Justice HECHT delivered the opinion of the Court, in which Chief Justice JEFFERSON, Justice MEDINA, Justice GREEN, Justice WILLETT, and Justice LEHRMANN joined.

The issue we address in this case is whether a statute that limits certain corporations' successor liability for personal injury claims of asbestos exposure violates the prohibition against retroactive laws contained in article I, section 16 of the Texas Constitution1 as applied to a pending action. We hold that it does, and therefore reverse the judgment of the court of appeals2 and remand the case to the trial court.

I

In 2002, petitioner Barbara Robinson ("Robinson") and her husband, John, Texas residents, filed suit alleging that John, age 63, had contracted mesothelioma from workplace exposure to asbestos products. As often happens, John had used several such products over the course of his life, and the Robinsons sued twenty-one defendants, including respondent Crown Cork & Seal Co., alleging that they were all jointly and severally liable. With respect to Crown, the Robinsons claimed that during John's service in the United States Navy from 1956 to 1976, he worked with asbestos insulation manufactured by the Mundet Cork Corporation, and that when Crown and Mundet merged, Crown succeeded to Mundet's liabilities.

Crown has never itself engaged in the manufacture or sale of asbestos products.3 It manufactures metal bottle-caps, known in the industry as "crowns", and other packaging for consumer goods. Crown and its affiliates have over 20,000 employees around the world, about 1,000 of whom work in Texas at facilities in Conroe, Sugar Land, and Abilene. In 2009, the parent company reported $1.193 billion gross profit on $7.938 billion net sales.4

In November 1963, Crown's predecessor, a New York corporation with the same name, which was then the nation's largest manufacturer of crowns, acquired a majority of the stock in Mundet, another New York corporation, which besides insulation, also manufactured crowns. Within ninety days, in February 1964, Mundet sold all its assets related to its insulation business. Two years later, in February 1966, the companies merged. In 1989, Crown's predecessor was reincorporated as Crown, a Pennsylvania corporation.

Crown acknowledges that under New York and Pennsylvania law, it succeeded to Mundet's liabilities, which, as pertaining to Mundet's asbestos business, have been hefty. Over the years, Crown has been named in thousands of lawsuits claiming damages from exposure to asbestos manufactured by Mundet. While Crown acquired Mundet for only about $7 million, by May 2003 Crown had paid over $413 million in settlements, and Crown's parent company estimated in its 2003 Annual Report that payments could reach $239 million more.5 Mundet's aggregate insurance coverage totaled $3.683 million.6

At first, Crown did not contest its successor liability to the Robinsons for any compensatory damages; consequently, the trial court granted the Robinsons' motion for partial summary judgment on that issue. But about the same time, the Texas Legislature enacted Chapter 149 of the Texas Civil Practice and Remedies Code, which limits certain corporations' successor liability for asbestos claims.7 Chapter 149 applies (with exceptions not relevant here) to "a domestic corporation or a foreign corporation that has . . . done business in this state and that is a successor which became a successor prior to May 13, 1968"8—a date by which, the Legislature appears to have thought, the dangers of asbestos should have been commonly known.9 For a covered corporation (again with some exceptions not relevant here), "the cumulative successor asbestos-related liabilities . . . are limited to the fair market value of the total gross assets of the transferor determined as of the time of the merger or consolidation",10 including "the aggregate coverage under any applicable liability insurance that was issued to the transferor . . . collectable to cover successor asbestos-related liabilities".11 This cap does not apply to a successor that continued in the asbestos business after the consolidation or merger.12 By restricting application of the cap to a corporation that had never engaged in selling asbestos products itself and had succeeded to another's liability for asbestos claims at a time when the extent of that liability was not fully appreciated, the supporters of Chapter 149 intended to protect only what they called the "innocent successor".

Chapter 149 contains a choice-of-law provision, making it applicable, "to the fullest extent permissible under the United States Constitution, . . . to the issue of successor asbestos-related liabilities" in Texas courts.13 Furthermore, the Legislature made Chapter 149 applicable to all actions:

Because the Act of which Chapter 149 was part, House Bill 4, passed by more than a two-thirds vote in both the House and Senate,15 it took effect immediately on approval by the Governor,16 which occurred on June 11, 2003.

House Bill 4 was massive tort reform legislation, of which Chapter 149 was a very small piece—two pages of a 52-page bill.17 Chapter 149 was not included in the bill as filed but was added when the bill came to the House floor by an amendment offered by the bill's sponsor. When asked which manufacturers "in particular" would be protected, the sponsor replied that he was "advised that there's one in Texas, Crown Cork and Seal".18 Although House debate on the whole bill took days, debate on Chapter 149 lasted just over an hour.19 Four unfriendly amendments,20 one of which would have made Chapter 149 inapplicable to "successor asbestos-related liabilities that were assumed or incurred before [its] effective date",21 all failed by wide margins. In the Senate, Chapter 149 was significantly revised but drew only one brief comment in that chamber, this observation by the committee chair as hearings commenced: "This, members, is the Crown Cork and Seal asbestos issue. What we have put in this bill is what I understand to be an agreed arrangement between all of the parties in this matter."22

No legislative findings or statement of purpose accompanied Chapter 149. But after the conference committee report on House Bill 4 was adopted in the House, the sponsor inserted a "statement of legislative intent" in the House Journal, which did not mention Crown and explained the policy basis for Chapter 149 as follows:

The statement described Chapter 149 as a "new concept" that was being tested "by taking one step at a time and providing realistic relief to those innocent successor corporations most at peril financially without limiting every type of asbestos liability."24 According to the statement, Chapter 149's restrictions had been crafted to ensure that "the benefits of this legislation should be limited . . . to those successor corporations who were the most innocent about the potential hazards of asbestos" and "were also at the greatest financial peril, especially those threatened with bankruptcy".25

Crown promptly moved for summary judgment under the new law, requesting that the prior order establishing its successor liability to the Robinsons be vacated and that their claims for asbestos exposure be dismissed. Crown asserted that the summary judgment evidence established that its merger with Mundet occurred before May 13, 1968, that it had never engaged in Mundet's insulation business, and that its successor asbestos-related liabilities, already more than $413 million, greatly exceeded the fair market value of Mundet's total gross assets determined as required by the statute26—about $15 million in 1966 (some $57 million in 2003 dollars). Thus, Crown contended, Chapter 149 barred the Robinsons from recovering on their claims. In response, the Robinsons argued that the record did not establish the applicability of Chapter 149,27 or if it did, the statute violated several provisions of the Texas Constitution.28

The trial court granted Crown's motion. Days later, John Robinson died.29 Barbara Robinson amended her petition to assert statutory wrongful death30 and survival actions31 against Crown and the other defendants still remaining in the case. (Several defendants had settled for amounts totaling $859,067 and been dismissed.) Without addressing these statutory actions, Crown moved to sever the summary judgment to make it final and appealable,32 and the trial court granted the motion. The court also stayed proceedings in Robinson's case against the other defendants.

On appeal, Robinson contends that Chapter 149 is a retroactive law prohibited by article I, section 16 of the Texas Constitution. The law is well-settled, she asserts, that the Legislature has no authority to extinguish vested rights, and that her accrued cause of action against Crown is a vested right. A majority of the court of appeals did not "find the law on vested rights to be as consistent and lucid as Mrs. Robinson claims"33 and concluded that it provides "no clear answer" to whether Chapter 149 is an invalid retroactive law.34 Relying on this Court's decision in Barshop v. Medina County Underground Water Conservation District,35 the court decided that whether a law is unconstitutionally retroactive depends not on whether it infringes upon a vested right but on whether it is a "`valid exercise of the police power by the Legislature to safeguard the public safety and welfare'".36 Whether an exercise of the police power is valid, the court of appeals determined, depends on

The court found that "the purpose for which [Chapter 149] was enacted—the financial viability of the State and businesses in the State—is a valid exercise of police power."38 The court further found that the restrictions in the statute left "the pool of potential defendants as large as possible for claimants having valid claims for damages resulting from asbestos products",39 thereby limiting the "detrimental impact on plaintiffs such as the Robinson so that [it] was not out of proportion to the end sought".40 Concluding that deference must be given to the Legislature in the exercise of its police power, the court held that Chapter 149 is not unconstitutionally retroactive because it is "(1) within the Legislature's police power and (2) narrowly tailored (a) to protect the most innocent corporations hard hit by asbestos litigation but (b) to leave the potential pool of asbestos defendants as large as possible."41 Accordingly, the court affirmed the summary judgment.42

The dissent disagreed with the majority's approach to assessing unconstitutionality. It argued that "the Legislature has no police power to enact retroactive laws in violation of section 16", even if reasonably exercised.43 Barshop notwithstanding, the dissent insisted, "the weight of precedent . . . requires the use of the vested-rights analysis."44 The dissent contended that "an accrued cause of action is a vested right",45 rejecting some caselaw that "an accrued claim is not vested until it is reduced to a judgment final by appeal".46 Thus, the dissent reasoned, "[b]ecause Mrs. Robinson's claims accrued and were pending in the trial court when [Chapter 149] took effect, Mrs. Robinson held vested rights in these claims that could not be destroyed",47 irrespective of the fact that Chapter 149 "does not bar all of Mrs. Robinson's remedy for the claimed injuries because she can sue other companies not protected".48 Without assessing the reasonableness of the Legislature's action, the dissent concluded that Chapter 149 is unconstitutionally retroactive because the "Legislature created a new substantive defense to successor liability and made it immediately effective in all pending cases, destroying Mrs. Robinson's vested rights in her accrued tort claims against Crown".49

We granted Robinson's petition for review.50 Another court of appeals, also divided, has since reached the opposite result from the court of appeals in this case.51

II

As a threshold matter, it is important to note the precise issue before us. The Robinsons' pleading on which Crown moved for summary judgment asserted common-law causes of action for negligence and strict liability, and claimed compensatory and punitive damages.52 For herself, Barbara claimed damages for John's medical expenses that she had incurred, as well as her loss of consortium and mental anguish, and punitive damages. Had John lived, the summary judgment would have disposed of all the Robinsons' claims against Crown.

But John died a few days after summary judgment was granted, and Robinson amended her petition to add statutory wrongful death and survival actions. The record does not reflect that Crown moved for summary judgment on Robinson's statutory claims, or that the trial court ever disposed of them specifically. The trial court and parties appear to have assumed, correctly, that the summary judgment was nevertheless final because Robinson's statutory claims are wholly derivative of John's common-law claims, and the adjudication of the latter effectively disposed of the former.53

But even though the summary judgment was final, an analysis of the retroactive effect of Chapter 149 on common-law claims and statutory claims presents different considerations. As we discuss more fully below, Crown argues that in determining whether the constitutional prohibition against retroactive laws applies in this case, it is significant that successor liability is a creature of statute. The same argument could be made to Robinson's wrongful death and survival claims, though not to the common-law claims the Robinsons previously asserted. Also, Robinson argues that it is important for our constitutional analysis that the common-law claims barred by Chapter 149 had both accrued and were the subject of a pending lawsuit before the statute was enacted. But neither is true of her statutory claims.

The parties have not briefed—or even mentioned—any of these issues but have confined their arguments regarding whether Chapter 149 is an unconstitutionality retroactive law as applied to the common-law claims the Robinsons asserted before John's death, which were adjudicated by summary judgment. These arguments are the only ones we address. We intimate no view on whether Chapter 149 limits Robinson's statutory wrongful death and survival claims except insofar as they are derivative of the claims specifically adjudicated by the trial court.

III

Before we can decide whether Chapter 149 is unconstitutionally retroactive, we must first resolve the parties' dispute over the proper standards to be applied in making that determination. Robinson, like the dissenting opinion in the court of appeals, argues that the test is simply whether vested rights have been impaired, period; if so, the law is prohibited, regardless of the Legislature's reasons for enacting it. Crown counters that the majority opinion in the court of appeals was correct in focusing instead on the reasonableness of the Legislature's exercise of its police power; the prohibition against retroactive laws does not invalidate a proper exercise of that power despite its impairment of private rights.54 As each position finds support in our case law, we begin by returning to first principles. We conclude that the history and purpose of the constitutional provision require a fuller statement of its proper application than we have previously given.

A

There exists in this country, as the United States Supreme Court observed in Landgraf v. USI Film Products, a "presumption against retroactive legislation [that] is deeply rooted in our jurisprudence[] and embodies a legal doctrine centuries older than our Republic. . . . [T]he `principle that the legal effect of conduct should ordinarily be assessed under the law that existed when the conduct took place has timeless and universal human appeal.'"55 In a concurring opinion in an earlier case, Justice Scalia noted that this principle

The United States Constitution does not expressly prohibit retroactive laws, but "the antiretroactivity principle finds expression" in its prohibitions of bills of attainder, ex post facto laws, and state laws impairing the obligation of contracts.57 The thrust of each is easily stated:

But the application of each prohibition must be measured by the object to be obtained. Thus, while the bill of attainder originated as an English parliamentary act sentencing to death someone who had attempted to overthrow the government,59

With respect to ex post facto laws:

And as for the prohibition against laws impairing contract obligations, Chief Justice Marshall observed:

Texas Constitutions have contained these provisions as well as a general prohibition against retroactive or retrospective laws.63 This prohibition against retroactive laws, like other constitutional bars, must be governed by its purpose, for "retroactive" simply means "[e]xtending in scope or effect to matters which have occurred in the past; retrospective",64 and "retrospective", even more simply, means "[d]irected to, contemplative of, past time".65 In our first case construing the retroactivity clause, DeCordova v. City of Galveston, Chief Justice Hemphill cautioned that applying this prohibition without regard to the objects to be achieved would have

The constitutional prohibition was not intended to operate so indiscriminately. "Mere retroactivity is not sufficient to invalidate a statute. . . . Most statutes operate to change existing conditions, and it is not every retroactive law that is unconstitutional."67

The presumption against retroactivity has two fundamental objectives identified by the Supreme Court in Landgraf. First, it protects the people's reasonable, settled expectations.

In other words, the rules should not change after the game has been played. Second, the presumption against retroactivity protects against abuses of legislative power.

As James Madison argued, "retroactive legislation also offer[s] special opportunities for the powerful to obtain special and improper legislative benefits."70

Still, not all retroactive legislation is bad. Landgraf also notes:

Constitutional provisions limiting retroactive legislation must therefore be applied to achieve their intended objectives—protecting settled expectations and preventing abuse of legislative power.

B

In DeCordova, Chief Justice Hemphill wrote that "[l]aws are deemed retrospective and within the constitutional prohibition, which by retrospective operation, destroy or impair, vested rights".72 For this formulation of the prohibition, he, like many judges since, cited Justice Story's statement in Society for the Propagation of the Gospel v. Wheeler, applying the New Hampshire constitution's prohibition against retroactive laws:

But as both cases explained, "impairs vested rights" has special meaning. "[A] statute merely regulating a remedy," Justice Story added, "and prescribing the mode and time of proceeding" does not impair vested rights.74 Chief Justice Hemphill agreed,

In other words, in applying the prohibition against retroactivity, a law that impairs a remedy does not impair a right, except sometimes. On further reflection, this Court conceded more than a century later: "Remedies are the life of rights. While our precedents recognize and apply the distinction [between a remedy and a right], they also recognize that the two terms are often inseparable."76

The obscurity in the right/remedy distinction typifies the problems in using "impairs vested rights" as a test for unconstitutional retroactivity, as our cases illustrate. In DeCordova, we held that a statute of limitations on suits for debt enacted after the defendant executed notes payable to the plaintiff but before they matured merely limited the plaintiff's collection remedy and therefore was not unconstitutionally retroactive.77 The idea that the debt had not been extinguished, only the means of collection, might be viewed by most creditors as a distinction without a difference. But the Court reasoned that the absence of a statute of limitations when the notes were executed did not give the plaintiff a vested right to sue forever. In Texas Water Rights Commission v. Wright, we upheld a statute authorizing forfeiture of a water permit after ten years of non-use, concluding that permit holders could reasonably expect enforcement of the "conditions inherently attached" to their permit, and that a permit included no right to be forever free of a remedy to enforce those conditions.78 Moreover, a retroactive use requirement was valid for the State "to assert and protect its own rights and interests in the water."79 In City of Tyler v. Likes, we held that a statute reclassifying a city's proprietary functions as governmental, thereby limiting liability, affected only a remedy, not a right, even though a claimant would recover less or perhaps not at all.80 And in In re A.D., we held that a statute removing the limitations period for enforcing child support decrees by ordering withholding of wages affected only a remedy, even though it expanded enforcement of the debt.81

In each of these cases, significant interests were adversely impacted by changes in the law, yet the Court held that vested rights were not impaired. The results of the cases seem entirely reasonable in a very general sense, although the claimants in the cases doubtless had a different view, but it is not clear how they were driven by a concern for protecting vested rights. In a recent case, Owens Corning v. Carter, we did not mention the right/remedy distinction in upholding a law that required application of the statute of limitations of the plaintiff's state of residence, even though doing so barred pending actions in Texas courts.82 We simply held that for a plaintiff who has not sued within the time permitted by the state in which he resides and in which the cause of action arose, barring suit in Texas "is not inequitable".83 Nevertheless, the plaintiff in a pending case had a viable claim that the change in the law extinguished.

