By Pete Barbrook-Johnson, co-author of Systems Mapping: How to build and use causal models of systems (Palgrave, 2022)
9th September 2022 (originally published by Oxford Smith School, reproduced with permission)
In the UK government, policy design and implementation is meant to follow something called the ‘ROAMEF’ cycle, which stands for Rationale, Objectives, Appraisal, Monitoring, Evaluation, Feedback. This imperfect framework is often the subject of ridicule among civil servants, but it serves a purpose by making us think about the different stages of the policy life cycle.
But it’s missing something vital: systems mapping. There are many types of systems mapping, but here I am referring to those that use diagrams to visually represent the causal structure or influences of a system. Policy system maps are simply versions of these diagrams which include government policy, placing it in the real systems it is trying to manage and influence.
I have been working with civil servants, analysts, and policy makers in the UK and abroad, to make use of different types of scientific modelling, for ten years now. The value of well-executed systems mapping exercises to the policy making process has been undeniable. Various public sector organisations are now beginning to advocate for it, for example the Government Office for Science. Systems mapping structures thinking, helps to build consensus around a shared problem and definitions, enriches understanding, and can play a valuable role in the design and analysis of almost any policy.
This includes climate and net zero. The policies needed to support the transformational change required to meet our climate targets, cut across different sectors, industry and the public, and connect into domains such as human health, biodiversity, economic growth, and social cohesion. These cross-cutting and deeply entwined policies and systems are crying out for systems mapping. Yet, at best systems mapping is haphazardly used in government.
Why systems mapping?
If we accept that government policy always operates in complex social settings, and we should, then any effort to make sense of that context before we intervene only stands to enhance the efficacy of the policies we design. We need to understand these complex systems before and whilst we act. We need to dive into their nuances and interdependences, to understand them as ecosystems of actors, power relations, ecological, physical and technical systems. If we are to do this, we need to work together, sharing mental models and assumptions and drawing a diverse range of expertise and opinion. Systems mapping allows us to do this.
In our work, we’ve been amazed at the response to our own particular type of systems mapping (‘Participatory Systems Mapping’). Civil servants have told us how it has transformed their approach to analysing large areas of policy, connecting teams, and developing novel insights. There are plenty of other recent examples, such as the system map for electric vehicles which found its way into the UK’s net zero strategy.
Yet, for every policy area that benefits from a comprehensive mapping effort, a far greater number go without. For every ‘converted’ policy analyst or policy lead keen to do more, there are ten who are not sure where to start, are uncertain about how systems mapping might help them, don’t know how to create space for a novel method in the face of pressing deadlines, or are nervous to push their seniors to do better.
To ensure this vital tool is not overlooked or side-lined, systems mapping should be made a core element of the government’s policy analysis processes. Giving system mapping this level of recognition would put it on a par with other key functions, such as evidence and literature reviews, policy appraisal (i.e. cost benefit analysis), and with policy evaluation (i.e. ex-post assessments of ‘did the policy work’?). This pride of place would pull a heightened recognition of complexity and systems thinking through the entire policy development process. In effect, allowing the method to drive a broader reconceptualisation of the policy-making process.
A core function: when, who and how?
So, if it were to be a core function, where would it typically operate? Since systems mapping helps so much with basic understanding and consensus building, it seems obvious it should be done at the beginning of the policy cycle. But we know from our experience how useful it can be during ex-post policy evaluation too. And it’s not hard to imagine how it might be helpful to frame an assessment of possible impacts of a policy, its costs and its benefits, before it’s rolled out. So, the answer seems to be, it would be helpful at all stages – it should follow the policy through from early development and rationale building, through setting objectives, appraising, monitoring and evaluating. It should be the first thing you see after you’ve read the policy name, summary and objectives.
The fact that systems mapping could and should be everywhere in the policy cycle, is why it needs to be a core function of government analysis. If only evaluators own it, then it will get stuck at the end of the cycle; if only economists own it, it will get stuck in appraisal. It needs to be jointly owned by evaluators, by social researchers, by economists, by geographers, and by operational researchers.
The desire to ensure consistent, coherent, and clear guidance on the various processes that make up the ROAMEF cycle may have had an unintentional adverse impact in encouraging siloed thinking amongst the various government professions, leaving them unfamiliar and unsure about one another’s methods. These groups sometimes work in silos and are sceptical of each other’s methods and worldviews.
A dedicated systems mapping function could be the glue that holds together the various processes that inform the policy design and implementation process – as well as the professions that typically ‘own’ them, closer together.
It’s time to map
So how should policymakers start? They can begin with what they’ve got. Analysts sometime use logic models and Theory of Change maps – they can keep those while ramping up their sophistication. There is no ‘right’ way to system map, the one essential element is a desire to try to build rich understandings of the system you are attempting to change, with as wide a group of people as you have resources to speak to. Myself and co-author Alex Penn have seen a variety of systems mapping techniques used in government. These should be encouraged and shared, and junior colleagues should have the confidence to lobby their seniors to embrace the technique.
With COP27 looming and politicians caught between their net zero pledges and the cost of living crisis, there has never been a better time to map the systems we need to change. Our hope is that the practice will be adopted as a core part of the UK’s policy design process. The decisions we take will be all the better for it.