dissenting:
It has been more than seven decades since the Supreme Court ended the “mischievous” regime of Swift v. Tyson, 41 U.S. 1, 16 Pet. 1, 10 L.Ed. 865 (1842), in which federal courts sitting in diversity disregarded state court decisions and independently determined the meaning of state law. Erie Railroad, Co. v. Tompkins, 304 U.S. 64, 74, 58 S.Ct. 817, 82 L.Ed. 1188 (1938). The Eñe doctrine has long required federal courts to “follow the decisions of intermediate state courts in the absence of convincing evidence that the highest court of the state would decide differently.” Stoner v. New York Life Ins. Co., 311 U.S. 464, 467, 61 S.Ct. 336, 85 L.Ed. 284 (1940); see also Ryman v. Sears, Roebuck & Co., 505 F.3d 993, 994 (9th Cir.2007). Today the panel majority returns us to the era of Swift v. Tyson, openly acknowledging that its opinion will lead to forum shopping and the inconsistent enforcement of state law, the very evils that the Eñe Court sought to eradicate. See Erie, 304 U.S. at 74-78, 58 S.Ct. 817.
The majority disregards not one but three intermediate California appellate decisions holding that California Civil Code § 2527 violates Article I, section 2 of the California Constitution. See ARP Pharmacy Servs., Inc. v. Gallagher Bassett Servs., Inc., 138 Cal.App.4th 1307, 42 Cal.Rptr.3d 256 (2006); A.A.M. Health Group, Inc. v. Argus Health Sys., Inc., No. B183468, 2007 WL 602968 (Cal.Ct.App. Feb. 28, 2007); Bradley v. First Health Servs. Corp., No. B185672, 2007 WL 602969 (Cal.Ct.App. Feb. 28, 2007). It does so not because of any convincing evidence that the state high court would rule differently, but because it has convinced itself that its interpretation of federal constitutional law is correct, that the three panels of the Second District Court of Appeal got it wrong, and that the California Supreme Court would side with the views of two federal judges over the seven state appellate judges and two state trial judges who have all ruled to the contrary.
In point of fact, the California Supreme Court denied review of the last of the appellate court decisions, leaving the precedent intact. The failure to follow the intermediate state courts violates the Eñe *1108doctrine and offends important principles of federalism and comity. Even worse, however, it is the majority that fails to correctly apply First Amendment principles to fact-based expression, while endorsing unfettered government authority to compel “objective” speech. Not only am I not convinced that the California Supreme Court would utilize the majority’s flawed analysis of the federal right of free speech to interpret the distinct, and more protective, state constitutional right, I find it highly doubtful. Therefore, I respectfully dissent.
I.
We confront in § 2527 an unusual law without clear analogies in existing precedent. The statute requires drug claims processors to undertake or obtain studies about pharmacy pricing, summarize the results, and transmit the material to their clients. § 2527(c)-(d). Essentially, it requires Business A to speak about Business B to Business C. Unlike a disclosure law, it does not require that regulated entities divulge information about themselves to the public, but rather that they privately produce information about third parties to their clients. Cf. Zauderer v. Office of Disciplinary Counsel of Supreme Court of Ohio, 471 U.S. 626, 651, 105 S.Ct. 2265, 85 L.Ed.2d 652 (1985). Moreover, § 2527 is a stand-alone law that does nothing more than mandate speech. It is not ancillary to any comprehensive economic regulatory scheme. Cf. United States v. United Foods, Inc., 533 U.S. 405, 411-12, 121 S.Ct. 2334, 150 L.Ed.2d 438 (2001).
