What Nuclear Verdicts Reveal—and Conceal—About Litigation Risk
By Marc Mayerson | November 5, 2026
“Nuclear verdicts” have become a defining concern for companies, insurers, reinsurers and the lawyers who advise them. Large awards can affect reserves, settlement decisions, insurance purchasing, investor perception and the management of litigation portfolios.
The concern is real. But the conventional explanation is often incomplete.
Consider Roundup. Juries in Philadelphia and Cobb County, Georgia, returned verdicts of more than $2 billion each against Monsanto Co. in 2024 and 2025. On Oct. 29, a St. Louis County jury in Aretino v. Monsanto Co. unanimously rejected the claims of three plaintiffs whose counsel had asked for as much as $30 million apiece. It was, as Bayer noted, the first Roundup trial since the U.S. Supreme Court held that federal law preempts the failure-to-warn theory that had been central to the litigation. The product and the alleged injury were the same. The venue, the record and—most conspicuously—the legal theory available to the plaintiffs were not.
The usual account attributes extraordinary verdicts to “social inflation”: changing jury attitudes, distrust of corporations, aggressive plaintiff advertising, litigation finance, venue effects and an increasing willingness to use damages awards to express disapproval of corporate conduct. Each factor may matter. In a particular case, several may matter at once.
But two connected developments deserve greater attention: the procedural reforms that have shaped which cases reach juries, and the capital-market innovations that help finance the development and maintenance of claims through a demanding pretrial process.
Modern litigation contains consequential mechanisms for testing pleading sufficiency, concrete injury, individualized claim support, and expert reliability before trial. Those mechanisms can screen out unsupported claims and resolve others before a jury is empaneled. They also can leave a trial docket more concentrated in severe injuries, developed evidence, and facts that are difficult for defendants to explain.
The same reforms can front-load the cost of pursuing potentially viable claims. Earlier investigation, records collection, claimant-specific documentation and expert development require resources before the claim’s value is established or any recovery obtained. Litigation finance, portfolio funding and specialized claimant-side infrastructure can help meet that need. Successful recoveries and favorable rulings can then support further investment, claimant identification and case development—a reinforcing cycle that helps sustain litigation even as procedural screens become more demanding.
The argument is not that financing determines liability or jury valuation, or that screening alone explains verdict inflation. It is that procedure, selection and capital interact. A verdict is not a random observation from the universe of injuries, disputes or filed complaints. It is the endpoint of a process that determines which claims can survive both legal scrutiny and the economic cost of reaching trial.
Understanding nuclear verdicts therefore requires looking beyond the jury room: at the cases selected to enter it, and at the resources that enabled them to get there.
Reading the Social-Inflation Evidence
A valuable contribution is the 2026 study by Tsz Chai Fung, Lie Ma, Liang Peng and Fang Yang, “Quantifying Social Inflation in Liability Insurance with Advanced Statistical Methods,” published as research by the Casualty Actuarial Society.
The authors examine U.S. verdict and settlement data, evaluating plaintiff-win rates, settlement behavior and severity while accounting for changing case characteristics. Their findings warrant attention: increased plaintiff win probability, declining settlement probability and substantial post-2020 growth in verdict severity, including after controlling for observable case characteristics.
But the study also emphasizes case composition and litigation behavior. Verdict data do not necessarily represent an unchanging population of disputes. Once those shifts are considered, the authors find that “nuclear” outcomes have not inflated systematically faster than more moderate awards.
That does not negate social inflation. It challenges an explanation centered exclusively on disproportionately rapid growth in the largest awards.
Marathon Strategies’ 2026 verdict survey fits that picture. It counted 190 verdicts of at least $10 million against corporate defendants in 2025—a record, and 40.7% more than in 2024—totaling more than $25.6 billion. Verdicts exceeding $100 million topped 40 for the second consecutive year. But the median nuclear verdict fell to $32.5 million, which Marathon attributes to an influx of awards just clearing the $10 million threshold and to a shift in case composition toward categories that historically produce smaller awards. On those figures, 2025 was more a story of frequency than of an accelerating upper tail, and case composition again does much of the explanatory work.
The distinction between verdict and settlement trends matters. The study finds stronger recent growth in verdict severity than in settlement severity. That divergence is consistent with selection into trial, although it does not prove that selection caused the increase. Changes in jury valuation, litigation strategy and other factors may contribute.
