Jackson v. Unocal Corp.

262 P.3d 874 (2011)

Richard JACKSON; Mary Jackson; Thomas Fehringer; Deborah Hradecky; Robert Hradecky; Dean Lousberg; and Lousberg Partnership, on behalf of themselves and those similarly situated, Petitioners.
v.
UNOCAL CORPORATION, Union Oil Company of California, and Unocal Pipeline Company, Respondents.

No. 09SC668.

Supreme Court of Colorado, En Banc.

October 31, 2011.

The Hannon Law Firm, LLC, Kevin S. Hannon, Justin D. Blum, Denver, Colorado, Inman Flynn Biesterfeld & Brentlinger, P.C., Richard P. Brentlinger, Eric Voogt, Denver, Colorado, Attorneys for Petitioners.

Holland & Hart LLP, John A. Ramsey, Rachel A. Yates, Greenwood Village, Colorado, Holland & Hart LLP, Stephen G. Masciocchi, Denver, Colorado, Attorneys for Respondents.

Robert F. Hill, John H. Evans, Nathan P. Flynn, Denver, Colorado, Thomas D. McFarland, Golden, Colorado, Attorneys for Amicus Curiae Colorado Trial Lawyers Association.

Kerr Brosseau Bartlett, LLC, Marc R. Brosseau, Catherine E. McDaugale, Denver, Colorado, Attorneys for Amicus Curiae Product Liability Advisory Council, Inc.

Justice MARTINEZ delivered the Opinion of the Court.

At issue in this land damages class action are the standards a trial court applies when deciding whether to certify a class pursuant to C.R.C.P. 23. C.R.C.P. 23 provides trial courts with a procedural tool for consolidating claims and managing them as a class action. The rule provides that the class certification decision must be made "[a]s soon as practicable" and "may be conditional." C.R.C.P. 23(c). In this appeal, we clarify three aspects of the class certification procedure. First, due to the fact-driven nature of class certification, a trial court must rigorously analyze the evidence and determine to its satisfaction that each C.R.C.P. 23 requirement is met. Second, a trial court may consider any factual or legal disputes relevant to the C.R.C.P. 23 requirements, including those disputes that incidentally overlap with the merits of the case. However, to the extent such disputes overlap with the merits, a trial court may not resolve factual or legal disputes to screen out or prejudge the merits of the case. Finally, a trial court's obligation to rigorously analyze the evidence relevant to the class certification decision extends to expert disputes. While the trial court need not determine which expert ultimately will prevail on the merits or whether an expert's testimony ultimately will be admissible at trial, it may consider expert disputes to the extent necessary to satisfy itself that the class advocate has established each C.R.C.P. 23 requirement.

We thus reverse the court of appeals' rulings that the trial court must apply a preponderance of the evidence standard to C.R.C.P. 23's requirements, that the trial court must resolve factual or legal disputes dispositive of class certification regardless of any overlap with the merits, and that the trial court must resolve expert disputes regardless of any overlap with the merits. We also conclude that the trial court rigorously analyzed the evidence in determining that Plaintiffs established an identifiable class and satisfied C.R.C.P 23(b)(3)'s predominance requirement. We therefore reverse the judgment of the court of appeals.

I.

Plaintiffs filed a class action alleging that Unocal Corporation, Unocal Pipeline Company, and Union Oil Company of California (collectively "Unocal") caused asbestos contamination from the removal of an oil pipeline. Unocal was the historic owner of a pipeline that was buried under approximately sixty-nine miles of easements in Logan and Weld Counties. The buried pipe contained a layer of asbestos wrap. From late 1996 through January 1998, Unocal hired a contractor to remove the pipeline. During the excavation and removal process, small pieces of the pipe's asbestos wrap were left on the easement properties.

In 2006, Plaintiff land owners brought this action, asserting claims for nuisance, negligence, trespass, respondeat superior, and unjust enrichment. Plaintiffs' amended complaint sought compensatory damages for diminution of land value, remediation efforts, and loss of use and enjoyment. Plaintiffs moved for class certification under C.R.C.P. 23(b)(3) on the basis that common issues of liability and damages predominated over individual issues. Plaintiffs sought to certify two classes: (1) an Easement Property Class that includes owners of properties containing the pipeline easement; and (2) a Contiguous Property Class that includes owners of properties that are contiguous to those containing the easement.

In determining whether to certify the Easement and Contiguous Property Classes, the trial court considered 146 pages of briefs with fifty-four exhibits, affidavits from seven experts, portions of deposition transcripts from twelve witnesses, wind and sampling data, and numerous other documents. The court also held a two-day hearing on class certification with additional oral and written testimony, including fifty-three more exhibits.

Among other things, the parties disputed three issues relevant to the class certification decision and to the C.R.C.P. 23 requirements. The first issue was whether asbestos fibers had migrated from the easement properties to the contiguous properties. Plaintiffs submitted as expert testimony the deposition of an industrial hygienist, Doctor Terry Spear. He relied on an air-diffusion model prepared by an environmental engineer, Doctor Kumar Ganesan, and opined that the asbestos fibers from the pipeline wrap migrated to contiguous properties to beyond five miles from the easement properties. In turn, Unocal called Doctor Brent Kreger, an expert toxicologist, who disputed Spear's opinion that asbestos fibers had blown onto the contiguous properties.

The parties also disputed the appropriate methodology for calculating the loss in property value attributable to the alleged asbestos contamination. Plaintiffs' real-estate appraiser, Wayne Hunsperger, opined that loss in property value could be calculated on a class-wide basis. In contrast, Unocal's appraiser, Michael Earley, explained that individual evidence, such as actual appraisals and paired sales, would be needed to reliably calculate loss in property value due to the alleged asbestos contamination.

Finally, Unocal presented eleven affirmative defenses, one or more of which will apply to fifty-nine of the sixty-six Easement Property Class members, as well as to an unspecified number of Contiguous Property Class members.

Based on the record, the trial court certified both classes under C.R.C.P. 23(a) and (b)(3). To begin with, the trial court cited Burkhead v. Louisville Gas & Electric Co., 250 F.R.D. 287, 291 (W.D.Ky.2008), for the proposition that C.R.C.P. 23 requires the class proponent to demonstrate an identifiable class where there is some evidence of a reasonable relationship between the class boundaries and the spread of contamination. It was undisputed that the easement properties had been contaminated with asbestos fibers and thus constituted an identifiable class. As noted, however, there was conflicting expert testimony regarding the spread of asbestos fibers to the contiguous properties. The trial court declined to resolve this expert dispute out of fear that it involved a "merits question that [is] not appropriate for determination at this stage of the proceedings." Instead, the trial court explained that it did not find Ganesan's air-diffusion model "so flawed to be inadmissible." The trial court thus ruled that there was "sufficient evidence" of a reasonable relationship between the spread of contamination and the Contiguous Property Class boundaries to satisfy the requirement of an identifiable class.^[1]

The trial court also determined that common issues predominated over individual issues for the purposes of C.R.C.P. 23(b)(3). As part of its analysis, the trial court sorted through a variety of legal and factual issues, identifying whether each issue was common to the class or individual to each class member. The trial court identified eleven issues common to all the class members. It noted that "the primary question of fact is whether the removal of the pipeline by the defendants resulted in the release of asbestos onto the class members' property." It also noted as a common question whether the asbestos contaminated the class members' properties. The court did not, however, resolve the expert dispute regarding whether Plaintiffs could rely on Ganesan's air-diffusion model as a class-wide method of proving asbestos contamination.

The trial court also declined to resolve the competing testimony from the two real estate appraisers regarding the potential need for individualized evidence of loss in property value. Instead, the trial court accepted both Unocal's position that there were individual issues regarding damages and Plaintiffs' position that whether the release of asbestos caused a diminution in value of the class members' properties was a common issue.

Finally, the trial court refused to consider Unocal's affirmative defenses that could potentially result in individual issues of proof predominating over common issues. The court explained that these individual defenses did not undermine predominance where common questions of liability otherwise predominated.

The court of appeals reversed the order granting class certification. Jackson v. Unocal Corp., 231 P.3d 12, 15-16 (Colo.App.2009). First, the court of appeals announced that a trial court must apply the preponderance of the evidence standard to the proof supporting each C.R.C.P. 23 requirement. Based on this standard, the court of appeals concluded that the trial court abused its discretion when it found that there was "some evidence" or "some reasonable evidence" of an identifiable class. The court of appeals then vacated the entire order and remanded the case for further findings, by a preponderance of the evidence, regarding the proof supporting each of the C.R.C.P 23 requirements.

