The Yale Law Journal


Nudges and Norms in Multidistrict Litigation: A Response to Engstrom

04 Nov 2019

A Response To

The Lessons of Lone Pine Nora Freeman Engstrom 30 Oct 2019

abstract. On paper, the Federal Rules of Civil Procedure apply equally to billion-dollar opioid allegations and claims for $75,000.01. In practice, however, judges and attorneys in high-stakes multidistrict proceedings like those over opioids have invented a smattering of procedures you will never find in the Federal Rules: plaintiff fact sheets, short-form complaints, science days, bellwether trials, census orders, inactive dockets, and Lone Pine orders, to name but a few. In a world where settlement is the prevailing currency, new norms take root. But as these norms blossom, the stabilizing features of the federal rules—balance, predictability, and structural protections—can wither. As safeguards atrophy and transferee judges actively nudge the parties to settle, the deals that emerge may reflect the scars of creative pruning, not the suit’s merits.

This Essay responds to Nora Engstrom’s article, The Lessons of Lone Pine, by introducing empirical data on her target (Lone Pine orders) and her remedy (plaintiff fact sheets). Examining these two innovations in the lifecycle of products-liability multidistrict litigation illustrates that norms can undermine a key aim of the Federal Rules: achieving procedural parity, where discovery burdens and the risks of erroneous decisions fall equally on both parties. Although procedural design choices should respond to case-management demands, both shortcuts can impact plaintiffs more harshly, upending protections for those who need them most.


The world of mass-tort multidistrict litigation that Nora Engstrom describes in her recent Yale Law Journal Article, The Lessons of Lone Pine, is foreign to most lawyers.1 These proceedings command not only their own lingo, like Lone Pine orders, census orders, and fact sheets, but also their own bespoke procedures, which an elite cadre of federal judges invoke to wrangle cases into settlement.2 That means that a key assumption behind the Federal Rules of Civil Procedure—that “a litigant can walk into any federal courtroom in the country and know that the same procedures will apply to her case”—does not ring true.3 For the uninitiated, Lone Pine orders require mass-tort plaintiffs to prove that the defendant’s product or device caused their injuries with an expert report, such as one from a surgeon or physician.4 As a judicial concoction, Lone Pine orders are the pluots of mass torts, grafting together snippets of federal rules while ignoring their procedural protections.5

In stark contrast to most existing literature lauding Lone Pine orders, Engstrom deftly exposes their deep flaws: Lone Pine orders circumvent jury trials, little precedent exists to guide them, and, as a case-management tool under Rule 16, they are afforded an abuse-of-discretion standard on appeal rather than the stricter de novo standard used for summary judgment.6 Put simply, Lone Pine orders bypass institutional constraints and are out of step with existing procedural rules. Like summary-judgment motions, they demand substantive support for claims, but they short-circuit Rule 56’s procedural protections and clear burdens of proof. And like Rule 11 sanctions, they hope to siphon off frivolous claims, but they demand that plaintiffs provide expert reports and prima facie evidence of their injury, whereas Rule 11 merely requires that lawyers certify that support exists or is likely to exist after discovery.7 Finally, like discovery sanctions under Rule 37, Lone Pine orders dismiss noncompliant cases, but contrary to this Rule they do so as a first step rather than a last resort.

Why do judges use them then? Some judges and defense lawyers claim that Lone Pine orders “represent salvation from the huddled masses of meritless plaintiffs’ claims lying in wait for eventual settlement checks.”8 With that in mind, Engstrom suggests a balance: judges should use Lone Pine orders sparingly and “only when (1) procedures explicitly sanctioned by rule or statute are unavailable or are patently insufficient, and (2) substantial evidence casts doubt upon plaintiffs’ entitlement to relief and/or the plaintiffs have displayed a marked and unjustifiable lack of diligence.”9 In lieu of Lone Pine orders, Engstrom contends that judges should winnow claims by requiring plaintiff fact sheets and disincentivizing noncompliance, “with a first step being that . . . noncompliant plaintiffs will face entry of a Lone Pine order, and those who fail to comply with that order will see their claims dismissed with prejudice.”10

