Insurance Appraisal When Coverage and Bad Faith Are in Dispute: In re ACE American Insurance Co.

My business has a property-insurance claim, and we disagree with the insurer about what is owed. If we believe the insurer has handled the claim in bad faith, can we still be required to go through the policy’s appraisal process?

The Supreme Court of Texas addressed that question this spring in In re ACE American Insurance Co., 734 S.W.3d 887 (Tex. 2026).

What appraisal is. Most property-insurance policies contain an appraisal clause. If the insurer and the insured disagree on the “amount of loss,” either side can demand appraisal. Each side picks an appraiser, the two appraisers pick an umpire, and a decision agreed to by any two of them is binding as to the amount of loss. Under longstanding Texas law, appraisal decides only how much the loss is. Whether the insurer is liable, and what the policy covers, remain questions for the courts. Id. at 891 & n.3, 892 (citing State Farm Lloyds v. Johnson, 290 S.W.3d 886, 890 (Tex. 2009)). Appraisal clauses are “generally enforceable, absent illegality or waiver.” Id. at 892.

What happened. In June 2022, a water line feeding the fire-suppression system at a Dallas food-distribution warehouse ruptured beneath the slab and caused extensive damage. The insurers paid part of the claim, including about $1.2 million for mold remediation. The insured contended it was owed more, including the full $10 million mold sublimit. The insurers demanded appraisal. The insured refused, calling the demand “premature and unwarranted.” After further negotiations failed, the insurers sued to compel appraisal. The insured counterclaimed for breach of contract, Insurance Code violations, and bad faith, alleging that the insurers had invoked appraisal to “coerce” it into accepting a “lowball” offer. Id. at 890–91, 893.

The trial court denied the motion to compel appraisal, and the Dallas Court of Appeals denied mandamus relief. The Supreme Court granted it. Id. at 891, 896.

Coverage questions do not block appraisal. The insured’s main argument was that the dispute was really about coverage, causation, and whether damage existed at all — matters outside an appraiser’s role. The Court disagreed. No one suggested the damage had been caused by anything other than the burst water line, a covered peril. What the insurers mainly contended was that the insured had spent more than necessary to restore the warehouse. On the Court’s framing, that is a dispute over the amount of loss. Id. at 893.

The Court applied that reasoning to each of the insured’s examples:

•           Mold. The difference between $1.2 million and $10 million was “over the amount of loss.” Any question about how mold losses should be split with another insurer could be decided by a court after the appraisers set the number. Whether remediation should have been billed on a time-and-materials basis instead of a fixed price was also an appraisal question. Id. at 893.

•           Building-code costs. The insurers did not dispute coverage for the added cost of complying with building codes. They argued only that the insured’s more extensive repairs triggered extra code requirements. The scope of necessary repairs was for the appraisers. Any narrower coverage question about a particular compliance cost could still go to a court afterward. Id. at 893–94.

•           Replacement cost and technical complexity. The insured argued that the dispute turned on “complex engineering and construction methodologies” that appraisers were not equipped to handle. The Court said that an appraiser deciding whether cheaper methods would have sufficed was “well within the ‘amount of loss’ lane,” and that the policy contained no exception for technically complex disputes. Id. at 894.

The Court was careful about what it was not deciding. It did not hold that no coverage disputes remain. It held only that such disputes “do not render an appraisal improper in the first instance.” Id. at 894. Quoting its 2009 decision in Johnson, it repeated that “unless the ‘amount of loss’ will never be needed,” appraisal should generally go forward “without preemptive intervention by the courts.” Id. at 893–94.

A disagreement is enough. The insured also argued that the insurers’ valuation had been “inconsistent and constantly shifting,” so there was no genuine disagreement to trigger appraisal. The Court rejected that. To “disagree,” it said, is simply “to fail to agree.” The insurers had consistently valued the loss well below the insured’s figure, and “[t]he appraisal provision requires nothing more.” Id. at 894–95.

