Top Banner
Electronic copy available at: http://ssrn.com/abstract=2332073 Vanderbilt University Law School Law & Economics Working Paper Number 13-27 The Labor Market for New Law Professors Tracey George Vanderbilt University Law School Albert Yoon University of Toronto Journal of Empirical Studies, Forthcoming This paper can be downloaded without charge from the Social Science Research Network Electronic Paper Collection: http://ssrn.com/abstract=2332073
44

Vanderbilt University Law School - The Wall Street Journalonline.wsj.com/public/resources/documents/lawschoolhiring.pdf · Vanderbilt University Law School . ... Understanding who

Jul 15, 2018

Download

Documents

lydang
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Vanderbilt University Law School - The Wall Street Journalonline.wsj.com/public/resources/documents/lawschoolhiring.pdf · Vanderbilt University Law School . ... Understanding who

Electronic copy available at: http://ssrn.com/abstract=2332073

Vanderbilt University Law School

Law & Economics Working Paper Number 13-27

The Labor Market for New Law Professors

Tracey George Vanderbilt University Law School

Albert Yoon

University of Toronto

Journal of Empirical Studies, Forthcoming

This paper can be downloaded without charge from the Social Science Research Network Electronic Paper Collection:

http://ssrn.com/abstract=2332073

Page 2: Vanderbilt University Law School - The Wall Street Journalonline.wsj.com/public/resources/documents/lawschoolhiring.pdf · Vanderbilt University Law School . ... Understanding who

Electronic copy available at: http://ssrn.com/abstract=2332073

1

The Labor Market for New Law Professors

Tracey E. George and Albert H. Yoon

Law school professors control the production of lawyers and influence the evolution of law. Understanding who is hired as a tenure-track law professor is of clear importance to debates about the state of legal education in the United States. But while opinions abound on the law school hiring process, little is empirically known about what explains success in the market for law professors. Using a unique and extensive data set of survey responses from candidates in the 2007-2008 legal academic labor market, we examine the factors that influence which candidates are interviewed and ultimately hired by law schools. We find that law schools appear open to non-traditional candidates in the early phases of the hiring process but when it comes to the ultimate decision—hiring—they focus on candidates who look like current law professors.

I. INTRODUCTION Every year in the United States, approximately a thousand people apply to become a tenure-track law professor.1 Nearly all hold law degrees, often from the most selective

Address correspondence to Tracey E. George, Vanderbilt Law School, 131 21st Avenue South, Nashville, TN 37203; email: [email protected]. George is Charles B. Cox III and Lucy D. Cox Family Chair in Law and Liberty, Professor of Political Science, and Director of the Branstetter Litigation and Dispute Resolution Program at Vanderbilt University. Yoon is Professor of Law, University of Toronto.

We are indebted to the Association of American Law Schools and especially former Executive Directors Carl Monk and Susan Westerberg Prager, Managing Director Jane LaBarbera, and Registration Manager Kai Baker for their assistance. The views expressed in this article are those of the authors and do not necessarily reflect the views of the AALS or its staff. We are grateful to Chris Bransford and Linda Reynolds for their assistance with creation and implementation of the survey, and to Ashley Dennis, Ellen Hunter, and Uros Petronijevic for research assistance. We presented earlier versions of this paper at Georgetown, Houston, Northwestern, and Vanderbilt law schools and benefited from their faculties’ and students’ thoughtful feedback. We also received valuable input from the American Bar Foundation Research Group on Legal Diversity. We thank Ronit Dinovitzer, John Goldberg, Mitu Gulati, Chris Guthrie, Joni Hersch, Lonnie Hoffman, Edward Iacobucci, Helen Levy, David Madigan, Tom Merrill, Richard Posner, Fred Tung, and Alan Wiseman for their helpful comments. This project benefited from the generous financial support of the Russell Sage Foundation and Vanderbilt Law School. All remaining errors are our own.

1 New law professors are hired either through a formal process organized by the Association of American Law Schools (“AALS”) or by informal applications submitted directly to law schools. From 2000-2010, the AALS market has averaged 967 applicants. AALS Statistical Report on Faculty, http://www.aals.org/resources_statistical.php. In 2012, 875 applicants participated in the AALS market. We do not have an accurate count of how many applications are sent to law schools by candidates who bypass the AALS process, but we do know that some new professors are hired outside the AALS process. Thus, 1,000 annual applicants is a reasonable estimate of the number seeking a tenure-track post each year.

Page 3: Vanderbilt University Law School - The Wall Street Journalonline.wsj.com/public/resources/documents/lawschoolhiring.pdf · Vanderbilt University Law School . ... Understanding who

Electronic copy available at: http://ssrn.com/abstract=2332073

2 The Labor Market for New Law Professors

schools.2 Many have served as judicial clerks, a fair number at the U.S. Supreme Court.3 Increasingly, applicants have earned a masters or doctorate in another discipline, or an advanced law degree.4 They are often published authors5 and experienced teachers.6 By most measures, the applicants are highly qualified to be law professors.7 Yet, in a typical year only one in seven applicants will join the tenure-track faculty at one of the roughly

2 See Richard E. Redding, “Where Did You Go to Law School?” Gatekeeping for the Professoriate and Its Implications for Legal Education, 53 J. LEGAL EDUC. 594, 595 (2003) (evaluating the impact of a small number of schools producing the majority of law professors); Deborah Jones Merritt & Barbara F. Reskin, Sex, Race, and Credentials: The Truth About Affirmative Action, 97 COLUM. L. REV. 199 (1997). See also PrawfsBlog, Entry Level Hiring: The 2012 Report, http://prawfsblawg.blogs.com/prawfsblawg/2012/03/entry-level-hiring-the-2012-report.html (presenting and slicing information reported by individual schools to the PrawfsBlog about each school’s entry-level hiring, and finding results consistent with those discussed here but emphasizing that the data is entirely self-reported).

3 See Kevin R. Johnson, The Importance of Student and Faculty Diversity in Law Schools: One Dean’s Perspective, 96 IOWA L. REV. 1549, 1559 (2011) (observing that a Supreme Court clerkship is a credential sought by many law schools); Trenton H. Norris, The Judicial Clerkship Selection Process: An Applicant’s Perspective on Bad Apples, Sour Grapes, and Fruitful Reform, 81 CAL. L. REV. 765, 767-768 (1993) (discussing the importance of a judicial clerkship to becoming a law professor).

4 In academic year 2008-2009 (the most recent year for which data is reported), more than one in three candidates had an advanced non-law degree (Ph.D., Masters, or M.D.) and almost one in three held an advanced law degree such as an LL.M. or J.S.D. One in ten had both. 2008-2009 AALS Statistical Report on Law Faculty: Educational Degrees. http://www.aals.org/statistics/2009far/degrees.html.

5 See, e.g., Brian Leiter, Information and Advice for Persons Interested in Teaching Law, Nov. 2002, http://www.utexas.edu/law/career/academic/Leiter_Teaching_Law.pdf (“it would be fair to say that the single best ticket to a job in law teaching is to have published at least one article since graduating law school. … Publications increasingly make and break candidacies.”); Jeffrey M. Lipshaw, Memo to Lawyers: How Not to “Retire and Teach,” 30 N.C. CENT. L. REV. 151, 158-167 (2008) (reviewing based on his experience the central role of a publication record and research agenda to the law school faculty appointment interview process); Tanya K. Hernandez, Placing the Cart Before the Horse: Publishing Scholarship Before Entering the Legal Academy, 7 MICH. J. RACE & L. 517 (2002).

6 Many law schools offer post-J.D. fellowships for lawyers aspiring to academic jobs. These positions usually include teaching, an experience not typically part of a law student’s experience (as contrasted to a doctoral student’s training). Professor Paul Caron maintains a list of such positions, which proliferated in recent years. Tax Prof Blog, “Fellowships for Aspiring Law Professors (2012-13 Edition),” http://taxprof.typepad.com/taxprof_blog/2012/09/fellowships-for-aspiring.html (listing .

7 See Lawrence B. Solum, Foreword: The New Realities of the Legal Academy, in BRANNON P. DENNING, MARCIA L. MCCORMICK, & JEFFREY M. LIPSHAW, BECOMING A LAW PROFESSOR: A CANDIDATE’S GUIDE ix, x (2010) (“The credentials of many entry-level candidates today would have qualified their possessors for tenure at almost any elite American law school two or three decades ago.”); Lipshaw, supra note 5, at 164 (“The competition for jobs is unbelievably strong. ... Look up the credentials of the youngest faculty [at the lowest ranked schools]... Those professors are, almost uniformly, among the elite law schools’ elite former students.”); Kevin H. Smith, How to Become a Law Professor Without Really Trying: A Critical, Heuristic, Deconstructionist, and Hermeneutical Exploration of Avoiding the Drudgery Associated with Actually Working as an Attorney, 47 U. KAN. L. REV. 139 (1998) (offering a humorous account of the competitiveness of the market).

Page 4: Vanderbilt University Law School - The Wall Street Journalonline.wsj.com/public/resources/documents/lawschoolhiring.pdf · Vanderbilt University Law School . ... Understanding who

The Labor Market for New Law Professors 3

200 accredited law schools.8 And, the number of hires is likely to decline, at least in the short term, because law schools face revenue shortfalls due to shrinking class sizes.9 The competitiveness of the legal academic market is hardly surprising given the competitiveness of nearly all aspects of the legal profession. Despite the dramatic drop in applications over the last three years,10 getting into a top 25 law school as a student is difficult.11 The battle during law school for accolades is well-known.12 And, getting into a top law school is even more important as the entry-level job market becomes tighter.13 Thousands of law students vie for prestigious post-graduate jobs including judicial clerkships14 and associate positions at top firms.15 The labor market for lawyers, from

8 Between 1991 and 2007, approximately 12 percent of AALS participants were “successful” according to the AALS’s own reporting, or about 100 participants each year. AALS, Statistical Report on Law School Faculty and Candidates for Law Faculty Positions, 2005-2006, Table 13A, http://www.aals.org/statistics/0506/0506_T13A_E_14yr-7yr.html. See also Frank T. Read & M.C. Mirow, So Now You’re a Law Professor: A Letter from the Dean, 2009 CARDOZO L. REV. DE NOVO 55, 57 (Read, who served as the dean at five different law schools over his career, observed that “[i]n any law school, no matter how large, it is very rare to have more than one or two new teachers each fall”); American Bar Association, ABA-Approved Law Schools, http://www.americanbar.org/groups/legal_education/resources/aba_approved_law_schools.html (reporting the current number of ABA-approved law schools (199) and provisionally accredited schools).

9 See, e.g., Boston Public Radio, Here & Now Radio Program, “Law School Enrollment Plunges,” July 18, 2013, at http://hereandnow.wbur.org/2013/07/18/law-school-enrollment.

10 Ethan Bronner, “Law Schools’ Applications Fall as Costs Rise and Jobs Are Cut,” N.Y. Times, Jan. 30, 2013, at http://www.nytimes.com/2013/01/31/education/law-schools-applications-fall-as-costs-rise-and-jobs-are-cut.html?pagewanted=all&_r=0.

11 “7 Things You Don’t Know About Law School Admissions,” Bloomberg Law, June 5, 2013, at https://www.youtube.com/watch?v=Jr-Els4z9Y0&feature=youtu.be (reporting Bloomberg Law’s computations, based on ABA and LSAC data, that the top 25 law schools’ had an average rate of acceptance of 23%). See also Jesse Rothstein & Albert Yoon, Affirmative Action in Law School Admissions: What do Racial Preferences Do? 75 U. CHI. L. REV. 649, 662 n.49 (2007) (describing how only three of the 185 law schools accepted more than half of their applicants in 2007, which is same year as our data); Law School Admissions Council, LSAT Volume Data, at http://www.lsac.org/lsacresources/data/lsac-volume-summary.asp (reporting annual figures on LSAT takers, JD applicants, and matriculants).

12 See, e.g., JOHN JAY OSBORN, JR., THE PAPER CHASE (1971); The Paper Chase (Twentieth Century Fox 1973); SCOTT

TUROW, ONE L: THE TURBULENT TRUE STORY OF A FIRST YEAR AT HARVARD LAW SCHOOL (1977); AMANDA BROWN, LEGALLY BLONDE (2001); Legally Blonde (MGM 2001).

13 See Karen Sloan, “Read This if You Want a Legal Job: It Matters Which Law School You Attend,” Nat’l L.J., Apr. 8, 2013, at http://www.law.com/jsp/nlj/legaltimes/PubArticleLT.jsp?id=1202595014134; see generally The Law School Transparency Project. LST Score Reports, http://www.lstscorereports.com/.

14 See Administrative Office of the U.S. Courts, “More Federal Judges Participate in OSCAR in FY 2012,” The Third Branch News, Feb. 22, 2013, http://news.uscourts.gov/more-federal-judges-participate-oscar-fy-2012. .

Page 5: Vanderbilt University Law School - The Wall Street Journalonline.wsj.com/public/resources/documents/lawschoolhiring.pdf · Vanderbilt University Law School . ... Understanding who

4 The Labor Market for New Law Professors

beginning to end, is competitive. The recent economic downturn and technological advances in law have only exacerbated this competitiveness.16 We know a great deal about the dynamics of the other segments of the law market, but we know surprisingly little about the market for law professors.17 The law school pipeline for students is a subject of regular study.18 Scholars have examined the variety of career paths of recent law graduates19 as well as their experiences in their work.20 The workings of law firms have been subjected to careful scrutiny,21 as has the work of specific segments of the bar.22 While we obviously need to know more about law school

15 See, e.g., “Ranking the Go-To Law Schools: A Special Report,” The National Law Journal, http://www.law.com/jsp/nlj/PubArticleNLJ.jsp?id=1202589189668&interactive=true (ranking law schools by the percentage of 2012 graduates who obtain associate positions at NLJ 250 firms).

16 See, e.g., BRUCE MACEWEN, GROWTH IS DEAD: NOW WHAT? (2013); RICHARD SUSSKIND, TOMORROW’S LAWYERS: AN

INTRODUCTION TO YOUR FUTURE (2013); RICHARD SUSSKIN, THE END OF LAWYERS? RETHINKING THE NATURE OF LEGAL

SERVICES (2008). For a more optimistic assessment of the legal labor market, see Michael Simkovic & Frank McIntyre, The Economic Value of a Law Degree (working paper) (2013), available at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2250585.

17Empirical studies have examined the composition of law school faculties, but have not looked at the market by which applicants are matched to jobs. See section II infra.

18 See, e.g., Charles E. Daye, Walter R. Allen & Linda F. Wightman, The Educational Diversity Project: Analysis of Longitudinal and Concurrent Student and Faculty Data (Law School Admission Council Grants Report 10-01, March 2010) [http://www.lsac.org/LSACResources/Research/GR/GR-10-01.pdf]; Rothstein & Yoon, supra note 11; William D. Henderson & Andrew P. Morriss, Student Quality as Measured by LSAT Scores: Migration Patterns in the U.S. News Rankings Era, 81 IND. L.J. 163 (2006); Charles Longley, Law School Admissions, 1985-1995: Assessing the Effect of Application Volume (Law School Admissions Council Grant Report 1998); Charles Longley, Who Gets the App? Explaining Law School Application Volume, 1993 to 1996 (Law School Admissions Council Grant Report 1998) [http://www.lsac.org/LSACResources/Research/GR/grants-reports.asp].