Engstrom’s article is a tour-de-force on Lone Pine orders and should be required reading for every transferee judge who handles a mass-tort multidistrict litigation (MDL). Her dataset includes a wide array of Lone Pine orders used in both state and federal courts, many of which were issued during discovery and before a proceeding settled.11

Volume-wise, state courts handle more mass-tort cases than federal courts do. But once the Judicial Panel on Multidistrict Litigation (the Panel) centralizes federal cases before a transferee judge, that judge and the attorneys she selects to spearhead the MDL proceeding tend to be in the driver’s seat when it comes to negotiating a global settlement. As Part I explores, global deals aim to include all plaintiffs, regardless of whether they sued in state or federal court.

Part I thus narrows the scope of our inquiry, shifting away from how all state and federal courts use Lone Pine orders to how transferee judges employ them in products-liability MDLs. Products liability encompasses the largest and thorniest mass torts, like asbestos and pelvic mesh, which tend to demand heavy-handed case management.12 Corralling dispersed plaintiffs into a single settlement framework takes some doing, and ethics rules can take a backseat to closure goals.13 Looking at the lifecycle of mass torts in that context shows that Lone Pine orders appear almost exclusively post-settlement—not as pre-settlement sieves, but as an additional means to urge non-settling plaintiffs to settle.14

Repositioning Lone Pine orders as settlement nudges means that they should fail the first prong of Engstrom’s test, for tried-and-true procedures already exist to resolve cases without them. When plaintiffs refuse a settlement offer, nothing about the status quo should change. Non-settling plaintiffs should be able to continue their path toward trial. If they have not completed discovery, then discovery should continue. If they don’t respond to discovery requests, then the court could consider whether to issue Rule 37 sanctions and dismiss the case. Alternatively, if pretrial matters have concluded, plaintiffs’ cases should be remanded to their original court for trial as the MDL statute contemplates.15

Part II explores Engstrom’s proposed solution—using plaintiff fact sheets in lieu of Lone Pine orders. Unlike Lone Pine orders, fact sheets do appear routinely before a global settlement. But, as yet another ad hoc invention that substitutes for formal discovery rules, they fall prey to some of the same criticisms that Engstrom levels at Lone Pine orders.

I. lone pine orders as settlement nudges

As Engstrom points out, there is widespread agreement that not all mass-tort plaintiffs have valid claims, even though no firm data exists on the topic.16 But if plaintiffs’ attorneys file dubious claims, it’s with the hope of getting some small payout from a settlement grid—a volume game that they would lose if they withheld those claims from a settlement program.17 So, if transferee judges issue Lone Pine orders after lead attorneys negotiate a settlement, those orders are unlikely to weed out suspicious claims and more likely to push reluctant plaintiffs to settle by raising litigation costs and signaling judicial approval of the deal.

Using a dataset of products-liability MDLs that settled over the course of fourteen years,18 I found that judges in sixteen of thirty-four proceedings issued Lone Pine orders, or 47%. Of those, fifteen (93.7%) issued them after the first settlement date. This suggests that, at least in these MDL proceedings, Lone Pine orders do not winnow meritless claims pre-settlement.19 It’s the settlement’s claims administrators who decide which claims warrant compensation, not judges. Of course, not all noncompensable claims are “meritless”; some may simply fall outside the strict parameters that settling parties negotiate. In the massive Vioxx settlement, for example, the claims administrator asked for additional information on 256 potentially fraudulent claims, and did not get a response on 194.20 Of the sixty-two who responded, thirty-nine overcame the initial suspicion (46%), and Judge Fallon later reported that only two out of 50,000 claims were fraudulent.21

Looking at the bigger picture, the first settlements in those thirty-four proceedings occurred, on average, 1,234 days (or three years and 4.5 months) after the Panel centralized them. And the judges who issued Lone Pine orders within those proceedings did so on average 1,411 days (or three years and 10.3 months) after centralization.22 Although not all of the proceedings have closed and the difference is not statistically significant in this sample, closed proceedings with Lone Pine orders ended no sooner than those without them. More specifically, the ten closed proceedings that issued Lone Pine orders concluded an average of 3,304 days (or a little over nine years) after centralization, whereas the eight proceedings that did not issue them closed an average of 2,904 days (or less than eight years) after centralization—over a year sooner.23

