Summary
noting that an insurer's decision to deny benefits to its insured must be evaluated based on the information before the insurer at the time of that decision
Summary of this case from Sipes v. Allstate Indem. Co.Opinion
No. 10SC77.
2012-01-17
STATE FARM MUTUAL AUTOMOBILE INSURANCE COMPANY, an Illinois corporation, Petitioner v. Pauline REYHER, and Dr. Wallace Brucker, on behalf of themselves and all others similarly situated, Respondents.
Faegre & Benson LLP, Michael S. McCarthy, Todd P. Walker, Sarah A. Mastalir, Denver, Colorado, Faegre & Benson LLP, Aaron Van Oort, Minneapolis, Minnesota, Mendenhall & Malouff, R.L.L.P., H. Barton Mendenhall, Rocky Ford, Colorado, Attorneys for Petitioner. Hagens Berman Sobol Shapiro, LLP, Robert B. Carey, Megan E. Waples, Colorado Springs, Colorado, Todd A. Travis, P.C., Todd A. Travis, Englewood, Colorado, Law Office of John Gehlhausen, John Gehlhausen, Lamar, Colorado, Attorneys for Respondents.
Faegre & Benson LLP, Michael S. McCarthy, Todd P. Walker, Sarah A. Mastalir, Denver, Colorado, Faegre & Benson LLP, Aaron Van Oort, Minneapolis, Minnesota, Mendenhall & Malouff, R.L.L.P., H. Barton Mendenhall, Rocky Ford, Colorado, Attorneys for Petitioner. Hagens Berman Sobol Shapiro, LLP, Robert B. Carey, Megan E. Waples, Colorado Springs, Colorado, Todd A. Travis, P.C., Todd A. Travis, Englewood, Colorado, Law Office of John Gehlhausen, John Gehlhausen, Lamar, Colorado, Attorneys for Respondents. Wheeler Trigg O'Donnell LLP, Malcolm E. Wheeler, Michael D. Alper, Denver, Colorado, O'Melveny & Myers LLP, Richard B. Goetz, Matthew M. Shors, Washington, D.C., National Chamber Litigation Center, Inc., Robin S. Conrad, Washington, D.C., Attorneys for Amicus Curiae Chamber of Commerce of the United States and the American Tort Reform Association.The Hannon Law Firm, Kevin S. Hannon, Denver, Colorado, Attorneys for Amicus Curiae Colorado Trial Lawyers' Association.Husch Blackwell Sanders LLP, J. Eric Elliff, Denver, Colorado, Attorneys for Amicus Curiae Colorado Civil Justice League.Justice MARTINEZ delivered the Opinion of the Court.
The class certification issue presented by this appeal arises from a dispute concerning the payment of medical bills under the Colorado Automobile Accident Reparations Act (“No–Fault Act”). Plaintiffs, Pauline Reyher and Dr. Wallace Brucker, filed suit against State Farm Mutual Automobile Insurance Company (“State Farm”) alleging that it failed to pay the full, reasonable amount of medical expenses in violation of the No–Fault Act and its contracts. Plaintiffs subsequently moved for certification of two classes that included all insureds and all providers, respectively, who submitted a medical bill to State Farm and were reimbursed less than the full amount.
The trial court denied the motion for class certification on the grounds that Plaintiffs had failed, among other things, to establish C.R.C.P. 23(b)(3)'s predominance requirement. The court explained that the central question in the case—determining the reasonableness of medical bills submitted to State Farm—would require analyzing the facts of each claim. As a result, the court determined that the individual issues of each claim would predominate over issues common to the class.
The court of appeals reversed and remanded the case to the trial court to enter an order certifying the class. See Reyher v. State Farm Mut. Auto. Ins. Co., 230 P.3d 1244, 1253 (Colo.App.2009) [hereinafter “ Reyher II ”]. The court of appeals explained that Plaintiffs had alleged two conceivable theories of proving liability on a class-wide basis. The court of appeals thus concluded that Plaintiffs had demonstrated a class-wide theory of proof, thereby satisfying C.R.C.P. 23(b)(3)'s predominance requirement. Id. at 1258 (citing Farmers Ins. Exch. v. Benzing, 206 P.3d 812, 820 (Colo.2009)).
