Artie's Auto Body, Inc. v. Hartford Fire Insurance

947 A.2d 320 (2008) 287 Conn. 208

ARTIE'S AUTO BODY, INC., et al.
v.
The HARTFORD FIRE INSURANCE COMPANY.

No. 17797.

Supreme Court of Connecticut.

Argued November 29, 2007. Decided June 3, 2008.

Linda L. Morkan, with whom, on the brief, were Craig A. Raabe, Edward J. Heath and Andrea Donovan Napp, Hartford, for the appellant (defendant).

David A. Slossberg, with whom were Allison Near, Milford, Alan Neigher, Westport, Robert J. Berg, pro hac vice, and Ronald J. Aranoff, pro hac vice, New York, NY, for the appellees (plaintiffs).

NORCOTT, PALMER, VERTEFEUILLE, ZARELLA and SULLIVAN, Js.

ZARELLA, J.

The defendant, The Hartford Fire Insurance Company (The Hartford), appeals from the ruling of the trial court granting the motion of the plaintiffs, three Connecticut auto body repair shops and a trade association of Connecticut auto body repair shops,^[1] for class certification. The plaintiffs, who seek money damages and injunctive relief, allege that The Hartford engaged in a pattern of unfair and deceptive acts and practices in violation of the Connecticut Unfair Trade Practices Act (CUTPA),^[2] General Statutes § 42-110a et seq. (counts one and two), and was unjustly enriched as a result thereof (count three). On appeal,^[3] The Hartford claims that the trial court abused its discretion in granting the plaintiffs' motion for class certification because common issues of law or fact do not predominate over questions affecting individual members with respect to proof of The Hartford's liability under CUTPA. We disagree and, accordingly, affirm the trial court's order granting class certification.

The following facts and procedural history are relevant to our resolution of this appeal. On July 16, 2003, the plaintiffs filed a complaint against The Hartford on behalf of themselves and all other persons and entities licensed to perform auto body repairs in Connecticut^[4] who had been substantially harmed by The Hartford's practices. The complaint alleged unfair and deceptive acts or practices on the following grounds. First, The Hartford improperly steered insureds^[5] to a closed network of preferred auto body repair shops that charged labor rates well below reasonable market value. The shops, which The Hartford describes as direct repair program shops (preferred shops), have a contractual relationship with The Hartford to repair damaged automobiles that The Hartford refers to them. Second, The Hartford improperly established an artificially low standard or prevailing hourly rate for reimbursement to shops that were not in the network of preferred shops (nonpreferred shops), including those of the plaintiffs and members of the plaintiffs' putative class. Third, The Hartford provided positive and negative incentives to purportedly independent insurance appraisers to encourage or pressure them into accepting monetary and other limits proposed by The Hartford.^[6] Moreover, the appraisers acceded to these incentives and regularly conformed to the proposed limits in their appraisals. The appraisers also advised and requested insureds to direct their business to the preferred shops and away from the plaintiffs and other members of the putative class. The plaintiffs thus claimed that they had lost business and were forced to charge below market labor rates that had been set by The Hartford's appraisers and charged by the preferred shops.

On August 15, 2005, the plaintiffs filed a motion for class certification. The parties briefed the issue, and the trial court heard oral argument in March, 2006. On August 30, 2006, the court determined that the requirements for class certification had been met and granted the motion. After noting that the plaintiffs had not provided a consistent definition of the proposed class during the proceedings, the court ordered certification of a class consisting of "Connecticut licensed auto body repair shops, or licensed individuals, that have performed physical auto body repairs paid for directly or indirectly, partially or in full, by [The] Hartford as a result of automobile insurance policies issued by [The] Hartford." Thereafter, the trial court denied The Hartford's motion for reconsideration or reargument. This appeal followed.

I

We begin by setting forth the standard of review and the legal principles that govern class certification orders. "A trial court must undertake a rigorous analysis to determine whether the plaintiffs have borne the burden of demonstrating that the class certification requirements of Practice Book §§ 9-7^[7] and 9-8^[8] have been met. . . . A trial court nonetheless has broad discretion in determining whether a suit should proceed as a class action. . . . As long as the trial court has applied the proper legal standards in deciding whether to certify a class, its decision may . . . be overturned [only] if it constitutes an abuse of discretion. . . .

"[I]n determining whether to certify the class, a [trial] court is bound to take the substantive allegations of the complaint as true. . . . That does not mean, however, that a court is limited to the pleadings when determining whether the requirements for class certification have been met. On the contrary . . . [t]he class determination generally involves considerations that are enmeshed in the factual and legal issues comprising the [plaintiffs'] cause of action . . . and . . . it [sometimes] may be necessary for the court to probe behind the pleadings before coming to rest on the certification question. . . . In determining the propriety of a class action, [however] the question is not whether the plaintiff or plaintiffs have stated a cause of action or will prevail on the merits . . . but rather whether the requirements of [the class action rules] are met. . . . Although no party has a right to proceed via the class mechanism . . . doubts regarding the propriety of class certification should be resolved in favor of certification. . . .

"The rules of practice set forth a two step process for trial courts to follow in determining whether an action or claim qualifies for class action status. First, a court must ascertain whether the four prerequisites to a class action, as specified in Practice Book § 9-7, are satisfied. These prerequisites are: (1) numerosity—that the class is too numerous to make joinder of all members feasible; (2) commonality— that the members have similar claims of law and fact; (3) typicality—that the [representative] plaintiffs' claims are typical of the claims of the class; and (4) adequacy of representation—that the interests of the class are protected adequately. . . .

