Boycott, Exit & Plurality

On 16 November 2022, American legal education did something nobody had dared do in twenty years: Yale Law School, perennial holder of the top spot, announced it was withdrawing from the U.S. News rankings, with dean Heather Gerken's statement calling them "profoundly at odds with the mission of this profession." Harvard followed within hours, and within days Berkeley, Georgetown, Columbia, Stanford and Michigan pulled out, with more than 63 law schools and a set of leading medical schools eventually joining. You saw in R01 and R04 how this ranking machine disciplines schools; this was the largest counterattack the disciplined have ever mounted, and the ending is brutal: two years later the ranking was unscathed, and had used the occasion to upgrade itself.

U.S. News responded in two steps. Step one: rank them anyway. Law schools' core data, employment rates, bar passage rates and entering credentials, are already subject to mandatory disclosure standards set by the American Bar Association, so the magazine switched to public data and kept ranking every school that had left. Withhold your data and the ruler measures you with whatever data it can get: not one of the leavers disappeared from the list, they only lost their say over their own numbers. Step two: change the rules. It cut the weight of the reputation survey and raised employment, bar passage and debt measures. Once the method changed, the standings swung violently: in April 2026 Yale was pushed off the top spot it had held for 36 years by Stanford, and UC Berkeley fell out of the T14. The schools that launched the boycott took the biggest hit from the redesign.

The medical school side went further. After the collective withdrawal in 2023, U.S. News simply abandoned numerical rankings for research medical schools in 2024, switching to four tiers, in use for three years by 2026; traditional powers such as Michigan are not mentioned at all on the new list, with tiering wiping out differentiation entirely. One structure of interest also came into the open: a New York Times investigation in January 2024 found that U.S. News sells "badge" licences to universities, so a school at any rank can pay to use a "Ranked #X by U.S. News" mark in its recruitment materials. The ranker under boycott was taking criticism from the same schools it was selling ranking merchandise to.

The ending was predicted thirteen years earlier. Sauder and Espeland's 2009 judgment (the core of R04): resistance is not the opposite of power, it is part of how power runs. Withdrawal was also far from the first act of resistance, and the prehistory is long: 173 deans once co-signed a warning to applicants that "these ranking systems are inherently flawed"; the Association of American Law Schools (AALS) commissioned research attacking the rankings' methodology; some schools mailed letters condemning the rankings to LSAT takers. Each act of resistance turned the ranking further into the reference point around which everyone's beliefs are organised. Being the reference point means it keeps getting checked against, cited and argued over; that recurring entanglement is what prolongs it and deepens the discipline. The collective withdrawal of 2022 was no different: it did not stop the audiences that matter, applicants, employers and donors, from using the rankings, but it did force the rankings to change. The change reshuffled the seating, and the reshuffling produced a new round of anxiety and response. Resistance, methodological change, rank turbulence, another round of reactivity, and the loop is reconnected. The root is the four adjectives from R04: a ranking is a commensurable, relative, widely circulated, zero-sum measure. As long as the audience is still watching, you are on the ruler, and cooperating only decides whose numbers get used to measure you.

The one time a counterattack actually worked happened where nobody organised one. Sauder and Espeland again, comparing the situations of law schools and business schools in 2006. Law schools faced a dominant U.S. News: it ranked every accredited law school, comprehensively, every year, with no credible alternative list. Business schools were entirely different: BusinessWeek, U.S. News, the Financial Times, the Economist and the Wall Street Journal, at least five high-profile lists coexisting, with different methods and conflicting results. The consequence was decisive: multiple rankings dilute each other, none holds the power of life and death, and a school can pick the most flattering one to tell a story about, since "we are 9th on this list" and "we are 4th on that one" can both be true. Educators got their discretion back, and rankings' disciplinary effect on business schools is markedly weaker than on law schools. The sociologist Healy later commented that this is the genuinely distinctive finding in all of Espeland and Sauder's work: the more rankings there are, the weaker any one ranking's disciplinary force, and with enough competing evaluators the whole ranking business might even lose its legitimacy. Degree of monopoly, zero-sum stakes and coverage are the first set of moderating variables for the intensity of reactivity. The law schools' hell was never being measured, it was having only one ruler.

So "pluralisation" looks like the real way out, but the evidence and the limits both need stating in full. Limit one is a paradox, pointed out by that same 2006 paper: hedging against U.S. News by making more rankings can weaken a single ranking's worst effects, but further establishes "ranking" as the legitimate form of accountability in higher education, so the means of resistance entrenches the thing resisted. Limit two: pluralisation restores discretion, not immunity, and gaming continues, just spread from one front across several. Parallel cases in other fields support the same judgment: the World Bank's Doing Business rankings were discontinued in 2021 after a data manipulation scandal, and the successor B-READY claims a more transparent method, yet in late 2025 the Hong Kong government still publicly protested that its rating was "outdated and unfair," so a cleaner method does not dissolve the incentive of the ranked to fight over their score. The list of leavers is also lengthening across fields, with Paris-Sorbonne University withdrawing from the THE world university rankings in 2025. The latest correction from inside the theoretical camp comes from Chun in 2025: early work focused too much on a single dominant ranking and overlooked the normal condition in which multiple measures coexist and contend. Put it together: the workable form of counterattack is not exiting the ruler, it is having so many rulers that none can define you alone. Grade the evidence honestly: the business school comparison is the only strong empirical support for this route, and what it shows is a markedly weaker effect, not no effect.

One open question: did withdrawal actually weaken the rankings' influence, or did it merely switch the data supply from voluntary surveys to mandatory disclosure while the list carried on? The D3 report's verdict: answering it requires three to five years of post-withdrawal applicant behaviour data. Until then the only certainty is that in the winter of 2022 the schools that walked out thought they were leaving the exam hall, when in fact they had only changed seats.

The one-line takeaway: you can refuse to hand in your paper, but as long as anyone is still reading the list, you are still sitting the exam.