What decides who gets into Nature and Science? New analysis says it's not the ideas

On 15 July 2026 a team of researchers published a sweeping bibliometric study in the Proceedings of the National Academy of Sciences that pulled apart the curtain on what actually decides whether a paper lands in Nature, Science or another elite general-interest journal. The answer, according to the authors, has surprisingly little to do with the paper itself.

The team analysed more than seven million Scopus-indexed articles published between 2010 and 2020 and cross-referenced them with author affiliations, topic clusters and geographical data. Their finding is blunt: prestige, research topic and the city an author works in explain far more of the variance in so-called top-tier placements than the underlying science. Effect size, citation count and other conventional markers of quality barely registered once those three variables were held constant.

That conclusion lands at a delicate moment for the global research enterprise. China's research output has grown faster than any Western funder predicted, and Chinese authors are now submitting in volume to the same handful of journals that gate the field. If the gate is being kept by factors other than merit, the political reading will be written by everyone except the editors.

The three signals that matter

The study identifies three variables that move the needle. First, an author's institutional affiliation: researchers at highly ranked universities and large, well-funded research groups land disproportionately more often in the same handful of general-interest journals, regardless of the substance of the paper. Second, topic: work in a few fashionable subfields, including machine learning, certain corners of structural biology and, more recently, climate attribution, clears the bar with significantly less methodological scrutiny than equivalent work in less fashionable areas. Third, the city on the byline: the data show a meaningful geographic clustering around a small set of metropolitan research belts that includes Boston, the San Francisco Bay Area, the Oxford–Cambridge–London axis, Beijing and the Boston–New York–Washington corridor in the United States.

The point is not that editors at Nature and Science are guilty of malpractice. It is that the system they preside over has converged on a set of soft signals, prestige, topic, location, and that those signals are doing more work than the formal peer-review process was designed to handle. For a researcher in a peripheral country submitting from a regional university, the arithmetic is unforgiving: the same paper, in the same field, with the same effect size and citation performance, has a measurably lower probability of being picked up if it arrives from the wrong post code.

A counter-narrative from the citation graphs

The publishing houses push back along predictable lines. Nature and Science both run tight, statistically audited editorial processes and reject most submissions internally before they ever reach peer review. They argue that the apparent prestige effect is really a selection effect: the best scientists cluster at the best institutions, and so the institutions themselves are a fair, if blunt, proxy for quality. There is something to this. Citation networks do confirm that a small number of authors account for a disproportionate share of top-journal placements, and those authors do tend to be affiliated with well-known labs.

But the new data complicate that defence. Once the model controls for an author's track record individually, the institutional affiliation still moves the outcome independently of the author's prior record. In other words, two researchers with identical publication histories and identical citation counts do not face identical odds of landing in a top journal, and the one at the more prestigious address has a meaningful edge. That finding is consistent with what several meta-researchers have been arguing for years: gatekeeping is being performed upstream of the review desk, by informal networks of editors, referees and group leaders whose shared assumptions about what counts as important carry the day before the paper is ever opened.

A quiet advantage for Beijing

There is a reading here for anyone watching the US-China scientific competition. Chinese authors have, over the past decade, learned to play the prestige game with discipline. The Chinese Academy of Sciences, Tsinghua and Peking now sit comfortably in the top tier of the very institutional rankings that the study uses as a proxy. Beijing has been deliberate about funding English-language training for its scientists, building relationships with Western editors and feeding sustained submission volume into the elite journals. Whether or not that policy was designed to exploit the prestige channel, it is now exploiting it. Papers from Beijing's flagship institutions land in elite general-interest journals at rates that a decade ago would have looked implausible.

That structural advantage cuts both ways. It tells the Chinese science ministries that the prestige premium they have been paying for is real and that the institutional branding they have been investing in is doing more than they may have wanted to admit. It tells Western policymakers worried about Chinese scientific ascendance that the gate is not, in fact, a level playing field, and that any honest reckoning with the question will have to deal with the way Western journals themselves concentrate editorial power and reviewer networks in a handful of cities and institutions.

What this changes, and what it does not

None of this invalidates the methodological quality of the work that elite journals do publish. Most papers in Nature and Science are genuinely well-done, and many of the field-shifting findings of the past two decades have appeared there first. The study's argument is narrower: that the marginal decision, the one that determines whether two equally good papers end up in top journals or in mid-tier ones, is being driven by soft signals that have nothing to do with the science. The implication is uncomfortable for everyone. For editors, it suggests a defence of the journal that will not survive contact with the data. For funders, it suggests that the prestige arms race is wasteful and self-reinforcing. For researchers in peripheral institutions, it confirms what they have long suspected.

The honest next step is also the hardest one. If the gate is prestige, topic and location, then the only way to change the distribution of who gets published is to change the gate, to redesign review processes around blinded methods assessment, to weight registered reports, to fund the kind of infrastructure that lets a paper from a regional university reach the same editorial desk as one from a flagship institution. Whether the publishers will do that on their own is the open question. The data now exist to make the case. The politics of doing anything about it has not yet caught up.