Historical Data: Total Number Of Law Schools And Students, 1964-2012

This information is available online directly from the ABA, under the heading legal education statistics (no paywall): http://www.americanbar.org/groups/legal_education/resources/statistics.html

Click on Enrollment and Degrees. The ABA also posts data on scholarships, tuition, debt, etc.

Thanks for this data, Dan and the link, Brian. Lots of interesting trends in those data — in addition to the growth in women attending law school (which are available at the link Brian provided). Also, the dramatic increase in students from about 1967 to 1971. I'm guessing the war had something to do with that, but the enrollments continued high after that. You could write a lot about American history through the history of law schools.

I'm also interested in attrition. It used to be a lot higher than it is now. Looking for a moment at the JDs awarded in the early 1960s and comparing it to the first year enrollment three years before, it seems like only around 60% of the people who were enrolled as 1Ls graduated three years later. It looks like the graduation rate is now closer to 86% (comparing the 1L enrollment in 2009 with graduations in 2012). Do we have a good sense of why students were leaving in greater numbers than they are now? Did people leave for academic reasons? Because of job opportunities without the JD? Something else?

One other question here — there's a dramatic increase in the number of ABA approved law schools from 1968 to 1975. Does this reflect schools opening or schools obtaining ABA approval? Partly both?

"Do we have a good sense of why students were leaving in greater numbers than they are now"

Yes – debt, or the sunk cost fallacy. As the majority of law students found themselves heavily indebted by the end of their 1st and 2nd year of law school, most felt that they had no option but to stay the course when before, with lower debt levels they might have dropped out.

I assume that schools were much more willing to flunk students out in past decades (the whole, "Look to your left, look to your right, one of you will not be here at graduation" phenomenon…). With grade inflation, and law schools' need to maintain enrollment for financial purposes, that phenomenon has largely gone by the wayside.

This is a delicate topic so I will post anonymously. Another factor reducing the attrition rate is the increased diversity of the students. About half of certain URM groups are in the bottom 10% of their class. Flunking out students based upon grades would lead to intolerable results.

Hi MacK, I'm skeptical of your explanation that the difference in attrition relates to debt load. The graduation rate three years later for those entering in fall 1979 was about 87% (more or less where it is now). And don't we look back on the late 1970s/early 1980s as an era of low tuition and debt?

Intolerable for whom? The public who are served by these officers of the court?

Would be interesting (and informative) to see these numbers along with cost of tuition as compared to median income. Did cost of tuition rise proportionally to income; did cost rise because demand increased? Did the "cut" of the university change over time? There are so many factors-along with debt load-that are meaningful.

I'd be interested to see this in relation to both population in general and number of people w/ an undergrad degree (and so able to go to law school, at least in principle.) I don't suppose that would make the numbers look better for now (the population is larger than 2000-01, and I assume the population w/ a BA is also larger) but I think it would make some of the comparisons more useful.

This discussion only considers half of the question — the supply side. We don't really care how many new JDs are created each year, so much as whether the market can absorb them (i.e., whether supply and demand are in equilibrium).

So, I pulled up the BLS CPS series for the legal sector. (For those interested, I used series ids LNU02032459, LNU03034025, LNU04034025, LNU02038291, and LNU02038292.) These series give us the total number of folks employed and unemployed in the legal sector since 2000. That includes most JD-required jobs (lawyers, judges, law clerks), but also includes law-related non-JD jobs (paralegals, legal services); numbers for these latter jobs are only available since 2010.

The bottom line is that the legal sector has been adding an average of about 32,000 jobs per year since 2000. Based upon the data since 2010, about 65% (about 21,000) of these are JD-required jobs. The industry totals are copied at the bottom of this comment.

The average addition of 21,000 jobs per year is net of new JDs and retired JDs. If we (generously) assume the average lawyer's career is 40 years long, 2.5% of the industry retires per year — or about 22,000 in 2000 and 29,000 in 2011.