In three of these five cases, DeCordova, Wright, and Likes, it was important that, as it happened, the people involved had ample opportunity after the change in the law to protect their interests: four years to sue in DeCordova,84 seven years to resume pumping water in Wright,85 and two months to sue in Likes.86 But in the other two cases, A.D. and Owens Corning, the persons affected by changes in the law had no time to respond. We have since held that a change in the law need not provide a grace period to prevent an impairment of vested rights.87

"Statutes of limitations are procedural",88 but sometimes a change may impair vested rights. In 1887, we stated in Mellinger v. City of Houston that when a law "shall operate in favor of a defendant as a defense against a claim made against him, then it must be said that a right exists, has become fixed or vested, and is beyond the reach of retroactive legislation".89 Thus, we said, a law extending a limitations period so as to resurrect barred claims would be unconstitutionally retroactive. But only two years earlier the United States Supreme Court had held in Campbell v. Holt that just such a law reviving claims did not offend due process under the United States Constitution because "no right is destroyed when the law restores a remedy which had been lost."90 Campbell arose out of Texas, and the Supreme Court cited this Court's 1870 decision in Bender v. Crawford,91 which held that a retroactive suspension of limitations statutes during the aftermath of the Civil War was not a prohibited retroactive law, even though claims that would have been barred were not.92 Mellinger cited Campbell, and "not wish[ing] to be understood as questioning its correctness", distinguished the due process guarantees in the state and federal constitutions from the prohibition of retroactive laws.93 But while due process and antiretroactivity may protect vested rights differently, Mellinger did not explain why a limitations bar is a vested right in one context but not in the other. In other words, a law that is prohibitively retroactive might not also offend due process, but not because a vested right for one is not a vested right for the other. Nor did Mellinger cite Bender.

A generation later, we held in Wilson v. Work that "it is the settled law that, after a cause has become barred by the statute of limitation, the defendant has a vested right to rely on such statute as a defense."94 We repeated that view more recently in Baker Hughes, Inc. v. Keco R. & D., Inc.95 The earlier confusion may be attributable to the time in which the issues arose. Bender offered this insight:

Bender dared to speak plainly: there are vested rights and then there are vested rights, and not all laws which may fairly be said to retroactively impair vested rights are constitutionally prohibited. The problem is not confined to the aftermath of the Civil War. Many years ago, one commentator lamented:

What constitutes an impairment of vested rights is too much in the eye of the beholder to serve as a test for unconstitutional retroactivity.

This can hardly be more vividly demonstrated than in today's opinions by JUSTICE WAINWRIGHT and JUSTICE MEDINA. The arguments and authorities ably marshaled in each show a deep division over whether a retroactive restriction on a cause of action impairs vested rights. Of course it does, if a claim, no matter how flimsy, is a vested right; or not, if a claim, even a strong one, must be reduced to judgment before it becomes a vested right. The dispute over whether to call something a vested right appears driven not so much by what the words mean as by the consequence of applying the label—that its impairment is prohibited. Or as one commentator has put it: "it has long been recognized that the term `vested right' is conclusory—a right is vested when it has been so far perfected that it cannot be taken away by statute."98 The "impairs vested rights" test thus comes down to this: a law is unconstitutionally retroactive if it takes away what should not be taken away.

C

In two cases this Court has held that retroactive laws were not constitutionally prohibited, despite their impairment of vested rights, because they were each a valid exercise of the Legislature's police power. The first, Barshop v. Medina County Underground Water Conservation District,99 involved a facial challenge to the Edwards Aquifer Act.100 Before the Act, withdrawal of groundwater from the Aquifer was unrestricted. The Act created an Authority to regulate groundwater withdrawals, capped annual withdrawals, required that wells be operated under permits, gave preference to existing users, and restricted withdrawals under a permit based on the owner's historic use.101 The Act operated retroactively in basing the right to groundwater on historic use and gave landowners no opportunity to preserve their prior right to unlimited water, but we stated that "article I, section 16 does not absolutely bar the Legislature from enacting such statutes."102 Acknowledging that "retroactive laws affecting vested rights that are legally recognized or secured are invalid",103 we nevertheless held that "[a] valid exercise of the police power by the Legislature to safeguard the public safety and welfare can prevail over a finding that a law is unconstitutionally retroactive."104 The Legislature had included in the Act findings that the Authority was necessary "`to protect terrestrial and aquatic life, domestic and municipal water supplies, the operation of existing industries, and the economic development of the state'"105 and that "the aquifer was `vital to the general economy and welfare of this state.'"106 "Based on these legislative findings," we concluded that the Act was "necessary to safeguard the public welfare of the citizens of this state" and therefore the Act's retroactive effect did not "render it unconstitutional" on its face.107

The second case, In re A.V.,108 involved section 161.001 of the Texas Family Code, which lists several grounds for terminating parental rights. An amendment had added subsection (1)(Q) to the list, thus providing for termination when a parent "has knowingly engaged in criminal conduct for which the parent is incarcerated and unable to care for the child `for not less than two years from the date of filing the petition'".109 The issue was whether the amendment was unconstitutionally retroactive as applied to a parent convicted before it was enacted. The amendment, we noted, was primarily prospective, focusing on "the parent's future imprisonment and inability to care for the child, not the criminal conduct that the parent committed in the past."110 But to the extent the amendment had a retroactive effect, we held it was not unconstitutional. Recognizing that "a parent's constitutionally-protected relationship with his or her children [is] a right that presumably cannot be altered through retroactive application of law",111 we stated, quoting Barshop, that a "`valid exercise of the police power by the Legislature to safeguard the public safety and welfare' is a recognized exception to the unconstitutionality of retroactive laws."112 Given the Legislature's declaration that "`[t]he public policy of this state [is to] provide a safe, stable, and nonviolent environment for the child'", we concluded that public policy justified the statute's retroactive effect.113 Furthermore, we said, "[a] law that does not upset a person's settled expectations in reasonable reliance upon the law is not unconstitutionally retroactive."114 In our view, a person "could not reasonably expect that the State would not act to provide a safe environment for his children while he was imprisoned."115

Robinson does not argue that Barshop and A.V. were wrongly decided but nevertheless insists that the test for unconstitutional retroactivity is not whether a law is a reasonable exercise of the Legislature's police power but whether it impairs vested rights. In her view, rights may be "vested for different purposes depending on the context",116 thereby affecting the constitutional provision's operation, and thus prohibiting retroactive laws limiting liability for asbestos claims but not laws preserving groundwater and protecting children. Stated differently: the right to sue is protected from retroactive impairment while the rights to groundwater or one's children are not. One might view this as backwards, that a parent's right to a child, which is "fundamental",117 "`one of constitutional dimensions'",118 and "`far more precious than any property right'",119 would be more deserving of protection from impairment by retroactive laws than a claim of injury that might not even result in recovery. But regardless of the three rights' relative importance, Robinson's argument that they are somehow vested differently for purposes of determining unconstitutional retroactivity establishes the fundamental failure of the "impairs vested rights" test.

We agree with Robinson, however, that Barshop and A.V. do not except a retroactive law from the constitutional prohibition merely because there was a rational basis for its enactment, or even because, on balance, it is likely to do more good than harm. The Legislature found the water-permitting scheme established in the Edwards Aquifer Act to be necessary to discharge its constitutional duty to conserve groundwater,120 and the necessity of providing for the welfare of children of incarcerated convicts is too obvious to require justification. But necessity alone cannot justify a retroactive law. The retroactive laws in Barshop and A.V. were not unconstitutional because they did not defeat the objectives of the constitutional prohibition. There can be no settled expectation that a limited resource like groundwater, affected by public and private interests, will not require allocation, or that a person unable to care for his children has greater rights if his inability is due to prolonged incarceration than for other reasons. And in both cases, the Legislature acted for the general public good.

D

We think our cases establish that the constitutional prohibition against retroactive laws does not insulate every vested right from impairment, nor does it give way to every reasonable exercise of the Legislature's police power; it protects settled expectations that rules are to govern the play and not simply the score, and prevents the abuses of legislative power that arise when individuals or groups are singled out for special reward or punishment. No bright-line test for unconstitutional retroactivity is possible. Rather, in determining whether a statute violates the prohibition against retroactive laws in article I, section 16 of the Texas Constitution, courts must consider three factors in light of the prohibition's dual objectives: the nature and strength of the public interest served by the statute as evidenced by the Legislature's factual findings; the nature of the prior right impaired by the statute; and the extent of the impairment.121 The perceived public advantage of a retroactive law is not simply to be balanced against its relatively small impact on private interests, or the prohibition would be deprived of most of its force. There must be a compelling public interest to overcome the heavy presumption against retroactive laws. To be sure, courts must be mindful that statutes are not to be set aside lightly. This Court has invalidated statutes as prohibitively retroactive in only three cases, all involving extensions of statutes of limitations.122 But courts must also be careful to enforce the constitutional prohibition to safeguard its objectives.

Under this test, changes in the law that merely affect remedies or procedure, or that otherwise have little impact on prior rights, are usually not unconstitutionally retroactive. But these consequences of the proper application of the prohibition cannot substitute for the test itself. The results in all of our cases applying the constitutional provision would be the same under this test. The cases that considered only whether the challenged statute impaired vested rights implicitly concluded that any impairment did not upend settled expectations and was overcome by the public interest served by the enactment of the statute. And the cases that focused on the propriety of the Legislature's exercise of its police power implicitly concluded that the exercise was not merely reasonable but was compelling, notwithstanding the statute's effect on prior rights.

The test the court of appeals distilled from the cases focuses too much on the reasonableness of legislative action and does not give full voice to the concerns addressed by the prohibition against retroactive laws. The court believed that one consideration in applying the prohibition is whether a statute is "appropriate and reasonably necessary to accomplish a purpose within the scope of the police power".123 But the necessity and appropriateness of legislation are generally not matters the judiciary is able to assess. In Barshop, for example, we did not undertake to determine whether the regulation scheme fashioned by the Legislature in the Edwards Aquifer Act was the only, the best, or even a good way to conserve and allocate groundwater among those claiming a right to it. The important considerations were that the Act discharged the Legislature's constitutionally mandated duty to conserve public resources, that some regulation was entirely to be expected, and that the burden of its retroactive effect in basing future withdrawals on historic use was shared by all those claiming a right to groundwater.

The second factor in the court of appeals' test was whether a statute is unreasonable, arbitrary, unjust, unduly harsh, or disproportionate to the end sought to be accomplished.124 But the intent of the prohibition against retroactive laws is to foreclose these kinds of considerations to the Legislature in enacting laws and to the judiciary in reviewing them. A retroactive law is not permissible merely because the end seems to justify the means. The presumption is that a retroactive law is unconstitutional without a compelling justification that does not greatly upset settled expectations.

Robinson would go further. She argues that because the prohibition against retroactive laws is part of the Texas Constitution's Bill of Rights, it is absolute, and any weighing of the government's interest in enacting a retroactive law is precluded by article I, section 29 of the Texas Constitution, which states:

But Robinson's argument begs the question. We do not disagree that the constitutional prohibition is absolute when it applies, as are the right to worship, the right to free speech, the freedom from unreasonable search and seizure, the guaranty of due course of law, and the other protections of the Bill of Rights. But section 29 does not determine whether and how the Bill of Rights' provisions apply. What Justice Oliver Wendell Holmes observed about all rights applies to the right to be free from retroactive laws:

IV

Using the standards we have set out, we must now determine whether chapter 149 is unconstitutionally retroactive as applied to Robinson.

A

We first consider the nature of the rights claimed by the Robinsons and Chapter 149's impact on them. Chapter 149 does not directly restrict the Robinsons' common law action for personal injuries due to exposure to asbestos in the workplace. Rather, it supplants the usual choice-of-law rules for determining what state's successor liability law should apply in asbestos cases in Texas by mandating Texas courts to apply Texas law, then for the first time prescribes limits on that liability, even if, as here, successor liability arose under the law of another state. Crown argues that by allowing for an expansion of liability beyond the tortfeasor to include a successor by merger, successor liability is largely remedial in nature, and in any event, is a creature of statute in which there can be neither right nor expectation. Crown cites Dickson v. Navarro County Levee Improvement District, where we gave immediate effect to a statute that repealed a special, statutory cause of action.126 Crown analogizes this case to Owens Corning, which upheld the change in Texas law to allow a plaintiff no more time to sue here than he would have had in his state of residence.127

But the successor liability in this case is not a creature of Texas law; the parties agree that without Chapter 149, New York or Pennsylvania law would apply, and that under the law of those states, Crown's successor liability is unquestionable. So this is not a case like Dickson, in which the Legislature abolished a cause of action it had itself created; Chapter 149 limits liability created under other states' laws. Nor is this a case like Owens Corning, in which the Legislature changed the statute of limitations so that a nonresident plaintiff would gain no advantage by suing in Texas rather than in his home state; Chapter 149 disadvantages Texas residents, as well as nonresidents, who sue Crown in Texas rather than New York or Pennsylvania. Nevertheless, Crown has a point that choice-of-law rules are purely procedural and subject to change, often by courts, but certainly by the Legislature if it chooses to do so.

However, Chapter 149 extinguishes the Robinsons' claim and all other such claims against Crown in Texas, and while it does so indirectly, extinction was the Legislature's specific intent. An interest in maintaining an established common-law cause of action is greater than an interest in choice-of-law rules. We have held that an unliquidated personal-injury claim was not a property interest under the common law,128 but it is now assignable as other property interests.129 The rights protected by the constitutional prohibition against retroactive laws are no more limited to those recognized at the time the prohibition was adopted than are the rights protected by due course of law. An unliquidated claim may have little or no value, as for example when the cause of action has not been recognized or the elements of recovery cannot be proved. But here, claims like the Robinsons' have become a mature tort, and recovery is more predictable, especially when the injury is mesothelioma, a uniquely asbestos-related disease. Discovery taken in the case shows that the Robinsons' claims had a substantial basis in fact. Their right to assert them was real and important, and it was firmly vested in the Robinsons.

Crown argues that when Mundet was still selling the asbestos insulation to which John Robinson was exposed in ship boiler rooms, the Robinsons could not reasonably have expected Mundet to be able to pay all the claims that would eventually arise, or that the company would merge with a deeper pocket like Crown. But those are not the expectations the prohibition against retroactive laws protects. The Robinsons could well have expected, then as now, that a rule of law that permitted their recovery, and many others' before them, would not be changed after they had filed suit to abrogate their claim.

Crown argues that the Robinsons have alleged that all the defendants they have sued are jointly and severally liable and that they are likely to recover all their damages from those who have already settled and the others than remain. We refuse to speculate about what might happen. If Crown would otherwise be responsible for the Robinsons' injury, then by insulating Crown, Chapter 149 either reduces the recovery to which the Robinsons are entitled or requires the other defendants to pay Crown's share. Either way, the statute disturbs settled expectations.

We therefore conclude that Chapter 149 significantly impacts a substantial interest the Robinsons have in a well-recognized common-law cause of action.

B

We next consider whether Chapter 149 serves the public interest. Crown argues that the statute helps alleviate the asbestos litigation crisis that has already bankrupted many companies, resulting in lost jobs and a burden on the State's economy. The Legislature has recognized the severity of that crisis in another context,130 but it did not do so in enacting House Bill 4 and Chapter 149. On the contrary, the legislative record is fairly clear that chapter 149 was enacted to help only Crown and no one else. Crown itself has been unable to identify to us any other company affected by Chapter 149. There is evidence that Crown has about 1,000 employees in Texas and about the same number of former employees on retirement, and that it operates three facilities here. Crown asserts that it continues to be sued on asbestos claims in Texas, but the record is silent concerning the number of those claims or the amount of Crown's probable exposure.

The Legislature made no findings to justify Chapter 149. Even the statement by its principal House sponsor fails to show how the legislation serves a substantial public interest. No doubt Texas will benefit from reducing the liability of an employer and investor in the State, but the extent of that benefit is unclear on this record. And in any event, there is nothing to indicate that it rises to the level of the public interest involved in Barshop and A.V.

Crown argues that the public interest has been recognized by other states' legislatures in enacting similar legislation. We are aware of ten other state legislatures that have enacted laws similar to chapter 149. In one state, Pennsylvania, the legislation was fully retroactive,131 just as chapter 149 is, but the Supreme Court of Pennsylvania has held the statute to violate the Open Courts provision of the Pennsylvania Constitution.132 Statutes adopted in three states—Florida, Indiana, and Wisconsin— apply to pending actions if trial has not commenced.133 Statutes adopted in three other states—North Dakota, Ohio, and Oklahoma—have the same application unless it is found to be unconstitutional.134 South Carolina's statute applies only to actions filed after the statute's effective date,135 and Georgia's applies only to actions that accrue after the statute's effective date.136 The effect of Mississippi's statute on accrued or pending claims is unclear from the text.137 Other states' perception of the public interest served by retroactive legislation is at best ambiguous.

It is tempting to think that the real burden of Chapter 149 on the Robinsons and other plaintiffs in their shoes will be light compared to the benefit to Crown, its current and former employees, and the State. The Robinsons' case, and most others like it, involves many defendants and large settlements funded from many pockets. The impact of Chapter 149 on individual cases may be slight, relative to the cumulative impact on Crown without Chapter 149. But we think that an important reason for the constitutional prohibition against retroactive laws is to preempt this weighing of interests absent compelling reasons. Indeed, it is precisely because retroactive rectification of perceived injustice seems so reasonable and even necessary, especially when there are few to complain, that the constitution prohibits it.

