‘Human Capital’ Theory Doesn’t Explain Law Grad Earnings

…Is on The American Lawyer.

You can take a wild guess as to its topic.

CLASS OF 2014 EMPLOYMENT REPORT (Updated)

The ABA released the class of 2014’s employment information last week, and there are enough differences to warrant an update to this post, namely that I’d forgotten that the ABA accredited Lincoln Memorial University late last fall. Most of the figures remain the same, but to keep Web traffic to one post, I’m retaining the original “leaked” version at the bottom but striking it out. I’ve also changed the title.

We have 43,195 people who graduated from an ABA-accredited law school outside of Puerto Rico between September 1, 2013, and August 31, 2014. The employment information is good as of March 15, 2015. Readers likely know that the ABA now collects data as of ten months from the typical graduation date rather than nine. I don’t think there’s too much of an impact by the change except to applicants this year who might have wanted to rely on the data.

The tables are below the fold to conserve blog space.

Continue reading

Michael Simkovic Has a Big Fan on Wikipedia

Wikipedia readers interested in higher education, student loans, and related topics in the United States might be interested to know why so many Wikipedia pages rely so heavily on law professor Michael Simkovic’s paper “Risk-Based Student Loans.” The same goes for those interested in the future of legal education, given Simkovic’s prominence in that field.

A few examples:

  • The page on the “National Defense Education Act” cites “Risk-Based Student Loans” as evidence that the act “followed a growing national sense that U.S. scientists were falling behind scientists in the Soviet Union, catalyzed, arguably, by early Soviet success in the Space Race, notably the launch of the first-ever satellite, Sputnik, the previous year.”
  • The first paragraph of “Bankruptcy Discharge” ends with the statement, “Discharge is also believed to play an important role in credit markets by encouraging lenders, who may be more sophisticated and have better information than debtors, to monitor debtors and limit risk-taking,” and then cites “Risk-Based Student Loans.”
  • The “Higher Education Act of 1965” ends with a citation to Simkovic’s work as well. “The HEA has been criticized for establishing statutory pricing of federal student loans based on political considerations rather than pricing based on risk.”
  • In “Human Capital,” “Risk-Based Student Loans” tells us, “According to signaling theory, education does not lead to increased human capital, but rather acts as a mechanism by which workers with superior innate abilities can signal those abilities to prospective employers and so gain above average wages.”
  • Student Loans in the United States” contains six references to “Risk-Based Student Loans.” Here’s one:

Both subsidized and unsubsidized loans are guaranteed by the U.S. Department of Education either directly or through guaranty agencies. Nearly all students are eligible to receive federal loans (regardless of credit score or other financial issues). Federal student loans are not priced according to any individualized measure of risk, nor are loan limits determined based on risk. Rather, pricing and loan limits are politically determined by Congress. Undergraduates typically receive lower interest rates, but graduate students typically can borrow more. This lack of risk-based pricing has been criticized by scholars as contributing to inefficiency in higher education. [Emphasis added.]

A variation on the higher education bubble theory suggests that there is no general bubble in higher education—that is, on average, higher education really does boost income and employment by more than enough to make it a good investment—but that degrees in some specific fields may be overvalued because they do little to boost income or improve job prospects, while degrees in other fields may in fact be undervalued because students do not appreciate the extent to which these degrees could benefit their employment prospects and future income. Proponents of this theory have noted that schools charge equal prices for tuition regardless of what students study, the interest rate on federal student loans is not adjusted according to risk, and there is evidence that undergraduate students in their first 3 years of college are not very good at predicting future wages by major. [Underlined & italicized emphasis added.]

As of Sunday, April 19, 2015, I’ve found at least 13 pages on Wikipedia that cite “Risk-Based Student Loans.” I don’t intend to find them all.

I discovered the edits after one of them popped out to me. Readers will probably agree that many of the citations appear to be either (a) statements for which “Risk-Based Student Loans” at best serves as a tertiary source, or (b) interpolations into the articles that promote “Risk-Based Student Loans” more than inform the reader on the topic. After sleuthing one of these page’s history sections, and then using Wikipedia’s cruelly titled “WikiBlame” feature to establish who altered these pages, I traced them to two anonymous IP addresses: 108.46.277.251 (May 5, 2012, through May 10, 2012, by my interpretation) and 71.190.187.89 (May 15, 2012, through June 11, 2012). Both are located in Midtown Manhattan, and the links include lists of the altered pages. Most of the edits occurred by May 15.