This suggests that the market is absorbing about 50,000 new JDs per year today. Based on Dan's data, this means that we're actually producing near the market equilibrium for new JDs per year. (This is _not_ the conclusion that I expected to draw when I started this analysis.) The JD retirement rate is, of course, very elastic — it is likely that during the 2007-2010 downturn folks were leaving their jobs at decreased rates (older JDs weren't retiring because the market hurt their savings; younger JDs weren't leaving their firms because the economy-wide unemployment rate was terrible).

Indeed, with JD enrollment falling well below 50,000 per year, and a possible glut of lawyers who delayed their retirements during the downturn, the new supply of JDs could actually be falling below demand! (Again, not at all what I'd expect to discover in these numbers — and I am very open to being proven wrong.)

Of course, these numbers are all a bit fuzzy and built on a number of assumptions. But it does appear that supply and demand are surprisingly close to equilibrium. Also, these numbers don't tell us much (if anything) about tuition questions. It's entirely possible that average salaries are going down, even as the number of jobs continues to increase.

Here's the summary BLS data. "Industry size" is the total # of folks employed (lawyers, judges, clerks, paralegals, legal services); "hiring year" is Aug-July; numbers are in thousands. I've included the unemployed numbers for reference/interest. Sorry about the weird formatting; I couldn't get tabs to work properly in the comment.

Hiring Year_____Industry size_______Net adds_______# unemployed______Unemployment rate
2000____________1430.85_________-11.03___________19.77___________1.35
2001____________1478.54_________47.69___________34.31___________2.28
2002____________1495.15_________16.62___________34.77___________2.25
2003____________1556.23_________61.08___________33.08___________2.08
2004____________1585.00_________28.77___________28.54___________1.76
2005____________1604.54_________19.54___________23.69___________1.45
2006____________1663.46_________58.92___________27.54___________1.64
2007____________1686.62_________23.15___________43.62___________2.51
2008____________1691.46__________4.85___________58.62___________3.34
2009____________1715.23_________23.77___________52.31___________2.97
2010____________1747.77_________32.54___________60.85___________3.35
2011____________1780.92_________33.15___________55.77___________3.03

I'm posting this anonymously, being a very new, very junior professor. But if anyone would like to play around with this data, post here and I'll mail you.

But that's not what ITLSS tells me to believe…..

JuniorProf,

If supply of JD's is below demand, as you suggest, why does NALP report that only 56.7 percent of 2011 graduates obtained full time jobs as lawyers? NALP gets its employment information from law schools, which are not understating their results.

Really Junior:

Thanks for digging up these numbers, but there's something that seems to be missing. You describe an estimated 29,000 retired JDs in 2011. Let's assume this is correct. For your rough equilibrium conclusion to be valid, it would have to be true that something like 29,000 retired JDs are being replaced. But, what if – due to shifts in the packaging and delivery of "legal services" rather than "lawyering", those retired JDs aren't being replaced in anything like a 1:1 ratio? Moreover, to simplify grossly, if illustratively, suppose many retired JDs are law firm partners. Replacing them with new associates isn't immediately possible (you can't make a partner out of new law graduate), and firms seem inclined to get rid of less- productive (older?) partners anyway, which may keep per-partner profits up. Doesn't this crude model better describe the current picture? After all, if that replacement was really happening, say on an assembly-line basis eventuating in partnership, we wouldn't have so many unemployed new JDs.

Btw, another way to read your data is that new JDs could be taking paralegal jobs, which is to say that firms are replacing retired JDs with presumably cheaper service providers. But, this wouldn't be an equilibrium either (because new JDs didn't need law degrees for those jobs) and I don't see any evidence that JDs are even getting these jobs in large numbers. Am I missing something?

Brian,

It's a very good question — I was very surprised by the numbers. We need to look at the supply and demand side of the equation. Right or wrong, these numbers hopefully make that clear.

