Docket No. 103287.
IN THE SUPREME COURT OF THE STATE OF ILLINOIS
BARBARA’S SALES, INC., et al., Indiv. and on Behalf of All Others Similarly Situated, Appellees, v. INTEL CORPORATION et al. (Intel Corporation, Appellant).
Opinion filed November 29, 2007.
JUSTICE FITZGERALD delivered the judgment of the court, with opinion. Justices Freeman, Kilbride, Garman, and Karmeier concurred in the judgment and opinion. Chief Justice Thomas and Justice Burke took no part in the decision.
OPINION
In the early part of this decade, Intel Corporation (Intel) engaged in a massive worldwide advertising campaign touting the high performance of its “Pentium 4” microprocessor. The alleged disappointment of a nationwide group of purchasers led to this class action filed in Madison County, Illinois, by named plaintiffs from Illinois and Missouri against Intel, a Delaware corporation with its principal place of business in California. With alternate counts under California and Illinois consumer fraud laws, plaintiffs alleged that Intel deceived the entire class with a false representation implicit in the name Pentium 4, that the microprocessor was the best and fastest processor on the market. The circuit court ruled that Illinois substantive law controls this case and certified a class of Illinois consumers only. Pursuant to Supreme Court Rule 308 (155 Ill. 2d R. 308), the circuit court certified questions for an interlocutory appeal of this ruling. The appellate court answered that California law governs and that the circuit court should reconsider its class certification order in light of California law. 367 Ill. App. 3d 1013. We allowed Intel’s petition for leave to appeal. 210 Ill. 2d R. 315(a). For the following reasons, we conclude that Illinois law governs this case and that class certification was improper.
BACKGROUND The record reveals the following background information which, although not directly relevant to the named plaintiffs, is pertinent to the motion for class certification. A computer’s microprocessor is often referred to as the “brain” of a computer. The performance of microprocessors has steadily increased since they were first introduced in the 1970s. A principal measure of performance for the common consumer is speed. As the speed of a processor increases, the more instructions a processor can process, resulting in less time a computer requires to open software applications, refresh screens, and depict ever more realistic video game characters. One historical measure of speed is called “clock speed,” which is measured in hertz. Intel’s first microprocessor, the 4004, ran at 108 kilohertz per second (108,000 hertz), compared to the Intel Pentium 4 processor’s initial speed of 1.4 gigahertz per second (1.4 billion hertz). Intel has marketed various performance advancements in succeeding generations of microprocessors under advancing brand names such as the 286, 386, 486, Pentium, Pentium Pro, Pentium II, Pentium III, and the Pentium 4. At issue is whether the initial version of the Pentium 4 microprocessor, known within Intel as the “Willamette” family (hereinafter “Pentium 4,” “P4” or “Willamette”), lived up to Intel’s explicit and implicit representations as to its advancement in performance over the Pentium III and the processors of a competitor manufacturer, American Micro Devices (“AMD”). Intel introduced the Pentium 4 in November 2000 and shipped its one millionth Pentium 4 processor sometime in the first quarter of
-2- 2001. Along with various new features in its architecture, these processors had higher clock speeds than the Pentium III and AMD processors. Further, computers with a Pentium 4 processor were priced, at least initially, at a premium over similarly equipped computers containing a Pentium III processor. However, the actual superiority of the Willamette Pentium 4 over the Pentium III and the AMD processor was in doubt. The record reveals various internet and mass media reports questioned the Pentium 4’s performance immediately after it was released. These reports noted that superior clockspeed does not tell the whole story as to the actual performance of a microprocessor. Several sources criticized the microarchitecture underlying the Pentium 4 as being “marchitecture.” In other words, Intel’s representations as to high clock speeds were a deliberate marketing attempt to make it appear faster to the uninformed consumer than the slower-clocked Pentium III and AMD processors. Performance tests, commonly known as benchmarks, showed that the slower clocked Pentium III and an AMD processor were “faster” than the Pentium 4. The extent of the speed discrepancy between the processors depended on the benchmark and was often measured in milliseconds. Other benchmarks noted the Willamette’s excessive heat dissipation and power usage. For example, the Pentium 4 was slower at some common office applications using older operating software, but faster at the video game Quake. Another common criticism of the Willamette Pentium 4 concerned its memory capabilities. Intel’s public response to these criticisms varied. According to Intel, testing software had not been optimized for the Pentium 4. Further, according to Intel, the Pentium 4’s greatest advances were in areas such as 3D gaming, digital video creation, MP3 encoding, and streaming video. Intel also emphasized that the new microarchitecture had a high potential for increased performance as the manufacturing process improved. Intel explained that differences in system hardware and software design may affect actual performance for particular users, apart from the performance of the actual microprocessor. Finally, Intel emphasized that the processor itself was not defective, and that it performed well even on those benchmarks that labeled it “slower.”
-3- On June 3, 2002, plaintiffs filed a nationwide class action complaint asserting consumer fraud claims against Intel, Gateway Inc., Hewlett Packard Company, and HP Direct, Inc.1 The plaintiffs’ original complaint alleged that Intel misled the public by asserting in public statements that the Pentium 4 was the “highest performance processor.” Intel also allegedly suppressed and concealed the Pentium 4’s lack of performance gains over the Pentium III. Plaintiffs brought consumer fraud claims under California’s Unfair Competition Law (Cal. Bus. & Prof. Code §17200 (Deering 2007)), the California Consumer Legal Remedies Act (Cal. Civ. Code §1750 et seq. (Deering 2005)), and, alternatively, under the Illinois Consumer Fraud and Deceptive Business Practices Act (Consumer Fraud Act) (815 ILCS 505/1 et seq. (West 2002)). Plaintiffs sought an award of actual damages, restitution, attorneys’ fees, prejudgment and postjudgment interest, and their costs of suit, amounting to cumulatively less than $75,000 per class member. Intel filed several motions to dismiss. The circuit court denied Intel’s first motion to dismiss the California counts premised upon choice-of-law principles. A motion to dismiss based upon forum non conveniens principles was similarly denied. Intel also moved to dismiss the Illinois Consumer Fraud Act counts, arguing, inter alia, that plaintiffs failed to state a cause of action because of a failure to allege proximate cause. Specifically, Intel argued that plaintiffs were required, under Oliveira v. Amoco Oil Co., 201 Ill. 2d 134 (2002), to allege actual deception of the named plaintiffs. As no named plaintiff was allegedly aware of any specific representation made by Intel, these representations could not have proximately caused plaintiffs’ injuries. The circuit court agreed and dismissed plaintiffs’ Illinois Consumer Fraud Act counts. Plaintiffs subsequently added allegations in their first amended and second amended complaints that they were actually deceived by Intel. The circuit court found, over yet another motion to dismiss based on Oliveira, that these new allegations adequately stated a cause of
1 The claims against Gateway were severed due to arbitration agreements, while Hewlett Packard and HP Direct remain named defendants in this lawsuit but have not filed briefs in this case.
