The powers of library consortia 1: How consortia scale capacity, learning, innovation and influence

This is the first of 4 blog posts about library consortia. I will link to the others here as they appear:

  1. The powers of library consortia 1: How consortia scale capacity, learning, innovation and influence
  2. The powers of library consortia 2: Soft power and purposeful mobilization: scaling learning and innovation
  3. The powers of library consortia 3: Scaling influence and capacity
  4. The powers of library consortia 4: Scoping, sourcing and scaling

Scaling capacity, learning, innovation and influence

Libraries and related organizations group together in a variety of ways to get their work done. They consort where there are scale advantages: to lobby, for example, to negotiate and license, to reduce costs, or to build shared infrastructure. The ‘soft power’ of such groups is also important – the relationship building, the trust, the sharing of learning and innovation that comes from working together over time is an important reason that such groups persist.

In each case, libraries recognize that collective action or building shared capacity delivers scale benefits.


In this context, it is interesting to think about collaboration for scale in four areas: influence, learning, innovation and capacity. Some collaborative organizations span several of these areas, some are more specialized. In all these areas consortial or collective activity potentially creates scale advantages.

Organizations which primarily scale influence have an advocacy or representative role, and are often at national (e.g. ALA, RLUK) or international (e.g. LIBER, IFLA) levels, or come together around particular issues (e.g. SPARC around open access). A regional dimension is often very important for organizations which scale capacity (e.g. OhioLINK or OCUL), either for logistical reasons or because there is public support (at the level of a state or province, or at national level). Other organizations may pull together peer groups (e.g. PALNI or Ivies Plus) or particular functional or interest groups (e.g. DPN or Duraspace). And in an environment of evolving user expectations and behaviors, there is a growing recognition that libraries need to think purposefully about shared approaches to learning and innovation which better prepare them to respond to new needs. Some groups are focused on learning and innovation (e.g. CNI or the Digital Preservation Coalition), while it is also increasingly an important part of all groups.

Of course, groups come into existence and persist for different reasons and in different ways. But they often have overlapping roles and individual libraries typically support several. The organizational context of collaborative or consortial activity is surprisingly little explored in the library literature, in terms of incentives, organizational patterns, successful models, etc. There is certainly a descriptive body of work about classic library consortia (such consortia are often organized around some combination of licensing, resource sharing and training, or sometimes manage a shared library system), but the more general perspective is missing. It is this gap which prompts this series of blog entries.

In four blog entries I am going to consider library consortia or collaborative organizations using this framework. I have two purposes:

  1. The first is to ask some questions about the shape and dynamics of consortia, in order to contribute to thinking about an important and underexamined topic. Clearly, this can only be preliminary and suggestive in this form; I offer some observations, rather than a full grounded analysis. I use some simple heuristic frameworks to structure my discussion (scaling influence/learning/innovation/capacity and scoping/sourcing/rightscaling), while acknowledging that a fuller treatment might generate better approaches. I would note in particular that there is a large literature on organizations across several disciplines which would provide a valuable context for the examination of collaborative library work.
  2. The second is to underline how scaling learning and innovation have become important shared concerns in complex times, and deserve increased investment of both attention and resources.

In this introductory post I put consortial activity in the context of other approaches to scaling, and I touch on some general issues that will be discussed in more detail in later posts. In particular I am interested in scoping, sourcing and (right)scaling, three ways in which choices about collaborative activity need to be made.

Patterns of scaling

Of course, collaborative activity is only one way in which libraries acquire shared services or connect to broader resources, or in other words scale their capabilities. Reductively, one can point to three patterns which deliver shared library services at a level above the institution.

