What we don't know about the July 2020 bar exam and Covid-19: a lot

My post last week offering the tentative conclusion that it appears no one contracted the coronavirus during the July 2020 administration of the bar exam languished in obscurity for most of the week until something spurred attention to it, offering voluminous feedback, in some ways the usual assortment of decidedly-helpful to decidedly-unhelpful responses.

There were a couple of tranches of responses I wanted to highlight. The first were a few people encouraged by the results, in particular thinking about how test-takers and bar exam administrators adopted some useful practices ahead of the exam. Test-takers were unusually cautious in self-isolation before the exam to avoid getting turned away from the exam; basic standards like mask or physical distancing requirements go a long way for a relatively short period of time in a room; test-takers traveled by driving in lieu of mass transit to avoid illness. In short, maybe these results shouldn’t be much of a surprise. (They would have been a surprise to me in March, when my robust “might” was all I could muster for an in-person exam!)

The second—and there were much more of these—were those critical of my conclusions. I want to focus on two of the more helpful points to highlight the weaknesses of my post.

The first bucket are questions about whether I asked the right question. As I wrote:

I reached out to several bar licensing authorities to see, as I put it, “whether any person contracted Covid-19 as a result of the administration of the July 2020 bar examination, or if any person contracted it as a result of their traveling to or from or lodging around the time of taking the bar exam.” (Should I have asked a different question? Well, I tried….)

Lots of people suggested I asked the wrong question. For instance, “as a result of” put too much of a causal emphasis, which bar licensing authorities would not necessarily be able to answer—or that made it easier to answer “no.” A positive test shared after the fact could still be reported as one not “as a result of” the exam. Fair critique! My goal was a narrow and specific concern about what we know. But my inquiry could of course have been broader. Additionally, as I noted, “some jurisdictions did emphasize they asked test-takers to disclose if they contracted the coronavirus within 14 days after the exam, and none did so. (That would sweep more broadly than contracting it during the bar exam, of course.)” For a follow-up inquiry (or, if you’re thinking about you, your next inquiry!), I might ask something more broadly, without the causal aspect.

Second, and relatedly about the scope of the question, was whether jurisdictions had contact-tracing protocols in place. Again, some did volunteer the post-exam 14-day reporting period, which, again, suggested to me some contact-tracing protocols in place. But, would it have been more useful to ask—before, during, after the exam—for bar licensing authorities to specify what, if any, contact-tracing protocol they had? Or what their negotiations with the local health officials looked like to do so? Absolutely.

Much of my encouragement came from Colorado's process, because (1) it appears people in the room were all made aware (although I cannot tell how much was word of mouth as opposed to formal notification), and (2) Colorado bar authorities apparently knew the negative results of the test-takers in the room who later took a test as a matter of peace of mind. (That was despite the fact that Colorado's local health authorities said that testing of everyone in the room wouldn't be required because of existing social distancing protocol.)

In short, are there more helpful questions? Absolutely. Is there more evidence we could gather? Definitely. Should we (err, should someone) do so? Totally.

We don’t know the results in a lot of jurisdictions. We don’t know who opted out of the exams. We don’t know how online administrations or administrations split across months look. We don’t know the psychological effect it had on test-takers and how it compares to previous administrations of the exam. We don’t know a lot. Indeed, I only got to seven jurisdictions in my original post! There are a lot of important questions still to ask. (Some I’m working on in the months ahead!)

So… do it! Ask good questions (or, questions better than mine!); engage in good faith with bar licensing authorities and public health officials; determine what we can do going forward!

The most helpful piece, I think, is from Professor Paul Horwitz over at PrawfsBlawg, with this bit I’d like to emphasize:

Voicing ex ante concerns about risks is perfectly understandable. One can hardly wait until after the event to express worries about future risks. And the outcome doesn't mean the concerns about risk were unwarranted. But it seems to me that, at least for those whose arguments were based on academic expertise, or invoked that expertise and appealed to past and ongoing empirical study of the issue, or otherwise invoked a kind of academic or data-driven or scientistic authority in making various arguments, there is an arguable duty to follow up and see what the data ultimately revealed about the accuracy of those warnings.