The court of appeals also ruled on the need to resolve the conflicting expert testimony regarding the spread of asbestos fibers. After reviewing Colorado and federal cases, the court of appeals explained that a trial court must rigorously analyze the evidence supporting each C.R.C.P. 23 requirement and resolve factual or legal disputes dispositive of class certification, regardless of any overlap with the merits of the class's claims. The court of appeals was "persuaded that when faced with conflicting expert testimony on a question potentially dispositive of class certification, a trial court must compare the relative weight of expert opinions in ruling on a motion for class certification.'" Id. at 22 (quoting In re Urethane Antitrust Litig., 251 F.R.D. 629, 637 (D.Kan.2008)). The court of appeals thus directed the trial court, on remand, to resolve the expert dispute regarding the spread of asbestos to the extent necessary to decide whether an identifiable class had been established.

Finally, turning to the C.R.C.P. 23(b)(3) predominance inquiry, the court of appeals held that the trial court abused its discretion by failing to address Unocal's affirmative defenses and the issue of individualized damages. The court of appeals thus directed the trial court to make specific findings regarding Unocal's affirmative defenses and the evidence concerning individual damages.

Plaintiffs timely appealed to this court and we granted certiorari on three issues.^[2] Plaintiffs contend that the court of appeals: (1) erroneously created a new preponderance of the evidence standard for each C.R.C.P. 23 requirement; (2) improperly required the trial court to assess the credibility of expert testimony at the class certification stage; and (3) invaded the trial court's case management discretion.

II.

We review a trial court's decision to certify a class under the highly deferential abuse of discretion standard. See Farmers Ins. Exch. v. Benzing, 206 P.3d 812, 817 (Colo.2009); Friends of Chamber Music v. City and Cnty. of Denver, 696 P.2d 309, 317 (Colo.1985). An abuse of discretion occurs where the trial court's decision is manifestly arbitrary, unreasonable, or unfair, Hock v. New York Life Ins. Co., 876 P.2d 1242, 1251 (Colo.1994), or when the trial court applies the incorrect legal standards, see Kuhn v. State Dep't of Revenue, 817 P.2d 101, 105 (Colo.1991).

III.

We begin our analysis with an overview of the procedural requirements for class certification as set forth in C.R.C.P. 23. We then address three issues central to this appeal and C.R.C.P. 23.

A.

In our system of civil justice, class actions serve a number of important functions. The basic purpose of a class action is "to eliminate the need for repetitious filing of many separate lawsuits involving the interests of large numbers of persons and common issues of law or fact by providing a fair and economical method for disposing of a multiplicity of claims in one lawsuit." Mountain States Tel. & Tel. Co. v. Dist. Court, 778 P.2d 667, 671 (Colo.1989). Class actions also provide plaintiffs with access to judicial relief when they might not otherwise have such access. See Amchem Prods., Inc. v. Windsor, 521 U.S. 591, 617, 117 S. Ct. 2231, 138 L. Ed. 2d 689 (1997) (explaining that class actions allow parties to aggregate "relatively paltry potential recoveries into something worth someone's (usually an attorney's) labor" (internal quotation and citation omitted)). And, they protect defendants from inconsistent obligations. See William B. Rubenstein, Alba Conte & Herbert B. Newberg, Newberg on Class Actions § 1:6 at 27 (4th ed. Supp. 2009). Given these important purposes, Colorado has a "policy of favoring the maintenance of class actions." Benzing, 206 P.3d at 817-18 (citing LaBerenz v. Am. Family Mut. Ins. Co., 181 P.3d 328, 333-34 (Colo. App.2007)).

C.R.C.P. 23 provides trial courts with a procedural tool for consolidating claims into a class action. See Mountain States Tel. & Tel. Co., 778 P.2d at 671 ("Rule 23 of the Colorado Rules of Civil Procedure provides the procedural standards for the filing and maintenance of class actions."). C.R.C.P. 23 offers "the trial court flexibility in shaping a class action," and "provides the court with ample powers, both in the conduct of the trial and relief granted to treat common things in common and to distinguish the distinguishable." Goebel v. Colo. Dep't of Insts., 764 P.2d 785, 794 (Colo.1988). Due to the case management nature of C.R.C.P. 23, we have consistently held that trial courts have "a great deal of discretion in determining whether to certify a class action." Id. (citing Friends of Chamber Music, 696 P.2d at 316-17); see also Benzing, 206 P.3d at 817-18; Air Commc'n & Satellite Inc. v. EchoStar Satellite Corp., 38 P.3d 1246, 1251 (Colo.2002); State v. Buckley Powder Co., 945 P.2d 841, 844 (Colo.1997).

In determining whether an action is appropriate for class certification, a trial court must ascertain whether the claims actually meet the preconditions of C.R.C.P. 23(a): numerosity, commonality, typicality, and adequacy of representation.^[3] If these conditions are met, the trial court may then consider granting certification where it "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." C.R.C.P. 23(b)(3).^[4] Ultimately, "whether a case should be certified is a fact-driven, pragmatic inquiry guided by the objective of judicial efficiency and the need to provide a forum for the vindication of dispersed losses." Medina v. Conseco Annuity Assur. Co., 121 P.3d 345, 348 (Colo.App.2005) (citing Amchem, 521 U.S. at 591, 117 S. Ct. 2231).

The burden is on the class action advocate to demonstrate that each C.R.C.P. 23 requirement is met. Benzing, 206 P.3d at 818. Because "C.R.C.P. 23 should be liberally construed in light of its policy favoring the maintenance of class actions[,]" trial courts generally should accept the plaintiff's allegations in support of certification. Id. There may, however, be disputes regarding the C.R.C.P. 23 requirements that will require the trial court to look beyond the pleadings and conduct "some inquiry into the plaintiff's theory of the case." Id. at 820; see also LaBerenz, 181 P.3d at 334 (encouraging trial courts to "probe behind the pleadings before coming to rest on the certification question" (citations omitted)); Medina, 121 P.3d at 348 (deciding whether to certify a class "requires more than a review of the pleadings" and often requires an evidentiary hearing). It is thus well-settled that a trial court must conduct a "rigorous analysis" of the evidence supporting each C.R.C.P. 23 requirement. Gen. Tel. Co. of Sw. v. Falcon, 457 U.S. 147, 161, 102 S. Ct. 2364, 72 L. Ed. 2d 740 (1982).

The plain text of C.R.C.P. 23 then requires a trial court to make findings regarding the factual predicates of each class certification requirement. See, e.g., C.R.C.P. 23(b)(3) (directing trial courts to "find[ ]" that common issues predominate over individual issues). For example, C.R.C.P. 23(a)(1)'s numerosity requirement turns on a factual predicate regarding the size of the class. Accordingly, a trial court must rigorously analyze the evidence and make a finding that "[t]he class is so numerous that joinder of all members is impracticable." C.R.C.P. 23(a)(1). Similarly, a trial court must rigorously analyze the evidence and make findings that each of the remaining C.R.C.P. 23 requirements is satisfied.

B.

The first issue in this appeal is whether the class advocate must establish C.R.C.P. 23's requirements by a preponderance of the evidence. The court of appeals was unable to find any Colorado cases setting forth a specific burden of proof to apply to the evidence supporting each C.R.C.P. 23 requirement. Jackson, 231 P.3d at 17. The court of appeals thus looked to federal and out-of-state decisions, and was persuaded to adopt a preponderance of the evidence burden of proof. Id. at 18-19 (collecting cases). We reverse the court of appeals' decision because C.R.C.P. 23 is a case management tool and neither the rule nor our caselaw imposes a specific burden of proof on the trial court's certification decision.

C.R.C.P. 23 contemplates a trial court's broad discretion to consider the evidence and determine whether each of the class certification requirements is met. Further, C.R.C.P. 23(c)(1) states that class certification may be "conditional" and must occur "as soon as practicable."

As soon as practicable after the commencement of an action brought as a class action, the court shall determine by order whether it is to be so maintained. An order under this section (c) may be conditional, and may be altered or amended before the decision on the merits.

C.R.C.P. 23(c)(1). Because C.R.C.P. 23(c)(1) requires a trial court to make its certification decision as soon as practicable, it does not envision a protracted mini-trial on certification; rather, the rule provides for a procedure in which a class may be conditionally certified, and then decertified if new evidence is revealed during discovery. Due to the preliminary timing and conditional nature of class certification, trial courts have significant discretion to find whether the evidence proffered by the class advocate satisfies each C.R.C.P. 23 requirement.

Our recent decision in Benzing rests upon a trial court's ongoing obligation to consider whether proceeding as a class action is appropriate. There, we explained that the burden is on the plaintiff to demonstrate that the requirements of C.R.C.P. 23 have been met. Benzing, 206 P.3d at 818. We further explained that a trial court's certification order is "inherently tentative" and that a trial court "retain[s] a continuing obligation to review whether proceeding as a class action is appropriate" in light of new evidence. Id. (citations omitted). Benzing thus supports the proposition that a trial court's rigorous and ongoing analysis of the evidence is sufficient to ensure that C.R.C.P. 23's requirements are satisfied.

