In 1945, Dr. Joseph Turner, director of the Mount Sinai Hospital, in New York City, complained to the Association of American Medical Colleges about “the present evils in intern selection.” Most appalling was the chaos, “a disorder which has, by now, become intolerable.” He called upon the deans of American medical schools to unite in approving his scheme to “replace today’s disorderly scramble for intern places with a decent and dignified system of selection.”
Or subscribe now to read with unlimited access for less than $10/month.
Don’t have an account? Sign up now.
A free account provides you access to a limited number of free articles each month, plus newsletters, job postings, salary data, and exclusive store discounts.
If you need assistance, please contact us at 202-466-1032 or help@chronicle.com.
In 1945, Dr. Joseph Turner, director of the Mount Sinai Hospital, in New York City, complained to the Association of American Medical Colleges about “the present evils in intern selection.” Most appalling was the chaos, “a disorder which has, by now, become intolerable.” He called upon the deans of American medical schools to unite in approving his scheme to “replace today’s disorderly scramble for intern places with a decent and dignified system of selection.”
The deans of American higher education today would do well to consider such a scheme as a solution to the present evils of the academic-job market. I propose instituting a computerized match system for faculty hiring. Beginning in 1952, an algorithmic match was successfully used to stabilize the labor market for medical interns. On the academic-job market today, as funds for tenure lines shrink and the pool of unemployed Ph.D.s grows, the process of selecting candidates for tenure-track positions is increasingly chaotic. Like the labor market for medical interns in the mid-20th century, universities are facing a state of “market failure.”
When Turner addressed the AAMC, the medical internship was a relatively new phenomenon. In the early 20th century, medical science was changing fast, and medical practice in the United States was becoming an increasingly regulated and specialized profession. To keep up with the times, newly minted doctors began to take an extra year of practical training, in a hospital setting, after completing the four-year university medical curriculum. During World War I, the 19th-century system of hospital placement was formalized into the “internship.” Medical and surgical internships persist today, with the addition of several years of specialty training known as “residency.”
Interns and residents perform much of the labor that keeps a hospital running, and they have always been rather poorly paid for their work. As the term “resident” indicates, these “house staff” were traditionally compensated with room and board. Because interns provided cheap labor, the number of internships approved by the Council on Medical Education boomed in the 1920s, yet the number of graduates from U.S. medical schools remained stable, fixed by the same accrediting body.
The imbalance between the number of positions and of candidates made for an unstable market. Competition among programs for the best candidates, along with the necessity, for candidates, of completing such a program to ensure future employment, and the perceived inequality between programs, conspired to produce a hysterical situation. Hospitals began contacting students earlier and earlier in their medical training in the hopes of securing future interns before other hospitals. Students responded in turn, hoping to secure their own positions before their classmates. Internship placement began to consume the process of medical education. This problem will sound familiar to graduate students who find themselves overwhelmed with worry about the job market long before they begin writing their dissertations.
ADVERTISEMENT
In 1945, medical-school administrators agreed to a uniform date at which they would release student information to hospitals, intending to limit the interruptions of the placement process to senior year. Establishing a standardized calendar helped candidates on the academic-job market. But it was only a Band-Aid for mid-century medical education. Hospitals responded to the new time constraints by increasing decision pressure on students. By 1949, they were issuing “exploding offers” — telegrams to top candidates followed by telephone calls demanding an immediate response.
If a young medical man (and they were almost all men) received an offer from his top program, say Massachusetts General Hospital, then it was time for champagne. But if such a telegram arrived from Mount Sinai Hospital, he was faced with a dilemma. If he accepted the offer, and Mass General called the next day, he would be out of luck. But if he rejected Mount Sinai and never received a more desirable offer, he would be in even worse straits.
The solution, proposed in 1950 and adopted, after a trial run, in 1952, was a computerized match. Alvin Roth and Lloyd Shapley were awarded the Nobel Prize in Economics in 2012 for their work on the algorithm behind the National Resident Matching Program. The theory of stable matches and the implementation of market design is complex. But the process is fairly simple. The match takes place at the end of the application cycle, when interviews are complete. Students rank internship programs in a strict order of preference. Programs do the same. Both submit their lists to a centralized clearinghouse that runs the match. The algorithm tentatively matches candidates to programs, and then revises the matched pairs as it processes the lists. The end goal is stable marriages, with no frustrated lovers, i.e., no hospitals and candidates who would prefer each other to their assigned match.
On the academic-job market today, an even greater imbalance exists than the disparity that plagued the intern market in the first half of the 20th century. The situation is reversed: Eligible candidates far exceed available positions. Yet a number of features are the same: The market is controlled by a university-run credentialing process, and the hiring structures are strongly hierarchical. The pressures on candidates are similar.
Proposals for solving the woes of the academic-job market often focus on controlling supply and demand. A Keynesian approach would limit the awarding of doctoral degrees, thereby restricting the supply of eligible candidates and increasing hiring demand. Place the right number of candidates in the invisible hand of the market, and the problem is solved. A Marxist version to the same problem would unionize doctoral candidates and contingent faculty, seeking to increase the number of tenure-track positions, thereby revising the terms of the demand to meet the needs of candidates. Both of these solutions seek to establish parity between the number of positions available and the candidates eligible to fill them.
If I didn’t get an interview request, I would know right away that I didn’t get a job. And that would be a relief, compared to the current situation.
ADVERTISEMENT
A computerized match would not solve that problem. It did not do so in medicine. The number of internships available in U.S. hospitals exceeded the number of eligible candidates until the mid-70s, and the balance was only achieved by significant market participation of graduates from non-U.S. medical schools. The number of candidates eligible for tenure-track jobs at U.S. institutions of higher education is likely to continue to exceed the number of available positions. A match process cannot ensure that more Ph.D. students will ultimately receive the jobs they were ostensibly trained for. But it can alleviate much of the pain of the hiring process for candidates and search committees.
