China's Five Onlys Reform & CAS Journal List

TL;DR: China’s research-evaluation reform has moved in two connected stages. First, starting around 2018 and formalized in Ministry of Education/Ministry of Science and Technology guidance (February 2020) and a joint Communist Party of China (CPC) Central Committee/State Council policy (October 2020), the government committed universities and research institutions to “break the five onlys” (破五唯, po wu wei) — ending evaluation that leans solely on scores, enrollment/advancement rates, diplomas, papers, and prestige titles. Second, on 27 March 2026, the Chinese Academy of Sciences (CAS) National Science Library announced it would stop updating or releasing its widely used Journal Classification Table (commonly called the CAS Journal Ranking) from 2026 onward, with the 2025 edition standing as the final one. This guide explains what each reform actually did, why the journal-ranking discontinuation is a meaningful downstream step rather than a standalone event, and what it means for institutions and researchers who collaborate with Chinese partners or evaluate work published under this system.

What “breaking the five onlys” actually targets

“Five onlys” is shorthand for evaluation practices that reduce a person’s or institution’s worth to a small number of proxy metrics rather than substantive judgment. The specific five items named vary slightly by source and by which sector of education the document addresses, but the commonly cited formulation from the October 2020 national policy is: only test scores (唯分), only enrollment/advancement rates (唯升学), only diplomas (唯文凭), only papers (唯论文), and only titles or “hats” (唯帝子) — the last referring to prestigious talent-program titles (such as Changjiang Scholar or Thousand Talents designations) that carried outsized weight in hiring and promotion decisions independent of a candidate’s actual body of work.

For research administrators, the version of this reform that matters most is the research-specific application: universities and research institutes were told to stop using publication count, journal placement, professional title, degree credentials, and honorary/talent-program titles as mechanical stand-ins for research quality in hiring, promotion, tenure, and funding decisions. In substance and intent, this is the same critique DORA and CoARA make internationally about journal-level metrics standing in for the assessment of individual research — see CASRAI’s guide on DORA and CoARA research assessment commitments for the parallel international framing.

The policy timeline

Around 2018: Criticism of “five onlys” evaluation culture entered Chinese higher-education policy discourse, laying rhetorical groundwork for the formal reforms that followed. (Exact originating documents from this period are less consistently dated across sources than the 2020 policies below; treat this as the reform’s acknowledged starting point rather than a single dated instrument.)
February 2020: The Ministry of Education (MOE) and Ministry of Science and Technology (MOST) jointly issued an opinion directive discouraging Chinese universities from using Science Citation Index (SCI) paper counts and related bibliometric indicators as the primary basis for evaluating researchers, conferring degrees, or allocating funding — explicitly targeting “only-SCI” (唯SCI论) evaluation logic specifically.
October 2020: The CPC Central Committee and State Council jointly issued the Overall Plan for Deepening the Reform of Education Evaluation in the New Era, which formalized “breaking the five onlys” as a comprehensive, top-level target spanning the entire education system, not research evaluation alone.
27 March 2026: The CAS National Science Library announced it would stop updating and releasing its Journal Classification Table from 2026 onward, with the 2025 edition as the final release — the development this guide focuses on.

Six years separate the formal “five onlys” policy from the CAS journal-list discontinuation. That gap matters: it illustrates how much slower infrastructure and incentive systems change than policy language does, a pattern research-assessment reform advocates elsewhere (DORA’s own case studies, CoARA’s implementation guidance) also describe as the harder half of reform.

What the CAS Journal Classification Table actually was

The CAS Journal Classification Table (also referred to as the CAS Journal Ranking or Journal Partition Table) was a citation-based system, produced by the CAS National Science Library, that sorted tens of thousands of international journals into ranked tiers within discipline-specific categories. It functioned, in practice, similarly to how Journal Impact Factor quartiles or CiteScore percentiles are used elsewhere — as a quick, journal-level proxy for where a given publication venue sits in perceived quality or influence within its field.

The tension this created is central to why the 2026 discontinuation is notable rather than a routine product sunset: despite the formal 2020 “break the five onlys” policy explicitly discouraging journal-based, single-metric evaluation, the CAS Journal Classification Table remained in widespread informal use across Chinese universities, hospitals, and research institutes as a benchmark for hiring, promotion, bonus pay, and degree requirements. It was never an official Ministry of Education policy instrument, but its influence on real evaluation decisions was, by most accounts, comparable to or exceeding many that were. Its continued use after 2020 was itself frequently cited as evidence that “breaking the five onlys” was easier to state as policy than to implement in day-to-day evaluation practice — the same gap between stated principle and institutional behavior that DORA’s own signatory-compliance research has documented among Western institutions.

The March 2026 discontinuation: what CAS actually announced

On 27 March 2026, the CAS National Science Library issued a formal statement that it would no longer update or release the Journal Classification Table starting in 2026, making the 2025 edition the final official version. The announcement was subsequently discussed in a guest post on the DORA (San Francisco Declaration on Research Assessment) blog, authored by Dr. Fang Xu, Deputy Director of the Research Evaluation Center at the Chinese Academy of Sciences, framing the decision explicitly as an effort to weaken “journal centrism” (期凮导向) and redirect evaluation attention toward the quality, innovation, and societal value of individual research outputs rather than the venue they appear in. The post also positioned the move as consistent with DORA’s own principles and with the broader global shift away from journal-level metrics as a proxy for individual research quality.