The trial court's discretion in deciding whether to certify a class is consistent with the case-management nature of the class certification decision.^[5] For example, the key inquiry under C.R.C.P. 23(b)(3) is the evidentiary and procedural question of what shape the trial will take, and whether the proof at trial will be susceptible to class-wide adjudication. Our caselaw has thus directed trial courts to focus on whether the claims are susceptible to class-wide proof, not the proof itself. Id. at 820; see also Medina, 121 P.3d at 348 ("Many courts consider whether a common nucleus of operative fact exists." (citations omitted)). In light of the predictive and pragmatic nature of class certification, a trial court retains discretion to find whether the evidence satisfies the C.R.C.P. 23 requirements, and ultimately, whether a class action lawsuit would provide the parties with a just, efficient, and economical resolution.

C.R.C.P. 23 and Benzing provide sufficient guidance regarding the standards that govern the class certification decision. A trial court must conduct a rigorous analysis of the evidence and find to its satisfaction that each C.R.C.P. 23 requirement is established. Leaving class certification to the discretion of the trial court without requiring a specific burden of proof squares with the pragmatic and flexible nature of the class certification decision, recognizes the trial court's ongoing obligation to assess the certification decision in light of new evidence, and preserves the trial court's case management discretion. It is also consistent with the direction that C.R.C.P. 23 should be liberally construed in light of Colorado's policy of favoring the maintenance of class actions. Benzing, 206 P.3d at 818.

Unocal asserts three reasons for adopting a preponderance of the evidence standard, none of which is persuasive. First, Unocal urges this court to follow the recent trend among the federal circuit courts of appeals and adopt a preponderance of the evidence standard for the proof supporting each C.R.C.P. 23 requirement. See Fener v. Operating Eng'rs Constr. Indus. & Misc. Pension Fund (Local 66), 579 F.3d 401, 407 (5th Cir.2009); In re Hydrogen Peroxide Antitrust Litig., 552 F.3d 305, 320 (3d Cir.2009); Teamsters Local 445 Freight Div. Pension Fund v. Bombardier Inc., 546 F.3d 196, 202-04 (2d Cir.2008).^[6] While we recognize that the preponderance of the evidence standard appears to be gaining momentum among the federal courts, we decline to follow this emerging trend due to the important differences between C.R.C.P. 23 and Fed.R.Civ.P. 23 and our view of C.R.C.P 23 as a case management tool.

In 2003, Fed.R.Civ.P. 23(c)(1)(C) was amended to remove the provision that class certification "may be conditional." Fed. R.Civ.P. 23(c)(1)(A) was also amended such that the language stating that the certification be decided "as soon as practicable" was replaced with the phrase "at an early practicable time." Three federal circuit courts of appeals adopted a preponderance of the evidence standard after these 2003 amendments. See Local 66, 579 F.3d at 407 (5th Cir.2009); Hydrogen Peroxide, 552 F.3d at 320 (3d Cir.2009); Teamsters Local 445, 546 F.3d at 202-04 (2d Cir.2008).^[7]

Unocal attempts to argue that the 2003 amendments to Fed.R.Civ.P. 23 had little effect on the federal circuit court decisions to adopt a preponderance of the evidence standard. The Third Circuit, however, explicitly cited the 2003 amendments as "[s]upport" for its adoption of a preponderance standard. Hydrogen Peroxide, 552 F.3d at 318. It explained that the 2003 amendments, while subtle, actually "altered the timing requirement for the class certification decision." Id. at 319. Because certification does not have to occur "as soon as practicable" under amended Fed.R.Civ.P. 23, the Third Circuit observed that federal district courts may now order limited discovery prior to making a certification decision. Id. The Third Circuit further explained that the 2003 amendments require federal district courts to "make a definitive determination that the requirements of Rule 23 have been met before certifying a class." Id. at 319-20.^[8] A preponderance of the evidence standard squares with the additional discovery and definitive nature of certification under amended Fed.R.Civ.P. 23.

Colorado did not amend its own rule, C.R.C.P. 23, in 2003. C.R.C.P. 23(c) still requires a trial court to make a certification decision "as soon as practicable." If trial courts were required to apply a preponderance of the evidence burden of proof, they would have to permit discovery at an earlier stage in the litigation and might even have to hold protracted and expensive mini-trials on the factual issues underlying the certification decision. Such an outcome would delay class certification beyond what C.R.C.P. 23(c) requires. It would also compromise the judicial efficiency of the class action mechanism by requiring plaintiffs to effectively prove the merits of their case at the class certification hearing.

Moreover, unlike amended Fed.R.Civ.P. 23, C.R.C.P. 23(c) still allows class certification to "be conditional." A trial court is not therefore bound by a "definitive" class certification determination, see Hydrogen Peroxide, 552 F.3d at 319-20, but rather has a duty to monitor, and the flexibility to alter, the class certification decision in light of any new evidence, see Benzing, 206 P.3d at 818. The conditional nature of class certification in Colorado thus counsels against a specific burden of proof and in favor of the trial court's discretion to determine to its satisfaction that C.R.C.P 23's requirements are met as the litigation proceeds.

Lastly, unlike the federal courts that have adopted a preponderance standard, Colorado has a policy of liberally construing C.R.C.P. 23 in favor of class certification. Benzing, 206 P.3d at 818. To the extent recent federal circuit court decisions are based on a policy of limiting class actions, see Hydrogen Peroxide, 552 F.3d at 310 ("the potential for unwarranted settlement pressure is a factor we weigh in our certification analysis"),^[9] they are not persuasive. Accordingly, we decline to follow the recent trend among the federal circuit courts of appeals that have adopted a preponderance standard.

Unocal also argues that a preponderance of the evidence standard is necessary to ensure a "rigorous analysis" of the evidence and "actual, not presumed, conformance" with C.R.C.P. 23's requirements. Falcon, 457 U.S. at 160, 161, 102 S. Ct. 2364. In Falcon, however, the U.S. Supreme Court held that a class action "may only be certified if the trial court is satisfied, after a rigorous analysis, that the prerequisites of Rule 23(a) have been satisfied." Id. at 161, 102 S. Ct. 2364. The Court's use of the term "satisfied" in the same sentence as "rigorous analysis" implies that a discretionary standard is compatible with a "rigorous analysis" of the evidence.^[10] We thus find that Falcon supports our standard requiring a trial court to rigorously analyze the evidence and determine to its satisfaction whether the class advocate has satisfied each of the C.R.C.P. 23 requirements.

Finally, Unocal argues that a preponderance of the evidence standard is necessary for appellate review of a trial court's certification decision. Anything less than a preponderance standard, Unocal claims, would leave trial courts with virtually unfettered discretion and thus would be inconsistent with appellate review of the trial court's decision to certify a class. Our caselaw has, however, consistently recognized the discretion afforded a trial court to certify a class action under C.R.C.P. 23. See Benzing, 206 P.3d at 817-18; Air Commc'n & Satellite Inc., 38 P.3d at 1251; Buckley Powder Co., 945 P.2d at 844; Goebel, 764 P.2d at 794. Due to the early timing and conditional nature of class certification, a trial court must retain discretion in the class certification decision. This discretion also squares with the case-management nature of the C.R.C.P. 23 determination. Accordingly, so long as the trial court rigorously analyzes the evidence, it retains discretion to find to its satisfaction whether the evidence supports each C.R.C.P. 23 requirement. This standard is not, therefore, inconsistent with appellate review of a trial court's decision to certify a class.^[11]

C.

Having identified the appropriate standard for a trial court to apply when deciding whether to certify a class, we now turn to the second issue in this appeal: whether a trial court may resolve factual or legal disputes regarding the C.R.C.P. 23 requirements where those disputes independently overlap with the merits.

There is often an overlap between the class certification decision and the merits of the case. In Coopers & Lybrand v. Livesay, the U.S. Supreme Court explained that

Evaluation of many of the questions entering into determination of class action questions is intimately involved with the merits of the claims. The typicality of the representative's claims or defenses, the adequacy of the representative, and the presence of common questions of law or fact are obvious examples. The more complex determinations required in Rule 23(b)(3) class actions entail even greater entanglement with the merits....'