Alleged bad faith is not an exception. Finally, the insured argued that the insurers’ alleged failures — slow and bad-faith adjustment, nonpayment, and unfounded coverage defenses — were prior material breaches that excused it from appraisal. The Court rejected that argument as well. Accepting it would require a court to decide whether the insurer breached the policy before appraisal could occur, which “puts the cart before the horse.” And if an insured could avoid appraisal by alleging a coverage or claims-handling dispute, the Court said, “appraisal clauses would be virtually a nullity.” Id. at 895–96. The Court held that “an insurer’s alleged bad faith in handling a claim does not constitute an exception to the general enforceability of an appraisal clause.” Id. at 896. It made no difference that the insured said it had evidence, not just allegations, of bad faith; at this stage, neither breach nor bad faith had been tried or decided. Id. at 896 n.7.

What it appears to mean. For commercial policyholders, the decision suggests that framing a disagreement as a coverage or bad-faith dispute may not keep it out of appraisal. That appears especially true where the real fight is over the cost or scope of repairs. Disputes about which repair methods were necessary, or whether costs were inflated, seem to fall squarely within appraisal, however technical they are.

The opinion also preserves several things. Coverage questions, and the insured’s claims for breach of contract and bad faith, remain for the courts. The policy itself reserved the insured’s right to sue and the insurer’s right to deny the claim. And the Court noted that an appraisal award can still be set aside afterward in limited circumstances, such as fraud, accident, or mistake. Id. at 891 n.3, 895 n.6.

See also our post on quantum meruit when a contract exists, another recent decision about holding parties to the terms of their contract.

Proximate Cause and Highway Collisions: Werner Enterprises, Inc. v. Blake

When one driver loses control and crosses into oncoming traffic, can the other driver — or that driver’s employer — share responsibility for the crash because he was going too fast for the conditions?

The Supreme Court of Texas addressed that question in Werner Enterprises, Inc. v. Blake, 719 S.W.3d 525 (Tex. 2025), which reversed a judgment of more than $100 million in a fatal trucking case. It is one of the most closely watched injury decisions of the last few years, and its reasoning about causation may reach well beyond trucking.

What happened. On an icy December afternoon in 2014, a pickup traveling east on I-20 near Odessa lost control. In about two seconds it crossed a 42-foot grassy median and collided with a Werner 18-wheeler in the westbound lanes. The truck was driven by a driver in training, Shiraz Ali. One child in the pickup was killed, and three other family members were seriously injured. Id. at 529–31.

There was evidence that the roads were “like a skating rink,” that Ali had passed several accidents in the hour before the crash, and that he was driving faster than was safe on the ice, though below the posted limit. The plaintiffs’ own expert testified, however, that Ali reacted “appropriate[ly]” in the two seconds after the pickup started across the median. Id. at 530, 534 & n.2.

Werner admitted that Ali was acting within the scope of his employment. The jury nevertheless found both Ali and Werner negligent, assigning 14% of the responsibility to Ali, 70% to other Werner employees, and 16% to the pickup’s driver. Judgment exceeded $100 million. The Fourteenth Court of Appeals in Houston, sitting en banc, affirmed over four dissents. Id. at 531.

But-for cause is not enough. Texas negligence law requires proximate cause. Part of that is “but-for” causation — the injury would not have happened without the defendant’s negligence. But the plaintiff must also show that the negligence was a “substantial factor” in bringing about the injury. On the Court’s framing, that requirement builds in “the idea of responsibility”: liability falls only on a party who is “actually responsible for the ultimate harm.” A party whose conduct “merely creates the condition that makes the harm possible” is not a substantial factor as a matter of law. Id. at 528, 532–33.

One proximate cause, not two. The Court assumed that Ali’s speed, and perhaps even his being on the road at all, was negligent. It also assumed that the crash would not have happened if he had been driving much more slowly. Id. at 534. But it held that this showed only but-for causation. “[T]here can be more than one proximate cause of an injury,” the Court said, but “[t]hat does not mean there is always more than one. On these facts, we hold that there was only one.” Id. at 534–35.

The Court reasoned that nothing Ali did contributed to the pickup’s losing control and crossing the median. His truck’s position on the other side of the highway at that moment was a “happenstance of place and time” — a condition that made the harm possible, not its cause. The Court added that if Ali had been going much faster, the collision also would not have happened. Id. at 535–36.

