Justice WAINWRIGHT delivered the opinion of the Court

George S. Christian, Austin, TX, for Amicus Curiae Texas Civil Justice League. Christopher J. Pruitt, Brown Pruitt Peterson & Wambsganss PC, Fort Worth, TX, for Other interested party Boyd, Larry.
Dryden Liddle, San Rafael, CA, for Other interested party LawFinance Group, Inc.

David N. Kitner, Strasburger & Price, L.L.P., Dallas, TX, Creighton R. Magid, Dorsey & Whitney LLP, Washington, DC, for Other interested party U-Haul Co. of Arizona.

Christopher J. Pruitt, Brown Pruitt Peterson & Wambsganss PC, David E. Keltner, Kelly Hart & Hallman LLP, James Estes Griffis, Brown Pruitt Peterson & Wambsganss, Fort Worth, TX, Creighton R. Magid, Dorsey & Whitney LLP, Washington, DC, David N. Kitner, Strasburger & Price, L.L.P., Thomas S. Leatherbury, Vinson & Elkins LLP, Dallas, TX, Lisa Bowlin Hobbs, Kuhn Hobbs PLLC, Matthew Ploeger, Vinson & Elkins L.L.P., Austin, TX, for Petitioners U–Haul International, Inc.

Charles E. Schuerenberg, Schuerenberg & Grimes, Ted B. Lyon, Jr., Marquette William Wolf, Ted Lyon & Associates, P.C., Mesquite, TX, Charles W. McGarry, Law Office of Charles McGarry, Jeffrey S. Levinger, Levinger PC, Joseph Carl Cecere, Hankinson LLP, Dallas, TX, for Respondents Waldrip, Talmadge.

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

After being seriously injured in an accident by a U–Haul truck, Talmadge Waldrip

sued several U–Haul corporate entities and an independent dealer on theories of negligence and gross negligence, alleging that the accident was the result of a systematic pattern of poor inspection and repair practices, and incompetence.

Following a three-week trial, a jury found U–Haul International, Incorporated d/b/a U–Haul (UHI) and U–Haul Company of Texas, Incorporated d/b/a U–Haul Company of Dallas (UHT) both negligent and grossly negligent and East Fork Enterprises, Incorporated d/b/a Jot 'Em Down (JED) negligent (collectively, U–Haul). The trial court awarded approximately $45 million in damages, including more than $23 million in punitive damages against UHI and UHT.

322 S.W.3d at 829, 839. The court of appeals reversed the trial court's exemplary damages award against UHI and affirmed the trial court's judgment in all other respects. At issue in this case is whether Waldrip presented legally sufficient evidence at trial to support the actual and punitive damages awards, and whether the court of appeals erred in overturning the gross-negligence award against UHI. We reverse the court of appeals' judgment in part and affirm in part and remand the case for a new trial on the negligence claims against the defendants.

I. Background and Procedural Posture

Talmadge Waldrip was severely injured while exiting a U–Haul truck that rolled backwards onto him, allegedly due to problems with the truck's parking brake and transmission. U–Haul argues that its maintenance and safety procedures were adequate and the problems with the truck were not the result of faulty inspection or maintenance. Waldrip contends that U–Haul's failure to discover the problems with the truck's parking brake and transmission systems was due to negligence and gross negligence in its inspection and maintenance program.

The truck at issue in this case, identified as JH6097T, was a standard transmission jumbo hauler. The vehicle could roll backwards when idle unless the transmission was in gear or the parking brake was applied, or both. Waldrip testified that the truck rolled, despite his putting it in first gear and applying the parking brake. The jury found for Waldrip on all grounds and awarded actual damages of $21,425,000, of which it apportioned 50 percent to UHI, 49 percent to UHT, and 1 percent to JED. The jury also found UHI and UHT grossly negligent and awarded punitive damages of $42 million against UHI and $21 million against UHT. The plaintiffs did not submit a question on the gross negligence of JED in their proposed jury charge. After applying the statutory cap on punitive damages, the trial court entered judgment against UHI and UHT, ordering each to pay $11,760,000 in punitive damages as well as its share of the actual damages award. The court of appeals overturned the exemplary damages award against UHI, but otherwise affirmed the trial court's judgment, including actual damages. U–Haul filed a petition for review in this Court.

II. Facts

A. U–Haul Organization and Maintenance and Inspection Program

UHI is an equipment rental company based in Phoenix, Arizona with 49 subsidiaries operating in the United States and Canada. These subsidiaries, including UHT, operate U–Haul-owned rental centers, as well as oversee the operations of independent dealers, including JED, that rent U–Haul equipment. Of the more than 15,000 U–Haul rental locations across the United States and Canada, 14,000 of those locations are independent dealers. At the time of trial, U–Haul's fleet comprised more than 100,000 trucks.

As U–Haul's vehicle-inspection and maintenance policies for the fleet's parking brakes and transmissions are at issue, we will briefly review them. UHI is responsible for developing and maintaining the company's policies for inspecting, repairing, and maintaining the vehicles. UHT is responsible for enforcing and implementing UHI's policies in Texas. UHI has promulgated maintenance and repair policies relevant to this case. First, the preventive maintenance (PM–5) inspection is scheduled for every 5,000 miles and includes a functional and visual parking brake inspection, an inspection of the transmission-fluid level, a check for signs of leakage, and a test drive of the truck by another person to include a check of the parking brake. A PM–5 inspection is performed by a trained mechanic.

Also relevant to this case is the vehicle-safety certification required by UHI in addition to the PM–5. UHI requires that all trucks undergo a safety certification at specified chronological intervals. The required frequency of these certifications is in dispute.

Safety certifications include a functional and visual parking brake test.

Area Field Managers (AFMs) were employed by UHT and other UHI subsidiaries and charged with performing safety inspections on the trucks at independent dealers. The record indicates that AFMs are not mechanics and perform operational functions distinct from U–Haul's mechanics.

A third relevant UHI policy is the receipt and dispatch tag program (R & D tag). With every rental, U–Haul records customer feedback on the truck's performance using the R & D tag to collect information. When a customer returns a truck to either an independent dealer or company-owned location, U–Haul agents

are required to ask the customer about any problems with the truck, including brake, engine, or electrical problems, and the agent records this information on the R & D tag, which is a UHI form. The agent must sign the tag to certify that the truck is ready for rental.

UHI also requires an annual inspection per the United States Department of Transportation (DOT) guidelines. The parties dispute, however, whether federal law requires U–Haul to conduct this inspection.

If any safety issues are discovered during an inspection or the R & D-tag process that cannot be repaired on-site, U–Haul policy requires that the truck be grounded and the information entered into the company computer system in the Downed Equipment Tracker (DET) database until the repairs are completed. UHI maintained this computer database for all of the trucks in U–Haul's fleet.

B. History of the Truck Waldrip Rented (JH6097T Jumbo Hauler)

Talmadge Waldrip was injured in September 2006 when the parking brake or the transmission on the truck his daughter rented malfunctioned and the truck rolled over him as he attempted to exit it. At the time of the accident, JH6097T was an eighteen-year-old U–Haul jumbo hauler with a manual transmission and an odometer reading of more than 233,000 miles. At the time of the accident, JH6097T had an inoperable parking brake and a damaged transmission. The parties disagree on the extent and cause of these conditions, and whether they caused the accident.

In the years before Waldrip's accident, JH6097T had been driven throughout North America. When the accident happened, the truck was on its third transmission. The rear transmission seal had been replaced three times since 2001. The entireparking brake system had been replaced as well, but evidence presented by Waldrip at trial indicated that parking brake problems persisted.

In September 2005, a year before Waldrip's accident, the truck was grounded and placed in the DET database after a customer reported that the parking brake light stayed on and the truck had “no parking brake,” according to language entered in the database. The parties dispute exactly what “no parking brake” means. U–Haul contends that the notation “no parking brake” meant only that the indicator light was not working, a conclusion it argues is buttressed by the truck passing subsequent parking-brake inspections. Waldrip asserts that “no parking brake” meant that the parking brake itself was not functioning. U–Haul did not perform repairs on the parking brake at that time. James Anderson, the U–Haul employee who repaired the parking-brake light, testified that he inspected the parking brake and found it to be functioning.