Accepting the legislative record as indicating the reasons for its actions, we conclude that the public interest served by Chapter 149 is slight.

* * *

For these reasons, we hold that Chapter 149, as applied to the Robinsons' common-law claims, violated article I, section 16 of the Texas Constitution. The court of appeals' judgment is reversed and the case is remanded to the trial court for further proceedings.

Justice MEDINA filed a concurring opinion.

Justice WILLETT filed a concurring opinion, in which Justice LEHRMANN joined.

Justice WAINWRIGHT filed a dissenting opinion, in which Justice JOHNSON joined.

Justice GUZMAN did not participate in the decision.

Justice MEDINA filed a concurring opinion.

I join the Court's opinion because I agree that a retroactive law is presumptively "unconstitutional without a compelling justification that does not greatly upset settled expectations" and that no such justification exists here. 335 S.W.3d 126, 147. I further agree that the "constitutional prohibition against retroactive laws does not insulate every vested right from impairment, nor does it give way to every reasonable exercise of the Legislature's police power[.]" Id. at 145. And finally, I agree that Chapter 149 here violates article I, section 16 of the Texas Constitution because it operates to retroactively abolish the Robinsons' vested property rights, or in the words of the Court—"significantly impacts a substantial interest the Robinsons have in a well-recognized common-law cause of action." 335 S.W.3d at 149. I write separately because I do not share the Court's disdain for traditional vested rights analysis nor the dissent's view of that analysis.

I

I begin, as the Court does, with the twin goals served by the Retroactivity Clause: (1) it protects individuals against legislative enactments that unfairly deprive them of legitimate expectations, Owens Corning v. Carter, 997 S.W.2d 560, 572 (Tex.1999), and (2) it ensures that legislative enactments do not single out individuals for preferential or arbitrary treatment. See In re A.V., 113 S.W.3d 355, 361 (Tex.2003) (upholding law against retroactivity challenge because "the State [was] not pursuing a retributive or punitive aim"); see also Landgraf v. USI Film Prods., 511 U.S. 244, 284-85 & n. 20, 114 S.Ct. 1483, 128 L.Ed.2d 229 (1994). As the United States Supreme Court has observed, retroactive lawmaking creates special opportunities for rewarding favored constituencies at the expense of disfavored ones. Landgraf, 511 U.S. at 266-67, 114 S.Ct. 1483.

When determining whether a statute violates the Retroactivity Clause, vested rights analysis poses three related questions. First, does the claimant have a vested right affected by the statute? Second, does the retroactive statute impair that vested right? And finally, does a compelling public interest justify impairment through the state's police power? See In re A.V., 113 S.W.3d at 361; Barshop v. Medina Cnty. Underground Water Conserv. Dist., 925 S.W.2d 618, 633-34 (Tex.1996).

Concluding that vested rights analysis was "difficult" and "inconsistent," the court of appeals declined to address whether the Robinsons had a vested right in their accrued tort claim or whether Chapter 149 intruded upon that right. 251 S.W.3d 520, 526. The court concluded instead that, regardless of whether Chapter 149 implicated vested rights, the law could be upheld as a reasonable exercise of the police power. Id. at 534 (citing Barshop, 925 S.W.2d at 633-34). Taking much the same tack, the Court begins with the last question but correctly rejects the court of appeals' rational basis analysis as the appropriate constitutional standard. Along the way, the Court grapples with the nature of the underlying property interest and its impairment, ultimately concluding that the Robinsons possessed a substantial interest in a well-founded claim (dare I say a vested property right) that Chapter 149 retroactively impaired. Although the Court is reluctant to use the term "vested rights," preferring instead to speak of "settled expectations," I believe we are talking about the same thing.

II

Whether a right may be regarded as vested depends on considerations of "fair notice," "reasonable reliance," and "settled expectations." Owens Corning, 997 S.W.2d at 572-73; see also McCain v. Yost, 155 Tex. 174, 284 S.W.2d 898, 900 (1955). Because the parties agree that Robinson's claim has accrued, see Pustejovsky v. Rapid-Am. Corp., 35 S.W.3d 643, 653 (Tex.2000), the first question is whether or not that accrued tort claim is a vested right. I conclude that it is.

While we have never invalidated a statute on the grounds that it retroactively abrogated an accrued cause of action, we have noted on several occasions that accrued causes of action enjoy constitutional protection under article I, section 16. To be sure, no one has a vested right in a "mere expectation ... based upon an anticipated continuance of the present general laws." Ex parte Abell, 613 S.W.2d 255, 261 (Tex.1981). At the same time, the Retroactivity Clause

Mellinger v. City of Houston, 68 Tex. 37, 3 S.W. 249, 253 (1887); see also Owens Corning, 997 S.W.2d at 572-73 (observing that "[c]onsiderations of fair notice, reasonable reliance, and settled expectations play a prominent role" when determining rights entitled to constitutional protection); Middleton v. Tex. Power & Light Co., 108 Tex. 96, 185 S.W. 556, 560 (1916) (observing that a vested common law right of action is a property right that the legislation at issue did not affect).

We have also held that the Legislature may affect remedies for accrued causes of action, so long as the remedy is not entirely taken away. See, e.g., City of Tyler v. Likes, 962 S.W.2d 489, 502-03 (Tex.1997); Phil H. Pierce Co. v. Watkins, 114 Tex. 153, 263 S.W. 905, 907 (1924) (orig. proceeding); see also DeCordova v. City of Galveston, 4 Tex. 470, 477-78 (1849) (explaining the remedial exception and endorsing a New Hampshire decision affording protection to accrued causes of action). If Texas law afforded no constitutional protection to accrued causes of action, there would be no need to permit the Legislature to modify attendant remedies, nor any need to bar the Legislature from stripping a plaintiff of all remedy.

Not all courts agree with this analysis, however. Several federal courts suggest that a plaintiff has no vested right in an accrued tort claim until the claim is pursued to final judgment.1 The majority of jurisdictions, however, appear to afford constitutional protection to accrued tort claims without a final-judgment requirement.

For example, the Kansas Supreme Court has rejected the notion that a property right in an accrued tort claim should not vest before final judgment. Resolution Trust Corp. v. Fleischer, 257 Kan. 360, 892 P.2d 497, 500-06 (1995). The court noted some disagreement about vested rights, particularly in the federal courts, but suggested that "the apparent conflicting holdings [were] not as divergent as they initially appear[ed]." Id. at 503. Instead, the court observed that the outcome in decisions espousing a final-judgment requirement could generally be explained by other factors, such as: "(1) the nature of the rights at stake (e.g., procedural, substantive, remedial), (2) how the rights were affected (e.g., were the rights partially or completely abolished by the legislation; was any substitute remedy provided), and (3) the nature and strength of the public interest furthered by the legislation." Id. The court then noted that most of the federal cases favoring a final-judgment requirement involved issues of federal preemption and the substitution of a federal remedy for the common law claim, while others involved either the clarification of defects in an existing ambiguous statute or retroactive legislation aimed at urgent problems of great public interest. See Id. (citing cases). In short, the cases generally turned on factors other than the existence of a final judgment.

The United States Supreme Court has hesitated to apply the due process clause in this area. See, e.g., Daniels v. Williams, 474 U.S. 327, 332, 106 S.Ct. 662, 88 L.Ed.2d 662 (1986) (rejecting reasoning that would superimpose the Fourteenth Amendment as a "font of tort law"); Paul v. Davis, 424 U.S. 693, 701, 96 S.Ct. 1155, 47 L.Ed.2d 405 (1976) (same); but see Logan v. Zimmerman Brush Co., 455 U.S. 422, 428, 102 S.Ct. 1148, 71 L.Ed.2d 265 (1982) (considering it "settled" that "a cause of action is a species of property"). Notions of comity and federalism may explain this hesitancy in part, particularly when state enactments are under review. David Richards & Chris Riley, Symposium on the Texas Constitution: a Coherent Due-Course-of-Law Doctrine, 68 TEX. L.REV. 1649, 1666 (1990). But these considerations do not similarly encumber state courts. Moreover, our state Constitution, unlike its federal counterpart, includes an independent anti-retroactivity provision, and Texas's Retroactivity Clause goes beyond federal guarantees of property and due process. Ex parte Abell, 613 S.W.2d at 260. As we explained in Mellinger:

3 S.W. at 252; see also Richards & Riley, 68 TEX. L.REV. at 1650 (noting that drafters of our 1836 Constitution held "a distinctly Jacksonian" concept of democracy and a "wariness of governmental authority"). And finally, every state whose constitution includes an independent anti-retroactivity provision concludes that accrued causes of action are vested rights.2

Though a plaintiff's ultimate right to recover is contingent upon success at trial, a plaintiff may nonetheless have a "settled expectation" that, once wronged, he or she will be able to pursue a claim against his wrongdoer under the substantive laws as they existed at the time his or her cause of action accrued. Mellinger, 3 S.W. at 253; see also Likes, 962 S.W.2d at 502 (indicating that retrospective law that entirely eliminates pending cause of action would be "unconstitutionally retroactive"). A plaintiff thus has a property interest in an accrued cause of action which the Texas Constitution protects like other property. Subaru of Am. v. David McDavid Nissan, Inc., 84 S.W.3d 212, 219 (Tex.2002).

III

The dissent, however, prefers the minority view that a property interest in an accrued cause of action not vest before final judgment is rendered. According to the dissent, no reasonable litigant has a "`settled expectation' of achieving monetary recovery" before judgment. 335 S.W.3d 126, 174 (Wainwright, J. dissenting). And as to this case, the dissent submits that the "Robinsons' expectation in the continued state of law [was] low" because of "numerous contingencies" and their failure to take "any action in reliance on the law at the time." 335 S.W.3d 126, 183, 179 (Wainwright, J. dissenting). In the dissent's view, the Robinsons' pending tort claims were nothing more than a valueless contingent interest in an uncertain future judgment. I disagree for several reasons.

First, I reject the notion that an accrued cause of action has no value apart from a judgment and is not itself a protected property interest. An accrued cause of action is clearly property under Texas law. See TEX. PROP.CODE § 12.014. It has value, even if that value is not always easy to measure. Ownership and control is vested in the holder of the claim, and those interests can generally be sold or assigned. Int'l Proteins Corp. v. Ralston-Purina Co., 744 S.W.2d 932, 934 (Tex.1988); see PPG Indus., Inc. v. JMB/Houston Ctrs. Partners Ltd. P'ship, 146 S.W.3d 79, 106 (Tex.2004) (tracing development of modern law allowing transfer of choses in action); State Farm Fire & Cas. Co. v. Gandy, 925 S.W.2d 696, 707 (Tex.1996) (noting that "[p]racticalities of the modern world have made free alienation of choses in action the general rule").

Further, an accrued cause of action is "constitutional" property, a vested property right, because the holder has a legitimate expectation that the claim will be recognized by state law. See Jack M. Beermann, Government Official Torts and the Takings Clause: Federalism and State Sovereign Immunity, 68 B.U.L.REV. 277, 302 (1988); see also Logan, 455 U.S. at 428, 102 S.Ct. 1148 (considering it "settled" that "a cause of action is a species of property"). The dissent's preoccupation with final judgments is misguided because the relevant expectations here involve the cause of action, not some future judgment. It is the right to sue itself—the lawsuit— that is being taken away, not the final outcome.

Once a lawsuit is filed, subsequent action by the sovereign interferes not with possible or potential rights that might accrue in the future, but with existing expectations and rights that have accrued—that have "vested"—and that constitute a present property interest. As one legal scholar explains: "a cause of action might be thought of as an entitlement to employ the state's adjudicatory machinery which can only be denied for cause, cause being the failure to establish the elements of the cause of action or to comply with reasonable procedural requirements." Beermann, 68 B.U.L.REV. at 305 n.121 (citing Logan, 455 U.S. 422, 102 S.Ct. 1148). Other authorities share the same view:

Jeremy A. Blumenthal, Legal Claims as Private Property: Implications for Eminent Domain, 36 HASTINGS CONST. L.Q. 373, 401 (Spring 2009) (footnotes omitted); see also Resolution Trust Corp., 892 P.2d at 502 (holding accrued tort claim to be vested property right and rejecting similar final-judgment argument because it "fails to recognize the distinction between a right of action and a right of recovery").

Finally, even if some manner of affirmative act is, as the dissent suggests, a necessary part of the "settled expectations" test, it is clearly established here. Contrary to the dissent's characterization, the Robinsons were not idle while Crown Cork labored to undo their accrued claim. The Robinsons filed suit, litigated their claim for several months, and obtained a partial summary judgment. Only after that did the Legislature enact Chapter 149, taking away the Robinsons' summary judgment and their underlying cause of action. Although the Robinsons had not yet obtained a final judgment (and thus had no vested property right in that non-existent judgment), they did possess a vested property right in their pending cause of action that included the right to prosecute the claim. See Washington-Southern Navigation Co. v. Baltimore & Philadelphia S.B. Co., 263 U.S. 629, 635, 44 S.Ct. 220, 68 L.Ed. 480 (1924) ("The right of a citizen of the United States to sue in a court having jurisdiction of the parties and of the cause of action includes the right to prosecute his claim to judgment.").

Although I emphatically disagree with the dissent's view that an accrued cause of action is too indefinite, and its owner's expectations too insignificant, to warrant constitutional protection, I readily concede that "no one has a vested right ... or a property right, in a mere rule of law." 335 S.W.3d at 171 (Wainwright, J. dissenting) (quoting Middleton, 185 S.W. at 560). The continuation of a rule of law in the abstract, however, is very different from the preservation of a claim that has already accrued under that law. Although a "person has no property, no vested interest, in any rule of the common law," nevertheless, "[r]ights of property which have been created by the common law cannot be taken away without due process." Munn v. Illinois, 94 U.S. 113, 134, 24 L.Ed. 77 (1877).

The distinction is the difference between the prospective or retroactive application of a law. Prospective laws that diminish or eliminate future causes of action (or defenses) do not ordinarily implicate vested property rights. However, where "a law changes the legal consequences of past actions, it interferes with vested rights, and courts have found that property ... is implicated." Olivia A. Radin, Rights as Property, 104 COLUM. L.REV. 1315, 1331 (2004). A cause of action vests upon the occurrence of an injury, and that "vested right of action is property in the same sense in which tangible things are property, and is equally protected against arbitrary interference." Pritchard v. Norton, 106 U.S. 124, 132, 1 S.Ct. 102, 27 L.Ed. 104 (1882).

Our opinion in Ex parte Abell, which the dissent quotes at length, is to the same effect. 335 S.W.3d at 171 (Wainwright, J. dissenting) (quoting Ex parte Abell, 613 S.W.2d at 261). Abell restates the rule from Mellinger that the Legislature can declare prospectively that the state of facts that once created a claim, no longer will; but once "the state of facts which the law declares shall give a right comes into existence," the right is "fixed" or "vested," and the Legislature cannot retroactively undo what has already accrued. Abell, 613 S.W.2d at 261 (quoting Mellinger, 3 S.W. at 253).

The dissent accepts the rule, but only to a point. It acknowledges that a defense of repose vests upon accrual and cannot thereafter be rescinded by the Legislature:

335 S.W.3d at 177 (Wainwright, J. dissenting). But the dissent refuses to extend similar protection to an accrued claim. I see no basis for limiting the Retroactivity Clause to defensive claims. See 1 GEORGE D. BRADEN, ET AL., THE CONSTITUTION OF THE STATE OF TEXAS: AN ANNOTATED AND COMPARATIVE ANALYSIS 58 (1977) (observing that the "prohibition concerning `retroactive laws' seems to spring from a general suspicion regarding all retroactive laws").

Rights acquired under existing law, whether defensive or offensive, are treated similarly under the Texas Constitution. Thus, once a statute of limitations has run or a cause of action has accrued, retroactive legislation that either revives an extinguished claim, or bars an existing one, affects a vested property right. The Retroactivity Clause applies in either instance. See Mellinger, 3 S.W. at 253 (observing that Retroactivity Clause applies both to vested claims and defenses). I would therefore conclude that the Robinsons' accrued claims are vested rights to which the protection of article I, section 16 may apply.

IV

Crown Cork argues, however, that even if a plaintiff has a vested right in an accrued tort claim, Chapter 149 does not intrude upon the Robinsons' vested rights for several reasons. First, Crown Cork argues that Chapter 149 does not infringe on this right because the statute merely works a change to procedure or remedy. It is well established that a party has no vested right in a remedy or rule of procedure. Subaru, 84 S.W.3d at 219. The lines that divide substance from procedure or remedy, however, are notoriously difficult to draw. Tex. Water Rights Comm'n v. Wright, 464 S.W.2d 642, 648-49 (Tex. 1971). But we need not parse these "superfine" distinctions here, Langever v. Miller, 124 Tex. 80, 76 S.W.2d 1025, 1029 (1934), because, whether regarded as remedial or substantive, Chapter 149 entirely eliminates the Robinsons' remedy for Mundet's torts. See Likes, 962 S.W.2d at 502-03; Phil H. Pierce Co., 263 S.W. at 907; DeCordova, 4 Tex. at 479-80. Though it is conceivable that Robinson may establish that one of the other defendants in the original action is jointly and severally liable for all of her damages, Robinson has a separate cause of action against each of these defendants that might properly be tried and determined as if it were the only claim in controversy. See Morgan v. Compugraphic Corp., 675 S.W.2d 729, 733-34 (Tex.1984). Chapter 149 extinguishes this separate action in its entirety, releasing Crown Cork from its obligation assumed under the merger agreement to pay for Mundet's torts. See Sam Bassett Lumber Co. v. City of Houston, 145 Tex. 492, 198 S.W.2d 879, 882 (1947) (distinguishing permissible changes to statutes of limitations from enactments releasing or extinguishing a debt).

