VINCE CHHABRIA, United States District Judge

PRETRIAL ORDER NO. 236: GRANTING IN PART AND DENYING IN PART MOTION TO ESTABLISH A HOLDBACK PERCENTAGE

Re: Dkt. No. 12394

VINCE CHHABRIA, United States District Judge

Lead counsel in this MDL has asked the Court to order that, whenever anyone in the country recovers money from Monsanto based on allegations that its Roundup product caused their cancer, 8.25% of their recovery be held back and placed into a fund to compensate lawyers who took the lead in litigating against Monsanto. Lead counsel's motion does not merely seek to tax the recoveries of plaintiffs in this MDL; it also seeks to tax the recoveries of state court plaintiffs, and the recoveries of people who settled with Monsanto before filing any lawsuit at all. The motion does not merely seek to compensate lawyers who performed "common benefit work" in this MDL; it also seeks to compensate lawyers for doing work in state courts.

The motion is largely denied. The fact that counsel is even requesting such a far-reaching order—a request that has some support from past MDL practice—suggests that courts and attorneys need clearer guidance regarding attorney compensation in mass litigation, at least outside the class action context. The Civil Rules Advisory Committee should consider crafting a rule that brings some semblance of order and predictability to an MDL attorney compensation system that seems to have gotten totally out of control.

I. BACKGROUND

In 2015, the International Agency for Research on Cancer (IARC) classified glyphosate as "probably carcinogenic to humans." IARC's conclusion stemmed from scientific studies that found an association between glyphosate exposure and a particular type of cancer : non-Hodgkin lymphoma (NHL). Glyphosate is the active ingredient in Roundup, the popular weedkiller manufactured and sold by Monsanto.

The IARC classification inspired a wave of lawsuits against Monsanto by plaintiffs with NHL, in federal and state courts across the country. As of today—roughly five years later—a few thousand cases have landed in federal court. In state courts, there are tens of thousands of cases. Many more potential cases wait in the wings, with people hiring lawyers but trying to settle before filing a lawsuit. Public reports filed by Bayer (which acquired Monsanto after the IARC classification) state that well over a hundred thousand cases have either been filed or are in the works.

When multiple similar cases are filed in federal courts throughout the country, those cases are sometimes transferred to a single federal judge for pretrial purposes. The transfer decision is made by the Judicial Panel on Multidistrict Litigation. When the panel decides that all cases of a particular type should be transferred to a single judge, lawyers describe it as the creation of an "MDL" (for multidistrict litigation). Relatively soon after IARC announced its glyphosate classification, the panel decided to create an MDL for federal Roundup cases and ordered that all such cases be transferred to the Northern District of California, and specifically to the undersigned judge.

Most of the state court cases are in Missouri, where Monsanto is headquartered. The California state courts have the second-highest number of Roundup cases. The judiciaries in those states have taken similar measures to group the cases together, with individual judges being assigned large chunks of cases.

Because all the cases in a federal MDL present similar or identical issues, the MDL judge's pretrial decisions—whether they involve case management or substantive legal analysis—will typically impact all the plaintiffs. But the MDL judge cannot manage or adjudicate the cases effectively if the lawyers from every case have an active role. Accordingly, an MDL judge's first order of business is often to decide which lawyers will take the lead in managing and litigating the cases. This is an important decision because the performance of those lawyers, and the strategic decisions they make, often affect the outcome for the entire group of plaintiffs. For lawyers, leading an MDL is a major responsibility and a major risk. Lead lawyers invest a lot of time and money into managing and litigating the MDL. And if they lose, there is no way to recover their investment from the attorneys and plaintiffs who sat back and watched.

In this MDL, it was not difficult to decide which lawyers should take the lead because only one group came forward, presenting themselves as a slate. There were eight lawyers in total, all from different firms. The Court adopted the slate as proposed. The lawyers assigned themselves various titles, but for the purposes of this discussion it suffices to refer to them all as "lead counsel." Upon being appointed, these lawyers became responsible for managing the MDL on the plaintiffs’ side, conducting discovery, communicating with other plaintiffs’ lawyers who have cases in the MDL, determining the substantive legal positions to be taken, litigating those positions, and engaging in settlement discussions.

Shortly after being appointed, lead counsel filed a motion to establish a "common benefit fund." Common benefit funds are often created in MDLs so that if lead counsel succeeds in a way that benefits all the plaintiffs, they can be compensated from the fund. A certain percentage is held back from the recoveries of plaintiffs and placed in the fund. Then, towards the close of the MDL, lead counsel typically makes a showing to the Court related to the value of their work and the amount they believe they're owed from the fund.

It bears emphasis that these MDL common benefit funds are not necessarily for the compensation of lead counsel alone. Any counsel who (in coordination with lead counsel) ends up performing "common benefit work" can typically make a claim for compensation. The judge, often with the assistance of a court-appointed "special master," assesses the various claims for compensation and decides how the money should be distributed. Furthermore, the requirement that the defendant hold back a percentage of a plaintiff's recovery does not necessarily preclude that plaintiff or their lawyer from ultimately getting it, or a portion of it. The money is simply "held back" to make sure that any lawyer who performed work on behalf of all plaintiffs is fairly compensated.

From the description so far, one might assume that lead counsel's common benefit fund motion sought a holdback from any plaintiff within the MDL who obtained a settlement or judgment from Monsanto. But the range of people lead counsel sought to affect through the proposed holdback was far broader. And to be candid, this Court did not adequately scrutinize lead counsel's proposal—the motion was unopposed at the time, and the Court was not very familiar with the nuances of MDL proceedings. Thus, the Court signed (with just one change) the proposed order submitted by lead counsel. The order requires Monsanto to hold back a percentage of a settlement or judgment from any of the following people who obtain a recovery:

• Any plaintiff in the MDL.

• Any person not in the MDL who recovers from Monsanto, if their lawyer also happened to separately represent a client in the MDL. This includes someone who filed suit in state court and obtained a judgment or settlement. And it includes someone who settled before filing any lawsuit at all.

• Any person not in the MDL who recovers from Monsanto, if their lawyer has signed an agreement with MDL lead counsel to be subject to the holdback order in exchange for access to "MDL work product." Again, this includes someone who filed suit in state court and obtained a judgment or settlement, and it includes

someone who settled before filing any lawsuit.

Pretrial Order No. 12: Common Benefit Fund Order, In re Roundup , No. 16-md-2741 (N.D. Cal. Feb. 22, 2017), ECF No. 161.

The one change to the proposed order involved the percentage of the recovery to be held back. Although lead counsel proposed that it be 7% of the gross amount recovered in a settlement or judgment, Monsanto argued that it was premature to set a percentage so early, before the Court could begin to conduct a meaningful assessment of the value of the work performed by lead counsel and others on behalf of other plaintiffs. The Court accepted Monsanto's suggestion. The result was an order that laid the groundwork for the creation of a common benefit fund and specified who would be subject to it, but that left for another day a decision about the amount to be held back.

