Sustainable Scholarly Publishing Carbon Accounting Standards Need a Shared Ledger

The most radical document in scholarly publishing right now is not a journal editorial, a funder mandate, or a new impact metric. It is the emissions ledger attached to a publisher's annual report, the place where carbon accounting stops being a corporate slogan and starts being an audit of who publishes what, at what cost, and at whose expense. Scholarly publishing spent a decade treating sustainability as a paper-and-conference problem. The next argument is about standards, because a carbon figure without a shared standard is not evidence. It is a marketing choice.

Sustainable scholarly publishing carbon accounting standards sound like a compliance backwater. They are closer to a mirror held up to the business model. If the ledger counts only the publisher's office electricity, it misses almost everything that makes academic publishing unusual. If it tries to count the emissions of the research itself, it drags the publisher into laboratory freezers, supercomputers, and a researcher's flight to a society meeting.

The carbon ledger nobody can compare yet

Commercial publishers have published sustainability reports for years. Universities have signed net-zero commitments. Yet two publishers' carbon disclosures are rarely comparable. One publisher may report tonnes of carbon dioxide equivalent, or CO2e, from its offices and data centres. Another may include author travel to editorial board meetings. A third may quietly leave out the cloud infrastructure that runs its submission platform.

The result is a shelf of reports that look impressive and reconcile with nothing. A procurement officer comparing two publishers cannot tell whether a lower figure reflects cleaner operations or a narrower definition of the word operations. Carbon accounting standards exist to close exactly that gap.

The three scopes leave plenty of room

The Greenhouse Gas Protocol's Corporate Standard, first published in 2001 and revised in 2004, defines the basic structure. Scope 1 covers direct emissions from sources a company owns or controls, such as a fleet of vehicles or a gas boiler. Scope 2 covers indirect emissions from purchased electricity, heat, and cooling. Scope 3 covers everything else in the value chain, from business travel and purchased goods to downstream distribution.

For a scholarly publisher, scope 3 is where the accounting gets strange. A journal article has no physical freight, but it has server time, editorial software, reviewer connectivity, and sometimes a print run. It may sit behind a paywall or move through an open access platform. The Scope 3 Standard, published in 2011, asks companies to report material categories. The problem is that materiality in scholarly publishing remains undefined. What is material: the electricity used to store a manuscript, or the emissions of the research that produced it?

Why scholarly publishing's scope 3 is unusually slippery

Most industries can draw a line around their value chain and call it done. A publisher of scientific journals cannot do that without making a philosophical decision. The research article exists only because a lab ran experiments, a research group analysed data, and at least one author travelled somewhere to present preliminary findings. Those emissions are not owned by the publisher. They are also not incidental to the product. The article is the condensed record of them.

This boundary problem explains why carbon accounting for scholarly publishing keeps stalling. Include research emissions and the publisher's report becomes a commentary on laboratory science. Exclude them and the report becomes a narrow audit of office buildings and manuscript systems. Neither version gives a reader a meaningful per-article footprint.

The per-article figure is where the standard would either become useful or collapse into noise. A single article can require months of high-performance computing, international collaboration, and repeated resubmission across journal platforms. The publisher controls very little of that. But if publishers do not count it, the carbon number attached to an article tells researchers nothing about their own choices.

A standard borrowed from finance, not climate activism

The International Organization for Standardization publishes ISO 14064-1:2018, which specifies principles for organisation-level greenhouse gas quantification and reporting. The Science Based Targets initiative, founded in 2015, pushes companies to commit to reductions aligned with climate science. These frameworks come from the world of financial audit. They assume that a verifiable number is better than a vague promise, and that a company should report its boundary, methodology, and assumptions in public.

Scholarly publishing has borrowed the vocabulary without always borrowing the discipline. A publisher can say it is aligned with the Greenhouse Gas Protocol while still choosing a narrow scope 3 boundary. A university press can publicise a net zero target while relying on purchased credits that do not change its underlying energy use. The standard does not fail there. The failure is the absence of an academic publishing-specific boundary that makes the number audit-proof.

The same procurement logic appears in APC pricing disputes, where rising APC costs and waiver activism have forced libraries to ask what exactly they are buying. Carbon accounting adds a second question. What exactly is being burned to sell it?

Here's the catch

Carbon offsets are the easy way out. A publisher can buy renewable energy certificates, claim a lower scope 2 figure, and leave its air travel untouched. That is not fraud. It is accounting. But it rewards the party with the largest travel budget and the most aggressive marketing department. A standard that accepts offsets without requiring absolute reductions becomes a pricing mechanism for guilt, not a tool for change.

Hype says publishing can paper over the problem with better reporting. Reality says reporting without a common boundary is a public relations exercise. The most accurate carbon ledger in the world still tells you nothing if every publisher draws the line in a different place.

A credible standard would have to do three uncomfortable things. It would define which scope 3 categories scholarly publishers must include. It would require independent verification of raw energy and travel data. It would require publishers to disclose what share of their reduction comes from operational change rather than purchased credits.

None of those is technically impossible. Financial auditors already verify more complex supply chains. The resistance is commercial, because a shared boundary would let librarians and funders compare publishers on a like-for-like basis for the first time.

What a credible standard would actually require

First, a recognised boundary. A scholarly publisher should account for office energy, data centre energy used by manuscript systems, business travel, and the distribution of digital content. It should state whether the emissions of peer review and editorial workflows are included. If a category is omitted, the omission should be justified, not buried in a methodology note.

• Scope 1 and 2 data should come from metered or invoiced energy use, not estimates.
• Scope 3 travel data should cover flights, rail, and hotel stays associated with editorial work and publisher-run events.
• Digital infrastructure should be reported by service provider, region, and renewable energy share.
• Offset-based reductions should not be presented as equivalent to operational reductions.

A standard built on those rules would give universities a basis for asking publishers the same question they now ask about subscription pricing. It would also expose a deeper tension. Open access publishing, often presented as the fairer model, has its own carbon profile. APCs, submission systems, and preprint servers all consume energy. The argument is not that one publishing model is clean. It is that none of them has been measured honestly enough to know.

The researcher's ledger

The researcher who reviews manuscripts and publishes articles rarely appears in the publisher's carbon ledger, because unpaid labour has no business travel line. That omission is a precise demonstration of the problem. Peer review consumes electricity, bandwidth, and working time. It is materially necessary to production, yet it sits outside the company's formal value chain. The researcher is simultaneously a supplier, a customer, and an externality.

One researcher put the stakes more plainly than any standard-setter. The question is not whether a publisher can print a lower number. It is whether the number changes what a research group decides to do next week: which conference to skip, which dataset to store, which article to publish under what terms. Until the standard reaches that decision, it is reporting. Not reform.

Photo by Oleg Avramenko on Unsplash