Swiss Re’s 2025 behavioral research likewise cautions against reducing large verdicts to a “deep pockets” story. Its findings identify injury severity, rather than defendant size, as the main driver of verdict outcomes, while documenting changed attitudes toward lawsuits and damages.
Insurer loss data point the same way. AM Best’s Oct. 19 special report on social inflation found that commercial auto liability loss payments in the first 12 months rose 6.4% annually from 2014 to 2024, against average economic inflation of 2.8%. The report notes that claims staying open longer face greater exposure to jury precedent and likely higher adjustment expense, and it observes that average paid severity has trended downward in Florida following recent tort reforms. Duration and legal rules, not jury sentiment alone, shape what claims ultimately cost.
Observed verdict severity thus reflects more than one jury’s conscience. It reflects injury, proof, settlement sorting, venue and the composition of cases selected for adjudication. The useful question is how much each contributes—not whether one explanation excludes all others.
Procedural Victories Change Case Selection
Federal pleading doctrine illustrates the mechanism. Bell Atlantic Corp. v. Twombly and Ashcroft v. Iqbal require sufficient factual allegations to state a plausible claim, rather than a conceivable theory followed by discovery in search of supporting facts.
Their aggregate effect remains debated. Dismissal with leave to amend does not necessarily end a claim, and studies differ about effects on represented plaintiffs—compare the Federal Judicial Center’s 2011 report with work by Jonah Gelbach and William Hubbard. Nevertheless, plausibility pleading can require earlier investigation and a developed account of the claim’s factual basis.
Standing doctrine supplies another screen. Lujan v. Defenders of Wildlife, Spokeo Inc. v. Robins and TransUnion LLC v. Ramirez require concrete injury for federal standing; a statutory violation does not necessarily suffice. In TransUnion, 1,853 of 8,185 class members had credit reports disseminated to third parties. The other 6,332 lacked the concrete injury necessary to recover damages on the misleading-credit-report claims in federal court.
This doctrine limits some claims involving technical violations without demonstrated harm. It changes which plaintiffs remain eligible for federal adjudication.
Expert screening can be more consequential. The Supreme Court’s 1993 decision in Daubert v. Merrell Dow Pharmaceuticals Inc. and Federal Rule of Evidence 702 require judicial scrutiny of expert testimony’s factual support, methods and application. The 2023 amendment reinforced that proponents must establish admissibility requirements by a preponderance of the evidence.
Where expert testimony is necessary to establish general causation, exclusion can dispose of a litigation cohort. Surviving that challenge removes a potentially dispositive barrier; it does not prove that plaintiffs will prevail.
The Second Circuit’s July 13 ruling in the acetaminophen multidistrict litigation, left intact on Oct. 16 when the court denied rehearing en banc, illustrates the stakes. In the consolidated appeals led by Rutledge v. Walgreen Co., a panel opinion by Judge Guido Calabresi reinstated three challenged plaintiff experts while leaving two exclusions intact. Whatever one thinks of its gatekeeping analysis, the decision shows that expert admissibility remains central to the trajectory and practical value of a litigation cohort. The decision does not establish that expert screening has become unimportant. It demonstrates the opposite.
Substantive federal law can screen as decisively as evidence law. In Monsanto Co. v. Durnell, decided June 25, the Supreme Court held 7-2 that the Federal Insecticide, Fungicide, and Rodenticide Act expressly preempts a state-law claim that would require a cancer warning the U.S. Environmental Protection Agency has not required on Roundup’s label. The decision did not end the Roundup litigation. It changed which theories could reach a jury, which is why the Aretino plaintiffs tried design-defect, negligence and fraud claims instead. Monsanto contends that Durnell forecloses those theories as well; plaintiffs’ counsel disagree, as Reuters reported. However that dispute resolves, preemption reshapes the trial docket before any juror weighs the evidence.
These screens are not interchangeable. Preemption tests whether state law may impose the asserted duty; standing tests judicial power; pleading tests sufficiency; expert rules test admissibility. None universally measures merit. Collectively, however, they influence which cases remain capable of reaching trial.