Thus began the litigation. Lead counsel's first big responsibility was to litigate the question of "general causation," which involved presenting expert testimony and legal argument about whether a reasonable jury could conclude that Roundup is capable of causing NHL at exposure levels that people currently experience. If the plaintiffs were to lose on that threshold question, every single case in the MDL would be dismissed. This phase necessitated a tremendous amount of fact discovery and expert discovery. Each side hired numerous experts, and many of those experts testified at an evidentiary hearing that lasted over a week. Most of the attorney work on the plaintiffs’ side was performed by people from the lead counsel group (or people from their firms). Ultimately, the Court concluded that several of the expert opinions put forward by the plaintiffs were admissible, which meant that Monsanto's motion for summary judgment on the issue of general causation was denied.

Next, the parties and the Court identified three cases in the MDL that would serve as "bellwethers"—cases to be tried in the hope of educating both sides about their prospects in future cases, which in turn could facilitate settlement. Additional discovery took place, primarily focused on specific causation—that is, whether a reasonable jury could conclude that Roundup caused the NHL of the bellwether plaintiffs. The Court again heard extensive testimony from the experts on specific causation. Monsanto sought summary judgment in all three cases, arguing that no reasonable jury could conclude that Roundup caused NHL for these three particular plaintiffs. Monsanto also made a legal argument that affected all cases in the MDL: it contended that the plaintiffs’ claims—which were based on California law—were preempted by federal law. The Court denied Monsanto's motion.

Meanwhile, a plaintiff in California state court, Dewayne Johnson, managed to get his case tried on an expedited basis. This was the first trial ever about whether Roundup caused someone's NHL. Johnson was represented by attorneys from the MDL's lead counsel group. A jury found in his favor, awarding him $289 million in compensatory and punitive damages. The trial court reduced both compensatory and punitive damages, and the court of appeal reduced those damages even further, leaving a total recovery of more than $20.5 million.

The following year, the first of the three federal bellwether cases went to trial in this Court. The plaintiff was Edwin Hardeman. One of the two lawyers who stood up at trial was a member of the lead counsel group; the other was not. The jury awarded Hardeman roughly $5 million in compensatory damages and $75 million in punitive damages. This Court reduced the punitive damages award to $20 million, leaving Hardeman with a total award of roughly $25 million.

In the aftermath of this trial, the Court determined that pressing ahead immediately with the other two bellwether trials would not be the best way to manage the MDL. Monsanto had suffered resounding defeats in two trials, and a third trial was set to proceed in California state court in the coming months. Thus, Monsanto and its opponents would soon have the benefit of three bellwethers. Furthermore, the Hardeman trial was structured in a manner that could not have been more fair to Monsanto. Among other things, the trial was bifurcated between a causation phase and a damages phase. This meant that the jury was asked to render a verdict on the scientific evidence relating to causation before it was presented with a good deal of damning evidence against Monsanto—evidence which suggested that Monsanto never seemed to care whether its product harms people. The fact that the jury found causation despite this trial structure, and the fact that the jury decided to award $80 million despite the fact that Hardeman had recovered from his NHL, suggested that Monsanto's prospects in future trials were exceedingly dim and that there would not be much more to learn from another trial. The Court thus vacated the trial date for the next federal bellwether and began focusing on pretrial proceedings for the remaining cases in the MDL, so that those cases could be transferred back to their "home districts" for trial. The Court also appointed renowned mediator Ken Feinberg to preside over settlement negotiations between Monsanto and the plaintiffs.

The next trial in California state court took place shortly thereafter. The plaintiffs were a husband and wife named Alva and Alberta Pilliod. Lead counsel from the MDL was involved in the trial. It was not bifurcated. The trial went even worse for Monsanto than the first two—the jury awarded each plaintiff $1 billion in punitive damages, to go along with $55 million in compensatory damages. The trial court ultimately reduced the total award to roughly $87 million.

Back in federal court, the parties and the Court continued to move the MDL cases through the pretrial stage. Settlement negotiations continued as well. Ultimately, in June 2020, Bayer issued a press release announcing "agreements to resolve major legacy Monsanto litigation," including and especially the Roundup litigation. The press release stated that Bayer would "make a total payment of $10.1 billion to $10.9 billion ... to resolve current and address potential future Roundup litigation."

From this press release, people might have assumed that Bayer had agreed to establish a fund that would compensate all plaintiffs (and future plaintiffs) who could make some type of threshold showing that they used Roundup and developed NHL. People might also have assumed that a deal was squarely in place. But in reality, there was there was no global settlement, and no plan for a global settlement fund. It's not even clear whether Bayer had formally reached settlements with anyone up to that point. Instead, Bayer was in the process of negotiating agreements with individual law firms—agreements to settle what people in the MDL world often refer to as a firm's "inventory" of cases.

From the June 2020 press release until now (a year later), Bayer has been working to finalize settlements with law firms. This includes settlement of cases in the federal MDL, cases in state court, and prospective cases for which no suit has been filed. But no money has yet changed hands. This may be partly due to the fact that a holdback percentage for the common benefit fund has not yet been set. So now, lead counsel has filed a motion to establish the percentage. The motion differs in two key ways from what lead counsel requested at the beginning of the MDL. First, lead counsel asks that the holdback percentage be 8.25% of the total amount of each recovery, rather than 7%. The motion posits that 8% would represent attorneys’ fees and be used to compensate attorneys who did common benefit work, while .25% would represent litigation costs and be used to compensate plaintiffs who incurred such costs to the benefit of everyone.

Second, lead counsel asks that the holdback order apply even more broadly than previously requested. They now ask the Court to issue an order requiring that, whenever anyone with NHL anywhere in the country obtains a recovery from Monsanto based on their use of Roundup, Monsanto hold back 8.25% of their total recovery and contribute it to this common benefit fund. This includes plaintiffs in the federal MDL, plaintiffs in state court, and people who haven't yet filed lawsuits but are negotiating deals with Monsanto. Any such person would be subject to the holdback order regardless of whether their lawyer was involved in the MDL or used MDL work product. While the original motion was unopposed, the new motion has drawn multiple objections from lawyers who have settled, or are in the process of settling, their cases. The objections are not limited to this recent attempt to expand the scope of the holdback order; they are targeted at several aspects of the original order as well.

II. THE SOURCE AND SCOPE OF A DISTRICT COURT'S POWER TO CREATE, AND ORDER PAYMENTS INTO, A COMMON BENEFIT FUND.

The breathtaking nature of lead counsel's request, along with the many objections filed in response to it, have prompted the Court to go back and ask: what is the source of a district court's power to create and require contributions to a common benefit fund, and how far does that power reach? It's helpful to break up the discussion into the following categories of people for whom lead counsel requests a holdback:

• Any plaintiff in the MDL who recovers from Monsanto following a settlement or judgment.

• People outside the MDL who recover from Monsanto following a settlement or judgment, if their attorney happens to also represent a client in the MDL. This could include people who filed lawsuits in state court, or people who settled before filing any lawsuit.

• People who recover from Monsanto (whether in a state court lawsuit or before filing any lawsuit), if their attorney signed an agreement with lead counsel in the MDL to be bound by a holdback order in exchange for access to MDL work product.

• All other people who have recovered, or will soon recover, from Monsanto, regardless of any connection to the MDL.

A. Plaintiffs In The MDL

There is not much disagreement about whether a district court has the power to create a common benefit fund in the MDL and order holdbacks from the recoveries of MDL plaintiffs. There is, however, some ambiguity as to where that power comes from. Identifying the source of the power is important because it is relevant to the question of how far that power can reach.

