Summary. Multidistrict litigation now accounts for a large majority of the federal civil docket, and almost none of what happens in it appears in the Federal Rules. This article explains the machinery: how the Judicial Panel decides whether to centralize, what a transferee judge can and cannot do, how leadership is appointed and paid, how plaintiff fact sheets and census registries replace ordinary discovery, how bellwether trials are selected and what they are for, and how global settlements are structured and administered. It covers the doctrinal limits from Lexecon and Gelboim, the general causation gatekeeping that ends entire litigations, and the ethical constraints on aggregate settlement.

Here is a fact that surprises most lawyers who do not practice in this space: cases pending in multidistrict litigation make up a large majority of the entire federal civil caseload. Not a large share of tort cases. A majority of all pending federal civil actions, in most recent years.

A single MDL can hold tens of thousands of cases. One judge — a district judge in Philadelphia, or Charleston, or the Northern District of Ohio — presides over all of them. That judge will decide the science, the corporate discovery, the trial structure, and, functionally, the value of every claim. And the procedure governing all of it is largely judge-made, because 28 U.S.C. § 1407 is a single statutory section, and the Federal Rules were not written with aggregate litigation in mind.

This article explains how it actually works.

Part I: What an MDL is, and what it is not

Section 1407 authorizes the Judicial Panel on Multidistrict Litigation — seven federal judges appointed by the Chief Justice — to transfer civil actions "involving one or more common questions of fact" pending in different districts to a single district "for coordinated or consolidated pretrial proceedings," where transfer "will be for the convenience of parties and witnesses and will promote the just and efficient conduct of such actions."

Three features define it:

It is a transfer, not a merger. Each case retains its separate identity, its own caption, and its own governing state law. The transferee judge does not decide fifty cases as one; the judge decides common issues once for all of them.

It is pretrial only. The statute directs that each action "shall be remanded" to the transferor court at or before the conclusion of pretrial proceedings. The Supreme Court enforced that limit strictly in Lexecon Inc. v. Milberg Weiss Bershad Hynes & Lerach, 523 U.S. 26 (1998), holding that a transferee court has no authority to transfer a case to itself for trial under 28 U.S.C. § 1404. The practical workaround is the Lexecon waiver — parties consent to trial in the MDL court for bellwether purposes.

It is not a class action. There is no certification, no adequacy or predominance analysis, no opt-out mechanism, and no binding effect on absent persons. Every plaintiff is an individual party with individual counsel. This distinction matters enormously and is constantly blurred in public discussion.

Why the distinction matters. In Amchem Products, Inc. v. Windsor, 521 U.S. 591 (1997), and Ortiz v. Fibreboard Corp., 527 U.S. 815 (1999), the Supreme Court rejected settlement-only asbestos classes for failures of adequacy and cohesion — holding that a class cannot bind future claimants whose interests diverge from present ones. Those decisions closed the class action route for most personal injury mass torts and, in doing so, made the MDL the dominant aggregation device in American litigation. The MDL succeeded precisely because it aggregates procedure without purporting to bind anyone's substantive claim.

One more jurisdictional point. In Gelboim v. Bank of America Corp., 574 U.S. 405 (2015), the Court held unanimously that a case dismissed in its entirety within an MDL is immediately appealable as a final decision under 28 U.S.C. § 1291 — the plaintiff need not wait for the rest of the MDL to conclude. That case retains its individual character even inside the aggregate.

Part II: Getting to the Panel

The motion. Any party may move for centralization, as may the Panel on its own initiative. Briefing is short, the Panel hears argument at bimonthly sessions in rotating cities, and arguments are limited to a few minutes per party.

What the Panel actually weighs:

  • Common questions of fact — usually product design, corporate knowledge, warnings, and general causation. Individual questions of exposure, specific causation, and damages do not defeat centralization.
  • The number of cases and the pace of filing, including anticipated filings.
  • Duplicative discovery and the risk of inconsistent pretrial rulings.
  • Whether informal coordination would suffice — the Panel denies centralization where a small number of cases can be handled by cooperating counsel.