"Second, if the foregoing criteria are satisfied, the court then must evaluate whether the certification requirements of Practice Book § 9-8 are satisfied. These requirements are: (1) predominance—that questions of law or fact common to the members of the class predominate over any questions affecting only individual members; and (2) superiority—that a class action is superior to other available methods for the fair and efficient adjudication of the controversy. . . . Because our class certification requirements are similar to those embodied in rule 23 of the Federal Rules of Civil Procedure,^[9] and our jurisprudence governing class actions is relatively undeveloped, we look to federal case law for guidance in construing the provisions of Practice Book §§ 9-7 and 9-8." (Citations omitted; internal quotation marks omitted.) Collins v. Anthem Health Plans, Inc., 275 Conn. 309, 320-23, 880 A.2d 106 (2005).

In the present case, The Hartford concedes that the plaintiffs satisfied the first four prerequisites of numerosity, commonality, typicality and adequate representation but claims that the trial court incorrectly determined that they satisfied the requirement of predominance. We therefore turn to the legal principles that guide our analysis of this issue.

II

"[T]he fundamental purpose of the predominance inquiry is to determine whether the economies of class action certification can be achieved . . . without sacrificing procedural fairness or bringing about other undesirable results. . . . [C]lass-wide issues predominate if resolution of some of the legal or factual questions that qualify each class member's case as a genuine controversy can be achieved through generalized proof, and if these particular issues are more substantial than the issues subject only to individualized proof. . . .

"In order to determine whether common questions predominate, [a court must] . . . examine the [causes] of action asserted in the complaint on behalf of the putative class. . . . Whether an issue predominates can only be determined after considering what value the resolution of the class-wide issue will have in each class member's underlying cause of action. . . . Common issues of fact and law predominate if they ha[ve] a direct impact on every class member's effort to establish liability and on every class member's entitlement to injunctive and monetary relief. . . . [When], after adjudication of the [class-wide] issues, [the] plaintiffs must still introduce a great deal of individualized proof or argue a number of individualized legal points to establish most or all of the elements of their individual[ized] claims, such claims are not suitable for class certification. . . .

"[N]umerous [federal] courts have recognized [however] that the presence of individualized damages issues does not prevent a finding that the common issues in the case predominate. . . . In assessing the predominance requirement in cases involving individualized damages, the [c]ourt's inquiry is limited to whether . . . the proposed methods [for computing damages] are so insubstantial as to amount to no method at all. . . . [The plaintiffs] need only come forward with plausible statistical or economic methodologies to demonstrate impact on a class-wide basis. . . . Particularly [when] damages can be computed according to some formula, statistical analysis, or other easy or essentially mechanical methods, the fact that damages must be calculated on an individual basis is no impediment to class certification.

"It is primarily when there are significant individualized questions going to liability that the need for individualized assessments of damages is enough to preclude [class] certification. . . .

"These standards inform us that a court should engage in a three part inquiry to determine whether common questions of law or fact predominate in any given case. First, the court should review the elements of the causes of action that the plaintiffs seek to assert on behalf of the putative class. . . . Second, the court should determine whether generalized evidence could be offered to prove those elements on a class-wide basis or whether individualized proof will be needed to establish each class member's entitlement to monetary or injunctive relief. . . . Third, the court should weigh the common issues that are subject to generalized proof against the issues requiring individualized proof in order to determine which predominate. . . . Only when common questions of law or fact will be the object of most of the efforts of the litigants and the court will the predominance test be satisfied." (Citations omitted; emphasis in original; internal quotation marks omitted.) Id., at 329-32, 880 A.2d 106.

A

Causes of Action

The plaintiffs allege that The Hartford engaged in conduct that constituted a violation of CUTPA and that, as a result, it was unjustly enriched.^[10] To prevail on a CUTPA claim, the plaintiffs must prove that "(1) the defendant engaged in unfair or deceptive acts or practices in the conduct of any trade or commerce; General Statutes § 42-110b (a); and (2) each class member claiming entitlement to relief under CUTPA has suffered an ascertainable loss of money or property as a result of the defendant's acts or practices. General Statutes § 42-110g (a).^[11] The ascertainable loss requirement is a threshold barrier which limits the class of persons who may bring a CUTPA action seeking either actual damages or equitable relief. . . . Thus, to be entitled to any relief under CUTPA, a plaintiff must first prove that he has suffered an ascertainable loss due to a CUTPA violation." (Citation omitted; internal quotation marks omitted.) Collins v. Anthem Health Plans, Inc., supra, 275 Conn. at 334, 880 A.2d 106.

An "ascertainable loss" is a loss that is "capable of being discovered, observed or established." (Internal quotation marks omitted.) Hinchliffe v. American Motors Corp., 184 Conn. 607, 613, 440 A.2d 810 (1981). "The term ``loss' necessarily encompasses a broader meaning than the term ``damage,'" and "has been held synonymous with deprivation, detriment and injury." Id. To establish an ascertainable loss, a plaintiff is "not required to prove actual damages of a specific dollar amount." Id. "[A] loss is ascertainable if it is measurable even though the precise amount of the loss is not known." Id., at 614, 440 A.2d 810.

A plaintiff also must prove that the ascertainable loss was caused by, or "a result of," the prohibited act. General Statutes § 42-110g (a); see also Abrahams v. Young & Rubicam, Inc., 240 Conn. 300, 306, 692 A.2d 709 (1997). When plaintiffs seek money damages, the language "as a result of" in § 42-110g (a) "requires a showing that the prohibited act was the proximate cause of a harm to the plaintiff. . . . [P]roximate cause is [a]n actual cause that is a substantial factor in the resulting harm. . . . The question to be asked in ascertaining whether proximate cause exists is whether the harm which occurred was of the same general nature as the foreseeable risk created by the defendant's act." (Citations omitted; internal quotation marks omitted.) Abrahams v. Young & Rubicam, Inc., supra, at 306, 692 A.2d 709.