The court of appeals also determined that Plaintiffs had satisfied C.R.C.P 23(a)'s requirements as well as C.R.C.P. 23(b)(3)'s superiority requirement. Reyher II, 230 P.3d at 1255–57. State Farm does not appeal those determinations.
State Farm now challenges the court of appeals' determination that common issues predominate over individual issues for purposes of C.R.C.P. 23(b)(3). We conclude that the trial court did not abuse its discretion when it determined that individual issues predominate. Therefore, we reverse the judgment of the court of appeals.
I. Facts and Procedure
Reyher was insured under a no-fault insurance policy issued by State Farm. The policy was governed by section 10–4–706(1)(b) of the No–Fault Act, which at that time required State Farm to pay for its insureds “all reasonable and necessary expenses for medical ... services” related to covered automobile accidents. Ch. 303, sec. ––––, § 10–4–706(1)(b), 1984 Colo. Sess. Laws 1071 (formerly codified as amended at § 10–4–706; repealed effective July 1, 2003, ch. 189, sec. ––––, § 10–4–726, 2002 Colo. Sess. Laws 649).
In October 2001, Reyher was injured in an automobile accident and received medical treatment from Dr. Brucker at the Arkansas Valley Regional Medical Center. Dr. Brucker then submitted bills for the treatment to State Farm for reimbursement. At the time, State Farm contracted with Sloans Lake Managed Care to review and handle claims through its Auto Injury Management (“AIM”) program. To determine the “reasonable and necessary” price for a patient's treatment, Sloans Lake's AIM program relied on a Medicode database (the “database”) to compare the price of the treatment with charges for like services in the same geographical area. Based on recommendations generated by the database, Sloans Lake suggested repricing eight of Dr. Brucker's bills. State Farm repriced those bills, compensating Dr. Brucker only for the amount it deemed reasonable.
Plaintiffs filed suit alleging that State Farm had failed to pay the full, reasonable amount of the medical bills. They asserted five separate claims, including breach of contract and violation of the No–Fault Act. Their complaint included allegations on behalf of a class of persons similarly situated.
State Farm moved for summary judgment on all five of Plaintiffs' claims, arguing that Plaintiffs could not prevail because the Colorado Division of Insurance (“DOI”) had issued an order finding that State Farm's use of the database did not violate a DOI regulation relating to the Unfair Competition–Deceptive Practices Act, section 10–3–1104(1)(h)(III)–(IV), C.R.S. (2009) ( “UCDPA”). The trial court granted the motion.
Plaintiffs appealed and in Reyher v. State Farm Mutual Automobile Insurance Co., 171 P.3d 1263 (Colo.App.2007) [hereinafter “ Reyher I ”], the court of appeals reversed. The court explained that “[t]he determination of whether medical expenses and treatment are reasonable and necessary' under the No–Fault Act presents a question of fact.” Id. at 1265. The court then identified a number of questions of material fact regarding Plaintiffs' claims, including: (1) whether an insured presented an insurer with “reasonable proof” of medical expenses and whether that proof reflected reasonable expenses; (2) whether the database is the only source used by State Farm to determine whether expenses are reasonable; and (3) whether the database accurately assessed the reasonableness of medical bills. Id. at 1266. Because the DOI did not resolve any of these factual issues, the court of appeals concluded that its order did not dispose of the lawsuit. Id. at 1267. The court thus remanded the case to the trial court for further proceedings, including a “fact-driven, pragmatic inquiry” into whether Plaintiffs had satisfied the C.R.C.P. 23 criteria for class action certification. Id. (quoting Medina v. Conseco Annuity Assur. Co., 121 P.3d 345, 348 (Colo.App.2005)).
Subsequent to the Reyher I decision, the trial court held a two-day evidentiary hearing on the motion for class certification. In addition to numerous exhibits, the trial court admitted the testimony of various witnesses, including Russell Kile, a State Farm Section Manager. Kile testified to State Farm's claim review process, emphasizing that claim adjusters individually reviewed each claim and considered everything that had transpired with the claim. In its findings of fact, the trial court summarized Kile's testimony that Sloans Lake's AIM program utilized a three step process for repricing claims: (1) a code review to determine how accurately providers were reporting their claims; (2) repricing by comparing bills with a database for like services; and (3) an appeal process for reviewing repriced claims. The trial court further noted that, according to Kile, the database “was a reference for [Sloans Lake] and adjusters to use in determining the amount which should be paid for a particular service.”