Claims about the future are often too easy to make. I really like to go back and see how often those (often dire) predictions come true. Sometimes these questions are hard to figure out. This question in particular seems very hard to figure out—hence, a post that opens with “appears,” turns exclusively on the causal questions, and includes a lot of caveats (go back and read all of them!). And, of course, even if true, it doesn’t answer the much bigger questions about the right approach in each jurisdiction (see, e.g., the caveats). It only helps us move marginally forward on one piece of a very complicated puzzle.

The Twelfth Amendment and the Libertarian Party ticket in South Carolina in 2020

The Libertarian Party nominated Jo Jorgenson and Jeremy Cohen as its presidential and vice presidential candidates. The Libertarian Party has ballot access in nearly every state, including South Carolina. But there’s a Twelfth Amendment wrinkle to this ticket this year: Ms. Jorgenson and Mr. Cohen are both inhabitants of South Carolina, as their filing reports.

The Twelfth Amendment requires that when presidential electors meet to vote for a prsident and vice president, on of those two names, “at least, shall not be an inhabitant of the same state with themselves.” That means South Carolina electors cannot vote for two South Carolina inhabitants. It means that the ticket, as printed, is an unconstitutional one.

Of course, either Ms. Jorgenson or Mr. Cohen might move before the electors convene. In Jones v. Bush, for instance, Dick Cheney moved from Texas to Wyoming ahead of the 2000 presidential election to avoid this very issue in Texas.

While it seems unlikely that South Carolinians would choose the Libertarian Party ticket this fall, I’ve repeatedly argued that decisions like these, even involving disputedly-eligible tickets, should be left to the political process. And it’s of course the case that the facts could change before, or here even after, Election Day to allow the ticket to become eligible—an extra reason to do so.

South Carolina law purports to bind electors to vote for the ticket and not vote “faithlessly,” or else electors are subject to criminal penalties. But the law also provides that the executive committee of the party “may relieve” electors of this duty if “in the opinion of the committee, it would not be in the best interest of the State for the elector to cast his ballot for such a candidate.” That might be a case where a candidate dies. It might also be a case where vote might be rejected by Congress. That might allow the Libertarian electors to vote for someone else for, presumably, vice president.

In any case, the mere prospective of a potential Twelfth Amendment problem is no reason to exclude a ticket from the ballot. But, there is a lurking Twelfth Amendment issue here.

Kanye West, Arizona, federalism, and party disaffiliation statutes in presidential elections

I haven’t weighed in much on the Kanye West presidential run since mid-July because, well, I haven’t much of legal interest to write. There have been some typical ballot access issues and some petitioning challenges. He’s gotten ballot access in a few places but not most others, and he’s been kicked off the ballot in several places. He’s not running a “serious” campaign, as I indicated in July. And so I haven’t spent much effort thinking about it.

But the challenge in Arizona to Mr. West’s candidacy piqued my interest, because a district court got the law, in my view, quite wrong.

Arizona law provides under Section 16-341, “Any qualified elector who is not a registered member of a political party that is recognized pursuant to this title may be nominated as a candidate for public office otherwise than by primary election or by party committee pursuant to this section.” (Emphasis added.) Mr. West is apparently a registered Republican in Wyoming. Plaintiffs sought to exclude Mr. West from the ballot by arguing that he sought ballot access through a nonpartisan route, which is forbidden if he’s a Republican. The district court agreed, saying that the “most sensible reading” of the statute is that Mr. West is a Republican.

But when we register to vote and affiliate with a political party, we do not affiliate with a “national” party. We affiliate with a state party. In Arizona, there is a Republican Party, recognized under Arizona law, with its headquarters in Arizona. But Mr. West is registered with the Republican Party of Wyoming, not of Arizona. Wyoming’s Republican Party is not recognized under Arizona law.

True, the Republican Party of Arizona holds a presidential nominating primary to send delegates to the Republican National Convention. And true, this November, the Republican Party of Arizona, like the other Republican Parties throughout the country, will name Donald Trump and Mike Pence as their presidential ticket. But how the state parties affiliate with the national apparatus is a different question.