Figure 1. Timeline of MDL Events Within the Dataset24


As Figure 1 illustrates, 97.3% of these Lone Pine orders applied post-settlement only to non-settling plaintiffs. If Lone Pine orders do not screen cases pre-settlement, then judges must issue them for a different reason entirely. As twilight orders that follow on the heels of a private deal, Engstrom aptly observes that these Lone Pine orders can have a “dark side”—“to strongarm claimants to accede to a settlement agreement that they would rather refuse.”25 Thus, any study of Lone Pine orders in mass-tort MDLs must begin with an understanding of how those private settlements work, a task to which we now turn.

A. Settlement Dynamics in MDLs

Settlement designers in both mass-tort deals and class-action settlements want to include as many people as possible to give defendants all-inclusive relief. Class-action settlements can work like sticky flypaper: dealmakers aim to keep class members from opting out through most-favored nation provisions, walkaway clauses, and even liens on the defendant’s assets in favor of settling class members.26 Inertia helps. Class members are bound by the deal unless they opt out, so the default position promotes closure for defendants. Yet, Rule 23 includes some safeguards for absent class members—notice, opt-out rights, objections, judicial settlement approval, and adequate representation.27

But imagine a scenario without those class-action protections, where a client becomes a number, not a name, and is represented by the same attorney who represents hundreds of others. Imagine further that only one common factual question is needed to centralize those cases before the same judge and that the judge does not appoint lead plaintiffs’ lawyers to ensure that they adequately represent the spectrum of competing interests. Instead, judges appoint leaders based on attorneys’ expertise, tendency to cooperate with others, and ability to fund the suits.28 That’s what happens before mass torts conclude in private, aggregate settlements, as 47.9% of all products-liability proceedings in the dataset ultimately did. Unlike class settlements, where everyone is automatically “in,” plaintiffs’ attorneys must convince their clients to settle—to affirmatively act—to satisfy defendants’ closure goals. That is where the ethical conundrums and the “dark side” that Engstrom alludes to begin.

Figure 2. How Do Product-Liability MDLs End?29

Understanding how these deals work requires a bit of terminology and explanation. In this context, the word “settlement” is a misnomer—the private agreements do not actually settle anything. They are deals between the lead plaintiffs’ lawyers and the defendant that set up settlement programs. This is key, because the terms require plaintiffs’ lawyers to do a little arm-twisting before an actual settlement occurs.30

For example, every publicly available private deal in the dataset included a “walkaway” clause, meaning that the defendant could abandon the settlement offer if too few plaintiffs signed up.31 To measure compliance and identify the denominator for the walkaway percentage, some leaders request a “census” order, as Figure 1 showed. Although lead lawyers could count federal cases on their own, they want a full roll call that identifies state-court plaintiffs and unfiled claims. So they enlist the court’s help: judges in 35% of the thirty-four proceedings required attorneys to register all of their clients, regardless of where (or whether) they filed those claims.32

Meeting walkaway percentages, which require between 85% and 100% of plaintiffs to enter the program, means plaintiffs’ lawyers must corral as many clients as possible or risk blowing up the offer and not getting paid.33 Entering the program, however, requires plaintiffs to dismiss their lawsuit with prejudice.34 Yet, in 60% of the private settlements, plaintiffs had little idea how much—if anything—they would actually receive.35 Informed consent is therefore a stretch.36

That takes out carrots and leaves us with the real stick: ethically dubious clauses that require plaintiffs’ attorneys to recommend that all of their clients settle and to withdraw from representing clients who refuse.37 Non-settling clients find themselves between a rock and a hard place.38 In corresponding with pelvic-mesh plaintiffs as part of an ongoing study,39 for example, one plaintiff told me “I let [my law firm] know that I do NOT want them to dismiss my case [to enter a settlement program]. They have told me that means I will likely be dropped from them . . . . Right now they are trying to bully me, and that just isn’t working!”40