Choosing the transferee judge and district turns on: where the largest number of cases is pending; the location of the defendant and the evidence; the judge's experience and availability; the docket conditions of the district; and, candidly, the Panel's assessment of who will manage it well. Parties advocate for particular districts intensely, because the choice of judge shapes everything that follows.

Common outcomes: grant and centralize; deny because coordination is unnecessary; or centralize a subset while leaving other cases out.

The strategic point for defendants. A defendant facing scattered filings usually wants centralization — one judge, one set of rulings, one discovery record, and a forum for global resolution. A defendant facing a small number of strong cases usually does not. That calculation is made early and it drives whether centralization is sought or opposed.

Part III: The first year — leadership, structure, and orders

Once centralized, the transferee judge builds an administrative apparatus that has no direct basis in the Federal Rules and that governs the litigation entirely.

Leadership appointment. The court appoints a plaintiffs' leadership structure: lead counsel or co-leads, a plaintiffs' steering committee, and frequently subcommittees for science, discovery, law and briefing, bellwethers, settlement, and state-court liaison. Applications are submitted; the court weighs experience, resources, willingness to commit time and money, diversity of the bar, and the ability to work cooperatively. Defendants organize their own liaison structure, which is simpler because there are usually few defendants.

What leadership does — and what it means for individual counsel. Leadership conducts the common discovery, retains and funds the general causation experts, briefs the common motions, negotiates the case management orders, and leads settlement negotiations. Individual counsel with a hundred cases in the MDL will typically do very little in the litigation itself and a great deal in client intake, records collection, and — eventually — settlement participation decisions.

Common benefit funds. Leadership work is financed through an assessment — commonly five to twelve percent — held back from every plaintiff's recovery, whether or not that plaintiff's lawyer did any common work. The court establishes the fund, sets the percentage, appoints a special master or committee to review time and expense submissions, and allocates at the end. Common benefit assessments are one of the most contested issues in modern MDL practice, particularly as applied to cases resolved in state court or by counsel who never appeared in the MDL.

The early case management orders, which together constitute the real procedural code of the litigation:

  • Appointment of leadership and establishment of the common benefit fund.
  • A master complaint and master answer, with short-form complaints by which individual plaintiffs adopt the master allegations and add their own facts.
  • A direct filing order, permitting plaintiffs to file directly in the MDL district while designating the district that would have been proper for later remand and choice of law.
  • Plaintiff fact sheets and defendant fact sheets, which substitute for ordinary interrogatories and document requests.
  • A protective order and a Rule 502(d) order for privilege.
  • An ESI protocol.
  • The bellwether process.
  • Science day — an informal, non-adversarial tutorial in which both sides teach the judge the underlying science with no ruling made.

Part IV: Discovery in the aggregate

Common discovery proceeds against the defendant on the questions shared across all cases: what the company knew, when it knew it, what testing was done, how the product was designed and manufactured, what warnings were given and considered, and how it was marketed and to whom. This is ordinary corporate discovery conducted once, at scale, producing millions of documents and dozens of depositions.

Individual discovery is compressed into the plaintiff fact sheet — a standardized questionnaire covering demographics, product identification, exposure or usage, medical history, treating providers, diagnosis, damages, and authorizations for records. Failure to serve a complete fact sheet is the most common route to dismissal in an MDL, and courts enforce it through show-cause orders and dismissal for failure to prosecute.

Census registries. A newer device: an early, pre-discovery registry in which counsel must submit basic claim information for every case and every unfiled claim, often supported by records demonstrating product use and diagnosis. Registries emerged in response to a recognized problem — MDLs accumulating large numbers of unvetted claims that inflate settlement expectations and consume court resources. Their use has expanded rapidly.