Maybe this seems too cute by half, but it’s a testament to how presidential elections just look different. For instance, Minnesota doesn’t have a “Democratic Party.” It has a Democratic-Farmer-Labor Party, but it’s affiliated with the national Democratic Party and participates in the Democratic National Convention.

We might want a disaffiliation statute if you’re an Arizona Republican running for an nonpartisan slot for, say, Congress or the state legislature. The Supreme Court in Storer v. Brown upheld such a requirement to protect the interests of the political party from sham candidacies or party raiding. But for presidential tickets, however, we’d need more express statutory clarity—at least, I think so, and my reading of the statute, I think, is the more persuasive view.

Puzzles when crossing state lines in election contests is hardly a novel problem. In 2015 in Arizona, for instance, a court threw out a criminal conviction for someone accused of voting twice, once in Arizona and once in Colorado. The court noted that the defendant hadn’t voted twice under the statute—the statute applied to Arizona elections, not other states’ elections.

It might be too late for an appellate court to correct this misunderstanding of state law. I don’t know whether Mr. West would appeal, or if there are other bases for throwing out his petition. But it’s simply to point out, I think, that disaffiliation statutes simply look different in presidential elections with out-of-state candidates.

UPDATE: On September 8, the Arizona Supreme Court affirmed the exclusion of Mr. West from the ballot, but not on this basis, which is good news. That said, it does appear that the Court used a novel rule previously inapplicable to other candidates, which may be its own problem….

It appears the July 2020 bar exam did not spread Covid-19 among any test-takers

There was much concern ahead of the July 2020 bar exam that administering the exam would be unsafe for test-takers. Indeed, it’s a reason why many states postponed their exams, shifted to an online exam, or offered diploma privilege to some cohort of would-be test-takers.

Several states, however, administered the July 2020 bar exam in person, on schedule. And it appears that there were zero instances of the spread of Covid-19 related to the administration of the exam in any of jurisdictions.

(Let me offer a few throat-clearing caveats here. Past performance is no indicator of future success in later administrations. Some set of test-takers had difficulty studying for the bar exam ahead of the July 2020 exam; some may have had difficulty taking it in masked environments. We’ll see what the results of these administration look like, of course—it might be that pass rates decline by some degree among some because of these factors. But, then again, they might increase if people had little to do or few places to travel and more time to study! Some people also probably withdrew from the administration of the July 2020 bar exam due to family commitments or high concern of the risk of contracting Covid-19 during the administration of the bar exam, which is not a trivial cost. Some set also may well have opted into these particular tests in UBE jurisdictions, as the score could transfer to other jurisdictions that had postponed administrations. In short, there are a lot of open coronavirus-related questions that this post is not about, but additional concerns to think about in evaluating costs and benefits.)

That said, if we are concerned about the risk of spreading Covid-19 during the bar exam, this should be encouraging. There are a couple of ways to think about Covid-19 spread, too. The first is during the administration of the exam itself—several hours indoors for a couple of days. The risk of transmission seems particularly low here if adequate precautions are put in place. The second is the broader concern of spreading Covid-19 during travel to and from the bar exam or lodging in another city during the administration of the bar exam itself. That’s a longer time period, a greater likelihood of interacting with others, and a different kind of risk much more difficult for a licensing authority to control, so I don’t want to minimize the risk of spread to “only” the several hours of test-taking.

My instinct is that if someone did contract the coronavirus during the administration of the bar exam, we’d probably know by now. That is, contact tracing would require contacting dozens of test-takers, and that news would presumably make it to some media outlet. Of course, someone might privately have become ill and never disclosed it to the state bar. Or there could have been a great deal of discretion in how these inquiries proceeded. Or maybe contact tracing simply never took place. There are, admittedly, limits to silence.

I reached out to several bar licensing authorities to see, as I put it, “whether any person contracted Covid-19 as a result of the administration of the July 2020 bar examination, or if any person contracted it as a result of their traveling to or from or lodging around the time of taking the bar exam.” (Should I have asked a different question? Well, I tried….) Not everyone got back to me, and I didn’t reach out to everyone, so it’s also limited in this respect, too.

