Key takeaways from 19 research and infrastructure institutions: Part 1 – Governance Structure

Over the past few months, we have taken a closer look at 19 research and infrastructure institutions from Germany, Europe and North America: their history, governance models, funding sources, strengths, potential blind spots and opportunities. These institutions vary widely: some have existed for decades, whilst others are still finding thier footing; some enjoy secure, long-term funding, whilst others move from one project phase to the next. In this six-part series, we examine the findings from this analysis most relevant to the establishment of a German Research Software Institution.

When designing a new institute, network or consortium, questions of governance and decision-making arise early on. Should there be a central board that concentrates power in the hands of a few? Or should decision-making be delegated as far as possible to member institutions, trusting that local initiatives will collectively add up to a coherent whole?

A compelling pattern emerges from the 19 institutions studied: those that have reliably fulfilled their mission over decades combine clearly defined central coordination with decentralised implementation within a federation that is sustainable for all involved. The central body is often not an executive body, but a hub where everything comes together, where common standards are developed and maintained, and where the entire community gains visibility. Decentralised implementation, on the other hand, remains with the partner institutions, namely where the knowledge, relationships and domain expertise actually reside.

An illustrative example is the Alliance for National High Performance Computing (NHR Alliance) in Germany. Here, a conscious decision was made to use a registered association as the legal framework to finance the investment in and operation of the participating university high-performance computing centres, amounting to over 60 million euros per year over a period of ten years. The association ensures implementation and focuses on tasks that no single centre could manage on its own. These include, amongst other things, a joint strategy, standardised interfaces, representation vis-à-vis the federal and state authorities, and the coordination of science-led computing time allocation.

A second example comes from the Network of University Medicine (NUM) in Germany. During the pandemic, this network demonstrated that subsidiarity can work as an operational principle. Decisions are taken by the committees responsible for dealing with the issue at hand. The central coordination office takes on tasks that only it can perform, such as organising financial flows, procedures, coordination processes and decision-making, as well as ensuring cross-organisational interoperability.

In some of the institutions examined, however, it becomes clear what happens when governance is too complex or unclear. Decision-making takes time when duplicate structures arise or their interfaces are not clearly defined. The same applies when the central budget is too limited to enable meaningful overarching coordination.

Sustainable governance for a German research software institution therefore requires a strong, highly competent central body that does not provide all services itself, but instead acts as a coordinating body, point of contact and convenor. In addition, strong partnerships are essential, contributing their respective operational strengths within a federation or network through clearly defined engagement tiers. Finally, supervisory bodies that do not interfere in day-to-day operations, but uphold the values of the federation and set the strategic direction, are equally necessary. In this context, there is an underestimated advantage to building on an existing institution as a legal basis: it avoids the start-up costs and conflicts that establishing a new legal entity might entail.

Shared responsibility is a sound architectural principle that consistently emerges as a key element in the most successful institutions in our analysis. An effective and efficient organisational structure rests on a carefully calibrated distribution of power, centralising what must be centralised, and deligating what decentralised units are best places to handle.