December 21, 2016
December 15, 2016
Professor Seto (Loyola/Los Angeles) asked me to share the following analysis of the ABA's new proposed bar passs standard:
There is a serious mathematical problem with the ABA’s 75% proposal.
Assume for illustration’s sake that bar results are normally distributed – that is, that they will cluster around the mean in a normal distribution. (There is no reason to believe that deviations from a normal distribution affect the analysis given below.)
In a state in which only one school provides the bulk of bar-takers, the bulk of that the state’s normal distribution will be reflected in the school’s bar passage rate. Assuming that the state bar passage mean is reasonably high, the school will always meet the ABA’s requirement, regardless of the quality of the education it provides.
In a state with multiple law schools, however, the normal distribution of bar results will instead be distributed unevenly across those schools. Some schools’ results will incorporate the lower end of that normal distribution. If there are enough schools in the state and students are sorted by bar-taker aptitude across those schools, the bottom schools will always fail the ABA’s requirement, again regardless of the quality of the education they provide.
Note that the difference between the two results – always meeting and always failing – is not because students are being admitted who shouldn’t be. If, in the state with multiple law schools, all such schools were to merge into one single mega-school, without any change in admissions policies or educational quality, that mega-school would absorb the bulk of the normal distribution in bar passage and would always meet the ABA’s requirement.
What this means is that the ABA’s proposal should chiefly impact lower-ranked schools in large states – not because the quality of the educations they provide or because they are admitting students who should not be lawyers, but because of the mathematics of normal distributions. To the extent those lower-ranked schools disproportionately service students from underserved communities, the ABA’s proposal will disproportionately impact diversity in the legal profession. And again, this is not because schools are admitting students who shouldn’t be admitted. It follows from the failure of the ABA to take the mathematics of normal distributions into account in structuring its proposed standard.
Conversely, the ABA’s proposal should have no significant impact on schools in states with few schools – again regardless of the quality of the educations they provide or the students they admit. And again, this follows from its failure to take into account the mathematics of normal distributions.
I question whether a rule that it likely to have a dramatically different impact in different parts of the United States for reasons having nothing to do with the rule’s ostensible purpose should be adopted. It is likely that the ABA is seriously considering the proposal in its current form in part because lawyers are, for the most part, innumerate. Unfortunately, this reinforces the argument that the ABA should not have regulatory jurisdiction over law school accreditation – precisely the argument the proposal is intended to rebut.
Comments are open for those who have ideas about how to address this difficulty. Post your comment only once, it may take awhile to appear.
December 10, 2016
From their various data sources, two tidbits: LSAT takers in September were up 1% compared to last year, while applicants were down about 5% in December compared to last year. The trend has been towards more applications later in the season, so I expect in the end this year will not be much different than last year in overall volume. We still do not have the December figure for test-takers.
December 01, 2016
U.S. LLM Programs Probably Benefit International Students (Part 2): Students who return home (Michael Simkovic)
In part 1 of this 2 part post, I noted that U.S. LLM programs may provide substantial financial benefits to students who remain in the United States, even if they do not necessarily pass a U.S. bar exam. But what about the LLM graduates who return to their countries of origin?
While good data is hard to come by, it is easy to imagine how studying U.S. law might benefit lawyers working outside of the United States and hard to imagine how international programs could continue to attract applicants if returning lawyers did not speak favorably of their studies abroad. Indeed, some countries explicitly encourage students to study abroad and return home. U.S. higher educational institutions often have far better resources than those available in international students’ countries of origin.
Many efforts to “harmonize” and “modernize” corporate, commercial, and regulatory law have historically been efforts to adopt a more U.S.-like approach. For example, the Uniform Commercial Code is greatly admired and appreciated by international bankers and lawyers who have grappled with other countries’ fragmented and inconsistent approaches to secured credit. The U.S. Bankruptcy code is thought to be more conducive to entrepreneurship and consumer finance than the approach in many foreign jurisdictions. Many securities are issued under U.S. law. Many international transactions explicitly choose New York law. U.S. anti-trust law can extend to companies based outside of the United States. U.S. tax law helps drive many international transactions.
Derek Muller recently noted evidence suggesting lower first-time bar passage rates for international students taking the bar exam compared to graduates of ABA approved 3-year JD programs. Although his headline is provocative, Muller is careful to avoid overstating the significance of this finding. Lower bar passage rates are expected, considering that for many LLMs, English is a second language, while for most ABA-approved-JD graduates, English is a first language. On the bar exam, LLMs compete without accommodation on a timed exam with a large essay component with students who either speak English as a first language or have had many more years to immerse themselves in the English language. It can take years for immigrants to become roughly substitutable for U.S.-born workers.
But as noted in my previous post, from a consumer protection perspective, the relevant comparison is not U.S.-born-citizens versus immigrants or foreigners. It’s immigrants or foreigners with more U.S. education versus immigrants or foreigners with less U.S. education.
1-year degrees appear to provide benefits. It seems likely that a 3-year degree would provide international students even more benefits than a 1-year degree. But a 3-year degree would also cost more and take more time to complete, and might be unappealing to those planning to return home.
The U.S. has a history of using English-literacy tests to exclude immigrants from Southern and Eastern Europe while permitting immigration from Northern Europe. Today, historians generally view these policies as discriminatory. It would be unfortunate if the ABA were to take a parallel approach today to deny access to U.S. legal education to lawyers from non-common-law countries on ill-considered consumer protection grounds. Many of those international students intend to return home rather than practice law in the United States. Some of those international students may face even lower odds of passing the bar exam in their countries of origin than in the United States, language barriers notwithstanding.
November 29, 2016
U.S. LLM Programs Probably Benefit International Students (Part 1): Students Who Stay in the U.S. (Michael Simkovic)
At a conference I recently attended, some law professors and administrators seemed willing to assume the worst about LLM and international JD programs. They seemed to think that LLM programs provide revenue to law schools but do little to help students. This stoked my curiosity about international law programs. It seems likely, as conference attendees suggested, that LLM admissions are less exclusive than JD admissions at comparable institutions. But lower selectivity does not imply that LLM programs fail to help their students.
Immigrants are generally at a disadvantage relative to those born in the United States because of language, culture, and legal issues. But comparing immigrants to U.S.-born individuals tells us nothing about the benefits of U.S. education for immigrants. Instead, we can either compare immigrants to those from their countries who stay home, or compare immigrants to each other by education level.
Decades of peer reviewed labor economics research indicates that additional education boosts earnings. Moreover, Immigration to the United States can often dramatically boost earnings for immigrants over the long term. Are foreign LLM programs or international JDs exceptions to widely observed trends regarding benefits of education and immigration?
While data is limited, the unsurprising answer appears to be: Probably not.
Using U.S. Census data (ACS), I found (in a very preliminarily, quick analysis intended primarily to satisfy my own curiosity) that an LLM might boost long term annual earnings by as much as $25,000 on average compared to a bachelor’s degree (depending on unobserved selection effects, the causal boost could be lower since these are cross-tabbed means by race sex and education level). The earnings boost from a JD for immigrants might be around two or two and a half times as high as the boost from an LLM.
November 25, 2016
November 22, 2016
That's the Blog Emperor's characterization of the latest results, though California still has many graduates of non-ABA-accredited law schools taking the California bar and passing at very low rates (1 out of 4 or less).