IN THE SUPREME COURT OF TEXAS

IN THE SUPREME COURT OF TEXAS

════════════

No. 03-0505

════════════

Citizens Insurance Company of America, Citizens, Inc., Harold Riley, and Mark Oliver, Petitioners,

v.

Dr. Fernando Hakim Daccach, on behalf of himself and all others similarly situated, Respondent

════════════════════════════════════════════════════

On Petition for Review from the

Court of Appeals for the Third District of Texas

════════════════════════════════════════════════════

Argued October 21, 2004

Justice Wainwright delivered the opinion of a unanimous Court as to sections I-III and V-VIII; the opinion of the Court as to sections IV-A, IV-C, and IV-D, joined by Justice Hecht, Justice O’Neill, Justice Green, Justice Johnson, and Justice Willett; and a concurring opinion as to section IV-B, joined by Justice Johnson.

Chief Justice Jefferson filed a concurring opinion, joined by Justice Brister and Justice Medina.

In this interlocutory appeal petitioners challenge a trial court’s order certifying a worldwide class. Dr. Fernando Hakim Daccach, the class’s representative, alleges that petitioners Citizens Insurance Company of America (CICA), Citizens, Inc., Harold E. Riley, and Mark A. Oliver (collectively Citizens) sold securities from Texas to nonresidents without complying with the registration requirements of the Texas Securities Act. The court of appeals modified the class definition and affirmed the trial court’s certification of the class. 105 S.W.3d 712. Because we conclude the trial court did not consider the effect of res judicata on the adequacy of the class representative, the superiority of litigating this case as a class action, the typicality of claims within the class, and the predominance of common issues over individual issues, we decertify the class and remand the case to the trial court for further proceedings consistent with this opinion.

I. Background

Citizens, Inc. and its wholly-owned subsidiary CICA are Colorado corporations with their principal places of business in Austin, Texas. Riley and Oliver are officers and directors of Citizens, Inc. Citizens sells life insurance policies (CICA policies) through foreign insurance agents exclusively to persons outside of the United States. The purchasers reside in over thirty-five countries including the United States. The CICA policies allow policyholders to assign policy dividends and other benefits to offshore trusts. The trusts use the assigned dividends and other benefits to purchase common stock in Citizens, Inc. Each year since 1996 there have been approximately 30,000 CICA policies in effect, with each policyholder paying an average annual premium of around $2,000. Id. at 717. At least seventy‑five percent of these policyholders have assigned their policy dividends and other benefits to the offshore trusts. Id.

The CICA policies are not registered with the Texas State Securities Board, Texas Department of Insurance, nor any other regulatory body in the United States, although the common stock purchased with policy dividends is listed on the American Stock Exchange. Similarly, neither Citizens nor its salespersons have registered with any regulatory body in Texas or elsewhere in the United States. Citizens also asserts that the CICA policies are not subject to regulation in the countries in which the policyholders reside.

On August 6, 1999, Delia Bolanos Andrade and Luis Martin Tapia Alberti, both residents and citizens of Colombia, South America, filed a class action against Citizens in Texas state court. The original petition alleged several causes of action related to the CICA policies, including (1) violations of the Texas Deceptive Trade Practices Act, (2) breach of contract, (3) fraud, (4) fraud in the inducement, (5) negligent misrepresentation, (6) breach of the duty of good faith and fair dealing, (7) violations of the Texas Insurance Code, (8) equitable reformation of the policies, (9) conspiracy to plan and implement this scheme, and (10) unjust enrichment and the imposition of a constructive trust. On December 15, 2000, the class plaintiffs filed a second amended original petition to add a cause of action under the Texas Securities Act for selling securities in this state without first being registered. See Tex. Rev. Civ. Stat. arts. 581-12A, 581-33A(1).[1] By this time seven new plaintiffs had been added to the lawsuit, including Daccach.

On June 29, 2001, Daccach filed a motion for class certification in which he sought designation as the class representative and alleged against Citizens only one class claim: selling or offering securities from Texas in the form of the CICA policies without registering with the Texas Securities Board. See Tex. Rev. Civ. Stat. arts. 581-12A, 581-33A(1), 581-33D(1), D(3). Daccach expressly disclaimed any intention to pursue the other causes of action in the class suit. In the sixth amended petition, filed the same day as the first amended motion for class certification, the other plaintiffs pled the original claims against Citizens as individuals, not as class representatives. The trial court has not ruled on Daccach’s motion to sever the class claims.

Challenging Daccach’s motion for class certification, Citizens argued that Texas law should not apply to this worldwide class action and that Daccach’s abandonment of claims defeats certification prerequisites. In response, Daccach presented alternate choice of law analyses all of which directed the application of Texas law, but none of his theories analyzed the laws of other jurisdictions.

After conducting a four‑day hearing, the trial court granted Daccach’s motion in a twenty‑page class certification order. The order’s nine‑page trial plan identified four class‑wide issues to be resolved at trial: (1) whether a CICA Policy is a “security” pursuant to the Securities Act (including the question of whether the CICA policies fall within an insurance exception to the Securities Act); (2) whether Citizens sold or offered for sale the CICA policies from Texas; (3) the calculation of the statutory remedy pursuant to the Securities Act; and (4) attorney’s fees. The order defined the class as follows:

The Class consists of all persons, who, during the Class Period (August 6, 1996 through the date the Class is certified): (1) purchased a CICA Policy and executed an assignment to a trust for the purchase of Citizens, Inc. stock, or (2) paid any money that, pursuant to a CICA Policy and assignment to a trust, was for the purchase of Citizens, Inc. stock, or (3) were entitled to any cash benefits from a CICA Policy that, pursuant to a CICA Policy and assignment to a trust, were for the purchase of Citizens, Inc. stock. Specifically excluded from the Class are all persons who, within the time period established by the judgment, do not surrender their CICA Policies and take the other actions required to obtain the relief awarded by the Court.

Citizens brought an interlocutory appeal challenging the trial court’s certification order. See Tex. Civ. Prac. & Rem. Code § 51.014(a)(3). Citizens argued that the trial court abused its discretion in granting Daccach’s motion for class certification. First, Citizens challenged the adequacy of the trial court’s class definition. Second, it argued that the trial court failed to conduct a proper choice of law analysis to determine whether common issues predominate over individual issues. Third, Citizens argued that the trial court failed to adequately establish the class certification prerequisites. The court of appeals rejected all three points, holding that: the class definition, after a one-word modification, precisely ascertains the class members;[2] the trial court was not required to engage in a “most significant relationship” choice of law analysis; and the trial court did not abuse its discretion by finding that the class certification requirements had been met. 105 S.W.3d at 729-30.

Citizens then petitioned this Court for review. Specifically, Citizens contends that the court of appeals erred in (1) affirming the certification of an improper fail-safe class, whose members are not presently ascertainable, and eviscerating material defenses; (2) not applying the “most significant relationship” test to resolve choice-of-law issues; (3) determining that common legal and factual issues predominate despite that calculating attorney’s fees will be an overwhelming task requiring the discovery and resolution of circumstances surrounding life insurance sales in over fifty foreign jurisdictions; (4) affirming that a class action is superior to individual claims despite the fact that certification requires dismissal of ten of the eleven original claims and that this is not a negative value suit;[3] (5) agreeing, without explanation, that the trial court will be able to implement the statutory remedy of rescission even though the beneficial interests of the policies are held in offshore trusts not parties to this case; (6) affirming that Texas Rule of Civil Procedure 42(a)’s typicality requirement was met despite the presence of a statute of limitations defense against the class plaintiff and the class plaintiff’s dismissal of ten of the eleven originally alleged claims; and (7) holding that previously asserted, and now abandoned, individual claims of class members would not be barred by res judicata, thereby reading Rule 42 as an exception to the Court’s transactional approach to claim preclusion. We granted Citizens’ petition for review.

II. Jurisdiction

Section 51.014(a)(3) of the Texas Civil Practice and Remedies Code allows the interlocutory appeal of class certification orders. Although interlocutory appeals are generally final in the courts of appeals, this Court has jurisdiction over an interlocutory appeal if the court of appeals “holds differently from a prior decision of another court of appeals or of the supreme court.” Tex. Gov’t Code §§ 22.001(a)(2), 22.225(b)(3), (c).[4] “[T]wo decisions hold differently or conflict when the rulings in the two cases are so far upon the same state of facts that the decision of one case is necessarily conclusive of the decision in the other.” Henry Schein, Inc. v. Stromboe, 102 S.W.3d 675, 687 (Tex. 2002) (internal quotation marks omitted) (citing Christy v. Williams, 298 S.W.2d 565, 567 (Tex. 1957)).

In Intratex Gas Co. v. Beeson, we held that the trial court abused its discretion by certifying a fail-safe class. 22 S.W.3d 398 (Tex. 2000). A fail-safe class is a class bound only by a judgment for the plaintiffs. Id. at 402. In such a case, “[a] determination that the defendant is not liable . . . obviates the class, thereby precluding the proposed class members from being bound by the judgment.” Id. at 405. We rejected that outcome because Rule 42(b) was never intended to be an exception to res judicata.

In this case, the court of appeals held that the class’s claim under the Texas Securities Act met the predominance requirements of Rule 42 even if the class abandoned other potential claims to meet that requirement. The court explained: “‘Clients who have claims not raised in this class action because the claims are unsuitable for class treatment can bring those claims on an individual basis, and res judicata will not bar those claims because absent class members had no opportunity to litigate those issues in this lawsuit.’” 105 S.W.3d at 725 (quoting Sullivan v. Chase Inv. Servs. of Boston, Inc., 79 F.R.D. 246, 265 (N.D. Cal. 1978)). The court of appeals’ creation of a special exception to established principles of claim preclusion conflicts with this Court’s holding in Beeson. Accordingly, we have jurisdiction over this appeal.

