Monday, November 28, 2011

The Difficulty of Quantifying the Effect of TwIqbal

When we took stock of things to be thankful for last week, we focused on family, friends, and health. Amidst the turkey, sweet potatoes, and cranberries (the real ones, not the canned stuff), there was little room for professional considerations (or, as our Texan friends call it, bidness). You see, despite our nerdiness, we actually do possess a sense of perspective. If we did make a list of legal-reasons-to-be-thankful, we'd include, along with the Washington Legal Foundation and Judge Posner, our old buddies Twombly and Iqbal.

But have those cases really made a difference? A couple of weeks ago we linked to an article by Professor Hubbard that concluded, after the requisite quantitative analysis, that Twombly might not have had much concrete effect. Color us surprised and somewhat disillusioned. We thought we were getting served up with stuffing and gravy, and instead someone passed us the turnips.

We got a peak at a forthcoming (volume 121) Yale Law Journal article, "Locking the Doors to Discovery? Conceptual Challenges in and Empirical Results for Assessing the Effects of Twombly and Iqbal on Access to Discovery." It's a fine article, well worth a read. We don't want to steal Mr. Gelbach's thunder, so we'll merely summarize a couple of the key ideas.

As a preliminary matter, pardon us for an aw-shucks moment as we note that the article refers to the debate in Pennumbra between some of our bloggers and Penn Professor Stephen Burbank over the merits of TwIqbal. But the focus of the new article isn't whether TwIqbal is good or bad, it's whether those cases have changed anything. Those cases have certainly caught the attention of the judiciary; as of July 2011, Twombly has been cited by courts over 45,000 times, and Iqbal over 25,000 times. The question is whether all of that is only lip service.

Gelbach canvasses the existing literature, which goes both ways. Gelbach's fundamental insight is that party reactions to the new doctrine - selection effects -- must be considered. Simply comparing before and after rates of dismissals can be misleading. If plaintiffs file fewer or different cases, or if defendants file more motions to dismiss, or if parties alter settlement strategies, the dataset against which motions to dismiss are measured is altered.

The dataset Gelbach mines is from the Federal Judicial Center, which seems to be the gold standard and which underlies most of the analytical literature. The Yale article then sprinkles a bunch of equations across the pages, producing an effect on us not unlike tryptophan. To his credit, the author bemoans "tedious algebra." But the result is far from tedious. TwIqbal appears to have had a profound influence. True, the percentage of dismissals has hardly budged, but it looks like defendants are filing more motions to dismiss. The bottom line, according to Gelbach, is that "at least 18 percent of all cases facing MTDs in the post-Iqbal period failed to reach discovery because of the switch to heightened pleading." Further, contrary to concerns expressed by some, TwIqbal seems to have had less, not more, effect on employment and civil rights cases. That is because there is less of a defendant selection effect for those cases, as defendants already usually filed motions to dismiss against them.

The article alludes to the issue of whether TwIqbal has increased social costs (more motions) or reduced them (sifting out meritless cases). Guess which way we vote? There is also a recommendation for further research: "If Twombly and Iqbal have culled mostly low-merit cases, then there should have been a noticeable drop in the number of cases where plaintiffs lose at defense summary judgment." We're not so sure about that. Sad to say, merit or lack thereof does not always animate the decisions of some judges. Specious notions of docket management, or deep-rooted hostility to dispositive motions, where judges think the whole game is about compelling settlement whether the cases have merit or not, taint the dataset. Meritless cases settle. Is that good or is that a pity?

In any event, Gelbach's article is undeniably interesting, and it restores our affection for TwIqbal. We might even put them back on our Holiday gift list.

.