The important part of Amir's post, besides focusing attention on a serious mismatch between (nondischargeable) student debt and salaries of recent graduates, is that it gives credence to Andy Morriss's idea about making this information more specific and thus more useful:
Some law-school academics are calling for the distribution of more-accurate employment information. “Prospective students need solid comparative data on employment outcomes, [but] very few law schools provide such data,” says Andrew Morriss, an Illinois Law professor who has studied the market for new lawyers.
In response, the ABA (and even USNWR) could ask law schools to post information along the following lines: median graduate starting salaries, standard deviation of starting salaries, or (for folks who prefer their info in graphics) a pie chart showing how many of the most-recent class of graduates earn $0-30,000/year, $31,000-60,000/year, $61,000-90,000/year, and $91,000+/year. A really useful comparison might include another pie chart showing how many students graduate with law school debt in the $0-30,000 range, the $31,000-60,000 range, the $61,000-90,000 range, the $91,000-120,000 range, and the $121,000+ range. An even more useful pie chart could show the relationship of law school GPA to starting salary. That last chart would sober everyone up.
Of course, such information is likely to hurt private schools disproportionately, as their tuition tends to be higher than the tuition of state schools, and there should be some way for schools to be able to demonstrate that the students taking low-paying (e.g., public service) jobs can make their debt service, with such help from schools as scholarships and loan forgiveness programs.
What is even more likely is that law school applicants will start doing some arithmetic. That in turn will lead to applications decreasing at schools at which an applicant's chance of graduating in the top of the class (thereby having access to more job options) is small and the hiring patterns below, say, the top 20% of the class predict small or solo practice for the other 80%. Decreasing applications will lead to decreasing median undergraduate GPAs and LSAT scores, which will cause those schools' USNWR rankings to drop. The vicious circle will continue and, at some point, some law schools may fail to survive.
I have no easy solutions, and in fact, I don't believe that the U.S. actually needs as many law schools--at least as currently envisioned--as it now has. If law schools actually differed dramatically from each other, then having 200+ ABA-accredited law schools makes sense. But the ABA standards tend to force schools into a fairly tightly bunched pattern, which USNWR then attempts to spread out in some ordinal way. Right now, the schools that USNWR ranks are (except for the truly exceptional schools at the very top) more alike than they are different.
Where does that leave graduates who can't pay their rent and service their educational loans? Scrambling for income and cursing the misleading information that they read when they were deciding among law schools.