So far, representatives in Colorado, Iowa, Mississippi, Montana, North Carolina, North Dakota, and West Virginia confirmed that, to their knowledge, no one contracted the coronavirus on account of the administration of the bar exam. (I’ll update if I have updates.) Some additionally confirmed that no proctors or staff contracted it, either.

Proving a negative is tough, but some jurisdictions did emphasize they asked test-takers to disclose if they contracted the coronavirus within 14 days after the exam, and none did so. (That would sweep more broadly than contracting it during the bar exam, of course.)

Colorado was of particular interest, as one test-taker learned immediately after taking the bar exam she tested positive for the coronavirus. No one reported contracting the coronavirus as a result of the Colorado exam. Indeed, according to a representative, about half of the test-takers in the room with this test-taker ultimately took a Covid-19 test, and all came back negative.

This should all be encouraging news for bar licensing authorities. Yes, there remain costs, as noted in the parenthetical above, about administering a bar exam during the pandemic. But, I think, if bar licensing authorities have the space to distance test-takers, it provides ample opportunity to continue administrations in February 2021 (or later this fall in jurisdictions that postponed the exam), it may well be that administering the test as scheduled rather than postponing was the better option. Jurisdictions like North Carolina and Colorado each nearly 700 test-takers, which shows that the administration doesn’t have to happen exclusively on a very small scale. (North Carolina and Colorado had the 14th- and 15th-most test-takers in the July 2019 administration of the bar exam, and neither is offering an additional administration later this fall.) That said, scaling to the 10,000 or more test-takers in New York could present very different challenges!

And, of course, we’ll wait for more data to roll in—whether pass rates change, whether the demographics of test-takers differed, and so on. Some states have preemptively and temporarily lowered their cut scores to account for some coronavirus-induced changes, for instance.

North Carolina is a great example—their total test-takers declined from 783 in July 2019 to 668 this July. But the cut score was lowered slightly, and the pass rate rose from 73% to 83% (perhaps also due in part to some non-random self-selection out of the exam). As of September 1, results are out for hundreds of new law school graduates—and many other jurisdictions languish about their plans.

I don’t purport, as I’ve said since March (!), to have a single answer for every jurisdiction on the bar exam, or what the best steps are for each jurisdiction. Some version of diploma privilege may make sense in some places, modified versions of the exam may look better elsewhere. But on this one little piece of information, we should be encouraged about the health and safety of test-takers for in-person administration of the exam.

It's not bad news: MBE scores rise for July 2020 bar exam

What a tepid title I can offer…. As longtime readers of this blog know, I track performance on the bar exam fairly closely. The July 2020 administration, however, offers only a limited window into what’s happening in legal education. Only some jurisdictions held a July 2020 bar exam; even in those jurisdictions, self-selection among test-takers this time around is probably not random (perhaps skewed toward better test-takers—those with ample resources to study and prepare, for instance).

So the NCBE’s update on performance on the July 2020 bar exam is of limited value. Scores went up. They went up a lot. In fact, they reached an all-time record high! It’s great news for those who took the July 2020 bar exam, of course! And pass rates will probably rise significantly in these jurisdictions that administered the exam.

View fullsize

Now, I won’t get overly excited about this chart—at least, I usually use this chart to make an overall evaluation of the state of legal education, and it really isn’t great to do that here. About 5700 test-takers took the July 2020 MBE. That’s compared to 45,000 or so in a typical July! (Removing New York alone, of course, accounts for 10,000 or so tests.) 5700 test-takers is a big sample, but it’s relatively a small cohort that tells us little about the overall state of legal education.

The MBE scores were up 5 points over July 2020, but they were up only 3.5 points over 2019 in the 23 jurisdictions that administered the July 2020 bar exam. That means if we added in other jurisdictions, we may well see a smaller spike.

Even within that cohort of states that administered a July 2020 exam, first-time test-takers were about 75% of all test-takers, about 6 points [I think] higher than the previous exam. So the pool of test-takers were stronger as it skewed toward first-time test-takers, which likely accounts for at least some of the improvement in scores. (It also suggests that those who failed the bar exam previously increasingly opted out of this administration of the exam.)