As our free speech jurisprudence treats “[e]aeh method of communicating ideas [as] ‘a law unto itself,’ ” so must it afford unique treatment to each different method of government mandated communication of ideas. Metromedia, Inc. v. City of San Diego, 453 U.S. 490, 501, 101 S.Ct. 2882, 69 L.Ed.2d 800 (1981) (adding that the “law must reflect the ‘differing natures, values, abuses and dangers’ of each method”) (quoting Kovacs v. Cooper, 336 U.S. 77, 97, 69 S.Ct. 448, 93 L.Ed. 513 (1949)); Wooley v. Maynard, 430 U.S. 705, 714, 97 S.Ct. 1428, 51 L.Ed.2d 752 (1977) (“The right to speak and the right to refrain from speaking are complementary components of the broader concept of ‘individual freedom of mind.’ ”) (quoting Bd. of Educ. v. Barnette, 319 U.S. 624, 637, 63 S.Ct. 1178, 87 L.Ed. 1628 (1943)). The parties have not identified any case that squarely controls the federal or state constitutional analysis of this unique brand of government mandated private speech about third parties.
The California Constitution provides that “[e]very person may freely speak, write and publish his or her sentiments on all subjects, being responsible for the abuse of this right. A law may not restrain or abridge liberty of speech or press.” Cal. Const, art. 1 § 2. This clause “enjoys existence and force independent of the First Amendment” of the United States Constitution. Gerawan Farming, Inc. v. Lyons, 24 Cal.4th 468, 101 Cal.Rptr.2d 470, 12 P.3d 720, 734 (2000). Indeed, “the California liberty of speech clause is broader and more protective than the free speech clause of the First Amendment.” Los Angeles Alliance For Survival v. City of Los Angeles, 22 Cal.4th 352, 93 Cal.Rptr.2d 1, 993 P.2d 334, 342 (2000); see also Kasky v. Nike, Inc., 27 Cal.4th 939, 119 Cal.Rptr.2d 296, 45 P.3d 243, 255 (2002). Therefore, we should be especially hesitant to tell the California courts how to apply their own Constitution to such a unique and unprecedented state law mandating speech.1
*1109II.
The majority identifies two “critical errors” in the Court of Appeal panel decisions that it believes the California Supreme Court would not make: giving insufficient weight to Rumsfeld v. Forum for Academic and Institutional Rights, Inc., 547 U.S. 47, 126 S.Ct. 1297, 164 L.Ed.2d 156 (2006) (“FAIR ”), and misinterpreting Riley v. National Federation of the Blind of North Carolina, Inc., 487 U.S. 781, 108 S.Ct. 2667, 101 L.Ed.2d 669 (1988). However, the California Court of Appeal panels reasonably interpreted both cases, and thus there is no convincing reason to believe that the California Supreme Court would rule differently.
In FAIR, the Court rejected a First Amendment challenge to the Solomon Amendment, a statute restricting federal funding to universities that do not grant military recruiters comparable access to other employers looking to hire at their law schools. 547 U.S. at 52-53, 126 S.Ct. 1297. The Court recognized that the “recruiting assistance provided by the schools often includes elements of speech” as “schools may send e-mails or post notices on bulletin boards on an employer’s behalf.” Id. at 61-62, 126 S.Ct. 1297. However, the Court concluded that this marginal compulsion of speech did not violate the constitution. The majority analogizes the pricing reports from § 2527 to the hypothetical e-mails and bulletin board postings in FAIR, and holds that because of its factual nature, such compelled speech does not give rise to First Amendment scrutiny. This dramatically overstates the holding of FAIR, as the Supreme Court specifically acknowledged that “these compelled statements of fact ... are subject to First Amendment scrutiny.” Id. at 62, 126 S.Ct. 1297.
The FAIR Court did not find a constitutional violation because the particular factual statements at issue were both hypothetical and ancillary to a comprehensive regulatory regime. The Court determined that the e-mails and bulletin board postings would only be “ ‘compelled’ if, and to the extent, the school provides such speech for other recruiters,” and that such compulsion would be “plainly incidental to the Solomon Amendment’s regulation of conduct.” Id. That stands in stark contrast to § 2527, where the compulsion is not contingent on any voluntary conduct by the regulated party, and where it is not ancillary to any comprehensive regulatory scheme. Rather, § 2527 is a direct, standalone government mandate of speech. Therefore, the Court of Appeal panels’ treatment of FAIR in this context was not erroneous, and it does not provide convincing evidence that the California Supreme Court would rule differently.2
*1110As for Riley, the Court of Appeal decisions relied on the case for the proposition that § 2527 warrants constitutional scrutiny even though it compels “essentially statistical information.” ARP, 42 Cal.Rptr.3d at 261. The majority argues that Riley stands for the narrower proposition that mandated factual speech only warrants scrutiny if the law has a direct chilling effect on other protected First Amendment speech. However, this confuses the initial inquiry into whether a regulation even implicates the First Amendment with the separate inquiry into whether it survives constitutional scrutiny.
