‘I told you so’ is not a nice phrase – it implies superiority and a lack of empathy with someone who may have made a mistake and should have known better. But I’m sorry it’s coming to coming to point where no other phrase will do. We have known since the early days of MATs that forming ‘clusters’ or hubs’ is important to creating sustainable multi-academy trusts (MATs) as they expand. I first wrote about this back in 2010 in a think piece for the National College, called ‘Chain reactions‘. In a report two years later my fellow researchers and I reinforced the significance of geographical proximity and the formation of clusters as a means to manage scale as MATs grew.
But parts of the sector and some decisions by RSCs and headteacher boards seem determined to ignore the advice.
Defining a cluster
Let’s first define what we mean by a cluster or hub (I prefer the term ‘cluster’ but for the purposes of this blog the terms can be interpreted as inter-changeable). I understand a cluster as being a group of schools located in a tight geographical area working closely together and led by overarching executive leader. The number of schools in the cluster may vary from two to around five or, perhaps, six depending on their size, their location in relation to each other and the scale of the challenges they face.
A ‘tight geographical area’ means schools within easy driving distance of each other – ideally no more than 20 minutes apart, though a slightly longer commute time can work. A group of schools does NOT constitute a cluster just because they are run by the same MAT and happen to be in the same region or even sub-region – not least because quite often the so-called cluster also includes academies that straddle different phases of education. Of course, there can be excellent learning and support across phases (for example, in developing an all-though curriculum, handling transition, managing behaviour and organising special needs) but the deepest impact is more likely to come from academies that serve the same age range collaborating together – and, as I shall explain and illustrate below, geographical proximity is, integral to this.
The lack of clarity on what is meant by a cluster bedevils our understanding of what is happening in the MAT landscape. On the face of it clusters are becoming an accepted feature of MAT growth, as the table below illustrates. But lumping together regional and hub management structures is not helpful for the purposes of knowing what is really happening in MATs. It conflates two rather different models for managing scale. Smaller MATs will not require a regional tier but may well use clusters. The largest MATs may require oversight at a regional level but this does not necessarily preclude the operation of clusters – i.e. they may well have clusters as well as regions.
Proportion of MATs, by different sizes, having a regional or hub management structure to support accountability
Source: DfE Academy trust survey 2017
The rationale for clusters
So why is a proper cluster strategy so important and why does geography matter so much? I would suggest three main reasons:
School improvement and professional development
Clusters provide MATs with the means to develop a new and powerful approach to school improvement and professional development. Going back to the early days of National Support Schools the concept of a strong school supporting a weaker one relied on a close geographical relationship between the two schools. That learning was borne out by the experience of charter school networks in the USA – as my colleagues and I wrote in our 2012 report on academy chains:
“Chartered management organisations (CMOs) in the US generally confine themselves to a particular geographic area (defined as a reasonable driving distance). They like to develop tight networks of schools that can be easily supported by CMO staff. Operating outside this network means fewer school visits, a greater challenge to organise professional development and share learning across schools and a potentially weaker bond between the CMO and schools.”
The model can be adapted to fit the context of stronger schools supporting weaker ones or stable schools working as group to improve pupil performance and progress. But proximity is central in both cases.
Leaders and staff with expertise can easily be deployed from one school to another for part or all of the week. Staff can share not just insets and twilight learning sessions they can crucially also work together to transfer that learning effectively into the classroom. Staff within a tight cluster can develop a common curriculum, plan lessons together, share expertise, observe each other’s practice, moderate each other’s other’s work, apply an agreed coaching model and run inquiry-led learning projects. A cluster can provide the stimulus and the means to tackle within-school variation by drawing on the challenge and support of a nearby school(s).
If clusters are to achieve their full school improvement potential MATs need to seize the moment and redesign procedures and timetables so that their staff have more time for this sort of intense collaborative professional development activity. Clusters also provide a perfect platform for giving staff in their early years of teaching experience of working in different contexts, by enabling or deploying them to work in different schools in a cluster.