19 See, e.g., AFTER THE JD II: SECOND RESULTS FROM A NATIONAL STUDY OF LEGAL CAREERS (2009); AFTER THE JD: FIRST

RESULTS OF A NATIONAL STUDY OF LEGAL CAREERS (2003) [http://www.americanbarfoundation.org/uploads/cms/documents/ajd.pdf].

20 See, e.g., Joyce S. Sterling & Nancy J. Reichman, So, You Want to be a Lawyer? The Quest for Professional Status in a Changing Legal World, 78 FORDHAM L. REV. 2289 (2010); Bryant G. Garth & Joyce S. Sterling, Exploring Inequality in the Corporate Law Firm Apprenticeship: Doing Time, Finding the Love, 22 GEORGETOWN J. LEGAL ETHICS 1361 (2009).

21 See, e.g., David Wilkins, Ronit Dinovitzer, & Rishi Batra, Urban Law School Graduates in Large Law Firms, 36 SW. U.L. REV. 433 (2007); LINCOLN CAPLAN, SKADDEN: POWER, MONEY, AND THE RISE OF A LEGAL EMPIRE (1993); CARROLL SERON, THE BUSINESS OF PRACTICING LAW: THE WORK LIVES OF SOLO AND SMALL-FIRM ATTORNEYS (1996); JOHN

P. HEINZ ET AL., URBAN LAWYERS: THE NEW SOCIAL STRUCTURE OF THE BAR (2005); JOHN HEINZ & EDWARD O. LAUMANN, CHICAGO LAWYERS; THE SOCIAL STRUCTURE OF THE BAR (1982); MARC GALANTER & THOMAS PALAY, TOURNAMENT OF

LAWYERS: THE TRANSFORMATION OF THE BIG LAW FIRM (1991).

Page 6: Vanderbilt University Law School - The Wall Street Journalonline.wsj.com/public/resources/documents/lawschoolhiring.pdf · Vanderbilt University Law School . ... Understanding who

The Labor Market for New Law Professors 5

admissions, lawyers, and legal practice, scholars are actively pursuing multiple lines of inquiry on the subject of the market for lawyers. The hiring of individuals who maintain this market—law professors—has not undergone careful and systematic examination. Law schools are in crisis as the New York Times, Wall Street Journal, and other media outlets now routinely remind us.23 Faced with shrinking classes and decreasing tuition revenue, law deans are talking about cutting faculty and freezing salaries.24 Last year, the American Bar Association created a “Task Force on the Future of Legal Education” to issue recommendations about the financing and provision of legal education.25 Even President Obama has weighed in, suggesting that “[l]aw schools would probably be wise to think about being two years instead of three years.”26 Many proposals imagine dramatic changes to the structure of legal education. But, we first should better understand the central player and single largest expenditure in the training of American lawyers: law professors. Any plan for addressing the costs of legal education and the future of law schools must look at how we have hired law faculty. That is what we do here. We seek in this article to explain success in the entry-level, tenure-track law teaching market from beginning to end.27 This process remains largely a black box. While many scholars have written about their market experiences (as candidates or recruiters)

22 See, e.g., David S. Abrams & Albert H. Yoon, The Luck of the Draw: Using Random Case Assignment to Investigate Attorney Ability, 74 U. CHI. L. REV. 1145 (2007); LYNN MATHER ET AL., DIVORCE LAWYERS AT WORK: VARIETIES OF PROFESSIONALISM IN PRACTICE (2001).

23 See, e.g., David Segal, “Is Law School a Losing Game?” N.Y. Times, Jan.9, 2011, at BU1; Steven M. Davidoff, “The Economics of Law School,” N.Y. times, Sept. 25, 2012, at F8; Jennifer Smith, “A Crop of New Law Schools Opens Amid a Lawyer Glut,” Wall St. J., Feb. 1, 2013.

24 See David Lat, “A Law School’s Possible Purge of Its Junior Faculty Ranks,” Above the Law, July 1, 2013, at http://abovethelaw.com/2013/07/a-law-schools-possible-purge-of-its-junior-faculty-ranks/.

25 American Bar Association, Task Force on the Future of Legal Education, http://www.americanbar.org/groups/professional_responsibility/taskforceonthefuturelegaleducation.html.

26 Dylan Matthews, “Obama Thinks Law School Should Be Two Years,” Wash. Post, Aug. 27, 2013, at http://www.washingtonpost.com/blogs/wonkblog/wp/2013/08/27/obama-thinks-law-school-should-be-two-years-the-british-think-it-should-be-one/.

27 The present study focuses on tenure-track hires for several reasons: (1) the tenure-track job market is largely distinct from the market for term appointments (such as clinical and research and writing posts at most institutions), for visiting professors, and for adjunct faculty; (2) professors on the tenure-track typically hold greater authority and power than untenured faculty; and (3) we would not expect qualifications to be the same and/or carry the same relative importance for non-tenured positions. See also Merritt & Reskin, supra note 2, at 206 (explaining their decision to focus on tenure-track professors).

Page 7: Vanderbilt University Law School - The Wall Street Journalonline.wsj.com/public/resources/documents/lawschoolhiring.pdf · Vanderbilt University Law School . ... Understanding who

6 The Labor Market for New Law Professors

and a handful have examined the composition of law faculties,28 no other study systematically analyzes the factors that influence how candidates perform on the teaching market.29 Just as importantly, because the law teaching market involves multiple stages, we do not know the factors that influence how law schools and candidates make decisions throughout this process. 30 The entry-level job market is the most important segment of the law teaching market to understand because the first academic job matters and tenured professors dominate legal education.31 An academic’s first position significantly affects her career opportunities and trajectory.32 A process of accumulative advantage exists such that the prestige of an academic’s first post gives greater opportunity and resources leading to advantages in the form of publications, citation, awards, upward mobility, and other measures of

28 Deborah Merritt and Barbara Reskin, for example, completed a substantial quantitative analysis of law professors, focusing on those who had been hired during a five-year period . See Merritt & Reskin, supra note 2 (presenting the primary multivariate analysis of their substantial database); Deborah J. Merritt, Barbara F. Reskin, & Michelle Fondell, Family, Place, and Career: The Gender Paradox in Law School Hiring, 1993 WISC. L. REV. 395 (1993); Deborah J. Merritt & Barbara F. Reskin, The Double Minority: Empirical Evidence of a Double Standard in Law School Hiring of Minority Women, 65 S. CAL. L. REV. 2299 (1992). See also Robert J. Bothwick & Jordan R. Schau, Gatekeepers of the Profession: An Empirical Profile of the Nation’s Law Professors , 25 U. MICH. J.L. REFORM 191 (1991); Donna Fossum, Law Professors: A Profile of the Teaching Branch of the Legal Profession, 1980 AM. B. FOUND. RES. J. 501.

29 Our database includes all job applicants who participated in the AALS faculty recruitment process as well as successful applicants who found jobs outside that process. We do not have data on unsuccessful applicants who did not use the AALS process.

30 We do not have data on prospective candidates who decided against applying for law teaching jobs. As discussed later, such self-selection is relevant to the interpretation of our empirical findings.

31 The ABA requires each law school require tenure or a comparable form of job security for its full-time faculty (excluding legal writing instructors and clinical professors). See ABA Standards and Rules of Procedure for Approval of Law School, available at http://www.americanbar.org/groups/legal_education/resources/standards.html.

32For example, several empirical studies have found that law professors at elite law schools have a comparative advantage in article placement in both their home institution’s law journal and in law journals generally. See, e.g., Albert H. Yoon, Editorial Bias in Law Reviews, working paper (2013); Leah M. Christensen & Julie A. Oseid, Navigating the Law Review Selection Process: An Empirical Study of Those With All the Power—Student Editors, 59 S.C.L. REV. (2008) (finding that law review editors reported that author credentials, specifically ranking of law school, influenced the selection process); Deborah Jones Merritt, Research and Teaching on Law Faculties: An Empirical Exploration, 73 CHI.-KENT L. REV. 765, 813-814 (1998) (finding that “[i]nstitutional prestige” after controlling for other variables “appears to affect a professor’s ability to publish in the top journals, but not his or her overall productivity”); Jordan H. Leibman & James P. White, How the Student-Edited Law Journals Make Their Publication Decisions, 39 J. LEGAL EDUC. 387, (1989) (learning in interviews with law review editors that author credentials played a significant screening or signaling role in selection); Ira Mark Ellman, A Comparison of Law Faculty Production in Leading Law Reviews, 33 J. LEGAL EDUC. 681, 692 (1983) (reporting that “the major law reviews publish the work of their own faculty disproportionately often”).

Page 8: Vanderbilt University Law School - The Wall Street Journalonline.wsj.com/public/resources/documents/lawschoolhiring.pdf · Vanderbilt University Law School . ... Understanding who

The Labor Market for New Law Professors 7

attainment.33 The entry-level institution’s rank or status is the key measure for success as a junior scholar as well as a key factor in her subsequent achievements.34 This article explores in detail the market for entry-level, tenure-track law professors through a unique dataset of entry-level candidates on the job market in the 2007-2008 academic year. Our primary results are both intuitive and counterintuitive. In the early stages of the job market, law schools are more willing to consider candidates who do not possess the traditional credentials of legal academia such as an elite law school education or a judicial clerkship. As law schools narrow their searches, they shift the focus of their recruitment efforts on candidates who possess these high-status credentials. In the later stages of the market, certain credentials explain whether a candidate lands a job but not where, while others explain where the candidate starts but not whether she will get a job. A doctorate in the social sciences or STEM is not statistically significantly related to a candidate’s chance of getting a call-back interview or an offer but significantly increases the odds of any offer coming from a higher ranked school. A law teaching fellowship, by contrast, substantially increases the probability of getting interviews and a job offer, but does not have a statistically significant effect on the ranking of the school which offered the job. Despite the ink spilled on race and gender in legal academic hiring, we find, with limited exceptions, these factors have little effect. After controlling for credentials, gender and race do not improve a candidate’s chances of getting a screening interview. The only stage where we find that race and gender have statistically significant effects are at the intermediate call-back interview stage where women and non-whites are statistically significantly more likely to be invited for a job talk interviews. But, women and non-whites are no more likely than similarly situated men and whites to get a job offer or, if they get an offer, for the offer to come from a more elite school. One factor looms large after the initial screening interview stage: where the candidate attended law school. Even controlling for other factors, graduates of top-50 law schools, and especially of Yale, Harvard, or Stanford, are much more likely than other candidates to receive call-back interviews and job offers. Top 50 law schools hire almost solely from

33 See, e.g., Paul D. Allison & J. Scott Long, Department Effects on Scientific Productivity, 55 AM. SOC. REV. 469 (1990); Barbara F. Reskin, Scientific Productivity and the Reward Structure of Science, 42 AM. SOC. REV. 491

(1977) (reporting that the prestige of Ph.D.-granting institution and entry-level institution affect productivity ten years after the doctorate); Diana Crane, Scientists at Major and Minor Universities: A Study of Productivity and Recognition 30 AM. SOC. REV. 699 (1965); see also STEPHEN COLE, MAKING SCIENCE: BETWEEN NATURE AND

SOCIETY (1992) (discussing the accumulation effect).

34 See, e.g., Ted I.K. Youn & Daniel Zelterman, Institutional Career Mobility in Academia, in ACADEMIC LABOR

MARKETS AND CAREERS 52, 56-58 (David W. Breneman & Ted I.K. Youn eds., 1988) J. Scott Long, Productivity and Academic Position in the Scientific Career, 43 AM. SOC. REV. 889 (1978).

Page 9: Vanderbilt University Law School - The Wall Street Journalonline.wsj.com/public/resources/documents/lawschoolhiring.pdf · Vanderbilt University Law School . ... Understanding who

8 The Labor Market for New Law Professors

other top 50 schools, and all law schools typically hire their faculty from schools ranked higher than them. This article proceeds as follows. In Part II, we explain the nature and functioning of the legal academic job market and describe our project collecting data on the candidates involved in that process during one academic year. In Part III, we construct a model of the market. Our hypotheses draw on existing economic and sociological theories of labor markets as well as studies of the market for professors in and out of law schools. In Part IV we present a detailed account of the candidates on the market and a statistical analysis of the factors that influence how individuals perform at each stage of the job market. Part VI discusses implications of our results to an understanding of law school hiring. We conclude by considering what this all means for legal education. II. A TOURNAMENT OF LAW PROFESSOR ASPIRANTS The entry-level law teaching market is unusually well-ordered. Prospective hires may run an independent search strategy, but many will choose to participate in an organized exchange created by the Association of American Law Schools or “AALS”.35 They do so because (almost) all law schools participate. The AALS process is the primary market and is governed by both rules and norms. In this section, we explain the characteristics of the market and our method for collecting data about it. The first section is divided into the four steps in the process for the successful job applicant: application, screening interview, job-talk or call-back interview, and job offer. The second section describes the detailed information we collected at each stage of this market, including how we administered the survey and compared respondents to non-respondents to disclose any possible response bias relevant to interpretation of our results. A. The Market 1. Submitting an Application The AALS process operates like an auction.36 Candidates participate by submitting biographical information to the AALS using a “faculty appointments register” form (or 35 The AALS process accounted for the substantial majority of tenure-track hires in 2008. In 2007-2008, law schools hired 220 new tenure-track professors, 167 were from the AALS process (or 76%). See text accompanying notes 35-40 infra (describing the search strategies of hires).

36 See http://aals.org/services_recruitment.php (website for the AALS describing faculty recruitment). Although the conference is dedicated to entry-level tenure-track hiring, some candidates register with the AALS for clinical or research and writing (typically non-tenure-track) rather than research (tenure-track) positions. In addition, some individuals who register with the AALS already have a tenure-track job and are looking to switch institutions. In our analysis, we focus on new hires to the entry-level tenure-track market.