Crown Cork also argues that Chapter 149 does not intrude on the Robinsons' vested rights because it affects vicarious liability. In Aetna Ins. Co. v. Richardelle, 528 S.W.2d 280 (Tex.Civ.App.-Corpus Christi 1975, writ ref'd n.r.e.), the court of appeals upheld a law retroactively repealing the statutory vicarious liability of some parents for the torts of their minor children. Though Aetna described vicarious liability as "remedial," it did not hold that the Legislature might alter vicarious liabilities at will without running afoul of the Retroactivity or Contract Clauses. Aetna merely held that statutory causes of action do not generally give rise to vested rights. Id. at 285 (citing Dickson v. Navarro Levee Imp. Dist., 135 Tex. 95, 139 S.W.2d 257, 259 (1940)). Unlike the parents' vicarious liability for the torts of their children, successor liability is not purely statutory. Successor liability, at least when it attends a formal merger, instead arises from the contractual merger agreement and from the plaintiff's underlying tort or contract claim, and was recognized at common law. See Tex. & P.R. Co. v. Murphy, 46 Tex. 356, 360 (1876); Stephenson v. Tex. & P.R. Co., 42 Tex. 162, 167-68 (1874); Turner v. Bituminous Cas. Co., 397 Mich. 406, 244 N.W.2d 873, 877-78 (1976) ("Most of [the rules of successor liability] may fairly be said to have arisen from case law"). Though successor liability is now governed by statute, the corporations law is a shield to common law liability, not a legislatively created right within the meaning of Dickson.

Crown Cork next argues that Chapter 149 does not intrude on the Robinsons' vested rights because it is akin to a borrowing statute or choice of law rule mandating that Texas, rather than Pennsylvania or New York, law apply to determine corporate successor liability in asbestos cases. In Owens Corning, we held that a plaintiff had no vested right in a borrowing statute that permitted out-of-state plaintiffs to file stale out-of-state claims in Texas courts under Texas' more permissive statute of limitations. 997 S.W.2d at 571-73. But Chapter 149 represents not only a choice of law rule but also a change to Texas' substantive law of successor liability. Cf. id. at 573. Before the passage of Chapter 149, no matter which state's law applied, Crown Cork would have faced liability for Mundet's torts. TEX. BUS. ORG. CODE § 10.008(a)(3)-(4); TEX BUS. CORP. ACT § 5.06(3); N.Y. BUS. CORP. LAW § 906; 15 PA. CONS.STAT. § 1929. Indeed, Pennsylvania has already invalidated a statute providing protections virtually identical to those found in Chapter 149. See Ieropoli v. AC&S Corp., 577 Pa. 138, 842 A.2d 919, 921 (2004).

Finally, Crown Cork cites Owens Corning to argue that Chapter 149 does not interfere with vested rights because Robinson had no legitimate expectation that Mundet would merge with a much larger corporation and because it is not inequitable to relieve Crown Cork of wholly unexpected and innocently acquired asbestos liabilities. 997 S.W.2d at 572-73. In approving the borrowing statute at issue in Owens Corning, we noted that it was not inequitable to require a plaintiff bringing an out-of-state claim to satisfy the statute of limitations provided by the law supplying the cause of action: "a plaintiff should not be able to gain greater rights than he would have in the state where the cause of action arose and where he lives simply by bringing suit in Texas." Id. at 573. But Chapter 149 goes much further and "creates an immunity where none existed before." Id. To be sure, Crown Cork probably did not expect the merger with Mundet to entail such extensive liability, and Robinson could hardly have a settled expectation that Mundet would be acquired by a much larger corporation. But it is not inequitable to require Crown Cork to pay for Mundet's torts because when two corporations formally merge, the law regards them as one. Though this rule may permit plaintiffs to recover where they otherwise would not, "[i]n substance, if not in form, the post-transfer entity distributed the defective products and should be held responsible for them." RESTATEMENT (THIRD) OF TORTS: PRODUCTS LIABILITY § 12 & cmt. b.

Thus, I conclude that the Robinsons' accrued tort claim here is a vested right that Chapter 149 has retroactively abrogated. But this is not the end of the analysis. As the Court observes, "the constitutional prohibition against retroactive laws does not insulate every vested right from impairment" and a compelling public interest may justify impairment, although "the heavy presumption against retroactive laws" makes instances of this quite rare. 335 S.W.3d at 146.

In fact, we have only twice recognized legislative interests of sufficient import to override vested private rights: Barshop, 925 S.W.2d 618, and In re A.V., 113 S.W.3d 355. At issue in Barshop was a statute regulating water use in the Edwards Aquifer basin. Before the enactment of the law, property owners were permitted, under the rule of capture, to extract as much water as they desired from the aquifer. Concerned that the rule of capture discouraged water conservation, the Legislature authorized local water districts to regulate water use through a permitting scheme that allocated use permits on the basis of historical use. Without conducting a vested rights analysis, we held that the Legislature's interest in water conservation trumped whatever interest landowners had in the continued existence of the rule of capture because "[c]onservation of water has always been a paramount concern in Texas, especially in times, like today, of devastating drought." Barshop, 925 S.W.2d at 626.

In In re A.V., 113 S.W.3d 355, we held that the Legislature could permissibly enact a statute terminating parental rights on the basis of a parent's future imprisonment for prior criminal convictions. Though we found that the provision did not intrude upon vested rights, we explained that, even if it had, the Legislature's interest in protecting "the safety and welfare of its children" trumped any individual interest in retaining parental rights while unable to care for the child. Id. at 361.

In contrast to the public interest at issue in those cases, the interest protected here is essentially a private economic one. As the Court observes, "the legislative record is fairly clear that chapter 149 was enacted to help only Crown and no one else." 335 S.W.3d at 149. The Legislature certainly has a valid interest in protecting Crown Cork's shareholders and pensioners, and in promoting business in the state. The Legislature also has a legitimate interest in protecting defendants from excessive liability. But the Legislature's interest in protecting the financial well-being of a favored defendant is not on par with the public interest in the avoidance of catastrophic drought or the protection of child welfare. Cf. Barshop, 925 S.W.2d at 626; In re A.V., 113 S.W.3d at 361. Private economic interests will generally not justify intrusions into the vested private rights of others. See, e.g., Travelers Ins. Co. v. Marshall, 124 Tex. 45, 76 S.W.2d 1007, 1011-12 (1934) (holding that the interests of homeowners during the Great Depression did not justify interference with private mortgage contract rights); Lucas v. United States, 757 S.W.2d 687, 701 (Tex. 1988) (holding that it was "simply unfair and unreasonable to impose the burden of supporting the medical care industry solely upon those persons who are most severely injured and therefore most in need of compensation."). Moreover, the Legislature's interest in protecting "innocent" defendants does not justify its assumption here of the judiciary's role of adjusting private obligations incurred under existing law. See Landgraf, 511 U.S. at 267 n. 20, 114 S.Ct. 1483 (quoting Richmond v. J.A. Croson Co., 488 U.S. 469, 513-14, 109 S.Ct. 706, 102 L.Ed.2d 854 (1989) (Stevens, J., concurring in part and concurring in the judgment)). As with the takings clause, one of the purposes of the prohibition on retroactive lawmaking is to ensure that the burdens of governance do not fall unfairly on a small number of citizens. See id. (citing THE FEDERALIST No. 44 (James Madison)). The legislative interest asserted here is simply insufficient to justify an intrusion into the Robinsons' vested rights, and thus I agree with the Court that the innocent successor provisions of Chapter 149 cannot be justified as a valid exercise of the police power and is, as applied here, prohibited under the Retroactivity Clause. TEX. CONST. art. I, § 16.

Justice WILLETT, joined by Justice LEHRMANN, concurring.

Litigants in our adversarial system are hard-wired for certitude, adept at insisting the law "clearly" or "plainly" favors their side or, as here, labeling the controlling analysis "straightforward and simple." If only. Today's case is both complex and consequential, and fiendishly so. The facts are compelling; the law is unclear; and the stakes are high, not just for these parties but also for our constitutional architecture that both confers and constrains governmental power. I concur that chapter 149 is an invalid exercise of legislative police power that cannot surmount our Constitution's ban on retroactive laws. But I write separately to stress that this case, at heart, implicates issues far beyond whether Barbara Robinson can sue Crown Cork & Seal.

Every case that reaches this Court concerns real people buffeted by real problems in the real world. This dispute, however, possesses a transcendent quality, touching not only these parties but also building-block constitutional principles that belong to all Texans. In that sense, it affords a whetstone on which to sharpen our thinking on some bedrock notions of government and how the Texas Constitution assigns democratic responsibilities. More to the point, it teaches a vital lesson about diminished liberty stemming from government overreaching: The Legislature's police power cannot go unpoliced.

The Texas Constitution looks unkindly on retroactive laws, but as a constitutional matter, retroactive is not always retrograde.

While it is axiomatic that the Legislature, through budgeting and lawmaking, has primacy in setting State policy, that power, though unrivaled, is not unlimited. One constraint is the Texas Constitution's Bill of Rights, including article I, section 16's prohibition against retroactive laws.1

Retroactive legislation is disfavored because, as the Father of the U.S. Constitution explains, citizens deserve protection from the "fluctuating policy" of the legislature.2 Robinson's position takes James Madison one leap further: Disfavored actually means disallowed, and "the police power may not be used to deprive citizens of their property retroactively by eliminating their vested rights in accrued claims." Robinson insists our Bill of Rights, including the Retroactivity Clause, is impregnable in this regard given this mandate from article I, section 29:

This admonition naturally commands judicial respect, but it cannot bear the weight Robinson places on it. We long ago crossed the Rubicon of declaring the Retroactivity Clause non-absolute (despite article I, section 29's seeming absolutism), recognizing that some retroactive laws "may be proper or necessary, as the case may be."4 Specifically—and this is one facet of retroactive-law analysis where a controlling principle (if not its application) is uncomplicated—such laws are constitutionally permissible if they are a "valid exercise of the police power by the Legislature to safeguard the public safety and welfare."5 Retroactivity in and of itself is not fatal,6 and nothing in the Bill of Rights handcuffs the Legislature from confronting urgent state priorities.

The notion that article I, section 29 hermetically seals off the Bill of Rights from all legislative attention invites myriad absurdities.7 The "all laws contrary ... shall be void" language may be facially inviolable, similar to the federal Bill of Rights' "Congress shall make no law" language, but it must accommodate legislation at the edges. The practical challenge for judges is to set that perimeter, and to do so in a principled, no-favorites fashion.

House Bill 4 was enacted against a backdrop of urgency, but with legislative police power, unfettered must never be unfretted.

As litigants often discover, in the Legislature a deal is sometimes a raw deal. But unfair does not always equal unconstitutional; even vested rights can be impinged if lawmakers have a good-enough reason.

Both the U.S. Supreme Court and this Court have lamented the "`elephantine mass of asbestos cases' lodged in state and federal courts,"8 branding it a "crisis"9 that "defies customary judicial administration."10 In response, the bipartisan civil-justice reforms enacted in 2003's House Bill 4 effected a sea change in the Texas tort landscape;11 likewise the omnibus asbestos-litigation reforms enacted in 2005's Senate Bill 15.12 Both measures sought to address perceived flaws in asbestos-related litigation—HB 4 by limiting so-called "innocent successor" liability (immediately and retroactively),13 and SB 15 via more sweeping reforms for asbestos and silica claims.14

In upholding a retroactive water regulation in Barshop, we expressly relied on formal and extensive findings that the Legislature made part of the statutory text itself: "Based on these legislative findings, we conclude that the Act is necessary to safeguard the public welfare of the citizens of this state. Accordingly, the retroactive effect of the statute does not render it unconstitutional."15 Chapter 149's enacted text includes no such findings. Instead, Crown Cork relies on the legislative record, contending it amply underscores an urgent public need: protecting imperiled-but-nonculpable companies in order to safeguard the livelihoods of endangered-but-innocent employees, pensioners, and local economies.16

Nobody disputes "the authority of the Legislature to make reasoned adjustments in the legal system."17 But lawmakers aiming to statutorily prescribe what is constitutionally proscribed must make a convincing case. As the Court carefully explains, the sparse record underlying chapter 149 falls short of what must be shown before someone is made to surrender a constitutional right.

This case concerns high-stakes issues far beyond chapter 149, principally how the Texas Constitution allocates governing power.

Today's case is not merely about whether chapter 149 singled out Barbara Robinson and unconstitutionally snuffed out her pending action against a lone corporation. Distilled down, it is also a case about how Texans govern themselves.

Delimiting the outer edge of police-power constitutionality has bedeviled Texas courts for over a century. The broader issue of a citizen's relationship with the State has confounded for centuries longer.

Gauzy definitions such as these—and laments over such imprecision—offer scant comfort in this enterprise. The issue is elemental, but not elementary. Fortunately, we are not entirely without guidance.

Appropriately weighty principles guide our course. First, we recognize that police power draws from the credo that "the needs of the many outweigh the needs of the few." Second, while this maxim rings utilitarian and Dickensian (not to mention Vulcan21), it is cabined by something contrarian and Texan: distrust of intrusive government and a belief that police power is justified only by urgency, not expediency. That is, there must exist a societal peril that makes collective action imperative: "The police power is founded in public necessity, and only public necessity can justify its exercise."22 Third, whether the surrender of constitutional guarantees is necessary is a legislative call in terms of desirability but a judicial one in terms of constitutionality. The political branches decide if laws pass; courts decide if laws pass muster. The Capitol is the center of policymaking gravity, but the Constitution exerts the strongest pull, and police power must bow to constitutional commands: "as broad as [police power] may be, and as comprehensive as some legislation has sought to make it, still it is subsidiary and subordinate to the Constitution."23 Fourth, because the Constitution claims our highest allegiance, a police-power action that burdens a guarantee like the Retroactivity Clause must make a convincing case.24 Finally, while police power naturally operates to abridge private rights, our Constitution, being inclined to freedom, requires that such encroachments be as slight as possible: "Private rights are never to be sacrificed to a greater extent than necessary."25

If judicial review means anything, it is that judicial restraint does not allow everything. Yes, courts must respect democratically enacted decisions; popular sovereignty matters. But the Texas Constitution's insistence on limited government also matters, and that vision of enumerated powers and personal liberty becomes quaint once courts (perhaps owing to an off-kilter grasp of "judicial activism") decide the Legislature has limitless power to declare its actions justified by police power. At that constitutional tipping point, adjudication more resembles abdication.

Whatever the police power's amorphous boundaries, we know these two things: (1) the Legislature may ask for private sacrifice, and receive it—provided the private rights sacrificed are outweighed in public good, burdened as little as possible, and amply justified on public-necessity grounds; and (2) the Legislature's police power is not infinitely elastic, able to extinguish constitutional liberties with nonchalance. Texans long ago and since have embraced constitutional, meaning limited, government. The judiciary thus has a superseding obligation to disapprove certain encroachments on liberty, no matter the legislative vote-count. Put another way, judicial review sometimes means thwarting today's majority from thwarting yesterday's supermajority—the one that ratified our solemn Constitution.26

Legislative police power is not constitutional carte blanche to regulate all spheres of everyday life; preeminence does not equal omnipotence.

The Texas Bill of Rights—enshrined to recognize and establish "the general, great and essential principles of liberty and free government"27—declares an emphatic "no" to myriad government undertakings: no religious test for office, no double jeopardy, no self-incrimination, no curtailment of free speech, etc. It is, like its federal counterpart, irrefutably framed in proscription. And, like its federal counterpart, its limitations are not exception-free; desperate times permit desperate measures (to a point). But we should steadfastly resist defining desperation down. Exceptions to constitutional guarantees are real but also rare, just like modern citations to Marbury v. Madison: "The powers of the legislature are defined, and limited; and that those limits may not be mistaken, or forgotten, the constitution is written."28

The "danger that liberty should be undervalued" necessarily implicates "the adjustment of the boundaries between it and social control."29 There must remain judicially enforceable constraints on legislative actions that are irreconcilable with constitutional commands. If legislators come to believe that police power is an ever-present constitutional trump card they can play whenever it suits them, overreaching is inexorable.

To be sure, constitutional analysis is nuanced and not prone to doctrinaire absolutes. It is easy to say the sovereign's shield must never become a sledgehammer, but it is more difficult—and every bit as important—to discern the moment at which it threatens to become a switchblade, carving quietly yet critically away at cherished rights.

It merits repeating that this Court and the U.S. Supreme Court have long permitted legislative bodies to burden constitutional freedoms upon a strong public-welfare showing. The reason chapter 149 offends the Retroactivity Clause is because it lacks that showing. Indeed, if chapter 149's meager record were sufficient, there would be scant defense against future police-power incursions—incursions that, while ostensibly well-meaning, shrink the sphere of protected liberty and erode bit by bit the notion of limited government. "Experience is the oracle of truth,"30 wrote Madison, and history teaches this is a ratchet that clicks only one way.