Leadership as a pooled resource
A cluster provides the basis for rethinking the leadership model for schools. Every school needs its own leadership team but given the challenge of recruiting sufficient high calibre headteachers it might equally well be led by a head of school rather than a headteacher or principal. We know that one of the big blocks to applying for headship is the scale of the step-up from deputy to full headship responsibility. The post of head of school, supported by an experienced executive head for a cluster, provides a great way to grow people into senior school leadership positions.
However, viewing leadership as a pooled resource within clusters goes further than this. Instead of each school having its own array of subject leaders, specialists and SENCOs, a cluster system enables schools to develop and retain highly skilled practitioners who can exercise responsibility across a group of schools. It means that MAT possesses in each cluster a breadth of expertise that can be shared with colleagues and used to support their development and accelerate school improvement. Of course, some things will be better organised at a school level and some at a MAT level but, as the graphic below illustrates, with clear thinking and definition of roles MATs can combine the intimacy of intensive working between schools with the benefits of being able to access broader programmes.
Agreeing respective roles and responsibilities across a MAT
The model not only makes sense in its own terms but it can also be key element of a MAT’s talent management strategy and contribute towards its leadership pipeline. It can also yield significant cost savings. Research studies and illustrative leadership frameworks produced by the National Schools Commissioner, Sir David Carter, have demonstrated how a cluster leadership model can save tens of thousands of pounds or, in the case of secondary school clusters, savings running into six figures. With school budgets are under such intense pressure this provides a further incentive to take cluster leadership seriously.
In my previous blog below I summarised the findings of a report from the Education Policy Institute which found that geography was a significant factor in MATs realising economies of scale. MATs that are more geographically dispersed tend to spend more per pupil on back office costs. In contrast, “standardising various practices (such as staffing contracts), or tendering certain services (such as cleaning) across schools, noticeably reduced marginal costs once clusters of schools reached a ‘critical mass’ of around 3-6 academies”.
The cluster might also provide the basis for delivering IT support, HR, financial administration and estates management – depending on the extent to which services are being run locally or centrally.
So there you have it: the case for clusters. To my mind it’s strong – not to say overwhelming. But it comes with two riders.
Governance of clusters
Establishing clusters provides an opportunity to rethink the governance model of MATs. As the chart below shows different MATs are taking different approaches. I suspect that most MATs are continuing with the ‘normal’ approach of the MAT board exercising formal oversight and accountability with individual local governing bodies (LGBs) or academy councils having a range of delegated responsibilities. In this model there may (or may not) be informal sessions where LGB chairs and/or other governors across a cluster meet together with the CEO and chair of the board to liaise with the MAT or hold joint professional development sessions. Some MATs have started to formalise a cluster governance role – but only on an advisory basis. However, a few MATs have gone further and delegate school oversight to a governance body constituted on a cluster basis. This is another area where we need more information and research about which governance model is likely to suit or work best in what circumstances.
Emerging practice on MAT governance at cluster level
Planting new clusters
A commitment to clusters does not preclude MATs starting or forming a cluster in a locality where they are not currently operating. We do have a number of examples of both CMOs in the USA and MATs in this country successfully moving beyond their initial geographical base and establishing groups of schools in new areas. We don’t have the research to validate the basis of how to do this effectively but from my work with MATs I would suggest the following 10-point checklist:
- Is the performance of the MAT strong and sustainable enough to justify the resources and effort required for expanding into a new area?
- Has the MAT sufficient leadership capacity and expertise to lead the outreach and creation of a new cluster?
- Is the MAT satisfied that it can create a new cluster and not just acquire one school in an isolated location (even if that is the initial starting point)?
- How will the model for delivering school improvement need to be adapted to be appropriate for the new context?
- Has the MAT identified local sources of school improvement support that a new cluster could tap into?
- How will the new cluster relate to and work with the rest of the MAT?
- What are the financial implications of establishing a new cluster – what does it mean for the MAT’s business model?
- What are the implications of a new cluster for the delivery of back office support functions?
- Has the MAT considered the governance model for the new cluster – and identified the right personnel to undertake key roles?
- Has the MAT talked through and gained ownership of its strategy with leaders and governors in its existing academies and clusters?