Page 10: Vanderbilt University Law School - The Wall Street Journalonline.wsj.com/public/resources/documents/lawschoolhiring.pdf · Vanderbilt University Law School . ... Understanding who

The Labor Market for New Law Professors 9

“FAR” form, as it commonly known among candidates) which is available online and then uploaded into a searchable database. The form asks for background information about the candidate, including education, work experience, publications, and teaching experience. The candidate also may list courses that they are interested in teaching and any geographical preferences. Many candidates choose to attach curriculum vitae to their profiles. AALS distributes these forms in four waves: the first and largest is in August and the next in time and size is in September. 2. Screening Interviews The market sorts and clears through a three-day national meeting sponsored by AALS in late October or early November. After reviewing the FAR forms, schools invite candidates to meet with a committee for a brief screening interview (typically 20-30 minutes) at the AALS national meeting, known officially as the AALS Faculty Recruitment Conference (and unofficially as the meat market).37 Schools may opt to screen a candidate outside the process, either to preempt other schools or to take advantage of geographic proximity. The norm is to screen a candidate prior to a full interview with the entire faculty. Schools, however, may choose to skip this step and go directly to the on-campus interview. 3. Job-Talk Interviews Following screening interviews (usually), schools invite a smaller number of candidates to campus to meet with the rest of the faculty and to give a job talk typically based on the candidates’ current or recent research.38 Where schools may interview thirty or more candidates at the screening stage, they typically will interview fewer than 10 at the on-campus interview stage. The call-back or job-talk interview is extensive, involving a day or more of interviews and meals with faculty (and occasionally students), tours of campus and environs, and a one-on-one meeting with the dean (or a designate). The job talk itself may be the most important aspect of the visit.39 4. Job Offers After the on-campus interview, schools decide whether and to whom to extend a tenure-track offer. Law school hiring typically works through a faculty appointments committee. The committee is charged with handling all steps of the process. While committees usually have complete discretion at the first two stages, they typically make only a

37 For a discussion of the conference, see, e.g., James D. Gordon III, An Insider’s Guide to the Faculty Recruitment Conference, 43 J. LEGAL EDUC. 301 (1993).

38 Our description is of the modal school. There are some exceptions. At least one law school is known to have two on-campus interview steps: the first with the full committee and the second with the full faculty.

39 See, e.g., Anne Enquist, Paula Lustbader, & John B. Mitchell, From Both Sides Now: The Job Talk’s Role in Matching Candidates with Law Schools, 42 U. TOL. L. REV. 619 (2011).

Page 11: Vanderbilt University Law School - The Wall Street Journalonline.wsj.com/public/resources/documents/lawschoolhiring.pdf · Vanderbilt University Law School . ... Understanding who

10 The Labor Market for New Law Professors

recommendation at the final stage. Generally, the full tenure-track faculty votes on whether to extend an offer. The most successful candidates routinely have multiple offers. The hiring process can be analogized to a tournament.40 For most candidates, the tournament represents a competition with one another for limited tenure-track openings at schools, where schools eliminate candidates in successive stages. Most elimination occurs in the first stage, with fewer in each successive round. For the most successful candidates, however, the competition is bilateral. They are competing for the best offers, while schools are competing with one another to convince these candidates to join the faculty. B. The Data Our data comes primarily from surveys of candidates. The survey asked detailed questions about the candidate and about her experiences on the entry-level law teaching job market.41 We supplemented these responses with limited biographical information--degrees obtained, subject areas of interest, and publications-- from other sources including the candidate’s AALS FAR form and online information.42 We surveyed two groups of individuals: 858 candidates who participated in the AALS faculty recruitment market and 53 new law professors who were hired without participating in the AALS market. Given the sensitive subject matter of our study, our response rate was good for both groups: 52% of the AALS candidates whom we surveyed completed all survey questions and an additional 8% completed at least part of the survey.43 In addition, 22 of the new hires outside the AALS market completed our survey, 40 For an interesting discussion of tournaments in a different legal environment, see MARC GALANTER AND

THOMAS PALAY, TOURNAMENT OF LAWYERS: THE TRANSFORMATION OF THE BIG LAW FIRM (1994).

41 Our surveys were approved through our respective Institutional Review Boards. The IRBs’ approval requires that our results are reported in a manner that ensures the anonymity of respondents. Consequently, we are constrained in our ability to use variables which would involve small numbers of candidates who might then be identifiable. We also followed numerous other steps designed to ensure that surveyed individuals knew that participation was voluntary and confidential, and it would not affect their candidacy for a position. For example, both authors were firewalled from the entry-level hiring process at our own institutions during the 2007-2008 academic year.

42 We worked with AALS in 2006, 2007, and 2008 to contact AALS FAR candidates to ask them to participate in the survey. AALS granted us permission to use the FAR database to supplement our survey results and also to provide a benchmark against which to evaluate the representativeness of the survey sample. During the 2007-2008 faculty recruitment process, candidates were allowed to opt out of having their FAR forms used as the basis of an academic study. A small number of candidates did so and are not included in our database. The AALS FAR data is owned by AALS, and as a result, we cannot share the data with third parties

43 For a discussion of issues related to response rates below 100%, see EARL BABBIE, THE PRACTICE OF SOCIAL

RESEARCH 272-274 (12th ed. 2010) (Babbie also observes that a return rate of 60% for mail surveys is good); DELBERT C. MILLER, HANDBOOK OF RESEARCH DESIGN AND SOCIAL MEASUREMENT (1991); Don A. Dillman, Mail and Other Self-Administered Questionnaires, in HANDBOOK OF SURVEY RESEARCH 359 (Peter H. Rossi et al. eds. 1983).

Page 12: Vanderbilt University Law School - The Wall Street Journalonline.wsj.com/public/resources/documents/lawschoolhiring.pdf · Vanderbilt University Law School . ... Understanding who

The Labor Market for New Law Professors 11

or 42%. We supplemented this survey data with information collected from multiple sources, including the AALS faculty appointments register, publicly available websites with biographical information, and reports of new hires posted on blogs and individual school’s websites. Whenever a survey achieves less than a perfect response rate, it is crucial to determine whether there is a non-response bias which will skew any resulting findings. In order to evaluate the possible nature of bias, we begin by explaining how the survey was administered and then by comparing the characteristics of respondents to non-respondents. 1. Survey Administration The survey data is based on individual surveys of candidates who were on the job market in academic year 2007-2008.44 We administered four surveys: (1) an October 2007 survey focused on screening interviews, (2) a May 2008 follow-up survey of those who completed the first survey and asking about subsequent developments, including call-back interviews and job offers, (3) a combined paper survey sent after the 2007-2008 hiring season ended to anyone who did not complete the online surveys, and (4) a combined online survey to new professors hired outside the AALS market. For the first survey, we contacted all AALS-registered individuals by email in late October 2007, which was shortly after the faculty recruitment conference.45 The email invited them to complete a confidential online survey by clicking on a link and offered a $20 Amazon gift card for completion. 46 The survey included questions on demographics, prior academic job search experiences, and details about their current search, including school inquiries and interviews. We followed up with non-respondents by email and postcard to encourage them to complete the survey. At the end of the 2007-2008 academic year, we again contacted respondents via email with a follow-up survey focusing on their experiences during the call-back and job-offer stages of the job market. We asked them which schools, if any, invited them to give a job talk, and whether they accepted the interview. We then asked what tenure-track job offers,

44 AALS provided us with email addresses for each of the four waves of applicants. Candidates were given the option of refusing to allow us to contact them, and a handful asked not to be contacted. The surveys also made clear that participation was voluntary and confidential, and that any resulting publications would protect the anonymity of respondents.

45 All surveys were retrospective, asking candidates about their experiences up to that point in the market. However, the first online survey was completed while many candidates were still participating in the market, while nearly all candidates had finished their search by the time of the subsequent surveys.

46 Incentives are an important component of increasing mail survey response rates. See, e.g., A.H. Church, Estimating the Effect of Incentives on Mail Survey Response Rates: A Meta-Analysis, 57 PUB. OP. Q. 62 (1993)

Page 13: Vanderbilt University Law School - The Wall Street Journalonline.wsj.com/public/resources/documents/lawschoolhiring.pdf · Vanderbilt University Law School . ... Understanding who

12 The Labor Market for New Law Professors

if any, they received, and whether they accepted an offer.47 The survey included additional questions about the candidate’s characteristics as well. Participants again received a $20 Amazon gift card for completing the second survey. We also followed up by email and postcard throughout 2008 and into 2009, and survey responses continued to come in. For candidates who did not participate in the first online survey, we mailed them a paper survey in the summer of 2008. The paper survey asked the same questions as the two electronic surveys. Participants who completed the paper survey were given a $40 Amazon gift card. We followed up with non-respondents, and survey responses continued to come during 2009. In the spring of 2010, we identified the complete list of individuals who were hired for tenure-track law teaching jobs during the 2007-2008 academic year. This information was drawn in part from Larry Solum’s blog on entry-level hires for each year. We verified Solum’s list by visiting each law school’s web site. For schools for which Solum did not report any information, we contacted the associate dean’s office directly for a list of tenure-track hires for that year. We believe that our dataset provides a complete set of entry-level tenure-track hires for this academic year. For individuals we identified as having been hired during the 2007-2008 academic year but who did not complete a survey, we made a final attempt in the summer of 2010 to garner their participation through a single, electronic survey. This included fifty-three individuals who did not register for FAR and accordingly did not participate in the AALS faculty recruitment conference. Participants who completed this survey received a $50 Amazon gift card. In an effort to obtain full information about all new law professors, we followed up with all new hires by individual emails and phone calls to encourage them to participate. 2. Comparing Respondents to Non-Respondents We compare respondents to non-respondents to uncover any sample bias. Of the 911 AALS market participants, 466 completed all survey questions. Men made up the same percentage of survey respondents as non-respondents (64%), and non-whites made up about the same percentage (24% of respondents compared to 23% of non-respondents).48 47 Specifically, we first asked respondents whether they received an job offer, and in a separate question asked the nature of the job offer (e.g., tenured, tenure-track, clinical, fellowship, etc.). Not surprisingly, the vast majority of offers were tenure-track.

48 This figure compares known race. We know the race of all survey respondents because every respondent answered our question about race. However, many AALS FAR applicants choose to leave the question blank on the form which is circulated to law schools. Comparing the representation of whites in the respondent and non-respondent groups, we find that 76% of survey respondents identify as white while only 63% of non-respondents identify as white. However, an additional 14% of non-respondents did not answer the FAR form race question. If the pools are otherwise similar on race, then most of the non-respondents are likely white. This would be consistent with anecdotal accounts that the majority of FAR candidates who do not disclose race are white. See sources cited in notes infra.

Page 14: Vanderbilt University Law School - The Wall Street Journalonline.wsj.com/public/resources/documents/lawschoolhiring.pdf · Vanderbilt University Law School . ... Understanding who

The Labor Market for New Law Professors 13

Candidates who graduated from the first tier of law schools were slightly more likely to complete the full survey, and those from the fourth tier were slightly less likely to do so. The survey respondent data includes a higher percentage of candidates who clerked (40% versus 35%), hold a non-law doctorate (14% versus 11%), and published academic writing (68% versus 62%). Thus, our sample has more candidates whose qualifications have traditionally been associated with market success. However, the differences are sufficiently modest that the sample is fairly representative of the prospective participants. We also received partial surveys from 73 candidates: they answered the first survey regarding screening interviews but did not complete the second survey regarding job-talk interviews and job offers. We expect that these candidates dropped out between survey 1 and survey 2 because they withdrew from the job market. Of those candidates who completed survey 1 but not survey 2, most had no screening interviews. We include these candidates in our evaluation of market success at the first stage, but do not include them in our statistical analyses of later stages. However, we do consider how our results would look if we assume that they dropped out of the market after failing to get any screening interviews. III. A MODEL OF SUCCESS ON THE ACADEMIC JOB MARKET Scholarship concerning the legal academic market has been largely anecdotal and ranges from the non-controversial to the contentious.49 The non-controversial writing is of the “how-to” variety.50 Current legal academics have drawn on their own and colleagues’ experiences to offer insights to potential job candidates.51 Prominent legal scholars, including Brian Leiter and Larry Solum, have maintained popular and influential blogs on the academic hiring market, providing up-to-date information on who has been hired and

49 Arguments regarding the role of ideology in law school hiring falls at the contentious end of the spectrum. See, e.g., Adam Liptak, Lawsuit Pits Political Activism Against Campus Diversity, N.Y. Times, Jan. 9, 2012. We asked respondents about their ideology, but the smaller response rate and issues of relative salience (i.e., whether ideology was easily discerned by schools) made it inappropriate for inclusion in the present study which looks at the entire sample.

50 See, e.g., Lucinda Jesson, So You Want to be a Law Professor, 59 J. LEGAL ED. 450 (2010); Brad Wendel, The Big Rock Candy Mountain: How to Get a Job in Law Teaching, http://ww3.lawschool.cornell.edu/faculty-pages/wendel/teaching.htm.

51 See, e.g., BRANNON P. DENNING, MARCIA L. MCCORMICK, & JEFFREY M. LIPSHAW, BECOMING A LAW PROFESSOR: A

CANDIDATE’S GUIDE (2010); RONALD EADE, HOW TO BE A LAW PROFESSOR GUIDE: FROM GETTING THAT FIRST JOB TO

RETIREMENT (2008); David W. Case, The Pedagogical Don Quixote de la Mississippi, 33 U. MEM. L. REV. 529, 568 (2003); James Gordley, Mere Brilliance: The Recruitment of Law Professors in the United States, 41 AM. J. COMP. L. 367 (1993); Don Zillman, Marina Angel, Jan Laitos, George Pring & Joseph Tomain, Uncloaking Law School Hiring: A Recruit’s Guide to the AALS Faculty Recruitment Conference, 38 J. LEGAL EDUC. 345 (1988); Jon W. Bruce & Michael I. Swygert, The Law Faculty Hiring Process, 18 HOUSTON L. REV. 215 (1981); William J. Prosser, Advice to the Lovelorn, 3 J. LEGAL EDUC. 505 (1951).

Page 15: Vanderbilt University Law School - The Wall Street Journalonline.wsj.com/public/resources/documents/lawschoolhiring.pdf · Vanderbilt University Law School . ... Understanding who

14 The Labor Market for New Law Professors

offering their insights for those seeking jobs.52 While these writers may disagree, the exchange is friendly and motivated by a desire to inform and advise prospective law professors. Scholarship on race and gender in faculty hiring is part of an often heated debate over the appropriate bases for selecting new law professors. 53 The competing analyses draw primarily on the authors’ first-hand experiences and informal observations,54 but sometimes cite the relative number of whites and minorities, and women and men who are successful on the entry-level law teaching job market.55 But, beyond these raw figures, the conclusions draw heavily upon the authors’ own experiences as part of the market. A noteworthy exception is research by Deborah Merritt and Barbara Reskin.56 They looked at professors who were hired between 1986 and 1991, and found that race and sex had little influence on relative prestige of the first job after controlling for other factors. Their research marked a substantial step forward in understanding how law schools hire new tenure-track professors. But, Merritt and Reskin limited their study to the fraction of applicants who were ultimately hired. Given the substantial winnowing from the initial pool of candidates to those hired, we clearly need to understand the process from the beginning.

52 See Brian Leiter’s Law School Reports, Advice for Academic Job Seekers, http://leiterlawschool.typepad.com/leiter/advice-for-academic-job-seekers/; Lawrence B. Solum, Legal Theory Blog, Entry Level Hiring Report, http://lsolum.typepad.com/legaltheory/entry_level_hiring_report/. See also PrawfsBlawg, Entry-Level Hiring Report, http://prawfsblawg.blogs.com/prawfsblawg/2011/04/rookies-hiring-an-open-thread.html.