While in Dallas prior to the accident, the truck passed safety certifications three times—in November 2005, January 2006, and March 2006. The November 2005 safety certification was performed by Reginald Benjamin, a UHT employee. The latter two certifications were performed by Jason Crews, an AFM hired by Lynn Buck, UHT's Marketing Company President in Dallas. Crews had no prior experience in maintaining, inspecting, or repairing trucks when he was hired. Buck, aware of Crews' lack of experience, testified that she believed mechanical experience was not necessary for the role because AFMs are not responsible for repairs and are merely responsible for identifying truck problems during safety certifications. Buck also testified that U–Haul provided the necessary on-the-job training for AFMs. Crews underwent two to three weeks of training, including one day in a “shop” in which he went through a typical day of work. However, Waldrip presented some evidence that, according to U–Haul's policies, Crews should have spent more than one day in a “shop.”

The JH6097T was transferred by UHT to JED in June 2006 and Crews remained responsible for safety certifying the truck. Crews certified the truck as safe again in July 2006. At trial, Crews did not specifically remember working on JH6097T, but stated he would not have placed the safety certification sticker on it if the parking brake had not passed inspection. Crews did not check the fluid levels of JH6097T during safety certifications because he believed that he was not required to do so on standard-transmission trucks.

At trial, Crews indicated that he did not know that it was possible to check the transmission-fluid levels on standard-transmission vehicles, and even testified that a standard transmission did not contain transmission fluid. He later acknowledged that the fluid could be checked in a standard transmission.

1. Customers who had parking brake and transmission problems with the JH6097T truck and reported those problems to U–Haul

Several people who rented and drove JH6097T in the year before Waldrip's accident testified at trial that they did or did not have certain problems while driving the truck. Because U–Haul's knowledge of these problems is at issue, we separate those customers who had problems and testified they reported them to U–Haul from those customers who either did not have problems with the truck or did not report problems.

We discuss first those customers who reported issues with the truck's parking brake or transmission. In October 2005, JH6097T passed a PM–5 inspection in Florida.

After this PM–5 inspection, Ignacio Reveles rented and drove JH6097T from Florida to Texas. Reveles testified that the transmission was “not in good condition,” that he had trouble starting the truck and shifting gears, and that the truck rolled backwards with the parking brake set while it was in gear. In order to keep the truck from rolling again, he placed wooden blocks behind the back tires of the truck for the duration of his trip. Reveles testified that he reported these problems to the U–Haul agent in Florida and was told to call U–Haul's 1–800 number. When he returned the truck to a Dallas U–Haul center, he said he told the agent that the parking brake did not work, he had problems shifting gears, and the truck leaked. The R & D-tag does not record any of the problems reported by Reveles and records of the toll-free call indicate only that Reveles reported a dead battery.

Steve Marco rented the truck on September 1, 2006. Unlike other renters of JH6097T, he had driven large standard transmission trucks. Marco testified to many problems with the truck, including difficulty shifting gears and lack of tension in the parking-brake lever. Because of this, he took precautions to make sure the truck did not roll, including turning the wheels and parking against a curb. Marco says he reported these problems, but the R & D tag for that rental does not reflect any reported issues.

Jonathan Simington, Waldrip's grandson, rented the truck on September 9, 2006.

Simington also had experience driving manual trucks. After Simington set the parking brake and put it into first gear, the truck rolled down a building ramp. He also reported that the transmission jumped out of gear while he was driving. Larry Boyd, Simington's uncle, was working at JED at the time Simington returned the truck and was 10 to 15 feet from Simington with his back turned. Simington reported these problems to Boyd, although he acknowledges that Boyd may not have heard him. Simington did not report any of these problems to Waldrip or any other U–Haul entity.

2. Customers who did not report parking brake and transmission problems with the truck to U–Haul

In September 2005, after the U–Haul jumbo hauler was placed in the DET database and the parking-brake light was fixed, Pablo Cassanova rented the truck in Florida and reported that the truck was in good working condition. He parked the truck overnight on an incline and did not put the transmission in gear, but he did set the parking brake and the truck did not move.

Derrick Bradley was the first customer to rent JH6097T after the July 2006 safety certification. He testified that he had difficulty getting the truck from second to third gear. JED was closed when he returned the truck, so he parked the truck near the entrance gate and set the parking brake. Bradley observed it roll backwards about a foot until it hit the gate.

Bradley says he called JED the next day to ask about the gate, but did not notify anyone of the problems he had with the truck.

Bree Adams rented the truck from JED in July 2006 after Bradley returned the truck. She did not have any transmission problems and the truck did not move when the parking brake was engaged, although Adams acknowledged that she parked on a generally flat surface.

Jeff Everett rented the truck from JED in August 2006. He parked on both flat and inclined surfaces, put the truck in gear, and set the parking brake. The truck did not move. Everett did not know whether the transmission gears or the parking brake held the truck in place. He also acknowledged that he had “no idea” if the parking brake was working because he did not test it.

Clay Vestal rented the truck on August 17, 2006, and had problems getting the truck into gear. The truck rolled down his driveway, among other unrelated problems. He did not report any of these problems to any U–Haul employees.

3. Waldrip's Accident

Waldrip's daughter, Anne Waldrip–Boyd, rented JH6097T from JED on September 20, 2006. Anne asked Waldrip to drive the truck because she could not operate a standard transmission. Waldrip drove the truck about seven miles to a warehouse in Forney, Texas. Waldrip testified that when he arrived at the warehouse, which was located on a slight slope, he turned off the ignition, put the truck in first gear, and set the parking brake. For the truck to have rolled backwards with the transmission in first gear, both the parking brake and the transmission must have malfunctioned. As he stepped out of the truck, it began rolling backwards, with the truck door hitting him in the back and knocking him to the ground. The truck then rolled over Waldrip and dragged him down the slope approximately sixty feet.

C. Expert Testimony on JH6097T's Condition

1. Waldrip's Experts

Waldrip sued U–Haul, alleging negligence and gross negligence. Extensive evidence on the state of JH6097T's parking-brake system and transmission was presented at trial. Waldrip presented Randy Reed, a mechanic with over twenty-three years of experience working with large trucks and specifically with U–Haul's jumbo hauler design. Reed inspected the truck's parking-brake system and transmission two days after the accident and again months later. He testified that he found damage to both. Reed testified that the parking-brake drum, brake shoes, and drive shaft were coated in oil, which led to a lack of friction and made the braking system ineffective. Reed had inspected hundreds of parking-brake systems on similar trucks and had seen “a lot of damaged components,” but had never seen “one of this factor, not with that kind of grease on it.”

Reed also stated that half of the transmission's six-quart capacity had leaked though the rear-transmission seal. The seal featured numerous indentations around its circumference, leading Reed to conclude that it had been installed with an improper tool, causing it to leak.

Reed opined that the transmission fluid leaked onto the parking-brake assembly, accelerating the brake shoes' wear and causing them to separate from the metal backing plates. A correctly maintained and adjusted parking brake locks the drive shaft so the truck will hold when the brake is engaged. However, if oil or any petroleum product gets on the brake shoes, the shoes must be replaced and cannot simply be cleaned. According to Reed, parking-brake failure was a direct result of the condition of the brake shoes and drum. After considering the witnesses' testimony regarding the parking-brake failures, Reed concluded the parking brake was not functioning by some point in 2005 and, more particularly, was not functional in October 2005 after the last PM–5 inspection, or after the last safety certification in July 2006.

Reed also believed that low transmission-fluid levels damaged the first and reverse gears. Reed said the damage created a “false gear” capable of deceiving drivers into believing the truck was in first gear when in fact it was not in any gear. If a driver was able to actually get the truck into gear, the transmission alone would hold the truck in place, regardless of whether the parking brake worked. If the truck was actually in the “fake out gear,” however, the gear would not hold the truck. Reed believed the rear transmission seal had been leaking “for a while,” given the amount of fluid in the transmission.

Reed rejected U–Haul's theory that the transmission seal leak was caused by someone driving with the parking brake engaged, stating that the physical evidence did not fit that theory because the necessary components did not overheat. Significantly, Reed claimed that with proper maintenance and inspection, the faulty condition of both the parking brake and the transmission would have been detected. Reed did not believe that the reported inspections on the truck were performed correctly and said maintenance of the truck was “for lack of better terms not—not there.”

