The practice of Institutional Architecture

| By | Articles

It’s very easy to make sweeping generalizations about institutions, but these are very hard to connect back to specific, tangible – perhaps actionable – guidance on designing or retrofitting effective, resilient, institutions for particular contexts and problems. 

What we’re trying to do at TIAL is to fill in the middle space between the abstract and the particular. We are seeking to learn from accumulation of specific experiences and lessons from institutional design to understand more general, perhaps widely applicable, principles for institutional architecture and the skills behind them. 

This is a long-run project. It will take the collective wisdom of a field, and a substantial investment in creating the structures for accumulating and articulating knowledge, to build these insights. 

What we’re going to do today is go through a few examples from each of our past work on institutional architecture. We’ll see if we can draw out – or if I can draw out between the three of us – some more general avenues to build up and accumulate knowledge. We’re going to start with examples that represent kind of run of the mill, everyday, ordinary, typical institutional architecture opportunities and cases, and then move into what each of us have seen as outliers. 

I. THE SPACES FOR INSTITUTIONAL DESIGN

Geoff: I think the first thing to say is that we are surrounded by lots of examples of past institutional design, whether they are things like libraries or museums or schools or hospitals. All of these, at some point, were designed by someone to meet a need, something was in their head. They persuaded some other people to provide support or money or what have you. And precisely because there are so many millions of examples around us, as you said, Jessica, it’s pretty hard to generalize the thinking behind their design. Every story is slightly different. 

The University for Industry, which I was involved in creating quite a long time ago, combines some lessons about success and failure. 

This simple idea, and I think we’ll often get this with new institutions, was partly a need. Lots of people in workplaces needed new skills, and often quite modular little bits of new skills, maybe in digital tech, or how to do marketing, or how to do accounting, or what have you. A that time, there were also new opportunities coming from things like satellite TV. (This was slightly before the arrival of the World Wide Web.) That combination led to some interest in creating a new institution that would leverage these options to meet the need. 

It was put into a political party manifesto, then given a budget. It was then created essentially as a public agency, given the name Learn Direct, which sort of summed up the idea that you would learn directly. It had a mixture of physical centers and an online platform. This was the late 1990s.

At its peak it had about half a million learners at any one time and in some ways was working quite well. That’s the sort of success story. And as I say, it was a partly designed in advance with a notion of how its economics would work, the technologies it would use, the different things it would provide skills access to. 

But then about 10 years later, there was a change of government. Learn Direct hadn’t had sufficient legitimation, I think, amongst the elite. Most of its users were quite poor people, and the real influencers in the civil service or the media very rarely knew people who used it. They knew lots of people who used the top universities and the top schools. But in a sense, the institution hadn’t worked out a way of creating broad enough legitimation. So when the government changed, the new government slightly froze its budget and then privatized it to private equity, who then, for a whole series of reasons ruined it and almost destroyed it. 

It’s a kind of sad story of both success and failure. One of the errors was probably not to have involved the other main political party in the process. Employers and business were pretty engaged, but it was mainly small employers, not the big ones who often did their own training. If there hadn’t been a financial crisis and a drive to extreme austerity in 2009-11, it might well have survived actually a change of government, but it didn’t have quite enough resilience to the combination of an economic shock and a political shock. 

In an ideal world both of those might have been thought through right from the beginning. But you know who knows maybe that would have been impossible and you can never foresee every eventuality when designing an institution, but ideally any public one tries to get reasonably broad support from different political parties, especially if they’re likely to be in the government in the near future.

I think many of the principles that the early Learn Direct embodied are ones that we actually need for lots of new institutions which can help with the upskilling, upgrading of labor forces all over the world. We’ll need things rather like what Learn Direct tried to do though increasingly using obviously the web and also smart AI to help personalize curriculum. 

In brief: The combination of a visible need (ongoing needs for new skills) and new options for meeting it (satellite TV, in this case) led to the seeds of a proposal to leverage the new platform to deliver skills. The idea of Learn Direct moved from concept to reality by being included in a party platform, given a budget, and then created. It functioned for about a decade before losing support when there was a change of government and economic context. One lesson could have been to design in more resilience to such changes.  