The Riley Court first held quite broadly that “mandating speech that a speaker would not otherwise make necessarily alters the content of the speech. We therefore consider the Act as a content-based regulation of speech.” 487 U.S. at 795, 108 S.Ct. 2667. The Court thus established that the compulsion of factual speech triggered First Amendment analysis before even considering whether the regulation burdened other protected expression. Only then did the Riley Court proceed to discuss Wooley and Barnette, and the broader question of whether the regulation burdened other protected speech, as part of the separate and subsequent inquiries into the precise level of scrutiny to apply and whether the regulation was sufficiently tailored to fit the state interest. Id. at 797-99, 108 S.Ct. 2667.
In other words, the Riley Court held that the particular law compelling speech failed exacting scrutiny because of its chilling effect; it did not hold that a chilling effect is a prerequisite to any First Amendment scrutiny at all. The FAIR Court made this clear when it cited Riley for the proposition that “compelled statements of fact (‘The U.S. Army recruiter will meet interested students in Room 123 at 11 a.m.’), like compelled statements of opinion, are subject to First Amendment scrutiny.” 547 U.S. at 62, 126 S.Ct. 1297. The FAIR Court made no mention of a chill requirement, and indeed it suggested that the potential compulsion of e-mails and bulletin board postings would draw constitutional scrutiny despite the nature of the content. Id.
In faulting the California courts for relying on Riley’s holding about factual speech, the majority makes the stunning assertion that § 2527 is not subject to any First Amendment scrutiny because it requires only the dissemination of “objective” data, and “Defendants are not compelled to convey a viewpoint or perform any subjective analysis of the numbers they report.” No authority is cited for the proposition that compelled speech must contain subjective analysis or overt opinion in order to implicate constitutional rights. Indeed, it is well established that “the First Amendment’s proscription of *1111compelled speech does not turn on the ideological content of the message that the speaker is being forced to carry. The constitutional harm — and what the First Amendment prohibits — is being forced to speak rather than to remain silent.” Axson-Flynn v. Johnson, 356 F.3d 1277, 1284 n. 4 (10th Cir.2004).
The majority’s narrow interpretation of the First Amendment contradicts decades of Supreme Court precedent extending constitutional protection to communications containing truthful information. For instance, the majority’s reasoning fails to account for fact-based news reporting, which is considered protected speech under both the First Amendment and the California Constitution. See, e.g., Lafayette Morehouse, Inc. v. Chronicle Publ’g Co., 37 Cal.App.4th 855, 44 Cal.Rptr.2d 46, 51 (1995) (explaining that it is a “faulty premise ... that news reporting activity cannot be characterized as ‘free speech.’ In fact, courts have consistently described such activity as ‘free speech.’”) (citing Philadelphia Newspapers, Inc. v. Hepps, 475 U.S. 767, 775-776, 106 S.Ct. 1558, 89 L.Ed.2d 783 (1986), Joseph Burstyn, Inc. v. Wilson, 343 U.S. 495, 501, 72 S.Ct. 777, 96 L.Ed. 1098 (1952), and Daily Herald Co. v. Munro, 838 F.2d 380, 384 (9th Cir.1988)); see also Bartnicki v. Vopper, 532 U.S. 514, 527, 121 S.Ct. 1753, 149 L.Ed.2d 787 (2001) (“[S]tate action to punish the publication of truthful information seldom can satisfy constitutional standards.”) (quoting Smith v. Daily Mail Publ’g Co., 443 U.S. 97, 102, 99 S.Ct. 2667, 61 L.Ed.2d 399 (1979)). Moreover, the Supreme Court has extended First Amendment protection to numerous forms of speech that communicate nothing but factual information. See, e.g., Linmark Associates, Inc. v. Willingboro Township, 431 U.S. 85, 96, 97 S.Ct. 1614, 52 L.Ed.2d 155 (1977) (striking an ordinance prohibiting the display of “For Sale” and “Sold” signs in front of houses); Virginia State Bd. of Pharmacy v. Virginia Citizens Consumer Council, Inc., 425 U.S. 748, 770, 96 S.Ct. 1817, 48 L.Ed.2d 346 (1976) (extending First Amendment protection to the communication of product price information).
