Posts
The Knowledge Toll
How Academic Publishers Captured the Distribution of Publicly Funded Science — and What It Costs Everyone Else
The Bundle
In the 1990s, commercial academic publishers invented a licensing structure that made cancellation economically irrational regardless of whether any individual journal justified its cost. They called it the Big Deal. A 2014 analysis found that at many institutions, more than half the journals in a Big Deal package received zero citations from local faculty in any measured year. The subscriptions continued anyway. This post examines how the bundle works, why it holds, and how the Article Processing Charge migration preserved the revenue stream when reform pressure threatened to break it.
Before the Big Deal, university libraries subscribed to journals individually. A library would assess which journals its faculty actually used, rank them by cost-per-use, and cancel subscriptions that fell below the threshold where maintenance costs exceeded the cost of obtaining articles on demand through interlibrary loan. The individual subscription model was expensive and cumbersome, but it had one property that the publishers found intolerable: libraries could cancel.
The Big Deal eliminated cancellation as a practical option. By bundling hundreds or thousands of journals into a single multi-year licensing package — priced as a percentage increase over the institution's prior subscription spend — publishers converted a collection of individually cancellable subscriptions into a single unchancellable contract. The library that wanted the ten journals its faculty genuinely needed had to pay for the two thousand it did not need, because the ten were not available unbundled at a price that made individual subscription rational.
The bundle is the architecture's lock instrument. It is the mechanism that keeps the door closed even when individual actors can see through it clearly.
How the Big Deal Works
The mechanics of the Big Deal were developed and standardized in the late 1990s and early 2000s as publishers made the transition from print to digital delivery. Digital distribution offered genuine efficiency: instead of shipping physical journals to thousands of institutions, a publisher could provide online access to its entire portfolio at marginal cost near zero. The question was how to price that access.
The pricing structure the publishers developed reflected the foundational insight from Post 1: academic library demand is price-inelastic because cancellation carries professional costs that exceed subscription savings. The Big Deal formula typically established a base price anchored to the institution's prior print subscription spend, then applied an annual escalation rate — typically 3% to 6% above inflation — to a package covering the publisher's full journal portfolio. The institution received expanded access to a much larger number of titles than it had previously subscribed to, at a price calculated to capture most of the surplus value of that expansion for the publisher.
The contract terms reinforced the lock. Multi-year agreements — typically three to five years — prevented annual renegotiation. Non-disclosure clauses prohibited libraries from sharing their contract terms with peer institutions, preventing the price transparency that would allow libraries to negotiate collectively. Cancellation penalties and must-take provisions made mid-contract exit expensive. The bundle, the multi-year term, and the NDA together produced a structure in which the library's choice at contract renewal was not "how much should we pay" but "how much more than last time."
The Usage Data: What the Bundle Actually Delivers
The most consequential analysis of Big Deal economics was published in 2014 by Ted Bergstrom and colleagues in the Proceedings of the National Academy of Sciences. Examining subscription data from a sample of research universities, they found that a significant fraction — at many institutions, more than half — of the journals included in Big Deal packages received zero citations from faculty at that institution in any given year. The institution was paying for access to journals that no member of its research community had cited, let alone read or used.
The cost-per-use calculation that the bundle structure makes impossible to perform on an ongoing basis would, if performed, reveal that the effective price per actual use of a bundled journal far exceeds any reasonable valuation of the content accessed. The journals that are used heavily are the prestigious titles that anchor the bundle — the ones that would command high prices in an unbundled market. The journals that are unused are the filler that justifies the bundle's size and annual price escalation, contributing nothing to research at the subscribing institution while contributing to the publisher's per-institution revenue.
The bundle packages value and non-value together at a price calibrated to the value, then makes it structurally impossible to separate them. This is the instrument in precise financial terms: the publisher sells ten journals' worth of value at two thousand journals' worth of price, and the contract makes separating the two categories economically infeasible.
The NDA: Price Transparency Suppression
The non-disclosure clause deserves specific FSA attention because it is the instrument that prevents the collective action that would otherwise constrain the bundle's pricing power. Academic libraries are not commercial competitors. They are public and nonprofit institutions whose aggregate subscription spending represents an enormous and largely coordinated potential market. If libraries could share their contract terms, they could identify above-market rates, negotiate collectively, and establish price floors below which publishers would lose enough contracts to threaten their revenue base.
The NDA prevents this. A library that signs a Big Deal contract agrees not to disclose its terms to peer institutions. The publisher negotiates with each institution individually, knowing its own full range of contract terms and knowing that no library knows any other library's deal. The information asymmetry is total and structural. The publisher holds complete price information across all contracts. Each library holds information about only its own. In a market where price discovery is the prerequisite for competitive negotiation, the NDA eliminates price discovery for the buyer while preserving it for the seller.
The Scholarly Publishing and Academic Resources Coalition maintains a Big Deal Cancellation Tracker and has advocated for NDA elimination. Some institutions have challenged NDA clauses on public transparency grounds. The publishers have resisted consistently, understanding precisely what price transparency would cost them.