1. The MDL Statute. The first place one might look is the statute that created MDLs in the first place: 28 U.S.C. § 1407. But section 1407 merely creates the Judicial Panel on Multidistrict Litigation and authorizes the panel to transfer similar cases from around the country to a single federal judge for pretrial purposes. It is a procedural statute—not one that expands the substantive powers of the district court assigned to handle the cases. In re Genetically Modified Rice Litigation , 764 F.3d 864 (8th Cir. 2014) (citing In re Showa Denko K.K. L–Tryptophan Products Liability Litigation–II , 953 F.2d 162, 165 (4th Cir. 1992) ). Accordingly, if a district court has the power to require that a percentage of recoveries be deposited into a common benefit fund in an MDL, that power must exist independently of section 1407. Which means, incidentally, that although this ruling speaks in terms of the power of an MDL court, the same concepts apply outside the MDL context (for example, if one district judge happens to be handling dozens of related cases filed within the same district).

2. The Common Fund Doctrine. One doctrine commonly invoked to support the creation of common benefit funds in MDLs is the "common fund" doctrine. This doctrine stems from cases holding that district courts may, in rare circumstances, invoke their equitable power to award attorneys’ fees to successful plaintiffs. Indeed, the proposed order signed by this Court at the outset of the MDL cited this line of cases. On reflection, despite the superficial similarity based on the name of the doctrine, the common fund cases are a questionable source of support for creating and requiring contributions to a fund in a situation like this.

The default rule in the United States is that each side pays its own attorneys’ fees, regardless of who ultimately wins the case. Sometimes Congress will adopt a statute authorizing federal courts to award fees to the winning side in a particular type of case, or in a particular type of situation. But absent a statute, courts have historically only been permitted to require the payment of attorneys’ fees in three circumstances: (i) when a party has acted in bad faith; (ii) when a party has willfully disobeyed a court order; or (iii) when the "common fund" exception applies. Alyeska Pipeline Service Co. v. Wilderness Society , 421 U.S. 240, 259, 95 S.Ct. 1612, 44 L.Ed.2d 141 (1975).

The common fund exception recognizes that a plaintiff who litigates and recovers a specific piece of property or pot of money—one that benefits a discrete group of people in addition to themselves—is entitled to reasonable attorneys’ fees from that recovery. The common fund line of cases typically involve a plaintiff recovering or preserving some money or property—for example, a trust—which can be loosely understood as a "fund." A similarly situated group of people typically has an interest in that fund, and the judgment obtained by the plaintiff paves the way for each person to realize that interest. In such circumstances, the Supreme Court has held that the prevailing plaintiff may seek, and the district court may order, reimbursement of attorneys’ fees and litigation costs from the fund produced by the plaintiff's litigation. Such reimbursement works to spread the plaintiff's fees and costs proportionately among those who benefited from the lawsuit, preventing the plaintiff from shouldering the entire cost of litigating alone. The authority to affect the rights of people who are not parties to the case comes from the equitable jurisdiction that the district court exercises over the fund, by way of the litigation. See, e.g., Boeing Co. v. Van Gemert , 444 U.S. 472, 478, 100 S.Ct. 745, 62 L.Ed.2d 676 (1980) ; see also Trustees v. Greenough , 105 U.S. 527, 26 L.Ed. 1157 (1882) ; Central Railroad & Banking Co. v. Pettus , 113 U.S. 116, 5 S.Ct. 387, 28 L.Ed. 915 (1885). One helpful example is Sprague v. Ticonic National Bank , 307 U.S. 161, 59 S.Ct. 777, 83 L.Ed. 1184 (1939). In that case, Lottie Sprague made a bank deposit that was backed by certain earmarked bonds. Fourteen other deposits with that bank were backed by the same bonds. The bank went into receivership, putting these deposits in jeopardy. Sprague sued and successfully imposed a lien on the bond proceeds. The consequence for Sprague was that the bank would be precluded from distributing the bond proceeds to unsecured creditors before satisfying its obligation to her. Similarly, her successful lawsuit established a claim from the same bond proceeds for the fourteen other depositors. In light of these circumstances, the Supreme Court held that the district court had the power to award Sprague reimbursement of her attorney's fees and litigation expenses from the bond proceeds: it was a situation in which the plaintiff effectively "produced" a fund through the successful result she achieved, and the fund was "for all practical purposes created for the benefit of others." Id. at 167, 59 S.Ct. 777.

This line of cases clearly justifies a fee award for common benefit work by MDL lawyers in one circumstance: when the MDL results in a settlement where the defendant establishes a fund for paying claims made by similarly situated people. Cf. Boeing Co. v. Van Gemert , 444 U.S. 472, 100 S.Ct. 745, 62 L.Ed.2d 676 (1980). With that type of global settlement, the lawyers (and their clients) have invested time and money to secure a fund for the benefit of the group, and the district court has the power to require that a portion of any payment from that fund be earmarked for the compensation of the lawyers who did the work that led to its creation. (Or, as typically happens, the court can make it simpler and compensate the lawyers by awarding them money directly from the fund.) Of course, questions could arise about whether the attorneys who did the common benefit work (or their clients) would get an unfair windfall from common fund payments. But as discussed later in this ruling, that is a matter of discretion; it seems clear that, pursuant to the common fund doctrine, an MDL court has the power to order such payments if discretion warrants it.

It seems like a stretch, however, to apply the common fund doctrine to the type of arrangement contemplated in this case (and in many other MDLs). Lead counsel has not asked this Court to reimburse them from some fund or property interest that was preserved or established as a result of a victory against, or settlement with, Monsanto. Rather, lead counsel asked the Court to create a fund at the outset—before any judgment was entered and before liability was established—and to require contributions from individual recoveries down the line, whenever they were reached. Exercising equitable jurisdiction over a fund ultimately secured by litigation to reimburse the plaintiff who secured it seems quite different from creating a fund at the beginning of litigation and requiring contributions to that fund from individual recoveries in anticipation of the MDL lawyers needing to be reimbursed.

The distinction between these two concepts is highlighted by Grupo Mexicano de Desarrollo S.A. v. Alliance Bond Fund, Inc. , a Supreme Court case involving a somewhat analogous situation. 527 U.S. 308, 119 S.Ct. 1961, 144 L.Ed.2d 319 (1999). There, the Court held that, although a district court has equitable authority to impose a lien on the defendant's funds or property after judgment is entered to ensure that the judgment can be effectuated, a district court lacks the equitable authority to enjoin a defendant from disposing of assets while the case is still pending—even if it's likely that the plaintiff will prevail, and even if there's a serious risk that the assets will disappear. Relatedly, the Court emphasized in Grupo Mexicano that district courts cannot exercise their equitable powers beyond situations where courts have historically invoked them, even if some of the same principles apply to the new situation. As noted, preemptively creating a common benefit fund and requiring holdbacks seems quite different from the historical practice of awarding fees from a fund secured by a successful plaintiff in the litigation.

The link courts often draw between the traditional common fund cases and the type of arrangement proposed here is that, in both contexts, there are concerns about free riding. Those concerns arise when people sit back and watch one plaintiff work hard (and spend hard) to obtain a favorable result, only to swoop in and benefit from that result in the end. See, e.g., In re General Motors LLC Ignition Switch Litigation , 477 F. Supp. 3d 170, 180-82 (S.D.N.Y. 2020). It's true that free rider concerns exist in both contexts. But free riding occurs all across our legal system. The presence of free riding, and the desire to address it, is not itself a source of judicial power. The power of a court to address free riding concerns—just like the power of a court to address any concern—must actually come from somewhere. As it relates to the type of holdback order contemplated in this MDL, that power does not seem to come from the common fund doctrine.