III. General Certification Requirements

All class actions must satisfy four prerequisites: (1) numerosity—the class is so numerous that joinder of all members is impracticable; (2) commonality—there are questions of law or fact common to the class; (3) typicality—the claims or defenses of the representative parties are typical of the claims or defenses of the class; and (4) adequacy of representation—the representative parties will fairly and adequately protect the interests of the class. Tex. R. Civ. P. 42(a); see also Sw. Ref. Co. v. Bernal, 22 S.W.3d 425, 435 (Tex. 2000). In addition, a class action must satisfy at least one of the requirements in Rule 42(b). Here Daccach argues the class action satisfies Rule 42(b)(3),[5] which requires that “questions of law or fact common to the members of the class predominate over any questions affecting only individual members” and that class treatment be “superior to other available methods for the fair and efficient adjudication of the controversy.” Rule 42 contains a list of nonexhaustive factors to aid a court in determining if (b)(3) certification is appropriate:

(A) the interest of members of the class in individually controlling the prosecution or defense of separate actions; (B) the extent and nature of any litigation concerning the controversy already commenced by or against members of the class; (C) the desirability or undesirability of concentrating the litigation of the claims in the particular forum; (D) the difficulties likely to be encountered in the management of a class action.

Tex. R. Civ. P. 42(b)(3).

In Bernal, we explained that to properly apply the certification prerequisites, a trial court must perform a “rigorous analysis.” 22 S.W.3d at 435. And to correctly determine these certification issues, a certifying court must “understand the claims, defenses, relevant facts, and applicable substantive law.” Id. This understanding requires a choice of law analysis at the outset anytime there is an issue of which of several jurisdictions’ laws should govern a case. Compaq Computer Corp. v. Lapray, 135 S.W.3d 657, 672 (Tex. 2004) (“[Variations in the laws of multiple jurisdictions] ‘may swamp any common issues and defeat predominance.’” (quoting Castano v. Am. Tobacco Co., 84 S.W.3d 734, 741 (5th Cir. 1996))). The threshold question, therefore, is whether the trial court conducted a proper choice of law analysis and correctly decided that Texas law governs this class suit. Compaq, 135 S.W.3d at 741.

IV. Choice of Law

A. Plain Language of Section 12

“[W]hen ruling on motions for class certifications, trial courts must conduct an extensive choice of law analysis before they can determine predominance, superiority, cohesiveness, and even manageability.” Id. This analysis arms the court with information necessary to determine if questions of law or fact common to the members of the class will predominate over any questions affecting only individual members. See Tex. R. Civ. P. 42(b)(3). Common questions of law may not predominate if class members’ claims are not governed by the same law. Schein, 102 S.W.3d at 695-99. Here, Daccach, as class representative, bears the burden of showing that Texas law applies to the class’s claim. Compaq, 135 S.W.3d at 672.

Daccach alleges that the defendants violated the Texas Securities Act by offering or selling securities from Texas without registering with the Texas Securities Board. He argues that the Texas Securities Act directs the application of Texas law notwithstanding the presumed interests of the forums in which the plaintiffs reside. He explains that the defendants were Texas residents doing business in Texas at all relevant times. Daccach argues that an analysis of other jurisdictions’ laws is unnecessary because (1) Citizens is estopped from arguing the applicability of foreign laws because of its previous position that the CICA policies were not subject to foreign regulation, and (2) the registration provision of the Texas Securities Act directs the application of Texas law in this case, rendering unnecessary a comparison of other potentially applicable jurisdictions’ laws.

Citizens argues that the pleadings show that the dispute implicates interests of over thirty-five jurisdictions where the putative class members reside. Citizens asserts that, at a minimum, the pleadings require the trial court to conduct an extensive analysis of potential conflicts among other jurisdictions’ laws and determine which jurisdiction has the most significant relationship to the class claim. See Restatement (Second) of Conflicts of Laws § 6 (1971). Citizens contends that failing to do so violates the Due Process Clause of the U.S. Constitution. See Phillips Petroleum Co. v. Shutts, 472 U.S. 797, 812 (1985).

In the class certification order, the trial court concluded:

The Class Plaintiff has asserted only one cause of action for which he seeks class certification, that is, violations of the Texas Securities Act. At the certification hearing, the Class Plaintiff presented evidence of numerous activities of the Defendants in Texas relating to the CICA policies. Accordingly, the Court concludes that for purposes of class certification, Texas law applies.

The court of appeals held that under section 6(1) of the Restatement (Second) of Conflict of Laws, Texas law applies because the Texas statute directs that it apply. 105 S.W.3d at 723-24.

As an initial matter, we reject Daccach’s estoppel argument. We have explained that “[a] court may not accept ‘on faith’ a party’s assertion that no variations in [other jurisdictions’] laws exist.” Compaq, 135 S.W.3d at 672-73. The court must determine which substantive law governs the case. See Shutts, 472 U.S. at 820 (“plaintiff’s desire for forum law is rarely, if ever controlling”); Tracker Marine, L.P. v. Ogle, 108 S.W.3d 349, 352 (Tex. App.—Houston [14th Dist.] 2003, no pet.). Which jurisdiction’s substantive law governs is ultimately a question of law for the court. Compaq, 135 S.W.3d at 672.

Daccach pleads a single cause of action on behalf of the class: Citizens violated section 12 of the Texas Securities Act that requires dealers in Texas who offer or sell securities to register with the Texas Securities Board. See Tex. Civ. Stat. arts. 581-12A, 581-33A(1). No one disputes that Citizens was doing business in Texas and physically present in Texas when it sold the CICA policies to class members.

Texas has a strong interest in regulating the sale of securities in and from the state. The Section 12 registration provisions indemnify investors victimized by violations of the Texas Securities Act, encourages compliance with its regulatory and disclosure provisions, creates an incentive for its private enforcement, and guards the integrity of the state’s securities industry by protecting resident sellers who operate in compliance with the law. See Tex. Civ. Stat. art. 581-33 cmt. background–1977; Rio Grande Oil Co. v. State, 539 S.W.2d 917, 921 (Tex. Civ. App.—Houston [1st Dist.] 1976, writ ref’d n.r.e.) (“A state is damaged if its citizens are permitted to engage in fraudulent [securities] practices even though those injured are outside its borders.”); see also Unif. Sec. Act–1956, § 414(a)-(f) cmt. 3, 7C U.L.A. 941 (2006) (noting that state Blue Sky laws are also intended to prevent use of a state as a base of operations to defraud persons in other states); Joseph C. Long, The Conflict of Laws Provisions of the Uniform Securities Acts or When Does a Transaction “Take Place in the State?”, 31 Okla. L. Rev. 781, 784 (1978); Louis Loss, The Conflict of Laws and the Blue Sky Laws, 71 Harv. L. Rev. 209, 225 n.50 (1957); Jack E. McClard, The Applicability of Local Securities Acts to Multi-State Securities Transactions, 20 U. Rich. L. Rev. 139, 142 (1985). Because the class lawsuit only alleges Citizens’ failure to register with the Texas Securities Board before allegedly offering and selling securities from Texas, Section 12 governs under any conflict of law principles that might apply.

Absent unique statutory circumstances, trial courts must conduct the extensive choice of law analysis described in Compaq before making a certification decision. Compaq, 135 S.W.3d at 672. In Compaq, the class alleged a Uniform Commercial Code breach of an express warranty claim occurring in all fifty states. See id. at 672-73. We held that the trial court erroneously applied Texas law after a cursory review and failed to analyze the relevant law of each state. However, a claimed failure to register as a dealer before offering or selling securities is different. Securities offered or sold in multiple states may be subject to the registration requirements of each state in which an offer or sale is made. See Lintz v. Carey Manor, Ltd., 613 F. Supp. 543, 550 (D.C. Va. 1985); Michael A. Hanzman, The Reach of State Blue Sky Laws—A Potentially Dangerous Trap for the Unwary Practitioner, 63 Fla. B.J. 16, 19 (1989) (“[J]ust as one transaction can violate both federal and state law simultaneously, it can violate several blue sky laws simultaneously.”); see also 69A Am. Jur. 2D Securities Regulation § 18 (2006) (stating more broadly than held here that “all of the blue sky laws of all the jurisdictions apply to the transactions which are within the bounds of the statute”). Multiple registration requirements of multiple states may govern the dealer’s conduct and give rise to several statutory violations. See Chrysler Capital Corp. v. Century Power Corp., 1992 WL 163006 (S.D.N.Y. 1992) (unpublished); Simms Inv. Co. v. E.F. Hutton & Co., 699 F. Supp. 543, 545 (M.D. N.C. 1988) (“The court has firmly concluded that the securities laws of two or more states may be applicable to a single transaction without presenting a conflicts of laws question.”) ; see also Unif. Sec. Act–1956, § 414(a)-(f) cmt. 3, 7C U.L.A. 941 (2006); Joseph C. Long, 12 Blue Sky Law §§ 3.01 n.2.2, 3.02[1] n.13 (rev. ed. 1988); Louis Loss & Joel Seligman, Fundamentals of Security Regulation 92–94 (3d ed. 1989); McClard, 20 U. Rich. L. Rev. at 141. Thus, a claim based on the failure to register with the Texas Securities Board before offering or selling securities from Texas does not present a classic conflict of laws problem. We do not hold that the Texas Securities Act directs the application of the Texas Blue Sky laws[6] in every securities case involving facts touching Texas or its residents. The question is one of legislative intent as to the particular provision at issue, subject to constitutional limitations.

We recognize that violations of other securities laws, such as those based on misrepresentation, may well be subject to a different analysis. See Tracker Marine, L.P., 108 S.W.3d at 359; Loss, 71 Harv. L. Rev. at 209 (indicating that unlike registration requirements, presumably the conflict of laws rules for anti-fraud aspects of the blue sky laws are not too different from “the rules for common-law deceit or rescission”). But those types of claims are not raised here.

