I don’t think plaintiffs should have to prove their case when they file.  But neither should they be empowered to drag out a meritless case.  To take Twombly as an example:  I think that the case should have survived a 12(b)(6) motion.  But I agree with Ms. Mitchell and Mr. Wallach that the district should have been empowered to limit discovery solely to the question of conspiracy.  Further, I think it appropriate for the district court to limit initial discovery so as to force plaintiff, especially in a claim the seems unlikely to win, to find evidence of conspiracy in a timely manner.

This proposal speaks to my general themes: lets not fix discovery by pleading reforms and lets have contextualized fixes aimed at where real harms exist as opposed to blanket pleading changes that may well do more more than good when applied to a whole host of cases.

 Additionally, discovery costs are not the only costs incurred in litigation.  Once a complaint proceeds beyond the pleading stage, the next exit for defendant is summary judgment.  That alone can be a very expensive production, requiring the amassing of declaration and evidence, as well as a carefully crafted brief.  Pre-trial preparations are even more costly for a defendant who foregoes summary judgment.  All of these matters and the uncertainties and costs they pose weigh on a defendant to settle a claim, meritorious or not, if it is not disposed of at the outset.  Oftentimes too, litigation can require pulling key decision makers and workers away from doing their daily duties to prepare for an participate in deposition and other court-related obligation.  Or in smaller claims, it can mean that the defendant has to miss work.  This too can take a toll on defendants.  All of these burdens and costs suggest that courts should be wary of liberalizing the pleading standard too readily.

 Although Professor Mulligan suggests that the discovery issue is raised in only a small number of cases, we believe the number of cases where a plaintiff with legitimate claims does not have access to sufficient facts to plead those claims is likely even smaller, if they exist at all.  Rather than letting all claims proceed through full-fledged discovery, it would be better to create an exception to Iqbal that allows a court to order limited discovery and the right to amend, under narrow circumstances.  If a plaintiff could show good cause why the key facts going to a particular element of a claim were under the sole control of defendant and inaccessible to plaintiff, then a limited right of discovery before amendment could be recognized.  Otherwise, the current Iqbal standard should control.  The benefit of such an approach is that discovery could be tailored to fleshing out whether plaintiff has a claim, rather than continuing the Conley practice of opening the floodgates to discovery on all subjects on the flimsiest of allegations.  

Drafting a narrow exception to Iqbal and Rule 8’s requirement that the plaintiff come forward with facts showing that the plaintiff is entitled to relief might well be an appropriate topic for amendment of the Federal Rules of Civil Procedure.  But no such amendment is necessary for the Supreme Court to interpret what the existing Rule 8 means.

I do not advocate meritless suits.  Rather I am noting that many claims require plaintiffs to prove information to which they cannot gain access until they get to discovery.  Hence making them plead such facts is a catch-22.  This is not saying such facts do not exist, rather it is saying they are not in possession of the plaintiff at the time of filing the answer.  If we put pressure to prove all facts before filing a complaint, the effects could be unappealing.  To name just one for example, this scheme would push plaintiffs to hire teams of private investigators, outside the supervision of the court, to gather information in order to justify allegation.  

I am signing off.  It’s been a pleasure. My thanks to Ms. Mitchell, Mr. Wallach, and the folks at PublicSquare.net.

I agree that discovery costs hit defendants in this context more than plaintiffs.  But my point was only that these costs are not a one-way street and that if discovery is to cost $7 million dollars, even sharing 1/7 of that cost is significant, especially given the payment structure for plaintiffs’ attorneys.  This remains a financial disincentive for plaintiffs to engage in discovery in a meritless case.