Lone Pine orders require plaintiffs, at a specified stage, to produce prima facie evidence supporting exposure, injury, and causation — typically an expert affidavit — or face dismissal. They are powerful, contested, and generally entered later in a litigation, after general causation has been resolved and the meritless portion of the inventory has become apparent.

The tension underneath all of this is real and worth stating plainly. Aggregation makes it economically possible to litigate against a well-resourced defendant, which is its purpose. It also makes it economically possible to file claims that would never be filed individually, because the marginal cost of adding a case to an inventory is small. Fact sheets, registries, and Lone Pine orders are the tools courts have developed to manage the second effect without destroying the first.

Part V: General causation — where litigations live or die

In a mass tort, causation splits into two questions:

General causation — can this substance or product cause this disease in humans, at these exposure levels? Specific causation — did it cause this plaintiff's disease?

General causation is decided once, for everyone. A ruling excluding the plaintiffs' general causation experts under Fed. R. Evid. 702 and Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993), ends the entire MDL. A ruling admitting them transforms the defendant's settlement posture overnight. It is, in most pharmaceutical and toxic tort MDLs, the single most consequential event in the litigation.

What the inquiry examines: epidemiological studies and their design, power, and confounders; the strength of association and whether relative risk exceeds the threshold some courts require for individual causation inferences; dose-response relationships; biological plausibility and mechanism; animal and in vitro data and their extrapolability; the Bradford Hill considerations; whether the methodology was developed for litigation or independently; and — after the 2023 amendment to Rule 702 — whether the opinion reflects a reliable application of the methodology to the facts, established by a preponderance of the evidence.

The 2023 amendment matters here. It was adopted precisely because courts had been admitting expert testimony on the view that questions of application go to weight rather than admissibility. In mass tort litigation, where a single ruling governs thousands of cases, that shift has real consequences. See Expert Witnesses After the 2023 Amendment to Rule 702.

Preemption frequently rides alongside causation as the other potentially dispositive common issue — whether federal regulatory approval displaces state-law failure-to-warn claims. It is briefed early and it can end a litigation as cleanly as a causation ruling. See FDA Regulation of Drugs, Devices, and Diagnostics.

Part VI: Bellwether trials

The purpose is information: to test theories, evidence, witnesses, and — above all — to generate verdicts that give both sides a data point for valuation. A bellwether verdict binds nobody. It informs everybody.

Selection methods, in ascending order of contentiousness:

  • Random selection from the full pool, which produces representativeness and mutual dissatisfaction.
  • Party picks — each side selects a number, producing extreme cases on both ends.
  • Stratified selection by injury type, exposure duration, or geography.
  • Court selection from a pool, sometimes after discovery on a larger group.

The recurring criticism of party-pick systems is that they generate outliers rather than representative cases, and therefore poor valuation information — which is precisely why courts increasingly use stratified or randomized pools.

The process: a discovery pool of perhaps twenty to fifty cases; case-specific discovery and depositions; specific causation experts; case-specific Daubert and summary judgment motions; and then a handful of trials.

The Lexecon problem. Because the MDL court cannot try a transferred case without consent, bellwethers proceed either on a Lexecon waiver, or in cases directly filed in the MDL district that could properly have been brought there, or by remand to the transferor court for trial.

How verdicts translate into settlement. A string of defense verdicts collapses plaintiff valuations. A large plaintiff verdict — particularly with punitive damages — moves defendants to negotiate. Mixed results, which are the norm, produce a settlement range keyed to injury severity and case characteristics. The bellwethers do not resolve the cases; they price them.

Part VII: Global settlement

Most MDLs end in an aggregate settlement, and the structures are more varied than outsiders assume.