  1. Public – Shared infrastructure or services is provided through a public agency or with public support, often involving national (or regional) coordination. Think variously of services historically provided by Jisc in the UK, or the regional bibliographic networks in Germany, or Libraries Australia, or Bibsys in Norway. This is a common pattern outside the US, although within the US there is often some support through state-supported agencies (e.g. OhioLINK).
  2. Collaborative – A group of libraries works together to get something done. Formality, size and scope vary enormously. This collaboration often takes the form of a consortium, and libraries support many consortia.
  3. Third party – A shared service can be bought from a provider (publisher, system vendor, aggregator, etc). Consultancy is another example of a third party service, as a way of scaling learning by tapping into broader expertise or experience. Of course, this is a major way in which libraries scale their services – for example this is largely how they acquire the journal literature, automation, and other central services.

These three patterns may be layered or blurred in various ways. A library may collaboratively source a service through a consortium, but the consortium may then contract with a third party to actually provide it (a shared library management system for example, or a digital preservation service). In fact, one important function of collaborative or public provision is to efficiently or cost-effectively procure shared services from third parties. Publicly supported groups may have a strong collaborative element, or have contributions from members; OhioLINK is an example here. Jisc in the UK describes itself as a ‘membership’ organization – it is moving from its centrally publicly funded model to one in which universities and others subscribe to benefit from services.

So, while library collaboration (typically through consortia) is my major focus here, intersections with public and third party provision are frequent.

Contrasts in consortial activity

The variety of library collaborations is great, and this translates into great variety of consortial organizations and directions. These also change over time. By way of example, I note three important contrasts about consortial activity here.

  • Single-purpose or multi-purpose. Some groups may specialize around one function or direction. SPARC for example is focused on an open agenda. HathiTrust is focused on preservation of and access to the digitized collections of its members, in support – more broadly – of the preservation of the scholarly and cultural record. However, other groups may do many things. Take PALNI for example. PALNI is a consortium of private academic libraries in Indiana, and provides a range of services around a deep collaboration focus – sharing infrastructure, expertise, and other capabilities of member libraries. PALNI will adapt to support the changing priorities of its members, and so, for example, it recently developed programmatic support for digital scholarship. In this way, there may be an interesting balance between economies of scope (where, for example, it makes sense for members to centralise a variety of different functions with a consortium, including shared infrastructure and attention to evolving service areas) and economies of scale (where the costs of providing a service, including infrastructure, marketing, R&D, etc, are spread across many participants, as in, for example, HathiTrust). Of course, many third-party providers exploit economies of scale to deliver services in focused areas (e.g. cloud system providers, aggregators, publishers). A related issue here is the relative homogeneity of members. The intersection of member interests bounds the scope of the consortium. Where a consortium comprises members of different types (large, small; academic, public) then the intersection of interests may need be quite tightly scoped (resource sharing, for example) to achieve desired advantages and the consortium may be focused on those. Where consortium members form more of a peer group, then the intersection of interests is greater, and the consortium may move shared concerns forward across a broader range.
  • Agent or provider. There is no one model of collaboration. Libraries may work together informally to get things done. They may add progressively more structure: convene member working groups, hire staff, hire an executive director, constitute the collaboration as an independent entity. Of course, once the collaboration is structured as an independent entity, it has to sustain itself and issues of alignment and governance come to the fore. Over time, this may become more important as original personalities or organizational contexts may change, new requirements may emerge, activities may grow or become more diverse. As the organization grows, it may outgrow the ability or interest of members to entirely support its activities it through dues and it may seek to diversify revenue streams by offering additional services, applying for grants, and so on. This means that the balance between the group as a direct agent of the member library interest, a collective resource to be directed, as it were, and the group as a provider to the libraries, independently developing potential products, can change. As a provider, the consortium needs to supply support and training, upgrade the service in line with user expectations, and generally offer the support that libraries might expect from third party providers. This means in turn that services are potentially compared with those of third party providers – think of, for example, of a shared institutional repository provided by the consortium, and the comparison to, say, bepress. This service relationship can change the dynamic between members and consortium. Size is an important factor here also, given that as an organization grows it may be difficult for all members to directly participate in decision making. One important dynamic is where the membership is no longer a mutually ‘known community’ with regular social interaction. Other factors are diversification of funding, or different types of library served. As size or complexity grows in this way, so likely does the focus on representative and decision-making structures in governance, and on the role, accountability and authority of the executive director. Typically, an executive director will be accountable to the membership for decisions through a representative Board. The balance between agent and provider is an important contributor to the sense of ownership and belonging that members will have towards a consortium, as is the participation in governance mechanisms.
  • Collective action and rightscaling. Consortia are effectively a response to a collective action question (is it more efficient to solve some questions in a group, or to turn to a group to acquire capabilities you cannot generate internally?), which in turn raises a right-scaling question (at what level is action desirable, or what is the optimum size of a group?). For many issues, a regional scale is important, where personal networks within that known community are an important part of the value a consortium delivers. For others, especially in a network environment, it makes sense to scale to much broader groups. For example, OCLC has built platforms on which tens of thousands of libraries collaborate at scale around cataloging, resource sharing and discoverability. Jisc provides infrastructure solutions which scale to the national level. Scale also becomes an important part of scoping decisions. There is one network-level HathiTrust, for example, which makes more sense than multiple regional ones. This again underlines the diversity of consortial activity, from locally-constituted groups to large providers of community infrastructure.