The concurrence would require a thorough comparison of the laws of the jurisdictions implicated by the pleadings, even though the only claim at issue is that Texas residents offered or sold securities from Texas without registering with the Texas Securities Board. The concurrence asserts that the Court’s choice of law determination “risks making Texas a magnet forum for national and international class actions.” __ S.W.3d __, __. We do not hold, contrary to the concurrence’s indication, that a class may gain application of Texas law by simply suing in Texas on a Texas statute or on any Texas Blue Sky provision. It is the rare class suit in which a Texas court reaches a permissible conclusion on choice of law without an extensive analysis.

The concurrence also argues that the Court allows “the simple institution of a multistate class suit” to create a “substantial threat to our constitutional system of cooperative federalism.” Id. at __ (quoting 4 Herbert B. Newberg & Alba Conte, Newberg on Class Actions § 13.37 (4th ed. 2002). As explained, the filing of suit in Texas is not the basis for the choice of law determination. The determinative question is not where the class suit was filed, but, under principles of statutory interpretation, whether the resident defendants’ actions in Texas constitute conduct the Legislature intended to regulate. This holding offends neither the U.S. Constitution nor principles of federalism.

Further, the concurrence fails to explain the parameters of its approach. Which of the other jurisdictions’ laws may govern the failure to register in Texas? Which jurisdictions’ laws should be studied and compared to Section 12(A), the only claim alleged? Neither the pleadings nor the class definition assert the violation of the common law or another jurisdiction’s statute prohibiting the sale of securities to or from Texas by a resident dealer not registered in Texas. We do not compel plaintiffs in individual suits to plead the violation of all potentially applicable laws, yet the concurrence would impose that burden on class plaintiffs. Daccach was not required to present a global overview of potentially applicable securities registration laws to pursue a claim against Citizens for selling or offering securities in or from Texas as an unregistered dealer. See Irving L. Faught & Z. Faye Martin Morton, Recent Developments in Securities Law: USA 2002—Something Old, Something New . . ., 60 Consumer Fin. L.Q. Rep. 58, 60 (2006); Larry Kramer, Choice of Law in Complex Litigation, 71 N.Y.U. L. Rev. 547, 549 (1996); Larry Kramer, Rethinking Choice of Law, 90 Colum. Law Rev. 277, 284–87 (1990) (reasoning that the choice of law analysis of the class’s claim should not be evaluated differently than it would be if brought by an individual). The trial court correctly concluded that the Texas Securities Act applies to this suit.

B. The Restatement Approach

In the alternative, the registration requirement in the Texas Securities Act contains a statutory directive compelling the application of Texas law. This Court uses the analysis described in the Restatement (Second) of Conflicts of Laws to resolve choice of law issues and select the particular substantive issue that governs a case. See Hughes Woods Prods., Inc. v. Wagner, 18 S.W.3d 202, 203-04 (Tex. 2000); Schein, 102 S.W.2d at 696-99. Section 6 of the Restatement provides:

(1) A court, subject to constitutional restrictions, will follow a statutory directive of its own state on choice of law.

(2) When there is no such directive, the factors relevant to the choice of the applicable rule of law include

(a) the needs of the interstate and international systems,

(b) the relevant policies of the forum,

(c) the relevant policies of other interested states and the relative interests of those states in the determination of the particular issue,

(d) the protection of justified expectations,

(e) the basic policies underlying the particular field of law,

(f) certainty, predictability and uniformity of result, and

(g) ease in the determination and application of the law to be applied.

Restatement (Second) of Conflicts of Laws § 6 (1971).

The Restatement identifies a framework many courts follow when deciding which jurisdiction’s law applies. The first question is whether the particular substantive law is subject to a clear choice of law determination by the Legislature of the forum state. See Marmon v. Mustang Aviation, 430 S.W.2d 182 (Tex. 1968). If there is such a directive, a court examines the directive in light of constitutional limitations that might preclude application of the local law. If answering the first two inquiries does not resolve the issue, a court can apply the forum law if it does not conflict with the laws of other interested jurisdictions. See Compaq, 135 S.W.3d at 672. If variation in the laws of several interested jurisdictions creates a conflict, then courts will apply the significant relationship guidelines of Section 6(2) and any other specific sections applicable to the substantive law at issue. See Hughes Woods Prods., 18 S.W.3d at 205; see, e.g., Restatement (Second) of Conflicts of Laws § 192 (1971) (relating to the validity or rights created by life insurance contracts).

Under this hierarchy, the factors in section 6(2) of the Restatement do not come into play if there is statutory guidance that the law is intended to govern the transaction. For instance, when the Fifth Circuit applied this analysis to project Texas’ choice of law rule for pendant state claims, it declined to look to the “most significant relationship” guidelines when a Texas statute provided clear choice of law guidance, and held that “‘a court should only resort to the § 6 guidelines in the absence of either a valid contractual agreement between the parties regarding the applicable law, or a local statutory provision controlling the disposition of the choice of law question.’” Sommers Drug Stores Co. v. Corrigan, 883 F.2d 345, 353 (5th Cir. 1989) (emphasis added) (quoting Am. Home Assurance Co. v. Safeway Steel Prods. Co., 743 S.W.2d 693, 697 (Tex. App.—Austin 1987, writ denied)). An examination of the provision of the Texas Securities Act at issue shows whether the Texas Legislature intended to direct the application of that provision to the facts alleged in this case.

The sole violation of law alleged by the class is embodied in sections 33A and 12(A) of article 581 of the Texas Securities Act concerning liability of sellers of securities who fail to register in Texas. The relevant precedent from this Court guides the determination of whether Section 12 contains a directive from the Legislature to apply Texas law, even though some acts may have occurred outside Texas. In Marmon, we stated:

Unless the intention to have a statute operate beyond the limits of the state or country is clearly expressed or indicated by its language, purpose, subject matter, or history, no legislation is presumed to be intended to operate outside the territorial jurisdiction of the state or country enacting it. To the contrary, the presumption is that the statute is intended to have no extraterritorial effect, but to apply only within the territorial jurisdiction of the state or country enacting it, and it is generally so construed. An extraterritorial effect is not to be given statutes by implication.

430 S.W.2d at 187 (citations omitted); see 73 Am. Jur. 2D Statutes § 250 (2006).[7] Determining if the extraterritorial reach of Section 12 is “clearly expressed” or otherwise “indicated by its language, purpose, subject matter, or history” begins with the language of the provision. See Marmon, 430 S.W.2d at 187.

The Texas Legislature prohibited the offer or sale of a security “in this state” by any company or person, who has not previously complied with the requirement to register as a securities dealer or satisfied a dealer, security, or transaction exemption from registration. Tex. Civ. Stat. arts. 581-12(A), 581-33A. The requirement in Section 12 to register before making offers or sales “in this state” attaches to both offers and sales of securities and includes both offers or sales from persons in Texas to nonresidents and those from out-of-state sellers to Texas residents. See __ S.W.3d at __, n.6. Therefore, section 12 requires that persons and companies register or satisfy an exemption from registration before making offers or sales of securities from locations in Texas to out-of-state purchasers. See generally Ennetex Oil & Gas v. State, 560 S.W.2d 494 (Tex. Civ. App.—Texarkana 1977, writ ref’d n.r.e.); Rio Grande, 539 S.W.2d 917.

The Texas Securities Board, empowered to administer the securities laws, determined in its rules that section 12 of the Texas Securities Act governs “an offer or sale from Texas.” 7 Tex. Admin. Code § 139.7. Section 139.7, entitled “Sale of Securities to Nonresidents,” provides that “[a]n issuer or selling agent who makes an offer or sale from Texas, by any means . . . is a dealer and must comply with the dealer registration requirements of the Securities Act.”

This interpretation of Section 12 is supported by the purpose of Texas Blue Sky laws. The commentary to Article 581-33 reiterates the long-standing purposes of the provision: to indemnify investors victimized by violations of the Texas Securities Act, encourage compliance with the Act’s regulatory and disclosure provisions, and create incentives for its private enforcement. Tex. Rev. Civ. Stat. art. 581-33 cmt. background–1977. Given the nature of securities transactions, achieving these purposes will ultimately require that the Act apply to situations that involve some out-of-state activities, as when an unregistered dealer in Texas sells securities to a nonresident.

The history of choice of law concerns arising in the subject matter of securities also supports our interpretation of the language of Section 12. The first Blue Sky laws were promulgated in 1910. Julian M. Meer, The Texas Securities Act—1957 Model: Facelift or Forward Look?, 36 Tex. L. Rev. 429, 430 (1957). By 1957, every state except Delaware and Nevada had enacted some form of Blue Sky law to regulate securities transactions. Louis Loss, The Conflicts of Laws and the Blue Sky Laws, 71 Harv. L. Rev. 209, 225 (1957); Meer, 36 Tex. L. Rev. at 430. Also by 1957, it had become apparent that courts were struggling to apply “traditional but unsuitable [common law] choice-of-law concepts” to the nationwide scheme of securities regulations. Loss, 71 Harv. L. Rev. at 248. Professor Louis Loss, the primary draftsman of the Uniform Securities Act of 1956, reported that Blue Sky decisions on choice of law in the securities arena “def[ied] generalization.” Id. at 216. The Eighth Circuit Court of Appeals referred to the “bewildering state of affairs in the case law governing transactions which crossed states lines.” Kreis v. Mates Inv., 473 F.2d 1308, 1311 (8th Cir. 1973). Professor Loss explained why:

When a whole area of “public” law owes its very existence to legislation, it is not merely anomalous that so important a segment of the area is left to the chance application of conflict-of-law concepts developed by the common law in quite different contexts; it would be amazing if the result were a reasonably satisfactory geographical allocation of the statutes. . . [T]he one solution to the multifarious and vexatious problems of the conflict of laws which no blue sky state has thus far adopted is the codification route.

Loss, 71 Harv. L. Rev. at 248.