The current pages have benefited from nearly three years of improvement (although it’s troubling that so many of the pages’ headers carry content alerts given the topics’ importance), but the initial edits evidenced surprising devotion to Simkovic’s work, especially since he wasn’t so prominent then. For instance “Student Loan” received as many as twenty-one citations to “Risk-Based Student Loans,” and twenty were added to Student Loans in the United States. Sometimes it appears as the very first authority listed.

Here’s “Student Loan” early on May 5, 2012:

Click to enlarge. It's unclear why the citation date is 2013.

Click to enlarge. It’s unclear why the citation date is 2013.

Click to enlarge.

Click to enlarge.

The intensity of the edits quickly caught the eyes of some Wikipedia editors. On the evening of May 5, 2012, editor ElKevbo contacted whom I’ll call “Simkovic’s Fan” to inform him (yes, he’s a dude) that using an unreviewed, “self-published source” ran afoul of Wikipedia policies. At the time, “Risk-Based Student Loans” had not been published in an academic journal. Simkovic’s Fan defended his namesake, claiming Simkovic was “an established bankruptcy and credit scholar” with numerous prominent publications to his name. ElKevbo asked to continue the discussion on the Talk page for “Student Loans in the United States,” and added, “I object to widespread use of an unpublished work that has not undergone any sort of peer review.”

There, the dialogue resumed with ElKevbo reiterating that if “Risk-Based Student Loans” stood for the propositions for which Simkovic’s Fan claimed, then more appropriate sources would be available. ElKevbo also condemned “the manner in which this unregistered editor is attempting to ram these edits through without consensus.” Editor Nstrauss also criticized Simkovic’s Fan for un-reverting the restored article to replace his numerous citations to “Risk-Based Student Loans.” (Nstrauss refers to Simkovic’s Fan as “the anon Virginia editor.” I believe that’s a reference to the location of Simkovic’s Fan’s Internet service provider, Verizon, which is in Virginia. The IP address was in New York City.)

Undaunted, Simkovic’s Fan continued inserting citations to “Risk-Based Student Loans” in various related articles. No other Wikipedia editors appear to have noticed the widespread changes.

I’ll be upfront, I suspect Simkovic’s Fan is in fact Michael Simkovic, given the location of the IP addresses and the fact that it’s highly implausible that anyone would have read “Risk-Based Student Loans” in mid-2012 and been so convinced by it as to regard it as a comprehensive, Wiki-worthy treatise on higher education, student loans, bankruptcy, etc. I also think only Simkovic would exaggerate his prominence as “an established bankruptcy and credit scholar.” I’m open to alternative candidates.

There are a few reasons I’m raising this discovery.

One, I have no desire to become an expert on Wikipedia policies, but “Risk-Based Student Loans” is probably inappropriate for nearly all the statements it purports to validate. Wikipedia prefers its articles to be based on secondary sources. Going by its title and abstract, “Risk-Based Student Loans” is not, for example, a secondary source on the National Defense Education Act or signaling theory. In these contexts, “Risk-Based Student Loans” is a tertiary source and it should be substituted for better ones. Moreover, as tertiary sources go, it may not be a neutral one.

Two, many of the remaining citations use passively voiced or weasely worded phrases that just advance Simkovic’s arguments in “Risk-Based Student Loans,” as in “Student Loans in the United States” or “Higher Education Bubble.” When properly rephrased these aren’t “scholars'” or “proponents'” consensus opinions; rather, they’re Simkovic’s, and his positions on these issues are probably too specialized to be relevant to the articles’ topics.

The only place “Risk-Based Student Loans” really ought to be cited is in articles on risk-based lending, and arguably Simkovic is not a sufficiently noteworthy scholar for his opinions to appear there—and certainly not as the first citation. Again, it may not pass Wikipedia’s neutrality requirements.

Three, Simkovic’s Fan has disserved Simkovic’s research efforts by concealing the sources “Risk-Based Student Loans” relies on to make its arguments. To the extent that those authors’ points are relevant to the altered articles’ topics, they are the ones who should be cited, not Simkovic. Simkovic’s Fan should not have treated “Risk-Based Student Loans” as a comprehensive secondary source on anything.

Four, it’s more believable than not that Simkovic himself is the one who in 2012 edited Wikipedia to promote his own expertise and scholarship. If so, he certainly overstepped Wikipedia’s conflict of interest policies on self-citation in a shamefully vain and unprofessional manner, particularly for the reasons stated in the previous point and given ElKevbo’s objection that “Risk-Based Student Loans” hadn’t even been published yet. If true, Simkovic should apologize, and Wikipedia’s editors should reevaluate all references to his article.