Juha: I’ll just pick up where Geoff started in terms of lots of the institutions starting out as social innovations. It’s important to understand the roots of those innovations. If you look at, for instance, the Finnish social delivery and social service system, that’s exactly the dynamic that you’d see. Many of the current institutions started out as social innovations, from baby box to nurseries, et cetera, so then institutional innovation is more of a question of codification. 

A typical case would be that you have a need for embedding new institutional capacity into a specific context. The case I have in mind is the launch of new policy experimentation approach in Finland from 2015 to 2019. 

Innovations started before that. The previous government (2011-2015) had essentially asked for our support in terms of how to utilize some of the more innovative methods in policy design – experimentation, design thinking and behavioural insights – in order to essentially have a more systematic way of moving forward. It was a really interesting exercise because it essentially led us into a role of defining a structure within an existing organization (Prime Minister’s Office of FInland) – what that institutional capacity within the government should look like. 

This unfolded on many different fronts. On the one hand, we had to engage with the narrative. In that case, it was whether there was a recognition that one needs to embed more transformative capacities into the government machinery in order to push forward more transformative policies. That was the diagnosis – the analytical and almost rhetorical side of things – that we had to align with the political demand. The actual design work, on the other hand, is of course very technical because you need to engage with the structures that are there. The processes that are there that are often quite detailed. 

One of the very concrete lessons from that phase was actually from a bit of a mistake.  We took quite a bit of time to put really, really beautiful illustrations in place. We tried to visualize, we tried to document the process in really, really nice ways – that did not help at all. The response and the reason was that civil servants typically don’t look at nice visualizations. Of course, some might, but their more typical approach was to look at things as a table in which you see the steps in a really structured manner. If you’re over-designing and over-enforcing your own way of looking at things to a community of practitioners or experts that have a different way of looking at things, you very easily alienate. Finding the right way of communicating and the empathy to be able to speak the language of your collaborators is very important for institutional design that is about embedding new institutions into existing structures. 

The new experimentative policy design approach was embedded [into the government program] in 2015 and 27 policy experiments were then actually launched under a centre-right government during 2015-19. There was a support mechanism within the Prime Minister’s Office to help with how to actually implement those experiments. One of those 27 experiments was the experiment on universal basic income, which captured all the attention – in fact one of the most significant impacts of the technical work may have been in terms of actually accelerating the narrative around UBI. 

And on that, we tried to do our best to ensure that there is continuation and legitimization of the transformative institutional capacity over time. We spent a lot of time on building the infrastructure around the institutional innovation. For instance: we engaged with academia, stakeholders, everybody in order to ensure that if one does policy experiments, the individual experiments have legitimacy. Obviously you’re dealing with high tension situations in which some portions of population may gain or lose some of advantages and there are constitutional factors, etc. to take into account. 

Still, I don’t think we are where we wanted to be when we started that work 10 years ago. There was a lot of cultural continuation. The next government that came to power in 2019 (a left wing coalition), actually doubled down on pledging to continue experimentation on policies as one of the ways of pursuing imaginative policy measures. We have potentially gained ground in terms of political parties’ ability to explore policies and suggest experiments. This is good. However, we do not yet have the institutional capacity to really systematically deliver and support experiments. We worked embedding it, but there are limits. 

In brief: Institutional design often requires embedding new capacities into existing structures – and that requires attention to both the narrative and the technical aspects of detailed existing processes. It requires empathy and attention to communication. The results and ultimate impact can be surprising – in this case technical work to embed policy experimentation in policy design seems to have advanced the narrative around universal basic income (one of 27 experiments that came out of the new approach). It can also be partial: in this case, cultural shifts took root to create space to explore policy experiments across successive governments in Finland, but without creating the full capacity to support and deliver innovations beyond the delivered experiments. 

Jessica: It’s striking that in both of these examples, there was a kind of invitation for a design. A visible need with a possible new answer in your case, Geoff, and a desire to build a new state capacity in yours, Juha. 