When plaintiffs seek only equitable relief, ascertainable loss and causation may be proven "by establishing, through a reasonable inference, or otherwise, that the defendant's unfair trade practice has caused the plaintiff [injury]. . . . The fact that a plaintiff fails to prove a particular loss or the extent of the loss does not foreclose the plaintiff from obtaining injunctive relief and [attorney's] fees pursuant to CUTPA if the plaintiff is able to prove by a preponderance of the evidence that an unfair trade practice has occurred and a reasonable inference can be drawn by the trier of fact that the unfair trade practice has resulted in a loss to the plaintiff." (Citations omitted; emphasis in original; internal quotation marks omitted.) Service Road Corp. v. Quinn, 241 Conn. 630, 644, 698 A.2d 258 (1997). When plaintiffs seek both monetary and equitable relief, proving that the unfair trade practice was a substantial factor in causing the plaintiffs' harm for the purpose of obtaining money damages also, by reasonable inference, serves to establish harm for the purpose of obtaining equitable relief.

B

Generalized Versus Individualized Proof

On appeal, The Hartford does not challenge the trial court's conclusion that generalized evidence may be used to prove deceptive acts and practices under the first prong of CUTPA^[12] but, rather, its conclusion that the plaintiffs may rely on generalized evidence to prove causation and ascertainable loss under CUTPA's second prong. The Hartford specifically claims that the plaintiffs' methodology is based on assumptions and estimates and that causation and ascertainable loss may be proven only by examining evidence obtained from individual insureds and repair shops. The plaintiffs respond that the bulk of the evidence regarding steering and the suppression of labor rates is drawn from The Hartford's own records and the testimony of The Hartford's employees, thus rendering individualized proof unnecessary. The plaintiffs also contend that the methodology proposed by their expert witness demonstrates that generalized evidence may be used to prove causation and ascertainable loss. We agree with the plaintiffs.

We begin by summarizing the proffered evidence, which consists of two affidavits from the plaintiffs' expert, Frederic B. Jennings, Jr., an economic consultant, and two affidavits from David A. Slossberg, the plaintiffs' counsel. Attached to three of the four affidavits are exhibits containing extensive documentation generated by The Hartford, including internal memoranda describing and evaluating its policies and programs, and several depositions by company employees.

The exhibits contain evidence on steering intended to show that, when insureds require auto body repairs, employees of The Hartford known as customer care team specialists (specialists) are instructed to direct the insureds to the closest preferred shop through The Hartford's customer repair service program (repair service program). The specialists are trained to "sell" the repair service program aggressively by informing insureds that, if they utilize the recommended preferred shop, they will receive $100 off of their deductible and a lifetime guarantee for the repair. The Hartford gives the specialists scripts to follow so that they communicate with the insureds in a uniform manner. Moreover, there is considerable pressure placed on the specialists to perform. Supervisors regularly monitor customer calls and coach the specialists to sell the repair service program, and The Hartford offers cash bonuses and other incentives to specialists who obtain the best results. Results are measured by "[repair service program] utilization" numbers, which represent the number of insureds who utilize the repair service program as a percentage of the total number of calls fielded by a particular specialist. Repair service program utilization is an important factor in determining whether a specialist receives a bonus, a raise or discipline for deficient performance.

The exhibits also contain evidence on the suppression of labor rates indicating that The Hartford has been eliminating the use of independent appraisers in recent years and increasing the number of its own automobile service representatives to perform appraisals so that the company can control their content, including labor rates. The Hartford reviews the work of the automobile service representatives for conformance with its expectations and has characterized shops that charge considerably more than its approved labor rate as "militant. . . ." (Internal quotation marks omitted.) When confronted with estimates that exceed what The Hartford views "to be within the normal limits," it negotiates to reduce them or may refuse to honor them altogether. Although The Hartford claims that it pays the prevailing labor rate in Connecticut, there appears to be a growing disparity between the labor rate that it pays for auto body repairs and the market rate for comparable work by automobile mechanics, which may be double the labor rate for auto body repairs. In a letter addressed to the state department of insurance and the office of the attorney general in February, 2002, four automobile service representatives expressed concern that continuing to follow The Hartford's appraisal practices would place them and other representatives in jeopardy of violating their legal and ethical obligations to conduct independent appraisals but that, if they failed to comply with The Hartford's guidelines, they would risk losing their jobs.

Jennings proposes a methodology to examine the effect, if any, of the foregoing policies and practices on nonpreferred shops in Connecticut. To measure the effect of steering, Jennings suggests a three step analysis pursuant to which the number of claims diverted to preferred shops initially is established by estimating the percentage of claims that would have flowed to preferred shops in the absence of any steering. Jennings uses the specialist with the lowest rate of success in referring insureds to preferred shops as the benchmark. The benchmark assumes that the specialist had no effect on the customer's decision and, therefore, represents a conservative approach because even the least effective specialist is attempting to steer claimants to preferred shops. Jennings notes, for purposes of the remaining analysis, that the referral rate for the least effective specialist in the month of February, 2003, was 19 percent. He then observes that the average rate of referral to preferred shops in 2003 was 47 percent. From this, he concludes that, if 19 percent is used as the benchmark for the number of insureds who would use preferred shops in the absence of steering, no more than 19/47, or approximately 40 percent, of all referrals to preferred shops would have gone to those shops without steering. Correspondingly, approximately 60 percent of all referrals to preferred shops would have gone to nonpreferred shops without steering. Jennings next applies these percentages to the 3693 repair service program jobs performed in 2003 by Connecticut shops and concludes that 60 percent, or approximately 2200 jobs, were diverted from nonpreferred shops and directed to preferred shops because of steering.