Waldrip also presented Dr. Kurt Marshek, a Ph.D. in mechanical engineering, to critique U–Haul's policies and practices regarding its vehicles' safety. Marshek noted that “it probably doesn't even make sense to have the truck on the road anymore with 225,000 miles on it or 228,000 miles, especially if it's having all sorts of mechanical problems, which this particular truck was having.” Marshek also testified that DOT inspections were required annually, but observed that the last DOT inspection on JH6097T occurred in May 2005, more than one year prior to the accident. Marshek rejected U–Haul's argument that conducting PM–5 inspections relieved U–Haul of its duty to perform the federal inspection, noting that the latter is a time-based inspection.

Based on the various witnesses' testimony that the parking brake failed after the last PM–5 inspection, including Reveles and Bradley, Marshek told the jury that if a proper DOT inspection had been performed before September 2006, the parking-brake failure would have been detected.He observed further that U–Haul's failure to perform that inspection was a proximate cause of the accident. Based on his review of U–Haul documents and the testimony of James Guinn, a former U–Haul center general manager, Marshek concluded that safety certifications were required on trucks such as JH6097T every thirty days, but that JH6097T's safety certifications did not adhere to such a schedule. Marshek reasoned that, had safety certifications been correctly administered every thirty days in at least the three months before the accident, the parking-brake failure would have been detected. Marshek also believed that the last safety certification—in July 2006—was not properly performed, based on Bradley's testimony that the truck rolled when the parking brake was applied.

2. U–Haul's Experts

U–Haul's experts examined the truck and agreed with Waldrip's experts that the parking brake was inoperable and the transmission was damaged when the accident occurred. However, they disagreed on the cause and the extent of this damage, as well as when it occurred. U–Haul presented Thomas Green, who holds a bachelor's degree in mechanical engineering and is a consultant in accident reconstruction, cause, and analysis, as well as vehicle-systems analysis. After examining the truck's transmission after the plaintiffs had inspected and dismantled it, Green observed damage to the reverse gear teeth, but noted that all other gears “looked to be in good shape” and had no abnormal signs of wear. Green concluded that the damage was caused by one or more drivers trying to shift into reverse while the vehicle was moving forward. While the damage could cause problems shifting into reverse, it would not cause problems shifting into first gear. Green also argued that there was no “absence of lubrication” that would “cause any problem.”

Green agreed with Reed that the oily substance coating the parking brake came from a leak in the rear transmission seal. He said the leak could have been caused by dirt or debris getting into the seal, improper installation of the seal, or heat generated by driving with the parking brake engaged. He did not believe the seal was improperly installed (although he agreed an improper tool may have been used), leaving heat or debris as the likely causes of the leak. Because the fluid drained from the transmission contained metal, he believed the seal began leaking after July 2005, when the truck's transmission fluid was drained and changed. Green acknowledged that the leak could have been discovered either by checking for leaks or checking the transmission-fluid levels. But, he believed that the transmission had been properly maintained, explaining that the gears were not abnormally worn and were adequately lubricated and that U–Haul records showed the transmission fluid was changed and checked when required.

Ken Sorenson, a Ph.D. in mechanical engineering, inspected the truck's parking-brake system for U–Haul after the plaintiffs had already done so. The system was contaminated with oil and debris and the brake shoes were worn down to the metal plate. He concluded the truck had been driven with the parking brake applied, wearing down the brake shoes.

He also inspected the transmission and, like Green, saw that the spline teeth on the reverse gear were deformed. Other than that damage, he said the rest of the gear teeth looked “okay.”

D. Canadian Evidence

Over objection to his testimony, Waldrip presented evidence about investigations of U–Haul's safety practices in Canada. Brian Patterson testified as a lay witness. Patterson was president of the Ontario Safety League (OSL),

a private entity that inspected U–Haul trucks coming into Canada from the United States in 2005 and 2006 and reached conclusions about U–Haul's routine safety. Patterson testified that he personally observed the inspections of approximately thirty U–Haul vehicles by licensed mechanics, and the evidence he reviewed “indicated a systemic disregard for public safety in the maintenance of vehicles in the Province of Ontario.” Patterson personally recalled only three vehicles with inoperable parking brakes. He also testified to other problems including carbon-monoxide poisoning, leaking brake lines, leaking transmission fluid, broken power steering hoses, bald tires, faulty suspensions, broken brake drums, faulty shocks, and rotted truck floors. As a result of the initial investigation, Canadian governmental entities initiated investigations of U–Haul and, as part of those investigations, approximately 1,400 vehicles were inspected. Brake issues were studied separately in British Columbia, where fifty percent of the trucks tested had faulty brakes. Patterson stated that the trucks with faulty brakes “would not have been allowed on the road.” Patterson offered no documentation in support of his testimony. Patterson explained that the inspection reports were retained by the garage, which was defunct at the time of trial.

Patterson also testified that he dealt directly with top U–Haul officials in Canada, one of whom was Claude Bouchet. According to Patterson, Bouchet told him that, “particularly in 2005 he had absolute difficulty getting any action from ... U–Haul International, and in fact, he had no real say over being able to correct the problems that [had] been identified in Canada.” Patterson also disagreed with statements made by UHI's board chairman, Joe Shoen, that U–Haul's safety policies and procedures were a “10 of 10.”

E. U–Haul's Safety Procedures

Shoen testified that of the 100,000 trucks in U–Haul's fleet, more than 4,500 were standard-transmission jumbo haulers with mileage exceeding 200,000. He was not concerned about the number of higher-mileage trucks still on the roads because the trucks were built for a 300,000–mile service life. Defending U–Haul's safety procedures, Shoen testified that UHI created an inspection-maintenance system allowing every participant in the system to stop a truck rental if a problem were revealed. The system operates on the presumption that the truck “doesn't rent” until someone in the system makes a “positive move.”

Shoen described U–Haul's safety-inspection process as a tiered, multi-step progression. The process begins with the R & D-tag, used on each rental and requiring the agent to go through eight steps before allowing the truck to be rented. The next level is the periodic safety certification. Shoen explained that U–Haul–owned stores rent three times more frequently than independent dealers; consequently, equipment at the dealers is inspected three times less frequently, although Shoen said it “ends up being about the same interval based on numbers of customers through the vehicle.” The third level of UHI's program is the mileage-based preventive maintenance inspections, such as the PM–5 inspection. Additionally, Shoen testified that although the company is not subject to the federal regulations relating to the DOT inspection, it voluntarily performs annual DOT inspections on its trucks.

Shoen stated that UHI's system is “as good or better” than its “truck rental peers.” His opinion was that “U–Haul is rated a 10 in safety.” He agreed that a truck with an inoperable parking brake would not be “a 10” and such a condition would, in fact, be a “material shortcoming” of the truck. Shoen testified that an inoperative brake would be the fault of “[e]verybody probably involved in the U–Haul system ... from myself down through and including the people at the actual rental location.” He disputed any suggestion that it was somehow to U–Haul's advantage to delay and avoid maintenance, explaining that a lack of maintenance could pose “worse financial consequences than the cost of the maintenance....”

Shoen believed Crews properly performed all safety certifications on the truck in the months before the accident because Crews signed off on the inspections, saying “if [Crews] certified that it worked, then it worked.” If the parking brake did not work the next day when the truck was rented to the first customer, he believed something must have happened in the intervening time to have caused the problem.

Lynn Buck testified that she knew that Crews had no mechanical experience when she hired him, but that mechanical experience was not a qualification she was looking for when hiring an AFM because they received training, would only perform “minor maintenance checks,” and more qualified people were brought in if a repair were needed.

F. Procedural Disposition

The jury found UHI, UHT, and JED negligent and the proximate cause of Waldrip's injuries, apportioning liability among them.

It also found both UHI and UHT grossly negligent and awarded Waldrip exemplary damages against both of them. The plaintiffs did not submit a question on the gross negligence of JED in their proposedjury charge. The trial court entered judgment on the verdict, with some modifications, and U–Haul appealed.