Robinson's case provides a real and important reminder of the limits of legislative power and the scope of judicial review. But after her case has come and gone, I hope what Edmund Burke called a "fierce spirit of liberty"31 will help steer a course with senses heightened to constitutional guardrails.

Police power is an attribute of sovereignty, but sovereignty ultimately rests in "the people of the State of Texas."32

The Texas Constitution places limits on government encroachments, and does so on purpose. Our Bill of Rights is not mere hortatory fluff; it is a purposeful check on government power. Everyday Texans, and the courts that serve them, must remain vigilant, lest we permit boundless police power, often couched in soaring prose, to abridge our Constitution's enduring "principles of liberty and free government."33 As Justice Brandeis warned in his now-celebrated Olmstead dissent: "Experience should teach us to be most on guard to protect liberty when the Government's purposes are beneficent."34

Shortly after the Federal Constitution was approved in September 1787, Thomas Jefferson wrote James Madison from Paris, advocating a Bill of Rights and also voicing confidence that the people would be the best sentries against overreaching government: "[I am] convinced that on their good senses we may rely with the most security for the preservation of a due degree of liberty."35 Jefferson was right. We are our own best lookouts against invasions, however well-intentioned, that siphon our "due degree of liberty"—siphoning that often occurs subtly, with such drop-by-drop gentleness as to be imperceptible.

To be sure, Members of the Texas Legislature have sworn to "preserve, protect, and defend the Constitution and laws of the United States and of this State,"36 and they doubtless believe their enactments honor basic constitutional guarantees. I never second-guess the Legislature's motives and goodwill (and have never needed to); we are blessed with 181 lawmakers who serve Texas with full hearts.37 But where the Constitution is concerned, the judiciary's role as referee—confined yet consequential—must leaven big-heartedness with tough-mindedness.

* * *

Summing up: Judges are properly deferential to legislative judgments in most matters, but at some epochal point, when police power becomes a convenient talisman waved to short-circuit our constitutional design, deference devolves into dereliction. The Legislature's policymaking power may be vast, but absent a convincing public-welfare showing, its police power cannot be allowed to uproot liberties enshrined in our Constitution.

Justice WAINWRIGHT, joined by Justice JOHNSON, dissenting.

The Legislature enacted Chapter 149 of the Civil Practice and Remedies Code to protect businesses, which acquired other entities, from financial disaster based solely upon the acquired entities' past, discontinued manufacture of asbestos products. The statute limits the liability of the acquiring business, which had not engaged in the asbestos business, to the fair market value of the acquired entity at the time of the acquisition. Through Chapter 149, the Legislature balances limitations on asbestos-related recoveries against protecting the assets and employees of businesses who did not cause the illness, while leaving intact the entirety of potential liability and damages proven against companies that were involved in the asbestos business and are, perhaps, more culpable. The Court's holding that the legislation is unconstitutional prevents the Legislature from addressing an injustice arising from a crisis that caused dozens of bankruptcies and the loss of thousands of jobs in this state and throughout the country due to asbestos-related litigation. See, e.g., Jonathan Orszag, The Impact of Asbestos Liabilities on Workers in Bankrupt Firms, Remarks at the Asbestos Litigation Symposium at the South Texas College of Law in Houston, Tex. (Mar. 7, 2003), in 44 S. TEX. L.REV. 1077, 1078-80 (2003) (describing results of a study indicating that sixty-one companies entered into bankruptcy and that 52,000 to 60,000 people lost their jobs due to asbestos litigation).

The Court's new balancing test reaches the wrong result. By holding that an unliquidated claim with "substantial basis in fact" is entitled to constitutional protection, it ignores an important principle. 335 S.W.3d 126. The constitutional retroactivity doctrine does not protect an asserted entitlement to property one does not own, and until a final judgment in a case, we do not know whether the claim will be vindicated or refuted. The Court's reasoning that the right to file a claim is protected by the retroactivity doctrine because, at least in part, the claim is well founded with a "substantial basis in fact" springing from a "mature tort" with "more predictable" recovery, is a troubling proposition. 335 S.W.3d 126. It is unclear what that means, but it suggests that the constitutional retroactivity protection is dependent on the perceived strength of a claim. The likelihood of success in litigation is dependent on a myriad of factors that make such predictions difficult at best. We have held that an unliquidated personal injury claim is not a protected property interest, and the contingent recovery from one should not be either.

While JUSTICE MEDINA, who writes separately, and I disagree on the result, we agree that the Court should not abandon vested rights jurisprudence in favor of a new and uncertain approach. The analysis in the Court's opinion is contrary to both the clear rule among the federal courts of appeals that have addressed the issue and the majority rule among our courts of appeals. The Court could rely on traditional police power jurisprudence in which, even if the Robinsons had a vested right in their unliquidated cause of action, courts consider whether the Legislature's action was justified by its constitutionally recognized police power to act in the interest of the health and welfare of Texas. Indeed, the Court's new balancing test for retroactivity analysis is similar to the police power balancing test I expound under existing law, but is newly incorporated into the retroactivity doctrine. For all these reasons, I respectfully dissent.

I. BACKGROUND

John Robinson served in the Navy for twenty years, and during that time he was exposed to steam pipes and boiler doors coated with insulation containing asbestos. Some of the insulation and other products were marked with a "big M," the trademark used by Mundet Cork Corporation. In August 2002, Robinson was diagnosed with mesothelioma. He claims the disease occurred as a result of his exposure to asbestos in, among others, insulation products produced by Mundet.

Crown Cork itself has never been in the business of mining, manufacturing, installing, selling, distributing, removing, or otherwise making asbestos or any asbestos-containing product. However, on November 7, 1963, Crown Cork's predecessor entered into an agreement to purchase the majority of Mundet's stock after the majority shareholder died and offered the shares for sale. Crown Cork paid approximately $7 million for the stock, a majority interest in the company.

Mundet ceased manufacturing insulation products prior to Crown Cork's acquisition of Mundet, but continued to hold insulation products in stock until early 1964, when a third-party entity purchased the assets of Mundet's insulation division, including its inventory, contracts, raw materials, and accounts receivables. On January 4, 1966, Mundet statutorily merged with Crown Cork's predecessor, and in 1989 Crown Cork was reincorporated in Pennsylvania.1

After he had been diagnosed with mesothelioma, Mr. Robinson and his wife filed suit in 2002 against Crown Cork and twenty other defendants for damages caused by Mr. Robinson's exposure to asbestos-containing products. The Robinsons sought to hold each defendant jointly and severally liable. On November 25, 2002, the Robinsons filed a motion for partial summary judgment to establish Crown Cork's liability for actual damages as Mundet's successor. Crown Cork did not contest its successor liability for compensatory damages, and on July 16, 2003 the trial court granted the Robinsons' motion, holding that Crown Cork "is liable and bears responsibility for the compensatory damages, if any, awarded to Plaintiffs that are attributable to the conduct, products, or torts of its predecessor Mundet Cork Corporation."

House Bill 4, a bill drafted to comprehensively address perceived crises in medical malpractice, asbestos, and other litigation issues in Texas, was introduced in the Texas House of Representatives on February 17, 2003, without any provision regarding successor asbestos liability. Tex. H.B. 4, 78th Leg., R.S. (2003). Its purpose was to operate as a "comprehensive civil justice reform bill intended to address and correct problems that currently impair the fairness and efficiency of our court system." House Comm. on Civil Practices, Bill Analysis, Tex. H.B. 4, 78th Leg., R.S. at 1 (2003).

In late March 2003, more than 100 amendments were submitted to the Bill, including Article 17, the asbestos successor-liability article. The article was debated on the floor of the House on March 25, 2003 and passed the House three days later. Both the House and Senate held hearings on the bill as a whole. In an April 30, 2003 meeting of the Senate State Affairs Committee, Senator Ratliff, the committee chair, introduced hearings on the Senate Substitute to House Bill 4. He described Article 17 as follows:

Hearings on the Proposed Senate Substitute for H.B. 4 Before the S. Comm. on State Affairs, 78th Leg., R.S. (Apr. 30, 2003) (Statement of Sen. Bill Ratliff, Chairman, S. Comm. on State Affairs). The act passed the Senate on May 16, 2003; the House accepted the Conference Committee compromise bill on June 1, 2003; both adopted corrections on June 2, 2003; and the bill was signed into law by the Governor on June 11, 2003. Act of June 2, 2003, 78th Leg., R. S., ch. 204, 2003 Tex. Gen. Laws 847, 899 (codified at TEX. CIV. PRAC. & REM.CODE §§ 149.001-.006). With a two-thirds vote in both chambers, the bill took effect immediately and was made retroactive to all cases "pending on that effective date and in which the trial, or any new trial or retrial following motion, appeal, or otherwise, begins on or after that effective date."2 Id. § 17.02(2), 2003 Tex. Gen. Laws at 895; see also TEX. CONST. art. III, § 39 ("No law passed by the Legislature, except the general appropriation act, shall take effect or go into force until ninety days after the adjournment of the session at which it was enacted, unless the Legislature shall, by a vote of two-thirds of all the members elected to each House, otherwise direct; said vote to be taken by yeas and nays, and entered upon the journals.").

The act limits the "cumulative successor asbestos-related liabilities" "incurred by a corporation as a result of or in connection with a merger or consolidation ... with or into another corporation or that are related in any way to asbestos claims based on the exercise of control or the ownership of stock of the corporation before the merger or consolidation that occurred" prior to May 13, 1968. TEX. CIV. PRAC. & REM.CODE §§ 149.001-.003.3 The asbestos liabilities of successor corporations "are limited to the fair market value of the total gross assets of the transferor determined as of the time of the merger or consolidation," id. § 149.003(a), and adjusted for inflation at a simple interest rate of the prime rate plus one percent, id. § 149.005(a). An "asbestos claim" is "any claim, wherever or whenever made, for damages, losses, indemnification, contribution, or other relief arising out of, based on, or in any way related to asbestos, including" property damage caused by asbestos, the health effects of asbestos exposure, or any claim made by or on behalf of any person exposed to asbestos. Id. § 149.001(1). The Legislature clearly intended to limit recoveries only against so-called "innocent" successor companies.

According to Crown Cork's experts, by May 2003, Crown Cork had paid or agreed to pay asbestos related claims, not covered by insurance, totaling more than seven times the present value of Mundet according to the statutory formula. On July 3, 2003, Crown Cork filed a Motion for Summary Judgment raising the affirmative defense of Chapter 149, introducing evidence of the value of Mundet and total asbestos-related payments made by Crown Cork to date. The Robinsons asserted that the statute was a "special law" in violation of article III, section 56 of the Texas Constitution, that it deprived the Robinsons of a vested property right in violation of article I, section 16 of the Texas Constitution, that the statute was an unconstitutional taking, violating article I, section 17 of the Texas Constitution and the Fifth and Fourteenth Amendments to the United States Constitution, that it constituted a deprivation of substantive due process rights under the Texas and United States Constitutions, that it deprived John Robinson of a contractual right, contrary to article I, section 16 of the Texas Constitution, and deprived John Robinson of his common law causes of action in violation of the Open Courts guarantee in article I, section 13 of the Texas Constitution. The Robinsons raise only the retroactivity and special law challenges before this Court. Implicitly finding that Crown Cork had established that the statute applied to it as a matter of law, and that Crown Cork had already paid liabilities in excess of Mundet's adjusted value, the trial court granted Crown Cork's motion for summary judgment on October 2, 2003. It issued an amended order nineteen days later, dismissing claims against Crown Cork brought by the Robinsons.4 The Robinsons nonsuited their remaining claims against Crown Cork and then appealed the summary judgment.5 The court of appeals affirmed. Characterizing the jurisprudence on vested rights as "inconsistent and difficult to use as a guide," the court instead balanced the Legislature's police power against the private rights impacted by the statute, and held that the statute was constitutional. 251 S.W.3d 520, 532-35 (Tex.App.-Houston [14th Dist.] 2006, pet. granted). One justice dissented, arguing that the court should have applied a vested rights analysis and concluded that the statute violated article I, section 16. Id. at 551-52 (Frost, J., dissenting).

II. ANALYSIS

In this Court, the Robinsons raise only two issues, and both are grounded exclusively in Texas law. They argue that Chapter 149 of the Texas Civil Practice and Remedies Code is an unconstitutional "special law" and that it is unconstitutionally retroactive when applied to the Robinsons' claims to effectively bar recovery.6 As the party challenging the constitutionality of the statute, the Robinsons must overcome the presumptions that "the Legislature intended for the law to comply with the United States and Texas Constitutions, to achieve a just and reasonable result, and to advance a public rather than a private interest." Tex. Mun. League Intergovernmental Risk Pool v. Tex. Workers' Comp. Comm'n, 74 S.W.3d 377, 381 (Tex.2002) (citing TEX. GOV'T CODE § 311.021; Spence v. Fenchler, 107 Tex. 443, 180 S.W. 597, 605 (1915)). The Robinsons also bear the burden of showing that the law is contrary to a provision of the state constitution. See, e.g., Walker v. Gutierrez, 111 S.W.3d 56, 66 (Tex.2003). The Robinsons' retroactivity claim is an as-applied challenge, which means that they must demonstrate that the statute is unconstitutional as it operates in practice against them. Tex. Mun. League, 74 S.W.3d at 381 (citing Tex. Workers' Comp. Comm'n v. Garcia, 893 S.W.2d 504, 518 n. 16 (Tex.1995)). Their special law challenge is a facial challenge, which means that the Robinsons must demonstrate there is no conceivable set of facts that could exist under which the statute would be constitutional. Garcia, 893 S.W.2d at 520.

In this case the Court determines that the law is unconstitutionally retroactive and thus does not reach the special law challenge. However, for the reasons that follow, I would hold that the law survives both challenges, but for reasons different from those articulated by the court of appeals.

A. Retroactive Law

Article I, section 16 of the Texas Constitution, part of the Texas Bill of Rights, declares that "[n]o bill of attainder, ex post facto law, retroactive law, or any law impairing the obligation of contracts, shall be made." TEX. CONST. art. I, § 16. A retroactive law "takes away or impairs vested rights acquired under existing laws...." Paschal v. Perez, 7 Tex. 348, 365 (1851). A retroactive law means a law applying to things that are past. DeCordova v. City of Galveston, 4 Tex. 470, 475 (1849).

Of course, not every law that affects relationships among parties based upon events occurring in the past is automatically unconstitutional, just as not every law that may affect a person's right to speak, that may affect a contractual obligation, or that may allow a search of a person's dwelling without a warrant, is unconstitutional. See Subaru of Am. v. David McDavid Nissan, Inc., 84 S.W.3d 212, 219 (Tex.2002). This Court has articulated three doctrines that further define the scope of the retroactivity prohibition. First, a law is not unconstitutionally retroactive unless it impairs a person's "vested rights." E.g., id. at 219. Second, a law is not unconstitutionally retroactive if it only modifies or reduces the person's remedy. E.g., City of Tyler v. Likes, 962 S.W.2d 489, 502 (Tex.1997); Holder v. Wood, 714 S.W.2d 318, 319 (Tex.1986). And finally, even if the law affects a person's vested rights, and not a remedy, a law may not violate the retroactivity prohibition if the government's interest in protecting society, based upon its police power, outweighs the individual's interest in his or her particular right. E.g., Barshop v. Medina Cnty. Underground Water Conservation Dist., 925 S.W.2d 618, 633-34 (Tex.1996). The first two tests are definitional—this Court has determined that a retroactive law does not implicate article I, section 16 of the Constitution unless the law both affects a vested right and impairs an actual right, not merely a remedy or a procedure. The third test may operate as an exception to the rule. Although related, the review of each doctrine is separate. E.g., In re A.V. & J.V., 113 S.W.3d 355, 361 (Tex. 2003) (describing "exceptions" to retroactivity); David McDavid Nissan, 84 S.W.3d at 219 (analyzing the procedural/remedial test as part of the vested rights exception because "procedural and remedial statutes typically do not affect a vested right"). Although the Court has not had occasion recently to address the specific meaning of article I, section 16's prohibition of retroactive laws, our precedents provide a useful roadmap.

1. Vested Rights

Vested rights derive from "[c]onsiderations of fair notice, reasonable reliance, and settled expectations." Owens Corning v. Carter, 997 S.W.2d 560, 572-73 (Tex. 1999). "A retroactive statute only violates our Constitution if, when applied, it takes away or impairs vested rights acquired under existing law." David McDavid Nissan, 84 S.W.3d at 219 (citing Ex parte Abell, 613 S.W.2d 255, 260 (Tex.1981)); McCain v. Yost, 155 Tex. 174, 284 S.W.2d 898, 900 (1955).

We explained "vested rights" in Ex parte Abell:

613 S.W.2d at 261 (quoting Mellinger v. City of Houston, 68 Tex. 37, 3 S.W. 249, 253 (1887)) (emphasis added). "A right cannot be considered a vested right unless it is something more than a mere expectation as may be based upon an anticipated continuance of the present general laws; it must have become a title, legal or equitable...." Id. (citation omitted) (emphasis added). This Court has clearly articulated that "no one has a vested right in the continuance of present laws in relation to a particular subject.... There cannot be a vested right, or a property right, in a mere rule of law." Middleton v. Tex. Power & Light Co., 108 Tex. 96, 185 S.W. 556, 560 (1916).

