages for the putative class as a whole through mathematical modeling, without showing that each member of the putative class actually sustained a diminution in value loss, or the amount thereof.  The majority seemed to sidestep that issue, rejecting an argument that the plaintiff had made a concession that some class members had no loss, and focusing on the abuse of discretion standard.  The court noted that the trial court had conducted a lengthy evidentiary hearing and that the result was not “unreasonable or untenable.”  Id. at *22-23.

A dissenting opinion by Justice Alexander, joined by Justice Pro Tem Seinfeld, took the class certification issue head on and found certification improper.  The key points made by Justice Alexander were:

In my view, the trial court abused its discretion when it found that common issues predominate over individual ones and that a class action is superior to other methods of adjudicating the disputes. My concern is with the plaintiffs’ intended method of arriving at the measure of damages. Moeller does not plan to prove his claim of breach of contract and resulting injury with evidence of the preaccident and postrepair values of class members' cars. Instead, Moeller intends to use a statistical methodology and data from car auction sales to prove that, on average, cars that are "wrecked and repaired" sell for lower prices than do cars that are "unwrecked" and therefore, as a statistical matter, diminished value exists. Moeller then plans to categorize and quantify the alleged average decreases in value associated with types or amounts of damage, multiply each alleged average amount by the number of class members in each damage category, and then tally the numbers to provide a class-wide “damage[s] estimate.”

The proposed measure of damages should not be permitted on a class-wide basis because individualized proofs of the preaccident and postrepair values of each damaged car should be required to ascertain which of the thousands of class members actually suffered damage caused by Farmers’ failure to tender a diminished value payment. Additionally, Moeller should be required to prove how much damage each individual class member sustained. The problem with the proposed method of calculating damages is that there is no way to ensure that each car included in the model actually sustained diminished value.

. . .

Additionally, the trial plan effectively converts the damages element of Moeller's claim into an affirmative defense. This impermissibly shifts the burden to prove damages from the plaintiffs to the defendant. This offends due process.

Id. at *43-46 (Alexander, J., dissenting) (emphasis added; citations omitted).

Again, trying to look at this objectively I think the dissent’s view is more consistent with fundamental principles of class action law.  While the dissent describes this as a damages issue, it’s really a question of whether liability for breach of contract (not merely damages) can be shown on a classwide basis or requires an individual analysis.  The dissent does not cite the U.S. Supreme Court’s opinion in Wal-Mart Stores, Inc. v. Dukes, 131 S. Ct. 2541 (2011), but it could have – the Supreme Court unanimously disapproved a similar statistical sampling methodology as a “novel project” that would improperly allow Rule 23 to affect substantive rights, and deprive Wal-Mart of its individual defenses.  That part of Wal-Mart might be a due process requirement applicable to state courts, and in any event most states do not allow their procedural rules to alter substantive rights.  But unless and until the Supreme Court takes a class action case on appeal from a state court and decides what due process boundaries there are, we will continue to see some state supreme courts take approaches to class actions that diverge substantially from federal law.

Several notable recent class action filings against insurers have come across my desk (or computer screen) and seem worthy of interest to readers of this blog.  I will summarize and comment briefly on them.  If you’d like a copy of any of the complaints, just e-mail me.