Five, I’m not knowledgeable about law schools’ or universities’ policies regarding their instructors’ activities on Wikipedia or similar venues, but even if Simkovic’s Fan happens to not be Simkovic, I really hope academics absorb the lesson established here of watching how their work is used in publicly available Web sites like Wikipedia. Hopefully Wikipedia will also do a better job of monitoring wanton edits to its articles and tighten its requirements on secondary and tertiary sources so other academics’ works aren’t misused this way again.

Finally, I acknowledge that this post might be construed as an “off-wiki attack” of Simkovic that Wikipedians may perceive as harmful to their community. I believe, however, that the harm is minimal given that the edits are nearly three years old and Simkovic’s Fan was not a registered user. I also will not edit any of Simkovic’s Fan’s citations myself.

********************

Even though I’m not the topic of this article, I stipulate that I have never and do not intend to alter any Wikipedia pages to promote this blog, my articles on The American Lawyer, or anything else I have written or will write. I also take this time to thank those who have linked to my work in Wikipedia pages. It means a lot to me that my readers consider my work noteworthy enough for citation for the masses.

Law School Diversity Improves at Schools With Worst Job Outcomes

…Is available for your reading on The American Lawyer.

As a side note, irrespective of what you think of Aaron Taylor’s research, please realize that his or those like will not be possible in the future if the ABA Section of Legal Education and Admissions to the Bar’s Data Policy and Collection Committee changes law schools’ entering credentials reporting requirements (pdf). The committee wants to replace matriculants’ 75th, 50th, and 25th percentile LSAT and GPA data with large tables. This change will make the new data incompatible with the years of previous information that was presented in the Official Guide. I’m in favor of backwards compatibility for data, and I sent the committee a comment saying as much, but if the committee decides to make the changes anyway, much will be lost.

GUEST POST—Smokin’ Bucketful of Awesome (Flow Chart Edition)

(Connecticut attorney Samuel Browning, a friend of The Law School Tuition Bubble, obtained permission from law professor Bernie Burk to create a flow chart version of a series of posts Burk wrote on The Faculty Lounge in June 2014 that characterized law school outcomes as between either “A Smokin’ Bucketful of Awesome” or “A Smoking Pile of Scrap.” Mr. Browning’s chart appears here with only minor proofreading on my part, so any unclear points, variances from Burk’s posts, or errors are his own. Actual hyperlinks to Burk’s articles are included at the bottom. Click on the flow chart to enlarge it in your browser.)

Smokin' Bucketful of Awesome (Flow Chart)

********************

http://www.thefacultylounge.org/2014/06/more-thoughts-on-self-delusion-in-the-legal-academy-and-an-effort-to-engage-the-aals.html

http://www.thefacultylounge.org/2014/06/self-delusion-spreads-from-professional-to-graduate-education-consternation-curiously-absent.html

http://www.thefacultylounge.org/2014/06/still-more-thoughts-on-self-delusion-in-the-legal-academy-or-accepting-the-difference-between-a-smok.html

71 Percent of Lawyers Work in the Legal Services Sector

71 percent of employed lawyers, that is. We’re not talking about people who are on the rolls but aren’t working.

I haven’t carefully read through all of Michael Simkovic’s and Frank McIntyre’s most recent analysis in law graduate earnings, but it looks like they’re still uninterested in exploring the possibility that law grads’ earnings are attributable to demand-side factors, like price or income elasticity of demand for lawyers’ services. Because they don’t show us that law students who complete all the required law school course work without graduating have the same earnings as law graduates, anything they say about a JD premium is premature. Such an analysis is crucial because one of their own citations, David Card’s 1999, “The Causal Effect of Education on Earnings,” indicates that law grads earn substantially more than the trend would suggest. This finding screams for testing, but Simkovic and McIntyre aren’t careful enough researchers to do that.

Thus, it follows that their comparisons between law grads and college grads in “Timing Law School” are equally inadmissible. Indeed, I may not bother commenting on “Timing” at length at all.

However, I did decide that a little procrastination is good for the human spirit (and entertaining to the reader), so I poked around “Timing” to see what errors I could find. I’ll showcase one.