In terms of explanation, I have two ideas. First, current-year JDs are competing with prior-year JDs. It's well known, and the numbers show, that in the 2007-2009 period firms weren't hiring much at all. This created a glut of JDs. Many have left the market; others have continue to compete for jobs. And, second, the attrition trends that affect the broader economy affect the legal economy as well. Junior lawyers are less likely to risk leaving their jobs due and many senior lawyers have delayed their retirement. Both of these push down demand, but only in the short run. (The retirement effect could marginally decrease demand in the long-run.)

NotSoNewProf,

I agree with both of these points. But I think the first one is answered by the data directly: the legal sector is adding jobs in absolute numbers. The second point is very well taken, and I'm frustrated that BLS only started separating out paralegal and legal services jobs in 2010. We can't tell from this data if there's a shift from JD-required to JD-preferred (or even non-JD) legal sector jobs; and we can't tell if there's a shift in salary, e.g., from full-time associate jobs to part-time contract jobs.

As mentioned in my reply to Brian, the key point is that we need to look at both the supply and demand sides of the equation. It's meaningless to say "50,000 new JDs/yr is too many" unless we can say "because the legal market is only absorbing 40,000 per year."

Will Brian Tamanaha join me in decrying the fact that academics must post their research anonymously to avoid retribution from the critics of the legal academy?

JuniorProf,

This passage is from BLS, "Lawyer Outlook:"

"Competition should continue to be strong because more students are graduating from law school each year than there are jobs available. As in the past, some recent law school graduates who have been unable to find permanent positions are turning to the growing number of temporary staffing firms that place attorneys in short-term jobs. This service allows companies to hire lawyers “as-needed” and permits beginning lawyers to develop practical skills."

Matt Leichter (Law School Tuition Bubble) indicates that BLS has repeated this passage for a decade. Data from NALP indicates that even before the 2008 crash, going back to at least 2001, about 30 percent or so of grads were not landing full time jobs as lawyers. Your findings do not match these indicators.

Perhaps you should email Leichter with your findings (see LSTB). He has lots of information from BLS and should be able to figure out what explains the disparity.

Steve,

I know junior profs who post anonymously, not because they fear retribution from critics, but because they fear their senior colleagues who don't like what they say will deny them tenure. They are prudent to remain anonymous.

Tenured law professors who have something to say, on the other hand, should put their name on it–demonstrating the courage of their conviction. The ire of critics is no excuse for a professor who has iron clad job protection.

Thanks Brian. I'll look into that (to the extent I can — I'm playing with these numbers during writing breaks.) That said, if the 50,000/yr steady-state absorption estimate is correct, given than law schools only began hitting that number of new JDs/yr as the 2007-2009 recession began, BLS's statement is surely correct.

Returning to NotSoJunior's early points, I've been looking at the LE data series for earnings data, but the data is too coarse to tell if there has been any effect. There does, however, appear to be about a 10% increase in the ratio of part-time to full-time lawyers between 2000 and 2012, and a similar increase in the ratio of paralegals to full-time lawyers. Additionally, the year-over-year growth in part time positions is slightly higher than that of full-time positions (about 2.6% vs. 1.9%); the year-over-year growth in paralegals has averaged 3.5%; it was about 2% from 2000-2007, and has shot up to about 5.5% since 2008.

All of this does tend to corroborate the intuitions that JDs are shifting from full-time positions to part-time and paralegal positions. That said, the BLS data is coarse; these data, at least to the cursory extent of my analysis, corroborate, but don't confirm, these intuitions.

Reallyjuniorprawf,

The BLS 22,000 prediction I understand to be net new hiring and rolls in the retirement of lawyers over the next decade. Moreover, you need to remember that the current profession 1 skews young, that is to say the massive growth in the last 20 years means that a lot of lawyers are only in their forties – and a lot of lawyers will retire later because of poor pension provision and the ability to work for longer

It may be of interest that the number of students at six schools called "elite" by people other than me (I went to NYU, which is not one the six) has not changed since 1970. See http://www.law.harvard.edu/programs/plp/pages/statistics.php. The share of all JDs at those schools shrank from 9% to 4% from 1970 to 2011.