In short, this is terrific news for the crop of test-takers who did take the exam. (Not to mention those in the four states with some version of “emergency” diploma privilege this fall.) Maybe the absence of, say, weddings and night life meant limited travel and greater focus on studying for the exam as an additional contributing factor to success. One could only guess. But it probably doesn’t tell us a whole lot about the overall state of legal education and the cohort of recent law school graduates.

Proportional representation not required in union representation under guarantee of "fair and democratic elections"

That’s the finding from the First Circuit in its recent decision in Conille v. Council 93. The challengers sought equal (something akin to “one person, one vote” or proportionate representation in the makeup of elected vice presidential positions. While some of the selection of representatives were not in dispute, here’s how the court framed the mechanics of the disputed election mechanism:

To choose the vice presidents, Council 93's constitution divides the locals into thirteen legislative districts -- nine geographic and four organizational. These legislative districts do not have independent governing bodies; rather, they function solely as a way to divide delegates to nominate and elect members of the executive board. The Council 93 constitution allocates a specific number of vice president positions to each legislative district. The number allocated to each district bears little, if any, relationship to the number of members in that district. Rather, the allocations are artifacts of agreements made over time as locals have joined the Council. For example, a single vice president is chosen by the delegates representing over 1,800 members in the Vermont district, while four vice presidents are chosen by the delegates representing 1,500 employees in a “Department of Mental Health” legislative district. It is this type of disproportionality that triggers plaintiffs' displeasure.

The core of the dispute turns on construction of terms in the union’s constitution, and courts defer to the union’s interpretation, so winning a “one person, one vote” or proportionality claim would be difficult. But here’s the court’s analysis:

In the alternative, plaintiffs contend that the structure of Council 93's Executive Board violates the AFSCME constitution, particularly Paragraph 4 of its “Bill of Rights,” which provides that “[m]embers shall have the right to fair and democratic elections, at all levels of the union. This includes due notice of nominations and elections, equal opportunity for competing candidates, and proper election procedures which shall be constitutionally specified.”

. . .

The merits of plaintiffs' claim under the AFSCME constitution turns on the question of the proper interpretation of Paragraphs 4 and 7 of the constitution's Bill of Rights. When reviewing a union's interpretation of its own constitution, we defer to that interpretation unless it is plainly unreasonable. . . .

We begin with the constitution's text . . . noting at the outset the absence of any express guarantee of equal or proportional representation on its executive board or on any of its or its subsidiaries' governing bodies, including the subsidiaries' executive boards. This silence contrasts with the text of paragraph 7 of the AFSCME Bill of Rights, which states that “[a]ll members shall have an equal right to vote and each vote cast shall be of equal weight” specifically on issues pertaining to the collective bargaining of contracts, memoranda of understanding, agreements affecting members' wages, hours, or any other terms of employment. The union thus knew precisely how to require equally weighted votes on an issue within the constitution if it wished to do so; instead, it chose to use only the term “fair and democratic” when referring to how elections of its officers must be conducted.

So, we ask if this term by itself renders supererogatory the need to include explicitly the right to proportional representation or an equal vote. See Ramos v. Louisiana, ––– U.S. ––––, 140 S. Ct. 1390, 1400, 206 L.Ed.2d 583 (2020) (suggesting that, with respect to certain rights, the absence of an explicit grant of that right in the text of the Constitution does not imply the right does not exist, but rather, supports the inference that the right “was so plainly included” that stating it explicitly would be “surplusage”).