Ultimately, the drafters of the Uniform Act rejected citizenship or residence within a particular state as the policy base for application of the Uniform Act to particular transactions. Joseph C. Long, 12 Blue Sky Law § 4:2 (rev. ed. 1988). Instead, they elected a territorial base as the foundation for the choice of law decision, requiring that a transaction have some physical nexus or acts within the state whose securities statute was alleged to govern. Id. The Uniform Act’s approach is that a statute governs a transaction and claims arising from it if wrongful acts in the transaction occurred “in this state.” Id. § 4:1. The Restatement elaborates:

The court should give a local statute the range of application intended by the legislature when these intentions can be ascertained and can constitutionally be given effect. If the legislature intended that the statute should be applied to the out-of-state facts involved, the court should so apply it unless constitutional considerations forbid.

Restatement (Second) of Conflicts of Laws § 6(1) cmt.(b) (1971). Choice of law in this area of the Blue Sky laws is now primarily a matter of statutory interpretation, except, of course, for those states that have not legislated choice of law instructions. See Benjamin v. Cablevision Programming, 499 N.E.2d 1309, 1316 (Ill. 1986) (reasoning that whether the Illinois Securities Act applied to a sale of a security from Illinois to a California purchaser was a question of statutory construction, and holding that statutory language referencing an offer or a sale “in this state” indicated the application of the Illinois securities statute); see also 69A Am. Jur. 2d Securities Regulation § 18 (2006) (In the area of securities transactions, state Blue Sky laws apply to the transactions which “are within the bounds of the statute.”).

The Texas Securities Act was adopted substantially from the Uniform Securities Act. See Tex. Rev. Civ. Stat. art. 581-33, cmt. background–1977 (noting that enactment of article 581-33 in 1963 was a modification of the Uniform Act). The Texas Legislature incorporated part of the Uniform Securities Act in Texas Blue Sky laws, including an important term of art in the particular substantive provision at issue here—“in this state”—used in connection with mandates to comply with specified regulatory requirements, like dealer registration.[8]

Based on the language, purpose, subject matter, and history of the Texas Blue Sky laws and the Uniform Securities Act, and the registration requirements in particular, we conclude the Texas Legislature intended section 12 of the Texas Securities Act to prohibit the unregistered sale of securities from Texas, even when the purchasers are nonresidents. This approach does not mean that the Texas Securities Act directs the application of the Texas Blue Sky laws in every securities case involving facts touching Texas or its residents. The question is one of legislative intent as to the particular provision at issue, subject to constitutional limitations. See, e.g., Yadlosky v. Grant Thornton L.L.P., 197 F.R.D. 292, 301 (E.D. Mich. 2000) (stating, in reference to the misrepresentation provisions of the Michigan Blue Sky law, that “it appears application of Michigan law to all of the 2811 investors would be contrary to the policies of other state ‘blue-sky’ laws”).

C. Constitutional Limitations on State Regulation of Extraterritorial Conduct

The trial court must also determine whether the Texas statute meets constitutional requirements before it is applied to extraterritorial conduct.[9] Due process requires that the application of Texas law be neither arbitrary nor fundamentally unfair. See Shutts, 472 U.S. at 818-19, 822. Although the Constitution imposes “modest restrictions” on the application of a forum state’s substantive law to conduct that occurs, at least in part, outside of the state, to constitutionally select a forum state’s law to apply to a class action, the state must have “a significant contact or significant aggregation of contacts” to the claims asserted by each member of the plaintiff class. Allstate Ins. Co. v. Hague, 449 U.S. 302, 313 (1981); see also Travelers Health Ass’n v. Virginia, 339 U.S. 643, 649 (1950); Shutts, 472 U.S. at 818. These constitutional limitations apply to choice of law determinations in class suits. See Shutts, 472 U.S. at 821-22.

Citizens contends that because the court of appeals chose to apply section 6(1) of the Restatement in lieu of the most significant relationship test, the court’s choice of law analysis did not satisfy the constitutional due process guarantee that the application of Texas law be neither arbitrary nor fundamentally unfair. Texas has an interest in transactions involving the purchase and sale of securities. The constitutional question in this case, then, is whether Texas has sufficient contacts with the class members’ transactions to satisfy constitutional due process.

In his pleadings and summary judgment evidence presented to the trial court, Daccach alleges that all defendants are Texas residents, Citizens maintains its principal place of business in Texas, advertising and sales materials were created and sent from Texas, a significant portion of the activities related to the marketing and creation of the instruments happened in Texas, and Citizens devised, implemented, and administered the securities “scheme” in Texas. Although Daccach admits that none of the class members are from Texas, he maintains that all CICA policies were sold from Texas. Citizens does not contest that these activities occurred in Texas, but only argues that these activities do not constitute the “sale” of a “security” in Texas. Citizens’ argument relates to a contested fact issue set for trial and does not controvert the facts alleged. Because Texas has a significant aggregation of contacts to the business activities alleged to have occurred within the state, we conclude that the application of Section 12 to this lawsuit falls comfortably within the constitutional constraints on the extraterritorial application of Texas laws. Making this determination does not resolve whether Citizens actually “sold” a “security” from Texas within the meaning of the Texas Securities Act; that is a matter to be determined on the merits.

To obtain class certification, we require an “extensive analysis” of choice of law. Compaq, 135 S.W.3d at 672. Here, Daccach alleges only that Citizens violated Article 581–33(A) by selling securities in or from Texas without registering as a dealer. No choice of law question is presented. The trial court was required to determine whether the application of the Texas statute at issue met constitutional requirements when applied to the allegations. The trial court did not abuse its discretion in determining that there was a significant aggregation of contacts with Texas to apply Article 581–33(A) constitutionally. Therefore, for different reasons, we affirm the court of appeals’ holding that the trial court properly determined that Texas law governs.

D. Impact of Contacts with other Jurisdictions

At this point we return to the reason for the choice of law scrutiny—to provide the context for a court’s rigorous analysis of the certification requirements. See id. at 672-73. The court must ensure that the class representative is adequately representing the rights of absent class members in all aspects of the class litigation. The class representative’s burden in this regard stems from the Due Process Clause, which demands “that the named plaintiff at all times adequately represent the interests of the absent class members.” Shutts, 472 U.S. at 812 (citing Hansberry v. Lee, 311 U.S. 32, 42-43, 45 (1940)). Thus, in this case, where there is a significant aggregation of contacts with Texas to apply Texas law constitutionally, the fact that other jurisdictions are implicated by the pleadings raises an issue of adequacy of representation. In a worldwide case like this, where a class representative abandons or chooses not to allege certain claims, including claims that may exist in other jurisdictions, the potential effect of claim preclusion on absent class members raises concerns about the prerequisites of predominance, superiority, typicality, and adequacy. If other jurisdictions’ laws could apply to the transaction, even though only a Texas violation is alleged, the class members who could assert those causes of action may be barred from later pursuing them in a different lawsuit. The laws of other interested jurisdictions may provide certain class members more beneficial remedies or causes of action arising from the same subject matter of the lawsuit. We therefore address the effect of res judicata or claims preclusion on later litigation of claims not alleged or abandoned and how the risk of preclusion may affect class certification.        

V. Res Judicata and Claim Abandonment

In the court of appeals, Citizens challenged the trial court’s class certification by arguing that because Daccach abandoned all claims but the Texas Securities Act claim, he was not an adequate representative of the class, common issues did not predominate over individual issues, a class action was not superior to other methods of adjudication, and Daccach improperly seeks to resolve a single issue instead of the entire controversy. The court of appeals rejected Citizens’ arguments, explaining that the Texas Securities Act claim was the entire controversy in itself and that certification was still appropriate even if other claims existed. 105 S.W.3d at 725. Specifically, the court of appeals affirmed the trial court’s certification order despite Daccach’s abandonment of numerous claims because “‘[c]lients who have claims not raised in this class action because the claims are unsuitable for class treatment can bring those claims on an individual basis, and res judicata will not bar those claims because absent class members had no opportunity to litigate those issues in this lawsuit.’” Id. (quoting Sullivan v. Chase Inv. Servs. of Boston, Inc., 79 F.R.D. 246, 265 (N.D. Cal. 1978)).

Citizens contends this holding amounts to a special exception to established principles of claim preclusion, and therefore, contradicts our holdings in Intratex Gas Co. v. Beeson, 22 S.W.3d 398, 405 (Tex. 2000), and Southwestern Refining Co. v. Bernal, 22 S.W.3d 425, 432 (Tex. 2000). Relying on Henry Schein, Inc. v. Stromboe, Citizens adds that res judicata precludes litigation of previously abandoned class claims arising out of the same transaction, and therefore, defeats class certification because Daccach’s willingness to abandon claims to the detriment of absent class members undermines the prerequisites of predominance, superiority, typicality, and adequacy. See 102 S.W.3d 675, 695 (“[I]t is not clear that a class action is superior . . . if it necessitates that plaintiffs give up substantial rights, nor is it clear that the willingness . . . to forego consequential damages is typical of the other 20,000 class members.”).

Daccach admits that for the class suit he abandoned all but the Texas Securities Act claim because the abandoned claims were not suitable for class treatment. Daccach contends, however, that because these claims were procedurally barred by Rule 42§s certification requirements, res judicata will not preclude subsequent litigation of the claims that cannot be litigated through diligence in this class action. See Barr v. Resolution Trust Corp. ex rel. Sunbelt Fed. Sav., 837 S.W.2d 627, 631 (Tex. 1992). For the following reasons, we agree with Citizens and conclude that the trial court erred in certifying the class without considering the adequacy of the class representative in light of the res judicata effect of the class representative’s decision to abandon claims.

A. Res Judicata

Generally, res judicata prevents a plaintiff from abandoning claims and subsequently asserting them when the claims could have been litigated in the prior suit. Jeanes v. Henderson, 688 S.W.2d 100, 103 (Tex. 1985); see also State & County Mut. Fire Ins. Co. v. Miller, 52 S.W.3d 693, 696 (Tex. 2001). For res judicata to apply, there must be: (1) a prior final judgment on the merits by a court of competent jurisdiction; (2) identity of parties or those in privity with them; and (3) a second action based on the same claims that were raised or could have been raised in the first action. Amstadt v. U.S. Brass Corp., 919 S.W.2d 644, 652 (Tex. 1996). The doctrine seeks to bring an end to litigation, prevent vexatious litigation, maintain stability of court decisions, promote judicial economy, and prevent double recovery. Barr, 837 S.W.2d at 629; Jeanes, 688 S.W.2d at 105.