Speaking of the free rider problem, it's worth noting another difference between the traditional common fund cases and the type of arrangement that lead counsel has proposed here. The common fund doctrine was created against the backdrop of the rule that winning plaintiffs typically must pay their own attorneys’ fees. Thus, if not for the common fund exception, if a plaintiff stepped up to secure a fund without being compensated by the other beneficiaries of the ruling, those beneficiaries would be unjustly enriched at the expense of the plaintiff. The plaintiff had to pay their attorneys; the other beneficiaries did not. Thus, the common fund exception is largely about making sure that the plaintiff is reimbursed for fees. Cf. United States v. Tobias , 935 F.2d 666, 668 (4th Cir. 1991) ("Generally, a fund claimant who is represented by counsel ... is deemed not to have taken a ‘free ride’ on the efforts of another's counsel."). With the arrangement proposed by lead counsel here, it does not appear that any actual plaintiffs would recoup any attorneys’ fees as a result of common benefit work their lawyers performed. As far as the Court can tell, regardless of whether a holdback is ordered in this MDL, plaintiffs whose lawyers performed common benefit work will see the same contingency fee extracted from their recovery pursuant to their contracts with their lawyers. Thus, the proposed common benefit fund and holdback order will do nothing to ease the financial pain that any plaintiff will feel when their lawyer takes a percentage (typically 40%) of the recovery. Rather, it is the lawyers who will receive additional money, over and above the 40% of their clients’ recoveries. That's not to say it's categorically inappropriate for lawyers who do common benefit work to get paid more than the contingency fee they negotiated with their own clients—more on that later. But it further highlights the degree to which this type of arrangement differs from the traditional common fund cases.

Notwithstanding all this, an old Ninth Circuit case arguably requires a district court to treat a holdback order of this type as authorized by the common fund doctrine. In Vincent v. Hughes Air West, Inc. , 557 F.2d 759 (9th Cir. 1977), the Court considered (among other important things discussed later in this ruling) whether an MDL court had the authority to create a fund and order a holdback from the recoveries of MDL plaintiffs. The Court ultimately concluded that there was no meaningful difference between a plaintiff securing a fund at the end of a case (like in Sprague ) and a court creating a fund and then using its existence to justify a holdback from individual recoveries. Rejecting the argument that conflating these two concepts is "bootstrapping of the rankest order in that an attorney and the court create by compulsion a fund which in turn they use to trigger application of the common fund doctrine," the court concluded that, "[t]aking a broader view ... lead counsel helped ‘create’ a fund consisting of the various settlements negotiated in the [MDL]." Id. at 770-71. Vincent ’s conclusion on this issue may have been indirectly invalidated by the Supreme Court's reasoning in Grupo Mexicano . But for purposes of this ruling, this Court will assume that the common fund doctrine gives it the authority to require a holdback for recoveries—at least when the recoveries are obtained by plaintiffs in the MDL itself.

3. The district court's inherent authority to manage its docket. Perhaps, in light of Vincent , there is no need for further discussion of the power to order holdbacks from the recoveries of plaintiffs in an MDL. But because Vincent ’s reasoning on this issue is arguably undermined by subsequent Supreme Court precedent, it seems worth exploring the matter further. Moreover, as discussed in the next section, understanding the true source of this power helps elucidate how far a holdback order can reach.

District courts may exercise power in furtherance of the efficient and speedy resolution of cases. This authority, the Supreme Court has explained, is "governed not by rule or statute but by the control necessarily vested in courts to manage their own affairs ...." Link v. Wabash R.R. Co. , 370 U.S. 626, 630-31, 82 S.Ct. 1386, 8 L.Ed.2d 734 (1962). "It has long been understood that certain implied powers must necessarily result to our Courts of justice from the nature of their institution, powers which cannot be dispensed with in a Court, because they are necessary to the exercise of all others." Chambers v. NASCO, Inc. , 501 U.S. 32, 43, 111 S.Ct. 2123, 115 L.Ed.2d 27 (1991) (citing United States v. Hudson , 7 Cranch 32, 34, 11 U.S. 32, 3 L.Ed. 259 (1812) ).

The Fifth Circuit—in a case decided the same year that the Ninth Circuit decided Vincent —relied primarily on this concept in concluding that a district court has the authority to require, for any plaintiff in the MDL who might settle with the defendants, a holdback to compensate attorneys who did common benefit work. In re Air Crash Disaster at Florida Everglades , 549 F.2d 1006 (5th Cir. 1977). As the Fifth Circuit noted, district courts handling a mass of related cases have the power to appoint lead counsel to manage the litigation. That power is rooted in a district court's inherent authority to efficiently manage the cases on its docket because, without lead counsel, hundreds of diligent lawyers would lobby the court for attention, leaving it unable to manage and adjudicate the related cases, not to mention the rest of its docket. See id. at 1012-15 ; see also Vincent , 557 F.2d at 773-75. It follows, the Fifth Circuit Court explained, that a district court can order additional compensation for lead counsel (beyond whatever contingency fee they receive from their own clients) based on the work they do for the collective whole. In re Air Crash Disaster , 549 F.2d at 1016.

In terms of finding the right source of authority for holdbacks, the inherent managerial authority is a much better fit than the common fund doctrine—at least when it comes to holdbacks for plaintiffs who are actually part of the MDL. As Judge Furman nicely put it, "A subset of plaintiffs’ lawyers do the lion's share of the work, but that work accrues to the benefit of all plaintiffs. If those other plaintiffs were not required to pay any costs of that work, high-quality legal work would be under-incentivized and, ultimately, under-produced." In re General Motors , 477 F. Supp. 3d at 174. Mass tort cases are difficult enough to manage and adjudicate when the common benefit work is being done by good lawyers; the cases would be impossible to manage if good lawyers lacked sufficient incentive to step up.

It's probably wrong to say—as the Fifth Circuit and others have—that lead MDL counsel would essentially be doing volunteer work absent a common benefit fund and a holdback order. In re Air Crash Disaster , 549 F.2d at 1016. After all, once you've been appointed lead counsel and have taken control of the MDL, you and your clients will typically have much greater bargaining power compared to other plaintiffs. So you can often expect to obtain a more lucrative settlement for your clients (and by extension, for yourself) than the lawyers who are waiting on the sidelines and not bringing bellwether cases to trial. The differential between what you recover and what the free riders recover may already be adequate compensation for the common benefit work that you did. But it remains safe to assume that if a district court lacked the ability to ensure that lead counsel can receive compensation from the other plaintiffs in the MDL when the situation warrants it, good counsel would be hard to find in at least some cases. Good counsel may not want to gamble that their recoveries will be so much greater as to justify doing a large amount of common benefit work.

Accordingly, when a district court is handling an MDL or some other group of related cases, it has the authority to compensate common benefit work by redistributing attorneys’ fees in those cases, to the extent necessary to ensure effective management and adjudication of the litigation. B. A Person With No Case In The MDL Who Recovers From The Defendant And Whose Attorney Happens To Represent Someone In The MDL.