Common architectures:

  • A settlement matrix or grid. Points are assigned for injury type and severity, duration of use or exposure, age, comorbidities, alternative causes, product identification strength, and treatment history. Points convert to dollars from a fixed fund.
  • A participation threshold. The defendant's obligation is conditioned on a stated percentage of eligible claimants — often 85% to 98% — opting in. This is the mechanism that makes global peace possible and the mechanism most criticized, because it exerts substantial pressure on individual claimants to accept.
  • A claims administrator and special master to evaluate submissions, apply the grid, and resolve disputes, frequently with a limited appeal to a neutral.
  • A qualified settlement fund under the Internal Revenue Code, to hold and distribute proceeds with defined tax treatment.
  • Lien resolution administration for Medicare, Medicaid, ERISA plans, TRICARE, VA, and provider liens — see 42 U.S.C. § 1395y(b) and, on Medicaid recovery, Gallardo v. Marstiller, 596 U.S. 347 (2022).
  • Common benefit assessment withheld from each recovery.
  • Walk-away rights for the defendant if participation falls short.

The ethical constraint that governs every plaintiff's lawyer. The aggregate settlement rule — Model Rule 1.8(g) — provides that a lawyer representing two or more clients shall not participate in making an aggregate settlement of their claims unless each client gives informed consent in a writing signed by the client, after disclosure of the existence and nature of all the claims involved, and of the participation of each person in the settlement.

Read carefully, that rule is demanding. Each client must be told what every other client is receiving. Each client decides individually. A lawyer with eight hundred cases must obtain eight hundred individual informed consents, and cannot bind a client who declines. Provisions purporting to require a lawyer to withdraw from representing non-participating clients, or to recommend the settlement to all clients, raise serious conflict problems. This is the most frequently litigated ethics issue in mass tort practice and the one most likely to produce discipline. See Legal Ethics in Practice.

Judicial review is limited. Unlike a class settlement, which requires court approval for fairness under Fed. R. Civ. P. 23(e), an MDL aggregate settlement is a private contract among individual parties. Courts exercise substantial practical influence — through fee caps, common benefit rulings, and management of the process — but the formal approval architecture of Rule 23 does not apply. Whether it should is one of the live debates in civil procedure.

Part VIII: The state court side, and bankruptcy

Parallel state proceedings. Only federal cases go to an MDL. Many mass torts run simultaneously in state courts, particularly in consolidated proceedings in New Jersey, California, Pennsylvania, and Delaware. Coordination between the MDL and state proceedings is informal — cross-noticed depositions, shared document depositories, coordinated trial settings — and depends on cooperation between judges.

Removal and CAFA. Defendants generally prefer federal court and remove aggressively; plaintiffs frequently structure filings to defeat removal by naming non-diverse defendants. The Class Action Fairness Act's mass action provision, 28 U.S.C. § 1332(d), permits removal of certain actions in which claims of one hundred or more persons are proposed to be tried jointly — a provision plaintiffs' counsel structure around by filing in groups below the threshold. See Removal and Remand.

Personal jurisdiction. Bristol-Myers Squibb Co. v. Superior Court, 582 U.S. 255 (2017), held that a state court lacked specific jurisdiction over the claims of non-resident plaintiffs who were not injured in the forum, even though resident plaintiffs asserting identical claims could proceed. That decision substantially constrained the consolidated state-court filings that had previously aggregated nationwide claims in favorable forums, and it pushed more litigation into the federal MDL system. See Personal Jurisdiction Over Online and Foreign Defendants.

Bankruptcy as a mass tort resolution device. A defendant facing overwhelming aggregate liability may seek Chapter 11, invoking the automatic stay under 11 U.S.C. § 362 to halt all tort litigation and resolving claims through a plan and a claims trust. Asbestos liabilities have been channeled this way for decades under the trust and injunction mechanism Congress created for that purpose. More recently, solvent defendants have attempted divisional mergers — the so-called Texas two-step — to place tort liabilities into a newly created entity that then files. Courts have divided sharply on whether such filings satisfy the good-faith requirement for a bankruptcy petition, and the issue remains genuinely unsettled. See Chapter 11 Reorganization.