The trial court acknowledged that the court of appeals had, in ruling on the summary judgment appeal, identified several common questions of law and fact, including two that correspond to Plaintiffs' class-wide theories of proof:
If [the database] was the only source [used by State Farm to determine whether expenses are reasonable], did State Farm violate DOI regulation no. 5–2–8(4)(E)(2), which requires insurers to “make decisions independent of the vendor's recommendations when appropriate[,]” and thereby presumptively also violate the Act?
Did State Farm compensate insureds for all “reasonable” medical expenses by using the AIM database? The trial court agreed that these questions appeared to be common to both insureds and providers across the spectrum.
The trial court then identified numerous questions regarding the reasonableness of the medical bills which could not be answered by evidence common to the class, including:
What process did [State Farm] use in deciding not to pay the bill as originally submitted?
Was the database used in the repricing program reliable?
Was the repricing decision reviewed?
At the time, was the AIM database the only source of ‘reasonableness' which was considered?
Was it unreasonable for the insurer to deny full payment?
Based on its review of the testimony and conclusion that these questions could not be answered with evidence common to the class, the trial court refused to certify either class, concluding that Plaintiffs had failed to establish any of the C.R.C.P. 23 requirements. Analyzing C.R.C.P. 23(a)(2)'s commonality requirement, the trial court explained that Plaintiffs' claims ultimately turned on the reasonableness of medical bills. The court further noted that
the question of the reasonableness of the medical bills comes down to the facts of each claim, and, despite the many common questions of fact and law, in the final analysis the individualized facts of each repriced bill will have to be examined to answer the question of what is reasonable. The court concluded that answering the question of whether a medical bill was reasonable would require individual claim-by-claim inquiries. Accordingly, the court explained, “it is difficult to envision that one trial—regardless of who represents each class—will produce formulaic solutions which can be applied to resolve the issue of reasonableness for all other policy holders.” As such, the trial court determined that Plaintiffs had failed to establish C.R.C.P. 23(a)(2)'s commonality requirement. Based on this analysis, the trial court also determined that Plaintiffs had failed to establish C.R.C.P. 23(b)(3)'s predominance requirement.
On interlocutory appeal, the court of appeals reversed on the basis that Plaintiffs had alleged at least two potential class-wide theories for proving liability. Reyher II, 230 P.3d at 1253. The court of appeals explained that
plaintiffs could, for example, prove that State Farm had a company-wide practice of unlawfully relying solely on the database without investigation to reprice the class members' bills. Plaintiffs could also conceivably prove that State Farm systematically failed to pay the class members' reasonable medical bills because it determined reasonable payments by using a database incapable of determining reasonableness, thus rendering the repriced payments necessarily not reasonable. Id. at 1257–58. The court of appeals thus identified two distinct class-wide theories of liability, one of which turned on State Farm's alleged failure to properly investigate claims and the other of which turned on State Farm's systematic repricing of claims based on an inaccurate database. Both theories of liability would obviate the need for individualized determinations of the reasonableness of each medical bill. Based on these class-wide theories of proof, the court of appeals determined that Plaintiffs had established C.R.C.P. 23(b)(3)'s predominance requirement. Id. at 1258. The court of appeals also determined that Plaintiffs had established all of the C.R.C.P. 23(a) requirements as well as C.R.C.P. 23(b)(3)'s superiority requirement. Id. at 1255–57. Accordingly, the court of appeals reversed the trial court and remanded with instructions to certify the proposed insured class. Id. at 1258.
The court of appeals also instructed the trial court to enter an order certifying the provider class if the court determined that Dr. Brucker had standing. Id.
State Farm appealed to this Court and we granted certiorari. State Farm only challenges the C.R.C.P. 23(b)(3) predominance determination of the court of appeals. Accordingly, our analysis focuses solely on this class certification requirement.
We granted certiorari on the following issue:
Whether the court of appeals erred in reversing the trial court's denial of class certification under C.R.C.P. 23.