UHI and UHT challenged the legal and factual sufficiency of the jury's gross-negligence findings against them. The court of appeals overturned the gross-negligence finding against UHI, holding that a reasonable juror could not form a firm conviction that UHI acted grossly negligent, mainly because Waldrip presented no evidence of Thomas Coffee's capacity with UHI and because no evidence showed that UHI had any knowledge of problems with JH6097T. Coffee, UHI's Director of Repair Analysis and Support, was employed to maintain UHI's computer records, including the DET. U–Haul also alleged that the trial court abused its discretion by allowing Patterson's testimony, arguing that the evidence was irrelevant, prejudicial hearsay tainting the entire trial. The court of appeals held that the trial court did not abuse its discretion in allowing this evidence. 322 S.W.3d at 852.

The court of appeals affirmed the gross-negligence award against UHT, based on Buck's hiring of Crews. Id. at 848. The court held that the jury could have formed a firm conviction that Crews was unfit and Buck was reckless in hiring him, which together caused Waldrip's injuries. Id. The court also affirmed the negligence findings against UHI, UHT, and JED. Id. at 842–43. In this Court, U–Haul argues that the court of appeals erred in affirming the gross-negligence award against UHT and affirming the negligence awards against all of the U–Haul entities. U–Haul also argues that the court of appeals erred in concluding that the trial court did not abuse its discretion by admitting Patterson's testimony.

III. Discussion

A. Canadian Evidence

The trial court allowed Patterson's testimony on the investigations by OSL, a safety advocacy group in Canada, and other Canadian entities about the safety of U–Haul's vehicles. The Waldrips argued that the evidence showed 1) that the “incidence and type of safety problems with the trucks inspected in Canada” were indicative of the “condition of the entire fleet, including the truck that injured Waldrip,” and 2) UHI's conscious indifference to the “system-wide failures of its maintenance and inspection program....” The court of appeals affirmed the trial court's admission of the Canadian evidence, holding that our opinion in Nissan Motor Co., Ltd. v. Armstrong, 145 S.W.3d 131 (Tex.2004), did not mandate exclusion of this evidence and, even if the trial court's ruling was in error, the error probably did not result in an improper judgment. 322 S.W.3d at 848–49. The court of appeals allowed Patterson to testify but interpreted the trial court as having granted U–Haul a running objection. The court was unpersuaded by U–Haul's argument that admission of the evidence violated its due process rights.

U–Haul argues that its objections were not waived because it timely and properly raised objections to the testimony. Although it appears U–Haul “clearly identified the source and specific subject matter of the expected objectionable evidence prior to its disclosure to the jury,” objecting to Patterson's testimony on the Canadian evidence, the trial court declined to grant a running objection. See Volkswagen of Am., Inc. v. Ramirez, 159 S.W.3d 897, 907 (Tex.2004) (specifying the requirements for a running objection). Instead, after a hearing on the matter, the judge ruled:

The Court's going to overrule the objections and allow Mr. Patterson to testify ... [a]nd let the record show that I'm allowing [U–Haul counsel] to insert these objections when actually Mr. Patterson is being called as opposed to excusing the jury and having to go through all of these objections again.
As required by the trial court, U–Haul objected on the grounds of relevance, hearsay, and lack of similarity a total of 12 times throughout Waldrip's direct examination of Patterson. U–Haul timely objected to the Canadian evidence, and we will consider U–Haul's challenges to Patterson's testimony. Some of the more pertinent objections are noted in the discussion.

Evidentiary rulings are committed to the trial court's sound discretion. Bay Area Healthcare Grp., Ltd. v. McShane, 239 S.W.3d 231, 234 (Tex.2007) (per curiam). A trial court abuses this discretion when it acts without regard for guiding rules or principles. Owens–Corning Fiberglas Corp. v. Malone, 972 S.W.2d 35, 43 (Tex.1998). Even if the trial court abused its discretion in admitting certain evidence, reversal is only appropriate if the error was harmful, i.e., it probably resulted in an improper judgment. See Nissan, 145 S.W.3d at 144;Tex.R.App. P. 44.1, 61.1.

Waldrip argues that if we consider objections to this evidence, we should only do so in relation to UHI, as UHT and JED waived objections to this evidence by their failure to request a limiting instruction under Texas Rule of Evidence 105. As U–Haul correctly argues, however, Rule 105 does not apply when the evidence in question is not admissible against any party for any purpose. SeeTex.R. Evid. 105(a) (advising that the court, upon request, should restrict evidence to its proper scope if evidence is admissible as to one party or for one purpose). Additionally, this evidence was presented to the jury without specifying against whom it was offered. Notwithstanding Waldrip's assertion that the evidence was only offered against UHI, U–Haul argues that it was used against and prejudiced all defendants. We agree that the questioning and Patterson's answers were not limited to UHI. Waldrip's counsel alluded to “U–Haul” safety problems throughout the questioning, and the jury was never advised by counsel or the court that it was only to consider the evidence against UHI.

1. Abuse of Discretion

U–Haul's principal argument is that Patterson's testimony did not meet our requirement that trial courts “carefully consider the bounds of similarity, prejudice, confusion, and sequence” in deciding whether to allow Waldrip to present Patterson's testimony to the jury. See Nissan, 145 S.W.3d at 139. In Nissan, Armstrong sued Nissan for injuries arising from a car accident caused by unintended acceleration. 145 S.W.3d at 134–35. At trial, Armstrong presented evidence of a Nissan-maintained database of 757 complaints of unintended acceleration from customers driving the same model vehicle as Armstrong's. Id. at 136, 140–41. The jury returned a verdict in Armstrong's favor and the court of appeals affirmed. Id. at 136.

We reversed the court of appeals and ordered a new trial because the database evidence was not relevant and constituted hearsay, explaining that none of the reports attributed the unintended acceleration to the throttle cable, which was the plaintiff's theory of her case. Id. at 141, 149. We concluded that the complaints were not sufficiently similar to the defect Armstrong alleged to be admissible. Id. at 141;see also Mo. Pac. R.R. Co. v. Cooper, 563 S.W.2d 233, 236 (Tex.1978) (“The plaintiffs were required to show that the earlier accidents occurred under reasonably similar but not necessarily identicalcircumstances.”) (citations omitted). Moreover, all of the reports were hearsay. Nissan, 145 S.W.3d at 140.

In Serv. Corp. Int'l v. Guerra, we applied Nissan 's reasoning to plaintiffs' case against a corporation that operated a cemetery (SCI Texas) and its parent corporation (SCI International) for alleged mental anguish caused by moving the decedent's body from one burial plot to another without the family's permission. 348 S.W.3d at 226. The trial court admitted evidence of other lawsuits, verdicts, and judgments against SCI International. Id. at 233. The SCI entities appealed, arguing that the evidence was irrelevant. Id. at 227, 233–34. We explained that evidence of other wrongful acts is not admissible to show “action in conformity therewith.” Id. at 235 (citing Tex.R. Evid. 404). It is admissible to show a party's material intent if the prior acts were “so connected with the transaction at issue that they may all be parts of a system, scheme or plan.” Id. at 235 (quoting Oakwood Mobile Homes, Inc. v. Cabler, 73 S.W.3d 363, 375 (Tex.App.—El Paso 2002, pet. denied)). The plaintiffs offered this evidence on a theory similar to that of the Waldrips in this case, arguing that the evidence was “relevant to show a pattern of indifference amounting to a common scheme” and that SCI “took no action to avoid recurrences of misconduct.” Id. at 236. We rejected the argument in Guerra because the plaintiffs failed to show a sufficient connection between the evidence offered and the events at issue, outside of the broad characterization of the incidents showing a “pattern of indifference.” Id.

Although the issue in Guerra was the relocation of a body without permission, only some of the evidence offered by the plaintiffs detailed similar behavior. Id. Waldrip's evidence suffers from the same pitfall, as very little of the Canadian evidence presented dealt with faulty parking brakes and none of it specifically addressed transmission issues. Rejecting in Guerra admission of other allegedly similar evidence of relocating a body, we explained that:

The events occurred at different cemeteries and there was no evidence that any of the same employees were involved or that they occurred under similar circumstances. The events also occurred more than a year apart. There is no evidence that the events were part of a system, scheme, or plan.
Id. Even under the guise of a similar “pattern of indifference” argument, Patterson's testimony suffers from similar pitfalls.