Part IX: A worked example

The litigation. A surgical mesh device. By month six, four hundred cases are pending in eleven districts. The manufacturer moves for centralization; plaintiffs' firms split, some supporting a district where their cases are concentrated.

Month 4 — The Panel. Centralization granted; transferred to a district with the largest concentration and an experienced judge.

Months 5–9 — Structure. Leadership applications; three co-leads and a fifteen-member steering committee appointed; common benefit fund set at eight percent plus a cost assessment. Master complaint, short-form complaint, direct filing order, plaintiff and defendant fact sheets, protective order, ESI protocol. Science day in month nine.

Months 9–26 — Common discovery. 2.4 million documents; forty-one corporate depositions; regulatory submissions, complaint files, internal safety reviews, and marketing materials. Filings grow to eleven thousand cases.

Month 22 — Census registry ordered after the court observes that a meaningful fraction of filed cases lack product identification. Two thousand cases are dismissed voluntarily within ninety days.

Months 24–30 — General causation. Plaintiffs' experts on mechanism and epidemiology; defense motions under Rule 702. The court admits opinions as to two injury categories and excludes them as to a third, eliminating roughly fifteen percent of the inventory.

Months 30–40 — Bellwethers. A pool of thirty, stratified by injury category and implant duration; case-specific discovery; six cases tried on Lexecon waivers. Results: two defense verdicts, three plaintiff verdicts of $1.1M, $2.8M, and $7.4M (the last including punitive damages, later reduced), and one mistrial.

Months 40–48 — Settlement. A matrix with four injury tiers and adjustments for revision surgery, implant duration, age, and comorbidities. Fund of $840 million; participation threshold of 95%; special master and claims administrator appointed; lien resolution program for Medicare, Medicaid, and private plans; common benefit holdback of eight percent.

The individual lawyer's experience. A firm with two hundred cases spent most of the litigation collecting medical records, completing fact sheets, and responding to registry orders. When the settlement arrived, it obtained two hundred separate signed informed consents disclosing the aggregate terms and each client's allocation — and had to be prepared for clients who declined, whose cases would return to the ordinary litigation track.

Part X: The economics that drive everything

Nothing about MDL practice makes sense without the money, and the money is unusual on both sides.

On the plaintiffs' side. Common discovery in a pharmaceutical or device MDL costs tens of millions of dollars — document review, expert retention, epidemiology, corporate depositions across the country, and trial preparation for bellwethers. That expense is fronted by leadership firms, unsecured, over four to seven years, against an uncertain recovery. It is financed through firm capital, bank lines, and increasingly third-party litigation funding, which brings its own privilege, disclosure, and control questions. See Third-Party Litigation Funding.

Client acquisition is its own industry, and it explains much of what critics observe. Advertising, lead generation, and case referral networks supply inventories, with referral fees flowing between firms under state rules that vary in what they require by way of disclosure and joint responsibility. The economics reward volume, and volume without vetting is exactly the problem census registries and Lone Pine orders were designed to address.

On the defense side. A defendant in a large MDL is running a multi-year program with a defined budget: national coordinating counsel, regional trial counsel, document review vendors, science and regulatory experts, and a settlement reserve disclosed to auditors and, for public companies, to investors. The reserve is itself consequential — an accrual signals valuation to the plaintiffs' bar and to the market, which is why the timing and size of reserves are handled with care.

Insurance. Product liability towers, excess layers, and disputes with carriers over trigger, allocation across policy years, exhaustion, and the duty to defend run alongside the MDL as a separate litigation, frequently in a different forum. See Business Insurance and Coverage Disputes and Insurance Coverage Toolkit.

The fee structure at the end. A settling plaintiff's recovery is reduced by, in order: the common benefit assessment (often five to twelve percent off the top), the individual contingency fee (frequently capped by the court at a percentage below the retainer's stated rate), case-specific costs, and lien repayment. A claimant awarded $180,000 on a matrix may net well under half of it, and the arithmetic should be explained plainly at the outset rather than discovered at disbursement.