• Use of Staff Counsel:  In Golden v. State Farm Mutual Automobile Insurance Company, Cause No. 02D01-1110-PL-363 (Indiana Superior Court, Allen County; removed to federal court), the plaintiff alleges that State Farm improperly fails to disclose to its insureds that it may use staff counsel to represent them in defending lawsuits under liability insurance coverage.  There are two proposed classes:  (1) insureds in Indiana that have purchased or renewed a policy with State Farm within the last two years, containing liability coverage; and (2) insureds in Indiana who were represented by State Farm staff counsel within the last two years.  The causes of action are breach of an alleged duty to disclose the use of staff counsel (at the time of policy issuance or renewal), breach of a duty of good faith and fair dealing by not disclosing the use of staff counsel at the time of policy issuance or renewal, unjust enrichment, and injunctive relief barring State Farm from continuing to issue or renew policies without disclosures regarding staff counsel, and barring State Farm from assigning staff counsel to represent insureds where no prior disclosure was made.  It will be interesting to see if this complaint survives a motion to dismiss.  There may not be any legal duty to inform insureds about the use of staff counsel absent a statute or regulation requiring it.  It also seems unclear that anyone is injured by a failure to disclose at the time the policy is issued or renewed, particularly if they have never been sued.  The complaint seems to suggest that the use of staff counsel is somehow a new or unusual practice not followed by other insurers.  I’m not sure what the practice has been in Indiana, but as far as I know all of the major insurers have been using staff counsel to defend in the vast majority of jurisdictions for some time (except for a few jurisdictions where use of staff counsel is prohibited).  The complaint also seems to suggest, without articulating any basis, that staff counsel is somehow inferior to private outside counsel.  On the other hand, there is probably no harm in disclosing the use of staff counsel, and some insurers probably are doing that.  It’s hard to imagine that people buy their auto or homeowners’ policies based on whether the insurer is going to use staff counsel in defending them in a lawsuit.
• Wildfire Claims:  In Abed v. Allstate Ins. Co., Case No. BC 473460 (Cal. Super. Ct., Los Angeles County), the named plaintiffs assert a variety of claims against Allstate arising from their claim for smoke damage to their house from the “Station Fire” in Southern California in August of 2009.  They assert various individual claims but only one cause of action on behalf of a putative class, alleging that Allstate’s policies failed to comply with California law on appraisal, and the efficient proximate cause doctrine.  The appraisal-related claim focuses on a provision in the California standard fire insurance policy providing that “In the event of a government-declared disaster, as defined in the Government Code, appraisal may be requested by either the insured or this company but shall not be compelled.”  Cal. Ins. Code § 2071.  The plaintiffs assert that the “Station Fire” was a “government-declared disaster” within the meaning of this provision.  They claim that Allstate improperly sought to compel a mandatory appraisal, and the appraisal clause in its policy failed to include this sentence.  The efficient proximate cause claim is a bit difficult to discern from the complaint.  That doctrine applies where a loss has more than one cause, and it appears that the claims at issue here were attributable only to the wildfire.  There is no suggestion that I can identify of another cause.  On the appraisal issue, although I think it involves the kind of individual issues that would not be appropriate for class treatment, insurers may want to check into their practices in California with respect to appraisal of claims for government-declared disasters given the unusual statutory language.
• Depreciation on Auto Claims:  In Silvin v. Geico General Insurance Company, Case No. 1:11-cv-24128-CMA (S.D. Fla.), the plaintiff seeks to certify a nationwide class on the question of whether a particular Geico policy form allows for deduction of “betterment” or depreciation on auto claims under either comprehensive or collision coverage.  The policy language that is quoted in the complaint does not appear to make any reference to a deduction for “betterment” or depreciation.  It will be interesting to see what happens with this case.  This also seems like an area in which insurers may want to check what their policies say and what their practices are.

The recent denial of class certification in Fosmire v. Progressive Max Ins. Co., 2011 U.S. Dist. LEXIS 117366 (W.D. Wash. Oct. 11, 2011) is the second opinion I’ve seen post-Wal-Mart that applies the new standards in detail in an insurance class action.  This putative nationwide class action alleged that Progressive improperly failed to pay for diminution in value on auto claims under uninsured/underinsured (UM/UIM) coverage.  That is, where a vehicle has been damaged in an accident and repaired, it may not be fully repairable, or it may lose some value because a repaired vehicle may be worth less than one that has never been in an accident.  (It seems that those “Carfax” reports make a difference to used car buyers.)  The plaintiff asserted that Progressive had what she labeled as a “don’t ask, don’t tell” policy, under which the company would not advise insureds of the potential availability of a payment for diminution in value, and would consider such claims only where the insured made a claim for it.  Id. at *4-5.