To make matters worse, these mandatory recommendation and withdrawal provisions tend to appear alongside simultaneous agreements for attorneys not to solicit or accept new clients.41 So, if that plaintiff declined to settle and her attorney withdrew from representing her, she may have to proceed pro se—not a simple task for injured plaintiffs laden with medical bills, especially when Lone Pine orders demand expensive expert proof.42

Withdrawing from representing a client midstream typically necessitates judicial approval.43 In 35% of the thirty-four proceedings, judges acquiesced. Others were “systematic in not permitting withdrawal.”44 Adaptive lawyers, however, have engineered workarounds in their contingency-fee agreements. Some build escape hatches: one firm incorporated an exit clause that allowed it to get out of its client obligations at any time.45 Others include deterrents: one firm required a client to reimburse it for her litigation expenses if she went against counsel’s settlement advice and insisted on going to trial but ultimately lost.46

As Judge Jack B. Weinstein recognized, “[t]heoretically, each client has the option of rejecting his share of a settlement . . . . In practice, the attorney almost always can make a global settlement and convince the clients to accept it.”47 But as the terms Lone Pine orders, census orders, and withdrawal orders suggest, attorneys are not alone in their settlement-inducing endeavor. Judges nudge too. In 64.7% of those thirty-four proceedings, judges formally appointed the private claims administrator or settlement master.48 And, to varying degrees, 52.9% “approved” the private settlements despite having no formal statutory authority to do so.49

Approving settlements and appointing claims administrators imprints the deal with a judicial imprimatur. In the In re Fresenius GranuFlo litigation, Judge Douglas Woodlock observed, “it is nice to be asked to provide an imprimatur,” but “[it is] to some degree not my responsibility to provide imprimaturs to various kinds of things [like appointing a settlement master] . . . . I try not to exercise my jurisdiction when . . . it is the judicial equivalent of a blurb on the back of a new best seller.”50

But 52.9% of judges felt differently. Judicially endorsing a private deal legitimizes it. Yet none of those proceedings adversarially aired the deal’s pros and cons. Judges did not write an opinion reasoning through whether the settlement was fair, reasonable, or adequate, as they would in approving a class settlement. Nor did they decide whether lead lawyers adequately represented plaintiffs with conflicting interests.51 In some instances, judges blessed the deal before ever ruling on the merits: judges in only 50% of those thirty-four proceedings had made at least three merits-related rulings (summary judgment, Daubert motions, class certification, or presiding over a bellwether trial) before the first settlement and any subsequent Lone Pine order occurred.52 In four proceedings, judges issued post-settlement Lone Pine orders without making a single merits-related ruling beforehand.53

B. Lone Pine Orders in MDLs

The Pradaxa suits aptly illustrate how Lone Pine orders enter into MDL case management and settlement dynamics. In Pradaxa, Judge David Herndon never held a bellwether trial nor ruled on summary judgment, Daubert, or class certification.54 He urged the parties to settle and, once they did, issued a flurry of orders. He stayed the proceeding, which meant nonsettling plaintiffs could not continue discovery or have their cases remanded.55 He issued a census so that all attorneys with a case before him had to include eleven categories of information for all of their clients—regardless of where (or even whether) they had filed suit, and regardless of whether the client wanted to settle.56

Judge Herndon then appointed the private claims administrator and imposed additional discovery burdens on non-settling plaintiffs via a Lone Pine order.57 Under the order, non-settling plaintiffs had approximately fifteen days after the settlement’s opt-in deadline to produce fact sheets, affidavits, and pharmacy and medical records dating back to five years before their alleged injury occurred. In total, plaintiffs had a little over two months to comply with the Lone Pine order by producing an expert report on both general and specific causation. If they failed, they had just twenty days to fix the deficiency—a deadline that could not be extended.58

That is the Lone Pine order’s power. Perhaps now we can see those twilight orders for what they are in MDLs: a cleanup device, not a screening tool. Colorfully described by plaintiffs’ leadership as “a post-settlement mop-up procedure,” and by defendants as a “put up or shut up” device, settlement designers use Lone Pine orders to signal that plaintiffs should either settle or prepare to prove their claims immediately without the procedural protections of summary judgment.59