-4- action. These allegations included: (1) “Plaintiffs and Class Members have been actually deceived by Defendant’s failure to disclose material information and/or by their affirmative misrepresentations,” and (2) “Intel has conditioned the consumer, through its marketing and naming practices, to believe that each of its high-performance processors is superior in speed and performance to the previous model.” On April 24, 2004, plaintiffs filed the current, third amended complaint, which included another relevant allegation that Intel, through its marketing practices, “conditioned the market to believe that megahertz measures relative performance and that a processor with a higher clock speed will deliver faster performance.” Plaintiffs thereafter moved for certification of the nationwide class. The evidence submitted by both parties is voluminous, and we highlight only those portions of the evidence submitted to the court that are necessary for this opinion. Plaintiffs relied on extensive evidence obtained in discovery concerning Intel’s marketing practices and the performance of Pentium 4 processors. Plaintiffs stressed Intel’s massive advertising campaign and its overall strategy for that campaign. Plaintiffs noted that as part of Intel’s worldwide billion dollar marketing scheme, the Pentium 4 campaign garnered roughly $300 million worldwide and $100 million in the United States. Those dollar amounts included Web site promotions, television, radio and print advertisements, informational brochures, product demonstrations, training of retail sales agents, and promotional events. Intel made a number of public statements that its processor was the highest performing processor on the market. Several statements made by computer manufacturers that the Pentium 4 was the “fastest” were also highlighted in the motion. Plaintiffs’ essential complaint common to all class members is that Intel was teaching the market that “4 is better than 3.” As such, plaintiffs highlighted several statements by computer manufacturers that the Pentium 4 processor was the “fastest” processor on the market. Plaintiffs repeatedly emphasized an Intel “consumer campaigns overview” slide shown to a group at Comp USA, which they assert summarized Intel’s core strategy. It stated, in part: “I would like to focus on what we will do to promote the Pentium 4 processor this year, which is at the core of this strategy.
-5- Specifically, why buy a P4P? The most important thing is our Brand and what it stands for. We have taught the market that four is better than three.” Plaintiffs also directed the circuit court’s attention to several other places in the record to demonstrate that Intel was trying to teach the market that “four is better than three.” The “four is better than three” scheme of which plaintiffs complain is also reflected, according to plaintiffs, in the deposition testimony of Intel employees Ann Lewnes and Pam Pollace. Lewnes, Intel’s vice- president of sales and marketing and the director of the “Intel Inside” program, testified in her deposition “that the name Pentium 4 is intended to communicate that it’s better than Pentium III.” Lewnes testified that she was unaware if Intel had communicated to the public that the Pentium 4 was not better for particular usage models. She was also unaware if the word “best” was ever used in connection with the marketing of the Pentium 4. Pam Pollace, head of Intel’s Corporate Marketing Group, stated in her deposition that: “Our intent was that 4 was better in many ways than III,” and that “it was better for many aspects.” Plaintiffs submitted expert opinions concerning the effect of Intel’s marketing campaign. One of plaintiffs’ marketing experts, Dr. Tulem Erdem, opined in an affidavit that processor performance matters to consumers in the personal computer market and is a key driver of consumer decisionmaking. Professor Urdem stated, “Intel strategically chose the Pentium 4 sub-brand to signal higher overall performance compared to Pentium 3. To amplify this effect, Pentium 4 was offered at higher clock speed levels than Pentium 3.” He also asserted, “Intel communications clearly positioned Pentium 4 as Intel’s ‘best,’ Pentium 3 as Intel’s ‘better’ and Celeron as Intel’s ‘good’ performance processors. Pentium 4 was also depicted as the best performing chip in the industry at the time.” According to another of plaintiffs’ marketing experts, Dr. Joel Cohen, Intel communicated to purchasers that the Pentium 4 processor was the fastest and highest processor available for PCs. In other words, according to Dr. Cohen, it is a fact that the Pentium 4 processor name universally communicates increased performance as compared with the Pentium III processor name, and that
-6- representation was material to purchasers. Dr. Cohen stated that it was reasonable for purchasers to believe the promises of increased speed and performance of the processor. Intel’s market segmentation of consumers on a spectrum of those having little computer knowledge, to those highly knowledgeable, does not diminish the fact that Intel’s implicit promise of increased performance associated with the Pentium 4 would be reasonably believed by all segments of the class. Further, Intel failed to properly qualify its representations and promises of increased performance associated with the Pentium 4 processor. Dr. Cohen concluded, “To the extent that Pentium 4 based personal computers do not deliver these expected results, the naming, advertising, marketing and promotion efforts by Intel are misleading and deceptive.” Each plaintiff testified in deposition form and in interrogatory answers that they were misled by the “four is better than three” marketing scheme. Several named plaintiffs submitted the following, virtually identical, interrogatory answer and affirmed these answers in deposition testimony: “Plaintiff states that while she remembers seeing advertisements, printed materials and commercials about Pentium 4/P4 processors in personal computers, she cannot specifically remember any particular advertisement or statement. She based her decision to purchase the computer on the label Pentium 4/P4 which meant to her that the computer was faster than the Pentium III or PIII.” Other named plaintiffs could not recall any specific advertisements, from any source, containing a specific statement made by Intel that the Pentium 4 microprocessor was faster than the Pentium III processor. Each named plaintiff took advantage of sources of information other than the mere name “Pentium 4.” These included advertisements from computer manufacturers; consultations with friends, family, and sales clerks; and/or feature and price comparisons at retail stores and online sources. Intel submitted a number of expert opinions in opposition to the motion for class certification, which comprise many pages of the record. The gist of these opinions was, however, that a computer purchaser does not make a decision simply based on the number 4 being higher than the number 3. These experts stated that the purchase