Patterson testified at length about a wide range of problems allegedly discovered in Canada when the OSL conducted an investigation into the “routine practices of U–Haul International and U–Haul Arizona.” He also testified that he personally saw the condition of about thirty of the fourteen hundred U–Haul trucks that were inspected in Ontario by licensed mechanics. The thirty trucks he observed being inspected in Canada were owned by U–Haul Arizona. Notwithstanding relevance and hearsay objections, which were overruled, Patterson concluded that U–Haul had a “systematic disregard for public safety in the maintenance of vehicles in the Province of Ontario.” After the trial court overruled an objection to relevance, he further testified that, of the fourteen hundred trucks tested, “brake issues” accounted for fifty percent of the problems. He specified that three of the vehicles had “emergency brake” (parking brake) issues. He also testified, notwithstanding objections that this testimony was irrelevant and prejudicial, that there were carbon monoxide poisoning issues in some U–Haul truck cabs because carbon monoxide could seep in through leaky floors. Without producing the government reports, he further testified that there were Canadian government studies of fourteen hundred trucks and that fifty percent of the vehicles tested had faulty brakes and would not have been allowed on the road in British Columbia. The hearsay objection to that testimony was overruled.

Like the evidence presented in Nissan and Guerra, Patterson's testimony is not sufficiently similar to JH6097T and distracted the jury from the relevant legal issues, for several reasons. See Guerra, 348 S.W.3d at 236;Nissan, 145 S.W.3d at 138. First, there is no evidence that the trucks tested in Ontario were of the same type or size, or even if their parking brake and transmission systems were the same or similar. There is no assertion or evidence that the vehicles inspected in Canada were jumbo haulers like the vehicle that injured Waldrip in Texas. U–Haul also takes issue with Patterson's inability to explain how the trucks in Canada were selected for inspection and whether the trucks were similar to JH6097T, were subject to the same or similar inspection and maintenance requirements as JH6097T, or if the deficiencies found in these trucks were attributable to problems in UHI's inspection and maintenance policies. No evidence showed whether the trucks had similar mileage to JH6097T.

The majority of the issues in Patterson's testimony are unrelated to problems with parking brakes or transmissions, the frequency with which the trucks were inspected, or how they were inspected. In fact, evidence presented in Nissan was arguably closer in similarity to the accident at issue in that case than Patterson's testimony is to the accident at issue in this case. All of the evidence in Nissan related to the same model automobile as the one driven by the plaintiff and involved the same alleged condition of unintended acceleration. See145 S.W.3d at 140–41. The proposition that the jumbo hauler was negligently maintained must be proven and cannot simply be inferred from even a large number of complaints unless the requisite similarity is established to make the prior incidents relevant, i.e., probative of the liability of U–Haul in this case. See id. at 142.

Waldrip also argues that Patterson's testimony was relevant to show that UHI, and specifically Shoen, had knowledge of a widespread disregard for its fleet maintenance and safety policies and that this resulted in deficiencies in fleet safety, including inadequate or inoperable parking brake or transmission systems. There was no attempt to identify the safety standards for the same or similar trucks in Canada or in the United States, and no opinion, if admissible, was offered on whether the trucks tested would have been allowed on the road in Texas or the United States. Our concerns about prejudice, confusion, relevance and delay are not alleviated simply because the evidence here was supposedly offered for a different reason than the evidence about product defects in Nissan. See id. at 138.

We do not pass judgment on the substance of Patterson's testimony, only on the predicate for its admissibility. All of the questions and shortcomings discussed above in the admissibility of his testimony do not necessarily have to be answered to properly admit it. However, Patterson's testimony falls well short of supporting a reasonable conclusion that safety problems in a foreign country indicated a disregard of inspection and maintenance programs causing an objectively extreme risk of serious injury in the United States; nor does it provide any reasonable basis to conclude that the accident in this case was attributable to a defective U–Haul safety program.

Shoen's testimony confirmed that Canadian regulations and standards were used in the investigations performed in Canada that Patterson identified. Without evidence on what those Canadian regulations and standards were and how they relate to those of the United States or Texas, this evidence is not probative of U–Haul's adherence to or failure to adhere to appropriate standards in the United States. Patterson did not testify, even if proper, that the policies or procedures in Texas or the United States were unsafe, he only offered his lay opinion that the trucks were not suitable for operation under unspecified standards in Canada. Waldrip did not carry his burden of showing the relevance of this evidence.

U–Haul contends it was an abuse of discretion to admit testimony of the contents of public records concerning the inspection results of fourteen hundred trucks in Canada and the braking and other problems allegedly found, when the actual reports were not themselves produced. If shown to be relevant, with a proper foundation, Patterson could conceivably surmount hearsay hurdles for the truck inspections he personally observed. That included thirty vehicles. He was not present for the inspection of the many other U–Haul trucks inspected. He had no personal knowledge of those inspections, but based his testimony on the government reports or studies that presumably contained the information. SeeTex.R. Evid. 602 (“A witness may not testify to a matter unless evidence is introduced sufficient to support a finding that the witness has personal knowledge of the matter.”). Patterson did not produce any of the public reports from which the findings he testified to originated. We have not addressed whether failing to produce or admit the public records renders inapplicable the hearsay exception of Rule 803(8) for public reports setting forth factual findings from an investigation made pursuant to lawful authority.

SeeTex.R. Evid. 803(8); cf. Benefield v. State, 266 S.W.3d 25, 34 (Tex.App.—Houston [1st Dist.] 2008, no pet.) (excluding public records for failure to properly authenticate them). The briefing does not directly address this issue of first impression and we reserve it for another day.

Waldrip also argues that this evidence was merely offered as “rebuttal” testimony in response to Shoen's comments about U–Haul's safety program being “as good or better” than its “truck rental peers” and that “U–Haul is rated a 10 in safety.” But Waldrip, not U–Haul, first offered Shoen's testimony in excerpts from Shoen's videotaped deposition. Waldrip chose to present Shoen's testimony to the jury, and U–Haul attempted to rebut its substance. The trial court abused its discretion by allowing Waldrip to present Shoen's testimony only to bootstrap Patterson's “rebuttal” testimony, when the evidence did not meet Nissan 's requirements.

2. Improper Judgment

Reversal of erroneously admitted evidence is warranted only if the error probably resulted in the rendition of an improper judgment. See Nissan, 145 S.W.3d at 144;Tex.R.App. P. 44. 1, 61.1. Although we have recognized the “impossibility of prescribing a specific test” for harmless-error review, McCraw v. Maris, 828 S.W.2d 756, 757 (Tex.1992), we must evaluate the entire case from voir dire to closing argument, considering the evidence as a whole, the strength or weakness of the case, and the verdict. Reliance Steel & Aluminum Co. v. Sevcik, 267 S.W.3d 867, 871 (Tex.2008) (citations omitted); Spohn Hosp. v. Mayer, 104 S.W.3d 878, 883–84 (Tex.2003). In determining whether the erroneous admission of evidence probably led to an improper judgment, courts look to the role the evidence played in the context of the trial and the efforts made by counsel to emphasize the erroneous evidence, as well as whether contrary evidence existed that the improperly admitted evidence was calculated to overcome. See Guerra, 348 S.W.3d at 236;Sevcik, 267 S.W.3d at 873–74;Nissan, 145 S.W.3d at 144. We believe that the erroneous admission of Patterson's testimony probably led to an improper judgment.

In affirming the trial court's decision to admit this evidence, the court of appeals focused on the page length of Patterson's testimony in relation to the voluminous trial record. See322 S.W.3d at 850. We have held, however, that even if improper evidence is only mentioned briefly, the nature of the evidence can have a profound impact on a trial. See Coastal Oil & Gas Corp. v. Garza Energy Trust, 268 S.W.3d 1, 25–26 (Tex.2008). Waldrip's counsel emphasized Patterson's testimony throughout the trial. Patterson was allowed to testify during direct examination that U–Haul's trucks had rotten floors, allowing carbon monoxide to seep into the passenger compartment and potentially causing carbon monoxide poisoning to the drivers and passengers, notwithstanding objections on the grounds of prejudice and irrelevance to the brake issue that was alleged to have caused the injuries. Waldrip's counsel argued in closing that U–Haul had “a rotten fleet.”