53 See, e.g., Michael Stokes Paulsen, Reverse Discrimination and Law School Faculty Hiring: The Undiscovered Opinion, 71 TEX. L. REV. 993 (1993); Richard H. Chused, The Hiring and Retention of Minorities and Women on American Law School Faculties, 137 U. PA. L. REV. 537 (1988)

54 See, e.g., John Hasnas, Affirmative Action and the New Discrimination: A Reply to Duncan Kennedy, 54 LA. L. REV. 263 (1993); Marina Angel, Women in Legal Education: What It’s Like to be Part of A Perpectual First Wave or the Case of the Disappearing Women, 61 TEMPLE L. REV. 799 (1988).

55 See, e.g., Angela Onwuachi-Willig, Complimentary Discrimination and Complementary Discrimination in Faculty Hiring, 87 WASH. U.L. REV. 763 (2010); Alfred C. Yen, A Statistical Analysis of Asian Americans and the Affirmative Action Hiring of Law School Faculty, 3 ASIAN L.J. 39 (1996). The Society of American Law Teachers (“SALT”) has collected independent data on law school faculty composition by race and gender. See Richard H. Chused, The Hiring and Retention of Minorities and Women on American Law School Faculties, 137 U. PA. L. REV. 537 (1988). For the AALS data, see, for example, Richard A. White, The Gender and Minority Composition of New Law Teachers and AALS Faculty Appointments Register Candidates, 44 J. LEGAL EDUC. 424 (1994) (compiling data from several years); AALS Statistical Report, supra note 1.

56 See Deborah Jones Merritt & Barbara F. Reskin, Sex, Race, and Credentials: The Truth About Affirmative Action, 97 COLUM. L. REV. 199 (1997) (presenting the primary multivariate analysis of their substantial database); Deborah J. Merritt & Barbara F. Reskin, The Double Minority: Empirical Evidence of a Double Standard in Law School Hiring of Minority Women, 65 S. CAL. L. REV. 2299 (1992).

Page 16: Vanderbilt University Law School - The Wall Street Journalonline.wsj.com/public/resources/documents/lawschoolhiring.pdf · Vanderbilt University Law School . ... Understanding who

The Labor Market for New Law Professors 15

Conventional wisdom and anecdotal accounts about the law teaching market abound, but we need a fuller explanation or theory to determine whether certain variables influence academic market actions. Inference requires a reason to believe that any observed statistical relationship reflects an underlying causal one.57 A large body of empirical research has explored other academic labor markets. This research is dominated by the economic theory of human capital and the sociological theory of status attainment and stratification.58 Human capital theory connects individual attributes, background, and endowments (ability, education, and training) to market results.59 Under this conception, individuals invest in developing valuable skills and abilities and as a result gain a market return for those investments. In the academic job market, publication of research papers and, to a lesser extent, experience in teaching are deemed important assets to gain a position that values scholarly productivity (especially at high-prestige institutions) and pedagogical skill (especially at teaching institutions). Thus, this theory—a credential acquisition model-- would predict that candidates will invest in career-relevant education and experience. In the institutional ascription or “queuing” model, candidates are placed in a job queue based largely on the prestige or quality of their graduate institution.60 These candidates may otherwise be equally qualified for the job, or the prestige graduate may be less qualified as measured by pre-doctorate productivity.61 The dominant variable for

57 Lee Epstein & Gary King, The Rules of Inference, 69 U. CHI. L. REV. 1 (2002) (discussing this and much broader set of issues related to the construction and interpretation of empirical research).

58Ted I.K. Youn, Studies of Academic Markets and Careers: An Historical Review, in ACADEMIC LABOR MARKETS AND

CAREERS 8 (David W. Breneman & Ted I.K. Young eds. 1988) (reviewing and evaluating the study of labor markets for professors).

59GARY S. BECKER, HUMAN CAPITAL (1964); Orly Ashenfelter & Joseph D. Mooney, Graduate Education, Ability and Earnings, 49 REV. ECON. & STAT. (1968); W. Lee Hansen, The Economics of Scientific & Engineering Manpower, II

J. HUM. RES. 191 (Spring 1967).

60 Joseph R. Stiglitz, The Theory of Screening, Education and the Distribution of Income, 65 AM. ECON. REV. 315 (1975) (finding that job prospects are determined by relative prestige of Ph.D.-granting institutions); BERNARD BERELSON, GRADUATE EDUCATION IN AMERICA (1960) (this early study of stratification in academia found a strong positive correlation between prestige of degree-granting department and first job department in the hard sciences); THEODORE CAPLOW & REESE MCGEE, THE ACADEMIC MARKETPLACE (1958); JONATHAN R. COLE &

STEPHEN COLE, SOCIAL STRATIFICATION IN SCIENCE (1973) (confirming and elaborating the early work of scholars like Berelson and Allison and Caplow).

61 J.R. Niland, Allocation of Ph.D. Manpower in the Academic Labor Market, 2 INDUS. REL. 141 (1972); E.P. Lazear & S. Rosen, Rank-order Tournaments as Optimum Labor Contracts, 89 J. POL. ECON. 135 (1981), Diana Crane, The Academic Marketplace Revisited, 7 AM. J. SOC. 953 (1970) (sociologist who concluded that despite a commitment to universalism, science department hiring is explained more readily by graduate department prestige than by scholarly productivity).

Page 17: Vanderbilt University Law School - The Wall Street Journalonline.wsj.com/public/resources/documents/lawschoolhiring.pdf · Vanderbilt University Law School . ... Understanding who

16 The Labor Market for New Law Professors

explaining initial placement is prestige of doctorate-granting institution.62 Allan Cartter, who published some of the earliest studies of faculty hiring in colleges and universities, concluded that “institutions tend to hire from the same array of graduate schools.”63 He hypothesized, based on studying the subject for nearly two decades, that this consistent hiring dynamic is self-reinforcing as current faculty hire from their alma maters and inter-institutional ties develop. We build a model based on hypotheses derived from theories of labor market success and informed by the personal observations and narratives published by numerous law professors. The variables in the model generally find support in existing empirical studies of the composition of law faculties, including the characteristics which distinguish faculty at higher ranked schools from those at lower ranked schools. We specifically consider education, professional experience, scholarship, teaching, and demographics. A. Educational Background We expect law schools will prefer to hire law graduates, especially those trained in the U.S., and we further hypothesize that all law schools will prefer to hire graduates from higher ranked law schools (both relative to the hiring school and to all schools generally). The majority of law professors have traditionally graduated from a small number of law schools.64 Most advice to prospective law professors emphasizes the benefits of an elite J.D. (and discusses the challenges posed by a degree from a lower ranked institution, based on the U.S. News and World Report law school rankings). And, studies of new law professors have found that the prestige of a professor’s J.D. institution was the best predictor of the

62 Ted I.K. Youn, Studies of Academic Markets and Careers: An Historical Review, in ACADEMIC LABOR MARKETS

AND CAREERS 8, 17 (David W. Breneman & Ted I.K. Young eds. 1988) (concluding based on a review of existing economic and sociological studies that “[t]he prestige of first career position is determined largely by the prestige of the graduate institution”); Robert McGinnis & J. Scott Long, Entry into Academia: Effects of Stratification, Geography and Ecology, in ACADEMIC LABOR MARKETS AND CAREERS (David W. Breneman & Ted I.K. Young eds. 1988) (finding that institutional or mentor prestige determines entry-level academic posts for biocchemists, leading to long-term stratification in academia due to differing reward structures); see also J. Scott Long, P.D. Allison, & Robert McGinnis, Entrance Into the Academic Career, 44 AM SOC. REV. 889 (1979) (finding that biochemistry academic positions are allocated according to Ph.D. institution prestige rather than pre-doctoral productivity, especially with respect to first position); Barbara Reskin, Academic Sponsorship and Scientists Careers, 52 SOC. EDUC. 129 (1979) (concluding that sponsorship by a prominent scholar matters more than personal achievement in a scientist’s career).

63 ALLAN M. CARTTER, PH.D.S AND THE ACADEMIC LABOR MARKET (1976); see also Allan M. Carter The Supply and Demand of College Teachers, 1965 AM STATISTICAL ASS’N SOCIAL STATISTICS PROCEEDINGS 70.

64 See, e.g., Bothwick & Schau, supra note 28, at 191 (finding, based on a random sample of full-time law professors in the 1988-1989 AALS Director of Law Teachers, that more than half graduated from the top 20 law schools); Fossum, supra note 28, at 507 (finding, based an analysis of all full-time law professors listed in the 1975-1976 AALS Director of Law Teachers, that the majority came from 20 “feeder” or “producer” schools).

Page 18: Vanderbilt University Law School - The Wall Street Journalonline.wsj.com/public/resources/documents/lawschoolhiring.pdf · Vanderbilt University Law School . ... Understanding who

The Labor Market for New Law Professors 17

relative ranking of a professor’s teaching institution.65 While these studies do not consider the relationship between J.D. prestige and success in the market, the rank of a candidate’s law school clearly plays some role in the hiring process. The queuing model hypothesizes that institutional ascription—the assumption that a candidate from a higher ranked school is a stronger candidate—will dominate other variables in predicting placement. Empirical research has tested the importance of graduate institution relative to scholarly productivity and institutional prestige in academic careers. Studies of entry-level job placement find that prestige of doctoral department is far more important than scholarly productivity to prestige of first job.66 This is true even in disciplines such as the hard sciences and math, marked by a greater consensus over the standards for evaluating work which should allow for merit-based selection and eliminate the need to rely on proxies for ability (such as institutional prestige).67 If high-consensus disciplines are disproportionately influenced by prestige of school, then low -consensus disciplines like social sciences and law should certainly show an effect.68 Low-consensus disciplines are those in which scholars actively debate what research questions are valuable, what methodologies are appropriate, and where scholars should publish their work.69 65 See, e.g., sources cited in note 28 supra.

66 Sociologists Long, Allison and McGinnis, for example, found that Ph.D. prestige was the most important explanatory variable and had a .28 regression coefficient as compared to -.02 for pre-employment publications (and citations), controlling for mentor’s reputation, undergraduate institution selectivity, and graduate program size. J. Scott Long, Paul D. Allison & Robert McGinnis, Entrance into the Academic Career, 44 AM. SOCIO. REV. 816 (1979). These sociologists, who have done much work in the area, concluded that “neither the quantity nor the ‘quality’ of one’s early publications has a significant influence on where one ends up in the prestige hierarchy” in biochemistry. Id. at 820.These studies focus primarily on elite research institutions (Ph.D. granting), and may not tell us much about hiring beyond the first tier of American law schools where teaching (and, in law schools, service to the bench and bar) grow in importance relative to research.

67 For a discussion of the high consensus/low consensus distinction and its implications for studies of academic labor markets, see Lowell L. Hargens & Warren Hagstrom, Scientific Consensus and Academic Status Attainment Patterns, 55 SOC. EDUC. 183 (1982).

68 See, e.g., Richard Carson & Peter Navarro, A Seller’s (& Buyer’s) Guide to the Job Market for Beginning Academic Economists, 2 J. ECON. PERSP. 137 (1988) (finding that economics departments, in response to a survey, claimed that many factors matter to the decision to interview and the evaluation of the interview itself); but see Wendy A. Stock, Richard M. Alston, & Martin Milkman, The Academic Labor Market for Economists – 1995-96, 28 ATL. ECON. J. 164 (2000) (finding in a later survey of economics department that school affiliation was very important to decision to interview and was weighted as more important than particularized characteristics).

69 See Stephane Baldi, Prestige Determinants of First Academic Job for New Sociology Ph.D.s 1985-1992, SOCIO. Q. 777, 778 (1995) (“Fields characterized by specific, standardized procedures for evaluation of contributions (such as the hard sciences) should put greater emphasis on individual performance for employment in an academic department than fields that lack standardized procedures”); RICHARD WHITLEY, THE INTELLECTUAL AND

SOCIAL ORGANIZATION OF THE SCIENCES 164-205 (1984).

Page 19: Vanderbilt University Law School - The Wall Street Journalonline.wsj.com/public/resources/documents/lawschoolhiring.pdf · Vanderbilt University Law School . ... Understanding who

18 The Labor Market for New Law Professors

Although the queuing model emphasizes the role of institutional prestige rather than individual merit, it does not necessarily follow that hiring is indifferent to merit. Indeed, the two are closely correlated: scholars who adhere to this model hypothesize that high-ability students will choose more prestigious institutions because of the greater job opportunities that these schools afford. Higher ranked schools also may provide a stronger network of potential collaborators—both faculty and classmates—which may be more important as law becomes more collaborative. But, institutional prestige is at best a proxy for quality whereas pre-employment publications would be a direct measure. Nevertheless, the queuing model supports the hypothesis that the relative ranking of a candidate’s law school will be the single most important factor in predicting law job market success, beyond more direct measures of potential such as publications and teaching. Academic achievement in law school also is a likely predictor of job market success. If law schools recognize classroom success in similar ways, then we would expect such success to be a signal to the market of a candidate’s abilities. Under the credential acquisition model, aspiring law professors will invest in doing well in law school, as reflected in class rank or graduation honors, because this will signal a skill at legal analysis and discourse which distinguish law school teaching and research. Older studies have found that law school honors such as membership in the law school honorary Order of the Coif or on the law review were statistically significantly correlated with the status of the first job of new law professors.70 Today, law school success is much harder to measure. Elite law schools are less likely than other schools to recognize relative academic performance in a readily observable way. Yale, Harvard, and Stanford, for example, have moved to pass-fail grading.71 For those elite law schools which have retained letter grades, they often do not report class rank. In a related move, law schools such as Columbia, Harvard, and Stanford closed their Order of the Coif chapters.72 Law review selection at most schools has moved away from a grade-on-only model, and the proliferation of specialized law journals has made it difficult to assess the value of journal membership. In sum, academic success has become difficult to measure directly. Another recent trend has been a significant increase in the number of candidates who possess both a J.D. and a doctorate in a non-law discipline. From a human capital perspective, a doctorate demonstrates both a capacity to research and write as well as a

70 See, e.g., Fossum, supra note 28, at 509, 517 (reporting that half of law professors in 1975-1976 were either members of law review and/or Order of the Coif).

71 See, e.g., Brian Leiter’s Law School Reports, Harvard Law School to Adopt Pass-Fail Grading System Like Yale and Stanford, http://leiterlawschool.typepad.com/leiter/2008/09/harvard-law-sch.html.