The court of appeals called the vested rights analysis "inconsistent and difficult to use as a guide." 251 S.W.3d at 526. Other courts of appeals have called the vested rights analysis "amorphous." Sims v. Adoption Alliance, 922 S.W.2d 213, 216 (Tex.App.-San Antonio 1996, writ denied); Ex parte Kubas, 83 S.W.3d 366, 369 (Tex. App.-Corpus Christi 2002, pet. ref'd). Courts from other states and commentators have also criticized vested rights analyses, preferring an analysis requiring a balancing of the nature and strength of the public interest served by the statute, the extent to which the statute modifies or abrogates the pre-enactment right, and the nature of the right the statute alters. See, e.g., Owen Lumber Co. v. Chartrand, 276 Kan. 218, 73 P.3d 753, 755-56 (2003); Peterson v. City of Minneapolis, 285 Minn. 282, 173 N.W.2d 353, 356-57 (1969); see also Charles B. Hochman, The Supreme Court and the Constitutionality of Retroactive Legislation, 73 HARV. L.REV. 692, 697 (1960). And the Court's opinion, in rejecting a "bright-line test for unconstitutional activity," and in recognizing that the Texas Constitution "does not insulate every vested right from impairment," seems to abandon the vested rights analysis altogether, or, at a minimum, detaches the concept of vested rights from its traditional significance in a retroactivity analysis. 335 S.W.3d 126. However, the doctrine's difficulty is not a justification to abandon it wholesale. For, at the core of the vested rights doctrine lies an extremely important principle—the constitutional retroactivity doctrine does not protect an asserted entitlement to property one does not own, and until a final judgment in a case, we do not know whether the lawsuit will prove or refute a claim to recover.

Applying our century-old jurisprudence, I would hold that an accrued, but unliquidated cause of action is not a vested right because: (1) the framers of the Texas Constitution would not have considered an unliquidated cause of action to be a vested property right entitled to protection under the Retroactivity Clause; (2) a lawsuit is not a right to recover anything but a contingent and unliquidated pursuit of a claimed injury that may or may not be successful; and (3) until and unless a final judgment is rendered in favor of the claimant, there is no right to recover damages on the claim against another. See Mellinger, 3 S.W. at 252; Graham v. Franco, 488 S.W.2d 390, 393 (Tex.1972); Ex parte Abell, 613 S.W.2d at 260.

In interpreting the Texas Constitution, our duty is "to ascertain and give effect to the plain intent and language of the framers of [the constitution] and of the people who adopted it." Wilson v. Galveston Cnty. Cent. Appraisal Dist., 713 S.W.2d 98, 101 (Tex.1986) (quoting Gragg v. Cayuga Indep. Sch. Dist., 539 S.W.2d 861, 866 (Tex.1976)). We look

Davenport v. Garcia, 834 S.W.2d 4, 30 (Tex.1992) (Hecht, J., concurring) (citations omitted).

Examining the state of "vested rights" and what constitutes a vested property right at the time of the framing of the 1876 Constitution provides important insight into what the Framers considered protected by the Retroactivity Clause. Prior to and at the time of the adoption of the Texas Constitution in 1876, it was well established that the doctrine of vested rights created an exception to the prohibition on retroactive legislation. See, e.g., DeCordova, 4 Tex. at 475; Paschal, 7 Tex. at 365 ("Mr. Justice Story defines a retrospective law to be, one which takes away or impairs vested rights acquired under existing law, or creates a new obligation, or imposes a new duty, or attaches a new disability in relation to transactions already past." (citing Soc'y for the Propagation of the Gospel v. Wheeler, 2 Gall. 105, 138, 22 F. Cas. 756, 767 (No. 13,156) (C.C.D.N.H.1814))).7 A vested right is now, and was then, considered some form of "property right." Middleton, 185 S.W. at 560. However, at the time of the framing of the Constitution of 1876, an accrued, but unliquidated, cause of action for personal injury, was not "property" in any sense. See G.H. & S.A. R.R. v. Freeman, 57 Tex. 156 (1882); Stewart v. H. & T.C. Ry. Co., 62 Tex. 246 (1884). Common law tort causes of action for personal injury could not be assigned and did not survive the death of the victim. As described by Chief Justice Greenhill:

Graham, 488 S.W.2d at 393 (emphasis added) (quotation omitted); see also State Farm Fire & Cas. Co. v. Gandy, 925 S.W.2d 696, 706-07 (Tex.1996) (discussing the role at common law regarding the assignability and survivability of personal injury tort causes of action). Legislation was required to amend both of those common law rules. E.g., Act of May 4, 1895, 24th Leg., R. S., ch. 89, § 1, 1895 Tex. Gen. Laws 143 (current version at TEX. CIV. PRAC. & REM.CODE § 71.021) (allowing survival of personal injury claims); TEX. PROP. CODE § 12.014(a) (allowing "an interest in a cause of action on which suit has been filed" to be "sold, regardless of whether the ... cause of action is assignable in law or equity"); Gandy, 925 S.W.2d at 707 (noting that personal injury claims only became assignable after they could survive the owner's death).

As in other circumstances, property is treated differently. In 1876, choses in action for injury to property were considered property, and they were alienable, assignable, and devisable.

Graham, 488 S.W.2d at 393 (quoting Freeman, 57 Tex. at 158); see also Gandy, 925 S.W.2d at 706 (noting that "[t]he pressures against the rule of inalienability were commercial and thus affected only debts and other contract rights that were not personal to the owner and could survive to his estate upon his death"). The common law in Texas did not consider tort causes of action for personal injury to be "property," and "vested rights"—a concept recognized in common law at the time of the framing of the 1876 Constitution—are a species of property. Therefore, under the Texas Constitution, ratified in 1876, an accrued, but unliquidated personal injury cause of action was not considered to be a "vested right" for purposes of the Retroactivity Clause. Gandy, 925 S.W.2d at 706. This reasoning applies with special force to the Robinsons' as-applied challenge, because at common law Mr. Robinson's claims would not have survived his death. His claims exist today only by virtue of statutes. The framers of the Texas Constitution would have not believed that there would be a settled expectation in allowing Mrs. Robinson to continue to prosecute these uncertain claims, either as Mr. Robinsons's personal representative or derivatively through a statutorily created wrongful death action.

The Court recognizes this historical disconnect, yet dismisses it in a single sentence, stating simply that "[t]he rights protected by the constitutional prohibition against retroactive laws are no more limited to those recognized at the time the prohibition was adopted than are the rights protected by due course of law." 335 S.W.3d 126. A court should be cautious in providing new protections for rights that were not part of the sphere of rights contemplated by the democratic institutions that enacted the constitution. See McDonald v. City of Chicago, ___ U.S. ___, 130 S.Ct. 3020, 3051-53, 177 L.Ed.2d 894 (2010) (Scalia, J., concurring) (criticizing the dissent's conceptual framework to "`do justice to [the Due Process Clause's] urgent call and its open texture' by exercising the `interpretive discretion the latter embodies'" and to hold that the Clause encompasses "new freedoms the Framers were too narrow-minded to imagine" (quoting Id., ___ U.S. ___, 130 S.Ct. at 3099-100 (Stevens, J., dissenting))).

The right to file a cause of action is not an entitlement to enforce the alleged claim, but a "mere expectation" subject to numerous contingencies. Ex parte Abell, 613 S.W.2d at 261-62; Mellinger, 3 S.W. at 252-53. A plaintiff's ultimate recovery is contingent upon more than just success at trial. For example, it is contingent upon finding—and serving with process— the right defendant, who may be an inaccessible foreign defendant, or, as in this case, may be a corporation long since out of business. See, e.g., TEX.R. CIV. P. 103-109a (discussing methods of service); GFTA Trendanalysen B.G.A. Herrdum GMBH & Co., K.G. v. Varme, 991 S.W.2d 785, 785 (Tex.1999) (per curiam) (holding special appearance by foreign corporation did not waive challenge to jurisdiction). A plaintiff's recovery may be contingent upon following particular pretrial procedures, such as the filing of an expert report or providing discovery. See TEX. CIV. PRAC. & REM.CODE § 74.351 (requiring the service of an expert report by the plaintiff in a health care liability claim and demanding dismissal of the claim if the report is not timely served); Cire v. Cummings, 134 S.W.3d 835, 841-42 (Tex.2004) (holding that "death penalty" sanctions of dismissing plaintiff's claim was warranted because of plaintiff's failure to produce audiotapes that would have proved or disproved plaintiff's legal malpractice claims). Any informed client knows that winning a lawsuit, even a seemingly "open and shut" case, is never certain, particularly when multiple defendants and multiple products may have caused the same injury, and no reasonable person has a "settled expectation" of achieving monetary recovery once she discovers a harm inflicted upon her.

Rather, I would hold, consistent with the jurisprudence of the United States Supreme Court8 a majority of the federal courts of appeals,9 a number of other states,10 and a majority of the courts of appeals to address the issue in this state,11 that a cause of action becomes a "vested right" for the constitutional retroactivity analysis when it has reached a final determination—that is, where it has been reduced to an enforceable judgment in the plaintiff's favor.12 As aptly put in an opinion of the Court of Appeals for the First District:

Houston Indep. Sch. Dist. v. Houston Chronicle Publ'g Co., 798 S.W.2d 580, 589 (Tex.App.-Houston [1st Dist.] 1990, writ denied).

This rule is most consistent with the understanding of vested property rights at the time of the ratification of the 1876 Constitution. It is consistent with our subsequent interpretation of the words of the Retroactivity Clause.13 It is consistent with our case law and the great weight of court of appeals opinions. And it is more predictable and avoids confusion and ambiguity when the Legislature attempts to constitutionally craft a law affecting past conduct.

This Court's first significant discussion of retroactivity occurs in Mellinger v. City of Houston, 68 Tex. 37, 3 S.W. 249 (1887). The City of Houston sued to recover taxes on property that would otherwise have been barred by a subsequently repealed statute of limitations. The Court ruled that the statute was not to be applied retroactively and thus did not specifically decide whether Mellinger had a vested right that would be violated by retroactive application of the law. Id. at 251-52. It then stated that "an action barred by the statute of limitations was forever barred" and explained that a law may be unconstitutionally retroactive "if a statute of limitations applied to existing causes barred all remedy, or did not afford a reasonable period for their prosecution; or if an attempt were made by law, either by implication or expressly, to revive causes of action already barred...." Id. at 253-55. Mellinger did not hold that an unaccrued cause of action was a vested right subject to protection, but it did indicate that a shortening of the statute of limitations would require a grace period to allow those who had not filed their cause of action to do so before the new limitations period would come into effect. Id.

Subsequent cases from this Court recognize that the Legislature cannot resurrect causes of action that have already been extinguished by retroactively lengthening the statute of limitations. E.g., Baker Hughes, Inc. v. Keco R. & D., Inc., 12 S.W.3d 1, 4 & n. 12 (Tex.1999); Wilson v. Work, 122 Tex. 545, 62 S.W.2d 490, 490-91 (1933) (per curiam). This rule makes sense because "[t]o permit barred claims to be revived years later would undermine society's interest in repose, which is one of the principal justifications for statutes of limitations." Baker Hughes, 12 S.W.3d at 4. In other words, when the statute extinguished a cause of action, a defendant received a vested right of repose barring the extinguished claim.

In City of Tyler v. Likes, 962 S.W.2d 489 (Tex.1997), a case upon which the Robinsons principally rely, this Court held that a modification of the Tort Claims Act to provide the city with sovereign immunity from the plaintiff's common law tort claims was not constitutionally retroactive. It recognized that the statute "affect[ed] a remedy" for the plaintiff, which usually does not implicate the Retroactivity Clause unless the "remedy is entirely taken away." Id. at 502 (citation omitted). We noted that "[t]he Legislature can affect a remedy by providing a shorter limitations period for an accrued cause of action without violating the retroactivity provision of the Constitution if it affords a reasonable time or fair opportunity to preserve a claimant's rights under the former law, or if the amendment does not bar all remedy." Id. (citing Tex. Water Rights Comm'n v. Wright, 464 S.W.2d 642, 649 (Tex.1971); Mellinger, 3 S.W. at 254-55). Because the statute became effective seventeen months after her action accrued, the Court held that the plaintiff had a reasonable time to preserve her rights, and thus the statute was not unconstitutional as applied. Id. Likes emphasizes (as discussed further below) that where the legislation affects the plaintiff's remedy without entirely taking it away, the legislation is not unconstitutionally retroactive. Id.

Finally, this Court has specifically held that the Mellinger retroactivity exception, requiring that a party receive reasonable time to preserve its rights, which was relied on in Likes, has an exception itself. In Owens Corning v. Carter, 997 S.W.2d 560 (Tex.1999), the Court upheld a retroactive application of an amended borrowing statute against a constitutional challenge. At the time the lawsuit underlying the case was filed, Texas's borrowing statute provided that a non-Texan who was injured in a foreign state could bring an action in Texas, even if the limitations period in the plaintiff's home state had run, so long as the action was begun within the time provided by Texas law. Id. at 565; cf. Igal v. Brightstar Info. Tech. Grp., Inc., 250 S.W.3d 78, 90-91 (Tex.2008) (holding that res judicata bars relitigation of administratively determined facts and distinguishing a rule where "a claimant whose action is precluded by limitations in one state court may still be able to pursue the same action in a different state with a longer limitations period"). In early 1997, while the plaintiffs' lawsuits were pending, the Legislature amended the statute to require, among other things, that the action is begun in Texas within the time provided both by Texas law and the law of the foreign state in which the wrongful act, neglect, or default took place. Carter, 997 S.W.2d. at 572 (citing TEX. CIV. PRAC. & REM.CODE § 71.031(a)(3)). The plaintiffs challenged the law as unconstitutionally retroactive, and we rejected that challenge. First, we recognized that the plaintiffs did not have any settled expectations in the continuance of the current law—the limitations period. Second, we noted that "requiring a grace period for otherwise time-barred claims would defeat the very purpose of the borrowing statute: a plaintiff should not be able to gain greater rights than he would have in the state where the cause of action arose and where he lives simply by bringing suit in Texas." Id. at 573. In other words, even if the statute of limitations "grace period" rule articulated in Mellinger were to apply, because Alabama plaintiffs applying Alabama law had no expectation in the continuation of the borrowing statute, "such concerns play a minimal role and do not justify the application of a grace period." Id. (citing In re TMI, 89 F.3d 1106, 1116 (3d Cir.1996)).

This Court has recognized that contingencies, future expectations, and mere rules of law do not constitute vested rights. We have upheld retroactivity challenges only when it interferes with a final judgment, involved the vested parent-child relationship, or when the statute attempts to revive a cause of action previously barred by the statute of limitations. E.g., Milam County, 54 Tex. at 168; In re A.V., 113 S.W.3d 355, 361 (Tex.2003); Baker Hughes, 12 S.W.3d at 5. Otherwise, we have held on many occasions that laws, even those that explicitly apply retroactively, do not violate the Retroactivity Clause in article 1, section 16. See, e.g., David McDavid Nissan, 84 S.W.3d at 219-20; Carter, 997 S.W.2d at 573; Likes, 962 S.W.2d at 502; Barshop, 925 S.W.2d at 634; Ex parte Abell, 613 S.W.2d at 262; Exxon Corp. v. Brecheen, 526 S.W.2d 519, 525 (Tex.1975); McCain, 284 S.W.2d at 900; City of Dallas v. Trammell, 129 Tex. 150, 101 S.W.2d 1009, 1012-13 (1937).

The Robinsons' expectation that they could recover damages against Crown Cork as one of the numerous defendants in their lawsuit was low at the time Mr. Robinson's common law causes action accrued. Numerous contingencies surrounded their litigation, not the least of which were the identity of the potential tortfeasors and proving causation against Mundet from among nine other defendants.14 If they knew that Mundet was one of the parties responsible for producing asbestos that Mr. Robinson was exposed to, it is unlikely that they knew that Mundet had been bought by Crown Cork decades prior. This is not a situation where the obligations of two parties are identified by contract, where the government seeks to interfere with the parent-child relationship, or a party seeks to resurrect a claim long extinguished by a statute of limitations. Our case law is consistently hesitant to void statutes outside those categories as retroactive, and this is not an area into which our jurisprudence should expand.15

Finally, a "brighter-line" view provides more certainty and predictability and avoids confusion and ambiguity. Causes of action accrue when claimants are on notice of their injury and have the opportunity to seek a judicial remedy, when the injury occurs, or at the death of a promisor. Quigley v. Bennett, 227 S.W.3d 51, 58 (Tex.2007); Provident Life & Accident Ins. Co. v. Knott, 128 S.W.3d 211, 221 (Tex.2003). Certainly, these accruals almost always occur prior to the filing of a lawsuit (otherwise the claim would not be ripe). Therefore, accepting the Court's position that a right to file a lawsuit is a vested right would, in effect, preclude the Legislature from taking any action to modify or restrict a cause of action for some lawsuits that had not even been filed yet. It would further lead to unnecessary uncertainty and confusion.

