Search the Lounge


« Pamela Geller Is Not "Morally Responsible" for the Terrorist Attack in Texas | Main | Call for Papers: The First SELS Global Workshop for Junior Empirical-Legal Scholars »

May 07, 2015


Feed You can follow this conversation by subscribing to the comment feed for this post.


I have no problem with conditional scholarships. Schools want a heavy hitter that will bring credibility to the school through impressive achievements when they pay out heavy scholarships. It is just pay for performance, which is common in this economy. To wit, students that do well often try to negotiate scholarships up after first year.

The real holy grail of reform is in the federal student loan system. First priority is getting rid of IBR/PAYE and any sort of public interest/government loan repayment. Then make schools purchase any deliquent loans issued to their students. That would rid this system of so many problems.


I disagree with Dean Yellen on his conclusion that such scholarships do not work when they are properly disclosed. Some schools have eliminated them, and many have not. For prospective students, there is nothing inherently irrational about accepting a one year scholarship, which is what a scholarship unlikely to be renewed is. If a student has a 20 percent or 70 percent chance of retaining the scholarship in year 2, they can and generally do take that into account in calculating the total cost of attendance.


Ideally I'd like to see an accreditation requirement that says "no conditional scholarships other than those requiring good academic standing." That's because I find preying (whether intentionally or not) on students optimism bias, special snowflake syndrome, sunk cost fallacy, and any social/familial pressure they may face not to drop out immoral. But transparency is a good start.

If the justification for these is "motivating" students to perform better, it's just willfully blind to the incentives that are already out there. Your incentive at the level of school that offers a harsh conditional scholarship is to work as hard a possible to get good grades so you can get a good job or transfer. And the nature of the law school grading system means that a student could be perfectly, sufficiently motivated and still fall below top 1/3 just based on the fact that their professor was having a bad day when he graded that half of the stack of papers.

It's possible to imagine a student who already has a guaranteed job when they come into law school, has no desire to have a different job, and is not otherwise sufficiently motivated to try hard, but this scenario can't happen very often.


The argument about the curve is telling: no matter "how hard" one works, there will only be so many "As" and "Bs" and so forth.

Thus, the top 10% might have been the top 15% but for the rule that only 10% are "allowed" by the curve - a rule that requires a prof to judge one conceivably indistinguishable essay against another, with drastic consequences.

Speaking of performance in hypothetical terms is simply ignorant.

No matter how well students do, the math dictates that a certain percentage will lose scholarships in most of these programs.

Justify that with nonsense about "working hard."

The Most Interesting Breh in the World

Just to pile on, there is also a small sample size issue in play too. I.e., there are only maybe 6-8 exams over the first year of law school. Those exams also happen to be (a) the first law school exams these students have ever taken, and (b) at least somewhat less than objectively scored.

This would be like trying to figure out who the best professional baseball players are based on their first 10 or so at bats of their career, and cutting players who fail to, I don't know, hit .300 or something (which, if you're into this sort of thing, I suppose BABIP could kind of be a metaphor for the subjectivity in grading?).

The point, I think, is that incentives are only going to play a minor part in who retains their scholarships (even assuming students meaningfully change their habits in response to these incentives). What you're going to see instead is statistical noise and the inconsistency that comes from a lack of experience.

Add in the optimism bias/special snowflake stuff, and our cultural negativity towards quitting, and you end up with a system that is going to have very few observable positive effects on students.

Derek Tokaz


I think the bar could be a little higher than "good academic standing." For instance, a school might give out 30 scholarships and only 1 students loses it, or 150 and only 2 students lose them. I don't think a school like that is taking advantage of optimism bias.

I could also see your proposed rule working. I think the question would be about where schools choose to draw the good academic standing line, and if it makes sense to have a second line a little bit above that. (For instance, in the class I teach there are two lines, one for failing, and a second for not passing. And it makes sense, but only because of the nature of the class and how grade distributions actually play out.)

Perhaps the important distinction to make is between conditional scholarships and competitive scholarships. The way curves work, if the condition is high, it will necessarily also be competitive. However, many curves leave more wiggle room at the bottom, so in this range something could be conditional without being competitive.

terry malloy

thinking a little more on this:

I have no problem with a conditional scholarship described in the following way: this is a 1 year scholarship. if you manage to attain X at the end of year one, this scholarship will continue for six months, and be re-evaluated against X until graduation. the probability of X based on the grade curve is Y.

That's not the way these were sold.

it was "you have a full ride!!*"

*there is a 70% chance you have a full ride for only 1 year, then you have to pay full freight.


I think it would be better to lower tuition for everyone a bit instead of having "merit" scholarships at all. If most people are borrowing, and you are not addressing need with the scholarships, then you are just enticing the better students to come.

Once you have gone down the road of rewarding "merit", it seems appropriate in the second year to reward those who have shown "merit" in their first year grades, rather than those who simply have come into school with higher LSATs or undergrad GPAs. So it might be a better system for the second year to just re-award the scholarships to those closer to the top of the class. Then the school can help keep its better students from transferring.


The most insidious practice (aside from section stacking) is a conditional scholarship mandating a certain GPA, but not disclosing what percentage of students obtained that GPA or higher under the school's curve. For example, is a 3.3 GPA top 50% or top 10%? There is no way to tell without knowing the school's grading curve. Many students would see a 3.3 GPA and think, based on their undergrad experience, "I can get that in my sleep" without understanding how the curve works at a particular school. And really, how would they know?

It's extremely misleading, especially considering the power and knowledge disparity between schools and 0Ls, and if it were done in any other industry would rightly bring outrage, government investigation, and successful class action lawsuits.