We think that it does not. We can assume the term “democratic,” by itself, implies a relatively equal right to vote on such matters as one is entitled to vote on, but it is too much of a stretch to say that it must also imply proportional representation on the executive board. The United States is generally considered to be a democracy in normal parlance, notwithstanding the effectively disproportionate representation in the Senate and the Electoral College. See Lyman v. Baker, 954 F.3d 351, 371 (1st Cir. 2020) (“The United States' system of representative democracy [includes] ... the Electoral College and ... Senate.”). Similarly, the term “fair” may suggest some restraint on the procedures used for voting. Paragraph 4 thus requires, as examples, “due notice of nominations and elections, equal opportunity for competing candidates, and proper election procedures which shall be constitutionally specified.” While this list is non-exhaustive, it would have been rather simple for the union to include within it the requirement of equal or proportional representation on all governing bodies, as it did in Paragraph 7 for labor concerns and within this paragraph for competing candidates. Yet, it did not.

The defendants' position that the voting system used to select officers of Council 93's executive board is fair finds further support in the fact that those procedures themselves are approved and subject to change by the convention, in which voting is weighted just as plaintiffs would have it be -- proportionate weight is assigned to the votes of convention delegates based on the number of members represented.

Moreover, the actual behavior being challenged is not precisely an undemocratic or unfair election as a result of an unequally weighted vote. In choosing delegates to the Council 93 convention, union member votes are weighted equally, and these conventions, like a parliament, carry out many of the important legislative powers of the union. Similarly, the five senior officers on Council 93's executive board are chosen according to an equally weighted vote of convention delegates, in much the way that a prime minister might be chosen by a parliament composed of equally weighted votes. It is only the selection of vice presidents to represent the legislative districts that is being challenged.

We have a difficult time saying that the members' right to “fair and democratic elections” necessarily guarantees equal representation on this subordinate body of executive officials. That a cabinet may be made up of appointed officials who do not proportionally reflect the full constituency does not mean that the underlying election was not fair and democratic. The constitutional clause at issue here guarantees only that, when members vote, the process is fair and democratic. The plaintiffs have not contended that their actual elections are otherwise, only that every officer must proportionally represent the constituency. As to that contention, they point to nothing in their constitution that imposes that requirement on Council 93 or any of AFSCME's subordinate bodies.

AFSCME itself also allocates seats and voting for positions on its own executive board in a manner that belies equal representation of every union member, as plaintiffs concede. Although plaintiffs suggest that the ASFCME executive board is somehow more proportional because it is a “hybrid house-senate-like” system, this is nothing more than a particular type of disproportionality. And plaintiffs do not offer any basis in the AFSCME constitution to suggest how much proportionality is enough to be “fair and democratic.” One would expect to find some good reason why AFSCME would view its constitution as outlawing a practice by the Council that it allows itself. Plaintiffs offer no such reason, and we are not willing to create one sua sponte, especially where we are to defer to AFSCME's reasonable interpretations of its constitution. . . . cf. Coleman v. Miller, 307 U.S. 433, 454-55, 59 S.Ct. 972, 83 L.Ed. 1385 (1939) (holding that a “lack of satisfactory criteria for a judicial determination,” inter alia, favors deference to other bodies in deciding what counts as a “Republican Form of Government”).

Of course, the district court reasoned that the voting structure need not be precisely proportional but should at least have “some neutral principle that justifies weighted voting,” and it found no such rationale besides a “hodge-podge of historic deals.” But negotiation and deal-work are the very heart of what unions do. In fact, plaintiffs imply the need to allow unions to make such compromises by suggesting that AFSCME's executive board representation is not problematic because it is like the U.S. House and Senate, which itself is nothing more than an historic deal. See Ronald J. Krotoszynski, Jr., Reconsidering the Nondelegation Doctrine: Universal Service, the Power to Tax, and the Ratification Doctrine, 80 Ind. L.J. 239, 252 (2005) (discussing how the apportionment of seats in the U.S. House and Senate was part of a complicated “deal” involving the balance of power between competing factions). While we recognize that Council 93 is constrained in the deals it can negotiate by the AFSCME constitution, we are hesitant to retroactively read the terms “fair and democratic” to invalidate the bargained-for exchanges that the union members agreed to over the years, especially when AFSCME has made no indication that it believes the term carries such weight and has opted not to restrict its councils in this way. If AFSCME had wanted to tie the hands of its councils in this matter, it certainly could have stepped in and chosen not to approve the provisions in the Council 93 constitution incorporating these deals. Its acquiescence is entitled to some consideration.

