The ARWU Ranking Has 500 Seats. China Brought 71 New Universities in 2020

Introduction

Every August when ARWU university rankings are published, the same headlines return. A German university has slipped out of the world top 500. An Italian university is no longer listed. A French flagship has fallen. Commentators reach for the obvious explanation: these institutions or national university systems are declining.

Most university rankings have issues with self-reporting, subjective reputation indicators, lack of meaningful measures of teaching effectiveness, and lack of audit or verification mechanisms. ARWU however is built almost entirely from objective bibliometric data (Nobel/Fields winners, highly-cited researchers, Nature/Science papers, citation counts), not self-reporting or reputation surveys. Those critiques apply much better to THE and QS, not ARWU.

The table has 500 seats

The Academic Ranking of World Universities (ARWU), commonly known as the Shanghai Ranking, publishes exactly 500 places in its main table each year (Liu & Cheng, 2005). This single design fact is easy to overlook and impossible to overstate. When a new university enters the list, an existing one must leave. Not because it failed. Because the list itself cannot grow.

Between 2010 and 2020, the number of Chinese universities appearing in the ARWU top 500 rose from 22 to 71 (Corrigan & Rodriguez, 2022). That is a near-tripling of China's presence on a table whose total size never changed. Every one of those additional seats had to come from somewhere. Mathematically, roughly seventy universities elsewhere in the world had to make room.

Who made room

The Center for Security and Emerging Technology's 2022 analysis, based directly on ARWU and QS data, quantified exactly where that room came from. Over the same decade, the number of ranked universities fell in France (from 22 to 17, a 23 percent decline), Germany (from 39 to 30, also 23 percent), Italy (from 22 to 17, another 23 percent decline), and Japan, which lost 44 percent of its ranked institutions (Corrigan & Rodriguez, 2022). The United States lost proportionally less, 14 percent, but in absolute terms lost the most seats of any country, since it had the most to lose (Corrigan & Rodriguez, 2022).

Country	ARWU Top 500, 2010	ARWU Top 500, 2020	Change
France	22	17	-23%
Germany	39	30	-23%
Italy	22	17	-23%
Japan	25	14	-44%
United Kingdom	38	36	-5%
United States	154	133	-14%
EU-27	153	134	-12%
China	22	71	+223%

Source: Corrigan & Rodriguez (2022), Table 6, based on Academic Ranking of World Universities and QS World University Rankings data, 2010–2020.

The CSET researchers examined whether this decline reflected any genuine deterioration in these countries' universities. It found no such evidence. Nobel Prizes, research output, and international collaboration among American faculty showed no meaningful decline over the decade in question (Corrigan & Rodriguez, 2022). What had changed was not the quality of the incumbents. It was the number of competitors above the cutoff line.

Nobody got worse. The table got smaller, relatively speaking

This is worth stating plainly, since it is the finding that headlines never carry: the universities that lost their place in the ARWU top 500 between 2010 and 2020 did not, as a rule, get worse. The ranking's own architecture, a fixed 500-seat table scored primarily on Nobel and Fields awards, highly cited researchers, and publications in Nature and Science (Liu & Cheng, 2005), guarantees that a rising tide of new high-scoring entrants will displace existing institutions regardless of whether those institutions' own output changed at all.

A university sitting near rank 480 in 2010 needed only to stand still while dozens of newly funded Chinese universities, several backed by concentrated state investment through initiatives like Project 985 and the Double First Class Plan (Corrigan & Rodriguez, 2022), climbed past it. Standing still, on a fixed-size table, is functionally the same as falling.

Why the displaced were probably not your flagship universities

Here a further, logical point follows, though it should be read as reasoning from the mechanism rather than as a name-by-name audit of who exactly left the list. If displacement strikes hardest at whoever sits closest to the cutoff, and if a country's oldest, best-resourced, most internationally recognized universities tend to sit well above that cutoff, then the universities most exposed to displacement are structurally more likely to be the ones with a comparatively narrow research profile: newer institutions still building a publication record, universities formed or reshaped through recent mergers whose ARWU scores had not yet stabilized, or smaller, subject-specialized institutions competing on a table that rewards breadth and volume of output (Liu & Cheng, 2005).

This is an inference from how the ranking is built and where the losses concentrated, not a claim that every dropped university fits that description. But it is a considerably more defensible starting hypothesis than "national decline," and it follows from publicly documented facts about the ranking's mechanics rather than from speculation about any single institution.

Conclusion: What this means for anyone reading a ranking headline

Three things follow from the 2010–2020 data. First, a country's count of ranked universities is not, by itself, a performance indicator, particularly for countries whose universities cluster near the cutoff rather than the top. Second, exposure to displacement is a structural feature of a fixed-size, research-weighted table, not a report card on any individual university. Third, the correct response to a falling count is to ask where in the table the losses occurred and why, not to reach immediately for a story about decline.

The next time a headline announces that a country's universities are "falling behind" because some of them dropped off a list, it is worth remembering the arithmetic. 71 additional Chinese universities did not appear because their competitors failed. They appeared because a table with 500 fixed seats had no seats to spare, and something had to give.

---

References

Corrigan, J., & Rodriguez, S. (2022). Chinese and U.S. university rankings: A lens into top universities and their graduates (CSET Data Brief). Center for Security and Emerging Technology. https://doi.org/10.51593/20210047

Liu, N. C., & Cheng, Y. (2005). The Academic Ranking of World Universities. Higher Education in Europe, 30(2), 127–136. https://doi.org/10.1080/03797720500260116