Part XI: The criticisms, taken seriously

MDL practice attracts sustained criticism from thoughtful people on both sides, and a practitioner should be able to state the objections accurately.

"There is no appellate check." Because MDL rulings are interlocutory and cases settle before final judgment, the transferee judge's most important decisions — on discovery scope, on case management orders, on general causation, on the settlement architecture — frequently escape appellate review entirely. Gelboim restored appellate access for fully dismissed cases, but it does not reach the ordinary case-management rulings that shape the litigation. Mandamus is available in theory and rarely granted.

"The settlement is not reviewed for fairness." A class settlement requires judicial approval under Rule 23(e), with notice, objections, and findings. An MDL aggregate settlement requires none of that. The counterargument is that each claimant is individually represented and individually consents. The rejoinder is that a 95% participation threshold, combined with a lawyer holding a thousand cases, creates pressure that individual consent does not fully answer.

"Unvetted claims inflate the inventory." Studies and judicial experience both indicate that a meaningful fraction of filed cases in some MDLs lack product identification, a qualifying injury, or any causal basis, and that these claims nonetheless affect settlement valuation. Census registries, early vetting orders, and Lone Pine orders are the response, and their expansion is one of the clearest trends in modern practice.

"Leadership entrenches." The same firms appear repeatedly in leadership positions across MDLs, which produces experience and also produces a closed circle. Courts have responded with structured application processes and explicit attention to including newer and more diverse counsel.

"Individual clients are functionally unrepresented." A claimant whose lawyer never appears in the MDL, whose case is governed by orders they never see, and whose recovery is set by a matrix has an experience quite unlike ordinary litigation. The honest defense is that the alternative — individual litigation against a defendant with unlimited resources — would produce no recovery at all for nearly all of them. Both things are true.

What a practitioner should take from this. These are not merely academic complaints. They shape how judges manage MDLs, and they are the reason a plaintiff's lawyer should document client communication carefully, obtain genuine informed consent, and be able to explain the fee and lien arithmetic in writing — and the reason a defense lawyer should push early for vetting rather than complaining about inventory quality at the settlement table.

Part XII: Beyond personal injury — the other MDLs

Product liability and pharmaceutical cases dominate the public image, but the Panel centralizes a much wider range of litigation, and the mechanics differ meaningfully.

Data breach and privacy MDLs. Consolidated after a large breach, these turn on standing under TransUnion LLC v. Ramirez, 594 U.S. 413 (2021) — whether risk of future misuse is a concrete injury — and on the availability of a Rule 23 class, which is far more viable here than in personal injury because damages are relatively uniform. Many of these proceed as class actions within an MDL, combining both devices. See Data Breach Response and Notification and State Consumer Privacy Laws.

Antitrust MDLs. Price-fixing and monopolization cases centralize routinely, with direct purchaser classes under federal law and indirect purchaser classes under state law running in parallel, plus opt-out actions by large purchasers. Discovery is document-intensive and frequently follows a government investigation. See Antitrust for Technology Companies.

Securities MDLs. Less common than in other areas because the PSLRA's lead plaintiff provisions and first-filed consolidation already aggregate most securities cases, but the Panel centralizes where parallel derivative and ERISA actions accompany the securities claims. See Securities Fraud Litigation Under Rule 10b-5.

Consumer product and false advertising MDLs. Typically class-based, with certification and predominance the central fight rather than general causation.

Employment MDLs. Wage and hour collective actions under the FLSA and state-law class claims, where the aggregation question is conditional certification and decertification rather than Rule 23. See Wage and Hour Law Under the FLSA.

Patent MDLs. Centralized where the same patents are asserted against many defendants in different districts, with claim construction — the Markman proceeding — as the common issue decided once. See Patent Litigation Toolkit.

The general lesson. In personal injury MDLs, the common issue is science. In antitrust, it is conduct. In securities, it is disclosure. In patent, it is claim construction. Identifying which common issue will actually be decided first — and whether a ruling on it is dispositive for everyone — tells you where the litigation is going and what the leverage points are.