Why the type of road mattered. The plaintiffs relied on Biggers v. Continental Bus System, Inc., 303 S.W.2d 359 (Tex. 1957), which upheld a verdict against a speeding bus after a car was pushed into its lane. The Court did not overrule Biggers, but it distinguished it on two grounds. Biggers involved a narrow two-lane road, where entering oncoming traffic is a more familiar risk than on a highway with a broad median. And the bus driver there had about 3.5 seconds to react, compared with roughly two seconds here. Id. at 536–38.

As a “practical” matter of “common experience,” the Court reasoned, drivers rely on one another to “maintain control of [their] vehicle[s] and to stay on [their] side of the road.” It added that it “struggle[d] to imagine a scenario” in which the law could fault a driver on one side of a broadly divided highway for a collision caused by an oncoming vehicle that crossed the median before he had time to react. Id. at 539.

The employer’s liability fell with the driver’s. The jury had also found Werner liable on two other theories: that it negligently trained and supervised Ali, and that it was “directly” negligent in sending an inexperienced trainee into winter weather without weather updates. The Court held that both theories depended on Ali’s driving having caused the injuries. Because it had not, neither could stand. Id. at 539–41. The Court reversed and rendered judgment for Werner and Ali. Id. at 541.

The separate opinions. The vote was 5–3 on the outcome. Chief Justice Blacklock wrote for the Court, and Justice Lehrmann did not participate.

•           Justice Bland, joined by Justices Boyd and Huddle, agreed that the judgment had to be reversed but would have ordered a new trial instead of rendering judgment for the defendants. In her view, there was some evidence that Ali’s speed made the injuries worse. She pointed to Werner’s own expert, who testified that at 15 miles per hour the truck would have stopped before the crash, and to a similar collision that day in which a truck moving at about 5 miles per hour left no one hurt. She also criticized the jury charge, calling the 84%-to-16% apportionment against Werner “upside-down.” Id. at 544–55 (Bland, J., dissenting in part).

•           Justice Young, joined by Justice Huddle, wrote separately about the “admission rule.” Under that rule, once an employer admits that its employee was acting within the scope of employment, separate claims such as negligent hiring or training generally should not go to the jury. The Court did not decide whether Texas follows the rule. Justice Young wrote that he was “inclined” to adopt it, and that a bare allegation of gross negligence should not be enough to get around it. Id. at 541–44 (Young, J., concurring).

What it appears to mean. On the Court’s reasoning, when a driver loses control and crosses a wide median into oncoming traffic, the driver who stayed in his own lane and had no real time to react is unlikely to be treated as a proximate cause of the crash, even if he was driving too fast for the conditions. Claims against that driver’s employer that depend on his driving appear to rise or fall with the claim against him.

The opinion also seems to set some limits on its own reach:

•           It distinguishes narrow two-lane roads and longer reaction times rather than overruling Biggers.

•           In a footnote, the Court said it was not holding that the at-fault driver is always the sole proximate cause. It noted that claims involving a vehicle defect, or conduct that made the injuries worse (such as a failure to wear a seat belt), may call for a different analysis. Id. at 536 n.4.

Whether Texas will adopt the admission rule, and whether a gross-negligence exception will apply to it, remain open questions that the concurrence suggests the Court may take up in a future case.

See also our post on shipper liability for motor-carrier accidents, another recent decision about responsibility for trucking accidents.

A Parent’s Signature and a Child’s Injury Claim: UATP Management, LLC v. Geter

 If I sign a contract with an arbitration clause, can it be enforced against someone I signed for — or by someone who never signed it at all?

The Supreme Court of Texas addressed those questions on October 16, 2026, in UATP Management, LLC v. Geter, No. 24-0905 (Tex. Oct. 16, 2026) (per curiam).

What happened. Before her daughter went into the play area at an Urban Air trampoline park in Beaumont, a mother signed a release and indemnity agreement on the child’s behalf. The agreement released and promised not to sue the local park and a list of related companies and individuals, which it called the “protected parties.” It also said that “[a]ny dispute or claim arising out of or relating to this Agreement” would be resolved by binding arbitration under the Federal Arbitration Act. The child was hurt, and the mother sued the local park and the protected parties on her behalf.

The defendants moved to compel arbitration, and the trial court denied the motion. The court of appeals in Houston split the difference. It held that the local park could compel arbitration, but that the protected parties could not. Its reasoning was that they had not signed the agreement and the arbitration clause did not mention them. The mother did not appeal the ruling for the local park, so only the protected parties’ position reached the Supreme Court. The Court ruled for them, deciding the case without oral argument.