What would a computerized match look like in faculty hiring? Let’s say that I am applying for a tenure-track assistant professor position in English. I would read job ads from universities and submit the requested application materials, just as I do now. The main difference would be timing. In order for a match to work, all the job ads would need to be posted by a given date, for example, September 1. It would work best if they were all posted in the same place, perhaps the website of the new Faculty Match Program. The application materials would likewise be due at a uniform time, let us say November 1.
Once I submitted my application materials, I would wait to hear back from employers, biting my fingernails, just as I do now. Interview requests would likewise be issued on a specified date, perhaps January 1. If I didn’t get an interview request, I would know right away that I didn’t get a job. And that would be a relief, compared to the current situation.
Between November 1 and January 1, search committees would read applicant materials and determine which candidates merit interview requests. Residency programs often use a standardized metric to determine who receives an interview and who does not, such as Usmle Step 1 scores. The reliance on standardized testing has received a lot of pushback in recent years, and rightly so. Yet it does decrease the burden on hiring committees, and it is significantly more feasible in a climate of uniform hiring practices.
ADVERTISEMENT
Internship programs typically do one round of interviews. In the past, interviews were conducted in person on the hospital campus. They included information sessions, designed to help students complete their rankings. This year the interviews were conducted virtually due to the pandemic. Colleges typically run two rounds of interviews: initial interviews at a disciplinary conference like the MLA, and secondary campus visits. In recent years, the Skype interview has come to replace the conference interview, and the pandemic solidified the trend. The place where it makes the most sense to run the match is between the initial interviews and the secondary campus visits.
iStock
After the first round of interviews, candidates would submit their preferences in rank order, and search committees would do the same. There would be just as much wrangling over preferred candidates as in the past, and the committee would retain the power to rank their top choices. But candidates would get a say too. Aspiring faculty members would rank programs based upon where they would, realistically, like to work. Candidates and committees would submit their lists to a centralized clearinghouse on a set date, say February 15. Match lists for campus visits would be released on March 1.
The algorithm would be designed to ensure a maximum distribution of candidates across openings. While the number of first round interview requests a candidate could receive would be unlimited, the number of campus visits would be limited to three. The process would prefer to provide each candidate with one campus visit before providing any candidates with a second. This would benefit both parties. A greater number of candidates would receive campus visits than in the current system. And the department conducting the search could rest assured that the candidates matched to their campus were actually interested in taking the job.
Colleges would then conduct campus visits and complete the hiring process as usual. Because of the imbalance of candidates and positions, the risk of unmatched candidates would be high (but that’s nothing new). The risk of unmatched positions is small, perhaps smaller than in the current system, under which searches not infrequently fail despite the superabundance of job candidates. Stable matches would mean fewer faculty members who go on the market after one or two years in a position, thereby decreasing the quantity of applications that search committees need to wade through in future cycles.
The institutions that grant doctoral degrees also seek to hire the recipients of those degrees, so their stake in resolving the crisis is twofold.
Such a system would significantly decrease the burden on applicants by increasing the uniformity of the process. It would provide for earlier and coordinated notification of applicants without increasing the burden on committees. And it would produce a more equitable hiring system, with fewer opportunities for nepotism and less room for chance. Much of the emotional energy spent in fruitless hunting of job ads across the internet could be saved by a centralized process, and the ceaseless anxiety waiting for responses would have a fixed end.
ADVERTISEMENT
Like the evils of intern placement in 1945, problems in university hiring have been brewing for decades, and the disorder has become intolerable, from the perspective of candidates. But it took at least seven years of concerted efforts by all interested parties to replace the disorderly scramble of intern selection with a dignified matching algorithm. Resolving the entropic chaos of the academic-job market would be an even greater feat.
Establishing a faculty match would require centralization, standardization, and coordination. Convincing colleges and universities to hand over a part of their hiring processes to a single central body would not be easy. The match would work best if there were standardization across positions and advertisements, which means that intuitions of higher education would have to come to some agreement about the type of faculty member they seek to hire. In order for it to work at all, the match would require temporal coordination across a unified market. But it would not require solving the intractable problem of supply and demand, the mismatch between eligible candidates and available positions.
As in medicine, the supply of candidates for entry-level positions in the professoriate is controlled. There are a limited number of degree-granting institutions, and those institutions have a stake in seeing their candidates placed. The deans of medical schools instituted the initial changes to the intern market in 1945, and coordination with hospital administrators provided a lasting solution in 1952. The academic-job market has the added benefit of being a closed system. There are individuals without doctoral degrees who get hired in tenure-track positions, and there are many sorts of employers, from publishing houses to biotech firms, who hire Ph.D.s. Yet the institutions that grant doctoral degrees also seek to hire the recipients of those degrees, so their stake in resolving the crisis is twofold.
The computerized match was ultimately instituted in medicine because the preceding system disadvantaged employers. It did not work well for medical schools or their students either, but it wasn’t student unions that drove the change. Hospital administrators, whose residency slots were left unfilled, had the greatest incentive to lobby the AAMC for a computerized match.
Colleges and universities have an equally pressing incentive to change. The failure of the academic-job market is evident to all those involved. The madness of the market is subsuming the process of doctoral education. Without substantial changes, the doctoral degree will lose its value and the market will collapse. Collective action among graduate students and contingent faculty members can draw attention to these issues, but only the unified, cooperative action of deans and presidents can solve them.