The Robinsons did not have a vested right in their accrued causes of action when Mr. Robinson was diagnosed with mesothelioma. At most, they had contingent belief that they might be able to recover against Crown Cork or the other defendants. At the time Mr. Robinson's cause of action accrued, the Robinsons had not taken any action in reliance on the law at the time, and they had no entitlement to the law as it existed. Even after they filed their action and received a partial summary judgment that Crown Cork was liable as a successor corporation, they had an unliquidated interest in a personal injury tort claim that was not recognized as a property right—vested or otherwise—at common law. The expectation further deteriorated when Mr. Robinson passed away, and Mrs. Robinson asserted new statutory survival and wrongful death claims. I would hold that, when the Legislature limited recovery for asbestos claims only against innocent successor corporations that had caused no injury to claimants, the Legislature did not deprive the Robinsons of a vested right of action against Crown Cork, and thus Chapter 149 is not unconstitutionally retroactive as applied to the Robinsons. The Robinsons are not foreclosed, however, from going forward with their claims against other entities, consistent with the Act's limitations on recovery.

2. Police Power Balancing

The Robinsons argue that there is no room for a balancing of interests in the retroactivity analysis. They contend that if a right is vested, it cannot be affected by retroactive legislation.16 Regardless of whether the "vested rights" threshold exists, a balancing of interests and expectations is an integral part of retroactivity analysis in Texas jurisprudence, the jurisprudence of other states, and commentators and scholars in this area. Although the Court also balances interests, much in the same way I believe our jurisprudence demands that we balance interests pursuant to the state's police power, the Court's analysis overlooks a few critical points.

Courts carefully recognized that a retroactive law affecting vested rights may nonetheless be constitutional if the overriding public purpose of the act and the Legislature's legitimate exercise of its police power outweigh the interests or expectations of the affected party. E.g., Barshop, 925 S.W.2d at 633-34. As Justice Oliver Wendell Holmes, Jr. recognized in the context of a takings suit based on a statute retroactively preventing a mining company exercising its contractual rights to mine coal under a house:

Pa. Coal Co. v. Mahon, 260 U.S. 393, 413, 43 S.Ct. 158, 67 L.Ed. 322 (1922); see also In re Marriage of Bouquet, 16 Cal.3d 583, 128 Cal.Rptr. 427, 546 P.2d 1371, 1376 (1976) (noting that vested rights may be impaired when "reasonably necessary to the protection of the health, safety, morals, and general well being of the people"); Phillips v. Curiale, 128 N.J. 608, 608 A.2d 895, 902 (1992); Hochman, 73 HARV. L.REV. at 697 (advocating the abrogation of the "vested rights" concept and instead analyzing U.S. Supreme Court jurisprudence on retroactivity balancing the nature of the public interest served, the extent to which the statute modifies the asserted pre-enactment right, and the nature of the right which the statute alters).

In considering the balancing test to be applied this case, the court of appeals balanced the proper exercise of the police power (weighing presumably not only the validity of the exercise, but the importance as well) against the "detrimental impact on plaintiffs such as the Robinsons," noting that the statute was narrowly tailored to protect the most innocent corporations but still "leaving the pool of potential [asbestos] defendants as large as possible...." 251 S.W.3d at 532-33. The Court, on the other hand, balances: (1) the nature and strength of the public interest served by the statute as evidenced by the Legislature's factual findings; (2) the nature of the prior right impaired by the statute; and (3) the extent of the impairment. 335 S.W.3d 126. Using this test, the Court determines that Chapter 149 is unconstitutionally retroactive as applied to the Robinsons.

The Court asserts that what "constitutes an impairment of vested rights is too much in the eye of the beholder to serve as a test for unconstitutional retroactivity ... [and there is] a deep division over whether a retroactive restriction on a cause of action impairs vested rights." 335 S.W.3d 126. So the Court vanquishes the vested rights jurisprudence because it is too hard to decide and it believes some cases applying it in the past were inconsistent. What areas of jurisprudence that span two centuries are not subject to the same criticisms? No one who has raised children doubts the statement that bathing a baby is challenging and risky and can be a tough chore, but it must be done. The Court throws out the baby it once embraced along with the bath water. It will come as no surprise that the new balancing test the Court establishes for evaluating retroactive legislation will be fraught with at least as many similar challenges, but have no precedents for guidance. The balancing test in Texas retroactivity jurisprudence is, candidly, a new baby in new bath water. Certainly, there are limits imposed by the Constitution on legislative power (as well as executive and judicial authority), but as Justice Scalia insightfully explained about a balancing test under the Commerce Clause of the U.S. Constitution:

Dep't of Revenue of Ky. v. Davis, 553 U.S. 328, 359, 128 S.Ct. 1801, 170 L.Ed.2d 685 (2008) (Scalia, J., concurring in part) (emphasis added).

Assuming that the Robinsons' accrued but unliquidated cause of action for personal injury is a vested right under the Retroactivity Clause, I consider whether the Legislature's exercise of its general police power outweighs the private interests at issue.

a. The Balancing Test to be Applied

We have not had the opportunity to fully discuss the contours of the police power exception vis-a-vis a retroactivity challenge. In Barshop v. Medina Underground Water Conservation District, we upheld the Edwards Aquifer Act against a retroactivity challenge where landowners above the Edwards Aquifer argued that the Act affected their vested right to withdraw unlimited amounts of water from the Aquifer. 925 S.W.2d 618, 634 (Tex.1996). Without deciding whether rights to groundwater were vested rights, we stated that because the authority was "required for the effective control of the [aquifer] to protect ... life, ... water supplies, the operation of existing industries, and the economic development of the state" and the aquifer itself was "vital to the general economy and welfare of this state," that the Retroactivity Clause in the Texas Constitution does not "absolutely bar the Legislature from enacting such statutes." Id. (quoting Act of May 30, 1993, 73d Leg., R. S., ch. 626 §§ 1.01, 1.06(a), 1993 Tex. Gen. Laws 2355, amended by Act of May 29, 1995, 74th Leg., R. S., ch. 261, 1995 Tex. Sess. Law Serv. 2505). In In re A.V., we upheld retroactive application of a statute allowing the termination of parental rights for those who are incarcerated for an extended period of time because the state has a duty to protect the safety and welfare of its children, and "[t]his `valid exercise of the police power by the Legislature to safeguard the public safety and welfare' is a recognized exception to the unconstitutionality of retroactive laws." 113 S.W.3d 355, 361 (Tex.2003) (quoting Barshop, 925 S.W.2d at 633-34). In Lebohm v. City of Galveston, we struck down a statute providing the City of Galveston a complete defense for injury caused by defective roads, streets, sidewalks, or other public places within the city limits, noting that "[n]o broad public policy or general welfare considerations are advanced to justify the charter provision as a reasonable exercise of police power [and w]e can think of none that could be advanced inasmuch as the operational effect of the provision extends only to the city limits...." 154 Tex. 192, 275 S.W.2d 951, 955 (1955).

Other states, however, have created a fuller rubric for examining the balance between the police power and the prohibition against retroactive laws. Each formulation seems to balance the nature of the public interest articulated by the Legislature, the extent to which the statute modifies or abrogates the vested right, the nature of the right the statute alters, and the fairness of the application of the new statute.17 The Robinsons' retroactivity challenge is an as-applied challenge, and thus the Robinsons must demonstrate that the statute is unconstitutional as it operates in practice against them. See Tex. Mun. League, 74 S.W.3d at 381. Therefore, it is appropriate to balance the expectations the Robinsons lost with the enactment of Chapter 149 against the degree of harm sought to be protected by the legislative enactment.

When considering the application of the police power, this case is a close one. It does not involve the potential shortage of water for millions of people, Barshop, 925 S.W.2d at 634, and it does not involve the state's duty as parens patriae to children, In re A.V., 113 S.W.3d at 361. But there are five reasons that Chapter 149 was a legitimate exercise of the police power, as applied to the Robinsons. The first three demonstrate that the Robinsons' expectations in the continued state of the law, as-applied, are low. The second two demonstrate that the Legislature's exercise of the police power was rational, justifiable, and reasonably limited.

First, at common law, Mr. Robinson's claims were not "property," were not assignable, and were extinguished when he passed away. It is only by statute that wrongful death claims continue to exist. The Legislature has broad authority to modify rights it creates by statute. "When a right or remedy is dependent on a statute, the unqualified repeal of that statute operates to deprive the party of all such rights that have not become vested or reduced to final judgment," and "all suits filed in reliance on the statute must cease...." Quick v. City of Austin, 7 S.W.3d 109, 128 (Tex.1998). This Court has further held that "[i]t is generally conceded that a right of action given by a statute may be taken away at any time, even after it has accrued and proceedings have been commenced to enforce it." Nat'l Carloading Corp. v. Phoenix-El Paso Exp., 142 Tex. 141, 176 S.W.2d 564, 568 (1944). Even assuming the Robinsons' acts of filing a lawsuit and receiving partial summary judgment resulted in some vested expectation, the Robinsons' claims, based in common law negligence and products liability, may continue only because of the statutory rights of survival, wrongful death, and successor liability through corporate merger. Accordingly, the Legislature retained discretion to modify the nature of their rights through Chapter 149's restriction on the amount of total damages recoverable against Crown Cork.

Second, Chapter 149 does not interfere with a claim sounding in contract or a claim for an injury to real or personal property, which was protected much more stringently at common law. E.g., Landgraf, 511 U.S. at 271, 114 S.Ct. 1483 (noting that the "largest category of cases in which [the Supreme Court of the United States has] applied the presumption against statutory retroactivity has involved new provisions affecting contractual or property rights, matters in which predictability and stability are of prime importance"). The Robinsons did not have an established relationship with Crown Cork (or even Mundet) with predetermined expectations that may have vested upon the occurrence of a contractual condition. Until this litigation, it is unlikely that the Robinsons even knew that Crown Cork was a successor to Mundet, or that Mundet manufactured asbestos products used in the ships on which Mr. Robinson was stationed. This weakens the expectancy the Robinsons may have had in their cause of action.

Third, Chapter 149, as applied, does not deprive the Robinsons of their cause of action against Crown Cork, and it does not deprive the Robinsons of real and substantial remedies for their alleged wrongs. The Robinsons sued twenty other defendants in this case and recovered approximately $850,000 from a number of the defendants for their injuries. They alleged that "[e]ach exposure to [asbestos-containing products] cause and/or contributed to Plaintiffs' injuries ..." and "[t]he actions of each and every Defendant are a producing and proximate cause of Plaintiffs' injuries and damages." Thus, the Robinsons lost only the right to recover against Mundet/Crown Cork, which had reached its maximum payout under Chapter 149. But the statute did not impair their right to seek substantial recoveries against other defendants, which were involved in the business of asbestos insulation for the same injuries to Mr. Robinson. There is no vested right in a remedy, and the Legislature may retroactively modify remedial laws, affect a court's jurisdiction, or provide alternative procedures or remedies. See Tex. Mun. Power Agency v. Pub. Util. Comm'n, 253 S.W.3d 184, 198 (Tex.2007); David McDavid Nissan, 84 S.W.3d at 219; Ex parte Abell, 613 S.W.2d at 260; Mellinger, 3 S.W. at 254.18 This case is a multi-defendant lawsuit where it is difficult to determine which asbestos products were the cause of Mr. Robinson's injuries.

Chapter 149 does not deprive the Robinsons of any cause of action or prohibit their right to sue any party. It simply cuts off recovery against innocent defendants at the point that the defendants have paid out for asbestos-related liabilities the fair market value of the assets of the company acquired. Importantly, Chapter 149 does not make any defendants immune from suit. Chapter 149 limits the remedy under prescribed circumstances. It is not disputed that if, for instance, Mundet's assets, acquired by Crown Cork, had a fair market value of $1 billion, Crown Cork could still be liable for damages in this suit. But because Crown Cork's asbestos-related liability payments exceeded the asset value of Mundet, it had reached the statutory limit for its liabilities as successor to Mundet. Even assuming for the sake of argument that the removal of recovery against one defendant in such a suit is not merely a change in remedy but a deprivation of a right, in this case the infringement was not a complete bar to all recovery for the wrongs alleged. Accordingly, the Robinsons were able to proceed against other defendants for the same claims based on admittedly the same injury.

Fourth, the Legislature rationally drew Chapter 149 to address a problem it perceived as very important—the effects on the Texas economy and employment because of the bankruptcy of companies that never manufactured, sold, or distributed asbestos-containing products. The asbestos litigation "crisis" had been well recognized in academic journals and even court decisions at the time the Legislature debated and enacted House Bill 4. E.g., Orszag, 44 S. TEX. L.REV. at 1078-81; Amchem Prods., Inc. v. Windsor, 521 U.S. 591, 598, 117 S.Ct. 2231, 138 L.Ed.2d 689 (1997) ("`The most objectionable aspects of asbestos litigation can be briefly summarized: dockets in both federal and state courts continue to grow; long delays are routine; trials are too long; the same issues are litigated over and over; transaction costs exceed the victims' recovery by nearly two to one; exhaustion of assets threatens and distorts the process; and future claimants may lose altogether.'" (quoting Judicial Conference Ad Hoc Committee on Asbestos Litigation, Report to the Chief Justice of the United States and Members of the Judicial Conference of the United States, at 2-3 (Mar.1991))); Humble Sand & Gravel, Inc. v. Gomez, 146 S.W.3d 170, 203-04 (Tex.2004) (O'Neill, J., dissenting) (recognizing the crisis and noting that "the solution to these problems is legislative, not judicial"); Paul F. Rothstein, What Courts Can Do in the Face of the Never-Ending Asbestos Crisis, 71 MISS. L.J. 1, 1, 4-9 (2001) (describing the "ever-expanding" crisis, and the filing of claims "[o]ver $20 billion and thirty bankruptcies later"). Others examined the potential for unfairness when a larger corporation's assets became susceptible to the stress of asbestos liability from a long-since acquired subsidiary. As stated by one commentator:

Mark H. Reeves, Note, Makes Sense to Me: How Moderate, Targeted Federal Tort Reform Legislation Could Solve the Nation's Asbestos Litigation Crisis, 56 VAND. L.REV. 1949, 1972-73 (2003); see also, e.g., Lester Brickman, The Asbestos Litigation Crisis: Is there a Need for an Administrative Alternative?, 13 CARDOZO L.REV. 1819, 1831-33 (1992) (recognizing that the asbestos litigators invoked successor liability laws "so as to reach into the deeper pockets of the companies that bought far smaller entities that manufactured asbestos-containing materials regardless of the culpability of the purchasing companies").

The Statement of Legislative Intent filed by Representative Nixon recognized an "unfairness" existing in corporate merger law where a "larger successor can easily be bankrupted by the asbestos-related liabilities it innocently received from a much smaller predecessor with which it merged may [sic] decades ago." H.J. of Tex., 78th Leg., R.S. 6042, 6043 (2003) (HB 4 Statement of Legislative Intent). The Statement also recognized that "Corporations actually in the asbestos business and their successors through merger have been financially drained by decades of litigation. As a result, nearly 70 such corporations have sought protection through bankruptcy. The cost in jobs and pension benefits, to cite just two examples, has been substantial." Id. at 6044. These findings were recognized in the House floor during debate, and were codified into the omnibus statute two years later that reformulated the method in which asbestos claims are litigated in Texas. See Act of May 19, 2005, 79th Leg., R.S., ch. 97, § 1(b)-(h), 2005 Tex. Gen. Laws 169, 169-70 (codified at TEX. CIV. PRAC. & REM.CODE §§ 90.001-.012). Protection of Texas's economy and jobs is certainly a rational basis for enacting legislation, and here there is a sufficient reason for the Legislature to enact the statute that it did.

Finally, the class of persons protected by the legislation has a rational relation to the legislative purpose of the legislation. The Legislature chose to relieve liability on "innocent successors," companies that did not manufacture or sell asbestos, but rather acquired a company that did. And the Legislature mediated the perceived unfairness not by foreclosing a remedy altogether, but merely limiting the remedy to the fair value of the acquired company's assets. TEX. CIV. PRAC. & REM.CODE §§ 149.001, .003. In this case, that is exactly what happened. Crown Cork's total liabilities for the asbestos sold and manufactured by Mundet far exceeded Mundet's present-day fair value. Had Mundet never been acquired by Crown Cork, its payouts for asbestos liability would have exceeded its value as a going concern, it likely would have been bankrupt, and, almost certainly, no money would have remained to pay the Robinsons' claims if they obtained a judgment against it. See In re Joint E. & S. Dists. Asbestos Litig., 237 F.Supp.2d 297, 302-06 (E.D.N.Y.2002) (discussing the factual and procedural background of the bankruptcy of the Manville Corporation, the establishment of the Manville Trust following its bankruptcy to pay asbestos claims, and its reformation once it was discovered that the trust was "deeply insolvent" and that beneficiaries would not be able to be paid in full, or even paid at all). Crown Cork chose to acquire Mundet through a statutory merger and not through an asset purchase, but it remains the purview of the Legislature to modify the legal effect of continuing liability of such mergers in Texas to avoid the ruin of businesses possessing assets that had nothing to do with asbestos production or manufacture. Importantly, the legislation restricts neither the right nor the remedy of plaintiffs who prove that Crown Cork itself caused them injury; it only addresses imputed successor liability.