437 U.S. 463, 469 n. 12, 98 S. Ct. 2454, 57 L. Ed. 2d 351 (quoting 15 Charles Alan Wright, Arthur R. Miller, & Edward H. Cooper, Federal Practice and Procedure § 3911 (1976)). Factual and legal issues relevant to the class certification decision may be independently relevant to the merits of the case. This overlap is particularly apparent in the context of C.R.C.P. 23(b)(3)'s predominance inquiry where a trial court must determine whether the plaintiff's claims are susceptible to class-wide proof. See Benzing, 206 P.3d at 820. As a result, a trial court's rigorous analysis of the evidence establishing each C.R.C.P. 23 requirement will frequently overlap with the merits of the plaintiff's claims. "That cannot be helped." Wal-Mart Stores, Inc. v. Dukes, ___ U.S. ___, 131 S. Ct. 2541, 2551, 180 L. Ed. 2d 374 (2011).^[12]

Our caselaw has permitted trial courts to rigorously analyze factual and legal disputes for the purpose of making a C.R.C.P. 23 determination, even where those disputes overlap with the merits. In Benzing, we explained that "the court may analyze the substantive claims and defenses that will be raised to determine whether class certification is appropriate...." 206 P.3d at 818; see also LaBerenz, 181 P.3d at 334 (permitting a preliminary hearing "addressed not to the merits of the plaintiff's individual claim, but to whether he or she is asserting a claim that, assuming its merit, will satisfy the requirements of" C.R.C.P. 23). Importantly though, a trial court's class certification decision may not "prejudge the merits of the case." Benzing, 206 P.3d at 818. C.R.C.P. 23 is not a tool for screening out and certifying only those claims that will prevail on the merits. Rather, C.R.C.P. 23 is a procedural tool for consolidating claims and permitting them to be tried as a class action before a jury. The text of the rule thus recognizes that it applies "before the decision on the merits." C.R.C.P. 23(c)(1). The trial court may not, therefore, resolve a factual or legal dispute that goes solely to the merits of the case. See Clark v. Farmers Ins. Exch., 117 P.3d 26, 31 (Colo.App.2004) (Class certification analysis "may include consideration of the merits of the claims, but without actually deciding them" (citations omitted)). Accordingly, a trial court may consider factual and legal issues that overlap with the merits only to the extent necessary to satisfy itself that the requirements of C.R.C.P. 23 have been met.

With these principles in mind, we now turn to the related issue of whether a trial court may resolve expert disputes relevant to the C.R.C.P. 23 determination.

D.

Expert opinions may be and often are offered to establish or refute the facts relevant to each C.R.C.P. 23 requirement.^[13] The concern, raised by Unocal, is that trial courts will uncritically accept an expert's opinion, "amounting to a delegation of judicial power to plaintiffs, who can obtain class certification just by hiring a competent expert." West v. Prudential Sec., Inc., 282 F.3d 935, 938 (7th Cir.2002). Unocal thus argues that a trial court must resolve expert disputes by weighing the credibility of each expert at the class certification stage. Unocal also promotes holding People v. Shreck hearings at the class certification stage to determine whether an expert's testimony will be admissible at trial. 22 P.3d 68 (Colo.2001).

To begin with, trial courts may not uncritically accept contested expert testimony offered in support of class certification. A trial court must rigorously analyze the evidence supporting class certification, including both an expert's testimony and the underlying evidence supporting that testimony. The trial court must then consider any evidentiary disputes—including expert disputes—that are relevant to the C.R.C.P. 23 requirements.

At the class certification stage, however, a trial court need not determine which expert will ultimately prevail for that is simply a merits decision best left for the jury. As the court explained in In re South Dakota Microsoft Antitrust Litigation,

The question at [the class certification] stage is not whether plaintiffs will be able to carry their burden of proving that their experts' analyses are reliable, but whether it appears that the differences between the experts can be intelligently presented and evaluated within the framework of the class action. On a motion for class certification, it is inappropriate to resolve a battle of the experts. Whether or not [the plaintiffs' expert] is correct in his assessment of common/impact injury is for the trier of fact to decide at the proper time.

657 N.W.2d 668, 677-78 (S.D.2003) (quotations omitted). For the purposes of C.R.C.P. 23, then, the issue is not whether the plaintiff's expert will ultimately prevail on the merits, but rather whether the plaintiff's expert offers evidence that can be intelligently presented and evaluated within the procedural mechanism of a class action. The trial court should therefore rigorously analyze all expert testimony offered by both sides and afford that testimony such weight as it deems appropriate. See In re Urethane Antitrust Litig., 251 F.R.D. at 637 ("The recent trend of authority is to permit the district court to compare the relative weight of expert opinions in ruling on a motion for class certification to the extent necessary to resolve the independent question of whether the plaintiff has shown that common questions will predominate.").^[14] It is, however, unnecessary for a trial court to declare a proverbial winner of battling experts at the class certification stage. See Hnot v. Willis Grp. Holdings Ltd., 241 F.R.D. 204, 210 (S.D.N.Y.2007); Wright v. Honeywell Int'l, Inc., 187 Vt. 123, 989 A.2d 539, 550 (2009).

Nor does our caselaw require a trial court to determine, at the class certification stage, whether the expert testimony will be admissible at trial. When analyzing expert testimony proffered in support of class certification, the issue for the trial court is whether the expert testimony establishes a C.R.C.P. 23 requirement to its satisfaction. See In re Urethane Antitrust Litig., 251 F.R.D. at 637 (Court was "satisfied" that expert's opinions were "sufficiently accurate" such that "the court will not disregard them"). This approach does not mandate a Shreck hearing with its full evidentiary and legal arguments at the class certification stage.

In this respect, our holding differs from at least two federal appellate court cases. In Sher v. Raytheon Co., the Eleventh Circuit held that a trial court erred by failing to determine the admissibility of expert testimony pursuant to Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 113 S. Ct. 2786, 125 L. Ed. 2d 469 (1993), at the class certification stage. 419 Fed.Appx. 887, 890 (11th Cir.2011). Similarly, in American Honda Motor Co., Inc. v. Allen, the Seventh Circuit held that "when an expert's report is critical to class certification, ..., a trial court must conclusively rule on any challenge to the expert's qualifications or submissions prior to ruling on a class certification motion." 600 F.3d 813, 815-16 (7th Cir.2010). That is, in the Seventh Circuit, district courts must hold a full Daubert hearing prior to class certification. Id. at 816.

We do not mandate such a requirement for trial courts in Colorado. As part of its rigorous analysis of expert testimony, a trial court may find it useful to borrow concepts from Shreck. The "flexible, fact-specific nature" of the Shreck analysis may help the trial court determine to its satisfaction that the expert testimony establishes a C.R.C.P. 23 requirement. 22 P.3d at 77. The trial court then has an ongoing duty to continue assessing its certification decision in light of new evidence, such as that which may emerge at a Shreck hearing. See Benzing, 206 P.3d at 818 (explaining that the trial court "retains a continuing obligation to review whether proceeding as a class action is appropriate" (internal quotations omitted)). Thus, after certification, a trial court may hold a Shreck hearing and determine whether the plaintiff has an admissible class-wide theory of proof that can be presented to a jury at trial.^[15] After all, if the expert testimony is ultimately deemed inadmissible under Shreck, the plaintiff may be unable to present common questions at trial, thereby requiring the trial court to decertify the class. In this light, because C.R.C.P. 23 affords trial courts broad discretion and flexibility to certify and subsequently decertify a class action, it is unnecessary to mandate a Shreck hearing prior to class certification.

IV.

We now review the trial court's determination that Plaintiffs established an identifiable class and satisfied C.R.C.P. 23(b)(3)'s predominance requirement.

A.

Although not specifically mentioned in C.R.C.P. 23(a), the definition of the class is an essential prerequisite to maintaining a class action. See Cook v. Rockwell Int'l Corp., 151 F.R.D. 378, 382 (D.Colo.1993) ("To satisfy the requirements of Rule 23(a), plaintiffs must first adequately define each class and then establish that each class is so numerous that joinder of all members is impracticable."); see also McElhaney v. Eli Lilly & Co., 93 F.R.D. 875, 877 (D.S.D.1982) ("Prior to a consideration of the criteria established by Rule 23, the Court must determine whether a class exists, and is capable of legal definition."). Prior to analyzing the C.R.C.P. 23 criteria, a court must determine whether the proposed class definition "specif[ies] a particular group that was harmed during a particular time frame," and "facilitate[s] a court's ability to ascertain [the class's] membership in some objective manner." Bentley v. Honeywell Int'l, Inc., 223 F.R.D. 471, 477 (S.D.Ohio 2004); see also Cook, 151 F.R.D. at 382 (requiring a proposed class definition that is "sufficiently definite so that it is administratively feasible for the court to determine whether a particular individual is a member [of the class]").

