[UPDATE: Since this posting, there have been a number of blog posts, replies, etc., on this topic at Balkinization, Leiter Law School Reports, and elsewhere. Those are the best places to get into the analysis. Fwiw, I'm not convinced that the 16-year era in the study is indicative of how the next 16 years will turn out. And as I understand it, the study is not designed to assist any particular law school applicant decide if any particular JD is worth the costs.]
The new study by Michael Simkovic and Frank McIntyre, which looks at the economic returns of a JD, "measured differences in annual earnings and hourly wages between those with law degrees and similar individuals who end their education with a bachelor’s degree." Subsequently, Stephen Diamond used the study to determine net present value of a JD, which he suggested was $330,000 at the start of law school. As I mentioned previously, the study will probably re-set the framework for discussing the economics of getting a JD. [Please read the UPDATE below, which discusses Simkovic's response to the central issue I discuss in this post.]
In a prior post, I wondered why the authors used as an historical baseline the extraordinary 16 year period from 1996 to 2011 when the famous bi-modal distribution of first year salaries was born, swelled, and then just recently thinned out. That post has graphs illustrating the history of that anamolous distribution. (Deborah J. Merritt asked the same question.) [I've also added a graph below, showing total law office jobs peaking well before the 2010 drop off.] Why would that 1996-2011 era with an unsustainable bubble at the high-end of the market make for a good baseline unless we believed it would be indicative of the future market? Would it still be a good baseline if we believed that a good chunk of the wealthy organizational clients have found ways to avoid paying high legal fees? [One of the principal ways was to analyze the functional tasks in any legal relationship and send each tasks to the lowest cost provider -- thus significantly bypassing the more expensive private practice firms. I had first noticed this "disaggregation" of legal services around 2005-06 and was discussing it inside law firms and with others in that era. That was before the economy turned in 2008 and the legal services market tanked a short time later.] Or if we believed that at the other end of the market, non-lawyers are finding ways to satisfy consumer demand for basic legal services? Simkovic responded with a comment about the difference between nominal vs real wages. [As noted below, Simkovic's respones on this point is that 1996-2011 happens to be the era with the best data and that he welcomes additional studies that include additional data from other years.]
I had also noted that the recent grads (e.g., 2008-11) presumably fell well below the Simkovic baseline and aren't likely to catch up. I realize that the model is built for the long haul and it would not necessarily be fatal to the model if any 2-3 years of grads was "off the model." But I've since seen the claim, at Brian Tamanaha's post discussed below, that the paper did not include the earnings of the graduating classes 2008-11 at all -- earnings which have been famously poor by historical standards and presumably would have led to a lower expected value for JDs across the chosen baseline. As I understand it, Simkovic will address those points in a future post.
Brian Tamanaha has now posted a long discussion of the Simkovic/McIntyre paper at Balkinization, making three critiques. First, Tamanaha argues that the choice of the 1996-2011 baseline and the exclusion of the 2008-11 grads distort the results. He argues that if the study had included four more years of available data (which might have been data generated in a different way), the model would have yielded much lower results. I assume Simkovic will respond to that. [UPDATE 7/24: Simkovic has responded by saying that the best data comes from that 16-year era and that using other, less reliable data would lead to questionable results. But even assuming that's true, that response doesn't directly tell us whether or not the best historical data are too rosy for predicting the future or for making decisions today. That fundamental question remains, doesn't it?]
Second, Tamanaha argues that the model does a type of misattribution, in that the undergrads who go to law school are from higher socio-economic strata, are more ambitious, get better grades, etc., and that those attributes would have a significant impact on earnings even if those undergrads hadn't gone to law school. (I wonder how that observation may play out over the next few years. Law schools admit people with lower credentials these days and recent studies have suggested that applicants with the highest LSATs have shown the largest decrease in enrollment. And we still don't know if the bar exam committees will lower their standards 2-4 years from now. If they maintain historical standards on the MBE, we could have unexpectedly high rates of bar exam failures and hence lower income results.) Tamanaha also suggests that the study does an apples-to-orange comparison when it compares the bottom quartile of law school grads to the bottom quartile of undergrads. I can't say either way if Tamanaha is right on those points. [UPDATE: Simkovic continues to respond to his critics at Leiter Law School Reports.]
Third, Tamanaha suggests that the study is affected by over-optimism.
No doubt, Simkovic will respond to those points. It's been an interesting debate and I've already learned quite a bit. Personally, I find the most useful metric to be the calculations of the net present value of the JD using the model Simkovic offers, which as I undersand it is something that Stephen Diamond added to the discussion. (To be clear, my intuition makes me doubt whether JDs from low-ranked schools have a NPV of a third of a million dollars, but I'm also a big believer that good data trumps intuition.) If we do begin to justify the cost of a law degree this way, we'd still need to discuss the distributional issues and look at who gets to be a lawyer.
[edited substantially since posting]