In short, for the reasons articulated above, the Robinsons' interest in their accrued, but unliquidated cause of action, is low. Their vested expectancy, if any, is minimal. Their right of recovery for the injuries complained of was not foreclosed. And their relation to Crown Cork was attenuated. The public interest in the legislation, and its retroactivity, is moderate. The Legislature acted in response to a known litigation crisis and acted with a reasonable and narrowly tailored response based on the current climate. Individuals may or may not personally believe in the wisdom of the particular legislation, but it is not our province to second-guess legislation because we do not agree with its policy. See McIntyre v. Ramirez, 109 S.W.3d 741, 748 (Tex.2003).

b. A Critique of the Court's Test

Although I disagree with the Court's analytical framework in arriving at its three-factor balancing test and the unfoundedly rigorous legal standards it applies, I do not wholesale disagree with the categories it has set up to determine whether a retrospective law is unconstitutionally retroactive. However, the Court's application of the law to the facts in this case creates more difficulties for the Legislature and the courts of our state in reviewing retroactive laws, and creates significant and unnecessary impediments to the Legislature's ability to correct law and make beneficial legislative changes in the future.

First, I disagree with the "compelling reason" standard applied by the Court. Nothing in our precedent, or any case law, requires such a heightened review of retroactive legislation. The Court repeatedly mentions the heavy presumption against retroactive legislation, but the presumption falls away in this case. The presumption is removed when a legislature "itself has affirmatively considered the potential unfairness of retroactive application and determined that it is an acceptable price to pay for the countervailing benefits." Landgraf, 511 U.S. at 272-73, 114 S.Ct. 1483. Not only did the Legislature "consider the potential unfairness" in this case, it voted to apply Chapter 149 retroactively by a supermajority. The Court's point that we should view fully retroactive legislation with skepticism is well taken; however, the presumption against retroactivity is unnecessary when the Legislature expressly concludes that the statute is to be applied retroactively. Id.; accord Lockheed Corp. v. Spink, 517 U.S. 882, 896-97, 116 S.Ct. 1783, 135 L.Ed.2d 153 (1996) ("[When] the temporal effect of a statute is manifest on its face, `there is no need to resort to judicial default rules,' and inquiry is at an end." (quoting Landgraf, 511 U.S. at 280, 114 S.Ct. 1483)); Tello v. Dean Witter Reynolds, Inc., 410 F.3d 1275, 1281-82 (11th Cir.2005) ("[The] presumption and analysis, however, are unwarranted when Congress states its unambiguous intention that the statute apply retroactively to pre-enactment conduct...."). Because it is for the Legislature to initially determine whether the benefits of retroactive legislation outweigh the detriments (at least to the statute as a whole), we are not commanded to review that decision to determine whether their justification was "compelling."

Second, the Court's evaluation of the Robinsons' interest seems to be focused on its pretrial evaluation of not only the existence of the Robinsons' claims, but their strength. The Court argues that the Robinsons' claims have "a substantial basis in fact" and that their claims are "mature tort[s], [such that] recovery is more predictable." 335 S.W.3d 126. I would not require courts in this state to evaluate plaintiffs' claims or defendants' defenses, under the Retroactivity Clause on whether the parties are likely to win or their claims have a "substantial basis in fact." As any experienced lawyer will acknowledge, the strength of a claim and the likelihood of success in litigation may be separate and independent things. This consideration is unwieldy, suggesting that the Legislature can enact retroactive legislation affecting substantive rights so long as there is a chance that it will not matter, at the end of the day.

Third, the statute does not affect settled expectations to the degree alleged by the Court. The Court alleges that the statute will affect the recovery "to which the Robinsons are entitled," once again presuming that the Robinsons' claims against Crown Cork will be successful. 335 S.W.3d 126. As discussed above, the Robinsons had no pre-tort contact with Crown Cork, and had no settled expectation that Mundet would be acquired by a richer company able to pay for Mundet's debts.

Fourth, the Court penalizes the Legislature because the legislation does not contain expressed "findings to justify Chapter 149." 335 S.W.3d 126. The Court does not consider the well-known facts about the asbestos crisis, Crown Cork's financial stake, subsequently codified legislative findings, or the possibility that other businesses may be subjected to financial ruin, as these facts were not included in the actual statutory language in House Bill 4. While I agree that such statutory findings are most helpful in determining legislative intent, United States v. Lopez, 514 U.S. 549, 562-63, 115 S.Ct. 1624, 131 L.Ed.2d 626 (1995) (concerning the Commerce Clause), I am aware of no Texas case that requires them. And, in fact, if the Legislature were to be so required for every bill in which their police power may be challenged, certainly the legislative process would be significantly burdened. Rational basis review does not require the Legislature to provide any particular purpose; the law will be upheld "if there is any conceivable state of facts which would support it." Carmichael v. S. Coal & Coke Co., 301 U.S. 495, 509, 57 S.Ct. 868, 81 L.Ed. 1245 (1937). The law may be valid even if the Legislature did not consider the valid purposes, but so long as the purpose "may have been considered to be true." Nordlinger v. Hahn, 505 U.S. 1, 11, 112 S.Ct. 2326, 120 L.Ed.2d 1 (1992) (citations omitted).

Thus, I believe it is imprudent to abandon our vested rights jurisprudence, and as applied, the Robinsons' do not have vested rights in their causes of action against Crown Cork. Even if the Robinsons' claims are vested rights, I would hold that, on balance, the Legislature's exercise of police power outweighs the Robinsons' rights, and thus Chapter 149 does not violate article I, section 16 of the Texas Constitution.

B. Special Law

Because the Court determines that Chapter 149 is unconstitutionally retroactive as applied to the Robinsons, it does not address the Robinsons' second argument, that Chapter 149 is an unconstitutional "special law." I would hold that it is not.

Article III, section 56(b) of the Texas Constitution provides that "where a general law can be made applicable, no local or special law shall be enacted." TEX. CONST. art. III, § 56(b). A "special law" is a statute that "relates to particular persons or things of a class," rather than the class as a whole. Clark v. Finley, 93 Tex. 171, 54 S.W. 343, 345 (1899) (emphasis added), cited in Lucas v. United States, 757 S.W.2d 687, 700 (Tex.1988); see also Ford Motor Co. v. Sheldon, 22 S.W.3d 444, 456 (Tex.2000) (defining a "special law" as one that "impermissibly distinguishes between groups on some basis other than geography" (citing Tex. Boll Weevil Eradication Found. v. Lewellen, 952 S.W.2d 454, 465 (Tex.1997))). The prohibition on special laws was added to the Texas Constitution of 1876 as one of many practical answers to the prevalent abuse of legislative and executive power that occurred in Texas following the Reconstruction. A.J. Thomas, Jr. & Ann Van Wynen Thomas, The Texas Constitution of 1876, 35 TEX. L.REV. 907, 915 (1957). In one session of the post-Reconstruction legislature five hundred special laws were passed. Id. Section 56 was thus seen to prevent "logrolling,"19 to ensure against the granting of special privileges, and to prevent lawmakers from trading votes "for the advancement of personal rather than public interest." Miller v. El Paso Cnty., 136 Tex. 370, 150 S.W.2d 1000, 1001 (1941); Sheldon, 22 S.W.3d at 456.

In the early twentieth century, the Court developed a test for reviewing whether a law providing a privilege to a particular class is in actuality a veiled attempt to provide a privilege to a particular member of the class. See Sheldon, 22 S.W.3d at 450-51; Maple Run at Austin Mun. Util. Dist. v. Monaghan, 931 S.W.2d 941, 945 (Tex.1996); Robinson v. Hill, 507 S.W.2d 521, 525 (Tex.1974); R.H.O, Recent Case, Statutes—Special Laws—Reasonableness of Classification, 11 TEX. L.REV. 134, 134-35 (1932) (collecting cases describing the legal standard for review of a special law). The Court first determines whether there is a reasonable basis for the classification made by the law, and then determines whether the law operates equally on all within the class. Rodriguez v. Gonzales, 148 Tex. 537, 227 S.W.2d 791, 793 (1950); Sheldon, 22 S.W.3d at 451. Only if the law fails both tests is it a special law and unconstitutional.

The determination of a "reasonable basis" for the classification is not an invitation for the Court to engage in weighing the relative pros and cons of a particular policy choice made by the Legislature. As stated by this Court over 100 years ago:

Clark, 54 S.W. at 345-46. We do not analyze the Legislature's classification to determine whether the classification is a good or bad idea. See Smith v. Davis, 426 S.W.2d 827, 831 (Tex. 1968). Rather we analyze to ensure that the classification is not made to "evade the prohibition of the constitution as to special laws by making a law applicable to a pretended class, which is, in fact no class...." Clark, 54 S.W. at 345. We presume the statute is valid, and "a mere difference of opinion" between the Court and the Legislature will not be sufficient to overcome the presumption of validity. Smith, 426 S.W.2d at 831.

The Rodriguez test's two-part structure provides the framework to determine whether a class is a "pretended class." The first part of the test examines the delineated class vis-a-vis the purpose of the legislation. Rodriguez, 227 S.W.2d at 793. For example, if the purpose of the law is to provide tax relief to businesses in the sports entertainment industry, but the tax relief is given only to businesses belonging to or supporting teams in leagues or conferences with "National" in their name but not with leagues or conferences with "American" in their name, the classification would likely have no rational relation to the purpose of the statute.

The second part of the test examines whether similarly situated parties are treated similarly under the classification, or whether the classification makes an irrational category considering the intent of the statute. See, e.g., Rodriguez, 227 S.W.2d at 794 (holding that statute setting out special procedures for collecting delinquent taxes on parcels of land greater than 1,000 acres situated in counties bordering Mexico and whose title emanated from the King of Spain as an unconstitutional special law, as there was "no substantial difference in the situation or circumstance of border counties relating to suits for delinquent taxes"); Miller, 150 S.W.2d at 1002-03 (holding as unconstitutional a statute providing an economic development tax only in counties meeting population requirements, due to the fact that the statute's classification was not distinct in any substantial manner from other counties in the state). Back to the example, the tax relief statute above would likely be unconstitutional, as its effect is to provide relief to the Houston Astros and the Dallas Cowboys and the businesses that support them (as the Astros are a member of the National League, and the Cowboys are a member of the National Football Conference), but would not provide relief to supporters of the Houston Texans and the Texas Rangers (as the Texans are a member of the American Football Conference and the Rangers are a member of the American League). The classification is a "pretended class" because the classification has no relation to the purpose of the law and treats similarly situated teams differently. Although the Court does not defer to the Legislature to determine whether a law is general or special, it does defer to the Legislature's policy choices and presumes that law is constitutional. See Smith, 426 S.W.2d at 831; McIntyre v. Ramirez, 109 S.W.3d 741, 748 (Tex.2003) ("Our role here, however, is not to second-guess the policy choices that inform our statutes or to weigh the effectiveness of their results; rather, our task is to interpret those statutes in a manner that effectuates the Legislature's intent.").

In this case, the purpose of the law has been clearly expressed by the Legislature—to eliminate the unfairness created when a corporation merged with a smaller corporation that had previously been engaged in the manufacture or sale of asbestos is exposed to asbestos liability exceeding the value of the acquired corporation, and to save such a corporation from bankruptcy. H.J. of Tex., 78th Leg., R.S. 6042, 6043 (2003) (HB 4 Statement of Legislative Intent). To address concerns in the Legislature, the measure was restricted in three ways. First, the original transfer of liabilities had to occur prior to May 13, 1968. This was the date in which the American Conference of Governmental Industrial Hygienists first adopted a change in the recommended threshold limit for asbestos in the air of a workplace. Second, to get the benefit of the legislation, the acquiring corporation could not continue in the asbestos business. Third, if the successor continued to control a premises after the merger, the successor would continue to be liable for any asbestos-related premises liabilities it received from the predecessor for injuries caused on those premises. Id. at 6043-44.

The Robinsons attack these limitations as pretexts to limit relief just to Crown Cork. However, it is clear that, regardless of the wisdom of the classifications, the classifications are rationally related to the objective of the bill. The act sought to protect "innocent" successor corporations. To define the most "innocent," the Legislature chose to limit mergers occurring prior to May 13, 1968. The Robinsons claim that this date was chosen arbitrarily and that the dangers of asbestos in the workplace were known prior to the ACGIH's modification. However, this is the date decided upon by the Legislature, and it has a rational relationship to the legislation—the Legislature could have, no doubt, chosen any number of cutoff dates to decide which successor corporations are the most "innocent," and while others may disagree as to the appropriateness of the date, such would merely be a "difference of opinion," and insufficient basis for overturning the statute. Smith, 426 S.W.2d at 831; see also Exxon Mobil Corp. v. Altimore, 256 S.W.3d 415, 420-22 (Tex.App.-Houston [14th Dist.] 2008, no pet.) (discussing, in the context of the basis for a punitive damages award "scientists' knowledge of the risk to refinery workers" of asbestos, and noting studies originating in the 1940s, 1950s, 1960s, and 1970s). Similarly, the second and third limitations also seek to limit protection to those businesses that were not involved with the manufacture or distribution of asbestos, or those that actually had asbestos on the premises. This is also a rational distinction: The Legislature sought to protect those businesses that had nothing to do with asbestos prior to a merger, had nothing to do with asbestos after the merger, and had no asbestos on its premises. The classifications are rational.

The Robinsons also argue that the law is a "special law" because it created a class of one—evidenced by (a) the fact that Crown Cork did not identify any other businesses to which the law applied, (b) Crown Cork's lobbying for the law in Texas and other states, and (c) statements by members of the Legislature that they were addressing "the Crown Cork and Seal Issue."

The Robinsons cite to Miller's statement that classification "must be broad enough to include a substantial class ..." to mean that it is the burden of the proponent of the law to prove that the law must apply to more than one person. Miller, 150 S.W.2d at 1001. On the contrary, the size of the class, itself, is not determinative. While courts must be more exacting in reviewing a law that appears only to apply to one party, a "substantial" class does not equate to a class with thousands, hundreds, or even dozens of members. There are no doubt many Texas laws that apply to a small subset of the population; rather, a "substantial" class is one that has substance—a real class of persons or entities, as opposed to a "pretended" class created as a pretext.

The Robinsons' evidence of pretext is no evidence at all. The Robinsons' bare argument that Crown Cork is a "class of one" is insufficient. First, it is not Crown Cork's, but the Robinsons' burden to demonstrate that the law is a special law. Second, even if the Robinsons could show that the law currently applied only to Crown Cork, that alone would not fulfill the burden that the law was special. As discussed above, the Robinsons must show that the classifications made by the Legislature were not rationally related to the objective of the law, and the Robinsons must show that the legislation has treated a similarly situated successor company differently from Crown. They have done neither.

The only other evidence the Robinsons provide is evidence of legislative history. The Robinsons argue that the law is special because Crown Cork lobbied for the act and that at least one legislator called the Act the "Crown Cork issue" in a committee hearing. This evidence is also unavailing. First, as a beneficiary of this law, Crown Cork would certainly lobby for its enactment. But then again, public interest groups, individuals, and businesses regularly lobby for legislation that affects them directly or as an industry, and lobbyists regularly draft legislation for legislators. See, e.g., Victoria F. Nourse & Jane S. Schacter, The Politics of Legislative Drafting: A Congressional Case Study, 77 N.Y.U. L. REV. 575, 583, 587, 591 (2002) (noting a number of responses by legislative aides that lobbyists regularly draft the text of bills debated in the Senate Judiciary Committee and discussing an account by a legislative aide where a companion bill was "negotiated and drafted by lobbyists and introduced with only `minor changes'"). Many involved in the "sausage making"20 task of developing law use lobbyists to draft the text of bills because lobbyists provide valuable information and perspective on the bills being introduced. Id. at 583. Cognizant as I am of the need to avoid the gifts given by the Legislature to favored individuals, the Robinsons must come up with more evidence than the mere fact that Crown Cork was involved in the passing, or even the drafting, of the act in question.

Likewise, the Robinsons' evidence of Senator Ratliff's statement is also not evidence of House Bill 4's "special law" status. The senator described Article 17 as "the Crown Cork and Seal asbestos issue." First, the statement is no evidence because, as this Court has repeatedly stated, a single statement by a single legislator does not evidence legislative intent and does not determine legislative intent. E.g., AT & T Commc'ns of Tex., L.P. v. Sw. Bell Tel. Co., 186 S.W.3d 517, 528-29 (Tex.2006); Gen. Chem. Corp. v. De La Lastra, 852 S.W.2d 916, 923 (Tex.1993). Second, to countenance this statement as even "persuasive authority as might be given the comments of any learned scholar of the subject," De La Lastra, 852 S.W.2d at 923, would be to do a disservice to the legislative process. Countless laws are either championed by a particular person or entity or arise out of the circumstances that will be or have been experienced by an individual or a business.21

In sum, the Robinsons meet neither of the factors in the Rodriguez test. The Robinsons have not shown that the Legislator's classifications are irrational or not related to the objective of the statute, nor have they shown that the Legislature has created a "pretended" class by excluding similarly situated entities.

III. CONCLUSION

I would hold that Chapter 149 is not an unconstitutional special law, and is not unconstitutionally retroactive as applied to the Robinsons because the law limited available remedies and did not destroy the Robinsons' vested rights. I therefore respectfully dissent.