The examples that I was going to bring start in a similar way with an invitation to address a publicly recognized problem. I was an advisor to a number of government-convened expert committees on infrastructure in India – one on public-private partnerships, one on transport development, one on urban infrastructure, one on setting up a strategic research network, etc. These commissions have a similar structure: there are objectives laid out in a terms of reference, a committee of prominent experts and representatives of key public agencies, and some kind of secretariat with technical advisors. This group produces a report with recommendations that sometimes get enacted, sometimes become a reference point for a longer policy conversation, and sometimes just disappear into the files. The intention in the start of the commission is usually to produce ideas that move into implementation, but they are often identified with one particular government and not picked up by the next. 

In my experience, a lot of the institutional architecture conversations start around precedents – how have other places solved a similar challenge: integrating transport across modes, financing urban infrastructure, reducing air pollution, etc. Occasionally I was able to make a case for considering a precedent for a certain capacity – say, interagency coordination – that might come from a different policy area, but often the examples of architectures for similar problems seemed more digestible. In transport policy, examples from transport; in city finance, examples from cities; etc. 

Even so – these precedents are useful not only for generating a sense of tangible possibilities, but also starting a conversation about what’s working and not in the current context – why are ports and rail not connecting up as they are being planned, for example, where’s the coordination being dropped? Then that opens up two tracks: possible ways to update what’s happening, and possible lessons from other sectors or places or times that might be adapted to the Indian constitutional, federal, economic, informational context. Each of those tracks ends up being a long conversation, over months with an exchange of ideas for comment (essentially stress-testing) by individuals who are usually leading key agencies that would have to be part of whatever new governance approach the committee recommends. 

The day -to -day really was a constant backing and forthing: here’s my specific problem, here’s a specific answer, that may not work, adjust back and forth. This approach is very, very communication intensive when it works well because there are often a lot of adjustments between a precedent or a design idea and what will actually work in practice in a particular context with lots of existing rules, culture, etc. This is the “technical part” that Juha mentions. There’s a boundary in working in this kind of committee space – inevitably each of the policy members of the committee represents a larger agency filled with people who may have their own views and expectations and narratives. Engaging with these is important for landing an architecture that fits in the context; but it’s not always possible in the committee space unless the committee member can bring them into the conversation. If there is no backing-and-forthing, just a presentation of “here’s how other places have addressed your issue,” the resulting report is not very useful. 

In brief: Having some tangible reference points – precedents and comparisons of different ways to structure public sector agencies or relationships to achieve particular goals – can help ground institutional architecture conversations in two ways. They provide a sense of the reality of other options – other structures exist in the world – and ideally also a basis for more abstract reflection on why the current institutional design is not working. The use of precedents comes with a downside, though, that they can anchor the conversation around designs that are not practical or applicable to the particular context. 

Geoff: I think there’s some different conditions which you’ve highlighted. One is that there’s a problem which matters to someone. They need a solution and they look out for solutions and precedents and examples. That happens quite often and in a way creates a certain kind of momentum because a minister or an official has to have some answer. It becomes matters to their career. There’s a second category which applies to the example I gave before  – the University for Industry. It had been put in a party manifesto. Again, that meant there was an obligation, there was a momentum to come up with something. You had to fill in the blank space with a proposal, a budget, and then people, et cetera. 

One other kind of example, which is in my mind,  is where actually there’s an idea which is being pitched. An example from the UK back in the 80s was a whole host of new regulators were created. Big utilities were privatized, some successfully, some with lots of problems. In a way, they were saying, well, electricity, water, railways, they’re all a bit similar. Let’s privatize them. That required them creating a new kind of regulator with quite a lot of power to regulate them.

Those partly drew on precedents, mainly from the US, but actually had to be designed anew. And there, the momentum, as it were, came, I think, really from a push of our ideas. There wasn’t an immediate problem that needed solving. Maybe there was a bit of a problem with public debt and so on, but it was more a push from ideas. 