-7- of a computer was a highly individualized decision based on many factors including price, memory, and the brand name of the original equipment manufacturer, and advice from various sources. Further, Dr. John Lynch, Intel’s marketing expert, disagreed with the implication that they were teaching the market that “4 was better than 3.” He explained, “I don’t think we’re actually teaching the market that 4 is better than III. I think we were telling people the benefits of the Pentium 4.” In other words, plaintiffs could not have purchased a computer based on the name “Pentium 4” alone. The parties also submitted contrary opinions on whether it could be proved that the Willamette Pentium 4 was outperformed by the Pentium III and AMD processors. Plaintiffs submitted the testimony from a computer expert, Dr. Edward Davidson, who opined that the speeds of the Pentium 4 and the Pentium III could be compared under one benchmark. Intel submitted an expert opinion disputing Dr. Davidson’s opinion that one benchmark could indicate the superiority of any particular version of the Willamette Pentium 4 processor over particular versions of the Pentium III processor and AMD processors. After discovery, briefing, and arguments, the circuit court entered its class certification order. The circuit court found that Illinois retained the more significant relationship to the controversy over California, thus requiring the application of Illinois law. The circuit court also found that Illinois law could not be applied to a nationwide class, and class certification extended only to consumers who lived in or purchased a computer in Illinois. The circuit court held the numerosity requirement was met because of the sheer size of the computer sales figures in the State of Illinois. The circuit court further held that, with respect to claims of Illinois purchasers under the Illinois Consumer Fraud Act, common questions of fact and law predominated over individual issues. The court found nine common issues, including: “[w]hether the Willamette 4 processor is in fact faster and more powerful than the Pentium III, Celeron, and AMD Athlon processor”; “[w]hether the Defendant intended the public to be misled into believing that the Pentium 4 processor was faster and/or more powerful and/or superior to the Pentium III and/or the AMD processor”; and “whether the Defendant’s conduct is in violation of the [Illinois Consumer Fraud Act].” The circuit court also noted that Intel raised several defenses that were common to all class
-8- members, including that its advertising, marketing, and naming practices were mere puffery that no reasonable consumer would rely on. Finally, the circuit court rejected Intel’s argument that “questions regarding consumer expectations, knowledge, technical sophistication and the alleged complexity of determining the performance of the Pentium 4 processor should weigh against certification.” Accordingly, the trial court certified the following class: “All persons or entities, who reside in Illinois or purchased in Illinois, and that are end use purchasers of ‘Willamette’ Pentium 4 branded processors or computers containing ‘Willamette’ Pentium 4 branded processors that are used for personal, family or household services.” Both parties sought interlocutory review of this ruling, and the circuit court certified the following question for review under Supreme Court Rule 308 (155 Ill. 2d R. 308): “Whether the circuit court erred in certifying a class of Illinois consumers under Illinois law, rather than certifying a nationwide or Illinois class under California law (as plaintiffs requested) or holding that the action should not proceed as a class action (as Intel requested).” The appellate court reversed the circuit court’s choice-of-law ruling–not only as to Illinois residents, but as to all putative class members nationwide–finding that California law applied. 367 Ill. App. 3d at 1021. The appellate court based its decision largely on section 6 of the Restatement (Second) of Conflict of Laws. 367 Ill. App. 3d at 1021, citing Restatement (Second) of Conflict of Laws §6 (1971). The appellate court also directed the circuit court to reconsider its class certification order in light of California law. It cautioned, “our decision today should not, however, be considered a new rule requiring our courts to always apply the consumer fraud law of a foreign state to adjudicate the claims of Illinois residents for products purchased in Illinois” and that the decision was “a narrow one.” 367 Ill. App. 3d at 1020. The appellate court also found our decision in Avery v. State Farm Mutual Automobile Insurance Co., 216 Ill. 2d 100 (2005), “not relevant,” as it did not deal with choice-of-law precedent. 367 Ill. App. 3d at 1021.
-9- We granted Intel’s petition for leave to appeal. 210 Ill. 2d R. 315(a). We subsequently granted leave to the “Conflict of Law Professors” to file an amicus curiae brief in support of the plaintiffs. DaimlerChrysler Corporation and the Product Liability Advisory Counsel, Inc., were also granted leave to file amicus curiae briefs in support of defendant.
DISCUSSION Our review of this certified question is governed by Supreme Court Rule 308 (155 Ill. 2d R. 308(a)). We are limited to the question certified by the trial court, which, because it must be a question of law and not fact, is reviewed de novo. Vision Point of Sale, Inc. v. Haas, 226 Ill. 2d 334, 342 (2007); Thompson v. Gordon, 221 Ill. 2d 414, 426 (2006). The certified question presents us with two major issues: (1) which law to apply to plaintiffs’ nationwide class action suit, and (2) under that law, whether to allow the motion for class certification.