Waldrip's advocacy for inclusion of this testimony, over objection, suggests it was a significant element of this case. See Guerra, 348 S.W.3d at 237;Sevcik, 267 S.W.3d at 874 (“[A] party's insistence on introducing inadmissible testimony ‘indicates how important he thought it was to his case.’ ”) (quoting Alvarado v. Farah Mfg. Co., 830 S.W.2d 911, 917 (Tex.1992)). Further, the court of appeals' opinion bolsters our conclusion, as that court overturned the gross-negligence holding against UHI because the jury found UHI grossly negligent without properly admitted evidence supporting that finding. See, e.g., Guerra, 348 S.W.3d at 237 (“Because there was no evidence to support a finding of compensable mental anguish, the jury's [award of mental anguish damages] must have been based on something other than properly admitted evidence, and we have no doubt that the extensive evidence of other suits, allegations in the suits, and similar evidence was a significant factor in the jury's damages findings, both actual and punitive.”) (citing Sevcik, 267 S.W.3d at 872). We agree with the court of appeals that Waldrip did not carry his evidentiary burden in the gross-negligence holding against UHI, and that Patterson's testimony was likely important in the jury's finding against UHI on this issue.

It also appears there was legally sufficient evidence of U–Haul's negligence. The jury could have reasonably believed that the truck's parking brake was inoperable in September 2006 when the jumbo hauler was placed in the DET. UHI designed and maintained the system for repairs and maintenance and the computer database records for the entire fleet. The jury reasonably could have viewed JH6097T's multiple brake and transmission problems as sufficient to place UHI on notice through its DET system of the danger that the truck presented to renters. The jury also arguably had a reasonable basis for believing that 1) UHT had notice of the parking brake and transmission problems with JH6097T and 2) the safety certifications were negligently performed. Furthermore, a reasonable juror could have believed that Crews negligently performed his safety certifications. We conclude there was legally sufficient evidence for a reasonable juror to believe that JH6097T's parking brake and transmission problems had persisted for some time and that JED violated its duty by not discovering these problems or by continuing to rent JH6097T knowing the risk it posed to its customers. On the record of the prior trial, there appears to be some evidence to support the negligence findings and justify the remand for a new trial.

B. Gross Negligence

In reviewing an award for exemplary damages, we conduct a legal sufficiency review under the “clear and convincing” evidence standard. Garza, 164 S.W.3d at 609 (citations omitted). “ ‘Clear and convincing’ means the measure or degree of proof that will produce in the mind of the trier of fact a firm belief or conviction as to the truth of the allegations sought to be established.” Tex. Civ. Prac. & Rem.Code § 41.001(2); In re J.F.C., 96 S.W.3d 256, 264 (Tex.2002). Even if the inclusion of the Canadian evidence was not an abuse of discretion, this Court does not believe that Waldrip presented clear and convincing evidence that UHI or UHT was grossly negligent.

Gross negligence consists of both objective and subjective elements. See Lee Lewis Constr., Inc. v. Harrison, 70 S.W.3d 778, 785 (Tex.2001). Plaintiffs must prove by clear and convincing evidence that 1) when viewed objectively from the defendant's standpoint at the time of the event, the act or omission involved an extreme degree of risk, considering the probability and magnitude of the potential harm to others and 2) the defendant had actual, subjective awareness of the risk involved, but nevertheless proceeded with conscious indifference to the rights, safety, or welfare of others. See id.; Tex. Civ. Prac. & Rem.Code § 41.001(11); State v. Shumake, 199 S.W.3d 279, 287 (Tex.2006) (citing Transp. Ins. Co. v. Moriel, 879 S.W.2d 10, 21 (Tex.1994)). In essence, Waldrip must establish that UHI and UHT were aware that JH6097T posed an extreme degree of risk and that each of them had actual, subjective awareness that the truck's parking-brake system was not functional, but nevertheless proceeded to allow the truck to be rented.

Under the objective component, “extreme risk” is not a remote possibility or even a high probability of minor harm, but rather the likelihood of the plaintiff's serious injury. Mobil Oil Corp. v. Ellender, 968 S.W.2d 917, 921 (Tex.1998); Harrison, 70 S.W.3d at 785. The subjective prong, in turn, requires that the defendant knew about the risk, but that the defendant's acts or omissions demonstrated indifference to the consequences of its acts. La.-Pac. Corp. v. Andrade, 19 S.W.3d 245, 246–47 (Tex.1999); Wal–Mart Stores, Inc. v. Alexander, 868 S.W.2d 322, 326 (Tex.1993).

The trial court instructed the jury that for UHI or UHT to be grossly negligent, 1) the person or persons responsible for the act or omission must be employed by UHI/UHT in a managerial capacity acting in the scope of that managerial capacity; or 2) UHI/UHT authorized the doing and the manner of the act; or 3) an employee of UHI/UHT was unfit and UHI/UHT was reckless in employing him; or 4) UHI/UHT or an employee in a managerial capacity of UHI/UHT ratified or approved the act. See Hammerly Oaks, Inc. v. Edwards, 958 S.W.2d 387, 391 (Tex.1997). U–Haul does not argue any objections to these instructions.

1. UHI

Waldrip's gross-negligence claim against UHI is based on three pieces of evidence: 1) a two-page affidavit of Thomas Coffee, 2) Patterson's testimony, and 3) federal motor-carrier regulations, with which U–Haul says it is in voluntary compliance. While this Court must view this evidence in the light most favorable to the trial court's gross-negligence finding, Waldrip must establish U–Haul's gross negligence by clear and convincing evidence. Garza, 164 S.W.3d at 609. Even if the Canadian evidence was rightfully admitted, we agree with the court of appeals that Waldrip has not set forth clear and convincing evidence of gross negligence against UHI. We take each of Waldrip's arguments in turn.

First, Waldrip notes that Thomas Coffee, UHI's Director of Repair Analysis & Support, was in charge of maintaining UHI's computer database, including the DET. Waldrip contends that the database gave UHI knowledge of when a truck was in the DET. He further argues that this knowledge establishes that UHI acted with conscious indifference by allowing the truck to be rented after it became aware that the parking brake was inoperable and was not repaired. However, we agree with the court of appeals that Waldrip presented no evidence that Coffee was employed by UHI in a managerial capacity. 322 S.W.3d at 845 (citing Hammerly Oaks, 958 S.W.2d at 391). An employee's title alone is not dispositive of whether he is a vice principal. Hammerly Oaks, 958 S.W.2d at 391. Waldrip urges us to hold that Coffee is a vice principal, yet offers no evidence or case law to support his claim.

Waldrip also offers no evidence establishing that Coffee or UHI had “actual awareness” of potential parking-brake problems with JH6097T. See Andrade, 19 S.W.3d at 246 (party asserting gross negligence must establish that the defendant had “subjective awareness” of a risk and showed “conscious indifference” toward the risk involved). Even assuming that the parking brake itself, and not merely the parking-brake light, was malfunctioning, absent positive proof of knowledge by management, actual knowledge cannot be imputed to UHI based on one entry in a massive database.

Waldrip also argues that the court of appeals erroneously required direct proof that UHI was aware of the specific threat posed to Waldrip by JH6097T. See, e.g., Harrison, 70 S.W.3d at 785 (holding that a foreseeable risk does not require proof that the defendant anticipated “the precise manner in which the injury will occur”); Mellon Mortg. Co. v. Holder, 5 S.W.3d 654, 655 (Tex.1999) (holding that it is enough that the “general danger” was foreseeable); see alsoTex. Civ. Prac. & Rem.Code § 41.001(11) (requiring “an extreme degree of risk ... to others,” not specific to the plaintiff). Waldrip argues that, given the Canadian studies, Shoen was aware of the general risk that safety inspections were not being properly performed on UHI's aging fleet and that brakes could fail at any time and will always eventually fail if inspections never reveal the need for repair.

We agree with Waldrip that awareness of an extreme risk does not require proof that the defendant anticipated the precise manner in which the injury would occur or be able to identify to whom the injury would befall. See, e.g., Harrison, 70 S.W.3d at 785;Holder, 5 S.W.3d at 655. However, Waldrip is not relieved of his burden of showing, by clear and convincing evidence, that UHI had knowledge of the risk of the truck rolling posed by JH6097T's brake and transmission problems. SeeTex. Civ. Prac. & Rem.Code § 41.001(2); see also Shumake, 199 S.W.3d at 287 (citing Moriel, 879 S.W.2d at 21). Waldrip's evidence did not demonstrate that anyone at UHI had knowledge of an extreme risk and showed conscious indifference to this knowledge. See Andrade, 19 S.W.3d at 246.