Ms. Mitchell and Mr. Wallach argue that Conley in essence read “showing that the pleader is entitled to relief” out of Rule 8(a)(2).  I think that position in error.  Any number of 12(b)(6) motions can be, and are with great frequency, under the Conley interpretation of Rule 8, arguing that the complaint fails as a matter of law.  For example, defendant can move under 12(b)(6) because the statute sued upon does not create a cause of action for plaintiff, Pegram v. Herdrich, 530 U.S. 211, 217 (2000) (affirming 12(b)(6) dismissal of ERISA claim because a breach of fiduciary duty cause of action did not arise against defendant under the statute) , or based upon the facts as stated in the complaint defendant necessarily prevails upon an affirmative defense, Jones v. Bock  549 U.S. 199, 215 (2007) (“A complaint is subject to dismissal for failure to state a claim if the allegations, taken as true, show the plaintiff is not entitled to relief. If the allegations, for example, show that relief is barred by the applicable statute of limitations, the complaint is subject to dismissal for failure to state a claim.”). Unfortunately, the Administrative Office of the United States Court does not provide a more detailed breakdown of case terminations (see here) so we cannot know how many claims were dismissed on 12(b)(6) pre-Trombly, but the numbers of dismissals before the pre-trial stage exceeds 50%.   Conley, then, did not commit the interpretation sin of rendering a portion of Rule 8(a)(2) a nullity, as Ms. Mitchell and Mr. Wallach suggest.  See Mountain States Tel. & Tel. Co. v. Pueblo of Santa Ana, 472 U.S. 237, 249 (1985) (“the elementary canon of construction that a statute should be interpreted so as not to render one part inoperative”).  Rather Conley (properly in my view) limited the “entitled to relief” language to legal challenges against the complaint that do not address the veracity of the allegations.  In fact, if one adopts Ms. Mitchell and Mr. Wallach’s interpretation of Rule 8(a), then Rule 9(b) is rendered a nullity—as there is nothing more to add in such claims that is not already required under their interpretation of Rule 8(a).  See Ricci v. DeStefano ,129 S.Ct. 2658, 2674 (2009) (“We must interpret the statute to give effect to both provisions where possible.”).  I think their reading of Rule 8(a)(2), not only contrary to 50 years of practice, but a strained interpretation of the Rules read as a whole.

Requiring plaintiffs to show a reasonable basis for believing that they may have a legitimate claim does not create a Catch-22.  It is simply not true that plaintiffs can discover whether such a basis exists only through discovery.  Taking Professor Mulligan’s example of a Title VII discrimination suit, if employees are discharged but have no grounds to plausibly believe that the discharge was driven by discriminatory motive, then they should not file lawsuits.  The mere fact that a person is fired should not be sufficient, without more, to set a lawsuit in motion.  On the other hand, if employees have a reasonable basis to believe that their discharge was the result of discrimination, they should allege the specific facts which lead them to that conclusion and they should have no trouble satisfying Rule 8.

Professor Mulligan is undoubtedly correct that plaintiffs incur costs in conducting discovery.  But that does not mean that the burden on plaintiffs and defendants is anywhere near symmetrical.  Plaintiffs will usually have fewer witnesses subject to deposition, especially where an individual sues a business.  While plaintiffs still must pay for attorney time in taking the deposition of a defense witness, defendants must pay both for the time of the attorney and the witness’s time.  Further, plaintiffs incur virtually no costs in demanding that defendants sift through thousands—and oftentimes millions—of documents to respond to requests for production.  It is simply beyond question that baseless lawsuits are often settled to avoid the costs of discovery.  

Professor Mulligan suggests that the a stratified system may be worth considering under which different types of cases are subjected to different pleading standards.  However, as Professor Mulligan acknowledges, such a system can be accomplished only by an amendment to the rules.  Therefore the issue goes beyond whether Iqbal is a proper interpretation of the existing rules.  Further, such a system would be accompanied by its own set of complications, inviting litigation over precisely what type of case a given complaint presents and whether it must clear the low or high hurdle.

Ms. Mitchell and Mr. Wallach also question the validity of the sample and study upon which I rely.  All the more reason for this discussion to have happened in the rules committee.  They could have looked to numerous studies, had public comment, commissioned a new study etc. 

If the imposition of fact pleading as to the 95% (or whatever the number is) of cases where discovery is closer to 3% of the amount at stake came at no cost, I would not complain.  The difficulty is the catch-22 I described earlier.  There are certain categories of cases, like intentional discrimination cases, where plaintiff’s claim requires allegations about facts that the plaintiff cannot obtain without discovery.  In the federal system until Twombly/Iqbal, plaintiff relied upon inference to survive a Rule 12(b)(6) motion in this regard.  This seems much less likely now.  As such, the Iqbal regime comes at the cost of dismissing more potentially meritorious cases pre-discovery. 

Hence, my view that a middle course is more appropriate.  Only in those cases, like anti-trust, where we know discovery is astronomically expensive, should we require fact-pleading.  We can do that by amending Rule 9.  But as to the vast bulk of cases, this cost-benefit analysis (as I understand the empirical evidence) does not weigh in favor of more rigorous pleading, because we will dismiss more potentially meritorious cases early even though we do not face a huge discovery cost problem here.

