In a November 29, 2006 New York Times article, Lawyers Debate Why Blacks Lag At Major Firms, Adam Liptak reported that many hiring partners at the "elite" law firms care about grades more than they care about any other hiring criterion. (Of course, the rank of the candidate's school is also important. . . .)
It's my understanding that very few clients actually ask their lawyers to sit for exams. Virtually 100% of clients would prefer that their lawyers think critically, research and write well, and have enough emotional intelligence to do their jobs without offending everyone in sight. Many law professors argue that grades on exams are superb measures of students' analytical skils. For courses in which grades are curved, though, at best the grades demonstrate RELATIVE analytical ability, not absolute analytical ability. For courses in which grades aren't curved, often there are other measures of the student's ability, such as seminar papers and in-class participation. Even if exams did demonstrate absolute analytical ability, they often don't demonstrate writing skill or emotional intelligence, and they certainly don't offer the types of real problem-solving situations that lawyers face. (I don't recall any exams in law school ever blending two or more substantive areas in a single question.)
So at best, exams are a shadow variable (and a faint shadow, at that) for the skills that lawyers need. Law schools need to provide a variety of measures of these skills--everything from seminars to clinics to joint courses with other graduate students--in order to give employers a better picture of the law students' talents. (See A "3D" JD: Stanford Law School Announces a New Model for Legal Education. I suggested something like this at Houston a few years ago and was almost laughed out of the room. It's nice to know that Stanford's doing the right thing.) Even these other ways of measuring a student's set of skills is artificial, but at least seminars, clinics, practica, and joint courses come closer to the types of tasks that junior lawyers have to perform.
Grades can become self-fulfilling prophecies if left unchecked. And if a law firm stresses grades and then "lets in" some "less qualified" candidates for other reasons, then those "less qualified" candidates come in with a stigma already solidly in place. Is there any wonder that, welcomed with less than open arms--for reasons that bear very little relationship to actual performance in the firm--some lawyers may decide to leave for other places that appreciate them more?
This emphasis on grades occurs during faculty hiring season, too, of course, and for the same silly reason: to use "predictors" that might be ten or twenty years old instead of using those predictors that mirror what law professors do every day. Does it really matter whether the person who wants to teach CivPro made an C in that course fifteen years ago (back when grade inflation was less prevalent, by the way) if the candidate has been in a federal courtroom every week since he received his degree? Goodness gracious (and I've always wanted to say that phrase in print)--at what point are we going to look at the person and not just his pedigree?