II. LANDING THE DESIGNS

Geoff: In all of those cases, I think it’s very important that any design is deliberately incomplete, leaves quite a lot of space for adaptation and iteration. The spaces for design each have quite different stories of momentum and the day -to -day iteration between the providers of the money and the support and the initial design. But usually a crucial moment is the first serious chief executive, who then in practice will always do quite a lot of redesign to make it fit an environment which is always slightly different from what was in the head of the people who initially formulated. And that’s where there’s again a lot of to and fro back and forth.

One of the other examples I was going to give, which was much smaller than the complete privatization of all infrastructures, was one in Australia where I worked on creation of an Australian Centre for Social Innovation. That got the backing of the Premier of South Australia and was created as a new institution, but it wasn’t driven by a massive urgent problem. It wasn’t in a manifesto, but there was a feeling that it was an idea whose time had come. It was important to create an institution, a home for an unknown series of future efforts – not just a series of programmes that might come and go. Twelve, 14 years later, it’s still going strong. It has a fantastic capability. 

I wish other countries had something like that. In a way, it’s a proof that often an institution is a much better way of orchestrating capabilities than just short-term programs. It created its own resilience, its own momentum, its own legitimacy by working in many other parts of Australia, getting the backing of many other governments. 

Jessica: I think what both of you are saying is that initial document design blueprint is heavily, necessarily heavily, incomplete. Institutional architects need to recognize that completeness may be both naive and potentially harmful because of the ownership and the requirement for the first implementers to have space to make it their own. 

In this parallel with architecture, it’s a half-finished building. We’re not just turning it over to the decorator – the next steward of the design has to do much more than pick the wallpaper and choose whether to hang drapes or not. 

Juha: Perhaps a different way of framing this point is that if institutions are codifications of practice; codification is utilization of power. That means that you need to engage with the actors that then make the decisions about how the institution will function. 

In that sense, when you provide initial sketches or ideas, whatever it is, those are artifacts through which you catalyze and incentivize those discussions forward in order to get reflections, get comments, get feedback. 

In the political context, especially in the kind of like coalition political context, you would frame that through negotiations. So you need to put the first designs in place in order to identify where the spaces are to negotiate both the demand and supply for new institutional designs. This has been very much the process in the cases in which at least I’ve engaged with. There will always be crunch moments – those moments in which you can lose everything. Even if you don’t lose anything and you actually move forward with a new design that gains traction, that doesn’t really mean anything because you will then end up in another crunch moment. The question of how long you stay within the implementation cycle will always be there because that then gets to the questions of capacity. 

Having an advisory role in these processes is really valuable because it essentially creates a little bit of legitimacy to negotiate being those different actors and create that space in the middle. Oftentimes when you talk about putting a new institution in place, those actors that have more formal accountabilities, they also have these incentives in place. In that sense, having a little bit of freedom from the outside often helps in terms of articulating, creating a shared language and vocabulary in between putting those early sketches in place that requires a high level of trust because those are bound to be naive, like you say. That’s at least the dynamic as I’ve seen it. 

I think, Jessica, what you touch upon with your example from India is a really interesting one, because it’s very true that part of that process is pulling in examples and showing and illustrating what it could look like and what the outcomes could look like. It is more meaningful if also that happens proactively and intentionally. If I think back of the example on policy experimentation, I think one of the keys in terms of where the potential impact of that story lies is in regards to our relationship and engagement with OECD, because they were then doing the case studies and documentations of the initial demonstration of a new approach to policy design. That was much more meaningful in terms of how the continuation and the final outcomes of that experiments took place. 

So in that sense, being more explicit in terms of understanding that in each geographic context, in each location, there are always risks. There’s always that situation where you can lose that individual institutional innovation. But if you’re able to create multiple experiments, if you’re able to understand what really happens in different contexts, then gradually you actually might move forward. The accumulation of those individual demonstrations can be a change in how we understand institutions in the context of a specific question at hand. 