72 See National Order of the Coif, http://www.orderofthecoif.org/.

Page 20: Vanderbilt University Law School - The Wall Street Journalonline.wsj.com/public/resources/documents/lawschoolhiring.pdf · Vanderbilt University Law School . ... Understanding who

The Labor Market for New Law Professors 19

body of work (the dissertation).73 Moreover, a Ph.D. is strong evidence of an interest in and commitment to academic employment whereas a J.D. is not. The number of professors with doctoral degrees has grown, and studies have found that they are associated with the employing law school’s ranking.74 Whether advanced degrees other than a doctorate are a strong signal is less clear. B. Professional Experience

Unlike faculty in other departments, law school faculty generally have worked outside of a university setting between graduate school and starting on the tenure track. This should come as no surprise: Law schools train lawyers, and we would expect them to look for candidates with legal practice experience, so that faculty can teach based on their own professional experiences and facilitate the placement of graduates in elite jobs.75 Hence, we hypothesize that candidates with practice experience will be more likely to be interviewed and hired. While an economic perspective would emphasize the potential returns to the school from hiring someone with such credentials, a sociological view would emphasize instead the signaling or institutional ascription value of elite practice experience. Law professors have long been drawn from the elite segment of the bar—top firms or competitive government posts. Current law professors would undoubtedly look to those same positions for new hires, assuming that those hired by the most selective employers must be the most desirable new lawyers. A judicial clerkship is often cited as a valuable credential for prospective professors. Clerking for a judge—especially a federal or state appellate judge—involves substantial research and writing, which would demonstrate a candidate’s capacity for this type of work. Given the competitiveness of the clerkship application process, a clerkship could

73 See, e.g., Thomas F. Bergin, The Law Teacher: A Man Divided Against Himself, 54 VA. L. REV. 637 (1968) (arguing that a Ph.D. would be better training for law professors because of the requirement that law professors be both scholars and teachers).

74 See, e.g., Joni Hersch & W. Kip Viscusi, Law and Economics as Pillar of Legal Education, Vanderbilt Law and Economics Research Paper No. 11-35, http://ssrn.com/abstract=1907760, at 3-4, 21-23 Table 1 Panels A-C (reporting on the number of professors with doctoral degrees at the top 26 law schools and finding a correlation between ranking and number of Ph.D.s on faculty); Tracey E. George, An Empirical Study of Empirical Legal Scholarship: The Top Law Schools, 81 IND. L.J. 141, 151-152 (reporting on the significant number of professors on elite law faculty with social science doctorates).

75 State bar membership, the formal mark of being a lawyer, is common for entry-level law teaching applicants. Almost every applicant—92%--had been admitted to a bar. And, those who were not admitted usually have a reason, such as a foreign law degree or no law degree. The lack of an American law degree should have independent effect on job prospects. Thus, we did not include bar membership as a variable because it was not well-supported theoretically as a proxy for legal experience and was unhelpful empirically in explaining the results.

Page 21: Vanderbilt University Law School - The Wall Street Journalonline.wsj.com/public/resources/documents/lawschoolhiring.pdf · Vanderbilt University Law School . ... Understanding who

20 The Labor Market for New Law Professors

also be used as indicator of academic ability.76 Many law professors, especially on top-20 law faculties, clerked before entering the academy, bolstering the view that a judicial clerkship could increase one’s chances of getting a teaching job.77 C. Scholarship and Teaching Law professors primarily engage in scholarship and teaching. Thus, we would expect that the best candidates would invest in gaining those credentials, and that law schools would focus on those credentials in their hiring decisions. Actual publications are better evidence of scholarly capacity than clerkships and good grades, which are simply proxies. Thus, we would expect a record of publications to be one of the most important predictors of academic market success.78 Holding a fellowship or visiting assistant professorship provides an opportunity for teaching and writing. Candidates with this experience should be at a comparative advantage over other candidates. Finally, in both endeavors, schools will consider subject matter of need. Academic jobs are typically posted by subject area, and schools claim to hire, at least in part, to fill needs. Candidates who are in high-demand fields should have a comparative advantage. D. Demographic Characteristics We control for demographic characteristics, such as race, sex, and years since graduation, which legal scholars believe influence faculty hiring as well. Scholars do not agree, however, about the direction of the effect. Some scholars have argued that women and minorities have an edge in academic employment at prestigious universities because of affirmative action programs.79 Law school hiring, too, has been the subject of great debate on this topic. Strong arguments have been made on both sides, but what little we know focuses only on relative placement of new hires. Merritt and Reskin found a very modest effect among new law professors: white women and men of color were likely to teach at slightly more prestigious schools than white men with comparable credentials.80 No one has studied the interview and job offer stages in law teaching hiring, but evidence from

76 See note infra.

77 See Bothwick & Schau, supra note 28, at 214 (finding that the percentage of faculty with clerkship experience had doubled since Fossum’s study); Fossum, supra note 28, at 511 (reporting that 16.6% of all law professors in her sample had clerked for a judge), 518 (finding that former clerks started at more elite law schools than other new professors who came directly from law school or law practice).

78 See, e.g., Ethan S. Burger & Douglas R. Richmond, The Future of Law School Faculty Hiring in Light of Smith v. City of Jackson, 13 VA. J. SOC. POL'Y & L. 1, 35 (2005).

79 See, e.g., ALLAN BLOOM, THE CLOSING OF THE AMERICAN MIND: ON RACE, BLACK POWER, BLACK STUDIES AND

AFFIRMATIVE ACTION (1987) (presenting perhaps the most famous such account).

80 Merritt & Reskin, supra note 2, at 248 Table 5, 274.

Page 22: Vanderbilt University Law School - The Wall Street Journalonline.wsj.com/public/resources/documents/lawschoolhiring.pdf · Vanderbilt University Law School . ... Understanding who

The Labor Market for New Law Professors 21

similar fields does not demonstrate a benefit or disadvantage as academic employers cull candidates.81 IV. RESULTS The analysis of our data involves a consideration of each step in the law teaching labor market. But, we begin with summary statistics to provide a picture of all candidates on the market in 2007-2008 and show how our survey respondents compare. We then use multivariate statistics to test our hypotheses at each stage of the hiring process, testing different versions of the model to focus on specific relationships and control for specific effects. A. Summary Statistics We begin by describing our data along the four dimensions discussed above: demographics, education, professional experience, and scholarship and teaching credentials. For each table, we report the characteristics of the survey respondents compared to the benchmark of the total 2007-2007 tenure-track law professor applicant population. Table 1 reports the demographics of 2007-2008 applicants for tenure-track law professorships. Survey respondents, like all candidates, were disproportionately men at a rate of nearly two to one. Whites are also more likely than non-whites to enter the law teaching market. The difference between the percentage of whites in the survey sample (76%) and the population of all candidates (70%) is likely due to the fact that seven percent of candidates leave the race question blank on the AALS FAR form. (By contrast, our data includes information on the race/ethnicity for all respondents.) Finally, half of all applicants graduated within the last nine (9) years and half of all respondents graduated within the last eight (8) years. Years since graduation can tell us whether there is an optimal time to go on the law teaching job market. Conventional wisdom suggests that law graduates benefit from educational and work experiences following law school but at some point may find it more difficult to enter legal academia.

81 Studies of other disciplines such as Economics and Sociology have not found an effect. See, e.g., Baldi, supra note 69, at 785; Debra A. Barbezat, The Market for New Ph.D. Economists, 23 J. ECON. ED. 262, 268-269 (1992).

Page 23: Vanderbilt University Law School - The Wall Street Journalonline.wsj.com/public/resources/documents/lawschoolhiring.pdf · Vanderbilt University Law School . ... Understanding who

22 The Labor Market for New Law Professors

Table 1. Demographics of Entry-Level Law Professor Applicants: 2007-2008

Survey Respondents All Candidates

Gender

Male 64% 64%

Female 36% 36%

Race/Ethnic Origin

White 76% 70%

Black 9% 9%

Asian 5% 5%

Hispanic 4% 4%

Native American 2% 1%

Other 4% 5%

Not Reported* 0% 7%

Years Since Law School Graduation**

75th percentile 5 5

Median 8 9

25th percentile 15 16

N 466 911

“Survey Respondents” completed the George & Yoon survey. “All Candidates” includes all AALS faculty appointments registrants and all new tenure-track hires who did not participate in the AALS process. “All Candidates” does not include individuals who failed to obtain jobs after applying directly to schools without participating in the AALS process. *Survey respondents were more likely to disclose race than were AALS registrants. **The variable “Years Since Law School Graduation” only includes law school graduates.

With respect to gender and race, the applicant pool appears to differ from the general population of law graduates. Historically, women and minorities are underrepresented among tenured or tenure-track research faculty.82 In our data, women also appear to be underrepresented in the law teaching market. Since 1997, women have comprised at least 45% of matriculated law schools students.83 In the survey and the AALS FAR process, they comprise 36% of applicants. By contrast, minority applicants appear well represented or

82 See, e.g., Richard H. Chused, The Hiring and Retention of Minorities and Women on American Law School Faculties 137 U. PENN. L. REV. 537 (1988).

83 The American Bar Association reports historical enrollment of men and women. See http://www.abanet.org/legaled/statistics/charts/stats%20-%206.pdf.

Page 24: Vanderbilt University Law School - The Wall Street Journalonline.wsj.com/public/resources/documents/lawschoolhiring.pdf · Vanderbilt University Law School . ... Understanding who

The Labor Market for New Law Professors 23

slightly overrepresented. Non-white student enrollment in law schools has been above 20% since 1995; non-whites represent 23-30% of the AALS applicant pool and 24% of survey respondents.

Table 2. Educational Background of Entry-Level Law Professor Applicants: 2007-2008

Survey Respondents All Candidates

Law School Attended*

Tier 1 67% 62%

Tier 2 16% 15%

Tier 3 7% 9%

Tier 4 4% 5%

Foreign Law School/Unranked 5% 7%

No law degree 1% 2%

Class Rank

Top 1% 4% 5%

Top 5% 13% 13%

Top 10% 14% 15%

Top 25% 5% 7%

Top 50 9% 8%

Bottom 50% 0% 0%

Not reported 54% 52%

Non-Law Ph.D.**

Humanities 5% 5%

Social Sciences 7% 5%

STEM 2% 1%

Other 1% 1%

N 466 911

“Survey Respondents” completed the George & Yoon survey. “All Candidates” includes all AALS faculty appointments registrants and all new tenure-track hires who did not participate in the AALS process. “All Candidates” does not include individuals who failed to obtain jobs after applying directly to schools without participating in the AALS process. * Law school tier is based on U.S. News & World Report 2007 rankings. **Non-law Ph.D. only includes Doctorates of Philosophy awarded in non-law subjects. STEM is science, technology, engineering, and mathematics.

While law schools do not formally require a law degree to join the faculty, it remains a widely shared credential. Table 2 shows that 98% of applicants graduated from an

Page 25: Vanderbilt University Law School - The Wall Street Journalonline.wsj.com/public/resources/documents/lawschoolhiring.pdf · Vanderbilt University Law School . ... Understanding who

24 The Labor Market for New Law Professors

American or foreign law school, and the majority attended a law school ranked among the top fifty (Tier 1) by the U.S. News and World Report law school rankings for 2007. More than three-quarters of applicants attended a law school ranked in the top 100 (Tier 1 or Tier 2). Survey respondents were more likely than non-respondents to have graduated from a Tier 1 law school (67% compared to 62%). We expected that academic success would be hard to measure directly, and our data bear that out. More than half of the candidates in our study did not report class rank. And candidates without a class ranking were disproportionately likely to have attended Tier 1 law schools. Thus, we do not include an academic success variable in our study because of the missing data problem. While research faculty in the arts and sciences typically possess Ph.D.s, law professors are unlikely to have earned a doctorate. Studies and stories suggest an increasing number of tenure-track law professors and job candidates have non-law Ph.D.s. While a single year of data does not allow us to observe trends over time, we note that a small but sizable fraction of the total applicant pool hold a non-law Ph.D. (Nearly all candidates with non-law doctorates also graduated from law school.) Applicants with doctorates were fairly evenly divided between the humanities and social sciences with a small number in STEM (science, technology, engineering, and mathematics) and other fields. Survey respondents were more likely than non-respondents to hold a Ph.D. in the social sciences or STEM.

Table 3. Professional Experience of Entry-Level Law Professor Applicants: 2007-2008

Survey Respondents All Candidates

Current Employment ( Fall 2007)

Private Practice 32% 35%

Government (non-clerkship) 10% 10%

Judicial Clerkship 4% 4%

Non-Profit/NGO 6% 7%

Law School Fellowship/Teaching 23% 21%

Other Fellowship/Teaching 9% 6%

Other Employment 14% 14%

Not reported 2% 2%

Judicial Clerkship (Post-Graduate)* 40% 35%

N 466 911

“Survey Respondents” completed the George & Yoon survey. “All Candidates” includes all AALS faculty appointments registrants and all new tenure-track hires who did not participate in the AALS process. “All Candidates” does not include individuals who failed to obtain jobs after applying directly to schools without participating in the AALS process. *Judicial clerkship includes any post-graduate clerkship with a judge.

Page 26: Vanderbilt University Law School - The Wall Street Journalonline.wsj.com/public/resources/documents/lawschoolhiring.pdf · Vanderbilt University Law School . ... Understanding who

The Labor Market for New Law Professors 25

Nearly all applicants reported that they were employed at the time they entered the law teaching market (fall of 2007), as noted in Table 3. Roughly one-third of respondents (and a slightly higher percentage of non-respondents) were working in private practice, predominantly in large law firms and disproportionately in cities with large legal markets. Nearly a quarter of respondents (and slightly fewer non-respondents) were already employed by law schools, working as teaching fellows, visiting assistant professors, and instructors. While this might appear to be a surprisingly high proportion, those positions often are promoted as a launching pad for tenure-track jobs. Most government employees, who comprise 10% of the population and sample, were working for the federal government. A judicial clerkship is often included on checklists for aspiring law professors. We find that while few applicants are judicial clerks at the time they apply for law teaching jobs, many previously worked for a judge. A higher percentage of respondents than non-respondents had this experience (40% compared to 31%). Many applicants have published writings as seen in Table 4. The variety of work is staggering, including editorials, blog postings, elite law review articles, magazine articles, university press books, and practice manuals. We focused principally on the type of writing emphasized in standard tenure requirements: scholarly articles published in American legal academic journals. (We excluded student notes and comments.). We report our findings categorized by Washington & Lee University Law Library’s ranking of law journals in 2007.84 If a candidate has more than one publication, we recorded the highest ranked article.85 Table 4 reports our findings dividing ranked journals into three tiers (top 100, 100-300, and below 300). We also report the percentage with other publications, including articles in unranked law reviews, non-law journals, and foreign periodicals as well as book chapters. Our final category is books which includes academic press and commercial press monographs as well as casebooks and practitioner guides (but does not include edited collections). Each candidate is listed only once in this category.

84 See Washington & Lee University School of Law, Law Library, “Law Journals: Submissions and Ranking, 2005-2012, available at http://lawlib.wlu.edu/LJ/. Under journal criteria, we selected “all subjects” in the United States. We coded the normalized “Combined Score” which “is a composite of each journal’s impact factor and total cites count.” We used the default setting which weights impact factor one-third and citations two-thirds. http://lawlib.wlu.edu/LJ/method.asp#methodology.