The Major Publishers: The Concentration That Makes Exit Impossible
| Publisher | Journals (approx.) | Bundle Structure | Notable Cancellations / Disputes |
|---|---|---|---|
| Elsevier (RELX) | ~2,500+ | ScienceDirect Freedom Collection; multi-year NDAs | UC system 2019–2021; MIT 2020; multiple European consortia |
| Springer Nature | ~2,900+ | SpringerLink packages; Nature portfolio as anchor | German DEAL consortium multi-year negotiation; Swedish consortium cancellation |
| Wiley | ~1,700+ | Online Library packages; multi-year agreements | Finnish consortium cancellation 2020; California State University system |
| Taylor & Francis | ~2,700+ | Online bundling across subject collections | Various consortium renegotiations post-2018 |
The concentration of this market means that a university library cannot exit the bundle structure without denying its researchers access to a substantial fraction of the literature in their fields. A library that cancelled all four major publisher packages would leave its faculty unable to access the majority of recently published peer-reviewed research in virtually every discipline. No research university has taken that step. The bundle's lock operates precisely because the concentration makes the exit cost professionally and institutionally prohibitive.
The APC Migration: When Reform Becomes Revenue
The open access movement — examined in detail in Post 3 — has produced a response from commercial publishers that FSA identifies as an adaptation instrument rather than a genuine reform. The Article Processing Charge, or APC, is a fee paid by authors, institutions, or funders to make a specific article freely available online at publication without an embargo period.
In principle, the APC represents a model shift from reader-pays to author-pays. In practice, at major commercial publishers, the APC has been layered on top of the subscription model rather than replacing it, producing what the library community calls the double-dipping problem: universities pay subscription fees to access the journal's paywalled content, and also pay APCs when their researchers want specific articles made open access within the same journal. The publisher collects twice from the same institution — once for access, once for openness.
APC prices at major commercial journals range from approximately $2,000 to $15,000 per article. At Elsevier's flagship journals, APCs for open access publication can exceed $10,000. The estimated production cost of formatting and hosting a digital academic paper runs well under $1,000 by independent analysis. The APC premium above production cost is, like the subscription premium above distribution cost, a return on channel ownership and brand prestige — not on service delivered.
Transformative Agreements: Reform as Repackaging
Between approximately 2015 and 2024, the dominant instrument through which universities and publishers attempted to resolve the tension between subscription models and open access mandates was the transformative agreement — a contract structure combining subscription access with a commitment to transition toward open access over time, funded through a combined read-and-publish fee.
The cOAlition S funders — the European research funding consortium behind Plan S — supported transformative agreements through December 2024, treating them as a legitimate transition mechanism. They ended that support explicitly because transformative agreements had not produced meaningful transformation. Publishers had used them to preserve and in some cases increase revenue while appearing to comply with open access mandates. The instrument had been captured by the architecture it was designed to reform.
The cOAlition S 2026–2030 strategy pivots explicitly away from transformative agreements toward diamond open access — where publication and access are both free, with costs borne by funders or institutions rather than captured by commercial publishers — and toward preprint infrastructure that bypasses the commercial distribution monopoly entirely.
Post 3 examines what that bypass looks like in practice: arXiv, bioRxiv, the NIH zero-embargo mandate, and the open infrastructure movement building the alternative system the commercial publishers' monopoly depends on not existing.
- Big Deal: multi-year bundle; individual cancellation unavailable at rational unbundled price
- Bergstrom et al. (PNAS 2014): zero-citation journals in bundles at majority of sampled institutions
- NDA clauses: standard in Big Deal contracts; suppress inter-institutional price transparency
- SPARC Big Deal Cancellation Tracker: institutional cancellations and NDA challenges (sparcopen.org)
- APC range: ~$2,000–$15,000+ per article; Elsevier flagship journals above $10,000
- Double-dipping: subscription plus APC from same institution — documented pattern
- Transformative agreements: cOAlition S support ended December 31, 2024
- cOAlition S 2026–2030: diamond OA, preprints, open infrastructure — published November 2025
- Market concentration: Elsevier, Springer Nature, Wiley, Taylor & Francis dominate peer-reviewed journals
The complete terms of Big Deal contracts — precise annual escalation rates, cancellation penalty structures, and the full scope of NDA provisions — are not publicly available for the majority of institutions. SPARC and individual institutional disclosures provide data points; a comprehensive national database of Big Deal terms does not exist in accessible public records.
The aggregate annual revenue Elsevier, Springer Nature, and Wiley collect specifically from Big Deal subscription contracts — as distinct from APC revenue, transformative agreements, or other streams — is not disaggregated in their public financial reporting in a way that allows precise calculation.
The internal publisher deliberations on APC pricing strategy — specifically whether prices above production cost are calibrated to maximize revenue extraction from institutions with open access mandates — are not public. The double-dipping pattern is documented in aggregate; the intent behind it is beyond the wall.
Primary Sources · Post 2
- Bergstrom, Ted C. et al., "Evaluating Big Deal Journal Bundles," PNAS 111(26), 2014
- SPARC Big Deal Cancellation Tracker — sparcopen.org; institutional cancellation data and NDA documentation
- cOAlition S announcement ending transformative agreement support, 2023 (coalition-s.org)
- cOAlition S Strategy 2026–2030, November 12, 2025 (coalition-s.org)
- UC Office of Scholarly Communication — Elsevier contract 2019 cancellation; 2021 renegotiation
- MIT Libraries — Elsevier contract cancellation 2020; public statement and rationale
- German DEAL consortium — Springer Nature and Wiley negotiations; outcomes documentation
- Elsevier APC rate schedules — elsevier.com; flagship journal open access pricing
- Frazier, K., "The Librarians' Dilemma," D-Lib Magazine (2001) — early Big Deal economics documentation
- Suber, Peter, "Open Access" (MIT Press, 2012) — subscription vs. OA economics analysis
No comments:
Post a Comment