II. Standard of Review
We review a trial court's decision to certify a class under the highly deferential abuse of discretion standard. Jackson v. Unocal Corp., 262 P.3d 874, 879 (Colo.2011). “An abuse of discretion occurs where the trial court's decision is manifestly arbitrary, unreasonable, or unfair, or when the trial court applies the incorrect legal standards.” Garcia v. Medved Chevrolet, Inc., 263 P.3d 92, 97 (Colo.2011).
III. Analysis
State Farm raises two challenges to the court of appeals' determination that common issues predominate over individual issues for the purposes of C.R.C.P. 23(b)(3). First, State Farm argues that the court of appeals erred in evaluating predominance based on Plaintiffs' mere allegations of a class-wide theory of proof, rather than on the evidence submitted with regard to class certification. Alternatively, State Farm argues that the need to individually analyze whether each class member was injured precludes a determination that common issues predominate over individual issues.
A. Class–Wide Theory of Proof
To satisfy the predominance requirement of C.R.C.P. 23(b)(3), a plaintiff must show that legal or factual questions common to the class predominate over questions affecting only individual members. “[T]he predominance prong of the C.R.C.P. 23(b)(3) inquiry usually involves liability, not damages.” Buckley Powder Co. v. State, 70 P.3d 547, 554 (Colo.App.2002); see also Jackson, 262 P.3d at 889. 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)). We have held that a trial court must rigorously analyze the evidence presented and “determine to its satisfaction whether the class advocate has established each of the C.R.C.P 23 requirements.” Jackson, 262 P.3d at 877.
The issue most relevant to the predominance requirement, and thus to the maintenance of the class in this case, is whether the Plaintiffs have advanced a class-wide method of proving that State Farm violated the No–Fault Act and its contracts with insureds. Plaintiffs claim that they have two class-wide theories of proving liability that obviate the need for individual inquiries into the reasonableness of each medical bill. First, Plaintiffs allege that they can establish liability with common evidence based on State Farm's failure to implement reasonable standards for investigation as required by the No–Fault Act and its contracts. Alternatively, Plaintiffs allege that they can establish liability on a class-wide basis by showing that State Farm necessarily failed to pay the reasonable amount of medical expenses because it relied solely on an inaccurate database, i.e. a database incapable of assessing reasonableness, to reprice properly submitted claims.
State Farm, in turn, argues that both of Plaintiffs' class-wide theories of liability are precluded by the trial court's identification of numerous questions that will turn on proof individual to each class member. Accordingly, State Farm argues that individual inquiries are necessary to determine whether State Farm is liable under the No Fault–Act or its contracts.
Our review starts with the trial court's analysis of the evidence offered with regard to class certification. 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. Jackson, 262 P.3d at 880. Our caselaw simply requires a trial court to rigorously analyze the evidence presented and determine to its satisfaction that the evidence supports each C.R.C.P. 23 requirement. Jackson, 262 P.3d at 882; Benzing, 206 P.3d at 818.
Applying that standard here, we find that the trial court rigorously analyzed Plaintiffs' proposed class-wide theories of liability as well as the evidence offered by State Farm to refute those theories. Plaintiffs claim that in Reyher I, the court of appeals found that Plaintiffs had introduced sufficient evidence, including letters and explanations of review, to permit a reasonable jury to infer that State Farm relied solely on the database to review medical claims. 171 P.3d at 1266. Plaintiffs argue that State Farm did not produce any evidence to rebut Plaintiffs' showing that State Farm relied solely on the database.
At the class certification hearing, State Farm offered the unrebutted testimony of Kile regarding State Farm's claim review process. Kile testified that State Farm's claim adjusters worked in conjunction with Sloans Lake to review medical claims. According to Kile, State Farm would receive a claim and assign it to an individual claim adjuster. That adjuster would then send the claim to Sloans Lake where it went through the AIM program. The AIM program included three components: a code review to ensure that doctors were accurately coding their bills, repricing by comparing claims with a database for like services, and an appeal process. After going through the AIM process, the claim was then returned to State Farm where it would be independently reviewed by the assigned adjuster. The adjuster would review all the contents in a claim file to determine what payment should be made on a bill. If the claimant had multiple claims with State Farm, the adjuster might even have considered information in other claim files. Kile thus explained that the database was merely a “reference” in State Farm's claim review process.
