The critics do not seem to realize that it is expensive to create an effective modern law school. The actual cost of doing it right is vastly underestimated. At HYS [Harvard, Yale, Stanford] for example sticker tuition is now north of 50K per year but that is, as far as I can tell from publicly available information, about one third of the actual cost spent per student each year. Other lower ranked schools have to try to get the job done with far less, of course, and most are effective in doing so. But it is no surprise, is it, that the schools with the most resources continue to dominate in the rankings?
Diamond has a Ph.D. in political science, but that experience doesn’t seem to have given him much in the way of empirical inclinations: There are many plausible and even screamingly obvious reasons why the richest law schools are the highest-ranked law schools, but “because they provide the most effective legal education relative to the original abilities of their students” would not be anywhere on that list, given that there isn’t a shred of evidence for the proposition that elite schools enhance human capital, as the economists say, more effectively than less gloriously endowed institutions.
Speaking of empiricism and endowments, let us turn to some actual law school budgets, past and present, and ask in a longitudinally-inclined way how much the cost of legal education has risen in recent years, and why. In what follows, all dollar amounts have been converted to constant, 2012 dollars.
School A is a hyper-elite law school at the very top of the legal academic hierarchy.
School B is a strong regional law school, which has always been ranked in the 30s and 40s since the advent of the pestilential USN rankings 25 years ago. (There are currently 202 ABA-accredited law schools).
The figures below are derived from the operating budgets for fiscal years 1996 and 2012 for School B. For School A, the figures are derived from its FY2012 operating budget, and from a reconstruction of its FY1996 budget, based on its tuition revenues and endowment income in that year, which I assume together made up the same percentage of its expenditures in FY1996 as they did in FY2012. (School A’s endowment tripled in real terms between FY1996 and FY2012. In 2012 School B’s total endowment was equal to 8% of School A’s. The schools are roughly the same size in terms of total student enrollment. Per student expenditures are calculated on the basis of all law students, both JD and non-JD.)
Total expenditures per student in FY2012: $101,902
Total expenditures per student in FY1996, in 2012 dollars: $49,750
Total expenditures per student in FY2012: $51,472
Total expenditures per student in FY1996, in 2012 dollars: $25,544
Basically, operating costs per student doubled in real terms at both schools over this 16-year stretch, so that Strong Regional was costing as much to run per student in 2012 as Hyper Elite had cost in 1996. During this time, Hyper-Elite’s operating costs per student have risen from 50% more than tuition to twice as much as tuition. Meanwhile, at School B, due to massive tuition hikes, the ratio between tuition and operating costs actually declined significantly between 1996 and 2012.
School B’s outputs, as measured by total legal employment percentage for new graduates, median salary for new graduates, bar passage rates, and law school ranking, did not improve between 1996 and 2012, either in absolute terms, or relative to other law schools. School A’s outputs did not improve between 1996 and 2012 by three of these four metrics, in either absolute or relative terms: median graduate salaries did improve in absolute (but not relative) terms.
What accounts for this extraordinary increase in operating costs at both schools?
This question can be answered with some precision for School B: by far the biggest factor has been the increase in personnel costs, and this in turn has been mostly a product of employing far more people.
For example, at School B, per capita tenure track faculty compensation increased by slightly less than 20% between FY1996 and FY2012, while non-tenure track faculty compensation increased by 16% (This means that if everything else had been held constant, operating expenses at School B would have been less than 10% higher in FY2012 than FY1996). But while per capita compensation increased relatively modestly, the number of faculty skyrocketed: the tenure track faculty was 45% larger in 2012, while the non-tenure track faculty increased by 64%. These increases were dwarfed, however, by the increase in non-teaching administrative staff, which nearly tripled. (For example, the office of career services grew from one full-time and one part-time employee, to six full-time people. By 2012 School B employed five people whose full-time job was various types of fund-raising; in 1996, one person had been engaged in this work).
While the same granular longitudinal comparison can’t be done for School A, School A did more than double the size of its teaching faculty between 1996 and 2012, so it seems likely that much of the doubling of operating costs at the school was driven by the same personnel factors that caused costs to double at School B. (Capital improvements also played a significant role at both schools, i.e., the so-called “amenities arms race.”).
What sort of improvements, if any, in regard to providing an “effective” legal education did this spending orgy produce? The answer of course is that we don’t know. As I noted above, increased spending at each school has seemed to have no effect on measurable outputs. Note too that a linear relation between increased spending and increased pedagogical effectiveness would mean that the legal education provided by School A in 2012 was four times as “effective” (whatever that is supposed to mean) as that produced by School B in 1996.
In sum, how plausible is it that these increased costs have been even minimally defensible, in terms of any conceivable cost-benefit analysis, from the perspective of the graduates of these law schools? To ask that question is to answer it.