Monday, April 3, 2006

Bar Pass Rates and Employment Rates: Some Curious Discrepancies

U.S. News prints both the bar passage rate for each school in the jurisdiction where most graduates took the bar, as well as the 9-month employment rate, as self-reported by the school.  The latter figure is, as noted before, a "work of the imagination," partly because of the loopholes in the reporting criteria that U.S. News employs (so, e.g., hiring unemployed grads as Research Assistants counts as "employment"; counting as "not seeking employment" any unemployed graduate who turns down an offer of a Research Assistantship by the school), and partly because U.S. News has no effective way to verify the reported data.

There is one public piece of data that might be thought a useful check on the self-reported employment figures, namely, the bar passage rate.  It turns out, however, that almost all schools report a somewhat higher 9-month employment rate than their bar passage rate.  Two obvious explanations for how this could legitimately be so are as follows:  (1) at many large firms, associates who failed the bar will be able to remain employed while preparing to take the bar again; and (2) the very best students are more likely to have opportunities to work outside the immediate jurisdiction, meaning the pool of students who take the "local" bar are likely to be somewhat weaker than the overall pool of students seeking employment.  (A third factor may be that the bar passage rates do include a small number of repeat test takers, folks who may not have been in the relevant employment pool.)

For most of the top law schools, the gap is modest between the precentage of all graduates reported as employed 9 months out and the percentage who passed the bar in the local jurisdiction:  Harvard reports 3.6% more graduates employed 9 months out than passed the New York Bar; Chicago reports .8% more; Stanford, 7.1% more; Duke, 6.1% more; Texas, 6% more; Vanderbilt, 6.6% more.  (It is striking how many of the top law schools now list New York as the bar exam that the largest number of its graduates take:  not just Columbia, NYU, and Cornell, but also Harvard, Yale, Penn, Michigan, Duke, and Georgetown.)  There are three exceptions among the top schools that stand out, all in California (which, of course, has the most difficult bar exam in the nation):  Berkeley reports 15.1% more graduates employed after nine months than passed the bar in California; UCLA reports 13.7% more; and USC reports 18.5% more.  The entire explanation in these cases may have to do with (2), above.

If we now look at the rest of the top 100 according to U.S. News, here are the schools with the largest gaps (20 points or higher) between employment rates after 9 months and bar passage rates in the primary jurisdiction in which graduates practice; in each case the figure, below, represents the additional number of graduates reported as employed compared to the bar passage figure:

University of Denver (32.8%)

Loyola Law School, Los Angeles (31.7%)

University of San Francico (29.5%)

McGeorge School of Law, University of the Pacific (27.6%)

Santa Clara University (26.6%)

Northwestern School of Law, Lewis & Clark College (25.6%)

Tulane University (24.8%)

Florida State University (21.9%)

Rutgers University, Newark (21.9%)

Seattle University (21.4%)

George Mason University (20%)

Seattle University, for example, reports 100% graduation after nine months, while only 79.6% of graduates pass the Washington Bar (the statewide average for Washington is 79%).  Given U.S. News reporting instructions, it is possible that most of those failing the bar were not "seeking employment" and so were excluded from the calculations.  Notice, of course, that California schools are also disproportionately represented among those with unusually large gaps. 

Any thoughts on how to explain these kinds of discrepancies?  No anonymous postings; comments may take awhile to appear, so post only once.  Thanks.

Rankings | Permalink

TrackBack URL for this entry:

Listed below are links to weblogs that reference Bar Pass Rates and Employment Rates: Some Curious Discrepancies:


While the statistic may be "employed 9 months out", they likely don't take the snapshot 9 months out.

Thus, anyone who received a job offer AT ANY TIME is likely counted. Thus, the employment numbers are calculated before the test is even taken, when employers are bullish that their selected candidate will, in fact, pass the bar.

This effect could be even more exaggerated if, after failing the bar, a previous law firm hiree gets hired as a research assistant. Now you have two jobs for one person, and the percentage goes up.

Posted by: Michael Risch | Apr 3, 2006 9:04:50 AM

The wording of the US News question does not distinguish between law or non-law jobs or between jobs held before or during law school and those acquired after graduation from law school. Bar passage would be irrelevant to the kinds of employment that count in this question. Slinging hamburgers counts as employment, so why try to make sense of this senseless question? Every school should report 100% to make it even more useless.