The diverted claims are then valued to determine the impact of steering on revenue. This is accomplished by valuing the steered claims at the 2003 state industry average repair cost per claim and multiplying that cost by the number of jobs, which results in a figure representing revenues lost by nonpreferred shops due to steering. Steered claims are valued at the 2003 average cost of repairs for preferred shops, which, as expected, is lower than the cost of repairs performed by nonpreferred shops. According to Jennings, this methodology can be applied to any period as well as to commercial claims using data provided by The Hartford during the discovery process.

The third step requires calculation of the profits generated by the shifted revenues to ascertain the respective gains and losses experienced by preferred shops and nonpreferred shops. Jennings assumes, on the basis of his knowledge of the industry, that the average net profit margin of Connecticut auto body repair shops is 5 percent. He then calculates the profits lost due to steering as 5 percent of lost revenues.

Jennings explains that the foregoing methodology is generally accepted among economists and is appropriate in the present context because it produces a uniform result for the class as a whole. He adds that, as discovery proceeds and more information is gathered, other methods may be utilized to make a final assessment of damages.

Jennings suggests a similar, multi-layered analysis to measure the effect of The Hartford's practices on labor rates. After determining the actual labor rates that The Hartford pays for repair work, which may be obtained from company records, Jennings estimates the prevailing labor rates in an uncontrolled market. He suggests that the prevailing rates may be calculated by examining comparable arm's-length transactions to derive a proxy for the labor rate in a freely competitive market (the comparability approach) or by performing a cost-based analysis to determine the minimum labor rate sufficient to keep pace with inflation and to support a sustainable level of profitability in the business. Both approaches would rely on data from public and industry sources and would yield a measure of economic loss to preferred shops and nonpreferred shops per repair hour of labor attributable to the suppression of labor rates, after adjusting for the impact of steering.

Jennings then applies this methodology to estimate the effect of labor rate suppression on the revenue of preferred shops and nonpreferred shops in Connecticut. Rather than performing a comparability or cost-based economic analysis, which apparently would require additional evidence, Jennings relies, for purposes of his affidavit, on testimony from appraisers formerly employed by The Hartford that the prevailing labor rate for auto body repairs in an uncontrolled market would be approximately two times greater than the rate approved of by The Hartford. Multiplying the hourly shortfall by the repair hours reimbursed, and adjusting for steering, generates a measure of overall loss due to The Hartford's suppression of labor rates.

Jennings distributes the aggregate losses to the putative class from The Hartford's practices on the basis of work performed by each class member. As he explains: "The Hartford's claimants . . . have already made (steering-constrained) choices about which shops to use. The market has thereby distributed each of those [repair] jobs (both those flowing to non[preferred] shops, and those steered into [preferred] shops) on the basis of (steering-constrained) considerations expressed through market decisions. These market decisions, as adjusted . . . for their steering and labor-rate distortions, can be used to guide distribution of damages from both sources of harm. The losses set forth above are calculated and expressed in proportion to work performed on . . . claims with this purpose in mind. A reasonable way to allocate damages across the class as a whole is in accord with [the] claimants' choices, even if their decisions were encumbered through steering and other pressures (which this damage assessment seeks to offset). Allocating damages to each class member on the basis of work performed for The Hartford over the relevant period balances the allocation in a manner commensurate to incurred losses, and in accord with [the] claimants' choices. . . . In any event, none of these estimates or their proposed distribution among those harmfully affected thereby involves individual differences in any essential way."

We conclude that the trial court was well within its discretion in finding that the plaintiffs satisfied their burden of demonstrating that generalized, class-wide evidence may be used to prove that The Hartford engaged in unfair or deceptive acts or practices that caused each of the putative class members to suffer an ascertainable loss. The plaintiffs rely on data from The Hartford's own records to describe these practices and to show that they were a substantial factor in steering insureds to preferred shops and in suppressing the labor rates charged for auto body repairs in Connecticut. Moreover, the proposed methodology was offered by a qualified expert with significant experience in the auto body repair industry who attested that, on the basis of his professional experience as an economist, such an analysis is a "generally accepted and appropriate methodology for this type of application. . . ." Accordingly, Jennings' methodology satisfies the standards for proving causation and ascertainable loss when plaintiffs in a CUTPA action seek monetary damages and equitable relief.

The Hartford nonetheless claims that the plaintiffs cannot rely on generalized evidence to prove that, as a direct and proximate consequence of its conduct, every class member suffered an ascertainable loss of money or property. The Hartford specifically contends that the plaintiffs cannot rely on generalized evidence to prove that it steered individual insureds to preferred shops because individual insureds ultimately make auto body repair decisions on the basis of factors beyond the insurer's control. These include recommendations from friends and colleagues, advertising, the customer's prior experience with the shop and, finally, the shop's location, reputation, specialty and curb appeal. The Hartford also claims that generalized evidence is insufficient to establish a causal link between its practices and the suppression of labor rates because individual shops charge different rates on the basis of shop specific considerations. The Hartford likens the present case to Collins v. Anthem Health Plans, Inc., supra, 275 Conn. at 336, 880 A.2d 106, and Macomber v. Travelers Property & Casualty Corp., 277 Conn. 617, 643-44, 894 A.2d 240 (2006), in which this court concluded that the predominance requirement had not been met. We disagree.