As I noted earlier, 15 states continue with a type of fact-pleading regime.  Nevertheless, these systems work reasonably well, even with the type of cases I describe above, in large part do the practice of pleading upon “information and belief.”  In essence this is a means of satisfying a fact-pleading requirement without actually attesting to the veracity of the fact. 

The Court has yet to adequately address this issue.  Allowing a liberal information-and-belief practice would essentially take the sting out of Iqbal and Twombly.  As such, I doubt this practice will settle in at the federal level.  Leaving us with the catch-22 I describe.

As to stigma harms for law suits, I agree that a falsely brought civil suit brings some stigmatic harm for defendants.  But losing the ability to bring, say, a meritorious Title VII claim because our pleading standard asks plaintiff to allege facts s/he cannot obtain brings costs too.    We should rationally balance those harms against the best evidence we have in a forum institutionally capable of engaging in the inquiry (i.e., the rules committee).

Now, our system limits jury power in many meaningful ways.  The jury must base its decision on admissible evidence, that evidence itself must be deemed sufficient (assuming it is true) to support the verdict, jury instructions play a big role, etc.  But absent truly outrageous claims, our system ultimately leaves the determination of whether a claim is true to the jury.  That is our tradition (and constitutional command in non-equity cases).

The set of truly outrageous claims that form the exception to this rule leaving the truthfulness of assertions to the finder of fact is limited.  Justice Souter, again the author of Twombly but dissenting in Iqbal, makes this point: “The sole exception to this rule lies with allegations that are sufficiently fantastic to defy reality as we know it: claims about little green men, or the plaintiff’s recent trip to Pluto, or experiences in time travel.”  Iqbal, 129 S.Ct. at 1959 (Souter, J., dissenting). 

Iqbal itself is not a little-green-men allegation.  Plaintiff alleges that John Ashcroft was motivated by racial and religious prejudice in enacting an enforcement regime.  Now this might be false.  (I hope so.)  But it is not an allegation of time travel.  Thus, traditionally speaking, it would be for the jury to ultimately decide whether the allegation is true.  Iqbal, however, empowers a trial judge to determine whether allegations are plausibly true based upon his or her common sense and experience.  Now, I am all for common sense and experience as the ultimate arbiter of liability — but traditionally it has been the jury who exercises that function.  I think that the proper balance.

We can already see this traditional balance shifting.   In a recent district court case in the Middle District of Georgia, Rhodes v. MacDonald, the court employed Iqbal’s common sense standard.  The court held, as a matter of common sense, that plaintiff’s allegation that President Obama was born in Nigeria is untrue.  While this is common sense to me too, I am not the jury.  Indeed, a jury might well agree with plaintiff, given that, according to some polls, 58% of Republicans thinks the President not a native born citizen. 

The point, I believe, is that while our system maintains many tools to ferret out meritless factual assertions, mere common sense of the trial judge has never been one of them.  This is a big change.  Take, again, my hypothetical Title VII claim in which the plaintiff alleges intentional racial discrimination.  Could a judge under Iqbal, dismiss the case because, based upon his/her common sense, employers just don’t discrimination anymore?  Is this any more implausible than that John Ashcroft was motivated by prejudice?  I think not.

Accordingly the court ruled in Twombly that “threadbare recitals of a cause of action’s elements, supported by mere conclusory statements” will no longer suffice. 550 U.S. at 555. In Iqbal, the Court made clear that Twombly’s ruling has general application. 129 S.Ct. at 1949. If Professor Mulligan intends to suggest that Iqbal signals a return to the pre-Rule 8 pleading requirements that mired parties in protracted debates about the technicalities of the complaint, that is not the case. Iqbal recognized that Rule 8 is a “notable and generous departure from the hyper-technical, code pleading regime of a prior era” but at the same time cautioned that the departure was not without bounds and that Rule 8’s requirements do not “unlock the doors of discovery for a plaintiff armed with nothing more than conclusions.” Id.