I. Choice of Law Choice-of-law problems arise when significant aspects of a lawsuit traverse state borders. These problems are numerous in this case. Intel is a Delaware corporation with its principal place of business in California. The Intel employees responsible for designing and marketing the “Willamette” version of the Pentium 4 microprocessor were primarily located in California, and also located in Oregon. Intel’s decision to brand the Willamette processors with the Pentium 4 name took place in California. Intel’s Press Relations group, the Intel Inside program, the Brand Campaign Strategy group, and Reseller Products Group are all located in California. Intel made its marketing decisions in California. It conducted marketing activities relating to the Willamette Pentium 4 to original equipment manufacturers and consumers throughout the nation. Intel retains offices and production plants throughout the world. Eight of the eleven named plaintiffs are Illinois residents, six of whom purchased their computers in Illinois. The remaining two Illinois plaintiffs purchased their computers over the telephone from the Dell Corporation in Texas. The other three plaintiffs are from Missouri, two of whom purchased their computers in Missouri, and
-10- one of whom purchased her computer over the telephone from Dell. All plaintiffs received and relied upon Intel’s communications–the alleged implicit representation of performance contained in the Pentium 4 name–in Illinois or Missouri. The named plaintiffs represent a nationwide class of consumers who have made purchases and received representations in all 50 states and the District of Columbia. We need not engage in a choice-of-law analysis if the consumer fraud laws of the states are all similar, i.e., the differences between the laws would not be outcome determinative. However, the parties agree as to the existence of substantial differences in the laws of the 50 states. See, e.g., Tracker Marine, L.P. v. Ogle, 108 S.W.3d 349, 352- 55 (Tex. Ct. App. 2003) (surveying cases). These substantial differences in the states’ laws are the varying degrees to which the law allows a private right of action and class actions, the limitations periods, what constitutes a violation, what form of scienter is required, what form of reliance is required, and damages. Tracker Marine, 108 S.W.3d at 352-53. As the Supreme Court has noted, “the States need not, and in fact do not, provide such protection in a uniform manner. *** The result is a patchwork of rules representing the diverse policy judgments of lawmakers in 50 States.” BMW of North America, Inc. v. Gore, 517 U.S. 559, 569-70, 134 L. Ed. 2d 809, 822-23, 116 S. Ct. 1589, 1596 (1996); accord In re Bridgestone/Firestone, Inc., 288 F.3d 1012, 1018 (7th Cir. 2002); Tracker Marine, 108 S.W.3d at 352-55; Lyon v. Caterpillar, Inc., 194 F.R.D. 206, 219-21 (E.D. Pa. 2000). Hence, the conflict among these laws dictates that we cannot affirm the appellate court’s application of California law on that basis alone. Plaintiffs’ framing of their cause of action narrows our analysis. Although the instant lawsuit implicates the class members and consumer fraud laws of all 50 states and the District of Columbia, the plaintiffs, as the masters of their complaint, seek relief only under Illinois or California law. Plaintiffs have not sought a separate cause of action under Missouri law. Indeed, the certified question asks only “Whether the circuit court erred in certifying a class of Illinois consumers under Illinois law, rather than certifying a nationwide or Illinois class under California law (as plaintiffs requested).” Further, we have recently held that the Illinois Consumer Fraud Act applies only to fraudulent transactions which take place “primarily and
-11- substantially” in Illinois. Avery, 216 Ill. 2d at 186. Therefore, while the Missouri and Illinois contacts are relevant to whether California law should be applied, transactions occurring entirely outside Illinois are not relevant to whether Illinois law should be applied because our decision in Avery precludes relief arising out of these transactions. Therefore, we need only concentrate on California and Illinois law. Importantly, the parties concede that Illinois and California laws conflict and this conflict may have an outcome-determinative difference. In short, plaintiffs need not prove actual deception of the named plaintiffs under California law as found in California’s Unfair Competition Law, as they would under Illinois law as found in the Illinois Consumer Fraud Act. Compare 815 ILCS 505/10a (West 2002); Shannon v. Boise Cascade Corp., 208 Ill. 2d 517, 525 (2005) (“deceptive advertising cannot be the proximate cause of damages under the Act unless it actually deceives the plaintiff”); with Cal. Bus. & Prof. Code §17200 (Deering 2007); Massachusetts Mutual Life Insurance Co. v. Superior Court, 97 Cal. App. 4th 1282, 1288, 119 Cal. Rptr. 2d 190, 193 (2002) (individualized proof of deception and reliance is not required under the Unfair Competition Law). Stated differently, specific class members proceeding under California law need prove only an inference of “common reliance” on the part of the class as opposed to actual reliance on any particular deception. Massachusetts Mutual Life Insurance Co., 97 Cal. App. 4th at 1293, 119 Cal. Rptr. 2d at 198. Because California and Illinois law conflict, we next look to the conflicts law of our forum state, Illinois. Esser v. McIntyre, 169 Ill. 2d 292, 297 (1996); Nelson v. Hix, 122 Ill. 2d 343 (1988); Restatement (Second) of Conflict of Laws §122 (1971). Illinois has adopted the approach found in the Second Restatement of Conflict of Laws. Morris B. Chapman & Associates, Ltd. v. Kitzman, 193 Ill. 2d 560, 568 (2000); Esser v. McIntyre, 169 Ill. 2d 292, 297 (1996). Following the Restatement, we apply the broad principle that the rights and liabilities as to a particular issue are to be governed by the jurisdiction which retains the “most significant relationship” to the occurrence and the parties. Kitzman, 193 Ill. 2d at 568; Ingersoll, 46 Ill. 2d at 44; Restatement (Second) of Conflict of Laws, Introduction, at VII-VIII (1971). We have jettisoned the lex loci delicti rule–also termed the place-of-the-injury rule (Ingersoll, 46 Ill. 2d at 47)–and do not merely count contacts, recognizing that other
-12- jurisdictions may have an interest in the controversy that are not adequately reflected by a simple tally. Rather, our approach in seeking the appropriate applicable law is informed by the issues raised. The parties agree that both Illinois and California have significant relationships to this consumer fraud action, but differ as to which relationship is paramount. Plaintiffs emphasize that Intel’s injury- causing conduct occurred in California and also that the false representations emanated from California. Intel asserts Illinois possesses the more significant contacts, as it is the place where the majority of the plaintiffs received the representations, the place where their reliance occurred, and the place of injury. The plaintiffs urge this court to follow the appellate court in relying on section 6 of the Restatement, which applies general principles to all causes of action. Plaintiffs additionally assert that section 145, which applies general principles to all torts, should also be used as the framework for our decision. Intel responds that section 148 of the Restatement provides the best means for analyzing the issues presented in this suit, as it applies specifically