Under the transactional approach followed in Texas, a subsequent suit is barred if it arises out of the same subject matter as the prior suit, and that subject matter could have been litigated in the prior suit. Barr, 837 S.W.2d at 631. We explained in Barr that “a final judgment on an action extinguishes the right to bring suit on the transaction, or series of connected transactions, out of which the action arose.” Id. at 631 (citing Restatement (Second) of Judgments § 24(1) (1982)). Determining the scope of the “subject matter” or “transaction” of the prior suit requires “an analysis of the factual matters that make up the gist of the complaint, without regard to the form of action.” Id. at 630. This should be done pragmatically, “‘giving weight to such considerations as whether the facts are related in time, space, origin, or motivation, whether they form a convenient trial unit, and whether their treatment as a trial unit conforms to the parties’ expectations or business understanding or usage.’” Id. at 631 (quoting Restatement (Second) of Judgments § 24(2) (1982)). “Any cause of action which arises out of those same facts should, if practicable, be litigated in the same lawsuit.” Id. at 630.

B. Class Actions

Texas Rule of Civil Procedure 42 was adopted in 1941 and patterned after Federal Rule of Civil Procedure 23. Ford Motor Co. v. Sheldon, 22 S.W.3d 444, 452 (Tex. 2000). Rule 42 was fully revised in 1977 to conform to the 1966 federal amendments. Thus, we rely on our precedents and persuasive federal decisions and authorities interpreting current federal class action requirements. Id. (citing RSR Corp. v. Hayes, 673 S.W.2d 928, 931-32 (Tex. App._Dallas 1984, writ dism’d)).

Rule 42 is a form of joinder, a procedural mechanism established to increase judicial economy and efficiency for suits with parties too numerous for conventional joinder. See Gen. Tel. Co. of the Sw. v. Falcon, 457 U.S. 147, 155 (1981) (discussing Federal Rule of Civil Procedure 23); Hansberry v. Lee, 311 U.S. at 42-43 (“[t]he class suit was an invention of equity to enable it to proceed to a decree [when parties are so numerous as to make joinder] in conformity to usual rules of procedure . . . impracticable”); see also Beeson, 22 S.W.3d at 404 (citing 5 James Wm. Moore et al., Moore’s Federal Practice § 23.02 (3d ed. 1999)); 7A Charles Alan Wright et al., Federal Practice and Procedure § 1751 (3d ed. 2005). Rule 42 is intended to eliminate or reduce the threat of repetitive litigation, prevent inconsistent resolution of similar cases, and provide an effective means of redress for individuals whose claims are too small to make it economically viable to pursue them in independent actions. Sheldon, 22 S.W.3d at 452 (citing The American Law Institute, Report: Preliminary Study of Complex Litigation 35 (1987)). Although intended to be an efficient device, “there is no right to litigate a claim as a class action.” Sheldon, 22 S.W.3d at 452-53. A Texas court may certify a class action only if the plaintiff satisfies the requirements of Rule 42. Id. at 453.

Moreover, nothing mandates that a plaintiff pursue a remedy through the procedures of Rule 42. It is the plaintiff who chooses to resolve a claim through the class action mechanism. Though perhaps inefficient, every claim fit for class certification could be litigated outside the confines of Rule 42, just as every claim not suitable for class treatment must be. Thus, despite their unique procedural requirements, class actions provide no greater substantive rights than other procedural mechanisms of litigation. See Bernal, 22 S.W.3d at 432 (“[O]ur procedural rules do not permit the form of the proceeding to determine whether substantive legal principles will control.”); see also Rules Enabling Act, 28 U.S.C. § 2072(b) (2000) (stating that the Federal Rules of Civil Procedure shall not “abridge, enlarge or modify” preexisting rights). Class certification under Rule 42 was never meant to be an exception to res judicata, Beeson, 22 S.W.3d at 405, or to exist “in some sort of alternative universe outside our normal jurisprudence,” Bernal, 22 S.W.3d at 432. Basic principles of res judicata apply to class actions just as they do to any other form of litigation. Cooper v. Fed. Reserve Bank of Richmond, 467 U.S. 867, 874 (1984); Bernal, 22 S.W.3d at 432; see also Matsushita Elec. Indus. Co. v. Epstein, 516 U.S. 367, 377-79 (1996). Accordingly, we hold that claims not pursued, or abandoned, in a class suit seeking damages that proceeds to final judgment on other claims arising from the same subject matter are subject to preclusion from relitigation by the principles of res judicata.

Although it has not unequivocally decided the preclusive effect on subsequent actions of a final judgment in a class suit, the United States Supreme Court has acknowledged the same approach. In Hansberry and Ben-Hur, the Supreme Court indicated that a judgment in a class suit with an adequate representative may bind absent members of a class. See Hansberry v. Lee, 311 U.S. at 42 (holding that a prior decree in a class suit did not bind a class member because the named plaintiff did not adequately represent the interests of the class of property owners); Supreme Tribe of Ben-Hur v. Cauble, 255 U.S. 356, 363-67 (1921) (specifically holding that a federal district court had jurisdiction of a class action); see also Smith v. Swormstedt, 57 U.S. 288, 303 (1853) (“[A] court of equity permits a portion of the parties in interest to represent the entire body, and the decree binds all of them the same as if all were before the court.”); Geoffrey Hazard, Jr. et al., An Historical Analysis of the Binding Effect of Class Suits, 146 U. Pa. L. Rev. 1849, 1925-26 (1998). The Court further stated that it is permissible to hold that a judgment rendered in a class suit would be res judicata as to members of the class, and the Fourteenth Amendment does not compel a different rule for conclusiveness of the judgments in class suits. Hansberry, 311 U.S. at 42.

This approach has been challenged as unfair to absent class members who do not opt out and are bound by the final judgment. The argument continues that these absent members should be entitled to pursue individual claims in the same or other forums if their class claims are unsuccessful. We view the matter in a fundamentally different light, allowing individual choice by the plaintiffs with their consequent ramifications, to govern the litigation in class suits as in other suits. We do not dictate how litigants should structure their cases or which legitimate legal strategies they will pursue. We simply emphasize that legal consequences attach to tactical and strategic decisions in class actions as in other lawsuits. For instance, outside of class action suits, litigants tailor their actions to seek positive results from proceedings. Parties often decide to drop claims to achieve a desired objective: to enter a particular forum or venue, to avoid removal to federal court, to avoid expense for claims with little likelihood of success, to refrain from opening evidentiary doors harmful to client or case, or to focus the case on claims most likely to be successful. Similarly, a class may decide to pursue certain claims, abandon some, or not plead others. In the context of class actions this is not per se inappropriate, but a class representative must be aware that there are consequences associated with such a decision that could undermine certification. For example, a specific issue may involve too little commonality to allow for a class to survive the predominance requirements. See Bernal, 22 S.W.3d at 435. Having given putative class members the opportunity to choose, however, we will ordinarily hold class actions to the same res judicata standards as other forms of litigation, including enforcing the preclusion on abandoned claims which could have been litigated in the suit.

C. Could the Claims have been Litigated?

Daccach concedes that res judicata applies equally to class actions. He contends, however, that the claims he abandoned are procedurally barred from litigation in the class action by Rule 42, and therefore, res judicata cannot apply to preclude subsequent litigation of the claims that cannot be litigated through diligence in this class action suit. See Barr, 837 S.W.2d at 631 (“A subsequent suit will be barred if it arises out of the same subject matter of a previous suit and which through the exercise of diligence, could have been litigated in a prior suit.” (emphasis added)). To achieve certification despite his abandonment of class claims, Daccach argues for a rule that would preclude later litigation of only those causes of action that could have been certified in the prior class action. For reasons we explain, we decline to adopt Daccach’s proposed rule.

Most courts agree with Daccach’s concession that the basic principles of res judicata apply to class actions. See Cooper, 467 U.S. at 874 (and authorities cited therein); Hansberry, 311 U.S. at 42; Ben-Hur, 255 U.S. at 367; Beeson, 22 S.W.3d at 405. However, only a few cases can be read to support Daccach’s contention that only claims that could have been brought in a class action will be barred from subsequent litigation. One line of cases, followed by the court of appeals below and two other Texas courts of appeals, holds that under Federal Rule of Civil Procedure 23(c)(4), or identical Texas Rule of Civil Procedure 42(d), parties may bring or maintain a class action with respect to specific issues and will not suffer the preclusive effect of res judicata for those claims not actually litigated as unsuitable for class treatment. See, e.g., Sullivan, 79 F.R.D. at 265 (holding that splitting claims that are amenable to class treatment is “perfectly appropriate” in order to realize the savings of resources of courts and parties that Rule 23 is designed to facilitate); Compaq Computer Corp. v. Lapray, 79 S.W.3d 779, 793 (Tex. App.—Beaumont 2002), rev’d on other grounds, 135 S.W.3d 657 (Tex. 2004); Microsoft Corp. v. Manning, 914 S.W.2d 602, 610 (Tex. App.—Texarkana 1995, writ dism’d); see also 5 Herbert B. Newberg & Alba Conte, Newberg on Class Actions §16.22 (4th ed. 2002). The reasoning is based on Federal Rule of Civil Procedure 23(c)(4)(A) and Texas Rule of Civil Procedure 42(d)(1): “an action may be brought or maintained as a class action with respect to particular issues.” Some commentators characterize this approach as a “sophisticated transactional approach” that limits the basic transactional approach of res judicata and “includes trial convenience in its calculus.” 18A Charles Alan Wright et al., Federal Practice and Procedure § 4455 (2d ed. 2002) (suggesting the approach is supported by Section 24(2) of the Restatement (Second) of Judgments, which requires that a “transaction” must “be determined pragmatically, giving weight to such considerations as whether the facts are related in time, space, origin, or motivation, whether they form a convenient trial unit, and whether their treatment as a unit conforms to the parties’ expectations” (emphasis added)).