When a plaintiff defines a class in geographic terms, courts often analyze whether there is a "logical reason" or "evidentiary basis" for drawing the class boundaries at a particular location. Burkhead, 250 F.R.D. at 291. All that is required is a "reasonable" relationship between the evidence and the class boundaries as proposed by the plaintiff. Id. "Usually, scientific or objective evidence closely ties the spread of the alleged pollution or contamination to the proposed class boundaries, as many mass environmental tort cases demonstrate." Id. Courts have thus found identifiable classes in environmental torts cases based on expert testimony and diffusion models. See, e.g., Boggs v. Divested Atomic Corp., 141 F.R.D. 58, 61-62 (S.D.Ohio 1991) (finding an expert's diffusion model sufficient to support a "reasonable relationship" between the evidence of record and the six-mile radius class definition); Boyd v. Honeywell Int'l, Inc., 898 So. 2d 450, 463 (La.Ct.App.2004) (Plaintiff's expert testimony, "[w]hile it may not be sufficient for purposes of proving causation on the merits, it may properly be considered as corroborative evidence supporting the trial court's decision on the geographic area in the context of determining class certification").

Here, Plaintiffs defined the Contiguous Property Class as "all Class Members who, as of June 8, 2006, owned real property that is contiguous with property that contains the Unocal Pipeline Easement." Plaintiffs further defined this class to encompass all properties within five miles of the easement properties. To support these class boundaries, Plaintiffs presented the expert testimony of Doctor Spear and an air-diffusion model prepared by Doctor Ganesan. Spear explained that the asbestos fibers, contained in the pipeline wrap, became airborne and mobile during the removal process. He further opined that the asbestos fibers had migrated to and contaminated contiguous properties to beyond five miles from the pipeline. To arrive at this opinion, Spear relied on his own experience observing the spread of asbestos at the Libby mine in Montana, literature describing the transport of asbestos fibers through the air, and the air-diffusion model prepared by Ganesan.

Unocal attacked Plaintiffs' Contiguous Property Class definition with rebuttal testimony from its own expert, Doctor Kreger. Kreger argued that Spear relied on the Ganesan air-diffusion model as a basis for concluding that asbestos fibers migrated to beyond five miles from the easement properties. Although Kreger conceded that Ganesan's model was appropriate for predicting the spread of asbestos, he explained that Ganesan relied on a set of inappropriate assumptions regarding the amount of asbestos fibers released during the removal of the pipeline. For example, Kreger explained, Ganesan failed to take into account the fact that the pipeline was removed from wet soil three feet below ground, thereby reducing the possibility that the asbestos fibers became airborne. Kreger also claimed that Ganesan overlooked the fact that the asbestos fibers were intact and covered in tar at the time of the pipeline removal. Finally, Kreger pointed out that Ganesan relied on an asbestos fiber count that was unrepresentative of the pipeline's actual condition when it was removed. Kreger thus asserted that the Ganesan model was based on fundamentally flawed factual assumptions regarding the amount of asbestos released and thus "makes no sense."

The trial court rigorously analyzed this expert dispute and properly determined that there was an evidentiary basis for Plaintiffs' Contiguous Property Class definition. During the two-day class certification hearing, the trial court heard Kreger's testimony critiquing Ganesan's air-diffusion model. The trial court also had deposition submissions from both parties on each expert's opinion regarding the spread of asbestos. Its order thus contains a detailed summary of each expert's competing opinion, the evidentiary bases for those opinions, and numerous citations to the record. Based on this record, the trial court stated that "this Court does not find, as defendants argue, that Ganesan's model is so flawed to be inadmissible." The trial court thus was satisfied that Plaintiffs had presented an evidentiary basis for the class boundaries, thereby establishing the requirement of an identifiable class. See Boggs, 141 F.R.D. at 61-62.

Nonetheless, the court of appeals reversed. The court of appeals identified two alleged errors in the trial court's analysis of the proposed class definition. First, the court of appeals observed that the trial court failed to apply the preponderance of the evidence standard to the proof supporting the class definition. Jackson, 231 P.3d at 19. Second, the court of appeals explained that the trial court improperly refused to resolve the expert dispute regarding the spread of asbestos fibers. Id. at 19-24. Because the court of appeals determined that both of these errors constituted an abuse of discretion, the court of appeals vacated and remanded the order certifying the class. Id. at 29.

There was no need for the trial court to resolve the expert dispute by a preponderance of the evidence prior to certifying the Contiguous Property Class. At the class certification stage, the trial court was not required to find whether Plaintiffs could prove by a preponderance of the evidence that the contiguous properties were in fact contaminated. Rather, for the purposes of class certification, the trial court simply had to determine to its satisfaction that there was evidence of a relationship between the class definition and the spread of asbestos contamination. It did just that by rigorously analyzing the competing expert testimony and finding that Ganesan's air-diffusion model was enough to establish an identifiable class. Because the trial court's rigorous analysis and factual findings satisfied the requirements of C.R.C.P. 23 and our caselaw, we reverse the court of appeals' judgment.

We note, however, that the trial court may subsequently hold a Shreck hearing to analyze the admissibility and reliability of Ganesan's air-diffusion model. In its order certifying the class, the trial court found that Ganesan's model was not "so flawed to be inadmissible." While this finding is sufficient for class certification, it does not constitute an admissibility determination by the standard set forth in Shreck, 22 P.3d at 77. Ultimately, though, whether Ganesan's model is admissible will be pivotal to the management of the case and likely will determine whether the Contiguous Property Class's claims can even be tried before a jury based on Ganesan's theory of contamination. Accordingly, given the pragmatic, case-management nature of C.R.C.P. 23, the trial court may hold a Shreck style hearing prior to proceeding to trial. If, at that hearing, the trial court finds Ganesan's theory inadmissible, it may find it necessary to decertify the Contiguous Property Class.^[16]

B.

To certify a class under C.R.C.P. 23(b)(3), a trial court must find that common questions "predominate over any questions affecting only individual members" and that class resolution is "superior to other available methods for the fair and efficient adjudication of the controversy." The predominance inquiry focuses on "whether the proof at trial will be predominantly common to the class or primarily individualized." Medina, 121 P.3d at 348. Often, the issue most relevant to this inquiry is "whether the plaintiff advances a theory by which to prove or disprove `an element on a simultaneous, class-wide basis, since such proof obviates the need to examine each class member's individual position.'" Benzing, 206 P.3d at 820 (quoting Lockwood Motors, Inc. v. Gen. Motors Corp., 162 F.R.D. 569, 580 (D.Minn.1995)).

Unocal challenges the trial court's predominance inquiry on three grounds. First, Unocal argues that the trial court could not have properly evaluated predominance without resolving the expert dispute over whether Plaintiffs have a scientifically-valid, class-wide method of proving that the contiguous properties were contaminated with asbestos. Unocal further argues that the trial court failed to rigorously analyze the evidence regarding the need for individualized damages. Finally, Unocal claims that the trial court improperly refused to consider the impact of its affirmative defenses on the predominance inquiry. We consider each argument in turn.

To begin with, the trial court was not required to resolve the expert dispute regarding the spread of asbestos contamination. The crucial issue was whether Plaintiffs had a "method to establish, on a class-wide basis" Unocal's liability. Benzing, 206 P.3d at 820. The trial court rigorously analyzed Spear's testimony and Ganesan's air-diffusion model. It found that Ganesan's model was not "so flawed to be inadmissible" and was satisfied that Plaintiffs had presented a class-wide theory of proving contamination of the class properties. The trial court also noted that the expert dispute regarding the factual assumptions underlying Ganesan's model was a merits question that could be presented to a jury. That is, "the differences between the experts [could] be intelligently presented and evaluated within the framework of the class action." See In re S.D. Microsoft Antitrust Litig., 657 N.W.2d at 678. The trial court thus declined to resolve this dispute at the class certification stage. In light of the trial court's rigorous analysis and factual findings, we hold that this analysis satisfies the requirements of C.R.C.P. 23 and Benzing, 206 P.3d at 820.

We also dismiss Unocal's second argument that the trial court failed to consider the impact of individual damages on the predominance inquiry. The predominance inquiry "usually involves liability, not damages[,]" and the "need for some proof of individual damages does not preclude certification under C.R.C.P. 23(b)(3)." Buckley Powder Co. v. State, 70 P.3d 547, 554 (Colo.App.2002). Here, the trial court was presented with conflicting expert testimony regarding the appropriate methodology for assessing loss in property value. On the one hand, Plaintiffs' expert appraiser, Wayne Hunsperger, opined that loss in value for the properties in the class area could be calculated based on class-wide proof.^[17] In contrast, Unocal's expert appraiser, Michael Earley, explained that individual evidence would be needed to calculate the loss in property value attributable to contamination in the class area.^[18] Contrary to Unocal's argument, the trial court did not ignore or decline to consider this expert dispute and the impact of individualized damages on the predominance determination. The trial court identified as a common question the diminution in value of class members' properties as a result of asbestos contamination. The trial court also identified as an individual issue the amount of damages suffered by each member as argued by Unocal's expert appraiser Earley. Nonetheless, despite the existence of some individual damages issues, the trial court determined that common issues predominated and that the case could proceed as a class action. This determination was within the trial court's case management discretion and recognizes the fact that the predominance inquiry often turns on common issues of liability, not damages. Id.

