Years ago, as I began to look at law student indebtedness metrics, I noted a number of reasons why debt figures should be construed with caveats.

Students with low debt loads could be independently wealthy or come from a wealthy family willing to finance the education. They could have substantial scholarship assistance. They could earn income during school or during the summers. They could live in a low cost-of-living area, or live frugally. They could have lower debt loads because of some combination of these and other factors.

Scholarship awards have, in recent years, appeared to be outpacing tuition hikes—which has been a several-year trend and places schools in increasingly precarious financial positions. Students are no longer purchasing health care due to the ability to remain on their parents' health insurance under federal law, a significant cost for students a few years ago. Schools have increasingly eased, or abolished, stipulations on scholarships, which means students graduate with less debt. Some schools have slashed tuition prices. We might simply be experiencing the decline of economically poorer law students, resulting in more students who need smaller student loans—or none at all. Students may be taking advantage of accelerated programs that allow them to graduate faster with less debt (but there are few such programs). As JD class sizes shrink, it's increasingly apparent that students who would have paid the "sticker" price as increasingly pursuing options at institutions that offer them tuition discounts.

These debt figures are only an average; they do not include undergraduate debt, credit card debt, or interest accrued on law school loans while in school. The average may be artificially high if a few students took out extremely high debt loads that distorted the average, or artificially low if a few students took out nominal debt loads that distorted the average.

Some borrowers will be eligible for Public Service Loan Forgiveness programs. That might make their debt loans appear high when they will ultimately have those loans paid off. And some borrowers will take out a high amount of debt only to earn very high salaries upon graduation.

In short? There are a lot of limitations to debt metrics.

Of course, I think, on the whole, a lower percentage of students taking out debt, the better; and a smaller debt load is better. Those are obvious points. But as USNWR includes two new indebtedness metrics worth 5% of the overall rankings, it’s worth considering the justification.

One reason is to say that “many new lawyers are postponing major life decisions like marriage, having children and buying houses – or rejecting them outright – because they are carrying heavy student loan debts.” True enough. But it’s also a reason, I think, to look at the debt to income ratios of graduates. That is, not all debts look the same—they are much less onerous when income levels are higher. Even this is a limited look, as cost of living matters dramatically, too. But there are alternative concerns, too—students may choose jobs they don’t really want or stick in loan forgiveness programs they don’t want to pay down debt. All important caveats.

Another is a racial disparity concern: “J.D. graduate debt is impacting Black and Hispanic students the most since they borrow more, according to the ABA.” That’s understating it: White students had average indebtedness of about $101,000 in 2018, compared to $150,000 for Hispanic students and $199,000 for Black students.

But more the point, why does USNWR think this metric will improve these disparities? The incentives for law schools are currently to maximize the median LSAT and UGPA scores, which focuses on “bounty” scholarships for high-performing LSAT scorers, which tend to be, at a given institution, white students. One way to both reduce debt levels and improve admissions metrics—both now valued by USNWR—is to increase these disparities. Independently wealthy students are also more attractive to a law school, those who take out zero dollars in loans—and, again, socioeconomic status interacts with racial characteristics, which is likely to increase these disparities. It’s strange, then, to cite race as a basis for including a metric, but to include the metric in such a way as to offer opportunities to exacerbate the very concerns raised.

Furthermore, USNWR already incentivizes schools for being expensive (and fails to disclose that data publicly). This means USNWR incentivizes a very expensive school with low reliance on tuition—pressing schools toward reliance on grants, endowments, or central administration support.

And Goodhart’s Law may come to debt metrics. Again, with my caveats above, I think fewer students with debt and smaller debt loads are, of course, on the whole, a good thing. It might be that schools really start to focus on financial health of students and provide greater counseling to students in law school. It might be that schools will take seriously these figures and do their best to reduce them for all students, not simply in ways that manipulate metrics at the margins. But with any newly-introduced metric, it’s not clear how it will play out.

I’ll have another post on the indebtedness metrics and how they are skewed to favor schools based on percentage of students who incurred debt instead of average debt—and why that’s the wrong approach.