The next question is whether the Court has the authority to require holdbacks from the recoveries of people who are not plaintiffs in the MDL but who happened to hire a lawyer with a client in the MDL. This could include someone who filed a lawsuit in state court and recovered from Monsanto following a judgment or settlement, and it could include someone who filed no lawsuit and settled out of court. The answer is that a district court cannot exert authority over the recovery of a person with so tenuous a connection to the MDL. Neither the common fund doctrine nor the district court's inherent power to control its docket justifies an order affecting the recovery of a nonparty merely because they happened to hire a lawyer with a client in the MDL.

As discussed earlier, the connection between the common fund doctrine and the type of holdback order contemplated here is tenuous at best. If it's tenable at all, it's because one similarity persists among many differences: the transaction takes place within the case itself. In the common fund cases, the district court is awarding money from a fund that is part of the case to a plaintiff who is part of the case. If the holdback order were extended to the recoveries of parties in state court or parties with no case at all, any plausible support from the common fund doctrine disappears entirely—the court is exercising authority over disputes that are not before it and over property that is not before it, without any clear limit on the court's reach. Cf. Alyeska Pipeline , 421 U.S. at 264 n. 39, 95 S.Ct. 1612 ("In this Court's common-fund and common-benefit decisions, the classes of beneficiaries were small in number and easily identifiable.").

To return to the Supreme Court's common fund decision in Sprague as an example, imagine if one of the other depositors had sued the bank in state court and then reached a settlement while Ms. Sprague's case was still pending. Imagine further that the bank paid the state court plaintiff with money from its general fund, not from the proceeds of the earmarked bonds. Now imagine what would have happened if Ms. Sprague asked the district court to order a portion of the settlement amount withheld and placed into a fund to potentially reimburse her for the fees she was in the process of incurring—just in case she was ultimately successful in her own lawsuit. It seems likely that the Court would have reached a different conclusion about the power of the district court in a situation like that.

If one understands the power to order holdbacks as coming from the inherent docket management authority, it becomes even more obvious that a district court lacks the authority to apply holdbacks to a dispute outside the MDL merely because the plaintiff in that dispute happened to hire a lawyer who represents a plaintiff within the MDL. Under this doctrine, district courts merely have the power to manage "their" dockets, and the purpose of this power is "the efficient and expedient resolution of cases." Dietz v. Bouldin , 579 U.S. 40, 136 S. Ct. 1885, 1892-93, 195 L.Ed.2d 161 (2016) ; see also United States v. Moussaoui , 483 F.3d 220, 237 (4th Cir. 2007) (explaining that the cases involving the inherent managerial power of a district court "stand for the proposition that a court has the inherent authority to control various aspects of the cases before that court so that they can ‘protect their proceedings and judgments in the course of discharging their traditional responsibilities.’ " (emphasis in original) (quoting Degen v. United States , 517 U.S. 820, 823, 116 S.Ct. 1777, 135 L.Ed.2d 102 (1996) )). It's one thing to require holdbacks in your own cases for the purpose of ensuring those cases are litigated and adjudicated properly; it's quite another to insist that you need to manage your docket by issuing orders affecting disputes that are not before you. See Degen , 517 U.S. at 823-24, 116 S.Ct. 1777 ("Principles of deference counsel restraint in resorting to inherent power, and require its use to be a reasonable response to the problems and needs that provoke it.").

Perhaps these are some of the reasons why the Eighth Circuit, in Genetically Modified Rice Litigation , 764 F.3d 864 (8th Cir. 2014), tersely affirmed a ruling by an MDL court that it lacked the authority to order holdbacks from state court plaintiffs’ recoveries. Lead counsel argued that the authority came from the fact that the court would be ordering the defendant in the MDL to withhold the money, rather than ordering the state court plaintiff to relinquish it. But the Eighth Circuit rejected this as a distinction without a difference:

But state-court cases, related or not, are not before the district court. The state-court plaintiffs at issue neither agreed to be part of the federal MDL nor participated in the MDL Settlement Agreement. Even if the state plaintiffs’ attorneys participated in the MDL, the district court overseeing the MDL does not have authority over separate disputes between state-court plaintiffs and [the defendant].

Id. at 874. The Court acknowledged the possibility that the lawyers for the state court plaintiffs had benefitted from lead MDL counsel's work but noted that "equity is insufficient to overcome limitations on federal jurisdiction." Id.

The same logic applies to disputes where the claimant has not filed a lawsuit in any court. This is underscored by the Ninth Circuit's opinion in Vincent , along with a similar opinion filed one year earlier: Hartland v. Alaska Airlines , 544 F.2d 992 (9th Cir. 1976). Both these cases stemmed from MDLs created following plane crashes, with the families of the deceased suing the airline. In Hartland , an MDL court insisted on a holdback from the recovery of one person who had sued the airline in state court, and one person who had not sued the airline at all. The Ninth Circuit issued a writ of mandamus to correct this "usurpation of power" by the district court, ordering that the money be returned to both people. Hartland , 544 F.2d at 1001. In Vincent , an MDL court similarly required a holdback from the recovery of someone who had never filed suit, and that person appealed. The Ninth Circuit held that the appeal was controlled by Hartland , and that although an MDL court has the power to order holdbacks from the recoveries of plaintiffs before it, it lacks the power to order holdbacks from the recoveries of claimants not before it. Vincent , 557 F.2d at 765-66.

To be sure, it's not immediately apparent whether the people who won their appeals in Hartland and Vincent had hired lawyers who also happened to represent clients in the MDL. But there is no reason to think that the presence of so tenuous a connection between those nonparties and the MDL proceedings would have changed the Ninth Circuit's views about the "usurpation of power" that the district courts had committed. The fundamental problem with lead counsel's argument is that they are thinking too much in terms of the lawyers, and not enough in terms of the clients. The lawyers may think of the recoveries as "their" recoveries—recoveries that come from their "inventory." But that "inventory" consists of people, and each person has their own case against the defendant. Authority over a person's case (and a person's recovery) cannot be created simply because that person's lawyer happens to have another case before the district court.

Incidentally, Hartland and Vincent , as well as the Eighth Circuit's ruling in Genetically Modified Rice , vaguely conclude that the MDL courts lacked "jurisdiction." And it appears they may have been speaking in terms of "subject matter jurisdiction." Hartland , 544 F.2d at 1001 (stating that the MDL court "had not even a semblance of jurisdiction—original, ancillary or pendent—to order anything or anybody, and least of all to compel lawyers who were not parties to the action to pay $3,250 into a fund"); Vincent , 557 F.2d at 765-66 (relying on Hartland and cautioning that "[i]f parties to a suit cannot confer subject-matter jurisdiction on a federal court by agreement, certainly ‘nonparties,’ people who have never been made parties to a suit anywhere, cannot confer such jurisdiction ...."); In re Genetically Modified Rice Litigation , 764 F.3d at 874 (affirming the district court's holding that it lacked "jurisdiction" based, in part, on Hartland ). But it's not clear that this is precisely the right way to think about it. Imagine that a federal case is proceeding down the hall in Judge Breyer's courtroom. Imagine further that Chief Judge Seeborg does not like what he's seeing and decides to issue an order in the case. There is subject matter jurisdiction in the case—it belongs in federal district court. There is jurisdiction "in the sense of addressing the power of the courts to entertain cases concerned with a certain subject." Sosa v. Alvarez-Machain , 542 U.S. 692, 714, 124 S.Ct. 2739, 159 L.Ed.2d 718 (2004). It's just that Chief Judge Seeborg lacks any authority over the case (no matter how badly the parties might prefer otherwise). It is in this more general sense that jurisdiction is lacking.