Finally, we conclude that the trial court rigorously analyzed Unocal's affirmative defenses. While courts should consider claims and defenses when ruling on certification, Benzing, 206 P.3d at 818, the fact that some affirmative defenses are particular to an individual class member does not defeat certification or require a finding, as a matter of law, that individual issues predominate. Unocal's answer pleaded eleven or more affirmative defenses, including statute of limitations, estoppel, laches, assumption of risk, and failure to mitigate damages. The trial court explained that even if Unocal's affirmative defenses raised individual issues, these individual issues "do not undermine predominance" because "liability issues and damages issues are common issues which predominate." Thus, the trial court determined that, despite Unocal's affirmative defenses, common issues of liability and damages predominated. The trial court's analysis is within its discretion to manage the claims as a class action. Indeed, as the trial court noted, "defendants made no showing that class certification would prevent defendants from being able to present evidence to support their affirmative defenses." Accordingly, because the trial court's analysis complies with C.R.C.P. 23 and our caselaw, we dismiss Unocal's challenge to the trial court's C.R.C.P. 23(b)(3) determination.

V.

For the foregoing reasons, we reverse the judgment of the court of appeals.

Justice EID dissents, and Justice RICE joins in the dissent.

Justice EID, dissenting.

Today the majority holds that the question of whether the plaintiff has produced sufficient evidence to meet the requirements of class certification is purely a discretionary matter for the trial court to decide. In my view, the majority's standardless approach makes class certification in Colorado essentially unreviewable by appellate courts and raises serious procedural due process concerns. I respectfully dissent.

This case requires us to determine what amount of evidence class proponents must put forth to meet the requirements of Rule 23. The majority declines to answer that question, however, and instead concludes that "trial courts have significant discretion to find whether the evidence proffered by the class advocate satisfied each C.R.C.P. 23 requirement." Maj. op. at 881. In other words, there is no "specific burden of proof" to be met. Id. at 881 (declining to "impos[e] a specific burden of proof on the trial court's certification decision"); id. at 882 ("[l]eaving class certification to the discretion of the trial court without requiring a specific burden of proof"). Instead, the discretionary burden of proof is apparently met when the trial court says it is. Id. at 882 ("A trial court must conduct a rigorous analysis of the evidence and find to its satisfaction that each C.R.C.P. 23 requirement is established") (emphasis added).

The majority's error is to confuse a trial court's discretion with the plaintiff's burden of proof. Discretion is, by definition, not a burden of proof. Discretion is what a trial court exercises in choosing from available options; it is, by definition, a dynamic concept. A burden of proof, by contrast, is a static, legal concept that is applicable to all cases in a particular category. By confusing the two, the majority defines the burden of proof as an ever-moving target of trial court discretion. In some cases, perhaps a plaintiff need only produce a scintilla of evidence; in others, a preponderance might be required; in still others, clear and convincing evidence. By adopting a regime of total evidentiary discretion, the majority makes it virtually impossible for plaintiffs and defendants to evaluate and present their case.

To compound this error, the majority repeatedly emphasizes how vast and subjective the trial court's discretion truly is. The majority begins by referencing the trial court's "significant discretion to find whether the evidence proffered by the class advocate satisfies each C.R.C.P. 23 requirement." Id. at 881; see also id. (trial court has "broad discretion to ... determine whether each of the class certification requirements is met"). The majority then stresses the subjective nature of that discretion, describing the trial court's task as "determin[ing] to its satisfaction whether the class advocate has satisfied" the requirements of Rule 23. Id. at 884 (emphasis added); see, e.g., id. at 877, 882, 884 (same).^[1]

After today's discretionary burden is put into place, class certification decisions will essentially be unreviewable in Colorado. In the majority's regime, the only person who needs to be "satisfied" is the trial judge; there is no objective standard of proof that a class proponent must produce. Therefore, there is very little for an appellate court to examine on review: if the trial court is "satisfied" when it enters an order certifying the class, the amount of proof produced to meet the trial court's discretion is, by definition, sufficient. One wonders what the purpose is of permitting interlocutory appeal of class action certification orders if there is nothing to review. See C.R.C.P. 23(f); § 13-20-901, C.R.S. (2011).

The discretionary burden regime adopted by the majority today is, in my view, contrary to Colorado law. The majority relies primarily on Farmers Insurance Exchange v. Benzing, 206 P.3d 812 (Colo.2009), in which we stated that a trial court's decision to certify a class is reviewed under an abuse of discretion standard. The fact that a decision to certify a class is reviewed for an abuse of discretion, however, does not translate into a regime under which every aspect of a class action case is discretionary. As several courts have explained, given the fact-intensive nature of the inquiry, a decision to certify a class is subject to abuse of discretion review. However, whether the trial court has applied the correct legal standard—including the proper burden of proof—is an issue that is reviewed de novo. See, e.g., Teamsters Local 445 Freight Div. Pension Fund v. Bombardier, 546 F.3d 196, 201-03 (2d Cir.2008) (noting that "[w]e review for abuse of discretion the district court's denial of class certification.... We review de novo any issues of law underlying the Rule 23 ruling, including the question of whether the district court applied the correct standard of proof"); Fener v. Operating Eng'rs Constr. Indus. & Misc. Pension Fund (Local 66), 579 F.3d 401, 406-07 (5th Cir. 2009) (same); In re Hydrogen Peroxide Antitrust Litig., 552 F.3d 305, 320 (3d Cir.2009) (same). To put it differently, a trial court automatically abuses its considerable discretion if it applies the incorrect burden of proof, as occurred in this case.

Secondly, the majority emphasizes that Colorado favors the use of the class action as a procedural device. But again, favoring the use of class actions in appropriate circumstances does not mean that class actions are appropriate in all circumstances. In fact, in the very case upon which the majority relies most heavily, Benzing, we held that class certification was inappropriate because there was no class wide method to show that the defendant had caused the putative class members' injury. 206 P.3d at 820-23.

Perhaps in recognition of the standardless nature of the discretionary burden it adopts, the majority repeatedly states that the trial court must perform a "rigorous analysis" of the evidence presented in support of certification, relying upon language from General Telephone Co. of Southwest v. Falcon, 457 U.S. 147, 161, 102 S. Ct. 2364, 72 L. Ed. 2d 740 (1982). But it is not enough for a trial court to simply consider the evidence and exercise its discretion through a "rigorous analysis," as the majority seems to think. See, e.g., maj. op. at 884. In other words, that "rigorous analysis" must lead to something—namely, a conclusion that a particular burden of proof has been met.

The trial court's opinion in this case perfectly illustrates this problem. The court began its analysis with the statement that "[i]n determining whether to certify a class, the court must accept as true all allegations set forth in the complaint and avoid inquiring into the merits," citing Eisen v. Carlisle & Jacquelin, 417 U.S. 156, 178, 94 S. Ct. 2140, 40 L. Ed. 2d 732 (1974). Trial Ct. Order at 3 (emphasis added); see also id. at 7 (refusing to consider defendants' expert's critique of plaintiffs' expert's opinion because "[t]hese are merits questions that are not appropriate for determination at this stage of the proceedings"); id. at 8. But in Wal-Mart Stores, Inc. v. Dukes (decided after the trial court issued its opinion in this case), the Supreme Court expressly disavowed the interpretation of Eisen adopted by the trial court, stating that "[f]requently [a] rigorous analysis will entail some overlap with the merits of the plaintiff's underlying claim. That cannot be helped," and that anything to the contrary in Eisen "is the purest dictum." 564 U.S. ___, 131 S. Ct. 2541, 2551-52 & n. 6, 180 L. Ed. 2d 374 (2011) (internal quotation marks omitted); see also maj. op. at 885 n. 12 (recognizing the Supreme Court's clarification of Eisen). Additionally, the trial court cited Blackie v. Barrack, 524 F.2d 891, 901 (9th Cir.1975), for the proposition that "the court must take the substantive allegations of the complaint as true." Trial Ct. Order at 8. But that statement in Blackie has been abrogated by Falcon's "rigorous analysis" requirement, and by the Ninth Circuit itself. See, e.g., Ilhardt v. A.O. Smith Corp., 168 F.R.D. 613, 617 (S.D.Ohio 1996) (recognizing abrogation by Falcon); Wang v. Chinese Daily News, Inc., 623 F.3d 743, 754 (9th Cir.2010) (noting that Blackie's statement had been overruled by the Ninth Circuit in Dukes v. Wal-Mart Stores, Inc., 603 F.3d 571, 581 (9th Cir.2010) (en banc), in which the court "firmly rejected any suggestion that a district court, in deciding a class certification motion, may not look behind the pleadings to overlapping merits issues").^[2] Even the majority implicitly disagrees with the trial court's premise, stating that a trial court "may consider" the merits to the extent necessary to determine class certification. See maj. op. at 877, 885. Because the trial court plainly misunderstood the task at hand, the fact that it exercised its discretion in a particular way should be accorded no weight.