Daccach suggests his proposed rule is consistent with Texas jurisprudence on res judicata, relying on some of our cases not involving class actions: Pustejovskey v. Rapid-Am. Corp., 35 S.W.3d 643, 651 (Tex. 2000); Getty Oil v. Ins. Co. of N. Am., 845 S.W.2d 794, 801 (Tex. 1992).[10] These cases are factually and legally distinguishable. In Pustejovskey, we addressed “whether a plaintiff may bring separate actions for separate latent occupational diseases caused by exposure to asbestos.” 35 S.W.3d at 644. In addressing the single action rule as a species of res judicata, Justice Gonzales noted that “the transactional approach set out in Barr does not necessarily penalize a plaintiff for not bringing a claim arising out of the same facts that nonetheless could not have been litigated in the initial action.” Id. at 651. But one of the reasons we adopted a separate accrual rule in that case—and, by implication, the reason the claim could not have been litigated in the prior action—was that the damage-causing injury had not yet been discovered. Id. at 652. In this case we are not faced with latent injuries giving rise to claims that could not have been litigated in a prior action due to lack of discovery. The claims in this case were discovered or discoverable and then abandoned by Daccach to try to achieve class certification.

In Getty Oil, we held that a third party’s claim against a tortfeasor’s insurers was not precluded by prior litigation against the tortfeasor because, under the “no action” clause of the insurance policy and Texas Rule of Civil Procedure 38(c), the third party could not sue the insurer until there was a judgment against the tortfeasor. 845 S.W.2d at 801. Res judicata did not bar the second suit because the third party was contractually precluded from litigating the claim in the prior suit. No similar contractual agreement or rule governs in this case.

Daccach also relies on the following statement made by the Fifth Circuit: “If the court rendering judgment lacked subject-matter jurisdiction over a claim or if the procedural rules of the court made it impossible to raise a claim, then it is not precluded.” Browning v. Navarro, 887 F.2d 553, 558-59 (5th Cir. 1989) (citing Restatement (Second) of Judgments § 26(1)(c) (1982)); see also Montgomery v. Blue Cross & Blue Shield of Tex. Inc., 923 S.W.2d 147, 150 (Tex. App.—Austin 1996, writ denied) (citing Browning, 887 F.2d at 558-59). Similarly, the United States Supreme Court cited the Restatement (Second) of Judgments, which states that a second action arising from the same facts may be brought if

“[t]he plaintiff was unable to rely on a certain theory of the case or to seek a certain remedy or form of relief in the first action because of the limitations on the subject matter jurisdiction of the courts or restrictions on their authority to entertain multiple theories or demands for multiple remedies or forms of relief in a single action . . . .”

Thomas v. Wash. Gas Light Co., 448 U.S. 261, 283 n.29 (1980) (quoting Restatement (Second) of Judgments § 61.2(c) (Tent. Draft No. 5, 1978)); see also Gunnells v. Healthplan Servs., Inc., 348 F.3d 417, 432 (4th Cir. 2003) (rejecting contention that plaintiffs’ individual direct claims would be barred because a class action, “of course, is one of the recognized exceptions to the rule against claim-splitting”) (citing 18 James Wm. Moore et al., Moore’s Federal Practice § 131.40[3][e][iii] (3d ed. 1999)); Restatement (Second) of Judgments § 26(1)(c) (1982). We are not persuaded by this argument.

First, the issue in Browning and Montgomery was whether the original decision-maker had subject matter jurisdiction to adjudicate the claim sought to be relitigated in district court. In Browning, the court barred litigation of a subsequent fraud claim because the bankruptcy court had subject matter jurisdiction to hear the claim in the prior suit by the party. 887 F.2d at 558-59. In Montgomery, the plaintiff was not barred from litigating extra-contractual claims because the administrative agency that presided over the prior suit did not have jurisdiction to hear those claims. 923 S.W.2d at 150. These rulings turned on a lack of jurisdiction and do not inform our reasoning in this case because rule 42 of the Texas Rules of Civil Procedure does not affect a trial court’s subject matter jurisdiction.

Second, we do not believe section 26(1)(c) of the Restatement speaks to the class action context. Nothing forces plaintiffs seeking damages into a class suit. They may decide to opt out and pursue their claims individually with separate counsel or decide that the size of the claim does not justify the cost of pursuing it. On the other hand, plaintiffs may choose to litigate their claims under Rule 42 because it provides a more efficient and perhaps less expensive means of litigating certain claims. It is the class representative’s choice to seek certification, and the putative class members’ decision not to opt out of the class, that restricts their ability to rely on certain theories of recovery that are unsuitable for class treatment. Any restrictions that class action requirements place on a trial court’s ability to entertain specific theories of recovery in a class suit arise solely because of the choice to seek class certification. By this choice class members may put at risk their ability to litigate certain other claims not suitable for class treatment. These restrictions follow the individual decisions of the class members and are distinct from the jurisdictional restrictions that may be placed on a bankruptcy court or administrative agency, to which we believe section 26(1)(c) of the Restatements (Second) of Judgments more appropriately applies.

We also are unpersuaded that an exception from res judicata principles for claims abandoned as unsuitable for class treatment is supported by the asserted precedent from the United States Supreme Court. In Cooper v. Federal Reserve Bank of Richmond, the Supreme Court announced that general principles of res judicata apply in class actions, but nevertheless determined that for the Title VII claims brought in a class suit under rule 23 of the Federal Rules of Civil Procedure, certain plaintiffs were not barred from subsequently bringing individual discrimination claims. 467 U.S. 867 (1984). Four employees sought certification of a class of employees alleged to have been discriminated against by a bank that engaged in “policies and practices” of racial discrimination in violation of Title VII Section 1981. Id. at 869-70. Upon receiving notice, six other employees joined the class. Id. at 870-71. After a trial in which all named plaintiffs testified, the district court found the bank had engaged in a pattern and practice of racial discrimination for certain levels of employees, but found as to other levels of employees that the discrimination was not pervasive enough to order relief. Id. at 870-72. The six joining class members moved to intervene to allege that each had been individually denied promotions for discriminatory reasons. Id. at 872. The motions were denied, and five of the six employees then filed a separate action against the Bank alleging violations of Section 1981. Id. On interlocutory appeal of the separate action, the federal circuit court concluded the doctrine of res judicata precluded the plaintiffs from maintaining their individual race discrimination claims because they were bound by the judgment in the class action. See EEOC v. Fed. Reserve Bank of Richmond, 698 F.2d 633, 674 (4th Cir. 1983). The U.S. Supreme Court reversed.

The Court began by stating “[t]here is of course no dispute that under elementary principles of prior adjudication a judgment in a properly entertained class action is binding on class members in any subsequent litigation.” Cooper, 467 U.S. at 874. The holding that basic principles of res judicata apply to class actions was part of a lengthy discussion of the “crucial difference between an individual’s claim of discrimination and a class action alleging a general pattern or practice of discrimination.” Id. at 876. The suggestion is that a class claim for a pattern or practice of discrimination involves factual issues distinct from those in a class member’s individual discrimination lawsuit. Id. at 876-77. The Court also expressly noted that the district court “pointedly refused to decide the individual claims” of the plaintiffs now seeking adjudication of the claims in a separate action. Id. at 881. According to the Court, therefore, the court of appeals erred in attaching preclusive effect to the class action because it was not dispositive of the individual claims alleged in the separate action. Id. at 880.

We read Cooper not as an exception to res judicata but as an application of its elements—a subsequent claim might not be barred if it does not involve the same factual issues that were litigated in the prior class action, a situation that can arise in the unique context of Title VII pattern and practice litigation. See, e.g., Munoz v. Orr, 200 F.3d 291, 307 (5th Cir. 2000) (“We note that the failure of proof on the class claim does not bar all individual class members from bringing their own suits, provided that they do not base their claims solely on issues already adjudicated in this action and that they can show individualized proof of discrimination.”) (citing Cooper, 467 U.S. at 880); Allison v. Citgo Petroleum Corp., 151 F.3d 402, 425 n.23 (5th Cir. 1998) (distinguishing Cooper and stating that a subsequent disparate impact class action will be barred by res judicata and collateral estoppel because it will inevitably contain the same factual issues as were litigated in the pattern or practice class action); see also Marshall v. Kirkland, 602 F.2d 1282, 1298 (8th Cir. 1979) (pre-Cooper case indicating that subsequent individual discrimination claims will not be precluded because the issues were not actually litigated in prior class action and there was no notice to the b(2) class that such claims might be waived); Tobias Barrington Wolff, Preclusion in Class Action Litigation, 105 Colum. L. Rev. 717, 727 (2005) (arguing that the result in Cooper “may represent the correct rule in a Title VII class action, [but] it does not flow inevitably from an application of basic claim preclusion principles”).

In addition, we find it significant that the U.S. Supreme Court emphasized the district court’s pointed refusal to decide the plaintiff’s individual claims. Cooper, 467 U.S. at 881. It would hardly seem appropriate to bar subsequent litigation of a dispute that a prior court refused to decide. In this respect, we find Texas Rule of Civil Procedure 42(d) instructive.

Rule 42(d) provides that “an action may be brought or maintained as a class action with respect to particular issues.” The rule, like its federal counterpart, “is a housekeeping rule that allows courts to sever the common issues for a class trial.” Castano v. Am. Tobacco Co., 84 F.3d 734, 746 n.21 (5th Cir. 1996). But, while we agree that Rule 42(d) allows a trial court to consider certifying a class whose representative has abandoned or split claims, we decline to take the further step of excepting a final judgment in such a class action from the principles of res judicata. Class members may be precluded from asserting those claims in subsequent individual litigation if they arose from the same transaction or subject matter and could have been litigated in the prior suit. See Barr, 837 S.W.2d at 631. Aggregation of claims in an appropriate class action is a more efficient way to resolve numerous disputes at once. However, efficiency is defeated if the tactfully structured dispute that is finally resolved in class suits may be relitigated in the same or other forums.