A survey of the broad scope of the proposed New York diploma privilege law

A10846 is a diploma privilege bill working its way through the New York legislature. After unsuccessful postponements and questions about how an online version would work, there’s been renewed urgency to help license attorneys in one of the largest and most significant legal jurisdictions in the United States.

I’m sympathetic to some calls for diploma privilege in some jurisdictions in the current environment—but I do think some tailoring is appropriate, as I’ve written. While Washington’s diploma privilege struck me as potentially overbroad, the current text of A10846 far exceeds it, and it’s worth looking at its scope—and what might be amended out of the bill later.

New York does not have a particularly difficult bar exam. And one could look at its July 2019 statistics as a window into what rules for a temporary diploma privilege might look like. Among first-time test-takers from ABA-accredited schools, 86% passed—4748 out of 5517 test-takers. That’s a high pass rate. Of course, it means 769 didn’t pass, and a diploma privilege for all first-time test-takers from ABA-accredited schools would sweep them in, too. As emphasized in my previous writing, however, maybe that’s a tradeoff New York is willing to make, and maybe one could have more robust follow-up of this cohort to try to minimize misconduct or malpractice.

If one took a broader view, more like Washington, to all test-takers, first-timers or not from ABA-accredited schools, it would bring in another 990 test-takers from the July 2019 bar exam. Only 28% (272) of those repeaters passed, so it would add 718 graduates who failed the bar exam multiple times. Again, maybe a tradeoff worth taking in the larger picture.

New York’s bill goes further still. It would extend to “any person who has graduated or will graduate from an American Bar Association accredited law school or who would otherwise be eligible to take the New York state bar examination.” (Emphasis added.) That last clause is significant because New York has a significant cohort of foreign-educated attorneys that take its bar exam each year.

In the July 2019 exam alone, for instance, 2398 foreign-educated test-takers took the bar exam for the first time, but just 1266 passed, a 53% pass rate. Another 1161 repeated the bar exam, and 250 passed, a 22% pass rate. That means over 2000 foreign-educated test-takers failed the July 2019 bar exam. All would be admitted under this rule.

I’ve pointed out in the past how non-JD legal education is on the dramatic rise in American law schools, and how foreign-educated LLM degrees are on the rise. But bar passage rates remain low, likely in part due to language barriers. These test-takers remain a significant cohort of overall test-takers.

Unlike other states that have had temporary “emergency” diploma privilege rules, New York’s is not limited to those who previously registered for the July 2020 bar exam. It extends to all prospective attorneys, through September 30, 2021 (or until the end of the Covid-19 disaster, whichever is later), who satisfy the other requirements, like character and fitness requirements.

That could potentially sweep in many more attorneys who’d otherwise be eligible but have failed the bar exam in the past—they might be in New York (the law is limited to those “who intend[] to primarily practice in the state of New York”) and practice. If they complete “100 hours under the supervision of a permanently admitted attorney,” they would be eligible for permanent admission to the bar. (One hundred hours is two and a half 40-hour weeks.) That’s much lower than, say, Utah’s 350-hour supervised practice requirement.

In short, if enacted as written, it’s possible to see thousands of new attorneys in New York—perhaps much more than the equivalent of the 3500 who failed the July 2019 bar exam if one looks at the scope of the rule that might extend to other would-be attorneys who could otherwise meet the rule in the next year.

It’ll be worth seeing if New York enacts the law as written or amends it. Maybe it won’t be enacted at all. But if it is, it’ll be an even more significant experiment in attorney licensing than Washington’s rule, and emphatically one to watch.

Recent dissental track records in the Ninth Circuit

The “dissental”—an opinion dissenting from the denial of rehearing en banc—is a popular tool in the Ninth Circuit. The circuit is large, and it has historically seen high reversal rates before the United States Supreme Court.

I thought I’d look at recent dissentals in the Ninth Circuit, given the arrival of a number of judges appointed by President Donald Trump, and who joined them.

A few methodology notes. A judge may write a “statement” concerning the denial of rehearing en banc, or an opinion concurring in the denial of rehearing en banc; I include neither in this tally. I do not include those judges who publicly voted in favor of rehearing en banc but did not join the dissental. Only those who joined the (or one of the) dissentals are included.