Despite the trial court's misapprehension of the inquiry, the majority affirms its conclusion on the ground that it conducted a "rigorous analysis." The court has thus turned the class certification inquiry into a purely procedural requirement. This is made clear by the fact that the majority relies on three factors to affirm the trial court's conclusion, including that the court: (1) held a two-day class certification hearing, in which it heard live testimony; (2) considered deposition testimony; and (3) included a detailed summary of expert opinions in its order, as well as "numerous citations to the record." Maj. op. at 888; see also id. at 878 (noting that the court considered "146 pages of briefs with fifty-four exhibits, affidavits from seven experts, portions of deposition transcripts from twelve witnesses, wind and sampling data, and numerous other documents. The court also held a two-day hearing on class certification with additional oral and written testimony, including fifty-three more exhibits"). "This analysis," the majority concludes, "satisfies the requirements of C.R.C.P. 23." Id. at 889 (emphasis added). In other words, as long as the trial court considers whether certification is appropriate, its certification decision will be upheld.

Significantly, the majority cites to no case in which a court has adopted a discretionary burden of proof. By contrast, numerous courts, both state and federal, have imposed a burden on class proponents to meet the certification requirements by a preponderance of the evidence.^[3] That is because the preponderance standard is simply an outgrowth of class certification requirements themselves. As the majority notes, Rule 23 requires the trial court to make certain "findings." Maj. op. at 881 (citing C.R.C.P. 23(b)(3), which mandates that the court "find[] 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. The matters pertinent to the findings include...."). The essence of making a "finding" is that the court is convinced that a particular proposition is true, and in the civil arena, that occurs when there is a preponderance of the evidence to support the proposition. § 13-25-127(1), C.R.S. (2011) (adopting a preponderance of the evidence standard in civil cases). As one court reasoned, the preponderance standard is required because "the plain text of Rule 23 requires the court to find ... the facts favoring class certification." Alaska Elec. Pension Fund v. Flowserve, 572 F.3d 221, 228 (5th Cir.2009) (per curiam) (citation and internal quotation marks omitted). The trial court cannot "find" something simply by concluding, as the trial court did here, that the particular air-diffusion model supported by the class proponents was not "so flawed to be inadmiss[i]ble." Trial Ct. Order at 7; see also maj. op. at 889. Rather, a finding is a conclusion, based on a complete examination (or "rigorous analysis") of the record, that a proposition has been established by a preponderance of the evidence. At bottom, the majority's discretionary burden of proof— which permits certification as long as a class proponent's proof is "not so flawed to be inadmissible"—is inconsistent with the "rigorous analysis" that the majority purports to require.

Lacking caselaw support, the majority rejects the preponderance standard on the ground that the Federal Rules of Civil Procedure were amended in 2003 to remove the language that class certification "may be conditional"—language that remains in the Colorado rule. Maj. op. at 883-84. The majority seems to reason that because Colorado continues to permit certification on a conditional basis, a trial court is permitted to certify a class on a preliminary basis with minimal evidence, and may decertify the class at a later time if the preliminary basis is ultimately discredited. Maj. op. at 881. But the majority misreads the purpose of a "conditional certification" under our rules. As the Advisory Committee to the Federal Rules Committee made clear, "[t]he provision for conditional class certification [wa]s deleted [in the federal rules] to avoid the unintended suggestion, which some courts have adopted, that class certification may be granted on a tentative basis, even if it is unclear that the rule requirements are satisfied." Hydrogen Peroxide, 552 F.3d at 319 (citation to Advisory Committee Report omitted) (emphasis added). Here, the majority adopts—erroneously in my view—the "unintended suggestion" that a class may be certified on the basis of a model that is "not so flawed as to be inadmissible" because any critique of the model may, at a later time, demonstrate that the class should be decertified. Maj. op. at 889. I would hold, instead, that a trial court cannot certify a class unless it finds, after a "rigorous analysis" of all of the evidence before it—including the critique of the proposed model—that each of the Rule 23 requirements has been met by a preponderance of the evidence.^[4]

Ultimately, the majority's opinion rests on its impression that, in contrast to Colorado, federal courts have adopted "a policy of limiting class actions." Maj. op. at 883; see also id. at 883 n. 7. What the majority refers to as a federal "policy," however, is simply the concern, grounded in procedural due process, that erroneously certified class actions may "create unwarranted pressure to settle nonmeritorious claims on the part of defendants." Newton v. Merrill Lynch, 259 F.3d 154, 162 (3d Cir.2001); see maj. op. 883-84 (citing Hydrogen Peroxide, 552 F.3d at 310 (citing Newton)). This is a concern that informs our Rule 23 jurisprudence as well. See, e.g., Jahn v. ORCR, Inc., 92 P.3d 984, 989 (Colo.2004) (noting that C.R.C.P. 23 "incorporates the due process concerns underlying class actions"). Indeed, the risk of improperly granted or denied class certifications led to the interlocutory appellate review of such decisions under the federal rules, see Newton, 259 F.3d at 162, and our rules as well, see C.R.C.P. 23(f). By adopting an unreviewable discretionary burden of proof, the majority has essentially forfeited this important role for appellate review.

In the end, the majority views class certification as just another "case-management decision" by the trial court, like deciding whether to permit an amendment to a pleading, or whether to stay or continue proceedings. Maj. op. at 881-82 & 882 n. 5 (referring to the "case-management nature of the class certification decision" and listing case-management decisions it finds to be analogous). But again the majority simply misunderstands the nature of the class certification decision. In considering whether to permit the amendment of a pleading, or whether to stay or continue proceedings, the trial court need not evaluate conflicting expert testimony. To put it somewhat differently, the trial court's evaluation of conflicting expert testimony in this case regarding the validity of another expert's air-diffusion model emphatically was not a case management decision. Rather, it was (or should have been) a decision as to whether the model demonstrates, by a preponderance of the evidence, that there is an identifiable class and that common issues predominate over individual issues in the class claims. After today, the decision is simply a matter of keeping the trial court satisfied. For the reasons stated above, I respectfully dissent from the majority's opinion.

I am authorized to state that Justice RICE joins in this dissent.

NOTES

[1] Alternatively, the trial court certified the Contiguous Property Class based on a stigma theory and proximity to the contamination.

[2] We granted certiorari on the following issues:

(1.) Whether the court of appeals erred by creating a "preponderance of the evidence" burden of proof in the certification of a class pursuant to C.R.C.P. 23.

(2.) Whether the court of appeals erred by requiring the trial court to assess the credibility of expert testimony at the class certification stage.

(3.) Whether the court of appeals' construction of C.R.C.P. 23 improperly invaded the trial court's case management discretion.

[3] C.R.C.P. 23(a) requires the class action advocate to demonstrate:

(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.

[4] Alternatively, a trial (b)(2) class action. In certification of a C.R.C. court may certify a C.R.C.P. 23(b)(1) or this appeal, Plaintiffs only moved for certification of a C.R.C.P. 23(b)(3) class action.

[5] In fact, case management decisions are generally left to the discretion of the trial court without requiring a party to satisfy any particular burden of proof. For example, a trial court has broad discretion to stay or continue proceedings. Todd v. Bear Valley Vill. Apartments, 980 P.2d 973, 976 (Colo. 1999). A continuance is granted "only for good cause." C.R.C.P. 121 § 1-11. A trial court has discretion to order separate trials under C.R.C.P. 20(b) and 42(b) "to prevent delay or prejudice or in the furtherance of convenience." Sutterfield v. Dist. Court, 165 Colo. 225, 231, 438 P.2d 236, 240 (Colo.1968). As long as the trial judge "finds that the necessary prerequisites to separate trials laid down by those rules exist," the decision will not be overturned. Id. A trial court also has discretion to grant or deny leave to amend a pleading after considering such factors as whether there was undue delay, bad faith or dilatory motive, a failure to cure deficiencies with previous amendments, undue prejudice, or futility. Polk v. Denver Dist. Court, 849 P.2d 23, 25 (Colo.1993). Finally, matters of pretrial discovery usually fall within a trial court's discretion. Bond v. Dist. Court, 682 P.2d 33, 40 (Colo.1984) (noting that discovery rules "allow[] a trial court to exercise broad discretion to manage the discovery process in a fashion that will implement the philosophy of full disclosure of relevant information and at the same time afford the participants the maximum protection against harmful side effects").