CAS has not named a single replacement metric or system. Coverage of the announcement describes a transition toward discipline-tailored evaluation emphasizing peer review and representative-work assessment — the same “representative works” (代表作) model that has increasingly featured in Chinese funder and institutional evaluation since the 2020 reforms, and that is conceptually close to the qualitative, narrative-CV-style assessment approaches CoARA promotes among its signatories. The STM Association separately confirmed the same 27 March 2026 announcement date and noted CAS Library’s clarification that any journal ranking published by other organizations under a similar name has no connection to the CAS Library’s own (now-discontinued) list — a detail worth knowing given how many unofficial “CAS ranking” mirrors and derivative lists circulate outside China.

Why this matters beyond China

For research administrators and institutions outside China, this development is relevant in three concrete ways:

Evaluating co-authored or collaborative work. If your institution assesses joint publications, grant applications, or tenure cases involving researchers based at Chinese institutions, criteria that referenced CAS journal tiers (directly or via journals’ own promotional claims about their CAS ranking) are losing their basis going forward. See CASRAI’s guide on US-China research collaboration and compliance concerns for the broader context of evaluating cross-border partnerships.
Understanding funder criteria. Chinese national funders, including the National Natural Science Foundation of China (NSFC), have been moving toward representative-work-based review consistent with the 2020 reform direction. If you support researchers applying for or reporting on NSFC funding, see CASRAI’s NSFC funding overview for international researchers.
Recognizing a genuinely global pattern. China’s reform is not an isolated national experiment — it sits alongside DORA, CoARA, the UK’s REF, and other reform movements documented in CASRAI’s research evaluation framework guide as part of a broad, multi-country move away from journal- and citation-count proxies toward direct, qualitative assessment of research contributions. Institutions building or revising their own promotion and tenure criteria may find it useful to compare how different systems (DORA/CoARA commitments, REF panel review, China’s representative-works model) have each tried to solve the same underlying problem — see the comparison of DORA vs. CoARA and the guide on how institutions are actually rewriting P&T criteria under DORA and CoARA.

It is also worth noting where China’s reform differs from DORA/CoARA rather than simply mirroring it: DORA and CoARA are voluntary, signatory-based commitments made by individual institutions, funders, and publishers, with no government mandate behind them. China’s “five onlys” reform is top-down state policy, issued jointly by the CPC Central Committee, the State Council, and relevant ministries, and applies across the national education and research system rather than to opt-in signatories. That structural difference affects both how quickly it can be implemented (a national directive can move faster than voluntary adoption) and how implementation is verified (compliance is an administrative/political matter rather than a self-reported signatory commitment).

What research administrators should watch next

CAS has not published a fully specified replacement methodology as of this writing, and “representative-work” and peer-review-based evaluation frameworks are, by design, harder to standardize and audit than a published journal tier list. For institutions and administrators tracking this space, the practical open questions are: what specific criteria individual Chinese universities and funders adopt to fill the gap left by the journal list; whether other Chinese organizations’ unofficial “CAS-style” rankings (which CAS Library has explicitly disclaimed) gain influence as informal substitutes; and whether this accelerates parallel moves away from journal-tier-based evaluation at the institutional level in other countries with strong bibliometric-ranking cultures.

Frequently asked questions

What is the difference between the “five onlys” policy and the CAS Journal Classification Table discontinuation?

The “five onlys” reform (2018-2020) is the underlying national policy commitment to stop evaluating researchers and institutions by single proxy metrics, including journal placement. The CAS Journal Classification Table discontinuation (announced March 2026) is a specific, later, concrete step by one influential organization — the CAS National Science Library — to retire a tool that had continued to function as exactly the kind of journal-centric proxy the 2020 policy was meant to end.

Is the CAS Journal Classification Table the same as the Chinese government’s official journal blacklist or “early warning” journal list?

No. The Journal Classification Table was a citation-based quality-tier ranking covering a broad range of international journals, produced by CAS’s own library and information center. It is a distinct instrument from China’s separate “early warning journal list” (a periodically updated list flagging journals of integrity or quality concern), which is unaffected by this discontinuation and continues to serve a different regulatory purpose.

Does this mean Chinese institutions will stop using journal-based metrics entirely?

Not necessarily, and not immediately. Discontinuing one influential ranking tool removes a widely used proxy, but it does not by itself change every institution’s internal evaluation criteria, particularly at institutions that had built promotion, hiring, or bonus formulas around specific tiers. As with DORA and CoARA adoption elsewhere, the gap between a stated policy commitment and actual changed practice at the department level is typically the slower part of the reform.

How does this compare to DORA and CoARA?

The stated goal is closely aligned — both reject using journal-level metrics as a substitute for direct evaluation of research quality. The mechanism differs: DORA and CoARA are voluntary institutional commitments with no government mandate, while China’s reform is state policy applied system-wide. See the DORA vs. CoARA comparison for how the two international frameworks themselves differ from each other before comparing either to China’s approach.

China’s “Five Onlys” Reform and the End of the CAS Journal Classification Table