CELS VII: Data is Revealing Part 2


Shouldn't it be "data are revealing?"

Shouldn’t it be “data are revealing?”

[This is part 2 of my recap of the Penn edition of CELS, promised here. For Part 1, click here.  For previous installments in the CELS recap series, see CELS IIIIVV, and VIVII.]

Where were we?  I know: throwing stink-bombs at a civil procedure panel!

At the crack of dawn saturday I stumbled into the Contracts II panel. Up first was Ian Ayres, presenting Remedies for the No Read Problem in Consumer Contracting, co-authored with Alan Schwartz.  Florencia Marotta-Wurgler provided comments.  The gist of Ayres’ paper is that consumers are optimistic about only a few hidden terms in standard-form contracts. For most terms, they guess the content right. Ayres argued that should be concerned only when consumers believe that terms are better than they actually are.  The paper  proposes that firms make such terms more salient with a disclosure box, after requiring firms to learn about consumer’s knowledge on a regular basis. Basically: Schumer’s box, psychologically-calibrated, for everyone.  Florencia M-W commented that since standard-form contracts evolve rapidly, such a calibrated disclosure duty might be much more administratively complex than Ayres/Schwartz would’ve thought.  A commentator in the crowd pointed out that since the proposal relies on individuals’ perceptions of what terms are standard, in effect it creates a one-way ratchet. The more people learn about terms through the Ayres/Schwartz box, the weaker the need for disclosure. I liked this point, though it appears to assume that contract terms react fairly predictably to market forces. Is that true?  Here are some reasons to doubt it.

Zev Eigen then presented An Experimental Test of the Effectiveness of Terms & Conditions.  Ridiculously fun experiment — the subjects were recruited to do a presidential poll. The setup technically permitted them to take the poll multiple times, getting paid each time.  Some subjects were exhorted not to cheat in this way; others told that the experimenters trusted them not to cheat; others were given terms and conditions forbidding cheating. Subjects exhorted not to cheat and trusted not to cheat both took the opportunity to game the system significantly less often than those presented with terms and conditions. Assuming external validity, this raises a bit of a puzzle: why do firms attempt to control user behavior through T&Cs? Maybe T&Cs aren’t actually intended to control behavior at all! I wondered, but didn’t ask, if T&Cs that wrapped up with different formalities (a scan of your fingerprint; a blank box requiring you to actually try to sign with your mouse) would get to a different result.  Maybe T&Cs now signal “bad terms that I don’t care to read” instead of “contract-promise.”  That is, is it possible to turn online T&Cs back into real contracts?

Next, I went to Law and Psych to see “It All Happened So Slow!”: The Impact of Action Speed on Assessments of Intentionality by Zachary C. Burns and Eugene M. Caruso. Bottom line: prosecutors should use slow motion if they want to prove intent. Second bottom line: I need to find a way to do cultural cognition experiments that involving filming friends jousting on a bike. I then hopped on over to International Law, where Adam Chilton presented an experimental paper on the effect of international law rules on public opinion. He used a mTurk sample.  I was a concern troll, and said something like “Dan Kahan would be very sad were he here.” Adam had a good set of responses, which boiled down to “mTurk is a good value proposition!”  Which it is.

After lunch it was off to a blockbuster session on Legal Education. There was a small little paper on the value of law degrees. And then,  Ghazala Azmat and Rosa Ferrer presented  Gender Gaps in Performance: Evidence from Young Lawyers. They found that holding all else equal, young women lawyers tend to bill somewhat fewer hours than men, a difference attributable to being less likely to report being highly interested in becoming partners while spending more time on child care.  What was noteworthy was the way they were able to mine the After the JD dataset. What seemed somewhat more troubling was the use of hours billed as a measure of performance, since completely controlling for selection in assignments appeared to me to be impossible given the IVs available.  Next, Dan Ho and Mark Kelman presented Does Class Size Reduce the Gender Gap? A Natural Experiment in Law. Ho and Kelman found that switching to small classes significantly increases the GPA of female law students (eliminating the gap between men and women). This is a powerful finding – obviously,it would be worth it to see if it is replicable at other schools.

The papers I regret having missed include How to Lie with Rape Statistics by Corey Yung (cities are lying with rape statistics); Employment Conditions and Judge Performance: Evidence from State Supreme Courts by Elliott Ash and W. Bentley MacLeod (judges respond to job incentives);  and Judging the Goring Ox: Retribution Directed Towards Animals by Geoffrey Goodwin and Adam Benforado.  I also feel terrible having missed Bill James, who I hear was inspirational, in his own way.

Overall, it was a tightly organized conference – kudos to Dave Abrams, Ted Ruger, and Tess Wilkinson-Ryan.  There could’ve been more law & psych, but that seems to be an evergreen complaint. Basically, it was a great two days.  I just wish there were more Twiqbal papers.



You may also like...