There are several notable aspects to this opinion:

1. The court excluded the plaintiff’s expert under Daubert.  The court found the more “relaxed” standard adopted by the Eighth Circuit in In re Zurn Pex Plumbing Prods. Liab. Litig., 644 F.3d 604 (8th Cir. 2011) (see my prior blog post) to be the correct standard, rather than the full Daubert analysis that has been required by the Seventh and Eleventh Circuits.  This “relaxed” standard does not decide on admissibility at trial of the expert’s testimony, but rather focuses on its reliability for purposes of class certification issues and limits the focus to the evidence available to the expert at the class certification stage, in light of the discovery taken to date.  Here, the plaintiff’s expert (Polissar) proposed to calculate classwide damages using a 10-year-old study conducted by a different expert (Siskin).  The court found this testimony inadmissible for class certification purposes because Polissar did nothing to ascertain the reliability of Siskin’s work, had never looked at Progressive’s data produced in discovery, had no idea what the proposed class definition was, and had made no attempt to determine how the 10-year-old study by Siskin would apply to the proposed class in the case at bar.  This is a classic case for exclusion of expert testimony under Daubert.  Despite the court’s characterization of its analysis as application of a “relaxed” standard, in the end I don’t think that mattered much – this was a thorough vetting of the reliability and admissibility of the testimony. 
2. The named plaintiff was atypical because of a defense raised regarding a material misrepresentation on the insurance application.  Progressive raised a defense based on the fact that, when she purchased the policy, the plaintiff failed to disclose that her fiancé was a driver in her household who would drive the car. He was driving when the accident occurred.  Although the plaintiff argued that she did not make a misrepresentation, the court was “concerned that litigation concerning this defense will preoccupy Ms. Fosmire to the detriment of class claims” and “it threatens to undermine her credibility at trial, which also undermines the element of typicality.”  Id. at *24.  This is a good reminder for insurers and their counsel to fully explore any potential defenses to a named plaintiff’s claim.  Even if you are not going to win summary judgment on a defense, it may be grounds to defeat certification. 
3. Adequacy of representation was not satisfied due to claim splitting.  The court explained that some jurisdictions allow a recovery for “stigma” damages, on the theory that the value of a vehicle has declined because a repaired vehicle is worth less to buyers than one that has never been in an accident (my “Carfax” example above).  Other jurisdictions, like Washington State, do not allow this type of recovery but allow a recovery on the theory that a particular vehicle cannot be fully restored to its pre-loss condition (e.g., because of weakened metal).  The plaintiff here did not pursue “stigma” damages, apparently because she thought that would improve her chances on class certification.  Putative class members who wanted to pursue “stigma” claims in jurisdictions that allowed such claims likely would not be able to do so if the class were certified.  The court found that this created a conflict of interest between the named plaintiff and the putative class, and therefore adequacy of representation was lacking.  This is a good example of how, in defending these cases, you need to put yourselves in the shoes of the putative class members and think about what kinds of claims they might want to bring but that the named plaintiff chose not to pursue.  This is somewhat counterintuitive because in most cases your focus as defense counsel or in-house counsel is on aggressively defending the claims that were brought, not thinking about what additional arguments a plaintiffs’ lawyer might have asserted.  
4. Predominance was not satisfied due to differences in state law.  Because only one jurisdiction had expressly ruled on the availability of diminution in value damages under UM/UIM policies, the court found that the need to decide numerous unsettled issues of state law demonstrated a lack of predominance.  The court explained that “determining whether each [UM/UIM] statute requires diminution damages would require an evaluation of the interplay between the statutory language and the tort law in each state.”  Id. at *30.  Despite the plaintiff’s argument that the policy language was essentially the same in all applicable jurisdictions and that breach of contract law is consistent, the court found that given the nuances that would need to be delved into for each jurisdiction, the plaintiff had not met her burden of demonstrating a method for addressing the differences in state law.  (See my prior post on the topic of differences in state law on breach of contract.)
5. Rule 23(b)(2) certification was unavailable because of a lack of cohesiveness and because monetary damages were not incidental.  The court found a lack of cohesiveness precluding certification under 23(b)(2) for essentially the same reasons that certification was denied under (b)(3).  It also concluded that, under Wal-Mart, the damages sought could not be determined in an objective, classwide manner and thus were not “incidental.”

In my view, this is the type of case that was unlikely to meet the standard for class certification even before Wal-Mart.  It demonstrates though that the bar has been raised and that there are additional lines of argument available to insurers and other class action defendants.

In recent years there has been a significant amount of class action litigation in various jurisdictions regarding labor rates for repairs on auto claims.  The California Court of Appeal, Second Appellate District, recently affirmed a denial of class certification in one of these cases, focusing on the fact that the insurer handled each claim in a case-by-case manner, negotiating the labor rate where appropriate.