to fraudulent misrepresentation cases. Section 6 of the Restatement sets forth seven elementary principles for choice-of-law determinations, the following of which are relevant here: (a) the relevant policies of the forum state, Illinois; (b) the relevant policies of California and the relative interests of California in the determination of the particular issue; and (c) the ease in the application of the law to be applied. Restatement (Second) of Conflict of Laws §6(2) (1971); Kitzman, 193 Ill. 2d at 568-69 (utilizing relevant factors); Esser, 169 Ill. 2d at 299 (same); Nelson, 122 Ill. 2d at 350-51 (same). As the Restatement notes, the general principles in section 6 “leave *** the answer to specific problems very much at large,” and therefore the Restatement provides “a secondary statement in black letter setting forth the choice of law” rules in a given situation. Restatement (Second) of Conflict of Laws, Introduction, at VIII (1971). The admitted generality found in this commentary leads us to conclude that we should apply the “secondary statement in black letter setting forth the choice of law” found in section 148. Further, we believe “the bench and bar have overemphasized the general sections of the Second Restatement of Conflict of Laws, and have undervalued the specific presumptive rules.” Townsend v. Sears, Roebuck & Co., No. 103858, slip op. at 12
-13- (November 29, 2007). Nevertheless, the employment of these section 6 principles do not compel the application of California law in the present matter. As to the relevant policies of the forum consideration, we have recently held that “Illinois courts have an interest in not being burdened with applying foreign law in the absence of strong policy reasons and a strong connection to the case.” Gridley v. State Farm Mutual Automobile Insurance Co., 217 Ill. 2d 158, 175 (2005) (rejecting the application of Louisiana law under forum non conveniens principles). We can identify no strong policy reasons or strong connection to direct us to apply California law. First, Illinois might prefer to apply its law if it would better protect Illinois consumers. However, because California does not require “individualized deception,” California law seems to better protect consumers. Yet, it is not apparent that any individual consumer will not be able to pursue a cause of action under Illinois law for his or her individual injuries caused by Intel or the specific computer manufacturer. Thus, there does not appear to be any interest on the part of Illinois in the application of its standards in preference or deference to California’s standards to protect Illinois consumers. Next, the majority of the named plaintiffs are Illinois residents who received Intel’s representations and purchased their computers in Illinois or over the phone or internet from their homes. The remainder of the named plaintiffs are Missouri residents. Conversely, Illinois has no connection with the nationwide class consisting of foreign residents purchasing computers containing the Pentium 4 in a foreign state. Further, we have recently held that the Consumer Fraud Act applies only to fraudulent transactions that take place “primarily and substantially” inside Illinois. Avery, 216 Ill. 2d at 186. Thus, the relevant policy interest of Illinois would be to apply Illinois law to the claims of Illinois consumers, while excluding those claims which do not have a strong connection to Illinois. Next, as to the policies and interests of California, it is undoubtedly true that California has an interest in regulating Intel, as its principal place of business is located there. Further, it is also true that California has a consumer-friendly consumer protection law–as a suit may be brought alleging “common reliance” rather than “actual deception”–which may inure to the benefit of plaintiffs. However,
-14- neither California consumers, nor the interests of California in regulating Intel, will necessarily suffer if Illinois law is applied in the instant matter. California has no interest in extending its laws to noncitizens and to actions that occurred outside of California borders. Norwest Mortgage, Inc. v. Superior Court, 72 Cal. App. 4th 214, 222, 85 Cal. Rptr. 2d 18, 23 (1999). Moreover, the courts of California will likely have the opportunity to decide this issue under its own law. The record reveals another putative nationwide class action with allegations similar to those advanced in this case. This class action seeks a nationwide application of California’s consumer protection statutes and is currently pending in California state court. Skold v. Intel Corp., Case No. RG 04 145635 (Cal. Super. Ct. Alameda County). Finally, as to section 6’s third relevant factor, ease of application, it is self-evident that while use of California law may not be difficult, the use by a forum of its own laws does not present any further difficulty. See, e.g., Tracker Marine, 108 S.W.3d at 358. The appellate court reversed the circuit court on a misinterpretation of another general factor from section 6, a factor which is not particularly relevant to our analysis. Without citation, the appellate court stated, “[t]he needs of the interstate system *** require one forum with one result rather than results in 51 jurisdictions with the distinct possibility of conflicting decisions.” 367 Ill. App. 3d at 1019. This declaration completely ignores the distinct interests of the differing states embodied in our federalist system and constitutional precedent. BMW of North America, Inc. v. Gore, 517 U.S. 559, 570-71, 134 L. Ed. 2d 809, 823-24, 116 S. Ct. 1589, 1596- 97 (1996); In re Bridgestone/Firestone, Inc., 288 F.3d 1012, 1018 (7th Cir. 2002). Further, to satisfy due process guarantees, the Supreme Court has explained that a court “may not take a transaction with little or no relationship to the forum and apply the law of the forum in order to satisfy the procedural requirement that there be a ‘common question of law’ ” Phillips Petroleum Co. v. Shutts, 472 U.S. 797, 821-23, 86 L. Ed. 2d 628, 648-49, 105 S. Ct. 2965, 2979- 80 (1985). Moreover, an examination of the reasoning behind this factor shows that it is not fit for the appellate court’s purpose to obtain “one forum with one result.” The goals of the “needs of the interstate
-15- system” principle are “to make the interstate and international systems work well,” to promote “harmonious relations,” and “to facilitate commercial intercourse between them.” Restatement (Second) of Conflict of Laws §6, Comment d (1971). The Restatement also directs courts to strive to adopt “the same choice of law” rules reflected in other states’ precedent. Restatement (Second) of Conflict of Laws §6, Comment d (1971). The application of California law or Illinois law, as plaintiffs urge in this nationwide class, to a citizen of Washington state who purchased his computer in Washington state does nothing to improve the harmonious relations between the states. Thus, we are not persuaded that section 6 of the Restatement compels us to apply California law. We now turn to section 145 of the Restatement. Section 145 sets forth principles to be applied to tort cases, but also is cast in “great generality” and directs that the “best way to bring precision into the field is by attempting to state special rules for particular torts and for particular issues in tort.” Restatement (Second) of Conflict of Laws §145, Comment a (1971); see also Townsend, slip op. at 23 (stating that a court should begin a choice-of law analysis by ascertaining