The plaintiffs in Collins, eight orthopedic surgeons and four groups of orthopedic surgeons, alleged breach of contract, tortious interference with business expectancies, breach of the implied covenant of good faith and fair dealing and a violation of CUTPA. Collins v. Anthem Health Plans, Inc., supra, 275 Conn. at 315-16, 880 A.2d 106. In their motion for class certification, the plaintiffs sought to serve as representative parties for all physicians and physician groups who had entered into contracts with the defendant, Anthem Health Plans, Inc. (Anthem), from 1993 to 2001 to provide medical services to persons enrolled in Anthem's health insurance plans. Id., at 316, 880 A.2d 106. The trial court ultimately granted class certification with respect to subparagraphs 20(b), (g), (j) and (m) of the plaintiffs' complaint.^[13] Id., at 320, 880 A.2d 106. On appeal, we reversed the trial court's class certification order on the ground that the predominance requirement had not been satisfied, explaining that "generalized evidence would not obviate the need to examine each class member's individual position to determine whether he or she suffered an injury in fact and whether the challenged business policies were the cause of that injury." Id., at 336, 880 A.2d 106.

We conclude that Collins is distinguishable from the present case because the relationship between Anthem and the individual members of the putative class in Collins was determined by the particular services rendered by each class member, the circumstances surrounding the denial of authorization, which depended on the type of coverage provided under the patient's benefit plan, and the motivation of Anthem, its employees and the primary care provider. Thus, with respect to subparagraph 20(b), we observed that the allegation that Anthem's employees and participating primary care providers had denied authorization for covered medical services because of their desire to receive a bonus would require review of each medical procedure for which Anthem had denied authorization to determine that authorization actually had been denied, that the primary care provider or employee who denied authorization was eligible to receive a bonus or incentive, that the denial was not attributable to another cause, and that the individual class member had suffered a loss as a result of the denial. Id., at 337, 880 A.2d 106. We further observed that proving that a particular procedure was covered by a patient's benefit plan would be a daunting task because Anthem had thousands of different benefit plans, and the availability of benefits to particular members depended on multiple factors, including varying and tiered copayments, annual or lifetime benefit limits or maximums and medical necessity review. Id., at 337-38, 880 A.2d 106.

We also concluded that substantial, individual fact-finding was required to determine the merits of the allegations in subparagraph 20(g) that class members received incorrect payments due to Anthem's failure to provide a complete list of codes because claim codes varied by specialty, and class members who performed a wide array of procedures would likely have suffered greater harm than members who performed a more limited range of procedures. Id., at 339-40, 880 A.2d 106. In addition, payment differences could have arisen because of clerical errors or reasonable differences of opinion as to how a particular procedure should be coded. Id., at 340, 880 A.2d 106.

With respect to subparagraph 20(j), we determined that the effect of Anthem's alleged practice of profiling class members for the purpose of permitting or disallowing payment on the basis of statistical averages also required individualized proof to establish that each class member was threatened by the profiling policy and suffered an ascertainable loss as a result of that policy. Id., at 345, 880 A.2d 106. In this regard, we noted the deposition testimony of one of the plaintiff physicians that, after comparing his utilization statistics with those of other physicians in his specialty, he had concluded that his profile was "good" and that he therefore had no problem with this policy. (Internal quotation marks omitted.) Id.

We finally concluded that subparagraph 20(m), which alleged that Anthem had failed to provide the resources necessary for preauthorization on nights and weekends, required individualized proof because, in order to show causation and harm, each class member would have needed to establish that he or she had performed a procedure on a weekend or an evening between 1993 and 1996, when the challenged practice was in effect, that the procedure was a bona fide emergency that could not be delayed until preauthorization had been obtained, and that payment was denied due to a lack of preauthorization. Id., at 343, 880 A.2d 106.

In contrast to Collins, the relationship between The Hartford and the individual class members in the present case is not complicated by numerous distinctions among class members that require individualized proof, such as the particular services that class members provide. To the extent that The Hartford argues that proof of steering and the suppression of labor rates requires evidence as to the reasons why individual insureds patronized certain shops, or why individual shops charged certain hourly labor rates, it fails to recognize that, under Jennings' methodology, this information is incorporated in the benchmarks themselves. Thus, the benchmarks represent the equilibrium that would have existed if all of the independent variables and forces to which The Hartford refers, including recommendations from friends, advertising, shop reputation, the cost to different shops of doing business and normal, unfettered competition among the shops, were left to operate freely in an uncontrolled market. Having quantified normal market conditions in this manner, Jennings concludes that any deviation from the benchmarks is due to The Hartford's practices of steering and labor rate suppression, and not to other factors. This conclusion is based on The Hartford's extensive records and the testimony of its employees indicating that The Hartford had the expressed intention of directing business to preferred shops and of suppressing the labor rates charged by both preferred and nonpreferred shops in Connecticut.

The allegations in Collins did not lend themselves to a similar kind of benchmark analysis because, although Anthem's general policies may have been applicable to all members of the class, its relationships with individual class members were defined by so many separate conditions and variables that they could not be quantified for the purpose of making a before and after comparison to demonstrate the effect of Anthem's policies and practices. In sum, The Hartford has presented this court with no good reason to dismiss Jennings' methodology at this point in the proceedings as inadequate for purposes of class certification.^[14]