With Lone Pine orders positioned as MDL settlements’ billy club, I would go further than Engstrom and suggest that they should fail her test per se in the MDL context. She suggests that judges must use Lone Pine orders only as a last resort when no other tools suffice, or when credible evidence casts doubt on some or all of plaintiffs’ claims (or dilatory plaintiffs persist).60 But ready-made rules and statutory remands exist for all non-settling MDL plaintiffs. So, there is no need for Lone Pine orders to function as a gap filler.61 And, as Engstrom readily notes, “a plaintiff’s refusal to acquiesce to a global settlement should inform neither prong of the analysis.”62

Claims that remain post-settlement tend to stem from one of two kinds of plaintiffs: nonresponsive plaintiffs63 or plaintiffs who want a trial. Lone Pine orders do not distinguish between the two. They hit both with a heavy procedural burden that mimics summary judgment but lacks its safeguards. Separating these categories shows that existing procedures aptly address both situations, meaning that Lone Pine orders would not survive Engstrom’s first hurdle.

First, when nonresponsive plaintiffs refuse to settle, their cases should still be in discovery.64 If plaintiffs refuse to respond to discovery queries, Rule 37 supplies a series of steps and sanctions: attempts to confer in good faith, an order compelling disclosure or a response, then an array of possible sanctions that range from deeming disputed facts established to dismissing the action.

Second, when non-settling plaintiffs want to try their cases, they will respond to discovery requests and should be able to return to their original district once pretrial matters conclude. After all, multidistrict litigation is supposed to be for pretrial purposes only.65 In practice, however, only around 3% of all actions transferred have ever been remanded.66

As Judge Eduardo Robreno, who handled and remanded many of the asbestos cases, explained, “As a matter of judicial culture, remanding cases is viewed as an acknowledgement that the MDL judge has failed to resolve the case . . . .”67 That stigma should change. But vanity is no excuse for a Lone Pine order, especially when remanding cases can perform the same function. Once remanded, a defendant can move for summary judgment on specific causation and, if the plaintiff prevails, she can insist on her day in court.68

II. a critical look at plaintiff fact sheets

Siphoning claims pre-settlement falls to another judicially concocted order: plaintiff fact sheets. Unlike party-driven discovery devices in routine cases, this judge-issued order requires plaintiffs to provide defendants with specific information about their individual claims, circumstances, and injuries. As Figure 1 illustrates, all but one of the thirty-four proceedings (97%) used a plaintiff fact sheet. Engstrom suggests that streamlined fact sheets “have effectively culled meritless claims” with less expense and fewer disadvantages than Lone Pine orders.69

As I explain in this Part, however, I have reservations: like Lone Pine orders, fact sheets short circuit tried-and-true discovery rules as well as their built-in protections and limits. As they have evolved over time, some have become increasingly burdensome. Once again, that burden tends to fall disproportionately on plaintiffs.

Like Engstrom, I see some upside to narrowly tailored fact sheets: they give both sides data on the proceeding’s scope in terms of claims, damages, and numbers. When centralized into a repository that the judge can access, they can help select representative bellwether cases. And, once a plaintiffs’ steering committee decides which claims to develop and pursue, they can identify cases falling outside those parameters that should be remanded without having to wait for the MDL to run its course.70 Perhaps they winnow cases, too. But apart from a preliminary report by the Federal Judicial Center, there has been no systematic study documenting fact sheets’ use or effect.71

Overall, I have two concerns. First, I worry about the cumulative impact ad hoc procedures have on plaintiffs. And second, I worry about Engstrom’s recommendation that courts strengthen the consequences for plaintiffs who do not comply with fact sheet orders. As I explore in this Part, plaintiffs’ attorneys may not communicate with their clients as much as we would hope and even diligent plaintiffs may face barriers to obtaining their medical records. So, raising the stakes for noncompliance by subjecting plaintiffs to a Lone Pine order and then dismissing their claims with prejudice may be unjust.72