85 Candidates are listed only once in the “Scholarly Publication” section of Table 4. If a candidate/respondent published more than one work of the types listed, we included the highest ranked law journal article.

Page 27: Vanderbilt University Law School - The Wall Street Journalonline.wsj.com/public/resources/documents/lawschoolhiring.pdf · Vanderbilt University Law School . ... Understanding who

26 The Labor Market for New Law Professors

Table 4. Scholarship and Teaching of Entry-Level Law Professor Applicants: 2007-2008

Survey Respondents All Candidates

Scholarly Publication*

Law Journal Ranked 1-100 17% 13%

Law Journal Ranked 101-300 21% 18%

Law Journal Ranked below 300 14% 14%

Book 5% 4%

Other Publications 11% 13%

Preferred Teaching Subject (Top 10)**

Criminal Law/Procedure 8% 7%

Constitutional Law 6% 5%

Civil Procedure 6% 5%

Comparative/International 6% 7%

Intellectual Property 5% 6%

Tax 5% 6%

Corporate/Securities 5% 5%

Not reported 5% 5%

Contracts 5% 5%

Clinical 4% 4%

N 466 911

“Survey Respondents” completed the George & Yoon survey. “All Candidates” includes all AALS faculty appointments registrants and all new tenure-track hires who did not participate in the AALS process. “All Candidates” does not include individuals who failed to obtain jobs after applying directly to schools without participating in the AALS process. *The scholarly publication variable reflects the highest ranked legal academic journal in which a candidate has published. Rankings are based on the Washington & Lee Law Journal Composite Rankings for 2007. “Other publications” includes unranked journals, non-law journals, book chapters, and foreign journals. A candidate is listed only once based on highest ranked publication. **Preferred Teaching Subject is based on the candidates’ list of topics as “preferred” after combining related subjects. We report only the top 10, in descending order of frequency of mention by survey respondents.

Many applicants have published writings as seen in Table 4. The variety of work is staggering, including editorials, blog postings, elite law review articles, magazine articles, university press books, and practice manuals. We focused principally on the type of writing emphasized in standard tenure requirements: scholarly articles published in

Page 28: Vanderbilt University Law School - The Wall Street Journalonline.wsj.com/public/resources/documents/lawschoolhiring.pdf · Vanderbilt University Law School . ... Understanding who

The Labor Market for New Law Professors 27

American legal academic journals. (We excluded student notes and comments.). We report our findings categorized by Washington & Lee University Law Library’s ranking of law journals in 2007.86 If a candidate has more than one publication, we recorded the highest ranked article.87 Table 4 reports our findings dividing ranked journals into three tiers (top 100, 100-300, and below 300). We also report the percentage with other publications, including articles in unranked law reviews, non-law journals, and foreign periodicals as well as book chapters. Our final category is books which includes academic press and commercial press monographs as well as casebooks and practitioner guides (but does not include edited collections). Each candidate is listed only once in this category. Table 4 also reports the applicant’s substantive areas of teaching interest. Each applicant could list up to four subjects that he or she would “prefer” to teach. Thus, the percentages exceed 100%. Traditional first-year and other core subjects are relatively well-represented. Topics which are “hot” or have gained currency are also among the top 10. Overall, no clear patterns emerge, in part because candidates were given over 80 subjects from which to choose. In table 4, we combine related categories which remain separate in the multivariate model. The descriptive statistics reveal that survey respondents are, in general, typical of the population of applicants. They are white males who attended Tier-1 law schools and now work in private practice or in a law school. However, we find a modest selection effect. Respondents are more likely than non-respondents to have attended a top-50 law school, to hold a non-law doctorate in the social sciences or STEM, to have clerked for a judge, and to have published an article in a top-300 law journal. That is, respondents were somewhat more likely to have earned those credentials which we hypothesize would improve their chances of getting a law teaching job. To the extent that weaker candidates were less likely to complete our survey, we may underestimate the size of the effect of these credentials on success. But, the response-rate difference is so small that we would still expect to find a statistically significant relationship if one exists in the population as a whole. B. The Dynamics of Interview Invitations Candidates usually must be invited for two interviews to be considered for a job offer: an initial screening interview and a subsequent call-back interview at the law school. We look at the characteristics of those selected at each stage and then test which factors account for

86 See Washington & Lee University School of Law, Law Library, “Law Journals: Submissions and Ranking, 2005-2012, available at http://lawlib.wlu.edu/LJ/. Under journal criteria, we selected “all subjects” in the United States. We coded the normalized “Combined Score” which “is a composite of each journal’s impact factor and total cites count.” We used the default setting which weights impact factor one-third and citations two-thirds. http://lawlib.wlu.edu/LJ/method.asp#methodology.

87 Candidates are listed only once in the “Scholarly Publication” section of Table 4. If a candidate/respondent published more than one work of the types listed, we included the highest ranked law journal article.

Page 29: Vanderbilt University Law School - The Wall Street Journalonline.wsj.com/public/resources/documents/lawschoolhiring.pdf · Vanderbilt University Law School . ... Understanding who

28 The Labor Market for New Law Professors

success at each step. Our analysis of the interviewing process is based on the 466 applicants who completed our survey. Table 5. Interview Invitations for All Survey Respondents

(1) Received

Tenure-Track Offer

(2) Did Not Receive

Tenure-Track Offer

(3) All Respondents

Initial Interviews

Number of Invitations 16.46 (13.900)

4.59 (6.049)

8.90 (11.210)

Number Accepted 12.68 (9.480)

3.59 (4.829)

6.88 (8.153)

On-Campus Job Talks

Number of Invitations 3.12 (2.420)

0.39 (0.898)

1.38 (2.087)

Number Accepted 2.93 (2.192)

0.37 (0.887)

1.30 (1.938)

Tenure-Track Job Offers 1.52 (1.435)

--- 0.55 (1.131)

N 169 297 466

The data is based on responses to the survey. Averages are reported for each category with standard deviations in parentheses below. Column 1 includes 15 respondents who received tenure-track offers but declined the offer(s).

Table 5 reports the average number of invitations received by each survey respondent at each stage of the hiring process: initial screening interview, on-campus “job-talk” interview, and tenure-track offer. We also report the average number of each which were accepted. For example, the first entry for “initial interviews” is the average number of screening interview invitations per candidate and the second entry is the average number accepted.88 We report these numbers for all 466 candidates and also break out the data conditional on whether the respondent received at least one tenure-track job offer (169 candidates) or not (297). Unsurprisingly, the experiences of candidates who received tenure-track offers differed from those who did not. In the screening interview stage, applicants who were ultimately hired (Column 1) received an average of 16.5 invitations for screening interviews. These applicants accepted most (12.7), but not all, of the interviews. At the next stage, they received an average of 3.1 job talk invitations, and accepted on average 2.9

88 Most screening interviews occurred at the AALS Faculty Recruitment Conference, but some occurred at the school itself. An on-campus interview was coded as only an “initial” or preliminary interview if it did not include a job talk.

Page 30: Vanderbilt University Law School - The Wall Street Journalonline.wsj.com/public/resources/documents/lawschoolhiring.pdf · Vanderbilt University Law School . ... Understanding who

The Labor Market for New Law Professors 29

of them. At the final stage, they received an average of 1.5 offers.89 Among those who received offers, 64% received only a single offer. By comparison, applicants who did not receive a tenure-track offer (Column 2) were less likely to receive initial interview invitations and less likely to receive on-campus job-talk invitations. They averaged fewer than 5 invitations for screening interviews and fewer than 0.4 job-talk invitations. This evidence reinforces the significance of the early stages in determining the opportunities available in the final stage. Law schools appear to focus on many of the same candidates at the interview stages of the market. These statistics reflect the likely selection effect in the survey responses in favor of candidates who were more successful on the market. Nearly 85% of respondents reported having at least one screening interview. While we cannot be sure what percentage of the total applicant pool had screening interviews, we strongly suspect that it is significantly lower, suggesting that the number in our response group is biased upward. It is also possible that even though we had information on the number of interviews from 77% (169/220) of participants who ultimately took a tenure-track job, these respondents may not be representative of all candidates who were ultimately hired. For example, we had several respondents who reported over 30 initial invitations for screening interviews, a remarkable number. These responses nonetheless provide insight into the dynamic selection process that occurs for both candidates and schools, particularly the more successful ones. 1. Screening Interviews Table 6 reports different model specifications where the outcome variable of interest is whether the candidate received screening interviews. We run a probit model, with the coefficients reporting marginal effects. For example, in Column 2, a candidate who had published in a top-100 law journal at the time he or she went on the market was 8% more likely to have received a screening interview, holding other factors constant, than candidates who had not.

89 Respondents in Column 1 include 15 candidates who received tenure-track offers but declined them. This group had, on average, fewer interviews, job talks, and offers than applicants who ultimately accepted tenure-track offers.

Page 31: Vanderbilt University Law School - The Wall Street Journalonline.wsj.com/public/resources/documents/lawschoolhiring.pdf · Vanderbilt University Law School . ... Understanding who

30 The Labor Market for New Law Professors

Table 6. Probit Model of the Probability of Receiving a Screening Interview

(1) (2) (3) (4) (5) (6)

Female 0.0309 0.0304 0.0171 0.0306 0.0164 0.0204

(0.0327) (0.0323) (0.0318) (0.0321) (0.0320) (0.0316)

Nonwhite 0.000584 0.00321 -0.00983 0.00130 -0.0126 -0.00983

(0.0375) (0.0369) (0.0368) (0.0367) (0.0372) (0.0365)

Graduated Law School ≤ 10 years 0.126*** 0.0969*** 0.0898**

0.0836** 0.0840**

(0.0352) (0.0357) (0.0351)

(0.0352) (0.0349)

Judicial Clerkship

0.0484 0.0428 0.0641** 0.0376 0.0363

(0.0335) (0.0325) (0.0317) (0.0326) (0.0329)

Published Scholarship

Top 100 Ranked Law Journals

0.0831** 0.0778** 0.0820** 0.0714** 0.0727**

(0.0354) (0.0334) (0.0332) (0.0347) (0.0343)

101-300 Ranked Law Journal

0.0235 0.0175 0.0260 0.0156 0.0154

(0.0377) (0.0372) (0.0371) (0.0374) (0.0372)

Non-Law Ph.D.

Social Science/STEM

-0.0227 -0.0106 -0.0315 -0.0121 -0.0126

(0.0589) (0.05529) (0.0581) (0.0559) (0.0553)

Humanities/Other 0.0810* 0.0782* 0.0855** 0.0773* 0.0774**

(0.0455) (0.0411) (0.0414) (0.0418) (0.0407)

Current Employment

Law Teaching/Fellowship

0.0988*** 0.116*** 0.0966*** 0.0965***

(0.0336) (0.0326) (0.0338) (0.0336)

Other Teaching/Fellowship

-0.0353 -0.0469 -0.0290 -0.0287

(0.0622) (0.0651) (0.0610) (0.0604)

Government (non-clerkship)

0.0786** 0.0834** 0.0739** 0.0774**

(0.0352) (0.0368) (0.0362) (0.0350)

Private Practice

-0.0288 -0.0210 -0.0308 -0.0239

(0.0404) (0.0398) (0.0405) (0.0399)

Law School Attended

Tier 2

-0.0737

(0.0492)

Tier 3

-0.0137

(0.0603)

Tier 4

-0.0948

(0.00793)

Foreign Law School

-0.0343

(0.0777)

Top 3 (Yale, Harvard, Stanford)

0.0523

(0.0394)

N

455 455 455 466 455 455

Dependent variable is whether candidate received one or more initial screening interviews. Probit model reports marginal effects with standard errors are reported in parentheses. Models 1-3 and 5 include all survey

Page 32: Vanderbilt University Law School - The Wall Street Journalonline.wsj.com/public/resources/documents/lawschoolhiring.pdf · Vanderbilt University Law School . ... Understanding who

The Labor Market for New Law Professors 31

respondents who hold a law school degree. Model 4 also includes 28 respondents who did not graduate from law school. (All had Ph.D. degrees.) For “Published Scholarship,” the baseline (omitted) category includes any publications which are not in a top-300 law journal as well as no publications. For “Non-Law Ph.D.”, the baseline (omitted) category is no non-law Ph.D. For employment, the baseline (omitted) category includes any type of employment not listed or no employment. ***p<0.01, **p<0.05, *p<0.10.

With respect to demographic characteristics, neither gender nor race is statistically significant. Because of the small numbers of individuals in the separate race/ethnicity categories, we used a single “nonwhite” variable which is equal to one if a candidate chooses any ethnicity other than White. The number of years since a candidate’s law school graduation, however, is a statistically significant factor and has a meaningful marginal effect. We found that applicants who had graduated within the past ten years (between 1997 and 2007) were over 8-12% more likely to get hired than those who had graduated before 1997.90 A top-100 publication or a Ph.D. in a non-quantitative subject increased by approximately eight percent a candidate’s prospects for a screening invitation. By contrast, publications in lower ranked journals did not have a statistically significant effect.91 (The model does not measure the effect of books due to problems inherent to the measure.92) And, not all non-law doctorates are positively related to screening interview invitations:

90 Years since graduation should have a curvilinear relationship to job market success: additional years of experience would be positively associated with job market success until some point at which the return would become negative. Stated differently, the early years would reflect investments in personal capital which are positively associated with success. But, a candidate who has worked past that point is likely to have invested in activities designed for a different career and thus is a weaker candidate. Because of the curvilinear relationship, we tested a quadratic specification of time since graduation, including both years and years-squared. The direction of the covariate as well as the lack of statistical significance led us to reject this specification in favor of a cut-point approach. For screening interviews, probabilities steadily increased until thirteen years out of law school (13-18%), after which they began to decline. For the next two stages, the decline began around year 10. For consistency and comparison, we used the same binary variable for all three models.

91 The probit model includes a publication variable based on the Washington & Lee ranking of law journals.

The relationship between law journal ranking and job market success is not linear; thus, we created tiers based on natural breaks. The three tiers include the two reported (Top 100 and 101-300) and the omitted category (no top-300 articles). 92 We ran the models with the book variable as a separate category, but the sign (negative) did not make sense theoretically. Few candidates have books. And, the books include university press books, practitioners’ guides, casebooks, jury instruction manuals, books only loosely related to law, how-to books, and so on. But the numbers of each are so small, it is not meaningful to divide the “book” variable into separate categories. Therefore, we do not control for books. Another complication: In the ordered probit model, the book variable is perfectly correlated with outcome in some law-school tiers, distorting the standard errors in our ordered probit model.