In Holzman v. Farmers Insurance Exchange, 2011 Cal. App. Unpub. LEXIS 7262 (Cal. Ct. App. Sept. 26, 2011), the plaintiff, a lawyer, had his Porsche 911 repaired at a Porsche dealer that charged $135 per hour.  Farmers concluded that the market rate in the relevant geographic area was $65 per hour, but agreed to $85 per hour for this repair.  Farmers refused, however, to pay $135 per hour.  The plaintiff sued for the difference and sought to represent a putative class of luxury vehicle owners who paid out-of-pocket for repairs in excess of the deductible.  Id. at *2-7.

The trial court denied class certification.  The Court of Appeal affirmed based on a lack of predominance.  The court focused on the need for a case-by-case determination of reasonableness, in light of the fact that Farmers negotiated the labor rate where appropriate:

Farmers's practice of using the predominant market labor rate does not cause a member of the putative class to incur damages unless he or she is forced to pay out-of-pocket expenses that are not required by the insurance policy. Whether an insured incurs out-of-pocket expenses requires an individualized, case-by-case analysis. In some cases, Farmers negotiates a rate higher than the predominant market labor rate. The court must determine for each individual whether Farmers agreed to a higher labor rate and, if so, whether that rate was reasonable, or whether it was still so low that it violates the insurance policy, Insurance Code and/or applicable regulations. Accordingly . . . even if Farmers's use of the predominant market labor rate were an improper claims practice, class certification is unwarranted because common questions of fact and law are not predominant.  (Id. at *27.)

As I’ve noted before on this blog, one thing insurance companies can do to reduce their class action exposure is to give front-line adjusters discretion.  While sometimes there are business reasons for a bright-line rule, or it might seem easier for adjusters to follow, when discretion is given on this kind of issue it increases the chances of defeating class certification.  If Farmers had taken a bright-line position that it would never pay more than $65 per hour in this geographic area for auto repairs, regardless of whether the vehicle is a Hyundai or a Porsche, it would have been in a more difficult position in defending against class certification.  Where adjusters have discretion to make determinations on a case-by-case basis, courts often must do the same, and that often makes class certification improper.

I recently came across a new class action filing in Albert v. Erie Insurance Exchange, in the Court of Common Pleas of Philadelphia County, Pennsylvania.  (I don’t have a link to the complaint online, but e-mail me if you would like a copy.)  The complaint focuses on a provision in auto policies that provides for reimbursement of up to $100 per day for loss of earnings where an insured is required to take time off from work to assist in the defense of a claim or suit (such as attending a deposition or trial, or preparing for such testimony).  The allegations are that the insurer does not make insureds aware of this provision (although it is in the policy, which insureds can read) and the insurer does not make a proactive effort to find out if insureds are incurring lost wages so that reimbursement can be made.  I’m not sure this has much legs given the fact that insureds generally are expected to read their policies, but I thought I would advise readers of this new filing since I have not seen other filings on this issue.

Class action trials are so rare that there is little guidance in court opinions on how these cases should be tried, other than hypothetical discussions in class certification decisions regarding how trials might be conducted.  In Strawn v. Farmers Insurance Company of Oregon, recently reported in Legal Newsline and the Soha & Lang Coverage Blawg, the Oregon Supreme Court recently decided an appeal from a judgment after a trial in an insurance class action.  Reading between the lines of the opinion teaches a few lessons about trying these cases.

The plaintiff claimed that Farmers improperly reduced benefit payments for personal injury protection (PIP) coverage under auto insurance policies, by using “cost containment software.”  This software analyzed medical expenses in comparison with a database of charges for particular services in the region, and allowed Farmers to select a particular percentile (it chose the 80th percentile) as a cutoff for what it would pay for a particular service.  The claim was that this process was arbitrary and a breach of the contractual obligation to pay “reasonable and necessary” medical expenses.  (The issue presented in this case is similar to one raised in a recent class action filing against Nationwide that I posted about.)  A class was certified and the case was tried to a jury.  A judgment in favor of the class was entered for approximately $900,000 in compensatory damages and $8 million in punitive damages.  The court of appeals found that the punitive award exceeded constitutional limits.   