A parent can sign for a child. The court of appeals had relied on a doctrine called “direct-benefits estoppel” to bind the child, reasoning that she had accepted the benefits of the contract. The Supreme Court said that step was unnecessary. Because the mother signed on the child’s behalf, the agreement bound the child directly. Relying on its 2023 decision in Taylor Morrison of Texas, Inc. v. Ha, 660 S.W.3d 529 (Tex. 2023), the Court repeated that under the Family Code “parents may sign arbitration agreements on behalf of their children.” It noted an exception only in rare cases, “such as when a parent’s interests are adverse to the child’s interests.” Slip op. at 5–6.

The child could not back out. Texas law has long allowed a minor to disaffirm, or walk away from, a contract. But on the Court’s framing, that right has a limit: a minor “is not permitted to retain the benefits of a contract while repudiating its obligations.” Id. at 6 (quoting Dairyland Cnty. Mut. Ins. Co. of Tex. v. Roman, 498 S.W.2d 154, 158 (Tex. 1973)). Once the child went into the park, the Court reasoned, she had received what the agreement offered and could no longer disavow it. Id.

A court, not the arbitrator, decides who can enforce the agreement. The protected parties argued that the arbitrator, not a judge, should decide whether they could invoke the arbitration clause. The Court disagreed. When the question is whether someone who did not sign the agreement can enforce it, “it is for the courts, not an arbitrator,” to decide. Id. at 7. The Court cited, among other cases, its 2025 decision in another Urban Air case, Cerna ex rel. R.W. v. Pearland Urban Air, LLC, 714 S.W.3d 585 (Tex. 2025).

The related companies could enforce it as third-party beneficiaries. Texas law recognizes several ways a non-signatory can enforce or be bound by an arbitration agreement. Jody James Farms, JV v. Altman Grp., Inc., 547 S.W.3d 624, 633 (Tex. 2018). The Court addressed only one: whether the protected parties were third-party beneficiaries, meaning that the people who signed the contract intended to benefit them. That benefit has to be more than incidental. Slip op. at 7–8.

The Court found that intent in the agreement itself. By naming the companies as “protected parties” and releasing claims against them, the agreement showed an intent to benefit them. The mother argued that the protection extended only to the release and indemnity, not to the arbitration clause, which did not mention the protected parties. The Court read the contract as a whole instead. The arbitration clause applied to “any dispute or claim” relating to the agreement and did not exclude anyone. In the Court’s view, arbitration was the parties’ chosen way to resolve disputes over the very release that protected these companies. Id. at 8–9.

The Court reversed the part of the court of appeals’ judgment that had denied arbitration for the protected parties. It sent the case back to the trial court. Id. at 9.

What it appears to mean. For families, the decision suggests that the online or front-desk agreements common at trampoline parks, gyms, camps, and similar venues may carry real weight in Texas. A parent’s signature may bind the child to arbitration. The child’s use of the facility may close off any later attempt to disaffirm. And companies named in the agreement may be able to require arbitration even though they never signed it. For businesses, the opinion seems to show the value of drafting: the outcome turned on the agreement naming the related companies and on an arbitration clause broad enough to reach them.

The opinion also leaves some questions open. It decides where the child’s claims will be heard, not whether the release bars them. Some Texas courts of appeals have held that a parent cannot release a child’s personal-injury claims in advance. Whether that rule survives, and how an arbitrator will apply it, are questions this opinion does not reach. Nor does the opinion address the “rare” situations in which a parent’s interests might be adverse to the child’s.

See also our post on independent contractors and open-and-obvious dangers.

Appeal Bonds in Multi-Defendant Cases: In re Greystar Development & Construction, L.P.

We won a large judgment against several related companies. They’ve appealed. Do they have to post one appeal bond between them, or one each?

The Supreme Court of Texas answered that question this spring in In re Greystar Development & Construction, L.P., 741 S.W.3d 32 (Tex. 2026) (orig. proceeding). The vote was 5–4, and the answer was: one each.