Macomber also is inapposite. In that case, the plaintiff, Lisa Macomber, moved for certification of a class comprised of individuals like herself throughout the country who, since 1982, had entered into structured settlements with the defendant insurance companies in personal injury actions that were funded with annuities whose cost and true value allegedly were misrepresented by rebating and shortchanging schemes. Macomber v. Travelers Property & Casualty Corp., supra, 277 Conn. at 620-21, 623-24, 894 A.2d 240. We determined that the trial court had "abused its discretion in certifying the case as a class action because it [had] not undertake[n] the requisite analysis to determine whether individual issues of law predominated. . . ." Id., at 638, 894 A.2d 240. The trial court had allowed the plaintiff to demonstrate predominance through an analysis of approximately thirty out of tens of thousands of possible files from jurisdictions throughout the nation; see id., at 641, 643, 894 A.2d 240; rather than through an analysis of "whether the various differing state laws [that governed the putative class members' claims] shared a commonality that predominated over any differences in such laws. . . ." Id., at 640, 894 A.2d 240. We also agreed with the defendant insurance companies that "the predominance inquiry must focus, not only on whether the [defendant insurance companies] failed to disclose the costs and values of the annuities . . . but also on whether they made the critical representations outlined in that case, namely, representations of the costs of the annuities, and representations that the values equaled the costs. These are necessarily individualized inquiries, the presence and predominance of which simply cannot be properly gauged on the basis of thirty files out of thousands." Id., at 643-44, 894 A.2d 240. We acknowledged that these concerns might cause us to rule that the case was inappropriate for class certification because it did "not contain the requisite predominance of common factual issues." Id., at 644, 894 A.2d 240. We nevertheless reversed the trial court's class certification order and remanded the case for further proceedings to give the plaintiff an opportunity to establish the predominance of common legal and factual issues on the basis of an adequate record. Id.

Although we recognized the challenge posed by the large number of jurisdictions and potential class members in Macomber; see id., at 638-40, 894 A.2d 240; we did not conclude that the predominance requirement could not be satisfied but, rather, that discovery had been limited, and, therefore, the record was inadequate to decide the issue. See id., at 644, 894 A.2d 240. Furthermore, even though we acknowledged that individualized inquiries might be needed to determine whether the defendant insurance companies had made misrepresentations to the plaintiff and putative class members regarding the proposed settlements; see id., at 643-44, 894 A.2d 240; we afforded the plaintiff additional time to develop a proper record in support of her claim. See id., at 644, 894 A.2d 240. We thus remained open to the possibility that the plaintiff might be able to meet the predominance requirement. See id.

The Hartford also claims that the present case bears a "striking" resemblance to Klay v. Humana, Inc., 382 F.3d 1241 (11th Cir.2004), cert. denied sub nom. UnitedHealth Group, Inc. v. Klay, 543 U.S. 1081, 125 S. Ct. 877, 160 L. Ed. 2d 825 (2005). In that case, the plaintiff physicians sought to represent physicians in all fifty states who had submitted at least one claim to the defendant health maintenance organizations (HMOs) between 1990 and 2002 alleging that the HMOs had conspired "to program their computer systems to systematically underpay physicians for their services." Id., at 1246. On appeal, the Eleventh Circuit Court of Appeals reversed the District Court's order granting class certification as to the plaintiffs' state breach of contract claims because, although the claims were based on contract questions common to the entire class, individualized issues of fact were likely to predominate. Id., at 1261-67. The court observed that the case involved the actions of many HMOs over a significant period of time and that each HMO, during the specified time, had used many different contract forms. Id., at 1263. In addition, each HMO had contracted with multiple care-providing entities, and each contract involved the use of different contract forms. Id. Moreover, the fact that the HMOs allegedly had conspired to underpay physicians by programming the HMO computers in certain ways did not establish that any individual physician had been underpaid on any particular occasion. Id., at 1264. The court concluded that each physician would have been required to prove, for each alleged breach of contract, that the request for reimbursement had been submitted, that the physician submitting the request was entitled to a specified amount, that the physician actually had received a lesser amount, and that the HMO's reasons for denying full payment were insufficient. Id. The court further concluded that the plaintiffs' claims regarding the improper procedures that the HMOs allegedly had used to determine entitlement to reimbursement required individualized proof due to their complexity and uniqueness. Id., at 1264-66.

We disagree with The Hartford that the facts in Klay bear any resemblance to the facts in the present case. Rather, we view Klay as similar to Collins, which we decline to follow because members of the putative class in that case did not stand in the same relationship to the defendant insurer, Anthem, and, therefore, common issues of fact did not predominate. See Collins v. Anthem Health Plans, Inc., supra, 275 Conn. at 335-36, 880 A.2d 106. Like the plaintiffs in Collins, the plaintiffs in Klay had contractual relationships with the defendant HMOs that were highly differentiated and dependent on a multiplicity of contract forms. See Klay v. Humana, Inc., supra, 382 F.3d at 1263-64. Accordingly, the plaintiffs in Klay, unlike the plaintiffs in the present case, could not rely on generalized evidence to prove their claims.^[15]

The Hartford finally contends that Jennings' methodology improperly relies on assumptions and estimates. These assumptions and estimates, insofar as they pertain to steering, are that (1) active steering occurs, (2) incentives by The Hartford, rather than other factors, causes the specialists to direct insureds to preferred shops, and (3) the specialist's recommendations are the only reason that insureds choose preferred shops. The Hartford also claims that Jennings made several improper assumptions pertaining to labor rate suppression, including that (1) the average net profit margin on sales by auto body repair shops in Connecticut is 5 percent, (2) the auto body repair market does not operate freely in Connecticut because of The Hartford's significant market share, (3) The Hartford's dominant position in the market allows it to exercise undue influence over labor rates, and (4) in the absence of The Hartford's influence, auto body repair shops would be paid the same hourly labor rate as automotive mechanics, whose labor rate is approximately two times greater.

In response to this claim, we begin by observing that any methodology constructed to measure whether an identified action has had a predicted effect must allow for the possibility that a causal relationship exists, without assuming that such a relationship does, in fact, exist. Insofar as The Hartford claims that Jennings' other assumptions are improper, it provides no facts demonstrating that they are incorrect. Furthermore, The Hartford does not question Jennings' qualifications or the analytical steps that he proposes to measure the effect of its practices on the auto body repair industry. Indeed, in oral argument before this court, The Hartford conceded, in responding to a question as to whether it was attacking the persuasiveness or validity of Jennings' methodology, that it had no problem with Jennings' methodology as a whole but merely was arguing that generalized proof could not answer the question of why individual insureds had chosen to patronize selected shops. This type of circular reasoning, namely, that generalized evidence cannot be used to prove the plaintiffs' claims because they can be proven only by individualized evidence, begs the question and reflects The Hartford's misunderstanding of the theoretical principles underlying Jennings' methodology. We therefore conclude that The Hartford's claims lack merit.

