Great questions — and ones where your position as an academic researcher gives you a useful vantage point to evaluate what I'm about to lay out. I've pulled in current sources to supplement what I know. Here's a thorough treatment.
1. Industry Landscape & Major Players
The academic publishing market generated roughly $34 billion in 2024, with North America accounting for about a third of that. Journals represent ~68% of the market by content type, and STM (science, technology, and medicine) dominates at ~77% by subject area.
The industry is a tight oligopoly. The "Big Five" — Elsevier (owned by RELX), Springer Nature, Wiley, Taylor & Francis (owned by Informa), and SAGE — collectively publish roughly half of all research output. Elsevier alone is by far the largest; RELX reported profits of roughly $3.6 billion in 2023. Beyond the Big Five, important players include Oxford University Press, Cambridge University Press, Wolters Kluwer, and various society publishers (ACS, IEEE, APA, etc.) that occupy influential niches.
A key structural feature: the industry is consolidating, not fragmenting. In May 2025, BC Partners launched a dedicated acquisition fund for Springer Nature to fuel further growth through mergers. Wiley announced six new journal launches for 2026. The big players are getting bigger.
On the other side, a growing counter-movement of scholar-led, non-profit publishing platforms (think PLOS, eLife's new model, various Diamond OA journals, and preprint servers like arXiv and SSRN) is gaining traction — but hasn't yet meaningfully dented the oligopoly's market share or prestige.
Confidence: High. Market concentration data is well-documented in bibliometric studies (e.g., Larivière et al., 2015, and subsequent updates). Revenue figures come from public financial filings.
2. Business Model & the "Moat"
This is where it gets interesting — and where the economics are genuinely unusual.
The core business model has historically been subscription-based: publishers package journals into large bundles ("Big Deals") and sell institutional access to university libraries. Subscription-based publishing still holds ~53% market share, though this is declining as open access grows. The emerging alternative is the author-pays model (Article Processing Charges / APCs), where researchers or their funders pay to publish, and the article is freely available. APCs range from a few hundred dollars to $12,000+ at top-tier journals.
Profit margins are extraordinary — averaging 30–40% across the industry, with Elsevier consistently near 40%. For context, that's higher than most tech companies, and comparable only to pharmaceuticals and tobacco. These margins have been increasing even as the cost structure has shifted dramatically toward digital (which is far cheaper than print distribution).
The moat has several reinforcing components:
Free labor from academics. This is the single most important structural advantage. Researchers write the articles for free, peer reviewers evaluate them for free, and many editors serve for free or for nominal honoraria. The publisher's value-add is essentially brand management, production infrastructure, and distribution — but the core intellectual labor is donated. No other industry I'm aware of has this structure at this scale.
Prestige lock-in and the tenure system. The "publish or perish" incentive structure means researchers need to publish in high-impact, well-known journals to advance their careers. This creates inelastic demand — researchers will submit to Nature, Science, or top field journals regardless of the publisher's practices, because the reputational signal is what matters for hiring, tenure, and grants. The prestige hierarchy is self-reinforcing: journals are prestigious because good researchers publish there, and good researchers publish there because the journals are prestigious.
Bundling and library lock-in. The "Big Deal" bundling strategy ties libraries into multi-year contracts covering hundreds or thousands of journals at a discount. This makes it very difficult for libraries to cancel individual titles or switch to competitors, and it creates massive barriers to entry for new publishers. Libraries face the unenviable choice of paying escalating fees or losing access to content their faculty need.
Copyright ownership. Traditionally, publishers require authors to transfer copyright as a condition of publication. This gives publishers perpetual control over the dissemination of publicly funded research — a structural advantage that open access mandates are now challenging.
Network effects in indexing and discovery. Major publishers control or are deeply integrated into the discovery infrastructure (Scopus is owned by Elsevier, for instance). Being indexed in the right databases matters enormously for visibility and citation, which reinforces the prestige cycle.
Confidence: High on the structural description. The profit margin data comes from public financial filings and has been confirmed by multiple independent analyses. The moat components are well-documented in the academic literature on scholarly communication (e.g., Buranyi 2017 in The Guardian, the PLOS ONE political economy paper, and various library science analyses). The one area of moderate uncertainty is how durable these advantages are — more on that below.
3. Additional Context You Should Know
The open access transition is real but not straightforwardly disruptive. The major publishers have proven remarkably adept at adapting to OA mandates by simply shifting who pays rather than reducing revenue. The move from subscription fees to APCs preserves (and in some cases increases) revenue. Springer Nature reported that 50% of their primary research articles were OA in 2024, and revenue forecasts are up. The publishers are not being disrupted by OA — they're co-opting it. This is a critical insight: the moat isn't just "subscriptions," it's the prestige hierarchy and the captured labor market.
The NIH/federal policy landscape is in flux. In 2025, NIH moved to require immediate public access to federally funded research (no embargo period) and proposed capping APCs at $2,000–$6,000. This is the most direct threat to publisher margins I'm aware of, because it attacks the pricing power on the APC side. Over 900 stakeholders submitted comments, many expressing concern that caps could block access to prestigious journals. Coalition S (the European funder consortium) has also softened its approach in its 2026–2030 strategy, retreating from strict requirements — which some interpret as an acknowledgment that hard mandates had the unintended consequence of entrenching the author-pays model.
AI is a wild card. Publishers are integrating AI into peer review triage, plagiarism detection, and production workflows. But AI also threatens the value proposition in subtle ways: if AI can generate competent literature reviews and even draft papers, the sheer volume of submissions will increase (it already has — a 23% increase in published articles over the last five years), putting further strain on the peer review system while potentially devaluing individual publications.
The peer review crisis is serious. Reviewer burnout is accelerating, and the volunteer model is under real strain. Finding reviewers takes longer, turnaround times are increasing, and quality control is slipping in some areas. Paper mills — operations that sell fraudulent authorship on fabricated papers — are a growing problem, and generative AI is making detection harder. Cambridge University Press recently found that only one-third of stakeholders believe current reward/recognition systems are working well.
The equity dimension matters. APCs create a two-tiered system where publication is increasingly contingent on institutional wealth. This disproportionately affects researchers in the Global South, at teaching-focused institutions, and in underfunded fields (which, given your focus on education research, is directly relevant — education journals generally have lower APCs than biomedical journals, but the trajectory is concerning).
An analogy that might be useful: I'd characterize the academic publishing oligopoly as structurally similar to credit rating agencies (Moody's, S&P, Fitch). Both industries feature a small number of firms whose power derives not from the quality of their product per se, but from the fact that their ratings/brands are embedded in institutional decision-making processes (tenure committees, library procurement, grant evaluations). Disrupting either requires not just a better product, but changing the incentive structures of the institutions that rely on them.
Confidence on the analogy: Moderate — it's my interpretation, not established analysis, but I think the structural parallels are strong.
Confidence on the policy/AI/equity points: Moderate to high. The policy facts are well-sourced from Science/AAAS and NIH announcements. The AI and equity assessments involve more forecasting and are therefore less certain.
Bottom line: The academic publishing industry is a remarkably profitable oligopoly whose moat is built primarily on prestige capture, free academic labor, and institutional lock-in — not on the quality or cost of the services provided. Open access is reshaping the revenue model but not (yet) the power structure. The most potent threats are federal APC caps, the peer review labor crisis, and the longer-term possibility that tenure/promotion committees begin to deweight journal prestige — though that last one has been predicted for decades without materializing.