On replicating and scaling

There is an unhelpful tendency in the social/community sector to focus on a narrow number of highly marketable community projects and say – they’ve proven the concept, let’s replicate them! To view replication at scale as a matter of sharing compelling case studies so that others can follow their example.

But it rarely works like that, at least not at any scale, and especially not in domains like property development and housing where projects take at least 5 years and must contend with a system hostile to any kind of new entrant.

(When we talk about scale, we rarely define what we mean. Do we want 100 similar projects? One in every city, town and village? 5% of the market share? The unspoken ambiguity lets us off the hook, permits us to celebrate small-scale success with challenge.)

Often, these success stories have been enabled by specific local factors – the secret sauce of a talented entrepreneur who is good at gaining profile; a trail of grants from funders that have limited budgets; an unusually supportive council, landowner or anchor organisation that was critical to its success. You can’t replicate these factors at a large scale.

Sometimes, our desire to praise the success hides the shortcomings – the lengthy and the expensive, that arose from missteps, from governance failures, from trying to do something very new. We may lack the skill and experience to even see and understand these shortcomings.

So wannabe copycats are forced to reinvent the wheel. To re-imagine how the impact might be achieved by some other means applicable to their context. Replication comes to mean: find your own trail of grants, find new supportive partners, make old and new mistakes, wade through the swamp of despair that lies behind many good ideas and their achievement. This is hard work, takes a long time, and requires a lot of those involved.

You might say, this is in the nature of innovation. But innovation and scaling aren’t the same thing, and it’s impossible to take a concept to scale if every instance has to undertake deep innovation.

One consequence of this approach to scaling is that places and people the stand to benefit the most from community-led development are the most left behind by it. Local Trust has patiently put together the evidence that these approaches let down the areas that most need it. Areas they identify as ‘left behind’ are not only deprived, they also lack the pre-existing social capital – the people with the skills and drive, the places for them to meet, the relationships with key stakeholders. Strikingly, during the COVID-19 pandemic these left behind places formed half as many mutual aid groups, and received 60% less charitable grants, than similarly deprived areas with better social infrastructure. Yes, community land trusts and community anchor organisations are disproportionately present in more deprived parts of the country. But those that exist are quite small in scale relative to the local challenges, they are still missing from most of these places, and it is a tall order to say to a disinvested community ‘copy this exemplar’ without the kind of patient, decade-long funding and support provided by a programme like Big Local. Funding that really needs to be made widely available to build community-level capacity.

The Community Land Trust Network is currently co-funding a piece of research looking at the relative lack of black-led Community Led Housing organisations. I’m eagerly awaiting the research findings, but my hunch is that there is a similar problem: asking so much of minoritised communities that are already undercapitalised and often time poor, and that already face structural discrimination in areas like charitable funding, local government support and the housing market.

The more general consequence of this copycat scaling strategy is that the concept grows slowly, in niches where the soil has just enough fertility for the concept to take root, and a lack of competition from the private and public sectors to crowd them out.

We need to contend with the reality that, in a capitalist market economy, social enterprise risks being terminally uncompetitive.

For example, communities wishing to build homes will find they cannot compete for land allocated through the planning system for three reasons: in aiming for high standards they will be unwilling to pay as much as a private developer for the land; generally being small and under-capitalised their finance costs are much higher than for an established developer or housing association; and they lack risk capital with which to move swiftly and make bids, or buy options. By the time a community housing group has been able to organise itself, that land has been sold and for a price they could never have afforded.

If our only response is government support then our scale is limited by the extent to which the public sector is able and willing to provide it – with all the uncertainty introduced by cycles of elections, budgets and internal changes in leadership. In recent years we have achieved a lot here, but it has still been choppy, funding switched off more often than on, falling far short of the kind of industrial strategy we need to diversify the housebuilding industry and give communities real agency and power. Successive secretaries of state have wanted to support us, and we have gained recognition across a wide range of policy initiatives. But clearing the path requires a systemic and cross-departmental approach, which governments are rarely capable of.

So community groups are left looking for niches – such as small, hard-to-develop sites; empty buildings with an owner looking for an exit route; or failing assets like ailing community centres and libraries that the council is willing to offload onto a community organisation. They wait for the Community Housing Fund to be open to bids, or cast around for social investors and crowdfunding to reach the point where they can get a bank loan.