whether a specific presumptive rule applies to the conflict). The Restatement commentary therefore directs us to consider section 148 because that is the most precise section involved. Nevertheless, plaintiffs cite to this passage from section 145, comment f: “the place of injury does not play so important a role for choice-of-law purposes in the case of false advertising *** as in the case of other kinds of torts. Instead, the principal location of the defendant’s conduct is the contact that will usually be given the greatest weight in determining the state whose local law determines the rights and liabilities that arise from false advertising.” Restatement (Second) of Conflict of Laws §145, Comment f (1971). However, this is a misleading discussion of the Restatement commentary discussing “false advertising” claims. The “false advertising” addressed in comment f to section 145 involves a single-business plaintiff injured in multiple states as a consequence of a competitor’s “false advertising.” As comment f explains, in such a circumstance the place of injury has less significance because the individual plaintiff’s injury occurs in multiple jurisdictions: “The injury suffered through false advertising is the loss of customers or of trade. Such customers or trade will frequently be
-16- lost in two or more states.” Restatement (Second) of Conflict of Laws §145 comment f (1971). Therefore, our reading of section 6 and section 145 reveals that section 148 is more appropriate because it is applied more precisely to claims based on false representations and thus provides the proper analytical framework for our “most significant relationship” approach. Section 148 applies to “actions brought to recover pecuniary damages suffered on account of false representations, whether fraudulent, negligent or innocent.” Restatement (Second) of Conflict of Laws §148, Comment a (1971). According to this section, “[w]hen the plaintiff’s action in reliance took place in whole or in part in a state other than that where the false representations were made,” we are to consider these relevant factors: (a) the state where plaintiff acted in reliance upon defendant’s representations, (b) the state where plaintiff received the representations, (c) the state where defendant made the representations, (d) the domicil, residence, place of incorporation, and place of business of the parties, and (e) the place where a tangible thing which is the subject of the transaction between the parties was situated at the time. Restatement (Second) of Conflict of Laws §148(2) (1971). Following these factors, the named plaintiffs’ lawsuit presents us with the majority of contacts in Illinois, while also providing contacts in Missouri, Delaware, California, and possibly Texas. As we do not resolve this issue by a mere counting of contacts, but rather by the issues raised, we examine the listed factors in conjunction with the comments accompanying Restatement section 148. Following factors (a), (b), and (c) above, plaintiffs acted in reliance in Illinois and Missouri; plaintiffs received the representations in Illinois and Missouri; and Intel made the representations in California. Comment g states that the place where plaintiff received the representations “constitutes approximately as important a contact as does the place where the defendant made the representations. On the other hand, this place is not so important a contact as is the place where the plaintiff acted in reliance on the defendant’s representations.” Restatement (Second) of Conflict of Laws §148, Comment g (1971). Here, it is undisputed that plaintiffs both received and relied on the representations in their home states, thus favoring the application of Illinois or Missouri law.
-17- Under factor (d) of section 148, plaintiffs were domiciled in Illinois and Missouri, while Intel’s place of incorporation is Delaware and principal place of business is California. Comment i explains the relative importance of the parties’ domiciles: “The plaintiff’s domicil or residence *** are contacts of substantial significance when the loss is pecuniary ***. This is so because a financial loss will usually be of greatest concern to the state with which the person suffering the loss has the closest relationship. *** The domicil, residence and place of business of the plaintiff are more important than are similar contacts on the part of the defendant.” Restatement (Second) of Conflict of Laws §148, Comment i (1971). The financial loss here was the Illinois and Missouri residents’ and businesses’ payment of money, which occurred in Illinois or Missouri, not California. We agree with the Restatement that this contact is of substantial significance and is the greatest concern to Illinois or Missouri because the persons suffering the loss have the closest relationship there. Further, the plaintiffs’ domicil, Illinois or Missouri, is “more important than are similar contacts on the part of” Intel in California. Restatement (Second) of Conflict of Laws §148, Comment i (1971). According to factor (e), the place where the tangible thing that was the subject of the dispute was located in Illinois and Missouri. The record is not clear as to where the computers purchased over the telephone from Dell were located at the time of the transaction, although Dell is headquartered in Texas. According to comment i, “place where the thing is situated at the time of the transaction is a contact of some importance provided, at least, that both parties were aware that the thing was situated in this place at that time.” Because it is not apparent that Intel had any specific notion where the named plaintiffs’ computers and microprocessors were located at the time of the purchase, the contact is of little significance. The general approach to section 148 is found in comment j, and favors the application of Illinois law. Expounding that no “definite rules” can be stated, it asserts, “If any two of the *** contacts, apart from the defendant’s *** place of business, are located wholly in a single state, this will usually be the state of the applicable law with respect to most issues.” Restatement (Second) of Conflict of Laws
-18- §148, Comment j (1971). Unquestionably, more than two of the contacts for each Illinois and Missouri named plaintiff are located in Illinois and Missouri, not California. As comment j further explains, when a plaintiff purchases goods in the plaintiff’s home state based on representations received in plaintiffs’ home state, the law of that home state–not the defendant’s principal place of business–will usually govern: “So when the plaintiff acted in reliance upon the defendant’s representations in a single state, this state will usually be the state of the applicable law, with respect to most issues, if (a) the defendant’s representations were received by the plaintiff in this state, or (b) this state is the state of the plaintiff’s domicil or principal place of business.” Restatement (Second) of Conflict of Laws §148, Comment j (1971). Here, comment j directs us away from California law, as Illinois and Missouri were the places of plaintiffs’ reliance, where plaintiffs received the representations, and the state of plaintiffs’ domicile or place of business. We reject several of plaintiffs’ arguments against the use of section 148. First, plaintiffs argue that section 148 was written with common law fraud cases, rather than consumer protection cases, in mind. Thus, according to plaintiffs, it should be discounted because “reliance” is not at issue here. That there was some loose form of reliance in this case is so unavoidable that plaintiffs even argue in their briefs that “Every plaintiff testified that he or she purchased Pentium 4 processors in reliance on, and was deceived by, Intel’s conduct.” The “actual deception” and “as a result of” requirement of Illinois law (see 815 ILCS 505/10a (West 2002)) is directly analogous to the broad use of “reliance” in the Restatement; i.e., to identify the place where the plaintiff “purchased” a “tangible thing” based upon the representation at issue. Restatement of Conflict of Laws §148, Comment c (1971). Further, it is undisputed that at least some form of reliance is required under the laws of California, Illinois, and various other states throughout the nation. Plaintiffs’ further attempt to tie this case to California law because Intel’s representation emanated from there does not accord with any previous precedent of any Illinois court. Thus, we are not persuaded by plaintiffs’ citation of our decision in Martin v. Heinhold