Which brings us to the subject of discovery. Professor Mulligan suggests it is not as burdensome as critics suggest and that it is but a fraction of the amounts at stake. Statistics on this issue can be misleading because issues such as how cases are selected for inclusion in a study and whether they were settled to avoid discovery costs will cause wide variations in the results. In antitrust cases, the genesis of Twombly and an area in which I frequently practice, discovery costs are staggering. Once into discovery, plaintiffs can seek every communication relating to pricing from an entire industry for a multi-year time period. That means that every company in the industry needs to advise all of its employees who have anything to do with pricing to preserve all of their documents, expensive e-discovery searches must be undertaken and oftentimes an individual’s files will have to be gone through page-by-page to find responsive documents. All to respond to just a single request. In such cases, discovery can keep substantial teams of lawyers busy for years, literally. This is extraordinarily costly.

Nor are antitrust defendants alone. Just last year the Institute for the Advancement of the American Legal System issued an interim report that was a joint project of the American College of Trial Lawyers Task Force on Discovery and the Institute for the Advancement of the American Legal System. Among its conclusions, “[d]iscovery costs far too much and has become an end in itself.” “Meritless cases, especially smaller cases, are being settled rather than being tried because it costs too much to litigate them.” In response to the Institute’s survey, “[n]early half of respondents said that notice pleading has become a problem because extensive discovery is required to narrow the claims and defenses and 57 percent said that with notice pleading, motions to dismiss on the pleadings are not effective in limiting claims and narrowing litigation issues.”

Beyond discovery, what I think often gets lost in the analysis is that real people’s lives are affected when they are accused of wrongdoing. A lawsuit gives a veneer of credibility to allegations of wrongdoing and often individuals, or companies, have to live under a cloud for years before the case resolves. When the allegations of wrongdoing are nothing more than conclusory statements asserted with the hope they will open the door to a fishing expedition during discovery, it is right to require more before allowing them to proceed.

Iqbal is a welcome response to these issues. It requires plaintiffs to fulfill their obligation to show that they are entitled to relief, not to just list the elements of a claim as allegations in a complaint. This demands only that a plaintiff show that they have some reasonable basis for believing that they have a legitimate claim. This requirement will focus litigation on the claims with merit and reduce the extraordinary burden of litigation expenses in unmeritorious cases.

In essence, Twombly  re-imposed the fact-intensive pleading system that was in place pre-1937.  This system is not without merit.  In 15 states (including my home state of Michigan) apply this scheme in their state courts.  Moreover, while the Federal Rules generally adopt a notice-pleading regime, the Federal Rules do impose a fact-pleading regime by exception to certain classes of cases under Rule 9.  This fact-intension pleading scheme has the benefit (as the Twombly Court noted) of restricting the ability of meritless cases from proceeding to the at-times costly discovery stage. 

While in most cases, the costs of discovery do not appear rampantly out of control, there are certain categories of claims, anti-trust being one (Thomas E. Willging et al., An Empirical Study of Discovery and Disclosure Practice Under the 1993 Federal Rule Amendments, 39 B.C. L. Rev. 525, 532 (1998)), where discovery costs do tend toward the extreme.    Thus, there are a class of cases in which, from a policy perspective, good reasons exist to believe that access to discovery should be limited or curtailed.

The Court,  however, by interpreting Rule 8(a)(2) under a plausibility standard, made a rule that is applicable to every case.  See Iqbal, 129 S.Ct. at 1957.  It would have been better to focus the cure of limited access to discovery to those areas that required it.   That is, if anti-trust suits (and a few others) are as, a matter of empirical fact, significantly more costly in discovery than the bulk of suits, we should treat them differently.  Perhaps they should be added under the fact-pleading regime of Rule 9. 

The institutional posture of the rules committee, not the Court disposing of a case, is the place to make these types of arguments as they reach across swaths of cases instead of just discussing the facts of one particular case.  Moreover, the rules committee has the power to make any number of nuanced changes to address the cost of discover in such cases.  For example, it could amend Rule 26 (the principle discovery rule) to narrow initial discovery in such cases and impose an early summary judgment motion under Rule 56.  Or it could beef up Rule 11 (the ethics rule) to address meritless pleading in anti-trust suits.  The Court disposing of a particular case, however, by this institutional posture is forced into making a one-size-fits-all rule.