Many wonderful projects have been developed in these niches. But if we truly believe in the power of community to build wealth, social cohesion and climate resilience, these niches scratch the surface of its potential. It’s a small scale, and a slow and tortuous way to built out projects.

So… if ‘copy my project’ is not a good strategy to achieve scale, what is?

I don’t pretend to have the answers. But I’m going to tell you about some tactics that have emerged from our research and practice in the community land trust movement, and in looking at adjacent sectors and countries.

Seeing the whole system

Legacoop is Italy’s equivalent of Co-ops UK, representing over 15,000 co-operative enterprises. In 1996, Piero Ammirato produced a detailed study of how this successful network grew and was able to sustain itself within a capitalist economy.

(Big thanks to my friend Ed Jones for first pointing me to this paper years ago, and digging it out again for me recently.)

Ammirato emphasises that, of course, one cannot take the recipe from Italy to England (for example) and hope it will taste as good. Our tomatoes are pretty bland, after all.

One of the specific factors in Italy was the strong anti-capitalist culture derived from the Catholic, Socialist and Liberal Socialist traditions. In his excellent study of the first wave of community led housing from 1879-1919, Andrew Bibby documents a similar intellectual and political seedbed of self-help mutualism in the late Victorian labour movements in England. This was quickly eclipsed by currents like Fabianism, and is a very thin soil among the mainstream left today. Despite the efforts of MPs in the New Social Covenant Unit, the idea of community-powered conservatism hasn’t captured the mainstream of Conservative thought either.

Ammirato compiled the Italian recipe for success from seven ingredients:

1. A legislative framework to enable and protect co-operatives;
2. Regional and national support structures within the co-operative sector;
3. A culture of collaboration and entrepreneurialism;
4. The creation of consortia and intermediaries to pool resources and achieve economies of scale;
5. Creating financial institutions that open the door to private and public capital, including risk capital (not just building a parallel economy of co-operative capital);
6. Balancing core social values with a focus on economic growth;
7. And developing political alliances and a Parliamentary lobby.

The focus was not on building up individual co-ops to struggle in a hostile environment. The network was not focused on peer learning and sharing great case studies.

No, they focused on changing the whole system, the flows of capital and political power. They created their own supportive ecosystem that enabled them to knit into and compete in the market economy. They took advantage of political and economic crises, when the private and public sectors seemed incapable of providing jobs and economic growth, to gain significant state support.

This is very much the territory of the Community Land Trust Network’s national work, but on a much larger scale. So, too, the work of enabler hubs we support at the regional level. In their three years studying the Homes in Community Hands programme, academics from the Centre for Regional, Economic and Social Research (CRESR) at Sheffield Hallam University found that enabler hubs were able to begin building supportive ecosystems which resulted in the city regions having 160 more homes in their pipeline at year 3 of the evaluation than they otherwise would have expected. I wrote about this research here. Maddeningly, several of these hubs have now closed after government and charitable funders withdrew early.

It’s a tall order to transform the system to the degree that the co-operative movement has managed in the north of Italy. We can’t really claim to have any of the seven ingredients in place, though we are working on all of them. It is even more ambitious to think about community power within the context of the wider existential social and ecological crises bearing down on us. But without seeing the whole system, and tackling the problem systemically, we don’t stand much chance of achieving scale.

From exemplars to patterns

One of the challenges in replicating at scale in a more conducive system remains – what do you replicate? If projects aren’t replicable because of unique or rare factors, how do you develop models or products that can be replicated at scale?

Last year the Community Land Trust Network and UK Cohousing Network ran a six month collaborative research project exploring how the community led housing sector can be more effective and scalable, facilitated by the design team deepr. We took working groups and then a fifty-strong group in a residential through the Design Council’s double diamond methodology – discovering, testing assumptions, and then trying to better define the problem and potential solutions.

The top-rated solution that popped out of this process was to focus more on replicable project types or products. That is, in contrast with always treating each community project as bespoke and starting from first principles.