In each of these cases, there is a spectrum of choices or evolving stages. One of the more interesting things about this space is how groups can evolve: consortia may change over time as circumstances and environment change.

So, for example think of a group that is set up to serve a single purpose – over time, this purpose may shift, or other activities may be added. In some cases, a consortium can evolve into a general collaborative apparatus that members reflexively turn to when new or shared needs arise. In this way they may reduce the cost of building a new service, partnership or network from scratch. However, it also means that the consortium needs to be clear about what the appropriate response is – developing a service, contracting for a service, ruling it out of scope, and so on. Consortia may also want to attract new members, and evolve services in that context. The evolution of Lyrasis in recent years is an interesting example of diversification.

Similarly, the transition from agent to provider may be prompted by various steps. For example a university hosted consortium is a common model. However, a decision may be made to independently incorporate to get more flexibility – to avoid university procurement or HR processes for example. If a group develops shared infrastructure, there may be a decision to diversify revenue or to grow membership to support greater capital investment for R&D, platform development, and so on.

Finally, it may make sense to carry out some activity at a broader level. The ‘O’ in OCLC originally stood for Ohio: it was an organization set up to provide shared services to Ohio colleges. It has grown to serve thousands of libraries around the world, as the services it provides benefit from network effects. HathiTrust was originally set up at a BTAA level. It quickly grew to operate at a national scale, and is also looking at an international membership model. In other cases, an organization may deliberately restrict growth; trusted interaction with a known local ‘known community’ may be an important attribute which the group wishes to maintain.

These three cases are examples respectively of what I later discuss under scoping (what is the appropriate range of activity?), sourcing (how should the activity be procured?) and rightscaling (at what level should the activity be carried out?).


For convenience in these posts, I use the terms ‘consortia’ or ‘collaborative groups’ to refer to organizations which support some form of general collective activity, and which are supported by libraries, acknowledging that this may not always be the term used in a specific context.

Given the scale of the US – in terms of both the size of the sector and geographic extent – and the comparative absence of the overarching national public provision which is present in many other countries, it offers many examples of collaborative activity at different levels. While I emphasize the US in my examples, I also include some developments from other countries. In general, where I provide examples, I do not comment on how effective or efficient a particular approach is.

Acknowledgements. I am grateful to the following people for generous comments on earlier drafts: Gwen Evans, Mike Furlough, Kevin Guthrie, Susan Haigh, Brian Lavoie, Kirsten Leonard, Rick Lugg, John MacColl, Constance Malpas, Barbara Preece, Judy Ruttenberg, Katherine Skinner, John Wilkin, Nicola Wright. Of course, I alone am responsible for the use to which I have put their advice and they do not necessarily share any of the views expressed here! I presented some of this material at the JULAC 50th Anniversary Conference in Hong Kong, and I benefited from the views outlined in the formal responses of Peter Sidorko and Louise Jones.