Lead counsel makes several arguments in response to all this, but they are mostly policy arguments rather than arguments about the scope of a district court's power. For example, lead counsel notes that the common benefit work performed in the MDL paved the way for plaintiffs with state court cases (and for claimants who have not yet filed cases) to receive settlements from Monsanto. That's true, but as discussed, it does not matter for purposes of assessing the scope of an MDL court's authority to reach nonparties. In any event, while much of the pretrial work was done in the MDL, and while much of that work laid the groundwork for the three trials, ultimately it was the three jury verdicts that brought Monsanto to the table, and two of those were not even in federal court. It seems strange to assert that a federal district court should issue an order affecting the recoveries of nonparties based on common benefit work that was performed in state court. Especially when, as Judge Furman has noted, there's nothing to prevent the lawyers who did common benefit work in the state courts from seeking a redistribution of compensation from the judges presiding over the state court cases. In re General Motors , 477 F. Supp. 3d at 180.

Relatedly, lead counsel observes that if a lawyer has a client in the MDL, that lawyer will have access to "common benefit work" performed in the MDL—for example, the transcripts of expert depositions, or documents produced by the defendant. If a lawyer in the MDL also represents a client in state court, presumably that client will benefit from the lawyer's access to this MDL work product. It's not clear why that would present a policy concern in most instances—in our system, lawyers regularly pick up where other lawyers left off, benefitting from the good work of their predecessors. So if a lawyer with access to an expert deposition becomes better prepared to deal with the expert testimony in a separate case, what's the problem with that? Moreover, if the transcript of that deposition is on the docket, as much of the expert materials are in this MDL, then by definition it's available for use by anyone, and lead counsel has no right whatsoever to impose a tax on its use. In any event, to the extent lead counsel has a genuine interest in preventing nonparties from benefitting from confidential MDL work product, it can ask the MDL court to issue an order preventing MDL attorneys from using confidential MDL work product in other cases without the consent of lead counsel. Perhaps even an order requiring payment of a fee for its use outside the MDL context. But the possibility that an MDL attorney will use MDL work product to the benefit of a client outside the MDL is not a basis for a federal district court to assert jurisdiction over the recovery of a party not before it.

Lead counsel also notes that Monsanto's "inventory settlements" with law firms were brokered by Ken Feinberg, the special master appointed by this Court. Those law firms, lead counsel correctly observes, represented plaintiffs in the MDL along with state court plaintiffs and people who had not yet filed suit. But the appointment of a special master is not a thread that can connect all these cases for jurisdictional purposes. The Court appointed Feinberg simply to mediate cases in the MDL; it was Monsanto and the law firms who wisely decided to hire him to broker deals in the other cases. Indeed, it's not clear how a district court could have the authority to appoint someone to mediate disputes that reside outside its jurisdiction.

Next, lead counsel argues that MDL courts can use their inherent power to regulate the conduct of attorneys to require all attorneys in the MDL to contribute a portion of their contingency fees (from all their clients’ recoveries) to a common benefit fund. But the power to regulate attorney conduct typically involves sanctions and other disciplinary measures, and it can only be exercised to control courts’ "own proceedings." Chambers v. NASCO, Inc. , 501 U.S. 32, 111 S.Ct. 2123, 115 L.Ed.2d 27 (1991) ; United States v. Williams , 504 U.S. 36, 50, 112 S.Ct. 1735, 118 L.Ed.2d 352 (1992) ; United States v. Wunsch , 84 F.3d 1110, 1115 (9th Cir. 1996) ; Wharton v. Calderon , 127 F.3d 1201, 1207 (9th Cir. 1997) ("As broad as the scope of the inherent power of a federal court may be, it does not encompass the protective order at issue here. It cannot be justified as a sanction, as an attorney disciplinary measure, or as necessary to control the courtroom or proceedings pending before the court." (citations omitted)); see also 30 Moore's Federal Practice , § 807.01 (Matthew Bender 3d Ed.). It would stretch the power beyond recognition to say that it authorizes MDL courts to order appearing attorneys to contribute from their contingency fees from non-MDL recoveries to compensate lead counsel.

Finally, lead counsel argues that broad holdback orders have become a regular occurrence in MDLs, citing various pretrial orders. It's true that district courts often adopt orders requiring holdbacks from recoveries of plaintiffs outside the MDL—and in particular from the recoveries of plaintiffs who hired an attorney with an MDL case. In re Juul Labs Inc. , No. 19-md-02913 (N.D. Cal. May 27, 2020); In re Bair Hugger Forced Air Warming Devices , No. 15-md-2666 (D. Minn. May 24, 2016); In re Ethicon Inc. , No. 15-md-2652 (D. Kan. May 2, 2016); In re Cook Medical Inc. , No. 14-ml-2570 (S.D. Ind. Mar. 16, 2016); In re Testosterone Replacement Therapy , No. 14-cv-1748 (N.D. Ill. Nov. 25, 2014); In re Toyota Motor Corp. , No. 10-ml-2151 (S.D. Cal. June 9, 2013); In re Nuvaring , No. 08-md-1964 (E.D. Mo. Dec. 9, 2011); In re Yasmin & Yaz , No. 09-md-2100 (S.D. Ill. Mar. 25, 2010). Perhaps the prevalence of these orders suggests that this Court is wrong about the limits of a district court's authority. But perhaps instead it's a sign that we are too quick to assume that MDL courts can do whatever they want. Few of those holdback orders meaningfully consider the power of a district court to issue them. Instead, they seem similar to the common fund order this Court adopted uncritically in this case: orders based on language proposed by lead counsel, which engendered little or no opposition and were adopted at the outset of an MDL.

C. Claimants Outside The MDL Who Are Represented By A Lawyer Who Signed A Participation Agreement.

The next category consists of people whose lawyers signed an agreement with lead counsel to be bound by this Court's holdback order in exchange for access to MDL work product. This agreement is described in the holdback order as the "participation agreement."

As discussed in the preceding subsection, the Eighth Circuit and the Ninth Circuit may have considered a district court's authority to order holdbacks from the recoveries of nonparties to be an issue of subject matter jurisdiction. In re Genetically Modified Rice Litigation , 764 F.3d at 874 ; Vincent , 557 F.2d at 765-66 ; Hartland , 544 F.2d at 1001. If that's the right way to interpret those cases, district courts in those circuits lack the power to require a holdback from nonparties whose lawyers signed an agreement that they would be bound by a holdback order, because parties cannot agree to confer subject matter jurisdiction where none exists.