But most participants struggled to describe any replicable models. More than this, I think people struggle to articulate what it means to be ‘replicable’. People there still fell back to saying – this project was fantastic, how can we replicate it? This project represents a blueprint for the future, let’s scale it up. But if that project depended on specific secret sauce, it isn’t replicable by definition. What might replicable be are some ingredients that lead to certain outcomes.

Christopher Alexander, Sara Ishikawa and Murray Silverstein developed the concept of a pattern language, publishing their seminal book ‘A Pattern Language’ in 1977. It broke the task of urban design and architecture down into 253 recurring problems, and then described a proven (or ‘perennial’) solution to each. Any layperson could use it to make a decent fist of designing a climbing plant, a child’s bedroom, a small family house, a green street, a neighbourhood, even the distribution of towns, by combining a mixture of patterns.

Patterns aren’t whole projects, nor are they just preformed parts that you stick together to make the whole – a bit of this CLT, a bit of that cohousing project, bob’s your uncle. They are a way of understanding the problems you face and to find a coherent solution that fits within the wider system – to see what the problems are in a given place, or for a given community, and to develop the project that best answers the challenge (by implication in a way that can actually be achieved). They are used in sequence, moving from the larger patterns to the smaller, from the structure to the features that embellish them, and then those which embellish the embellishments, and so on.

In short, no pattern is an isolated entity. Each pattern can exist in the world, only to the extent that is supported by other patterns: the larger patterns in which it is embedded, the patterns of the same size that surround it, and the smaller patterns which are embedded in it.

Christopher W. Alexander, A Pattern Language

The concept has since been widely adopted, including in computer science where it influenced the development of programming languages and systems like UNIX that underpin the internet. That has been a big influence on me – once a geeky teenager. Programmers don’t start from scratch to design software; they build on, and knit together, existing libraries and functions, and build new patterned frameworks. Each part of the system does one thing, and does it well, and they interact effectively.

I think we need to look at successful projects and break them down into a language of patterns (and along the way note, but discard, the specific solutions that are annuals rather than perennials, one-offs that few can copy, as well as those that on reflection were actually rather expensive, slow or ineffective). If it turns out that a feature of a project depended on a grant from a funder with shallow pockets, we conclude it is an annual and discard it, or look for a perennial version.

We can then think about which sequences of patterns are important in bringing about the outcomes we want.

We can look to design new solutions, new products, that are replicable because they draw on patterns, and we can understand the contexts they work in.

This doesn’t make them identical, cookie-cutter models. When applied to place they become bespoke.

They may be less impactful, less exciting, than the pioneers. But they are accessible for communities and can be deployed at scale, and so have a far greater impact in sum.

Working with a pattern book and replicable products also makes it easier to develop consortia and intermediaries. Common problems with common solutions can be solved at a higher level. If you design a solution to a problem like “how can flat shares own and manage their own home?” that works across Greater Manchester, and if you can sequence that solution with others that create a replicable product, you can then get on and work at scale. This would be in contrast to starting out with a group of frustrated renters that want to start a co-op and trying to help them build or buy a home.

The one example of replicability that participants in this research project could cite is the Middlemarch model of partnership between a Community Land Trust and a Housing Association. This model accounts for 55% of homes completed by CLTs so far.

The Middlemarch back-story is more interesting than the product. Steve and Alison looked at what people were trying to achieve with CLTs in Dorset and Somerset, and why they were failing. They didn’t just try to help communities copy the pioneers like High Bickington and Broadhempston. They developed a new product that would solve problems of access to skills and development finance in a replicable way by partnering the CLT with a housing association; doing so in a way that would still achieve most of the communities’ top priorities, albeit at the expense of compromising on a few. They developed an enabling service that held the community’s hand through the process, and won the confidence of housing associations and councils. They now have almost 100 projects on their books,  an almost-100% success rate, and are working with the councils to develop a more supportive local policy environment. Together with the national network, we are also able to mobilise MPs from the region to advocate for changes to national policy.

It would be interesting to try and describe this ‘model’ in the form of a sequence of patterns.