Importantly, this one-size-fits-all rule comes at a cost.  Again in the vast bulk of cases the cost of discovery is not radically out of control.  Nevertheless, the Court has applied this plausibility requirement as to inferences upon all plaintiffs.  This requires them to plead facts to survive a pre-discovery dismissal that, often times, they simply cannot obtain without discovery.  For example, many federal anti-discrimination claims require  allegations about the defendant employer’s motivations.  Until one gets the boss into a deposition, allegations about motivation will necessarily be done by inference.  But now, skeptical judges are empowered to dismiss, say, gender or racial discrimination cases at the earliest stage because the judge does not find the plaintiff’s inference plausible. 

This puts plaintiffs in a catch 22.  They must plead discriminatory motivation with facts under Twombly.  But they cannot get those facts until they get to discovery.  But they cannot get to discovery until they plead discriminatory motivation with facts.  And on and on.  Now in those 5% of cases where discovery really is exceptionally expensive, such a scheme—all things considered—might be a sound policy choice.  But in a typical Title VII cases, say, where discovery is not so costly, I am hard pressed to believe this a sound policy.

The real issue in both these cases is not, actually, an abstract fight about the detail needed in a complaint (although that is how the Court frames the issues).  Rather both of these cases are about discovery (i.e., the process during which both sides in a lawsuit gather evidence in preparation for trial).  More precisely, these cases are concerned with the costs of discovery.  The assumed truth (I question this is a bit) is that discovery always costs large sums of money and that this disproportionately affects defendants. 

The Court, in a moment of clarity, noted this as well:  the Court fears that “the threat of discovery expense will push cost-conscious defendants to settle even anemic cases before reaching” summary judgment.  Bell Atl. Corp. v. Twombly, 127 S. Ct. 1955, 1967 (2007).  The Court also noted that allowing a weak case proceed to discovery “represent[s] an in terrorem increment to settlement value.”  The operative theory, then, is that discovery costs too much and that this cost hits defendants the hardest.  This huge cost, the argument goes, pushes defendants to settle cases that are unmeritorious.  This is unjust.   Therefore, we should prevent weak cases from proceeding to discovery by tightening pleading rules.  The key planks here are that discovery is always too expensive and that this cost disproportionately affects defendants.  I dispute these in turn.

As two, non-partisan Senior Researchers at the Federal Judicial Center observed “the debate over discovery reform has proceeded largely, but not entirely, with reference to salient personal experiences and not with benefit of empirical evidence.”  See Glenn. S. Koppel, Reflection on the Chimera of a Uniform Code of State Civil Procedure, 58 DePaul L. Rev. 971, 1001-02 (2009).  Anecdotes and political rhetoric have filled the knowledge vacuum created by the dearth empirical information.   Id.  In fact, recently the Advisory Committee recommended the scope-narrowing amendment to the discovery rules, despite the lack of an empirical case for the amendment, “largely because of the political preferences of the leadership of the American College of Trial Lawyers and the ABA Litigation Section.”  Id.

The question is: do we have evidence to support the notion that discovery is usually too costly.  The answer is we do not.  Rather the evidence supports the opposite conclusion.  In the overwhelming majority of cases, discovery comes to about 3% of the amount at stake.  Thomas E. Willging et al., An Empirical Study of Discovery and Disclosure Practice Under the 1993 Federal Rule Amendments, 39 B.C. L. Rev. 525, 531 (1998).  This is not a picture of discovery run amok or of a scary weapon to be wielded against hapless defendants.  To be sure, in 5% of cases filed discovery may run as much as 33% of the amount at stake.  Id.  But we know, by and large, which cases these are and can account for them by a separate rule.  (I will discuss this in a future post.)  There is no need to have a one-size-fits all approach, which is what Iqbal creates.  Indeed, the Federal Rules already allow for exceptions to the notice-pleading approach by way of Rule 9, where cases such as fraud are required to conform to the old fact-pleading scheme. 

I end by noting that the costs of discovery are not a one-way street.  In those few cases where discovery is very expensive, these cost must be borne by plaintiffs as well.  Moreover, given the typically contingency payment scheme for plaintiffs’ attorneys, this cost must be financed with the mere hope that a judgment or settlement is forthcoming.  Defendants, and defense attorneys, pay monthly, typically out of pocket.  Thus, the costs of discovery are certainly daunting and scary for plaintiffs too.  Proceeding to expensive discovery on a dog of a case will likely be a financial disaster for plaintiffs’ attorneys.  Thus, the in terrorem threat of discovery is not one-sided.  There is already a financial disincentive for plaintiffs to proceed to discovery on meritless cases.