[6] Unocal also cites to a number of federal district court cases that have applied recently a preponderance of the evidence standard. See In re Puerto Rican Cabotage Antitrust Litig., 269 F.R.D. 125, 130 n. 4 & 139 (D.P.R.2010); In re Health-South Corp. Sec. Litig., 257 F.R.D. 260, 272 (N.D.Ala.2009); Reed v. Advocate Health Care, 268 F.R.D. 573, 578 (N.D.Ill.2009).

[7] We recognize that a few federal district courts began to apply a preponderance of the evidence standard before the 2003 amendments to Fed. R.Civ.P. 23. See, e.g., Shepherd v. Babcock & Wilcox of Ohio, No. C-3-98-391, 2000 WL 987830, at *1 n. 5 (S.D.Ohio March 3, 2000); Ilhardt v. A.O. Smith Corp., 168 F.R.D. 613, 617 (S.D.Ohio 1996); Gibb v. Delta Drilling Co., 104 F.R.D. 59, 71 (N.D.Tex. 1984). However, because these federal district court cases were not cited in any of the recent federal appellate court decisions adopting a preponderance of the evidence standard, they do not appear to have had any persuasive effect. Accordingly, we focus our attention on the reasons provided by the federal circuit courts for adopting a preponderance standard, namely, the 2003 amendments to Fed. R.Civ.P. 23 and a policy limiting class actions. See Hydrogen Peroxide, 552 F.3d at 318.

[8] The Second Circuit also relied on the 2003 amendments as a basis for adopting the preponderance standard. In Teamsters Local 445, the court explained that it had effectively adopted the preponderance standard in its prior decision In re Initial Public Offerings Securities Litigation, 471 F.3d 24 (2d Cir.2006) [hereinafter In re IPO]. 546 F.3d at 202. The reasoning in In re IPO was based in part on the 2003 amendments to Fed.R.Civ.P. 23. 471 F.3d at 39.

[9] The Second Circuit in In re IPO also expressed concern about the coercive effect of class actions as a basis for adopting a preponderance of the evidence standard. 471 F.3d at 38 n. 9 ("Every class action defendant wants its evidence disputing Rule 23 requirements considered in order to try to fend off the enormous settlement pressure often arising from certification."). In re IPO, in turn, provided the foundation for the Second Circuit's subsequent decision adopting a preponderance of the evidence standard. See Teamsters Local 445, 546 F.3d at 202.

[10] The Court actually uses the term "satisfied" twice in the same sentence. Falcon, 457 U.S. at 161, 102 S. Ct. 2364. Some courts have viewed the Court's double-use of the word "satisfied" as perplexing. See In re IPO, 471 F.3d at 33. It is equally possible, however, that the Court used the word "satisfied" twice to emphasize that the discretionary nature of class certification corresponds with a rigorous analysis of the evidence.

[11] When considering purely legal theories supporting class certification, appellate courts should apply a de novo standard of review. See Benzing, 206 P.3d at 814.

[12] In Wal-Mart Stores, Inc., the Court dispelled any notion that its prior decision in Eisen v. Carlisle & Jacquelin, 417 U.S. 156, 94 S. Ct. 2140, 40 L. Ed. 2d 732 (1974) prohibits trial courts from analyzing issues that overlap with the merits. 131 S. Ct. at 2552 n. 6.

[13] An expert who testifies, for example, that every plaintiff has suffered injury is in effect testifying that injury may be established by common proof.

[14] A mere dispute among experts is not sufficient on its own to deny class certification. See In re S.D. Microsoft Antitrust Litig., 657 N.W.2d at 679 ("Production of a self-professed expert is simply not enough to meet the certification requirements under our rigorous analysis standard.").

[15] The Shreck hearing with full evidentiary and legal arguments comes later in the judicial proceeding than a class certification hearing. The two are, however, interrelated given that a Shreck hearing may convince the trial court that its initial certification of the class can no longer be justified, thus resulting in decertification of the class.

[16] The trial court's order certifying the class recognizes that "[t]he defined classes are certified, subject to later modification, if necessary."

[17] Hunsperger explained that a paired sales analysis which compares sales of homes in the class area with similar sales of homes outside the class area could be used to estimate the loss in property value due to contamination.

[18] Earley explained that he would have to analyze completed transactions in the class area and interview buyers and/or sellers to determine the actual impact of contamination on property value.

[1] The majority relies on language from General Telephone Co. of Southwest v. Falcon, 457 U.S. 147, 161, 102 S. Ct. 2364, 72 L. Ed. 2d 740 (1982), which states that "a Title VII class action, like any other class action, may only be certified if the trial court is satisfied, after a rigorous analysis, that the prerequisites of Rule 23(a) have been satisfied." (Emphasis added). As noted below, the majority's emphasis on the "satisfied" language to support its subjective burden gives little effect to the "rigorous analysis" requirement.

[2] The U.S. Supreme Court granted, vacated, and remanded the Wang case in light of Wal-Mart Stores, Inc. v. Dukes, 564 U.S. ___, 131 S. Ct. 2541, 180 L. Ed. 2d 374 (2011). Chinese Daily News, Inc. v. Wang, No. 10-1202, ___ U.S. ___, 132 S. Ct. 74, ___ L.Ed.2d ___, 2011 WL 4529967, at *1 (Oct. 3, 2011). While this "accepted as true" language has appeared in our caselaw, it has been qualified. Benzing, 206 P.3d at 818 (stating that "a court must generally accept as true the allegations in support of certification") (emphasis added).

[3] Federal circuit courts adopting the preponderance standard include: Teamsters Local 445 Freight Division Pension Fund v. Bombardier, 546 F.3d 196, 201 (2d Cir.2008); In re: Hydrogen Peroxide Antitrust Litig., 552 F.3d 305, 320 (3d Cir.2009); Fener v. Operating Eng'rs Constr. Indus. & Misc. Pension Fund (Local 66), 579 F.3d 401, 407 (5th Cir.2009). See also West v. Prudential Secs., Inc., 282 F.3d 935, 938 (7th Cir.2002) ("A district judge may not duck hard questions by observing that each side has some support, or that considerations relevant to class certification also may affect the decision on the merits. Tough questions must be faced and squarely decided, if necessary, by holding evidentiary hearings and choosing between competing perspectives."). The Tenth Circuit has imposed an even higher burden. See Reed v. Bowen, 849 F.2d 1307, 1309 (10th Cir.1988) (a party seeking class certification must show "under a strict burden of proof" that the Rule 23 requirements are met). Federal district courts adopting the preponderance standard include: In re HealthSouth Corp. Sec. Litig., 257 F.R.D. 260, 272 (N.D.Ala.2009); Reed v. Advocate Health Care, 268 F.R.D. 573, 578 (N.D.Ill.2009); In re Safety-Kleen Corp. Bondholders Litig., 2004 WL 3115870, at *2 (D.S.C.2004); Ilhardt v. A.O. Smith Corp., 168 F.R.D. 613, 617 (S.D.Ohio 1996); In re Puerto Rican Cabotage Antitrust Litig., 269 F.R.D. 125, 130 n. 4 & 139 (D.P.R.2010). State courts adopting the preponderance standard include: Warner v. Waste Mgmt., Inc., 36 Ohio St. 3d 91, 521 N.E.2d 1091, 1098 n. 9 (1988); Whitaker v. 3M Co., 764 N.W.2d 631, 638 (Minn.App.2009); Bourgeois v. A.P. Green Indus., Inc., 939 So. 2d 478, 487 (La.App.2006).

[4] The majority also points to the fact that the Second and Third Circuits cited to the 2003 federal rule amendments as supporting a preponderance standard. Maj. op. at 883 n. 7 & 883 n. 8 (citing Hydrogen Peroxide, 552 F.3d at 318, and In re Initial Public Offerings Securities Litigation, 471 F.3d 24, 39 (2d Cir.2006)). Both cases, however, referenced the changes in the federal rules as support for the proposition that examination of the merits of the case may be necessary to determine whether certification is appropriate, Hydrogen Peroxide, 552 F.3d at 316-20; In re IPO, 471 F.3d at 38-42, a proposition, as noted above, that the majority and the United States Supreme Court have adopted. In addition, as one commentator has noted, the Second Circuit's reliance on the 2003 rule changes in In re IPO gave it a way "to disavow its own precedents counseling against the weighing of competing expert submissions," Richard A. Nagareda, Class Certification in the Age of Aggregate Proof, 84 N.Y.U. L.Rev. 97, 114 & n. 54 (2009).