After considering this testimony, the trial court identified numerous questions regarding the reasonableness of the medical claims which, the court determined, could not be answered by evidence common to the class. Although the trial court recognized several issues common to the class, the court concluded that individual issues predominate and refused to certify the class. In light of the trial court's rigorous analysis of the evidence, we cannot say that this decision was an abuse of discretion.
This conclusion conforms with our recent decision in Garcia v. Medved, Inc. There, class certification turned on the plaintiffs' ability to establish a class-wide theory for proving their Colorado Consumer Protection Act claims in order to satisfy C.R.C.P 23(b)(3)'s predominance requirement. Based on the plaintiffs' allegations that they were not relying on individual sales interactions, and without holding an evidentiary hearing, the trial court certified two classes. We reversed because the trial court had failed to consider individual evidence offered by the defendant to refute the plaintiffs' class-wide theories of liability and, therefore, had failed to conduct the rigorous analysis required in making the class certification decision.
In this case, the court of appeals applied the incorrect standard to determine whether Plaintiffs had established the class certification requirements. The court of appeals did not dispute the trial court's analysis of the evidence presented, but rather believed it was bound to accept Plaintiffs' allegations as true. The court of appeals explained that any factual dispute regarding State Farm's reliance on the database was “an attack on the merits” that could not be resolved at the class certification stage. Reyher II, 230 P.3d at 1258 (quoting Burgess v. Farmers Ins. Co., 151 P.3d 92, 99–100 (Okla.2006)). The court of appeals thus accepted at face value Plaintiffs' allegations that State Farm had a practice of relying solely on the database to assess the reasonableness of claims and reprice them accordingly. As a result, the court of appeals concluded that Plaintiffs could “conceivably prove” State Farm's liability on a class-wide basis. Id. This constituted error.
In Jackson, we explained that a trial court may consider disputes “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.” 262 P.3d at 885. The U.S. Supreme Court has similarly permitted district courts to analyze issues that overlap with the merits for the purpose of determining whether the plaintiff has established the class certification requirements. See Wal–Mart Stores, Inc. v. Dukes, ––– U.S. ––––, 131 S.Ct. 2541, 2552 n. 6, 180 L.Ed.2d 374 (2011). A trial court may not, however, go a step further and prejudge the merits of the case or otherwise screen cases at the class certification stage. Jackson, 262 P.3d at 885.
In the instant case, whether State Farm relied solely on the database was an issue relevant to Plaintiffs' class-wide theories of proof and the merits of the case. The trial court only considered this issue to the extent necessary to satisfy itself that Plaintiffs had failed to establish State Farm's sole reliance on the database with common proof. Moreover, because the trial court undertook this analysis for the purpose of determining whether common issues predominate over individual issues, it did not violate Jackson or otherwise impermissibly prejudge the merits of the case.
The court of appeals also cited Houck v. Farmers Insurance Co., a case with largely identical facts, as a basis for concluding that the trial court in the instant case impermissibly prejudged the merits. 229 P.3d 551 (Okla.Civ.App.2009). In Houck, the trial court made a finding that an insurer “essentially abandoned an individualized approach to assessment of med-pay claims....” Id. at 556. The Oklahoma appellate court explained that “[t]he trial court erred in making this particular finding because it constitutes a[sic] improper determination of the merits.” Id. at 556 n. 6. Houck is, however, unavailing given our decision in Jackson that trial courts may consider factual disputes relevant to the C.R.C.P. 23 determination, regardless of an overlap with the merits. 262 P.3d at 885. Moreover, in the instant case, the trial court only considered the dispute regarding State Farm's use of the database to the extent necessary to satisfy itself that Plaintiffs would be unable to proceed on a class-wide basis.
Ultimately, the trial court concluded that individual issues predominated over common issues because Plaintiffs lacked a class-wide theory of proving liability. This determination is in accord with our statement in Benzing that “[a]bsent a class-wide method to prove causation, individual issues would predominate over common ones because separate inquiries would be necessary to determine the defendants' liability as to each class member.” 206 P.3d at 820. Accordingly, we defer to the trial court's analysis of the evidence regarding Plaintiffs' class-wide theories of proof and resultant determination that individual issues predominate over common issues. See Jackson, 262 P.3d at 890.