Most recently, in Sorrell v. IMS Health, 564 U.S. ——, 131 S.Ct. 2653, 180 L.Ed.2d 544 (2011), the Supreme Court applied heightened First Amendment scrutiny to a Vermont statute that restricts how certain entities can use medical prescription information. The Court approvingly quoted the Second Circuit, which had held that the “First Amendment protects even dry information, devoid of advocacy, political relevance, or artistic expression.” IMS Health, 131 S.Ct. at 2666-67 (quoting IMS Health Inc. v. Sorrell, 630 F.3d 263, 271-72 (2d Cir.2010)). The Vermont law in IMS Health is the flip side of California’s § 2527; they involve similar speech that Vermont prohibits and California compels.
The majority asserts that the compulsion of factual speech is “quite different from” the prohibition of such speech, but in fact, “in the context of protected speech, the difference is without constitutional significance, for the First Amendment guarantees ‘freedom of speech,’ a term necessarily comprising the decision of both what to say and what not to say.” Riley, 487 U.S. at 796, 108 S.Ct. 2667. As the First Amendment protects the “concomitant” rights to speak and refrain from speaking, it follows that § 2527 does not avoid all constitutional scrutiny merely because it mandates factual speech. Wooley, 430 U.S. at 714, 97 S.Ct. 1428. Facts, including statistics, convey messages. See IMS Health, 131 S.Ct. at 2667 (“Facts, after all, are the beginning point for much of the speech that is most essential to advance human knowledge and to conduct human affairs.”). The record in this case clearly “suggests that the governmental purpose in enacting section 2527 was to urge third *1112party payors, by the use of statistical information, to compensate pharmacists at a fairer rate for providing pharmaceutical services to their insureds.” ARP, 42 Cal.Rptr.3d at 265. The California Court of Appeal panels properly applied the holding of Riley to this scenario in which the Defendants complain that § 2527 compels them to disseminate a message with which they disagree. Nothing about these decisions provides convincing evidence that the California Supreme Court would decide the question differently.
III.
“[WJhere there is no convincing evidence that the state supreme court would decide differently, a federal court is obligated to follow the decisions of the state’s intermediate appellate courts.” Vestar Dev. II, LLC v. Gen. Dynamics Corp., 249 F.3d 958, 960 (9th Cir.2001). “This is especially true when the Supreme Court has refused to review the lower court’s decision.” See State Farm Fire & Cas. Co. v. Abraio, 874 F.2d 619, 621 (9th Cir.1989). The majority gives insufficient weight to the California Supreme Court’s denial of review here, relying on an inapposite citation about the meaning of the denial of a certification request from this court. See In re K F Dairies, Inc. & Affiliates, 224 F.3d 922, 925 n. 3 (9th Cir.2000). Here, in the context of the Erie doctrine, denial of review by a state high court is an appropriate and important factor to consider. See Tenneco West, Inc. v. Marathon Oil Co., 756 F.2d 769, 771 (9th Cir.1985).