As explained in the preceding section and by Judge Furman in General Motors , it's probably not an issue of subject matter jurisdiction. But it's still questionable whether a district court has authority (that is, "jurisdiction" in the more generic sense) over the recovery of someone whose lawyer signs a participation agreement. On the one hand, the Third Circuit has held, in an unpublished opinion, that a district court might be able to enforce that agreement through ancillary jurisdiction. See In re Avandia Marketing, Sales Practices & Product Liability Litigation , 617 F. App'x 136, 141-44 (3d Cir. 2015). According to the Third Circuit, an MDL court's power to appoint lead counsel allows it to issue pretrial orders governing lead counsel's compensation, and because an MDL court can incorporate into its pretrial order an agreement permitting lead counsel to trade MDL work product for a contribution into a common benefit fund, it can then exercise ancillary jurisdiction to enforce the agreement and require a contribution. On the other hand, it seems strange to exercise power over a nonparty's recovery from a state court judgment or settlement when the same purpose could be achieved through action that is far less intrusive: authorizing lead counsel to directly charge a lawyer who wishes to use confidential MDL work product, and then allowing that lawyer to decide, with their client and without involvement of a federal court, who should shoulder that cost. After all, to the extent this is in furtherance of the district court's efforts to manage the MDL cases, the court must take pains not to extend its inherent managerial authority beyond its reach. See Degen , 517 U.S. at 829, 116 S.Ct. 1777 ("[Inherent power] is limited by the necessity giving rise to its exercise."); United States v. Moussaoui , 483 F.3d 220, 237 (4th Cir. 2007).

In any event, there is no need in this ruling to take a position on whether an MDL court has the authority to require a holdback from a nonparty's recovery as a result of a lawyer participation agreement. As discussed in Section III, even if the Court had the authority to do so here, it would decline to exercise that authority.

D. People With No Connection At All To The MDL

By now it should be obvious how badly lead counsel overreached when they asked the Court to tax every Roundup-related recovery by anyone in the country, regardless of any connection to the MDL. Vincent , 557 F.2d at 765-66 ; Hartland , 544 F.2d at 1001 ; see also In re Showa Denko , 953 F.2d at 166 ("The order thus compels contributions from plaintiffs in state or federal litigation who are not before the court and by claimants who have chosen not to litigate but to compromise their claims outside of the court ... Claimants who have not sued and plaintiffs in state and untransferred federal cases have not voluntarily entered the litigation before the district court nor have they been brought in by process. The district court simply has no power to extend the obligations of its order to them."). That lead counsel was even willing to make such a request is perhaps reflective of a more widespread problem in the MDL world—a pervasive assumption that MDL courts have the power (or should have the power) to solve every problem or imperfection imaginable relating to mass tort litigation. One example of this—the example most relevant to this ruling—relates to the free rider issue. What began as a healthy concern about the free rider problem seems to have evolved into an obsession. From reading lead counsel's papers, you begin to get the feeling that any lawyer who secures for their client a quick settlement from Monsanto has committed some great injustice that must be rectified at all costs. Poor lead counsel does all the hard work killing the prey, only to see the "TV lawyers" and other vultures swoop in and pick at the remains.

This issue needs to be put in perspective. Yes, district courts must be able to redistribute attorney compensation when the situation warrants it. And in an MDL, the situation will often warrant it—to a degree. But there is nothing inherently wrong with lawyers (and by extension their clients) benefitting from the work of other lawyers without paying them. That happens all the time in our legal system. Against this backdrop, the idea of a lawyer in the MDL getting better results for a state court client as a result of access to "MDL work product" seems unremarkable—hardly an event that district courts should reach to the outer limit of their powers (or beyond) to address. And yes, it might strike some as unseemly that many lawyers obtain clients as a result of mass TV advertisements and do little other work to settle their cases. But perhaps that ad was the only thing that caused those clients to seek redress for an actual injury. There is no need to redistribute attorney compensation in mass tort litigation based on value judgments, or based on the assumption that any "benefit" lawyers and clients receive through the efforts of hard-working attorneys must be taxed. Rather, the concern is simpler: that lawyers doing common benefit work be compensated adequately for that work. They should be compensated enough to make the work worth doing—to make sure that the difference between taking the lead and sitting on the sidelines is meaningful enough to prevent the good lawyers from running to the sidelines.

With this in mind, the next question is whether the Court should exercise what limited power it does have to redistribute attorney compensation in this MDL.

III. WHETHER THE COURT SHOULD EXERCISE ITS DISCRETION TO REQUIRE HOLDBACKS IN THIS MDL.

Given the discussion in Section II, the only plausible candidates for a holdback order are plaintiffs in the MDL who will recover from Monsanto, and possibly nonparties whose lawyers signed the participation agreement. But in the interest of thoroughness, this section will also address whether the Court would, if its power were broader, order a holdback for the other people who lead counsel propose to tax.

A. The Recoveries Of MDL Plaintiffs

One could imagine an argument that there should be no holdback at all for this MDL. The argument would go something like this: Yes, the lead lawyers invested a great deal of time and money in these cases, but now they're likely making so much from settling their own "inventories" that they can each afford to buy their own island. Yes, much of the "common benefit work" was done in connection with general causation, which directly impacted all plaintiffs in the MDL, but what really opened the floodgates for settlement were the three jury verdicts, two of which took place in state court, not the MDL. And yes, lead counsel's hard work helped lay the groundwork for other lawyers in the MDL to get settlements for their clients, but the settlements obtained by those lawyers were likely far lower than the settlements obtained by lead counsel for their "inventories," thus diminishing the need to address the free rider problem. No doubt that the distribution of attorney compensation in this MDL is imperfect, but perfection can almost never be achieved in the area of attorney compensation—that would be a game of whack-a-mole. And there's no indication that the distribution is so out of whack as to justify the effort and time involved in holding back money from people's recoveries and figuring out how to redistribute it.

Tempting as it may be to adopt that argument (particularly after watching lead counsel argue to expand an already overbroad holdback order), the Court is convinced that a holdback is justified for recoveries obtained by plaintiffs in the MDL. This is so for several reasons. Perhaps most importantly, it's worth reiterating that ordering a holdback does not necessarily mean that more money will be distributed to lead counsel. Rather, a portion of each MDL plaintiff's recovery will be placed into a fund, and then Special Master Feinberg will prepare a recommendation to the Court for how that money should be distributed. Perhaps some of the lawyers who performed valuable common benefit work are not part of the lead counsel group, and perhaps the compensation they stand to receive from their own clients’ recoveries is not so great. Or perhaps some lawyers from the lead counsel group will be able to make a case for additional compensation despite the money they already stand to receive. But perhaps all or most of the money will be sent back to the people who were subject to the holdback, in which case delay would be the only real consequence.

Relatedly, there is a reliance interest at play here. After all, the Court signed lead counsel's proposed order at the outset of the MDL, thereby creating an expectation that a common benefit fund would be created. Lawyers who did common benefit work (whether they were part of the lead counsel group or not) did so on the expectation that they would have the opportunity to make a case for extra compensation—beyond what they stood to receive from their own clients’ recoveries. Thus, even in the face of doubt about whether any redistribution of attorney compensation is really necessary in this MDL, it's worth going through the trouble and expense of analyzing the issue carefully. This is so even if the Court ultimately concludes that all MDL plaintiffs should simply get their money back from the fund.

Finally, the common benefit work performed in the MDL indeed conferred a significant benefit on MDL plaintiffs and their lawyers. The work on general causation, on summary judgment in the bellwethers, and on the Hardeman trial helped pave the way for substantial settlements for other MDL plaintiffs, and thus for substantial contingency fee recoveries by the lawyers representing them. Thus, it's worth double checking to make sure that the lawyers in the MDL who performed common benefit work are adequately compensated in relation to the lawyers who did not.