Page 33: Vanderbilt University Law School - The Wall Street Journalonline.wsj.com/public/resources/documents/lawschoolhiring.pdf · Vanderbilt University Law School . ... Understanding who

32 The Labor Market for New Law Professors

Social Science and STEM doctorates have a negative but small and statistically insignificant effect.93 With respect to their employment at the time of going on the market, we found that applicants who were in law teaching or a fellowship were 10-12% more likely to receive screening interviews compared with the baseline omitted group of individuals who did not come from one of the categories listed in Table 6. Similarly, those working in government were 8% more likely to receive these interviews. Lastly, we include in our specification the tier of law school the applicant attended. The summary statistics show that a disproportionate number of applicants come from Tier 1 law schools, suggesting that graduates from these schools would fare better in the initial screening process. Our results do not support this finding: compared with the omitted baseline group of Tier 1 law graduates, graduates from other tiers have similar probabilities of receiving screening interviews. Graduates from the highest ranked schools within Tier 1 may perform differently than graduates from other schools in Tier 1. Although not reported in the main tables, we found a tipping point, where marginal positive effects start to get smaller, when we moved beyond the top-3 law schools for the second two stages of the job market (on-campus interview and job offer). The tipping point is the top 9 for screening interviews. We chose to use top-3 for comparison purposes across the three models. The choice is also supported by the fact that the top-3 schools--Yale, Harvard and Stanford—have consistently ranked among the top 3 in the U.S. News and World Report, and also annually send many graduates into law teaching.

93 Although not reported in Table 6, we also consider interactions of some of these terms. For example, we considered the joint effect of having both published and possessing a Ph.D. In our sample, candidates with Ph.D.s are more likely to have published (77%) than non-Ph.D. candidates (65%). We found, however, that among the subset of applicants who had published, those with a Ph.D. had roughly the same probability of receiving a screening interview (86%) as those without a Ph.D. (86%).

Page 34: Vanderbilt University Law School - The Wall Street Journalonline.wsj.com/public/resources/documents/lawschoolhiring.pdf · Vanderbilt University Law School . ... Understanding who

The Labor Market for New Law Professors 33

2. On-Campus Job Talk Interview In Table 7, we examine the characteristics that influence whether the survey respondent received invitations to give one or more call-back interviews. Call-back interviews are usually second interviews, and they involve a full-day of meetings and meals highlighted by a presentation to the full faculty. We first observe that women and non-white candidates do better than expected at this stage. Female candidates are 8-10% more likely on average to receive job talks than similarly situated male candidates, a statistically significant difference. Likewise, nonwhite candidates are 10-13% more likely on average to receive job talks than white candidates with comparable credentials, also a statistically significant finding. These results contrast with the results for screening interviews (Table 6), where female and nonwhite candidates did not appear to differ from their male and white counterparts. Together these results suggest that committees are considering women and minorities who share characteristics in common with other interviewed candidates at the initial interview stage, but are more aggressively pursuing both groups at the call-back stage. We see other changes from the screening stage: judicial clerkships, which were statistically non-significant at the screening stage (Table 6), emerge as statistically significant. Respondents who clerked were 11-18% more likely to have job talks. Non-law doctorates, regardless of subject, are statistically significantly related to the probability of being asked for a call-back, however. A top-100 law journal continues to be an important plus: such candidates were 18-20% more likely to have job talks. The largest factor came from professional employment: respondents from a law school teaching or fellowship position were 34-35% more likely to have job talk offers than the baseline group. In pair-wise comparisons, they were also at least 25% more likely to have job talk offers than respondents from any other employment. The magnitude of the factor reflects the importance law schools attach to candidates who are already in a legal academic environment. The mechanisms by which this job provides advantages can be both direct – e.g., signal of scholarly potential, opportunities to write articles – as well as indirect – e.g., developing closer ties to faculty or better understanding the law teaching market. The prevalence of law teaching/fellowships may also explain why the coefficient for bar membership was negative: these positions provide are designed to facilitate entry into law teaching. As a result, some recipients forego legal practice altogether, obviating the need for taking the bar.

Page 35: Vanderbilt University Law School - The Wall Street Journalonline.wsj.com/public/resources/documents/lawschoolhiring.pdf · Vanderbilt University Law School . ... Understanding who

34 The Labor Market for New Law Professors

Table 7. Probit Model of the Probability of Receiving a Call-Back (Job Talk) Interview

(1) (2) (3) (4) (5) (6)

Female

0.100** 0.0994** 0.0825 0.0959* 0.0864* 0.0897*

(0.0495) (0.0504) (0.0522) (0.0509) (0.0525) (0.0524)

Nonwhite

0.1000* 0.125** 0.120** 0.121** 0.119** 0.122**

(0.0553) (0.0557) (0.0572) (0.0559) (0.0578) (0.0573)

Graduated Law School≤10 Years 0.286*** 0.238*** 0.223***

0.223*** 0.201***

(0.0460) (0.0493) (0.0514)

(0.0520) (0.0526)

Judicial Clerkship

0.128** 0.129** 0.176*** 0.126** 0.109**

(0.0516) (0.0533) (0.0504) (0.0539) (0.0545)

Published Scholarship

Top 100 Ranked Law Journals

0.195*** 0.196*** 0.218*** 0.182*** 0.179***

(0.0654) (0.0676) (0.0643) (0.0695) (0.0692)

101-300 Ranked Law Journal

0.0247 -0.0130 0.00126 -0.0202 -0.0186

(0.0620) (0.0648) (0.0633) (0.0652) (0.0652)

Non-Law Ph.D.

Social Science/STEM

-0.00404 0.0581 0.0530 0.0679 0.0519

(0.0876) (0.0903) (0.0866) (0.0916) (0.0911)

Humanities/Other

-0.0172 -0.00649 0.00823 -0.0161 -0.0102

(0.105) (0.107) (0.103) (0.108) (0.107)

Current Employment (Fall 2007)

Law Teaching/Fellowship

0.340*** 0.348*** 0.335*** 0.340***

(0.0607) (0.0581) (0.0614) (0.0609)

Other Academic Teaching/Fellowship

-0.0934 -0.120 -0.0774 -0.0735

(0.0975) (0.0953) (0.100) (0.0992)

Government (Non-clerkship)

0.0870 0.0659 0.0855 0.0844

(0.0874) (0.0860) (0.0877) (0.0880)

Private Practice

0.0159 0.0370 0.00987 0.0326

(0.0663) (0.0639) (0.0667) (0.0669)

Law School Attended

Tier 2

-0.120*

(0.0718)

Tier 3

-0.00880

(0.0982)

Tier 4

-0.106

(0.123)

Foreign Law School

0.0490

(0.115)

Top 3 (Yale, Harvard, Stanford)

0.181***

(0.0666)

N

455 455 455 466 455 455

Dependent variable is whether candidate received one or more call-back interviews. Probit model reports marginal effects with standard errors are reported in parentheses. Models 1-3 and 5 include all survey respondents who hold a law school degree. Model 4 also includes 11 respondents who did not graduate from law school. (All had Ph.D. degrees.) For “Published Scholarship,” the baseline (omitted) category includes any publications which are not in a top-300 law journal as well as no publications. For “Non-Law Ph.D.”, the baseline (omitted) category is no non-law Ph.D. For employment, the baseline (omitted) category includes any type of employment not listed or no employment. ***p<0.01, **p<0.05, *p<0.10.

Page 36: Vanderbilt University Law School - The Wall Street Journalonline.wsj.com/public/resources/documents/lawschoolhiring.pdf · Vanderbilt University Law School . ... Understanding who

The Labor Market for New Law Professors 35

Given that the candidates who receive job talks are a subset of the candidates who receive screening interviews, the differences in the relevant factors for the screening interview (Table 6) and job talk (Table 7) stages are open to a couple of interpretations. The first is that law school appointments committees may be looking for different things in these stages. For example, in the first stage, their search may include candidates that strike the committee’s interest, broadly define. These criteria may include traditional indicators, as well as other factors (e.g., subject matter of their writing) not captured in the specifications. In the second stage, the criteria that compel the committee to invite candidates for a job talk more closely correlates to these traditional indicators. An alternative explanation is the selection effect of the participants in the survey. It could be that among the total applicant pool, the same factors predict both whether a candidate receives a screening interview as well as a job talk. The differences in the screening interviews and job talk stages may reflect that the survey respondents are disproportionately those who were successful in the first stage of the job market. For example, it could be that having a clerkship matters as much at the screening stage as at the job talk stage, albeit not a requirement. Because several respondents without a clerkship received screening interviews, it may understate the true effect.

C. Tenure-Track Job Offers Most applicants who went on the market in the Fall of 2007 did not end up with a tenure-track job. In defining tenure-track job, we asked the respondents directly whether their job offer(s) were tenure-track or non-tenure track. In most instances, tenure-track corresponded to research-track hiring, but for a few schools, tenure-track also included clinical hiring. In our results section, we have a single category of tenure-track hiring.94 Of the 911 participants (which represents the entire list of FAR participants, not simply those who completed our survey), 220 accepted a tenure-track job. In addition, 15 individual who completed the survey reportedly received at least one tenure-track offer but turned it down.95 Table 8 summarizes the outcomes for these individuals, on the tier of law school they attended and the tier of the school – both based on the 2007 U.S. News law rankings – that hired them. The modal outcome is that the tenure-track hires took a job at a lower-tiered

94 We recognize that the criteria for clinical-track tenure-track hiring likely differs in some respects from research-tenure-track track hiring. We do not, however, know with specificity what these differences are.

95 The total number of applicants who received a tenure-track job and declined it is likely higher, given the number of candidates who did not participate in the survey.

Page 37: Vanderbilt University Law School - The Wall Street Journalonline.wsj.com/public/resources/documents/lawschoolhiring.pdf · Vanderbilt University Law School . ... Understanding who

36 The Labor Market for New Law Professors

school than the one they attended. This result is perhaps unsurprising, given that – as shown in Table 2- the majority of applicants attended Tier 1 law schools. Table 8. Summary of Tenure-Track Hires by Relative Ranking of Hiring School and J.D. School

Compared to Rank of Law School Attended N

Rank of

Law School Job

Employed at Higher Tiered School 3

Employed at Same Tiered School 72

Employed at Lower Tiered School 126

Attended Foreign Law School 10

Did Not Report Attending Law School 9

Total 220

Data is based on survey responses, public websites, and authors’ communications directly with individual law

schools.

Table 9 provides a detailed breakdown of tenure-track hires by tier of J.D. law school and hiring law school for all tenure-track professors hired during the 2007-2008 hiring season. Like Table 8, Table 9 includes all FAR participants. For each tier, schools nearly always hire only from the same or higher tier of schools, and all law schools hire principally from Tier 1 schools. Only three hires joined a law school of higher tier than the one they attended. Focusing only on hires with American J.D.s, Tier 1 graduates took 98% of Tier 1 jobs, 90% of Tier 2 jobs, 68% of Tier 3 jobs, and 63% of Tier 4 jobs. Tier 1 graduates accounted for only 68% of the applicants with American law degrees but 82% of the tenure-track hires.

Table 9. Tenure-Track Hires By Tier of Hiring Law School and J.D. Law School

Law School Attended

Tier

1

Tier

2

Tier

3 Tier 4 Foreign Not

Reported Totals

Law

School

Hired

Tier 1 60 1 0 0 6 2 69

Tier 2 53 5 0 1 4 1 64

Tier 3 15 3 3 1 0 2 24

Tier 4 37 12 6 4 0 4 63

Totals 165 21 9 6 10 9 220

Data is based on survey responses, publicly available websites, and authors’ communications directly with

individual law schools.

Only Tier 1 and Tier 2 (schools ranked between 51-100 in the U.S. News rankings) schools hired applicants with foreign law degrees, while all tiers hired at least one applicant without a law degree. For reasons unknown to us, Tier 3 schools (schools ranked between 100-150 in the U.S. News rankings) hired fewer candidates for tenure-track jobs in 2007-08 compared with other tiers. Among the tenure-track hires who completed our survey and had multiple offers, nearly everyone took a job at the highest tier school that

Page 38: Vanderbilt University Law School - The Wall Street Journalonline.wsj.com/public/resources/documents/lawschoolhiring.pdf · Vanderbilt University Law School . ... Understanding who

The Labor Market for New Law Professors 37

gave them an offer.96 Seven candidates, however, turned down an offer from a higher-tiered school to accept an offer at a lower-tiered school. These summary statistics provide an incomplete picture. While certain factors – such as where one attended law school – appear to influence an applicant’s success on the legal academic job market, we do not know whether other characteristics, such as gender, race, publications, etc, influence outcomes. To better understand the potential role of these factors, we turn to regression analysis which allows us to examine how the aforementioned factors collectively influence how applicants fare on the entry-level job market. We look separately at the factors which predict receiving an offer and those which predict the ranking of the school where new professors start. The multivariate analysis allows us to calculate the independent effect of those specific factors which are often cited as critical to law teaching job market success. As reflected in Table 10, conventional wisdom is sometimes based on empirical reality. Candidates who graduated within the last ten years were 30-37% more likely to receive an offer than those who graduated outside of this window. A judicial clerkship improves the probability of being hired by roughly 9-19% (the effect diminishes sharply once controls are added for rank of law school attended). Prior experience in law teaching increases the probability by 25-27%. (But non-law teaching experience does not improve one’s chances.) An article in a top-100 journal increases the probability by 18%-20%. These effects are all statistically significant. Yet, other common assumptions are not supported by the evidence. The returns to coming from a government job are small and only weakly statistically significant. Consistent with the job-talk stage, having a Ph.D. does not increase the probability of receiving a tenure-track offer. Being a man or white did not hurt your chances for receiving a tenure track offer in any of the models. The returns to schooling are more pronounced at the job-offer stage than at the interview stages. Applicants from Tier 2 and 3 schools and those from foreign law schools fare worse but not dramatically than Tier 1 law graduates in getting a tenure-track offer (Column 5). (Graduating from a Tier 2 or Tier 3 law school had similar marginal effects.) However, separating Harvard-Yale-Stanford graduates from all other law schools (Column 6) reveals that graduates from these schools were 27% more likely to receive an offer. This stark contrast reflects that among the 169 candidates who received job offers in our survey sample, 77 (46%) came from one of these three schools.

96 It is possible to examine this question more closely, whether candidates with multiple offers took the offer from the highest-ranked school (again, based on the U.S. News law rankings).

Page 39: Vanderbilt University Law School - The Wall Street Journalonline.wsj.com/public/resources/documents/lawschoolhiring.pdf · Vanderbilt University Law School . ... Understanding who

38 The Labor Market for New Law Professors

Table 10. Probit Model of the Probability of Receiving a Tenure-Track Job Offer

(1) (2) (3) (4) (5) (6)

Female

0.0662 0.0661 0.0609 0.0676 0.0730 0.0712

(0.0487) (0.0496) (0.0504) (0.0490) (0.0509) (0.0513)

Nonwhite

0.00384 0.0302 0.0268 0.0378 0.0304 0.0307

(0.0547) (0.0566) (0.0572) (0.0552) (0.0576) (0.0580)

Graduated Law School ≤10 Years 0.373*** 0.332*** 0.324***

0.324*** 0.296***

(0.0397) (0.0428) (0.0438)

(0.0441) (0.0458)

Judicial Clerkship

0.128** 0.125** 0.185*** 0.123** 0.0933*

(0.0502) (0.0509) (0.0485) (0.0515) (0.0523)

Published Scholarship

Top 100 Ranked Law Journals

0.202*** 0.209*** 0.239*** 0.177** 0.187**

(0.0701) (0.0719) (0.0676) (0.0730) (0.0730)

101-300 Ranked Law Journal

0.0508 0.0388 0.0532 0.0325 0.0375

(0.0604) (0.0616) (0.0605) (0.0618) (0.0622)

Non-law Ph.D.