Like other ad hoc inventions, fact sheets lack standardization and are only loosely tethered to the discovery rules. They thus suffer from two of the pathologies that Engstrom identified with Lone Pine orders—they are inconsistent and unpredictable, and may be out of step with the formal procedural scheme.73

First, plaintiff fact sheets fluctuate in timing, name, compliance, content, and consequences. As plaintiffs face ever-changing burdens and deadlines, concerns arise about treating like plaintiffs alike. Although nearly all judges in my dataset ordered fact sheets, they did so at different times, ranging from forty-five to 1,084 days after the Panel centralized the proceeding. On average, judges required plaintiff fact sheets 248.8 days post-centralization,74 but called them different things. Some referred to fact sheets as plaintiff profile forms,75 which generally required less information (and were sometimes a precursor to a fact sheet), while others termed follow-on fact sheets a “supplemental discovery obligation,” which required an expert report (like a Lone Pine order).76

Plaintiff fact sheets likewise varied substantially on compliance deadlines, which spanned from thirty to 120 days.77 Although rolling deadlines make sense given that cases enter a multidistrict proceeding at different times, deadlines depended on everything from the first letter of the plaintiff’s last name,78 what “wave” or “phase” the plaintiff’s case fell into,79 and when the plaintiff filed her complaint,80 to the transfer date81 and whether the plaintiff entered a tolling agreement.82

As one would expect, fact sheets sought different information. In general, however, they requested (1) medical and pharmacy records, (2) personal background information, and (3) a plaintiff’s litigation history—from previous tort claims to employment lawsuits to bankruptcy.83 The first two categories appear tied to a hodgepodge of discovery rules, but the third makes little sense. Prior lawsuits unrelated to the current claim are not “relevant to any party’s claim or defense” under Rule 26, nor do they have a tendency to prove or disprove a fact that matters under substantive tort law, likely rendering them irrelevant under evidentiary rules.84 The only plausible inference a jury could draw from that information would be a forbidden and often unsupportable one: that this is an overly litigious person who probably has a frivolous claim.85

These idiosyncrasies raise the second concern: plaintiff fact sheets, like Lone Pine orders, can be out of step with the formal procedural scheme. Their requested content reads like a pancaked version of initial disclosures, interrogatories, requests to produce documents, and deposition queries, but without each rule’s prepackaged protections and limits. For instance, some judges deem fact sheets “interrogatories” under Rule 33, which imposes a twenty-five-question limit, including subparts. But fact sheets often exceed 100 questions and seek information that one would ordinarily expect to convey when deposed or to send as part of Rule 26’s initial disclosures.86

In a run-of-the-mill case, parties must send initial disclosures at least fourteen days after the Rule 26(f) conference, which occurs “as soon as practicable.”87 And then, once served with interrogatories and requests to produce documents, parties have thirty days to respond, unless the court orders otherwise.88 In the usual sequencing, then, most plaintiffs will know several months in advance what lies ahead and what documents they will likely need to gather. But in an MDL, their case might be transferred as a “tag-along” action such that an order for fact sheets greets them upon arrival. Most judges take care to give these new arrivals longer deadlines, and rightfully so, but there are no guarantees.

Timing can matter, especially when it comes to medical records and device injuries, which can entail multiple explant surgeries, specialists, and years of follow-up care. One pelvic-mesh plaintiff explained that she had seen over seventy-five different doctors in thirteen years and had eight mesh-related surgeries.89 And a 2018 study on U.S. News & World Report’s top-ranked hospitals found that even patients with less-extensive medical histories can face “long waiting periods and unclear request processes” for their records.90 While some hospitals made electronic and paper records available almost immediately, others took four to eight weeks and charged over $280 for a 200-page record.91 Even among these top hospitals, the report found “a lack of transparency” in the request process, noncompliance with HIPAA’s recommended fees, and “processing times longer than the state-required time.”92