In terms of the percentage of an MDL plaintiff's recovery that will be held back, there is a certain arbitrariness to lead counsel's request for 8.25%. The request assumes (contrary to the discussion in this ruling) that even if the attorneys who did common benefit work have already been compensated obscenely compared to the free riders—they should automatically be entitled to more because those who benefitted from their work should be taxed no matter what. That is neither necessary nor desirable. Their request also assumes that they should be compensated for work performed in state court, which, as previously discussed, is not appropriate. On the other hand, when lead counsel proposed 8.25% (and when they originally proposed 7%) they assumed that the order would extend far beyond plaintiffs in the MDL, which is not happening. It appears, based on the Court's general understanding of settlement amounts discussed in the filings in this case and in public reports, that an 8% holdback is likely more than adequate to make any necessary redistribution of attorney compensation. And it would not be surprising to see all plaintiffs receive all or part of that money back.

To be clear, the required holdback is 8% of a plaintiff's gross recovery, but it is ordered to come out of the portion of the recovery designated for attorney compensation. It would be a violation of this order for any attorney of a plaintiff in this MDL to reduce a plaintiff's net recovery on the basis of this ruling. When attorneys make claims for compensation from the fund, it must be based only on work performed in the MDL; state court work cannot be compensated. As previously mentioned, to the extent lead MDL counsel did common benefit work in state court, they can seek redistribution of attorney compensation there. To adequately assess whether attorney compensation should be redistributed, the Court will likely need to be informed by the Special Master of the amounts attorneys have already recovered—those who performed common benefit work and those who did not. And all counsel in the MDL should be aware that, even if they performed no common benefit work, they may have a strong claim to ultimately recover the amount withheld from their clients’ recovery, based on an argument that lead counsel's relative compensation is already so high that no further redistribution is warranted.

As previously noted, lead counsel proposed that an additional .25% of each plaintiff's gross recovery be taken from the plaintiff's share, for the purpose of redistributing litigation costs. However, lead counsel has made virtually no effort to show that this is necessary. In passing, lead counsel has stated that their clients’ litigation expenses amounted to $20 million. But that appears to be for all their clients, not just the ones involved this MDL. And it is not obvious why lead counsel's MDL clients, who are already no doubt being compensated more handsomely than the typical MDL plaintiff, need to be reimbursed for those litigation costs. Accordingly, this request is denied without prejudice to making a stronger showing, within 21 days of this ruling, that a redistribution of litigation expenses is necessary. It is ordered that no money shall change hands in any settlement in this MDL until the deadline has passed to file a renewed motion for a holdback of costs or until any such motion is resolved, whichever is later.

B. The Recoveries Of Non-Parties Whose Attorneys Signed A Participation Agreement

Assuming for argument's sake that the Court has the power to order a holdback from the recoveries of nonparties whose attorneys signed the participation agreement, it would not do so here, for two reasons.

First, the participation agreement contemplates that the nonparty clients will be subject to this Court's holdback order in exchange for giving their attorneys access to MDL work product. As noted earlier, much of the MDL work product—for example, expert reports and expert deposition transcripts—was rightly placed on the docket and thus was available to any member of the public for free. It would not be appropriate to take money from anyone's recovery based on access to that information. Moreover, to the extent that attorneys for nonparties received access to MDL work product that was actually confidential, it's doubtful that access to that work product is what truly advanced the ball for those attorneys and their clients. Lead counsel themselves imply that there was little difference, from the standpoint of reaching meaningful settlements, between firms that did and did not receive confidential MDL work product. They admit that "only a few non-leadership firms have requested access to the document depository." Plaintiffs’ Co-Lead Counsel's Reply at 14, In re Roundup , No. 16-md-2741 (N.D. Cal. Feb. 18, 2021), ECF No. 12615. Nevertheless, "most law firms were able to resolve their Roundup cases without hiring any experts, without accessing or reviewing discovery documents, without taking any depositions, and really without doing any pre-settlement work at all." Plaintiffs’ Co-Lead Counsel's Motion at 9, In re Roundup , No. 16-md-2741 (N.D. Cal. Jan. 11, 2021), ECF No. 12394. What truly advanced the ball for attorneys who settled their cases without doing much work—as lead counsel repeatedly trumpet in their papers—was the three jury verdicts and the work that led up to them. There is little reason to believe, at least on this record, that access to confidential work product increased anyone's settlement value.

Second, as lead counsel candidly explained at the hearing on this motion, there was no real correlation between the lawyers who signed the participation agreement and the lawyers who received access to the work product from the MDL. See Transcript of Oral Argument at 9, In re Roundup , No. 16-md-2741 (N.D. Cal. Mar. 8, 2021), ECF No. 12737. Things were moving so fast in the MDL, and lead counsel was working so hard, that they understandably did not take the time to ensure sure that attorneys who requested access to MDL work product signed the agreement.

Under these circumstances, it would not be reasonable to require a holdback from the recoveries of non-parties whose attorneys happened to sign the participation agreement, particularly where the authority to impose this requirement is questionable in the first place.

C. The Recoveries of All Other People

Even if the Court had the power to order holdbacks from the recoveries of all other people who would be affected by lead counsel's motion, it would not do so. This is so for many of the reasons already discussed in this ruling—reasons that are as relevant to the Court's exercise of its discretion as they are to an assessment of the Court's power. For example, it's likely that the lawyers who did common benefit work in the MDL have already been handsomely compensated, such that if a redistribution is warranted at all, it can likely be accomplished with a fund created from the recoveries of MDL plaintiffs. From the standpoint of exercising the Court's inherent managerial power, there is a much weaker nexus between nonparties and the MDL, and courts should exercise caution before extending that power broadly (even when they have it). And much of the overall common benefit work was performed outside the MDL—in particular, the two state court jury verdicts.

But there are further reasons not to extend the holdback order beyond recoveries of plaintiffs in the MDL. This is especially true as it relates to state court plaintiffs. Recall that lead counsel has asked the Court to impose a holdback on the recoveries of state court plaintiffs—not just from settlements, but from judgments. One wonders if this would be constitutional, but in any event it raises serious comity concerns, as explained by Judge Furman in General Motors . 477 F. Supp. 3d at 180. Moreover, it will be challenging enough to police the holdback requirements and adjudicate the potential redistribution of attorney compensation for plaintiffs in this MDL. Including cases that are not before this Court would make the task that much taller.

IV. CONCLUSION

For all these reasons, lead counsel's motion to establish a holdback percentage is granted in part and denied in part. Monsanto must hold back 8% of the gross amount of any recovery it pays an MDL plaintiff and place that money into the common benefit fund. The attorneys for the MDL plaintiffs are ordered that this 8% must come from the attorney fee portion of the recovery, not their clients’ portion. The request for a holdback from the recoveries of people who are not plaintiffs in the MDL is denied. The Court lacks the power to order holdbacks from the recoveries of most or all of those people. And even if the Court had the power, it would decline to exercise it in light of the specific facts relating to this MDL. The request for a .25% holdback to potentially redistribute the burden of litigation costs is denied without prejudice to filing a renewed motion within 21 days of this ruling.

IT IS SO ORDERED.