Social Science/STEM

-0.00134 0.0367 0.0392 0.0261 0.0268

(0.0843) (0.0891) (0.0852) (0.0898) (0.0902)

Humanities/Other

-0.0581 -0.0402 -0.0169 -0.0619 -0.0458

(0.0952) (0.0990) (0.0989) (0.0956) (0.101)

Current Employment (Fall 2007)

Law Teaching/Fellowship

0.256*** 0.272*** 0.245*** 0.265***

(0.0735) (0.0683) (0.0741) (0.0748)

Other Academic Teaching/Fellowship

-0.0180 -0.0623 0.0132 0.0176

(0.0972) (0.0917) (0.103) (0.103)

Government

0.176* 0.122 0.176* 0.182*

(0.0948) (0.0898) (0.0952) (0.0970)

Private Practice

0.0584 0.0832 0.0509 0.0914

(0.0672) (0.0641) (0.0672) (0.0695)

Law School Attended

Tier 2

-0.0894

(0.0643)

Tier 3

-0.122

(0.0852)

Tier 4

-0.23***

(0.0760)

Foreign Law School

0.0380

(0.119)

Top 3 (Yale, Harvard, Stanford)

0.267***

(0.0682)

N

455 455 455 466 455 455

Dependent variable is whether candidate received one or more tenure-track job offers. Probit model reports marginal effects with standard errors are reported in parentheses. Models 1-3 and 5 include all survey respondents who hold a law school degree. Model 4 also includes 11 respondents who did not graduate from law school. (All had Ph.D. degrees.) For “Published Scholarship,” the baseline (omitted) category includes any publications which are not in a top-300 law journal as well as no publications. For “Non-Law Ph.D.”, the baseline (omitted) category is no non-law Ph.D. For employment, the baseline (omitted) category includes any type of employment not listed or no employment. ***p<0.01, **p<0.05, *p<0.10.

Page 40: Vanderbilt University Law School - The Wall Street Journalonline.wsj.com/public/resources/documents/lawschoolhiring.pdf · Vanderbilt University Law School . ... Understanding who

The Labor Market for New Law Professors 39

D. Placement Measuring success on the academic job market by whether an applicant is offered a tenure-track job is, by itself, incomplete. Most schools make at most only a few tenure-track offers in a given year, suggesting that they have preferences over candidates. Given that we observed fifteen candidates who declined one or more tenure-track offers, it is reasonable to infer that candidates also have preferences over schools. We could have ordered schools along many dimensions because law schools are heterogeneous on many salient employment characteristics, including geography, size, and prestige. We focus on differences in prestige, defined by tier ranking from the U.S. News law rankings for 2007. The question that we now ask is whether the same factors that influenced whether a candidate receives any tenure-track offers also affect the relative prestige of the school hiring the candidate,. We focus on all new tenure-track law professors hired during the 2007-2008 hiring season (203 new law professors). We run an ordered probit, in which the outcome variable of interest is the tier of school hired and the coefficients report the expected ordered log odds of a candidate taking a job at a higher-tiered school. In this model, we are looking only at the subset of applicants hired for a tenure-track job. In other words, conditioned on an applicant being hired, we are examining how the applicant’s characteristics affect the tier of the hiring school. In our ordinal ordering, Tier-1 schools have a value of 4, Tier-2 have a value of 3, Tier-3 have a value of 2, and Tier-4 a value of 1, to allow for the interpretation where higher values correspond to placement at higher tier schools. The results, reported in Table 11, reveal that some of the statistically significant factors for predicting a job offer are not significant when considering the ranking of the offering law school. For example, graduating within the last 10 years from law school did not influence school prestige, nor did having a law teaching fellowship position. While these factors helped applicants distinguish themselves from the general applicant pool in receiving a tenure-track offer, those credentials do not help distinguish among the set of tenure-track hires. This result is perhaps unsurprising, given that candidates in Table 11 are a subset of candidates in Table 10.

Page 41: Vanderbilt University Law School - The Wall Street Journalonline.wsj.com/public/resources/documents/lawschoolhiring.pdf · Vanderbilt University Law School . ... Understanding who

40 The Labor Market for New Law Professors

Table 11. Ordered Probit Model of Relative Prestige of Tenure-Track Job

(1) (2) (3) (4) (5) (6)

Female

-0.204 -0.142 -0.144 -0.154 -0.0461 -0.122

(0.158) (0.161) (0.161) (0.155) (0.165) (0.162)

Nonwhite

-0.275* -0.0971 -0.0629 -0.153 -0.0447 -0.00431

(0.163) (0.176) (0.178) (0.171) (0.180) (0.181)

Graduated Law School ≤10 Years 0.328 0.182 0.127

0.150 -0.0394

(0.214) (0.223) (0.232)

(0.233) (0.238)

Judicial Clerkship

0.297* 0.298* 0.337** 0.252 0.229

(0.173) (0.174) (0.168) (0.175) (0.175)

Published Scholarship

Top 100 Ranked Law Journals

0.657*** 0.626*** 0.608*** 0.569*** 0.647***

(0.210) (0.216) (0.213) (0.217) (0.217)

101-300 Ranked Law Journal

0.121 0.0758 0.0601 0.0844 0.0782

(0.200) (0.203) (0.199) (0.204) (0.204)

Non-law Ph.D.

Social Science/STEM

0.875** 0.905** 0.977*** 0.854** 0.795**

(0.357) (0.362) (0.351) (0.364) (0.368)

Humanities/Other

0.236 0.282 0.289 0.222 0.253

(0.305) (0.307) (0.305) (0.307) (0.307)

Current Employment (Fall 2007)

Law Teaching/Fellowship

0.221 0.245 0.252 0.261

(0.181) (0.177) (0.182) (0.183)

Other Academic Teaching/Fellowship

-0.00566 -0.0322 0.327 -0.00992

(0.480) (0.474) (0.495) (0.480)

Government

-0.0687 -0.0956 -0.143 -0.202

(0.258) (0.245) (0.261) (0.263)

Law School Attended

Tier 1

0.581***

(0.211)

Top 3 (Yale, Harvard, Stanford)

0.604***

(0.178)

Cutpoints

Cut1

-0.544** -0.227 -0.211 -0.313* 0.269 -0.182

(0.221) (0.239) (0.260) (0.185) (0.315) (0.261)

Cut2

-0.234 0.110 0.126 0.0244 0.616* 0.163

(0.218) (0.238) (0.259) (0.185) (0.316) (0.261)

Cut3

0.572*** 0.969*** 0.990*** 0.860*** 1.502*** 1.064***

(0.220) (0.244) (0.264) (0.189) (0.324) (0.266)

N

203 203 203 220 203 203

Ordered probit model reporting Z-scores. The model is limited to applicants who took a tenure-track job. Models 1-3 and 5 include all 203 new hires who hold a law school degree. Model 4 also includes 17 new hires who did not graduate from law school. (All had Ph.D. degrees.) For “Published Scholarship,” the baseline (omitted) category includes any publications which are not in a top-300 law journal as well as no publications. For “Non-Law Ph.D.”, the baseline (omitted) category is no non-law Ph.D. For employment, the baseline (omitted) category includes any type of employment not listed or no employment. Private practice is part of the baseline (omitted) category for “Current Employment” in the Ordered Probit model but not the Probit models because a separate category violates the proportional odds assumption. ***p<0.01, **p<0.05, *p<0.10.

Page 42: Vanderbilt University Law School - The Wall Street Journalonline.wsj.com/public/resources/documents/lawschoolhiring.pdf · Vanderbilt University Law School . ... Understanding who

The Labor Market for New Law Professors 41

Other factors meaningfully influence the tier of hiring school, even if they do not appear to affect whether a candidate receives a tenure-track offer. In Table 10, having a social science or STEM Ph.D. had a statistically insignificant effect on whether a candidate was hired; in Table 11, having a social science or STEM doctorate increased the expected ordered log odds by roughly 0.8-1.0 as one moves to the next higher tier of school, and is statistically significant. (Perhaps relatedly, candidates with non-law teaching experience were less successful in earlier stages but more successful if hired.) The difference varies by tier of school. Expressing the Ph.D. effect another way – using the results from Column 5 in Table 11 – the predicted probability of a candidate being hired at a Tier-4 school, conditioned on being hired, is 26% without a Ph.D., and 7% with one. At a ier-3 school, the predicted probabilities are 12% without a Ph.D., 6% with one. For Tier-2 schools, the difference in predicted probabilities is smaller: 34% without a Ph.D. and 27% with a Ph.D. For Tier-1 schools, however, the direction switches: the predicted probability is 28% without a Ph.D., and 61% with a Ph.D. This means that conditioned on being hired, applicants with a Ph.D. are less than half as likely to end up at Tier-3 or Tier-4 schools as applicants without Ph.D.s, but more than twice as likely to end up at a Tier-1 school. In other words, having a Ph.D. may not affect one’s chances of getting a tenure-track job, but appears to have a significant and positive effect on placement. These differences may reflect the preferences of the schools or of the candidates themselves, something we intend to explore more fully in future work. Where applicants attended law school affects their placement. Graduating from a Tier-1 law school increases the expected ordered log odds also by approximately 0.6. This coefficient translates into Tier-1 graduates, conditioned on being hired, as having a 19% probability of being hired at a Tier-4 school, compared with law graduates from all law schools having a 27% probability. At Tier-2 schools, the probability of being hired is 11% for Tier-1 graduates and 12% for graduates from other tiers. Tier-1 graduates have a slightly greater probability than graduates of other tiers of being hired at Tier-2 schools: 34% to 30%. Finally, Tier-1 graduates have a 34% probability of being hired at a Tier-1 school, compared with 16% for graduates of other tiers. In the final specification (Column 6), attending a top-3 law school (Yale, Harvard, or Stanford) has statistically significant effect on the expected ordered log odds of placement at a more prestigious school. Conditioned on being hired, the predicted probability of a top-3 law graduate ending up at a Tier-1 school is 45%, compared with 23% for all other law graduates. At a Tier-2 school, the predicted probability for top-3 graduates is 33% and for other law graduates is 34%. For Tier-3 law schools, the predicted probabilities are 9% for top-3 graduates and 13% for other law graduates. Finally, at Tier-4 law schools, the predicted probabilities are 14% for top-3 graduates and 29% for other law graduates. These figures show that Tier-1 schools are twice as likely all other law schools combined to hire top-3 graduates, and Tier-4 law schools are twice as likely to hire from law schools other than Yale, Harvard or Stanford.

Page 43: Vanderbilt University Law School - The Wall Street Journalonline.wsj.com/public/resources/documents/lawschoolhiring.pdf · Vanderbilt University Law School . ... Understanding who

42 The Labor Market for New Law Professors

Table 12. Comparing Successful to Unsuccessful Entry-Level Teaching Job Applicants: Survey Respondents

Tenure-Track Job Offer

No Job Offers

Female 41% 33%

Non-White 25% 24%

Median Years Since Law School Graduation 6 11

Law School Attended

Tier 1 78% 60%

Tier 2 11% 19%

Tier 3 4% 9%

Tier 4 2% 5%

Foreign Law School/Unranked 5% 5%

No law degree 1% 1%

Published Scholarship

Top 100 Ranked Law Journal 26% 11%

101-300 Ranked Law Journal 23% 20%

Non-Law Ph.D.

Social Sciences/STEM 9% 8%

Humanities/Other 6% 5%

Current Employment ( Fall 2007)

Private Practice 31% 33%

Government (non-clerkship) 10% 10%

Law School Fellowship/Teaching 36% 16%

Other Fellowship/Teaching 5% 11%

Judicial Clerkship (Post-Graduate) 54% 31%

N 169 297

Data includes all applicants who completed the George & Yoon survey. Some survey respondents did not participate in the AALS market. This table breaks out respondents based on whether they received at least one tenure-track job offer (defined as “successful”). The variable “Years Since Law School Graduation” only includes law school graduates.

Page 44: Vanderbilt University Law School - The Wall Street Journalonline.wsj.com/public/resources/documents/lawschoolhiring.pdf · Vanderbilt University Law School . ... Understanding who

The Labor Market for New Law Professors 43

V. CONCLUSIONS Ultimately, our findings confirmed some common perceptions and dispelled others about who secures a tenure-track law teaching job. Women and non-whites fared the same as similarly situated men and whites. As expected, Tier-1 graduates and law teaching fellows fared better than other applicants, holding other characteristics constant. Conversely, the experience of applicants with social science and STEM Ph.D.s – who were no more likely to be hired, but conditioned on being hired, more likely to end up at a Tier-1 law school – suggests that the attributes that schools are seeking in candidates may differ depending on the relative prestige of the school. The characteristics which predict ultimate success (i.e., a job offer), however, do not tell us as much about success early in the market. In the first stage – screening interviews – it appears that law schools took a relatively expansive approach. For example, schools at this stage were most willing to interview candidates who had not clerked, had not published, or had attended a Tier-2 or lower school. By the second stage – job talks – these percentages declined. The remaining candidates without the traditional credentials were highly unlikely to get a job offer and any offer was likely to be at a lower ranked school. The queuing theory holds true for the tenure-track law teaching market as it does for other tenure-track academic markets. Graduating from a Tier-1 law school, or better yet, a top-3 law school, dramatically improves a candidate’s chance being offered a tenure-track job, holding other qualifications constant. This strong path dependence in law school hiring provides stability but possibly at the price of innovation and adaptation. The current study can help us to evaluate the law school crisis through its focus on how law schools hire tenure-track faculty. As law schools are called upon to reevaluate our goals and pedagogy, they rely on law professors who have a limited range of experiences. Nearly all new hires in our study attended a small set of schools which are more likely to emphasize theory over practice. And, the new hires have a shared set of professional experiences, especially time as a law teaching fellow and a judicial law clerk. Missing from those experiences is substantial time outside of a law school. The lack of real-world experience affects the capacity of new hires to teaching skills courses and to assist students making the transition to those real-world jobs. The challenges facing legal education, of course, cannot be fixed simply by hiring different tenure-track professors. Our study casts doubt on the ability of law schools to innovate in entry-level hiring due to the self-reinforcing nature of the selection process. Even if law schools could respond to the call for more skills training or better career advising by altering their tenure-track hiring, the reasons against doing so—the value of research and the core curriculum—may outweigh dramatic changes in tenure-track hiring. But, continuing to hire tenure-track professors who share the same credentials and experiences as tenured faculty may be a poor investment for law schools. The solution may be greater diversification in the types of law-teaching positions in law schools.