C

Predominance of Common Issues

We finally consider whether common issues of law and fact subject to generalized proof predominate over issues that require individualized proof. The trial court determined that common questions predominated on the issue of liability because almost all of the proposed evidence on whether The Hartford had engaged in unfair or deceptive acts or practices consisted of data and information provided by The Hartford's own documents, records and employees. The court also determined that the evidence that Jennings used to prove causation and ascertainable harm consisted of data from The Hartford's records and sources and, therefore, would be common to the class.

We agree with the foregoing conclusions, which were based on an analysis of the burden of proof with respect to The Hartford's liability. There being no issue on appeal as to whether the plaintiffs may rely on generalized proof for the purpose of allocating damages among members of the class, a key consideration in determining whether the predominance requirement has been met; see McLaughlin v. American Tobacco Co., 522 F.3d 215 (2d Cir.2008) (decertifying class in part because proffered method of calculating damages masked prevalence of individual issues); we conclude that the plaintiffs have satisfied the predominance requirement.

We note that, in the event that circumstances change as discovery proceeds and the trial court determines that class certification is improper, it may issue an order modifying its prior certification order or decertifying the class altogether. "Our courts . . . have stated that [a trial court] has broad discretion in determining whether a suit should proceed as a class action. . . . Nonetheless, despite the absence of a requirement under our class action rules that trial courts monitor developments that may bear [on] certification . . . such a procedure is prudent and sensible when a trial court considers it warranted under the circumstances of the particular case. Such an approach not only protects the resources of the courts . . . but also may protect the parties' interests — defendants may be protected from frivolous class action lawsuits and plaintiffs may be permitted to adjust the class definition when necessary to conform to the changing circumstances." (Citation omitted; internal quotation marks omitted.) Rivera v. Veterans Memorial Medical Center, 262 Conn. 730, 739, 818 A.2d 731 (2003).

For all of the foregoing reasons, we conclude that the predominance requirement has been satisfied with respect to The Hartford's liability under CUTPA and that the trial court did not abuse its discretion in granting the plaintiffs' motion for class certification. See, e.g., Collins v. Anthem Health Plans, Inc., 266 Conn. 12, 23, 836 A.2d 1124 (2003) ("In reviewing a decision of the trial court for abuse of discretion, every reasonable presumption should be given in favor of the trial court's ruling. . . . [R]eversal is required [only when] the abuse is manifest or [when] injustice appears to have been done." [Citation omitted; internal quotation marks omitted.]).

The order granting class certification is affirmed.

In this opinion the other justices concurred.

NOTES

[1] The plaintiffs include Artie's Auto Body, Inc., A & R Body Specialty, Skrip's Auto Body and the Auto Body Association of Connecticut. The first three plaintiffs are businesses licensed to perform auto body repairs in Connecticut. The fourth plaintiff is a not-for-profit association comprised of more than 100 auto body repair shops in Connecticut dedicated to the advancement of the auto body repair industry.

[2] Count one specifically alleges that The Hartford violated (1) the Connecticut Unfair Insurance Practices Act, General Statutes § 38a-815 et seq., (2) Connecticut law prohibiting appraisers and insurers from steering insureds to specific auto body repair shops; see General Statutes § 38a-354; (3) state regulations mandating fair and impartial appraisals of damaged property and prohibiting appraiser requests that appraisals or repairs be made in specified shops; see Regs., Conn. State Agencies §§ 38a-790-6 and 38a-790-8; and (4) well established federal policy regulating the auto body repair industry. See United States v. Assn. of Casualty & Surety Cos., Docket No. 63 Civ. 3106, 1963 U.S. Dist. LEXIS 9949, *2-*3 (S.D.N.Y. November 27, 1963) (federal court consent decree that enjoined certain automobile insurers from "placing into effect any plan, program or practice which has the purpose or effect of [1] sponsoring, endorsing or otherwise recommending any appraiser of damages to automobile vehicles; [2] directing, advising or otherwise suggesting that any person or firm do business or refuse to do business with [a] any appraiser of damage to automotive vehicles with respect to the appraisal of such damage, or [b] any independent or dealer franchised automotive repair shop with respect to the repair of damage to automotive vehicles; [3] exercising any control over the activities of any appraiser of damage to automotive vehicles; [4] allocating or dividing [insureds], territories, markets or business among any appraisers of damage to automotive vehicles; or [5] fixing, establishing, maintaining or otherwise controlling the prices to be paid for the appraisal of damage to automotive vehicles, or to be charged by independent or dealer franchised automotive repair shops for the repair of damage to automotive vehicles or for replacement parts or labor in connection therewith, whether by coercion, boycott or intimidation or by the use of flat rate or parts manuals or otherwise").

[3] The Hartford appealed to the Appellate Court, and we transferred the appeal to this court pursuant to General Statutes § 51-199 and Practice Book § 65-1.

Although the parties raise no claim regarding the appealability of the trial court's order granting class certification, we previously have stated that, when nonCUTPA counts are "inextricably intertwined" with CUTPA counts, the trial court's order granting class certification as to any nonCUTPA count also is reviewable. (Internal quotation marks omitted.) Palmer v. Friendly Ice Cream Corp., 285 Conn. 462, 475 n. 12, 940 A.2d 742 (2008).