Multidistrict Litigation and the Individual Case
The same individualized-proof and expert-screening mechanisms are recapitulated in federal multidistrict litigation. Under 28 U.S.C. Section 1407, related federal cases can be centralized for coordinated pretrial proceedings, but each claimant generally retains an individual case and is not automatically bound by another claimant’s merits outcome. Restrictions on mass-tort class certification have helped make MDL proceedings an increasingly important alternative for managing and resolving aggregate litigation—what Abbe Gluck and Elizabeth Chamblee Burch have called the MDL revolution.
MDL practice adds claimant-specific mechanisms. Courts may employ plaintiff fact sheets, census registries, individualized discovery and Lone Pine orders requiring information concerning product use, exposure, injury, diagnosis and causation. Noncompliance may lead to dismissal or other consequences.
Federal Rule of Civil Procedure 16.1, effective December 1, 2025, reinforces early MDL management, including consideration of how and when parties exchange information concerning the factual bases of claims and defenses. It does not mandate a particular screening regime.
Consolidation does not dissolve the individual case. On Oct. 2, the Sixth Circuit granted mandamus in the opioid MDL and vacated an order that had allowed about 800 plaintiffs to add claims against pharmacy benefit managers almost two years after the scheduling-order deadline. Rule 16, the court held, required the district court to assess each plaintiff’s diligence individually.
The firefighting-foam MDL shows these mechanisms working together. On Nov. 3, 3M Co. agreed to pay up to $7.35 billion over nine years—$5.8 billion in present value—to resolve the PFAS personal-injury claims of roughly 115,000 to 120,000 plaintiffs inside and outside the MDL. No personal-injury bellwether trial had been held. Screening, not a verdict, shaped the deal. Judge Richard Gergel required plaintiffs alleging any disease outside a short list to produce complete medical records and Rule 26 expert reports on general and specific causation or face dismissal, and plaintiffs’ leadership narrowed roughly 200 alleged injuries to six: kidney, testicular, liver and thyroid cancer, ulcerative colitis, and thyroid disease. When a large backlog of unfiled claims surfaced in 2025, the court vacated the first bellwether trial and imposed records and expert-disclosure deadlines on later-filed cases; Rule 702 challenges to the general-causation experts for two of the six diseases remain pending. The settlement compensates those six diseases and no others.
The settlement also carries the screen forward. It is an opt-in agreement that 3M may terminate if participation falls short, and it depends on a new case-management order for plaintiffs who decline it or sue later: 3M owes nothing from a $680 million reserve fund if that order ceases to be in effect, as Reuters explained. As entered on Nov. 4, the order requires those plaintiffs to serve case-specific expert reports on general and specific causation within 120 days, on pain of dismissal with prejudice, and is to remain in effect for at least nine years. Procedure becomes part of the price. Any PFAS claim against 3M that reaches a jury will have survived screening by disease, by proof and by its owner’s decision to decline the settlement.
The selection thesis requires care. Settlement can remove strong claims as well as weak ones. Substantial cases may never reach trial precisely because their value is predictable. Roundup shows the sorting at work: Monsanto’s proposed $7.25 billion class settlement, now under review in St. Louis, would resolve tens of thousands of claims; the three Aretino plaintiffs were not part of it. Screening does not guarantee that survivors are meritorious or unusually severe.
Still, eliminating unsupported claims and resolving routine disputes can leave a trial population weighted toward severe, complex and contested exposures.
The available evidence does not quantify how much these particular reforms contributed to verdict inflation; it does help explain why higher verdict severity can coexist with successful screening of unsupported claims.
Screening Begets Demand for Capital
The procedural story has an economic consequence: Earlier proof requires earlier investment.
Pleading investigation, records collection, product identification, exposure evidence, medical review, fact sheets and expert development require work before counsel knows whether a claim will survive or produce a recovery. Across hundreds or thousands of claims, those costs become a working-capital and organizational challenge.
A contingency-fee firm must finance that work while awaiting recovery. It may need intake personnel, claimant communications, records systems, medical-review capacity, expert consultants and lawyers capable of meeting individualized requirements over years of litigation.
Pleading, proof, and expert thresholds curb unsupported claims while requiring investment to document potentially stronger ones.
Those requirements often are essential safeguards. Information-intensive litigation increases the importance of resources capable of generating, evaluating and managing that information. In mass-claims practice, claimant acquisition, intake, record collection and case development can require an ecosystem of service providers. Lead generators, digital advertisers, intake vendors, medical-record retrieval companies, data-management providers and expert consultants may all become part of the process by which an individual allegation becomes a filed and documented claim.