On page 17 Simkovic and McIntyre write:

Based on initial outcomes for recent graduates and qualitative factors, Henderson and Zahorsky argue that the legal profession is experiencing a “structural shift” due to globalization and technological change.34 Others point to a decline in the size of “legal services” (law firms) relative to GDP.35 What this means for law school graduates is uncertain, since most legal services workers are not lawyers,36 and many law graduates work in fields other than legal services.37

Footnote 36 uses the Bureau of Labor Statistics’ Occupational Employment Statistics (OES) database to show that “out of more than 1.1 million legal services workers, only 375 thousand were lawyers. Other occupations include paralegals, secretaries, bookkeepers and computer support and business specialists.”

And footnote 37 says:

Around 60 percent of law graduates practice law. Simkovic and McIntyre, supra [“Economic Value of a Law Degree”] at 252. Of those working as lawyers, around 65 percent work in “legal services.” United States Department of Labor, Bureau of Labor Statistics, supra note 36. Some of the non-lawyers working in legal services have law degrees.

In other words, the authors bury in a footnote the fact that 65 percent of lawyers work in legal services, so they can claim that it’s unclear how economic swings affecting the legal services sector would in turn affect law grads because most workers there aren’t lawyers. Being mindful of the distinction between law grads and lawyers, it’s nevertheless pretty bizarre to believe that the one industry law school prepares people for most would have a trivial impact on their earnings. The only alternative interpretation is for Simkovic and McIntyre to show that the legal sector is laying off everyone but its lawyers—and admittedly (again, in a footnote) malemployed law grads.

The foregoing aside, their math is still incorrect. It’s true that 375,000 legal sector lawyers out of the 592,670 total in the OES equals 63 percent, but that’s not the full number of lawyers. Why? Because the OES omits self-employed workers, which feature prominently in the legal profession. This is an pitfall that I either first noticed or was pointed out to me when I started writing on law schools nearly five years ago, so it’s amusing to see Simkovic and McIntyre make it.

In 2012, the BLS’s Employment Projections program found 759,800 employed lawyers, of which 374,900 were legal sector wage-and-salary employees. According to the BLS’s estimate of the distribution of lawyers among industries (xls), 165,700 lawyers were self-employed workers. It’s just about impossible for these folks to not be working in the legal sector, and indeed, if one looks at the Bureau of Economic Analysis’ National Income and Product Accounts tables, one finds that self-employed workers are included in the category “Persons Engaged in Production by Industry” (Table 6.8D).

As a result, 540,600 lawyers out of 759,800 lawyers—71 percent—work in the legal services sector, not 63 percent. These scant 8 percentage points sure make it look more persuasive to me that what goes on in the legal sector influences law grads’ earnings. (Oh, and I add that another 17 percent of all lawyers work in government. Is that sector robustly hiring lawyers?)

I don’t expect those 8 percentage points to persuade Simkovic and McIntyre, though. They’ve gotten plenty of mileage asking the legal profession to accept on an untested, pure human capital hypothesis that law school pays off even if the legal sector implodes. They can at least include self-employed lawyers in their adverse footnotes.

LSAT Tea-Leaf Reading: February 2015 Edition

BOUNCE!

No. LSAT Takers, 4-Testing Period Moving Sum

20,358 people took the LSAT in February, up 859 (4.4 percent) from 2014. Notably, that’s growth in two consecutive testing administrations. Wow indeed.

This ends our LSAT year with 101,689 total LSAT takers, which is a record low going back to 1986. Back in those days, you said you were listening to Springsteen’s Born in the U.S.A. or Heart’s eponymous album, but we all know couldn’t take Barbra Streisand’s The Broadway Album and the soundtracks to Miami Vice and Rocky IV out of your tape deck. (Cultured readers from my age bracket will recognize how Rocky IV‘s villain’s theme closely resembles that of Unicron’s from the Transformers movie of the same period: Both were crafted by Vince DiCola.)

Back to less exciting 2015, I think there’s a little room left for LSAT takers to drop, but applicants aren’t shying away from law school as they were in the past. They’re down a mere 7 percent from this time last year.

No. Applicants Over App Cycle

It’s unlikely they’ll weigh in at fewer than 50,000, so it looks like we’re pretty close to the bottom if we’re not there already. Who knows, though, maybe it’s just people thinking they could get into elite law schools.

For some perspective on the law school crunch, here’re the trends since the 1960s.

Enrollment, LSAT, & Application Data

The only unobvious insight I can give you from this chart is how amazing it is that peak LSAT in 2009-11 just did not translate into peak applicants. Much of it is due to non-first-time test-takers, but it’s a real harbinger for how things will look going forward. We may be at the applicant trough, but folks, they ain’t coming back.

Follow

Get every new post delivered to your Inbox.

Join 171 other followers