What happened. In 2019 a construction crane collapsed during a storm in Dallas and struck an apartment building, killing a resident, Kiersten Smith. Her parents sued three affiliated Greystar entities and the crane company. The jury found that the three Greystar entities were operating as a joint enterprise, held them responsible for Smith’s death, and awarded roughly $360 million in non-economic damages plus $500 million in exemplary damages, which the trial court reduced under the statutory caps. The judgment made the three entities jointly and severally liable for the compensatory award. 741 S.W.3d at 37.

The bond. A defendant who appeals a money judgment in Texas can stop the plaintiff from collecting while the appeal is pending by posting security — usually a bond — in the amount of the compensatory damages plus interest and costs. The Legislature has capped that amount, however: the bond need not exceed the lesser of 50 percent of the judgment debtor’s net worth or $25 million. Tex. Civ. Prac. & Rem. Code § 52.006(a)–(b). The three Greystar entities posted a single $25 million bond, jointly, to suspend the judgment as to all of them.

The plaintiffs objected. Their position was that the $25 million cap applies to each defendant, so one bond could cover only one of the three. The trial court agreed, ruled that the joint bond could suspend the judgment against one entity at most, and declared that until the defendants filed individual bonds, no valid bond was in place — which meant the plaintiffs could begin collecting from the other two. The Dallas court of appeals upheld that ruling. Id. at 37–38.

The Court’s holding. The majority, in an opinion by Justice Busby, held that the cap applies per judgment debtor. The reasoning started with the statute’s definitions: “security” is defined as “a bond … posted … by a judgment debtor to suspend execution of the judgment,” and the cap says “the amount of security … must not exceed” $25 million or half “the judgment debtor’s net worth.” §§ 52.001, 52.006(b). Reading those together, the Court concluded that the cap limits the bond a debtor posts to suspend the judgment against itself, not the total of all bonds posted by everyone. Id. at 36, 38.

The Court also pointed to the surrounding law. Execution, garnishment, appellate rights, and supersedeas practice are all party-specific — each defendant files its own notice of appeal, each is subject to its own writ, and each debtor who wants to stop collection has long been required to post security for the full amount it owes. A per-judgment cap, the Court said, would be in tension with that structure and would force courts to invent exceptions the statute doesn’t contain. Id. at 39–47. As for policy, the majority noted that the $25 million cap is only one of several routes the Legislature has provided for reducing a bond, and saw no reason to read this one as broadly as possible. Id. at 48–49.

The remedy. Although the trial court was right about the cap, the Court held it abused its discretion by invalidating the joint bond immediately as to two of the entities, without giving them time to post their own. Drawing on the 20-day cure periods elsewhere in Rule 24, the Court said parties should presumptively be allowed 20 days to fix a bond found insufficient under the cap, and sent the case back with instructions to allow the two entities a reasonable time to post sufficient bonds. Id. at 50.

The dissent. Justice Huddle, joined by Justices Bland, Young, and Sullivan, agreed the trial court shouldn’t have invalidated the bond without notice but disagreed with the majority’s reading of the cap. In the dissent’s view, the statute caps neither per debtor nor per judgment but per bond: joint supersedeas bonds have been permitted in Texas for well over a century, the Government Code says the singular includes the plural, and nothing in § 52.006 says “$25 million per judgment debtor.” The dissent warned that the majority’s rule “drives up the cost of appeals in multi-defendant cases,” may require defendants to post more security than the judgment itself, and will “reverberate most loudly in nuclear-verdict cases.” Id. at 51–54 (Huddle, J., concurring in part and dissenting in part).

What it appears to mean. For a plaintiff holding a judgment against multiple defendants — related entities in particular — the decision seems to increase the security that must be posted before an appeal can proceed without collection, and to foreclose the use of a single capped bond to cover a family of companies. For defendants, it suggests that the way a business is structured carries consequences on appeal as well as at trial: separate entities get separate bond obligations. The 20-day cure period gives a defendant whose bond is challenged some room to respond. And the 5–4 split, together with the dissent’s direct appeal to the Legislature’s purpose, leaves open the possibility that the question will be revisited in Austin.

Rehearing was denied on September 18, 2026.

See also our overview of the tools for collecting a Texas judgment.

What Is a Trade Secret in Texas?

 A former employee took our customer list, our pricing, and some internal process documents to a competitor. Is that information a “trade secret” under Texas law?