Part XIII: Frequently asked questions

"Is an MDL a class action?" No. Each plaintiff is an individual party with individual counsel and an individual claim. Nobody is bound by anyone else's outcome, and there is no opt-out because there is nothing to opt out of. A class action may exist within an MDL, but the two devices are distinct.

"Will my case be tried?" Almost certainly not. A handful of bellwethers are tried out of thousands of cases. The overwhelming majority resolve through the aggregate settlement, and a small number are remanded to their original districts.

"Who is my lawyer?" The firm you retained. Leadership counsel act for the common benefit and do not represent you individually, which is why the informed consent obligations at settlement fall on your own lawyer, not on leadership.

"Why does my recovery get reduced by a percentage for lawyers I never hired?" The common benefit assessment funds the discovery, experts, and briefing that made any recovery possible. It is court-ordered, its percentage is set by the judge, and it is one of the more contested features of the system.

"How long will this take?" Large MDLs commonly run four to eight years from centralization to settlement distribution. Individual claimants often wait years after settlement for lien resolution and claims administration to complete.

"What happens if I don't accept the settlement?" Your case continues on the litigation track — potentially through remand to your original district for trial. In practice, very few claimants take that path, which is both the point of the participation threshold and the reason it is criticized.

"Can a defendant just declare bankruptcy?" It can file, and the automatic stay under 11 U.S.C. § 362 halts the litigation while the case is pending. Whether a solvent enterprise may use a divisional merger to place tort liabilities in a new entity and file has divided the courts, and the good-faith requirement for filing is where the fight occurs.

"How do I know if my case is in an MDL?" Ask your lawyer for the MDL number and the district, then look at the court's MDL page, where transferee judges publish case management orders, hearing dates, and often a public docket summary. Those orders are the operative procedural law of your case, and they are public.

Part XIV: What good management looks like

Judges who run MDLs well share a set of practices, and lawyers who understand them work more effectively inside the structure.

Early vetting. A census registry or fact-sheet deadline in the first year, enforced with show-cause orders, produces an inventory of real claims and prevents the settlement negotiation from being conducted over an unknown denominator. Judges who defer vetting until the settlement stage spend the settlement stage litigating who is actually in the case.

A science day before the science fight. An informal, non-adversarial tutorial — each side teaching, no ruling made — gives the court a working command of the discipline before the Daubert briefing arrives. It costs a day and improves every subsequent ruling.

A published, predictable order structure. Numbered case management orders on a public docket page, so that a lawyer joining the litigation in year three can read the procedural law of the case in an afternoon.

Bellwether pools selected before the parties know which cases will be tried. Stratified or randomized selection from a defined pool, with case-specific discovery on the whole pool, prevents both sides from cultivating outliers and produces verdicts that actually inform valuation.

Explicit attention to remand. Section 1407 contemplates remand, and an MDL in which nothing is ever remanded has quietly become something the statute did not authorize. Judges who suggest remand of unsettled cases at a defined point create a real alternative to the settlement, which improves the settlement.

Transparency about common benefit. Setting the assessment percentage early, appointing a neutral to audit time and expense submissions, and publishing allocation criteria reduces the litigation-within-the-litigation that otherwise consumes the final year.

And for counsel on both sides: the single most useful posture in an MDL is credibility with the court. A judge managing eleven thousand cases relies on the lawyers' representations to an unusual degree, and the lawyer whose statements have proven reliable gets discretionary rulings that a technically stronger but less credible opponent does not. That is not a legal doctrine, and it is nonetheless the most consequential fact about practicing in these proceedings.

Primary authority

Related documents

This article is educational and not legal advice. MDL practice is governed substantially by case management orders that differ in every proceeding, and by ethical rules that vary by jurisdiction. Consult the standing orders of the specific MDL and qualified counsel before acting.