Conditional scholarships making explicit the class rank required to retain are less objectionable, and have probably fallen out of favor only because many schools are willing to give unconditional scholarships to attract students, so using conditional scholarships is no longer a viable option if a school wants to attract quality applicants.



I think those 1/30 or 2/150 would be what you'd see under a good academic standing rule. The other floor I see is usually top 75-80% keeps the scholarship. The point is that the number should be consistently low enough that nobody can credibly claim the primary purpose of the conditional scholarship program is to recoup revenue.

I had many professors and admins tell us some variation of "grades don't reflect how hard you worked or your potential as a lawyer." One even went so far as to say he thought that 1st year grades were essentially random among the middle 80% of the class- a tough thing to admit when everyone understood how much grades mattered. I think most people who have been through law school know that two students could be putting in X amount of effort each and see many percentile points differential in rank during their first year of law school. I'm not sure how to reconcile those experiences with 50% conditional scholarship rates to reach the conclusion that those students just didn't try hard enough.

John Steele

Just out of curiosity, what would it take for students at a particular school to crowd-source the issue of whether their school is section stacking? (AFAIK, that practice is an urban myth, but there's no harm in verifying that.) I suppose the students could (1) ask the dean for a written answer as to whether it's done; (2) set up a website for students to identify which section they are in and whether they have a conditional scholarship; and (3) at public schools issue a FOIA. Presumably, the percentage of CS's would be the same across all sections. Finally, I suppose that the most likely targets would not be the schools where almost all students retain their scholarships.

By way of example, the law students at Berkeley crowd sourced the issue of whether their priority numbers for enrolling for courses each semester were truly random as the administration had said. Turned out, they weren't (due to a software glitch and nothing nefarious, as I understand it).

When we are pondering whether the students were given adequate information about scholarship retention, we might head over to Contract Law Prof Blog to read the recent comment by a Valpo grad. The post is May 4th and the comment is May 8th.



That's a great point. Assume in a 1L class with 100 students, honors grades are given to the top 10%, B's to the middle 80%, and Cs or below to the bottom 10%. The qualitative difference between the "best" B earning exam and the "worst" B earning exam (i.e. the difference between the 11th best and 90th best exam) would certainly be quite pronounced, but no one would ever know except for the professor doing the grading. To make things more pronounced, essay exams are inherently subjective and vulnerable to the grading whims of the professor.

Especially at T2 schools and below, the grading curve exists for two purposes: first, to separate out the very top of the class who will become eligible for biglaw firms, and second to separate out the very bottom of the class, those students likely to fail the bar and not succeed in practice so the school can fail them out before they hurt the US News ranking.


Is this group unfamiliar with the vigororous scholly/stip debates that flourish among law school applicants? I think its arguable that applicants are far better at ferreting out useful information than law faculty and that makes sense since they have the appropriate incentive to do so.

Kyle McEntee

I'm astonished by the need for this post, Bernie, but alas...

Excellent work as always. The legal profession depends on trust and still too many people are making it difficult to restore that trust to its entrance point.



THe typical law school curve is more like 10/30/60.

Doesn't affect your point, but that is the more typical forced distribution.


First, the vast majority of financial aid are not scholarships. I would define a scholarship as a source of money devoted to financially assisting students with school. This is simply a discount, a sale, - the money never EXISTS!

Second, in a macro sense, why are these a good idea? You are subsidizing the very students who are more likely to have a better outcome. There are already programs which subsidize new law debt (IBR, etc.) You would think that the ultimate master who provides the money (the federal government) would insist on a broader distribution for better credit risk.

Since in total, these types of discounts are not in the interest of law students I would propose that the ABA ban them. The only scholarships allowed will be real money. The stated tuition gets paid.

What would be the impact of this? First, tuition goes down. But how much? I'm sure law schools will figure out a revenue maximizing point soon enough.

Second, law applications probably drop somewhat. The lower price will help but the "free riders" will find something else to do with their life and their high LSAT/GPA combo.

Finally, the lower ranked law schools are going to have some real problems as it it likely that the "full pay" cohort at these schools might not be large or frankly bright enough to sustain the school.


jess : "If a student has a 20 percent or 70 percent chance of retaining the scholarship in year 2, they can and generally do take that into account in calculating the total cost of attendance."

First, since an example of just how hard to get that information has already been mentioned in the last post on this topic, your basic premise is incorrect.

Second, since by now it's clear that 0L's are far more ignorant than informed, your basic premise - cited without evidence, is 100% wrong.


Professor Simkovic and others have made the claim that law schools are reporting employment rates the same way that the government does. They are incorrect, in that the government reports several unemployment rates, and gives definitions and breakdowns of them.

In addition, the federal government gives other background information, including sampling methods, response rates, and corrections/adjustments/weighting.

As listed in my previous comment, the federal government also monitors ***applicable*** employment rates. For vocational schools, 'applicable' means employment in the field.

Law schools, of course are for the most part vocational. They are not general education programs.


"When we are pondering whether the students were given adequate information about scholarship retention, we might head over to Contract Law Prof Blog to read the recent comment by a Valpo grad. The post is May 4th and the comment is May 8th."

Posted by: John Steele

And as usual, the defense finally retreats to 'we didn't break the law'. 'Acting ethically' has long since been abandoned.


Kyle: "The legal profession depends on trust..."

Note that by now it's clear that the legal education industry depended (and depends) on misplaced trust, trust that this industry is not more accurately place with the used car sales industry.

The comments to this entry are closed.


  • StatCounter
Blog powered by Typepad