Bibby’s book on co-op housing tells another story of top-down led growth that might resemble a pattern book. While there were some co-ops set up by tenants, organically and bottom-up, they seemed to be the exception. Rather, a network of well-connected co-operative entrepreneurs set out to build different forms of co-operative housing; they raised a lot of the finance from the great and the good; they set up a national enabling body, which in many cases directly acquired land and commissioned the building; they acted at scale, building out well-known garden communities in places like 136-home Burnage Garden Village in the time it has taken many of today’s CLTs to build 6 homes. Today, the closest equivalent would be the way that GreenSquare Accord develops co-operative homes in Redditch and supports the tenants to take control of them.

With UKCN, we are now piloting a growth programme to help the sector to – if I can put it this way – replicate these stories. To discover, define and codify new pattern-based products and service models, and develop business growth strategies to take them to scale. Owen Jarvis and I will be sharing more on this very soon.

When bees meet trees

The housing association partnership has been key to the success of the Middlemarch model. Partnerships with private developers and landowners may also prove central to other replicable models.

There’s a joy and an integrity in pioneers who take the DIY route – who act as developer with traditional build contracts; who become regulated social housing landlords; who directly cultivate their land.

But partnering with other larger organisations can take a lot of the heavy lifting off the shoulders of community organisations. This frees them up to focus on the things they’re good at.

Why make every community group become experts in development viability and finance when what we value is their expertise about their local place, the community’s needs, the ability to steward projects and assets to maximise community benefit? Some will want to become experts in all of it from day one. Others may, in time, develop these other areas of knowledge and skill. But many will be quite content to partner with another organisation that can save them the trouble. They will be happy to work with an experienced enabler that can ‘drive from behind’ – moving the project forward, bringing key decisions to the table and guiding the group in arriving at decisions.

This is the UNIX philosophy again – each part of the whole doing one thing, and doing it well. Not trying to do everything, poorly. Patterns can describe these forms of partnership, as well as the form of project they deliver.

We may find that partnership patterns are fundamental to some sequences – the key to their success. Or, at least, a factor that will significantly increase the chance of success, and the scalability of the sequence.

We have found that so long as values and business objectives are aligned, and the partnership is facilitated by a skilled enabler, partnerships can help communities quickly achieve a scale they’d struggle with alone.

Owen Jarvis and Ruth Marvel wrote a report while on the Clore Social Leadership Programme capturing this theme, called When Bees Meet Trees. Large existing organisations – trees – bring considerable capacity and expertise, but feel they have to do everything themselves – including social innovation. This doesn’t play to their strengths. Meanwhile, smaller entrepreneurial organisations bring nimble, fast-moving cultures, but lack size and impact.

If the bees meet the trees – if trees see their role as partnering with and investing in the bees – then you can get the best of both worlds. Trees can not only help individual projects, they can help a sector achieve scale more quickly. You can then start to map out an ecosystem that can achieve collective impact. Where we lack the trees, our focus might be on building up consortia or regional intermediaries to be able to act at scale and partner with individual community projects.

I’d also like to turn this concept on its head. Rather than communities finding a developer to partner with, what if we saw developers seeking a community partner? What if it could become commonplace for landowners, developers, housing associations and planners to specify that a community organisation can, and should, be involved? One could inject the CLT model quite easily into many housing and non-residential developments – the community as a landowner, co-creator and long-term steward. This is starting to happen.

A scaling strategy might focus on services to the private sector, on demonstrating the business case for partnerships with community, and on supporting communities to then engage constructively and exercise genuine citizen power.

Concluding thoughts

In trying to see the whole system; in trying to identify and develop a pattern book for replicable products, services and partnerships; we hope to achieve the Community Land Trust Network’s mission to make community ownership of land and affordable housing commonplace.

It’s our best stab at the strategies and tactics to bring about our vision – that every community in England and Wales can play an active role in making itself more socially, economically and environmentally sustainable through community ownership.

We are trying to learn from other sectors, and from research that we actively participate in, advocate for, and sometimes fund. What I take from that work takes me back to defining scale. If our ambition is to gradually grow – for example, from 79 CLT homes in 2008 to 1,711 in 2023, and then to 3,000 in a decade’s time – then a strategy based around peer learning, small grants and innovation may suffice. But if our ambition is to reach the potential we’ve estimated of 278,000 community owned homes, we clearly need something different.