On the fundamentals of demand, meanwhile, the population grew from 200 to 300 million, roughly, and GDP from $1 to $15 trillion. Litigation grew significantly from 1970 to 1990 in both state and federal courts, including business litigation,, but federal court cases at least has been flat since 2000. M&A activity has gone through cycles but continues to much higher today than in 1970.

But for disruptive technological innovation, which has increased productivity of legal service providers and substitution to non-JDs, demand would have outstripped the supply of JDs. Because no one can confidently estimate how far those IT-driven trends can go, I am not surprised people have very different intuitions on what the average future value of the JD will be. Being lawyers, however, commentators don't let that stop them from making strong claims in opposite directions.

reallyjuniorprof there is a very simple flaw in the analysis – which was incidentally seems to have been made to achieve a particular objective – showing demand for new law graduates to be at least 50,000 per annum – the current capacity of US law schools. If your analysis was correct there would not have been shortfalls in hiring that predate the recession. But indeed hiring has been poor for over a decade. Employment at 9 months has been according to NALP's very optimistic dataset:

2001 68.3 percent;
2002 67 percent;
2003 65.5 percent;
2004 65.1 percent;
2005 66.7 percent;
2006 68.3 percent;
2007 70.7 percent;
2008 67.2 percent;
2009 62.5 percent.

You also need to know a key fact about BLS unemployment data – it bases the occupation of the unemployed person on what their last job was. If they were driving a cab, a barista, waiting tables, they are not counted as an unemployed lawyer. There are about 1.5-1.6 million million extant JDs in the US workforce, but BLS reports only 570,000 lawyers. Where are the missing million?

Veryjuniorprof, it's excellent that you're examining the numbers, but I think there are some errors in your calculation. Here are some suggestions:

1. In looking at an extensive time period, I would take numbers from the beginning and ending months–not the average annual figures (as it seems you did). In January 2000, there were 1,470,000 people working in "legal occupations." In January 2013, there were 1,704,000, an increase of 234,000. Dividing by thirteen years, we get an average increase of 18,000 new positions in the legal industry, not the 32,000 you originally posted. (The 32,000, I think, may itself have been a calculation error. If you take the annual numbers and do the math, you get an average of 29,167.)

2. So now we have to figure out how many of these positions were for lawyers rather than paralegals and other support workers. As you acknowledge, there's some danger in extrapolating backward but, if we take the 65% as the best approximation, that gets us to an average of 11,700 new positions a year.

3. Even this is a little high because "legal occupations" also include hearing officers, mediators, and conciliators. Most of those positions do not require either a JD or bar admission–although some law grads fill them as JD advantage positions. To get a better number, we would have to exclude those categories as well as the paralegals and "other support." But I haven't tried to do that in this calculation.

4. The replacement rates you assume are too high. We can't assume that everyone who leaves a position is replaced. BLS has more sophisticated formulas for calculating replacement.

BLS actually does the full calculation, incorporating still more variables, through its 10 year projections. http://www.bls.gov/emp/ep_table_102.htm. Those show that the economy will support about 228,400 new positions for lawyers, judicial law clerks, or judges/magistrates over the next ten years–or 22,840 per year. That figure includes all retirements. There is other supporting documentation on how BLS reaches these projections. See http://www.bls.gov/opub/hom/pdf/homch13.pdf.

Keep looking at the data–it's very important for law professors and others in the legal industry to do that.

Thanks Deborah — I had not seen that BLS analysis. I readily concede that theirs is more reliable than mine. And I am happy that folks seem to understand my underlying point, viz., that we need to consider both sides of the equation, not just the supply side, if we are to meaningfully discuss rightsizing the academy's output of JDs.

One other point on the BLS data:

The BLS distinguishes a number of occupations (e.g. compliance) that employ JDs.

This is part of why there is a lower number of legal jobs in BLS stats than there is in, e.g., ABA data.

And one other driver of demand: WIPO data show patent applications rose by 500% from 1980 (the first year they give data) to 2011.