Alex Reinert's article in a forthcoming Indiana Law Journal finds that, from 1990-99, "thin complaints" had just as much chance of success as "bulkier" complaints. Thus, Reinert suggests, Twombly and Iqbal may be producing false negatives, and there may not be the benefits to heightened pleading we imagine: "the data here suggest that the costs imposed by heightened pleading may be substantial and may not create the assumed benefits."
Reinert thanks fifteen people and two workshops in his opening footnote, and it is frustrating to think that perhaps not one of them pointed out to him that this is a complete non sequitur. Leaving aside problems with the dataset (which Reinert acknowledges), there are two issues that completely obliterate any ability to infer causality.
First, Reinert measures "success" to include settlements. But, as Twombly itself points out, once a case gets past the pleading stage, the expense of litigation may force a defendant to settle a meritless suit. A "thinly pleaded" suit that resulted in a settlement pre-Twombly, but in a dismissal post-Twombly, may be a meritorious suit, but it's much more likely to have been a nuisance settlement. Reinert pooh-poohs this possibility by arguing that someone inclined to make a nuisance settlement isn't likely to move to dismiss and defend an appeal, but that's a poor understanding of how litigation works: discovery is far more expensive (both in terms of legal expense and non-pecuniary distraction to the defendant) than dismissal and appeal. The fact that these cases would have been dismissed under Twombly is likely a feature, rather than a bug. Reinert has no access to the amounts these cases settled for (more or less than litigation expenses?) and has made no qualitative assessment of the merits of the cases.
Second, Reinert necessarily falsely assumes that a pleading that was thin pre-Twombly would be thin post-Twombly. That static assumption ignores how attorneys respond to incentives. Pre-Twombly the thinness of a pleading was a matter of a stylistic choice. An attorney could choose to draft a thin complaint to avoid accidentally making a concession that would adversely affect the case, and count on it surviving a motion to dismiss. As Judge Posner has written, "A plaintiff who files a long and detailed complaint may plead himself out of court by including factual allegations which if true show that his legal rights were not invaded." The fact that a plaintiff filed a thin complaint may just mean that the lawyer was being prudent, not that the complaint could not have been more detailed should courts require more. Again, this would require a qualitative assessment of the cases that Reinert does not perform.
The paper suffers from a fundamental problem of empirical studies of the legal world: the existing quantitative data, no matter how arduously compiled, does not provide the qualitative data necessary to draw public-policy conclusions, because much of what happens in the legal system is never recorded in court.