We caution, also, that Rule 42(d) cannot be used to manufacture compliance with the certification prerequisites. See Castano, 84 F.3d at 745 n.21 (“A district court cannot manufacture predominance through the nimble use of [Federal Rule of Civil Procedure 23(c)(4)].”). As explained below, the splitting or abandoning of certain claims may affect certification of the class in other ways.

D. Effects on Certification

The different procedural posture of the Cooper case raises another important issue. There the Court was faced with an interlocutory appeal of the actual subsequent claims being asserted, as opposed to this case in which we are asked to predetermine the preclusive effect of claims that may or may not be asserted in later litigation. In the only other case in which the United States Supreme Court has addressed res judicata in the class action context, a dissenting justice noted that “[a] court conducting an action cannot predetermine the res judicata effect of the judgment; that effect can be tested only in a subsequent action.” Matsushita Elec. Indus. Co. v. Epstein, 516 U.S. 367, 396 (1996) (Ginsburg, J., concurring in part and dissenting in part) (citing 7B Charles Alan Wright et al., Federal Practice and Procedure § 1789 (2d ed. 1986)). We generally agree with this maxim, but hasten to address a due process concern that this temporal distinction may raise.

Some courts have applied the principles of res judicata, but refused to hold that subsequent claims would be precluded due to a lack of adequate notice to class members regarding the claims being litigated in the class action. See, e.g., Wright v. Collins, 766 F.2d 841,847 (4th Cir. 1989) (no preclusion because no notice); Aspinall v. Phillip Morris Cos., 813 N.E.2d 476, 488-89 n.19 (Mass. 2004) (allowing member of class certified on economic damages theory to pursue individual claim for personal injury not suitable for certification in part because no “opt-out” provisions in state rules). The same reasoning has been the basis for court holdings that mandatory class actions for injunctive relief certified under Federal Rule of Civil Procedure 23(b)(2) cannot preclude subsequent individual claims for damages, even if based on the same events. See Hiser v. Franklin, 94 F.3d 1287, 1291 (9th Cir. 1996); Fortner v. Thomas, 983 F.2d 1024, 1031 (11th Cir. 1993); Brown v. Ticor Title Insurance Co., 982 F.2d 386 (9th Cir. 1992); Norris v. Slothouber, 718 F.2d 1116, 1117 (D.C. Cir. 1983); Johnson v. Gen. Motors Corp., 598 F.2d 432, 437-38 (5th Cir. 1979) (finding that due process requires notice to absent class members before individual monetary damages could be barred and, though an absent class member could be bound by the res judicata effect of a Rule 23(b)(2) class action judgment as to injunctive or declaratory relief, he could not be barred from pursuing his individual monetary claim); Coleman v. Gen. Motors Acceptance Corp., 220 F.R.D. 64, 80-84 (M.D. Tenn. 2004); In re Jackson Lockdown/MCO Cases, 568 F. Supp. 869, 888-89 (E.D. Mich. 1983); Jahn ex rel. Jahn v. ORCR, Inc., 92 P.3d 984, 985 (Colo. 2004) (en banc). Although we are not faced here with a b(2) class, notice and due process still demand our attention.

To have preclusive effect a prior judgment cannot be “constitutionally infirm.” Kremer v. Chem. Constr. Corp., 456 U.S. 461, 482 (1982). Due process requires “that the named plaintiff at all times adequately represent the interests of the absent class members,” as well as “notice plus an opportunity to be heard and participate in the litigation.” Shutts, 472 U.S. at 812; see also Tex. R. Civ. P. 42(a)(4); Amchem Prods., Inc. v. Windsor, 521 U.S. 591, 625‑26 (1997). We noted as much in Compaq Computer Corp. v. Lapray, where we stated that due process may require that class members be given notice of the class action and an opportunity to opt out and preserve claims that a class representative has abandoned. 135 S.W.3d 657, 668 (Tex. 2004); see also Gen. Motors Corp. v. Bloyed, 916 S.W.2d 949, 953 (Tex. 1996) (“The United States Supreme Court has made it clear that due process requires adequate representation of the interests of absentee class members that the judgment will bind.”). Although a certifying court cannot precisely predetermine the res judicata effect of a class action, it initially must protect the due process rights of absent class members by ensuring that the class representative adequately represents their interests. See Epstein v. MCA, Inc., 179 F.3d 641, 648 (9th Cir. 1999) (“[A]bsent class members’ due process right to adequate representation is protected not by collateral review, but by the certifying court initially, and thereafter by appeal within the state system and by direct review in the United States Supreme Court.”).

Some courts have reconciled the tension between the trial court’s inability to predetermine res judicata and its burden to protect class members’ due process rights by requiring the trial court to assess the “risk” that uncertified claims may be forever barred. See Clark v. Experian Info. Solutions, Inc., No. Civ. A. 8:001217-24, 2001 WL 1946329, at *4 (D.S.C. Mar. 19, 2001) (stating that offering only some claims for class certification when other, more lucrative claims could not be certified “defeats adequate representation since it places absent class members at the risk of having other claims forever barred by res judicata”); Zachery v. Texaco Exploration & Prod., Inc., 185 F.R.D. 230, 243 (W.D. Tex. 1999) (assessing the risk of abandoned monetary claims that members of the b(2) pattern-and-practice class may face in trying to bring later individual claims); Thompson v. Am. Tobacco Co., 189 F.R.D. 544, 550-51 (D. Minn. 1999) (refusing to certify because the “possible prejudice to class members” resulting from claim preclusion in the future “is simply too great”); Feinstein v. Firestone Tire & Rubber Co., 535 F. Supp. 595, 606 (S.D.N.Y. 1982) (refusing to certify class action for economic losses where plaintiffs also had personal injury claims because of significant risks that class members would “later [be told] that they had impermissibly split a single cause of action”); Millett v. Atl. Richfield Co., No. Civ. A. CV-98-555, 2000 WL 359979, at *9 (Me. Super. Ct. Mar. 2, 2000) (explaining that asserting claims for injunctive relief while leaving personal injury claims unraised places class members at risk of subsequent claim preclusion defense); Small v. Lorillard Tobacco Co., Inc., 679 N.Y.S.2d 593, 601-02 (N.Y. App. Div. 1998) (stating that paring down class claims to avoid certification problems creates impermissible “risk” of adverse preclusive effect). We agree with this approach.

A class representative’s decision to abandon certain claims may be detrimental to absent class members for whom those claims could be more lucrative or valuable, assuming those class members do not opt out of the class. Abandoning such claims, or claims “reasonably expected” to be raised by class members, could undermine the adequacy of the named plaintiff’s representation of the class. See City of San Jose v. Super. Ct. of Santa Clara County, 525 P.2d 701, 711-13 (Cal. 1974). But see Regions Bank v. Lee, 905 So. 2d 765, 772-73 (Ala. 2004) (rejecting adequacy challenge based on effect of res judicata because abandoned claims would involve a different cause of action against a different defendant than that involved in the class action). We hold, therefore, that Texas Rule of Civil Procedure 42 requires the trial court, as part of its rigorous analysis, to consider the risk that a judgment in the class action may preclude subsequent litigation of claims not alleged, abandoned, or split from the class action. The trial court abuses its discretion if it fails to consider the preclusive effect of a judgment on abandoned claims, as res judicata could undermine the adequacy of representation requirement. See Wolff, 105 Colum. L. Rev at 722 (“[T]he preclusion inquiry would sometimes reveal significant obstacles to class certification . . . .”).

A trial court could, however, determine that the risk of preclusion is not high enough to refuse certification. For instance, the abandoned claims may be insignificant, unlikely to succeed in any proceeding, or not valuable. Some abandoned claims may be alleged against different defendants or may not be ripe for litigation, in which case res judicata would not apply. But, because we hold class actions seeking damages to the same res judicata standards as other forms of litigation, including enforcing the preclusion on abandoned claims which could have been litigated in the suit, it is critical that putative class members be given adequate notice and an opportunity to exclude themselves from the class form of proceeding so that they may preserve individual claims that may otherwise be barred from subsequent litigation. See Richard A. Nagareda, Preexistence Principle and the Structure of the Class Action, 103 Colum. L. Rev. 149, 216 (2003) (contending that the ability to opt out respects the rights of class members to control their claims).[11]

Under Rule 42, notice must be given to the class, and class members given an opportunity to opt out, before the trial court addresses the merits of the class claims. See Bally Total Fitness Corp. v. Jackson, 53 S.W.3d 352, 360 (Tex. 2001) (Owen, J. dissenting); see also Am. Pipe & Const. Co. v. Utah, 414 U.S. 537, 548 (1974) (explaining that Federal Rule of Civil Procedure 23 was amended to avoid “one‑way” intervention issue arising when class members were not identified before court made decisions going to merits). To properly protect absent class members, a trial court must rigorously analyze Texas Rule of Civil Procedure 42’s prerequisites prior to sending any necessary class notice, as this analysis will likely affect the class definition and requisites for the notice. McAllen Med. Center, Inc. v. Cortez, 66 S.W.3d 227, 232 (Tex. 2001). Rule 42 sets out the following requirements for notice in a b(3) class action:

For any class certified under Rule 42(b)(3), the court must direct to class members the best notice practicable under the circumstances, including individual notice to all members who can be identified through reasonable effort. The notice must concisely and clearly state in plain, easily understood language: (i) the nature of the action; (ii) the definition of the class certified; (iii) the class claims, issues, or defenses; (iv) that a class member may enter an appearance through counsel if the member so desires; (v) that the court will exclude from the class any member who requests exclusion, stating when and how members may elect to be excluded; and (vi) the binding effect of a class judgment on class members under Rule 42(c)(3).

Tex. R. Civ. P. 42(c)(2)(B). Ultimately, to certify a class in which the representatives have abandoned claims in favor of pursuing certain class claims, raising a risk of preclusion for absent class members, effective notice must be given to these absent members of an identified class regarding the preclusive effect that may attach to their individual claims. The unnamed members may then exercise independent judgment and chose to remain in the class or opt out.

