Grassroots Trials

The co-design process should result in enough of a shared vision and understanding amongst the core project team for them to start working steadily together. Although careful analysis should mean that the project is strongly grounded in theoretical and local realities, it is essential to establish the richest possible set of feedback loops as it progresses. This comes from thoughtful engagement with a variety of stakeholders, and allows ongoing improvement and reduction of uncertainty.

Being able to do this is an essential characteristic of the kinds of local experts that Mutual Credit Services looks to work with. Typical stakeholders may include business networks, anchor institutions (local government, universities, hospitals etc.), accountants, and of course the intended users. Each should be met on their own terms, addressed in their own language, and presented with facets of the project that speak to their priorities. Whilst MCS has experience in developing communication strategies that describe what we can offer in terms of ‘adjacent possibles’, it is essential that the local team develop their own way of talking about the work they are doing and the intended outcomes.

If engaged with effectively, these stakeholders offer different forms of input and guidance. Anchor institutions may offer policy insight, extensive data sets, and credibility. Business networks may offer endorsements, promotion, and of course access to potential users. All of these can ensure the project’s strategy is aligned with local priorities (of which the local part of the co-design team, no matter how well-embedded, will inevitably only have a partial view of).

Because none of MCS’ collaborative finance tools are particularly complicated, the digital technology that supports their usage can (and should) be kept simple. As well as reducing costs, this also makes it possible to start engaging with potential users well before launching a commercial service. In the Local Loop Lancaster & Morecambe project, the demo app was presented as a survey: users simply entered estimates of their typical spending with local suppliers. Multilateral obligation set-off could be performed on this trial data, finding the trading loops. This had several desirable outcomes:

Users could be onboarded in a low-pressure manner with no ongoing commitment;
We could provide individualised estimates of who would benefit from participation in the full service;
Support operations could be developed and practiced;
User interactions, feedback, and requests informed further app design and functionality;
We could monitor the trading network as it developed and loops started to emerge.

This learning has allowed thorough de-risking of the follow-on Local Loop Merseyside project, with trial outputs – especially the network visualisations – serving as compelling communication materials. We also gained experience with using data to guide onboarding efforts in a manner consistent with the research carried out by network scientist and sociologist Damon Centola around how ‘complex’ behavioural contagions spread.

A common intuition about how behaviour spreads is based on the notion of ‘virality’. In this conception, what matters are the ‘weak ties’ that connect huge numbers of people – who themselves typically have meaningful relationships with only Dunbar-sized groups – into ‘small-world’ networks. This is embodied by the idea of ‘six degrees of separation’, which has become the received wisdom for how everything spreads. For spreading actual viruses, or simple information (memes, job openings) to remote parts of a network, leveraging weak ties can indeed be an optimal strategy, since transmission can be effectively instantaneous via simple mechanisms (infection, social media shares/likes), and is often involuntary.

Given the prevalence of this intuitive model, it is easy to assume that when attempting to diffuse a new behaviour or technological innovation that the focus should be on weak ties, as opposed to what appear to be the redundant, dense, overlapping connections that are characteristic of e.g. friendship groups, or indeed the dense trading networks within which MCS’ collaborative finance tools (especially mutual credit) work best.

However, there is plenty of empirical evidence that desirable innovations often fail to diffuse as hoped, and Centola demonstrates (through computational models, experiments, and case studies) that for behaviours that meet some minimal threshold for complexity – which can be as low as providing contact details to a sign-up form – the viral model of diffusion is misleading, and strategies based on it can be actively counterproductive.

For these ‘complex’ behaviours, it is precisely the mutually-reinforcing ‘strong ties’ found in clustered social groups that lead to adoption and maintenance. The contrast is encapsulated by the spread of a virus versus the spread of attitudes to vaccination.

Social change is about norms rather than information. Social networks should be viewed as prisms through which people understand ideas, beliefs, and behaviours, not as mere pipes for transmitting them. This network bias can reinforce people’s existing biases, but can also be used to trigger enthusiasm for an innovation or social movement.

This network approach to behaviour change therefore focuses on collective rather than individual factors. Although tailoring the messages people are exposed to can be effective, working with social norms and structures can be far more so. In the same way that schooling behaviour in fish could never be predicted from studying one in isolation, many behavioural diffusion phenomena only make sense when complex interdependent social relations are understood.

Specific applications examined by Centola include how a novel technology can be incubated in populations in which an alternative is already established, the use of wide bridges (exemplifying a distinction between the structural and relational importance of ties) for encouraging diffusion across organisational boundaries, how to design networks to increase diffusion amongst strangers, and how to control the kinds of behaviour that spread using incentives for social support or comparison.

MCS makes careful use of these insights when building networks, collaborating with partners, and developing tech. A successful trial brings a project closer to serving as a model to be adopted and adapted by others; for example, it will be possible to replicate Local Loop Merseyside as a generic Clearing Club, and connect to other clubs via the Credit Commons Protocol to achieve network effects. MCS is developing itself into a social franchise to enable such ‘replication and federation’ with maximum efficiency, quality of service, and local autonomy, whilst ensuring that learning and best practices are shared back into the core team.