The district court issued its ruling in this case on May 15, 2007, after the three California Court of Appeal panels had rendered their judgments that § 2527 violates the state constitution. However, in opting not to afford Erie deference to those judgments, the district court did not have the benefit of the California Supreme Court’s decision to deny review of Bradley, which occurred on June 13, 2007. See Bradley, 2007 Cal. LEXIS 6365. The panel majority is at no such disadvantage, but it still substitutes its own analysis for that of every state court to consider this matter.3
Two important “aims of the Erie rule[are] discouragement of forum-shopping and avoidance of inequitable administration of the laws.” Kohlrautz v. Oilmen Participation Corp., 441 F.3d 827, 831 (9th Cir.2006) (quoting Gasperini v. Center for Humanities, Inc., 518 U.S. 415, 428, 116 S.Ct. 2211, 135 L.Ed.2d 659 (1996)). As the majority acknowledges, its opinion will encourage forum shopping by creating a disparity in the administration of California law. The proffered justification for this unfortunate result is that the three California Court of Appeal panels made critical analytical errors in holding that § 2527 violates the state constitution, but in fact they properly interpreted the relevant state and federal law, and there is no convincing evidence that the California Supreme Court would rule differently. As this is a classic case requiring deference to state court judgments about a state law matter, I would reverse the decision of the district court.
. The case law relied on by the majority illustrates the dangers of disregarding state court decisions and imposing our own interpretations of legal gray areas. The first case cited by the majority for the proposition that we *1109can refuse to follow intermediate state appellate decisions that make "analytical errors” is Briceno v. Scribner, 555 F.3d 1069 (9th Cir.2009). There a divided panel of our court declined to follow two California Court of Appeal decisions interpreting a gang sentencing enhancement statute. Id. at 1080-82. However, the California Supreme Court subsequently disagreed with the Briceno majority, concluding that the Court of Appeal panels did not erroneously interpret state law. People v. Albillar, 51 Cal.4th 47, 119 Cal.Rptr.3d 415, 244 P.3d 1062, 1074-75 (2010).
. The majority’s suggestion that § 2527 may primarily regulate conduct because it requires the compiling and transmission of a document, is contrary to explicit Supreme Court precedent. See Bartnicki v. Vopper, 532 U.S. 514, 527, 121 S.Ct. 1753, 149 L.Ed.2d 787 (2001) ("It is true that the delivery of a tape recording might be regarded as conduct, but given that the purpose of such a delivery is to provide the recipient with the text of recorded statements, it is like the delivery of a handbill or a pamphlet, and as such, it is the kind of ‘speech’ that the First Amendment protects.”). All government compulsion of speech requires some conduct incident to the expression. Labeling § 2527 as conduct-based on this ground is akin to considering the laws in Wooley and Miami Herald Publ’g Co. v. Tomil*1110lo, 418 U.S. 241, 94 S.Ct. 2831, 41 L.Ed.2d 730 (1974) as primarily regulating conduct because they require the physical display of a license plate and the tangible allocation of newspaper column inches. The majority's attempt to distinguish "the actual performance of the pricing studies” from "the transmission of their results" is unavailing, as both "the creation and dissemination of information are speech within the meaning of the First Amendment.” Sorrell v. IMS Health, 564 U.S.-, 131 S.Ct. 2653, 2667, 180 L.Ed.2d 544 (2011); see also Brown v. Entm’t Merch. Assn, 564 U.S.-, 131 S.Ct. 2729, 2734 n. 1, 180 L.Ed.2d 708 (2011) ("Whether government regulation applies to creating, distributing, or consuming speech makes no difference.”). Finally, the majority asserts that § 2527’s compulsion of speech is merely ancillary to the legislature’s ultimate goal of reducing information costs. However, the majority contradicts itself, for elsewhere it conectly notes that the statute’s purpose "is of limited significance." What matters is the effect of the statute on speech, and just as in Wooley and Tornillo, the sole tangible directive of this regulation is the compulsion of speech.
. That the California Supreme Court denied review in Bradley, leaving state precedent intact in the face of a contrary federal district court holding, only further indicates that the intermediate state appellate courts correctly applied state law, and certainly does not count as convincing evidence that the California Supreme Court would uphold the statute.