Courts have noticed. Confronting tens of thousands of unfiled firefighting-foam claims, Judge Gergel reminded counsel that, to the extent the backlog resulted from “aggressive case acquisition advertising and marketing efforts,” they “should not accept more cases than they can reasonably investigate and prosecute.”
Litigation finance is one response. Westfleet Advisors reported approximately $2.8 billion in new U.S. commercial-litigation-finance commitments in 2025, a 23% increase over the preceding year. Portfolio transactions represented 64% of commitments.
Those figures concern commercial litigation finance, not all consumer, employment or mass-tort financing. They illustrate the institutional organization of litigation capital; they neither measure all capital supporting personal-injury inventories nor establish that funding causes larger verdicts.
Whether funding must be disclosed remains unsettled. On Oct. 21, the federal Advisory Committee on Civil Rules again stopped short of committing to a nationwide disclosure rule, notwithstanding a September letter from more than 200 companies, including insurers, and a same-day letter from Senate Judiciary Committee Chairman Chuck Grassley. The incoming subcommittee chair, U.S. District Judge Lewis J. Liman, identified as priorities “the definitional question of what constitutes litigation finance” and “the particular issues that we’re trying to solve”; further discussion is expected in April. A Federal Judicial Center study of five years of District of New Jersey disclosures, released the day before the meeting, reportedly found few trends. Local and state measures continue to multiply—the Western District of Louisiana adopted a disclosure standing order on Sept. 8, and Ohio’s funding statute took effect Oct. 6—but the debate still proceeds largely without data capable of testing claims about funding’s effect on outcomes.
The interaction produces a litigation flywheel.
Procedural requirements front-load investment. That investment creates demand for financing and specialized services. Capital enables claimant identification, investigation, records collection, expert development and inventory management. Recoveries can replenish resources; favorable rulings, bellwethers and settlement programs can supply information supporting further investment and claimant outreach. (Of course, adverse preemption or expert rulings, unfavorable bellwether-trial results, or insufficient documentation can reduce or destroy the value of a claim portfolio.)
The cycle can repeat, sustaining plaintiff-side capacity even while weak claims are screened out and few cases reach verdict.
Financing does not determine liability, injury severity or jury valuation. Its immediate role is to affect whether potentially viable claims can bear the cost of remaining in the system. Funding decisions reinforce selection, as financiers assess expected recoveries, litigation risk, and carrying costs.
The proposition that procedural requirements help create demand for capital is an economic mechanism, not a quantified causal finding. Funding demand can arise from many sources, including litigation duration, discovery expense and business expansion. Procedural reform is part of that account, not necessarily its exclusive or dominant cause.
A Better Framework for Verdict Risk
For defendants and insurers, declining claim counts should not be treated as evidence of proportionately declining trial exposure; the remaining cases may present a different concentration of severity, proof and litigation risk. The converse also holds: a single defense verdict does not reprice a mass tort. Aretino tested only the theories Durnell left standing, on one record, before one jury.
None of this denies changing jury attitudes or excessive awards. Some verdicts may reflect error or excess; others compensate genuine and serious injuries. Much of what appears systemic may be specific to a jurisdiction, judge, plaintiff firm, industry or evidentiary record.
The task is to disaggregate social inflation: Identify the contributions of jury valuation, case composition, procedural selection and litigation economics rather than assign every large award to one culprit.
Defense-side reforms have constrained important categories of claims. They also may have helped concentrate the trial docket in severe and difficult disputes, while increasing the resources needed to develop surviving claims.
The right question is not simply why a verdict was so large. It is also what had to happen—legally, factually, organizationally and financially—for that dispute to reach a jury, and on what theory.
This post provides general information about litigation and insurance and is not legal advice or a comprehensive survey of the law in any jurisdiction.
Related: Liability Coverage · Environmental, Pollution & Toxic Tort Coverage · Settlement Counsel · Expert Witness
Related posts: Expecting the Run-Around: Juries and Insurance-Coverage Cases · Trigger and Allocation for Asbestos, Other Bodily Injury, and Property Damage · Suing Excess Insurers Before Exhaustion
Topics: Nuclear Verdicts · Social Inflation · Mass Torts · Multidistrict Litigation · Litigation Finance
Comments