Posted by: stephen burnett | Apr 3, 2006 10:40:14 AM

With respect to the California law schools on the above list, one significant development is the shift by law grads in recent years from legal to non-legal employment. The interest on the part of non-legal employers (insurance companies, business firms, etc.) in law graduates has grown. And I think this trend is particularly important in California for two reasons: (1) It's a huge, diverse economy, and thus opportunities for law grads are more plentiful; (2) Grads seldom want to leave California and will often take less impressive jobs in order to maintain their Cali lifestyle.
The intriguing question is whether career services offices at these schools are aggressively encouraging their students to "drop out" of traditional law jobs in order to pad their employment stats.

Posted by: Dan Rodriguez | Apr 3, 2006 11:12:44 AM

Regarding Mr. Risch's comments: I believe the factual assumptions about timing in his posting may be mistaken, and that the 9-month figure is for roughly February following the year of graduation, by which time summer bar results are known. It is true that the prior winter's bar results are also included, and that, of course, is a different pool of job seekers. It is also true that the 9-month figure includes many folks employed much earlier than that, but "in theory" they aren't supposed to be counted if unemployed at nine months. One may wonder, of course, how accurate the reporting is on that score.

Posted by: Brian Leiter | Apr 3, 2006 12:34:07 PM

Especially for expensive private schools that are not a dominant player in their local legal markets--which I think describes several of the schools on your list--the prospects for local legal employment of the bottom 20% of the class may not be so good. And as Dan Rodriguez pointed out, the strength and diversity of the California economy may offer better paying non-law jobs for that group, for which student loan repayment may be an important concern.

Posted by: Fred Tung | Apr 3, 2006 12:37:01 PM

Brian--Another factor that may be at work is judicial clerkships. Many judges don't require their law clerks to take the bar exam prior to the start of a clerkship. And many clerkships begin in June, which makes taking the bar prior to starting virtually infeasible or at least very difficult. As a consequence, schools that have a high percentage of students who go on to clerk are apt to have higher post-graduation employment rates than bar passage rates would suggest should be feasible.

Posted by: Keith Sharfman | Apr 3, 2006 4:08:37 PM

Some discrepancies might be the result of how US News measures bar passage.

It might be that, say, 50% of a school's graduating class of 200 took the bar in, say, Ohio and the other 50% did not take the bar, took it elsewhere, etc. Since US News counts only bar passage in the state where the largest number of graduates took the bar, the other 50% affect only the employment numbers and not the bar passage stats.

With a 70% pass rate on the Ohio bar and 100% pass rate elsewhere, that would mean that 170 of the 200 passed the bar. If all of those were employed at 9 months, there should be an 85% employed at 9 months statistic and a 70% bar passage stat.

People taking the bar in "other" jurisdictions are likely to be disproportionately better students (people who got NY jobs, for example, despite going to an Ohio law school are more likely to be higher in class rank) but that's not 100% true. Further, the "other" bars may be harder than the "home" bar, or preparation may be reduced (e.g. if taking a class in local law helps). So it would be unusual to see a 100% pass rate among students on the "other" bar exams.

I would therefore expect the 9 months number to be somewhat higher than the bar passage rate. Whether it could be as much higher as the reports suggest without some creative accounting, however, is a different question and I share Brian's suspicions on this point. Bill Henderson (IU) and I are working on a paper on post-graduation measures in US News and we'll dig into this further and report back what we find.

Posted by: Andrew Morriss | Apr 4, 2006 4:52:57 PM

Several of the law schools on your list have "part-time" programs, e.g., McGeorge, Loyola and USF, and many, perhaps most, of the students in those programs were employed throughout their law school careers. Thus, even if the failed the (horrendous) California bar on their first try, those students will be employed nine months after graduation.

Of course, we should not overlook "optimistic" estimates from law schools to USNWR . . .

Posted by: Kneave Riggall | Apr 5, 2006 9:46:23 AM

Concerning the timing point, I think I have to agree with the statement that the snapshot for ascertaining employment is often taken before the 9-month period is up. Moreover, even temporary jobs count apparently, and the OCI sometimes will not change employment status from "employed" to "unemployed" after the short-term job is over. For instance, I am currently in a graduate program, but I am listed as "employed" on the OCI website. Perhaps this is because I had a job this past summer? If this phenomenon is reoccuring in other cases, then I imagine the figures are skewed by a few percentage points.

Posted by: Ben Love | Apr 11, 2006 7:33:44 AM

Post a comment