VI. Class Definition

Citizens challenges the court of appeals’ approval of the class definition on grounds that the definition fails to identify a presently ascertainable class from objective criteria and creates a “fail-safe” class. Citizens specifically points to the definition’s exclusionary language, which it contends creates a future contingency that grants each plaintiff a post-judgment opportunity to exclude himself from the class.

A class is properly defined only if its members are presently ascertainable by reference to objective criteria. Intratex Gas Co. v. Beeson, 22 S.W.3d 398, 403 (Tex. 2000). A class cannot be defined by subjective criteria or require analysis of the merits of the case. Id. A class definition that “rests on the paramount liability question” is not based on objective criteria because “the trial court has no way of ascertaining whether a given person is a member of the class until a determination of ultimate liability as to that person is made.” Id. at 404. In other words, the class is defined as members who succeed on the ultimate liability question. Such a “fail‑safe class” is also impermissible because it binds members only by a judgment favorable to them but not by a judgment favorable to the defendants. Id. at 405.

The trial court’s certification order defined the class as follows:

The Class consists of all persons, who, during the Class Period (August 6, 1996 through the date the Class is certified): (1) purchased a CICA Policy and executed an assignment to a trust for the purchase of Citizens, Inc. stock, or (2) paid any money that, pursuant to a CICA Policy and assignment to a trust, was for the purchase of Citizens, Inc. stock, or (3) were entitled to any cash benefits from a CICA Policy that, pursuant to a CICA Policy and assignment to a trust, were for the purchase of Citizens, Inc. stock. Specifically excluded from the Class are all persons who, within the time period established by the judgment, do not surrender their CICA Policies and take the other actions required to obtain the relief awarded by the Court.

Because the exclusionary language of the trial court’s class definition partially defined the class by actions taken after the judgment, it failed to create a class that could be objectively ascertained before judgment. Id. at 403-04. Although the contours of the class did not “rest on whether the CICA policies qualify as securities or whether the policies were in fact sold or offered for sale from Texas,” and thus was not invalid as a traditional “fail-safe” class, it did however allow putative class members to essentially opt out of the suit after the judgment and thus escape the binding effect of the judgment. This class definition was improper.

The court of appeals revised the definition, substituting the word “remedy” for the word “Class” in the definition’s last sentence:

Specifically excluded from the remedy are all persons who, within the time period established by the judgment, do not surrender their CICA Policies and take the other actions required to obtain the relief awarded by the Court.

105 S.w.3d 722 n.7 (emphasis added). We conclude that this corrects the defective class definition. This sentence simply states what is true is any case: a litigant, or in this case, a class member, may elect not to exercise a right to a remedy rendered in a judgment. Instead, it reiterates the obvious fact that even in the event of a favorable judgment, a class member may elect to keep his or her policy and decline the remedy. Regardless, the class member would still be bound by the judgment.

VII. Attorney’s Fees

Citizens argues that the class’s claim for attorney fees involves individual questions of fact because the statute allows recovery if “the court finds that the recovery would be equitable in the circumstances.” Tex. Rev. Civ. Stat. art. 581-33D(7). The class claim—that Citizens offered or sold securities in or from Texas without registering with the Texas Securities Board—implicates Citizens’ overall business scheme. The class makes no allegation of conduct varying from buyer to buyer with regard to this claim. We agree with the court of appeals that because “the heart of the dispute turns only on whether the jury decides [whether] the CICA policies constitute securities and whether they were sold from Texas,” attorney’s fees could be awarded based on Citizens’ marketing conduct in general.

VIII. Conclusion

As part of a trial court’s rigorous analysis for certification of a Rule 42(b)(3) class, a trial court must assess all of Rule 42’s requirements with awareness of res judicata’s preclusive effect on abandoned claims. See Bernal, 22 S.W.3d at 435. Although we hold that res judicata principles are applicable in class suits and could bar claims abandoned by the class representative, we do not dictate how plaintiffs should structure their case or which legitimate legal strategies they will pursue. We simply note that legal consequences attach to tactical and strategic decisions in class actions as in other lawsuits. While it is not per se inappropriate to abandon claims or for the trial court to certify a specific-issue class, the requirements of class certification must still be met. As we have cautioned above, a class representative’s abandonment of claims can affect the class representative’s ability to satisfy these requirements. Here the trial court failed to evaluate Rule 42’s prerequisites in light of the claims abandoned by the class representative. Therefore, we reverse the court of appeal’s affirmance of the trial court’s class certification order, decertify the class, and remand the case to the trial court for further proceedings consistent with this opinion. Tex. R. App. P. 60.2(d).

________________________________________

J. Dale Wainwright

Justice

OPINION DELIVERED: March 2, 2007

---

^[1] Article 581-33A(1) of the Texas Securities Act provides:

A person who offers or sells a security in violation of Section 7, 9 (or a requirement of the Commissioner thereunder), 12, 23C, or an order under 23A or 23-2 of this Act is liable to the person buying the security from him, who may sue either at law or in equity for rescission or for damages if the buyer no longer owns the security.

Tex. Rev. Civ. Stat. art. 581-33A(1). Section 12A states “no person, firm, corporation or dealer shall, directly or through agents, offer for sale, sell or make a sale of any securities in this state without first being registered as in this Act provided.” Id. art. 581-12A.

^[2] The court of appeals modified the last sentence of the class definition by substituting the word “remedy” for the word “Class.” 105 S.W.3d at 721-22.

^[3] A negative value suit is one in which the stakes to each member are too slight to repay the cost of suit. Sw. Ref. Co. v. Bernal, 22 S.W.3d 425, 439 (Tex. 2000).

^[4] In 2003, the Legislature amended sections 22.225(b) and (d) to give this Court jurisdiction over interlocutory appeals of orders certifying or refusing to certify a class. See Act of June 2, 2003, 78th Leg., R.S., ch. 204, § 1.02, 2003 Tex. Gen. Laws 847, 848-49. The amendments apply to petitions filed on or after September 1, 2003. Act of June 2, 2003, 78th Leg., R.S., ch. 204, § 23.02(a), (d), 2003 Tex. Gen. Laws 847, 898-99. Because Citizens filed its petition for review in June 2003, the amendments do not govern our jurisdiction in this case. See Hoff v. Nueces, 153 S.W.3d 45, 48 n.2 (Tex. 2004).

^[5] On July 31, 2002, the trial court certified the class pursuant to Rule 42(b)(4). Effective January 1, 2004, however, the Court deleted as unnecessary subparagraph (b)(3) from Rule 42 and substituted in its place—with minor changes not pertinent here—former subparagraph (b)(4). Tex. R. Civ. P. 42 cmt.–2003. Our references here are to current subparagraph (b)(3), which includes former subparagraph (b)(4).

^[6] The term “Blue Sky laws” was used by Justice McKenna writing for the U.S. Supreme Court in Hall v. Geiger-Jones Co., 242 U.S. 539 (1917). He stated: “The name that is given to the law indicates the evil at which it is aimed, that is . . .’speculative schemes which have no more basis than so many feet of “blue sky”‘; or, as stated by counsel in another case, ‘to stop the sale of stock in fly-by-night concerns, visionary oil wells, distant gold mines and other like fraudulent exploitations.’“ Id. at 550. Thus, Blue Sky laws were promulgated by states to protect investors from nefarious securities schemes.

^[7] In a similar fashion, the commentary to section 6 of the Restatement provides the following:

b. Intended range of application of statute. A court will rarely find that a question of choice of law is explicitly covered by statute. That is to say, a court will rarely be directed by statute to apply the local law of one state, rather than the local law of another state, in the decision of a particular issue. On the other hand, the court will constantly be faced with the question whether the issue before it falls within the intended range of application of a particular statute. . . . If the legislature intended that the statute should be applied to the out-of-state facts involved, the court should so apply it unless constitutional considerations forbid. On the other hand, if the legislature intended that the statute should be applied only to acts taking place within the state, the statute should not be given a wider range of application. . . . When the statute is silent as to its range of application, the intentions of the legislature on the subject can sometimes be ascertained by a process of interpretation and construction.

Restatement (Second) of Conflicts of Laws § 6(1) cmt. b (emphasis added). While we generally agree with this comment, the emphasized sentence does not fully explain the approach we follow in Texas to determine the extraterritorial affect of Texas statutes. See Marmon, 430 S.W.2d at 182.

^[8] The words “in this state” first appeared in Texas securities statutes in 1925 in a registration provision substantially different from the current version. Tex. Rev. Civ. Stat. arts. 579-600. In 1935, the words “in this state” were used in a securities registration provision more similar to the current version. Act of April 16, 1935, 44th Leg., R.S., ch. 100, _ 2, 1935 Tex. Gen. Laws 255, 256-59. The language of current Section 12 was adopted in very similar form in 1955, then re-adopted in its current form as section 12 of the Texas Securities Act of 1957.

^[9] The U.S. Supreme Court has identified two primary constitutional limitations on the application of a state’s substantive law to conduct occurring, at least in part, outside the state—the Due Process Clause and the Interstate Commerce Clause. Shutts, 472 U.S. at 818-22; Travelers Health Ass’n v. Virginia, 339 U.S. 643, 649 (1950). Citizens only raises a due process violation.

^[10] We reject the argument predicated on Van Dyke v. Boswell, O’Toole, Davis & Pickering that the class could circumvent this conclusion by obtaining a severance of its Texas Blue Sky claim into a separate action. 697 S.W.2d 381, 384 (Tex. 1985).

^[11] Because counsel and class representatives may have little or no interest in seeing absent class members opt out of a class, the trial court ensures that notice is effective under Texas Rule of Civil Procedure 42(c)(2)(B). See Linda S. Mullenix, No Exit: Mandatory Class Actions in the New Millennium and the Blurring of Categorical Imperatives, 2003 U. Chi. Legal F. 177, 245 (2003) (lamenting that opt-out claimants may be “fungible hostages” in a “class action game”).