[4] The plaintiffs' counsel estimated at oral argument that the putative class consisted of approximately 750 auto body repair shops and individuals licensed to do business in Connecticut, of which fifty are preferred shops that have a contractual relationship with The Hartford to repair damaged automobiles that The Hartford refers to them. These fifty shops are affected by the labor rate claim but not the steering claim.

[5] In its brief, The Hartford explains that, at the time the plaintiffs commenced this action, it offered its customer repair service program only to insureds. At some point in 2006, it began offering the program to claimants as well.

[6] During oral argument before the trial court on March 6, 2006, the plaintiffs' counsel further argued that The Hartford had prohibited insureds from obtaining automobile damage appraisals from independent appraisers.

[7] Practice Book § 9-7 provides: "One or more members of a class may sue or be sued as representative parties on behalf of all only if (1) the class is so numerous that joinder of all members is impracticable, (2) there are questions of law or fact common to the class, (3) the claims or defenses of the representative parties are typical of the claims or defenses of the class, and (4) the representative parties will fairly and adequately protect the interests of the class."

[8] Practice Book § 9-8 provides: "An action may be maintained as a class action if the prerequisites of Section 9-7 are satisfied and the judicial authority finds that the questions of law or fact common to the members of the class predominate over any questions affecting only individual members and that a class action is superior to other available methods for the fair and efficient adjudication of the controversy."

[9] Rule 23 of the Federal Rules of Civil Procedure provides in relevant part: "(a) Prerequisites. One or more members of a class may sue or be sued as representative parties on behalf of all members only if:

"(1) the class is so numerous that joinder of all members is impracticable;

"(2) there are questions of law or fact common to the class;

"(3) the claims or defenses of the representative parties are typical of the clams or defenses of the class; and

"(4) the representative parties will fairly and adequately protect the interests of the class.

"(b) Type of Class Actions. A class action may be maintained if Rule 23(a) is satisfied and if:

* * *

"(3) the court finds that the questions of law or fact common to class members predominate over any questions affecting only individual members, and that a class action is superior to other available methods for fairly and efficiently adjudicating the controversy. . . ."

[10] On the last page of its reply brief, The Hartford argues that the plaintiffs' unjust enrichment claim fails for the same reasons that its CUTPA claim fails. It is "a well established principle that arguments cannot be raised for the first time in a reply brief." (Internal quotation marks omitted.) Statewide Grievance Committee v. Burton, 282 Conn. 1, 19 n. 7, 917 A.2d 966 (2007). We therefore do not address it.

[11] General Statutes § 42-110g (b) provides: "Persons entitled to bring an action under subsection (a) of this section may, pursuant to rules established by the judges of the Superior Court, bring a class action on behalf of themselves and other persons similarly situated who are residents of this state or injured in this state to recover damages."

General Statutes § 42-110g (a) provides in relevant part that only "person[s] who [suffer an] ascertainable loss of money or property . . . as a result of the use or employment of a method, act or practice prohibited by section 42-110b, may bring an action" under CUTPA.

It follows that the "``ascertainable loss' threshold applies with equal force to class actions." Collins v. Anthem Health Plans, Inc., supra, 275 Conn. at 334-35 n. 9, 880 A.2d 106.

[12] We note that, in oral argument before this court, The Hartford expressly denied allegations in the plaintiffs' complaint that it had engaged in deceptive acts or practices.

[13] Subparagraph 20(b) specifically alleged that primary care providers and Anthem's employees improperly had denied authorization for covered medical services because of their desire to receive bonuses. Collins v. Anthem Health Plans, Inc., supra, 275 Conn. at 337, 880 A.2d 106. Subparagraph 20(g) alleged that Anthem's policy of sending to each health care provider a representative, rather than comprehensive, fee schedule that listed only those fees corresponding to the most commonly used billing codes in the provider's area of specialty caused the plaintiffs to submit claims based on inaccurate codes and, therefore, not to receive the correct payment for services rendered. Id., at 339, 880 A.2d 106. Subparagraph 20(j) alleged that Anthem had made " ``payment for services dependent on profiling,' a practice under which ``treatment and/or payment for covered services for the patient is permitted/disallowed in whole or [in] part by the use of statistical averages for the treating physician.' " Id., at 344, 880 A.2d 106. Finally, subparagraph 20(m) alleged that Anthem had failed to provide an on-call physician for weekends or nights to preauthorize procedures, and, therefore, plaintiffs who required authorization from a physician during those times could speak only to clerks and could not obtain preauthorization. Id., at 343, 880 A.2d 106. Consequently, plaintiffs who performed procedures without preauthorization ran the risk that Anthem would deny payment. Id.

[14] We note that, although the benchmark analysis proposed in this case is sufficient to overcome The Hartford's objections that common issues do not predominate on the matter of liability, such an analysis is not necessarily required in all cases to show that common issues predominate.

[15] The Hartford also argues, citing Agrella v. Ford Motor Co., Superior Court, judicial district of Waterbury, Complex Litigation Docket, Docket No. CV-02-0184712-S (May 18, 2006), that Connecticut trial courts regularly have refused to certify classes, particularly in cases involving CUTPA claims in which individual issues of law or fact have predominated over issues requiring generalized proof. In Agrella, the court declined to certify a class of Ford sport utility vehicle owners alleging financial harm due to Ford Motor Company's deceptive advertisements extolling the safety of its sport utility vehicles because individualized proof would have been required to show that each consumer had purchased or leased the vehicle as a result of Ford Motor Company's deceptive advertising. We first note that an appeal from the trial court's denial of class certification in that case is pending before this court. We also note that Agrella is factually distinguishable because the plaintiffs in Agrella are consumers who purchased the Ford Motor Company's product, whereas the plaintiffs in this case are not insureds of The Hartford but auto body repair shops that performed the repairs. We therefore reject The Hartford's attempt to compare Agrella with the present case.