The issues that Farmers raised before the Oregon Supreme Court were: (a) that it was improperly precluded from presenting individualized evidence to show that class members’ medical expenses were unreasonable; and (b) that plaintiffs failed to demonstrate class-wide reliance.  The Oregon Supreme Court rejected Farmers’ arguments on appeal, finding that the trial court had not actually barred individualized evidence, and that reliance could be inferred from the purchase of the auto policy where the terms of PIP coverage were statutorily mandated.  The supreme court agreed with the plaintiff that the constitutional issue should not have been reached by the court of appeals because no appeal was taken to the court of appeals from the trial court’s waiver ruling.  The trial court had ruled that Farmers had waived the constitutional issue by not asking for a jury instruction limiting the amount that could be awarded, and not asking for the jury to be sent back for re-deliberation after it made its award, and Farmers did not appeal the waiver ruling to the court of appeals.

The key lesson I see here is that a defendant may be better off making every attempt to try a class action as if it were a trial of thousands of individual cases.  Even if the plaintiff has the burden of proof and you think they cannot meet that burden with aggregate, class-wide evidence, consider putting on your own individualized evidence.  Ideally the trial court will recognize how individualized evidence is necessary and relevant, and decertify the class.   If the trial court does not allow such individualized evidence, this may well preserve good arguments (including due process arguments) for appeal.  Sometimes defendants may think that because they lost on class certification, the trial must be limited to only the named plaintiff’s claim or that evidence can only be presented in a global, aggregated fashion.  Not necessarily.  There are really no “rules” on how class actions must be tried.  Here, the court seems to suggest that Farmers should have tried to introduce individual evidence on how it handled particular PIP claims where it had a good defense.  Similarly, on the reliance issue, Farmers might have been able to develop evidence showing that some class members did not rely on what was in the policy (the Oregon Supreme Court disavows making any presumption of reliance).  This may have required depositions of a sample of putative class members in order to work up that issue for trial.  Thinking outside the box and almost disregarding the certification of a class potentially can be a good strategy in trying an insurance class action.  

The Illinois Appellate Court recently issued a decision reversing a class certification in Bemis v. Safeco Insurance Company, which involved medical payments under auto insurance policies.  The case was filed by a chiropractor who claimed that Safeco used computer software to improperly reduce the cost of services, in breach of the contract requirement to pay "the usual and customary charges incurred for reasonable and necessary medical *** expenses because of bodily injury caused by an accident."

The court held that the trial court abused its discretion in certifying a class on this issue, and its decision highlighted the problem with certification in many proposed insurance class actions:

Bemis argues that the predominant issue in this case is whether the computer database utilized by Safeco is inaccurate and whether the use of the database justifies a limitation of payment under the medical payments provision of the policy. However, if Bemis is successful in proving this, it does not follow that all the other class members submitted usual and customary charges representing reasonable and necessary medical expenses. Whether the usual and customary charge for a class member's reasonable and necessary medical expenses went unpaid is not a common question but is an individual question that will have to be answered for each claimant. Because proof that the usual and customary charge for one class member's reasonable and necessary medical expenses went unpaid would not establish a right of recovery for any other class member, common issues do not predominate.

This decision is particularly useful to insurers because insurance class actions commonly allege that the use of computer software in some respect in analyzing claims was improper, and was done across-the-board in a manner that could be appropriate for class treatment.  Some of those cases have been successful.

The concurring judge, however, suggested that perhaps the case could be appropriate for class treatment with certain expert testimony:

In this case, Safeco used expert testimony to determine the amount it paid to Bemis and the other potential class members. Its decision to reimburse according to "UCR 80th" percentile data was not arrived at by an individual analysis of each class member's bill. On remand, Bemis should be allowed to make a showing that he can present expert testimony, based on data relied on by experts in the field, that would establish that Safeco did not pay the reasonable or usual and customary charges for treatments he and the other class members provided to their patients. Bemis should be allowed to prove reasonableness in the same fashion Safeco determined reasonableness.

The concurrence highlights another important lesson for insurers -- be careful about how you frame your expert testimony, because what is good for the goose may be good for the gander.