-19- Commodities, Inc., 117 Ill. 2d 67 (1987), where we applied Illinois law to a nationwide class. As we explained recently, the decision in Martin was “specifically based” on certain key facts not present in this case. Avery, 216 Ill. 2d at 189. In Martin, defendant’s place of business was in Illinois and was “made manifest to each member” of the plaintiff class. Specifically, this court noted that Martin was based on the fact that: “(1) the contracts containing the deceptive statements were all executed in Illinois; (2) the defendant’s principal place of business was in Illinois; (3) the contract contained express choice-of- law and forum-selection clauses specifying that any litigation would be conducted in Illinois under Illinois law; (4) complaints regarding the defendant’s performance were to be directed to its Chicago office; and (5) payments for the defendant’s services were to be sent to its Chicago office.” Avery, 216 Ill. 2d at 189, citing Martin, 117 Ill. 2d at 82-83. Indeed, apart from the simple fact that both Intel and the defendant in Martin were headquartered in the state whose law plaintiffs sought to apply, there are no other similarities in the facts. This conclusion is indistinguishable from our proclamation in Avery, where we stated, “The appellate court’s conclusion that a scheme to defraud was ‘disseminated’ from State Farm’s headquarters is insufficient.” Avery, 216 Ill. 2d at 189. Plaintiffs’ reliance on Ingersoll v. Klein, 46 Ill. 2d 42 (1970), is also inapposite. That case concerned a traditional tort: a drowning in the Iowa territorial boundaries of the Mississippi River. We stated there that “the local law of the State where the injury occurred should determine the rights and liabilities of the parties, unless Illinois has a more significant relationship with the occurrence and with the parties, in which case, the law of Illinois should apply.” Ingersoll, 46 Ill. 2d at 45. Here, Illinois and Missouri are the locations where plaintiffs were allegedly deceived into purchasing a product, and this transaction is both the place of the “injury” and the place with the most “significant relationship with the occurrence and the parties.” While either Illinois or Missouri law could possibly apply separately to the named plaintiffs’ individual causes of action according to the individual facts, plaintiffs have only sought relief under Illinois law. Further, the certified question presents us with a choice between only Illinois and California. We therefore answer that Illinois law applies here. Hence, we agree with the circuit court’s
-20- ruling and reverse the appellate court’s ruling on this issue. After finding that Illinois law applied to this case, the circuit court certified a class solely consisting of Illinois consumers. Plaintiffs make no further argument that Illinois law should be applied to a nationwide class. Nevertheless, we agree that the circuit court was correct to limit the class to Illinois consumers only, as the Illinois Consumer Fraud Act applies to transactions that occur “primarily and substantially” in Illinois. Avery, 216 Ill. 2d at 186. We now review the propriety of this certification ruling under Illinois law.
II. Class Certification The Code of Civil Procedure allows the maintenance of a class action only if: (1) the class is so numerous that joinder of all members is impracticable; (2) there are questions of fact or law common to the class, which common questions predominate over any questions affecting only individual members; (3) the representative parties will fairly and adequately protect the interests of the class; and (4) the class action is an appropriate method for the fair and efficient adjudication of the controversy. 735 ILCS 5/2–801 (West 2002). The party seeking certification bears the burden of demonstrating that it is appropriate. Gridley, 217 Ill. 2d at 167. We need not determine whether plaintiffs satisfy class requirements because, as a threshold matter, the representation identified by the plaintiffs does not form the basis of an actionable claim under the Consumer Fraud Act. Section 10a(a) of the Consumer Fraud Act authorizes causes of action for deceptive business practices prohibited by the Act. Section 10a(a) states, in relevant part: “Any person who suffers actual damage as a result of a violation of [the] Act committed by any other person may bring an action against such person.” 815 ILCS 505/10a(a) (West 2002). A private cause of action brought under section 10a(a) requires proof of “actual damage.” 815 ILCS 505/10a(a) (West 2002). Section 10a(a) requires proof that the damage occurred “as a result of” the deceptive act or practice. 815 ILCS 505/10a(a) (West 2002). To adequately plead a private cause of action for a violation of section 10a of the Act, a plaintiff must allege: (1) a deceptive act or practice by the defendant, (2) the defendant’s intent that the plaintiff rely on the deception, (3) the occurrence of the deception in the course of conduct involving trade or commerce, and (4) actual damage to the
-21- plaintiff, (5) proximately caused by the deception. Avery, 216 Ill. 2d at 180. The “as a result of” language in section 10a(a) imposes an obligation upon a private individual seeking actual damages under the Act to “demonstrate that the fraud complained of proximately caused” those damages in order to recover for his injury. Oliviera, 201 Ill. 2d at 140, citing Zekman v. Direct American Marketers, Inc., 182 Ill. 2d 359, 373 (1998). Plaintiffs argue that the uniform representation implicit in the name “Pentium 4”–allegedly that this processor was the best and fastest on the market–was sufficient to afford recovery under the Consumer Fraud Act. Intel responds that this implicit representation is nothing more than puffery, and therefore is not a “deceptive act” within the purview of the Act. We agree with Intel. Puffing denotes the exaggerations reasonably expected of a seller as to the degree of quality of his or her product, the truth or falsity of which cannot be precisely determined. Avery, 216 Ill. 2d at 173-74. “Puffing in the usual sense signifies meaningless superlatives that no reasonable person would take seriously, and so it is not actionable as fraud.” Speakers of Sport, Inc. v. Proserv, Inc., 178 F.3d 862, 866 (7th Cir. 1999), citing Noll v. Peterson, 338 Ill. 552 (1930); see also Black’s Law Dictionary 1269 (8th ed. 1999) (“the expression of an exaggerated opinion–as opposed to a factual misrepresentation–with the intent to sell a good or service. • Puffing involves expressing opinions, not asserting something as fact. Although there is some leeway in puffing goods, a seller may not misrepresent them or say that they have attributes that they do not possess”). In Avery, this court held that the statement that the product was a “quality replacement part” was puffing. We also cited to a string of cases that identified the following phrases as puffery: “high-quality,” “expert workmanship,” “custom quality,” “perfect,” “magnificent,” “comfortable,” and “picture perfect.” Avery, 216 Ill. 2d at 174. Importantly, we further noted that words such as “best” are not actionable representations of fact. Avery, 216 Ill. 2d at 174 (stating “ ‘A general statement that one’s products are best is not actionable as a misrepresentation of fact’ ”), quoting State v. American TV & Appliance of Madison, Inc., 146 Wis. 2d 292, 302, 430 N.W.2d 709, 712 (1988). Plaintiffs contend that Intel conditioned the market to believe that each generation of the “Pentium” processor would be better than the