Jessica: In that sense the Indian model of expert committees are actually quite useful venues because the typical structure of one of these committees is to have a prominent individual with both the expertise and the political or civil service standing to bring people together, force some negotiations, and have some reputational capital. In the case of the Transport Committee, it was appointed by the Prime Minister and included the Joint Secretary for Power because the transport of coal uses a lot of railway capacity – do you move coal or do you move electricity? 

So these committees have the people who know the context on the committee. Among the consultants – some of the consultants were directly hired by the committee, but there were sometimes stakeholders involved in this kind of advisory role. The World Bank would usually help out because in the end, some of these things would be financed through World Bank loans. The Infrastructure Development Finance Company (IDFC) also supported some secretariats. Public sector arms of different consultancies would sometimes do pro bono work. 

The problem that this thing would run into, is that there was almost this expectation that the point was to produce a blueprint in waiting for the right moment. The terms of references – the political space – may imply immediacy – but in practice I think few of these committees anticipated immediate uptake of everything in them. And I think that’s why there are all these volumes that sit there. 

Geoff: This is very common in architecture. There are volumes of books of the buildings which were never built, the amazing designs for cities and buildings, some of which then sat waiting for the moment to shift or change of demand, but most never did. What you described is actually fairly rare, I think, in the public sector. Usually, if there is a design process, someone needs it, someone wants it. 

I actually think we need more of what you described, which is a little bit more speculative design to create options, which are then ready for a moment of need, like a pandemic. In a sense, the United Nations was an example of that a lot of work done on different variants of it. It was only when the conditions after the Second World War had shifted that those could be mobilized. I’m struck in most environments, there isn’t, as it were, a library of options to draw on. There isn’t a sort of the multiple volumes waiting for a new need. And a good example of that, which we are working a bit on is artificial intelligence regulation and governance, which has been clearly a coming issue for 10 – 20 years. 

When it suddenly became really urgent in the end of the last decade and Europe created its new Act and China started working on cyberspace administration, there were simply no designs out there. The work hadn’t been done to prepare. 

This I think we need a little bit more of what is common in physical architecture, which is the creation of the maquettes, the models, the examples, the potential blueprints, which are ready in case there is a shift of demand, a shift of need. That’s one of the weird purposes, I think, of a practice of institutional design is to have more things on the shelf, ready to go as it were when the need becomes more acute.

III. BUILDING THE SPACE FOR DESIGN

Jessica: That’s actually a good jumping off point to this question of how to build the space for even having an institutional design conversation.

If there is not a “power that be” that recognizes an opportunity and a problem and invites a design, how have you individually or with collaborators helped to create that moment? If we say that’s needed, we are trying to create the invitation to create the library of options, what should we do? 

My first instinct at this point would be a commentary in something like Nature plus a personalized advocacy strategy around that commentary to make sure that the points both have some standing (the platform) and are accessible (the advocacy). Then see who bites and has the power to create space for a sketch or more.  That’s just based on what I’ve seen happening in these kind of vague problems, no one has an owner, and it moves from idea into reality. What would you do and what have you done?

Geoff: It is, as you say, very context dependent. So three different quick different examples in the late 2000s. 

First, I and others worked on blueprints for “What Works” centers – publicly supported bodies to orchestrate evidence on fields like education or policing and so on. It took a little bit of proposing them, socializing the idea. Then after 2010, a new center -right government decided to back these. About 15 were set up, but the work had been done on promoting them, advocating them in a language which fitted, as it were, the new regime. 

A second example on artificial intelligence, I did do a proposal on an AI regulator about 10 years ago, which did persuade a government to create something a bit like it called the Center for Data Ethics and Innovation, but not on the scale with the powers I’d envisaged. My surprise was that more didn’t then materialize in other countries in those years. 

A very live example of the value of these libraries now is we have more elections happening this year than ever in human history, half the world’s population is voting. All of those elections are incredibly vulnerable to attacks, misinformation, deep fakes, and so on. No country on the planet has an institution capable of defending or acting quickly enough. And I think that’s one of the cases where one needs to be working on some options and proposals and socializing them. And we don’t quite know who will then be at the demand or who will have the authority and the will to implement them, but someone certainly will.

