The Role of University Consortia in the 21st Century
Ask an international office director why their university belongs to three or four different consortia at once, and the answer is rarely a tidy one. Some memberships were inherited from a predecessor. Some exist because a rival institution across town joined first. A few are genuinely load-bearing — they generate real mobility numbers, real joint grants, real research output. Most sit somewhere in between: not dead weight, not essential, just there.
That ambiguity is worth taking seriously, because the multi-institution consortium has become one of the default organisational forms in international higher education, and yet it is rarely examined on its own terms. It gets discussed as a subset of "internationalisation strategy" or folded into rankings methodology, but the consortium itself — as a structure, with its own incentives and failure modes — deserves closer attention.
A Structure Built for a Specific Problem
The basic problem a consortium solves is coordination cost. A university that wants to run a joint PhD programme, a research exchange, and a student mobility scheme with twenty different partner institutions faces twenty different negotiations, twenty different administrative relationships, twenty different sets of paperwork. A consortium collapses that into one relationship with a shared secretariat, shared governance, and (usually) a shared calendar of forums and calls for proposals.
This is not a new idea — regional and disciplinary networks existed well before the term "internationalisation" entered university strategic plans in the 1990s. What changed over the following two decades was scale and ambition. Networks moved from being loose alumni-style associations to structures with dedicated staff, formal MOUs, member fees, and measurable output targets. AC21, founded in 2002, sits squarely in that second generation: a consortium built around a small, deliberately curated membership rather than an attempt to maximise reach.
The contrast with larger networks is instructive. Universitas 21, founded a few years earlier, now counts 27 research-intensive universities across six continents. The Coimbra Group, focused on European research universities, has grown to 41 member institutions. Both models work, but they solve a different problem — breadth of reach and collective bargaining power with governments and funders — than a consortium built around fifteen or twenty institutions that know each other well enough to run genuinely joint programmes rather than parallel ones.
What Consortia Actually Produce
Strip away the language of "strategic partnership" and most consortium activity falls into three buckets. First, mobility programmes: structured, often short-form, exchanges for students and early-career researchers that would be logistically painful to arrange bilaterally. Second, seed funding for joint research, usually modest amounts intended to generate preliminary data for a larger external grant application rather than to fund a project outright. Third, recurring forums — conferences, symposia, leadership summits — that exist partly for academic content and partly, frankly, to keep the relationships warm between formal collaborations.
None of these functions is glamorous. That is precisely why consortia are so easy to underrate from the outside and so hard to walk away from once you are inside one. The university networks that make up a consortium like AC21 rarely justify their membership fee through any single flagship project. The value, when it is real, tends to be cumulative — a PhD student who found a co-supervisor at a forum three years earlier, a grant proposal that succeeded because two labs had already run a pilot study together.
This is also where consortia become vulnerable to a fair criticism: it is difficult to measure a benefit that shows up as "the relationship existed when we needed it" rather than as a line item. University leadership, especially under budget pressure, tends to ask for numbers. Consortia that cannot produce them — publication counts, mobility statistics, external funding leveraged — are the first to be quietly deprioritised at renewal time.
The Forum as Load-Bearing Infrastructure
It is worth dwelling for a moment on the recurring forum, because it is the part of the model most often dismissed as ceremonial. In our view, that dismissal misses what the forum is actually doing structurally. A biennial gathering of senior academics and administrators, of the kind AC21 ran through its International Forum series, is not primarily a venue for presenting research. Most of what gets presented at these events has already been published, or will be published somewhere with a higher citation count than the forum proceedings.
What the forum actually does is refresh the informal trust network that everything else depends on. Grant collaborations, joint supervision arrangements, and ad hoc research partnerships are built on personal relationships between individual academics far more than on institutional MOUs signed by a rector. When those relationships lapse — because people change jobs, because a global disruption cancels three years of travel, because budget cuts shrink international office staff — the consortium's practical value erodes even if the legal structure remains intact.
This is arguably the central lesson of the pandemic period for consortium-style networks generally: the ones that had built durable habits of in-person contact took the longest to recover their working relationships once travel resumed, precisely because so much of the coordination had depended on people simply seeing each other regularly.
Why Some Consortia Wind Down
Consortia do not usually collapse dramatically. They wind down the way most institutional structures wind down: budget lines get reallocated, a founding champion retires or moves institutions, and a gap opens between the administrative cost of running the secretariat and the visible output it produces. AC21's own wind-down after 2022, discussed in more detail on our about page, followed roughly this pattern rather than any single dramatic event.
That is not, to be clear, an argument that the consortium model has failed as an approach to international cooperation. It is closer to an argument that consortia have a natural lifespan tied to the institutional energy of the people who built them, and that treating a twenty-year run as evidence of failure rather than as a reasonably long institutional life is the wrong frame. Very few voluntary, fee-funded, secretariat-run networks in any sector last much longer than that without either formalising into something closer to a permanent international organisation or dissolving once their founding cohort moves on.
What Comes Next
The underlying need that consortia address — coordination cost across institutional borders — has not gone away, and if anything has grown as research becomes more international and more expensive to run bilaterally. What has changed is the range of tools available to meet that need. Some of the coordination that once required a formal consortium and an annual forum can now happen through lighter-weight, discipline-specific networks, or through the kind of ad hoc collaboration that digital tools make easier to sustain without a shared secretariat at all.
Whether that makes the traditional consortium model obsolete, or simply changes what it needs to deliver to justify itself, is not a question with a clean answer yet. What does seem clear from looking at a case like AC21's is that the model's value was never really about the size of the network. It was about whether the relationships it maintained were strong enough to survive the gaps between forums — and specific enough that individual researchers and students could point to something concrete they got out of it.