-22- last and that the Pentium 4 was “the best,” as it was the latest generation processor. In support of that position, plaintiffs highlight a number of statements made by Intel showing that it sought to teach the market that “four is better than three.” Plaintiffs also point to their interrogatory answers and deposition testimony which state that the individual named plaintiffs relied on the name “Pentium 4” in making their purchases. But plaintiffs can only identify one statement, the name “Pentium 4,” that was communicated to the entire class. Because this is indistinguishable from the use of the term “best,” a statement we found not actionable as puffery in Avery, we also find that the implicit representation inherent in the name Pentium 4 is mere puffery. Avery, 216 Ill. 2d at 174, quoting American TV & Appliance of Madison, Inc., 146 Wis. 2d at 302, 430 N.W.2d at 712. Our precedent disallowing such actions premised on puffery is based on the sound reasoning that no reasonable consumer would rely on such an implicit assertion as the sole basis for making a purchase. A reasonable consumer would not rely on it because there is nothing specific or explicit about the name “Pentium 4,” or even its alleged implicit meaning of “4 is better than 3” or “best,” that can be said to have a specific attribute other than the actual processor itself. One can say almost anything is the “best” because it is a mere suggestion, a belief, or an opinion. It is not, standing alone, a representation of fact. Though plaintiffs may have correctly felt that “4 meant better than 3,” it is merely, as plaintiffs call it, “implicit” that it was best. No concrete inferences may be made. Further, an implicit representation is not an affirmative one, and plaintiffs point to no affirmative representations made uniformly to the class as a whole. Indeed, no named plaintiff solely consulted the name “Pentium 4” before making the purchase. For example, plaintiff Donald Braddy consulted a Dell advertisement touting the Pentium 4’s high performance as well as viewed different computers in retail stores before making his purchase on Dell’s Web site. Plaintiffs mistakenly make much of their expert’s opinion that the “performance” of any Pentium 4 can be compared to any Pentium III according to a single benchmark. This is an issue the parties dispute and, if this case were to go forward, would need to be resolved at trial. However, there is nothing in the “4 is better than 3” marketing formulation that presents more than a vague promise of something
-23- better. Notably, the word “performance” was not a word that was communicated to the entire class. As such, plaintiffs’ theory requires vague suggestion upon vague suggestion: that “best” means “performance,” that “performance” means “speed,” and that “speed” can be tested for all P4s and P3s according to one benchmark. An example illustrates our point. The parties agree that it can be determined whether the 1.4 Ghz Willamette Pentium 4 is “faster” at the video game Quake than a specific Pentium III run on the same computer model with the same configuration. But this stands in fierce relief with a claim that the Pentium 4 is “better” or “best,” a claim so vague as to leave the standards for interpretation open to a number of plausible criteria for judgment. This could mean that the Pentium 4 is cheaper, smaller, more reliable, of higher quality, better for resale, more durable, creates less heat, uses less electricity, is more compatible with some versions of software, or is simple the latest in a temporal line of processors. Because the term “better” as a mere suggestion in the name “Pentium 4” is not capable of precise measuring, it is mere puffery and therefore not actionable under our precedent. That is true even if, as the appellate court concluded, “Intel specifically set out to show [the market] that the Pentium 4 was the best processor to date.” 367 Ill. App. 3d at 1015. Tangentially, we also note that a number of additional statements made by Intel are not actionable because no plaintiff was aware of these statements. Under Oliveira and its progeny, plaintiffs must prove that each and every consumer who seeks redress actually saw and was deceived by the statements in question. Avery, 216 Ill. 2d at 200, citing Oliveira, 201 Ill. 2d at 155; see also Zekman, 182 Ill. 2d at 373; Shannon, 208 Ill. 2d at 525. In their briefs, plaintiffs cite internal documents relating to Intel’s role in developing new benchmarks as allegedly deceptive to consumers. Internal Intel documents also show that Intel was aware of the Pentium 4’s inferiority on certain tests. Nevertheless, it is not apparent that Intel ever made any false public claims as to benchmarking. Further, plaintiffs do not argue that Intel made any uniform misrepresentation regarding clock speed, or falsely claimed that its product possessed a clock speed that it did not have. Intel merely asserted that its product was the highest performance microprocessor, a representation which the class members do not allege they all received in common.
-24- Accordingly, the class members’ uniform representation–plaintiffs’ allegation that “Intel has conditioned the consumer through its marketing and naming practices, to believe that each generation of its high-performance processors is superior in speed and performance to the previous generation” and the name “Pentium 4” was an implicit representation of processor performance that deceived all consumers–is not actionable under the Illinois Consumer Fraud Act. Therefore, we overturn the circuit court’s decision regarding class certification. Avery, 216 Ill. 2d at 125-26; Smith v. Illinois Central R.R. Co., 223 Ill. 2d 441 (2006). Because this is an adequate basis to resolve this dispute, we do not consider Intel’s other argument regarding the individualized purchasing decisions of plaintiffs preventing any predominant questions and damages. We find that the lawsuit as presently constituted should not proceed as a class action.
CONCLUSION For the foregoing reasons, we answer the certified question as follows: Illinois law governs the issues of liability and damages in the present case; the action should not proceed as a class action. Accordingly, we reverse the judgment of the appellate court and remand the cause to the circuit court.
Certified question answered; appellate court reversed; cause remanded.
CHIEF JUSTICE THOMAS and JUSTICE BURKE took no part in the consideration or decision of this case.
-25-