Waldrip next attempts to impute gross negligence to UHI based on its alleged violation of DOT requirements governing the inspection of its trucks. The parties argue at length whether U–Haul is subject to such laws, but we do not need to answer this question. The mere existence of federal regulations does not establish the standard of care or establish gross negligence per se. See Omega Contracting, Inc. v. Torres, 191 S.W.3d 828, 842–43 (Tex.App.—Fort Worth 2006, no pet.) (citing Supreme Beef Processors, Inc. v. Maddox, 67 S.W.3d 453, 457 (Tex.App.—Texarkana 2002, pet. denied)). The fact that a motorist does not have a current DOT inspection does not mean she has been grossly negligent in the inspection and repair of her vehicle. After considering all of the evidence against UHI, we agree with the court of appeals that Waldrip did not offer legally sufficient evidence of UHI's gross negligence.

2. UHT

UHT is the entity responsible for implementing the policies in Texas for inspecting, repairing, and maintaining U–Haul vehicles. Although Waldrip asserts several theories to support the gross-negligence finding against UHT, the court of appeals based its affirmation of this part of the award on Lynn Buck's hiring of Crews. The court of appeals reasoned that the jury could have formed a firm conviction that Crews was unfit and that Buck was reckless in hiring him. 322 S.W.3d at 848. U–Haul alleges that Waldrip did not offer clear and convincing evidence that Crews was unfit or that Buck was reckless in hiring him. We address each of these points in turn.

Viewed in the light most favorable to the jury's finding, it is likely that Waldrip presented ample evidence from which a jury could form a firm conviction that Crews was unfit to be an AFM. While U–Haul is correct that Waldrip did not prove that Crews was unfit merely because of his lack of experience, Crews' testimony at trial could, and most likely did, lead a jury to believe that he was unfit to be an AFM. While Waldrip did not present any evidence on whether U–Haul's training program was inadequate, the jury could have reasonably formed a firm conviction that Crews was grossly negligent in performing his duties as an AFM. He testified at trial that the transmission fluid could not be checked in a standard-transmission truck and that such a transmission did not have transmission fluid, though he later admitted that he was wrong.

However, Waldrip did not carry his burden of proof in establishing that Buck was subjectively aware of the risk of hiring Crews and consciously chose to disregard it. The court of appeals upheld the gross-negligence finding against UHT based solely on the reckless-hiring-of-an-unfit-employee theory. Id. This theory derives from the Second Restatement of Torts, which Texas has adopted on this issue. SeeRestatement (Second) of Torts § 909 (1977); Hammerly Oaks, 958 S.W.2d at 391 (confirming adoption of Restatement's “exceptional liability” test for imputing gross negligence to a corporation). But, in light of our 1994 decision in Moriel and revisions to Chapter 41 promulgated by the Legislature in 2003, the standard for gross negligence based on this theory has changed. See Moriel, 879 S.W.2d at 23;Tex. Civ. Prac. & Rem. Code § 41.001(11) ( amended by Act of June 2, 2003, 78th Leg. R.S., ch. 204, § 13.02, 2003 Tex. Gen. Laws 887).

Moriel emphasized the “exceptional” nature of punitive damages and limited them to situations in which the defendant knew of an extreme risk of serious injury but consciously disregarded it. 879 S.W.2d at 18–19, 22. In that case, an employer covered by workers' compensation with a punitive damages award entered against it sought clarification of the gross-negligence standard and the imposition of punitive damages. Id. at 12, 14. In holding that the plaintiff had not presented legally sufficient evidence to establish the employer's gross negligence, we reasoned that a plaintiff must “ ‘establish’ the defendant's actual conscious indifference, rather than raise the mere belief that conscious indifference might be attributable to a hypothetical reasonable defendant.” Id. at 20. After Moriel, the Legislature raised the standard of proof for punitive damages from a gross-negligence award to “clear and convincing.” Tex. Civ. Prac. & Rem.Code § 41.003(b) (amended by Act of April 6, 1995, 74th Leg., R.S., ch. 19, § 1, 1995 Tex. Gen. Laws 110).

While the court of appeals addressed Crews' unfitness as an AFM, it did not elaborate on Buck's mental state. Viewing the evidence in the light most favorable to the gross-negligence finding, Waldrip did not prove by clear and convincing evidence that Buck was aware, yet consciously indifferent to, the extreme danger caused by hiring Crews or hiring an AFM with no mechanical experience. Buck testified that she knew Crews had no mechanical experience, yet did not feel that hiring him was an extreme risk because 1) the U–Haul training program would provide basic competence and 2) AFMs were not charged with performing work that required a mechanic's skill or knowledge. Without evidence that a mechanical background was required for the AFM role or evidence that U–Haul's training program was inadequate and that Buck knew that, Waldrip fell short of establishing by clear and convincing evidence that Buck knew, and was consciously indifferent to, the risk associated with hiring Crews. In fact, the court of appeals' opinion noted that Buck believed that an AFM needed no mechanical experience, indicating that she was not consciously indifferent to the risks posed by a non-mechanic. 322 S.W.3d at 846–48 (“Buck hired a person with no mechanical experience, and who she believed needed no mechanical experience, to be directly responsible for checking the transmission fluid level and check [sic] for transmission fluid leaks on a heavy-duty truck.”) (emphasis added).

Further, we are concerned about the impact the court of appeals' reasoning may have on future gross-negligence cases involving alleged reckless hiring. Under the court of appeals' reasoning, any time an employer hires a previously inexperienced employee requiring training in specific safety tasks, the employer conceivably may be found grossly negligent and subject to punitive damages if the employee acts negligently in performing her tasks. However, a party cannot be liable for gross negligence when it actually and subjectively believes that circumstances pose no risk to the injured party, even if they are wrong. See Andrade, 19 S.W.3d at 248.

We agree with the court of appeals' rejection of Waldrip's other gross-negligence theories. We do not believe Waldrip offered clear and convincing evidence establishing Guinn's subjective awareness of or conscious disregard of any problems with JH6097T's parking brake or transmission, or any evidence that any U–Haul employee was grossly negligent and UHT ratified that gross negligence.

IV. Conclusion

We reverse the court of appeals' judgment in part and affirm in part. Because the trial court's admission of the Canadian evidence was an abuse of discretion that probably led to the rendition of an improper verdict, we reverse the judgment of the court of appeals and remand the negligence claims for a new trial against all defendants. We also reverse the court of appeals and render a take nothing judgment on the gross-negligence claims against UHT. We affirm the court of appeals' judgment rendering a take nothing judgment on the gross-negligence claims against UHI. In sum, we render a take nothing judgment on the gross-negligence claims against UHI and UHT and remand the negligence claims against all defendants for a new trial.

Justice LEHRMANN filed a dissenting opinion.

Justice LEHRMANN, dissenting.

I agree with the Court that the evidence supporting the jury's punitive damages award is legally insufficient. I also agree that there is some evidence of negligence. But I believe that the Court's conclusion that the trial court abused its discretion in admitting evidence of widespread problems with U–Haul's trucks in Canada is questionable, at best. Even if that conclusion were correct, though, I agree with the court of appeals that the court's error was harmless. Accordingly, I dissent from the Court's judgment to the extent it reverses and remands the plaintiffs' negligence claim for a new trial.

I.

At trial, the plaintiffs, Talmadge and Bernice Waldrip and their daughters Dinah Simington and Annabeth Boyd, introduced the testimony of Brian Patterson. Patterson, president of the Ontario Safety League, testified that inspections of U–Haul trucks entering Canada revealed a “systemic disregard for public safety in the maintenance of vehicles in the Province of Ontario.” According to Patterson, an investigation by various Canadian governmental entities detected problems with the brakes in more than fifty percent of U–Haul's trucks. Based upon our decision in Nissan Motor Co. v. Armstrong, 145 S.W.3d 131 (Tex.2004), the Court concludes that the trial court abused its discretion in admitting Patterson's testimony, and that its admission probably resulted in a wrongful judgment. 380 S.W.3d at 136. I respectfully disagree.