Jessica: When you say “proposal” and “socialize,” both of those are very high-level terms. Do you have a couple of more specific ways, like where does the proposal go? 

Geoff: For any specific topic, there will be a community of practice. There’s a landscape of stakeholders, people who have the power to make it happen, people who care about it, people who are influencers, and those will be different on any particular topic. One definition of a stakeholder is someone who can kill things as well as who can make things happen. It’s important to be aware, and you were both talking about this earlier, about who might be the negative stakeholder you need to neutralize or co-opt, et cetera. But that community will vary depending on the topic.

Jessica:  Juha – do you have thoughts on this and is there some more specific advice to make “propose” and “socialize” more tractable for someone who’s listening and says, “I have an idea to solve a problem and we need to create the space for design”?

Juha: I would split that in three. The first thing, and this is actually really topical because right now we’re working with European cities missions – what the mission governance and implementation should look like. In that context, when we talk about missions and governance of missions, my very typical statement is that we need to simulate tangibly how that would look like, because the conceptual side of things is one side, but it’s difficult to move from the conceptual into the concrete and tangible. How would the institutional idea or innovation actually operate and work and what would be the everyday realities of those people who need to make the decisions if the new institutional architecture would be in place. 

That’s something that actually has a really strong analog to architecture. Architecture generates those examples through competitions because it’s really difficult to imagine a building just conceptually. Institutional architecture needs to create a method of coming up with examples that essentially are able to capture different conditions. I think that’s a good method for institutional innovation as well. That’s one potential tactic to move forward that at least I found to be rather useful.

The second is what we’ve been already talking a lot about: a repository, basically illustrating examples and creating comparative cases. There’s a gap in terms of documentation of those repositories. I think one needs to create a little bit of a way of creating heuristics to identify the commonalities better because yes, everything is really contextual at the same time around questions of legitimization or consensus in terms of creating demand for an institution, there are still methods that work better than others. To be able to illustrate what those look like is rather useful. 

And then the third one is, in a way, creating scenarios. Two things in terms of actually getting the decisions is that if you’re able to make the case in terms of where the risks and failures lie, unless there’s enough emphasis on institutional capacity and institutional innovation, or where the gaps in success will lie is another way, because oftentimes those questions come at a point in which it’s too late to react in terms of the actual delivery of, for instance, political goals or an agenda. Doing a bit of work in terms of actually exploring how that pathway could or should look like can be rather valuable. 

With all of these three, and this is, I guess, where we are heading with this as well – they require quite a lot of work. It’s much easier to talk about this conceptually than actually do the work in terms of putting the simulations, repositories or tangible scenarios in place.

Jessica: There is a fairly broad spectrum of ideas here – on the one end, approaches to creating space for institutional designs for specific issues; on the other broadly priming a relevant group to have a sense of the possibility of design. It’s little bit like your book, Geoff, Another World Is Possible – priming people for a sense that there is the possibility of design. 

Are there things in the middle that you’ve seen work or have made work in the past?

Geoff: There will always be a number of layers. Maybe to make this concrete, take something like the circular economy which has been around as an idea for 30 or 40 years at least. There you want people to be brought into a broad agreement that this is a different way of thinking about the economy, the flows of waste and plastic and paper and electronics, and so on in a city or in a country. Usually you quickly come to the need for new institutions. 

I was involved in 20 years ago in creating RAP, which was meant to be the agency, the development agency for the circular economy and waste in the UK. It was a mixture of capability and experiments and nudging and so on. I think there’s increasingly a need for what is beginning to exist as sector specific circular economy partnerships – collaborations for, let’s say,  the electronics industry to help them reduce the mountains of e -waste – and at a city level because a lot of the practical management of waste is usually under municipal authorities. Again, they need specialist agencies, which may be slightly different from the ones they’ve inherited. So with any topic, there will be a number of layers from this broad and strategic to the very practical and implementation oriented. 