When ongoing treatment and records requests pose obstacles to gathering the information that fact sheets seek, lenience—not stricter penalties for noncompliance—should be the norm. Yet the consequences for noncompliance fluctuated substantially.93 Most judges gave plaintiffs a chance to “cure” any errors or omissions; but others were more draconian, allowing defendants to move to dismiss plaintiffs’ claims (without prejudice) thirty days after notifying them.94

As a discovery tool, fact sheets should be governed by Rule 37, which sets up a predictable—yet flexible—sequence: defense counsel would move to compel only after certifying that she first conferred with opposing counsel in good faith. This, one hopes, eliminates motions to compel in cases where plaintiffs encounter roadblocks beyond their control. If defendants do move to compel, noncompliant plaintiffs’ counsel would have an opportunity to respond and be heard. Afterward, if a plaintiff still does not comply, the court may choose from a menu of options ranging from striking pleadings to dismissing the action in whole or in part.95

As fact sheets evolve through norms that operate in the shadows of formal discovery rules, it is not just their variety and unpredictability that is worrisome.96 Over time, as lawyers simply add to what is already there, the questionnaires can look more like a museum storeroom and less like a streamlined discovery tool. An early fact sheet in the Silica litigation sought just a few pages of basic questions,97 but some recent examples span forty-eight pages (with forms and all), exceed 100 questions, and seek fifteen years of medical history, ten years of employment history, and information on everything from divorces to children’s names, addresses, and birthdays.98

With increased length, fact sheets’ aim may shift from discovery to harassment—something that Engstrom cautions against and that the federal rules committee explicitly considered when limiting parties to twenty-five interrogatories.99 Plaintiffs’ education levels vary and not all receive attorney assistance. As representing mass-tort clients has turned into a wholesale business for many lawyers, some plaintiffs have never even spoken with their attorney. Third-party vendors (often owned by plaintiffs’ lawyers) have popped up to fill the gap, promising that “our Physicians in India prepare Plaintiff Fact Sheets for $35 an hour as a billable expense.”100 So, by using a third-party service, the money eventually comes out of plaintiffs’ pockets as a cost rather than attorneys’ contingency fees, thereby impacting plaintiffs’ bottom line. In short, while plaintiffs’ fact sheets aim to streamline discovery, in practice, they can bypass built-in discovery protections and impact parties disproportionately.


It is true that multidistrict litigation should not abrogate defendants’ ability to discover plaintiff-specific information or the basic tenet that plaintiffs must prove their claims. But, as Engstrom acknowledges, judges should be mindful about overtaxing plaintiffs too—whether through Lone Pine orders or plaintiff fact sheets.101 Procedural justice dictates that procedural burdens should not fall disproportionally on one party. Rather, rules should distribute the risks of error and the costs of access as evenly as possible.102 Rule 26 strives to accomplish this by requiring that discovery remain proportional to a case’s needs. But as hand-me-down procedures like plaintiff fact sheets and Lone Pine orders spread through common-law norms and escape formal rules’ rigorous vetting process, parity may dissolve.103

Where does all of this leave us? On the one hand, transferee judges are right: every mass-tort proceeding differs in some respect, which suggests the need for flexibility over rigidity. On the other hand, when judges and parties invent procedures for a particular proceeding, they risk focusing on the pragmatics of that case, overlooking normative issues, and importing the baggage of their biases and thoughts about the proceeding. Achieving neutrality, where discovery burdens and the risks of erroneous decisions fall equally on both parties, may be hard in the heat of battle. Although design choices should respond to what is happening on the ground, issuing a Lone Pine order or a plaintiff fact sheet does not occur in a vacuum; those decisions have ripple effects that can impact plaintiffs more harshly than defendants.104 If judges experiment with bespoke procedures, they should take care to ensure a fair fit for both parties that works with—not against—off-the-rack rules.

Fuller E. Callaway Chair of Law, University of Georgia School of Law. I am grateful to Thomas Burch, Nora Freeman Engstrom, and Margaret S. Williams for their feedback on previous drafts. Most of the court documents I cite in this Essay are available at MDL Docs, U. GA. SCH. L., [].


table a1

event occurrence dates within the mdl products-liability dataset


table a2

judicial assignments and resolutions in the mdl products-liability dataset105106