Nonetheless, Plaintiffs argue that the trial court erred as a legal matter for two reasons, both of which we find unpersuasive. First, Plaintiffs claim that the trial court failed to understand its two class-wide theories of proof. Plaintiffs claim that they can prove liability by simply showing that State Farm's process for reviewing claims was inadequate or that it relied solely on a flawed database to reprice claims. Then, if State Farm breached its contractual or statutory obligations, Plaintiffs claim that it would be liable for the amount established by properly submitted medical claims. See § 10–4–708(1), C.R.S. (2002) (“Benefits for any period are overdue if not paid within thirty days after the insurer receives reasonable proof of the fact and amount of expenses incurred during that period[.]”). Plaintiffs thus argue that there is no need to conduct individual inquiries into the reasonableness of each medical bill. Accordingly, they urge us to conclude that the trial court erred as a matter of law when it denied class certification on the grounds that the central issue of liability is the reasonableness of each medical claim.
The trial court did not misunderstand or disagree with the legal basis for Plaintiffs' class-wide theories of proof. In its order, the trial court cited two common questions that correspond with Plaintiffs' class-wide theories of proof:
If [the database] was the only source [used by State Farm to determine whether expenses are reasonable], did State Farm violate DOI regulation No. 5–2–8(4)(E)(2), which requires insurers to “make decisions independent of the vendor's recommendations when appropriate[,]” and thereby presumptively also violate the Act?
Did State Farm compensate insureds for all “reasonable” medical expenses by using the AIM database? The trial court assumed that Plaintiffs' interpretation of the No–Fault Act was correct and recognized that these two questions could be common to the class. However, after rigorously analyzing Plaintiffs' class-wide proof, namely the nature of State Farm's claim review process, the trial court was satisfied that State Farm did not have a class-wide practice of relying solely on the database. The trial court then determined that proof at trial would be predominantly individual—a determination within the trial court's discretion. We defer to this case management decision and recognize that Plaintiffs' interpretation of the No–Fault Act and theories of proving liability can be tested in individual trials on the merits.
Plaintiffs also argue that State Farm is effectively barred from relying on evidence other than the database. Plaintiffs claim that State Farm cited only the database as the basis for denying claims, and thus is limited to that explanation at trial. As a result, Plaintiffs claim that determining liability pivots solely on evidence regarding the database which, by definition, is common to the class. However, “[a]n insurer's decision to deny benefits to its insured must be evaluated based on the information before the insurer at the time of that decision.” Peiffer v. State Farm Mut. Auto. Ins. Co., 940 P.2d 967, 970 (Colo.App.1996). Here, as the trial court found, State Farm did not rely solely on the database to review every claim, but rather had a three-step claim review process. It is therefore entitled to present all the information it used in reviewing each claim. Accordingly, because the trial court conducted a rigorous analysis of the evidence in making its class certification decision, we discern no abuse of discretion in its C.R.C.P. 23(b)(3) determination.
B. Individual Evidence of Injury
State Farm also argues that individual issues predominate over common issues because Plaintiffs cannot prove liability without demonstrating that each insured and provider suffered injury. Because we agree with State Farm's first argument and reverse the court of appeals' determination that Plaintiffs have a “conceivable” class-wide theory of proof, we find no need to address this alternative argument regarding the need to individually analyze whether each Plaintiff was injured.
IV.
For the foregoing reasons, we reverse the judgment of the court of appeals.
Justice RICE concurs in the judgment.
Justice EID does not participate.
Justice RICE, concurring in the judgment.
The trial court concluded that although the reasonableness of the medical bills could, in part, be demonstrated by classwide evidence, “ in the final analysis the individualized facts of each repriced bill will have to be examined to answer the question of what is reasonable” (emphasis added). In other words, the trial court determined, by a preponderance of the evidence, see Jackson v. Unocal, 262 P.3d 874 (Colo.2011) (Eid, J., dissenting), that individualized issues would predominate over common issues, thus precluding class certification. Because I find that the trial court did not abuse its discretion in denying certification in this case, I concur in the result reached by the majority.