This is the question every trade-secret case starts with, and it’s the one this blog first addressed in 2013. The law has changed since then — Texas adopted a trade-secret statute later that year and expanded it in 2017 — so this post replaces the earlier one.

The statute. Since September 1, 2013, trade secrets in Texas have been governed by the Texas Uniform Trade Secrets Act, found in Chapter 134A of the Civil Practice and Remedies Code. Before that, the rules came from court decisions. The statute largely tracks those decisions but puts the definition in one place.

As amended in 2017, the Act defines a trade secret as “all forms and types of information, including business, scientific, technical, economic, or engineering information, and any formula, design, prototype, pattern, plan, compilation, program device, program, code, device, method, technique, process, procedure, financial data, or list of actual or potential customers or suppliers,” regardless of how it is stored, provided two things are true:

• the owner “has taken reasonable measures under the circumstances to keep the information secret,” and
• the information “derives independent economic value, actual or potential, from not being generally known to, and not being readily ascertainable through proper means by, another person who can obtain economic value from the disclosure or use of the information.”

Tex. Civ. Prac. & Rem. Code § 134A.002(6).

Breaking that down. The definition is broad on what kind of information qualifies — business and financial information count, not just technical formulas — and the 2017 amendment made that explicit by adding “business, scientific, technical, economic, or engineering information” and the phrase “list of actual or potential customers or suppliers.” What narrows it are the two conditions.

Reasonable secrecy measures. The owner has to have treated the information as secret. Courts look at things like whether access was limited, whether employees and vendors signed confidentiality agreements, whether documents were marked, and whether the information was kept off public channels. Absolute secrecy isn’t required; the standard is reasonableness “under the circumstances.” A company that posts its pricing on its website, or emails its customer list to anyone who asks, is unlikely to meet it.

Independent economic value from secrecy. The information has to be valuable because competitors don’t have it and can’t easily get it through legitimate means. Information that a competitor could assemble from public sources, or figure out by reverse-engineering a product that is sold on the open market, generally won’t qualify.

The six factors. Before the statute, Texas courts weighed six factors to decide whether information was a trade secret: how widely the information was known outside the business; how widely it was known inside the business; what measures were taken to guard it; its value to the business and to competitors; the effort or money spent developing it; and how easily it could be properly acquired or duplicated by others. In re Bass, 113 S.W.3d 735, 739 (Tex. 2003). The statute doesn’t list those factors, but courts applying it continue to consult them, and they map fairly closely onto the two statutory conditions.

Some examples. Information that Texas courts have treated as a trade secret, depending on the facts, includes customer lists that were compiled with effort and kept confidential; pricing and margin data; manufacturing processes; formulas; software source code; and compilations of otherwise-public information arranged in a way that isn’t public. Information that generally does not qualify includes anything already public, an employee’s general knowledge and skill, and information the company itself didn’t bother to protect. A unique combination of known components can be a trade secret even if each piece is public on its own.

What the statute does with a trade secret. The Act defines “misappropriation” — acquiring a trade secret by improper means, or using or disclosing it when you knew or should have known it was obtained improperly or under a duty of confidence — and “improper means,” which include theft, bribery, misrepresentation, breach of a confidentiality duty, and espionage. § 134A.002(2)–(3). It provides for injunctions, damages, and in some cases attorney’s fees, and it includes a presumption in favor of protective orders during litigation. §§ 134A.003–.006. One limit worth noting: an injunction under the Act may not prohibit a person from using the “general knowledge, skill, and experience” acquired during employment. § 134A.003(a).

The federal overlay. Since 2016 there has also been a federal statute, the Defend Trade Secrets Act, 18 U.S.C. § 1836 et seq., with a definition that closely resembles the Texas one. It gives trade-secret owners the option of suing in federal court and, in narrow circumstances, obtaining an order for seizure of the misappropriated material. The two statutes coexist; many Texas trade-secret cases are filed under both.

A note on the 2013 post. The original version of this post described the common-law six-factor test as the governing standard. That was accurate when written. The statute now controls, and the earlier post has been retired in favor of this one.

For related topics, see our earlier posts on customer lists as trade secrets, remedies for misappropriation, the statute of limitations, and unique combinations of known components.