I track 17 dissentals from January 1, 2019 to August 24, 2020. (UPDATE: Thanks to the astute observation of “@fedjudges” on Twitter, I missed a few the first time around!) Most Trump-appointed judges joined the court after January 1, 2019 (but there are some interesting late 2018 ones I exclude!). I include the dates they assumed office, but it’s not clear when, for instance, how quickly some would begin to participate on en banc votes (or feel comfortable joining a dissental after taking the bench). Those who have retired or taken senior status before today are listed as “others.” If the judge did not join a dissental, it’s in gray; if they did, it’s marked with an “x” and is in orange. I cleared out any shading for opinions released on the day a judge took office or earlier.

View fullsize

You can click on the image to zoom in. Historically, Judges Conseulo Callahan and Sandra Ikuta have been among the most reliable dissental participants, with some regular voting from Judge Milan Smith, the only three President George W. Bush appointees left active on the Ninth Circuit. Among these fourteen cases with dissentals, there was one dissental joined by Judge Ronald Gould (a President Bill Clinton appointee) and one by Judge John Owens (a President Barack Obama appointee), and no other instances of a Democratic-appointed judge joining a dissental.

The bulk of joining (or writing!) dissentals has come from Trump appointees (with recognition that some judges joined the Ninth Circuit during this window).

Ryan Nelson: 12

Mark Bennett: 10

Dan Collins: 10

Bridget Bade: 9

Dan Bress: 7

Lawrence VanDyke: 6

Kenneth Lee: 5

Patrick Bumatay: 5

Danielle Hunsaker: 1

Eric Miller: 0

Details of the Biden campaign's concerns about Tammy Duckworth's eligibility as a "natural born citizen"

From this New York Times deep-dive:

Other candidates rose and faded in the process: Senator Tammy Duckworth of Illinois powerfully impressed Mr. Biden’s search team, but his lawyers feared she would face challenges to her eligibility because she was born overseas.

…

Ms. Duckworth was regarded by Biden advisers as among the candidates likeliest to help him achieve a smashing electoral victory in November. But legal advisers to the campaign expressed urgent concern that Ms. Duckworth could face challenges to her nomination in court: She was born overseas, to an American father and a Thai mother. While Mr. Biden’s team believed Ms. Duckworth was eligible for national office, campaign lawyers feared that it would take just one partisan judge in one swing state to throw the whole Democratic ticket off the ballot.

I don’t have strong thoughts on the merits of whether Ms. Duckworth is a “natural born citizen.” And even if 2020 was relatively quiet on natural born citizen challenges (until recently!), it’s interesting to see the political process play out here.

I’ve highlighted that there are several bodies that can ascertain candidate eligibility outside the judiciary: the voters, presidential electors, and Congress. Political parties are a good addition. But perhaps it’s worth adding a category for vice presidential candidates, as the nominee and his team can make judgment calls about eligibility, too.

But, disappointingly, the campaign lawyers made a pessimistic call. My work strongly resists the call for other bodies—particularly judges—to review candidate eligibility. Indeed, I’ve written extensively about that. I think many times there are not rules in place to even allow courts to review a candidate’s qualifications. A declaration from Congress could go a long way to ensure that a candidate’s eligibility is not questioned elsewhere. I think most states don’t have statutes in place to authorize review of candidate qualifications, and, indeed, I think they shouldn’t. And there’s always appellate review—granted, one challenge of late-breaking litigation (like naming a vice presidential candidate in August when ballot printing begins in a matter of weeks) increases risk and uncertainty.

Furthermore, it’s also a fundamental weakness of the National Popular Vote. Could one state exclude a candidate from the ballot? And if so, doesn’t that throw off the “national” popular vote total? Absolutely.

In short, it’s a highly cautious, litigation-avoidance strategy that kept (at least in part) Ms. Duckworth off the ballot. It’s worth considering whether a veepstakes that took place earlier could have successfully resisted litigation that might have challenged her eligiblity.