A. Doubtful that the trial court abused its discretion

First, it is questionable whether the trial court abused its discretion in admitting Patterson's testimony. Unlike the evidence that was improperly admitted in Nissan, the Canadian evidence was at least arguably probative of the plaintiffs' theory that Waldrip's injury resulted from U–Haul's inherently deficient and haphazardly followed maintenance procedures. See Nissan, 145 S.W.3d at 140–42. Rule 406 of our rules of evidence expressly recognizes that “[e]vidence of ... the routine practice of an organization ... is relevant to prove that the conduct of the ... organization on a particular occasion was in conformity with the ... routine practice.” Tex.R. Evid. 406.

In Nissan, the plaintiffs asserted products liability claims against a car's manufacturer after the vehicle unexpectedly accelerated. Nissan, 145 S.W.3d at 136. In particular, they claimed that the acceleration was caused by a defective throttle cable or boot, but the allegedly faulty parts had been removed and destroyed after the accident. Id. at 134, 136. We held that the trial court abused its discretion by admitting a database of more than 700 incidents of unintended acceleration, the testimony of four witnesses who had experienced the phenomenon, and several narrative reports regarding other incidents of unintended acceleration without proof of similarity. Id. at 144. In doing so, we emphasized that a claimant's testimony that a vehicle unintentionally accelerated, without more, cannot prove that a defect caused the acceleration. Id. at 137. The plaintiff, we said, “had to present evidence that her ... car was defective, not just that other owners experienced unintended acceleration.” Id. at 138. We determined that the disputed evidence was improperly admitted because there was no evidence that most of the incidents were sufficiently similar, and the sheer volume of reported incidents likely caused confusion and prejudice. Id. at 141, 147. Clearly, evidence of dissimilar incidents of unintended acceleration is irrelevant to whether a particular vehicle accelerated as the result of a defect. But evidence that more than fifty percent of U–Haul's trucks entering Canada had improperly functioning brakes is indicative of routinely shoddy maintenance practices.

B. The error was harmless.

Even if the trial court did abuse its discretion in admitting Patterson's testimony, the error was harmless. As the Court observes, there is no specific test for determining whether the erroneous admission of evidence probably resulted in the rendition of an improper judgment. 380 S.W.3d at 136 (citing McCraw v. Maris, 828 S.W.2d 756, 757 (Tex.1992)). Instead, a reviewing court must examine the entire record “from voir dire to closing argument, considering the ‘state of the evidence, strength and weakness of the case, and the verdict.’ ” Reliance Steel & Aluminum Co. v. Sevcik, 267 S.W.3d 867, 871 (Tex.2008) (quoting Standard Fire Ins. Co. v. Reese, 584 S.W.2d 835, 841 (Tex.1979)). “[W]hether erroneous admission is harmful is more a matter of judgment than precise measurement.” Nissan, 145 S.W.3d at 144. In exercising that judgment, courts should bear in mind that “a jury's decision is not to be tampered with lightly.” In re Columbia Med. Ctr. of Las Colinas, 290 S.W.3d 204, 210–11 (Tex.2009) (parenthetically describing Wal–Mart Stores, Inc. v. Seale, 904 S.W.2d 718, 722 (Tex.App.—San Antonio 1995, no writ)). My review of the record leads me to conclude that any error the trial court may have committed in admitting Patterson's testimony does not warrant a new trial. The jury's verdict did not turn on that evidence; to the contrary, there was abundant evidence of U–Haul's negligence.

First, erroneous admission of evidence “is likely harmless if the evidence was cumulative, or if the rest of the evidence was so one-sided that the error likely made no difference.” Reliance Steel, 267 S.W.3d at 873 (footnote omitted). “[T]he complaining party [must] demonstrate that the judgment turns on the particular evidence admitted.” Nissan, 145 S.W.3d at 144. U–Haul has not met that burden.

Patterson's testimony was both cumulative and far overshadowed by other evidence of U–Haul's negligence. There was evidence that U–Haul had more than 4,000 trucks with more than 200,000 miles on them. Moreover, the evidence in this case shows that, after the truck had been rented by one customer, U–Haul had been made aware that the parking brake on the truck Waldrip later rented had not been working properly and the brake light was staying on. The brake light was repaired, yet the problem with the parking brake itself was not. In his opening statement, U–Haul's attorney expressly admitted that the problem should have been fixed, but was not, due to a “miscommunication.” Joseph Shoen, U–Haul International's chairman of the board, acknowledged that an inoperative parking brake would be a “material shortcoming” and would be U–Haul's fault. He also acknowledged that U–Haul might permit vehicles with expired safety certifications, significant components of U–Haul's safety program, to be rented “[a]t the discretion of the person doing the renting.”

The Court's determination that the trial court's admission of Patterson's testimony was harmful is based almost entirely on the emphasis supposedly placed upon it by Waldrip's counsel. 380 S.W.3d at 136. But Waldrip's counsel's efforts to gain the admission of Patterson's testimony were far from extraordinary; Waldrip's counsel dispassionately explained his view of its relevance to the trial court. Waldrip's counsel did not mention the Canadian evidence in conducting voir dire or in his opening statement. And any reference to it in his closing argument was elliptical, at best. He did not mention Patterson's name or expressly refer to the Canadian evidence. Instead, he merely alluded to U–Haul's “rotten fleet.” That reference could as easily refer to the evidence of the age and mileage of U–Haul's fleet and its slipshod maintenance procedures. In light of the record as a whole, that vague reference does not justify casting aside a three-week jury trial.

The emphasis placed upon the Canadian evidence by Waldrip's counsel in questioning Lynn Buck is equally insignificant. Buck's testimony occupies 155 pages of the record. She was questioned at length about U–Haul's maintenance practices and safety procedures. The plaintiffs' examination of her focused on whether U–Haul's procedures were faulty for not requiring leasing agencies to test vital safety systems on returned trucks before they were rented again, and the fact that several renters before Waldrip had experienced and reported problems with the truck's parking brake. Her testimony regarding the Canadian evidence spans less than three and a half pages. During that brief interval, she testified that Patterson's statement that more than fifty percent of the trucks entering Canada had defective brakes was shocking, but questionable, and that Shoen had knowledge of the reported problems. Patterson's testimony was no more emphasized in the plaintiffs' questioning of Buck than it was in closing argument.

In other cases where we have required a new trial, the conclusion that the case turned on erroneously admitted evidence rested on much firmer ground than the Court finds here. In Nissan, for example, we were persuaded that the erroneous admission of evidence of other incidents of unintended acceleration was harmful in large part because of the court of appeals' reliance on the large volume of incidents. Nissan, 145 S.W.3d at 146–47. “As the experienced justices of the court of appeals made this mistake, it is probable that the jurors did as well.” Id. at 147. We also noted that the plaintiffs' counsel had mentioned the evidence “[a]t every opportunity.” Id. at 144. In Reliance Steel, the fact that the jury awarded more damages than the evidence supported or that the plaintiff's counsel asked for strongly suggested that improperly admitted evidence of the defendant's $1.9 billion in gross sales probably led to the rendition of a wrongful judgment. Reliance Steel, 267 S.W.3d at 871–73; see also Coastal Oil & Gas Corp. v. Garza Energy Trust, 268 S.W.3d 1, 25 (Tex.2008) (noting that the jury's award of more than three times the damages plaintiff requested showed that erroneous introduction of an internal memo was harmful). And in In re M.P.A., 364 S.W.3d 277, 289 (Tex.2012), rather than a single vague allusion to improperly admitted evidence, counsel referred to it throughout closing argument. See also Coastal Oil & Gas, 268 S.W.3d at 24–25 (noting that erroneously admitted internal memo was mentioned by both sides throughout their closing arguments). If the admission of Patterson's testimony was error, its impact does not compare with these cases.

II.

Regardless of whether the trucks described in Patterson's testimony were similar to the ones that injured Waldrip or what Canada's safety standards require, the fact that more than half of them had brake problems is probative of U–Haul's routine, systemic disregard for its own maintenance policies. For that reason, I think that it is doubtful that the trial court abused its discretion in admitting the evidence. Even assuming it was error, however, the error was harmless in light of the entire record. I respectfully dissent.