Part of the task of design is knowing which bits of those are about broad strategies and which actually need a very focused institution. And probably the lesson of every public sector around the world is usually things are only done well if they’re both. If there’s both a broad purpose, let’s say educate children, and then there’s institutions who really live and breathe nothing other than teaching and education and have a depth of knowledge about just doing that. It’s nested intelligence is the more sort of technical way of thinking about a series of nested capabilities which add up to running a system in a new way. 

On the circular economy, in a way, the striking thing is 40 years on, we’re still missing most of those institutions, I would say, compared to the much more familiar ones like industry bodies, or traditional utility regulators, and tax bodies. So these things take a lot of time.

Juha: I think the story of the circular economy illustrates many of the pain points in this process in many ways. My sense is that many different countries and cities did their own experiments, but there was very little learning in between those. The accumulation of institutional innovation and learning was much too slow. If there had been a bit of a support capacity to imagine those pathways in terms of how the more specific and more abstract would converge and where the actual institutional needs are that would have accelerated that a lot. I think right now with circular economy we’re in exactly that moment. 

The issue is that when we have done like multiple different experiments on institutional innovation in the context of circular economy, you have actually have a really wide ecosystem of actors. So then in order to consolidate, the engagement process will also be fairly heavy. So circular economy in many ways, I think is not necessarily a good example as a success, but really good example in terms of learnings around the importance of institution based on consolidation of learnings.

IV. BRINGING IT TOGETHER – CREATING THE SPACE FOR DESIGNS THAT CAN LAND

Jessica: We started with a set of spaces where there had been a recognized invitation to create a new institutional design. The conclusion from that is that the blueprints need to be there – but that doing the blueprint is in some ways not the hard part. The harder part is leaving enough openness to have it land.  

In the second half of the conversation, we have brought up many more examples where there should be more space for design than there actually is. 

On the circular economy, where you two are essentially arguing that there should be a reconfiguration into more of a nested architecture, order to move from experimentation to consolidation. What would be your first bet if you say had an invitation from an individual who has financial resources who said, get this started? How would you go about creating the space for a real conversation about designs that might make the leap from blueprint into practice? 

Geoff: Again, on the circular economy, in a way, what’s good there, because it’s 40 years on, a lot of people understand the need for a completely different model of how industry works. There are lots of examples from Japan to Finland to Canada, which can be drawn on, even if some of that mutual learning is weak. What I would do if someone said, “Here’s a lot of money to do something new in this field” is take one of the sectors which is furthest behind, like electronic waste or clothing – clothing recycling is dramatically lower than paper or glass – and see what are the points of leverage, what are the points where you can bring together a coalition. You can achieve momentum through success, through action, not just through paper blueprints. And then maybe do so in a way which is potentially replicable, which could be reproduced at a city or national level. 

Jessica: So a big part of design is basically creating these moments in which there’s openness to a blueprint from enough people who could then take it forward. And a big part of that is creating the sense of possibility. It seems to be the message that’s coming out of these examples.

Geoff: Yeah, and just one very simple addition to that. In some ways, this is exactly the same with physical architecture. If you’re wanting to create a new architectural movement, you have to both create the demand and the supply simultaneously. You have to make people want your new idea, and you’ve got to be able to build it and orchestrate the resources. But whereas an architect can often work with a single client, they only need to persuade one super rich person, let’s say, to buy their building, the sort of things we’re talking about there will nearly always be many more players it will be much more social, much more political, much more complex, in its nature.

Juha: Well, one of the current work streams that I’m engaged with is with the Finnish Ministry of Environment on circular economy. And there, where it easily ends up is that no single actor has a mandate around institutional innovation, or actually solving, if you want to use that frame, the issues around circular economy. Because even in the national context, actually the EU is a very heavily legislative body in this sense or the private industries have a significant sway in terms of how they’re doing things. 

It becomes more of a question of how to create a space in which there is sufficient alignment between different actors that there is actually demand towards moving to the collective direction in order to move forward with the circular economy agenda. 

In reality, in this case, for example, those actors are aligned. They really want to move towards the same direction. For example, in the national domestic landscape – construction is by far the most impactful domain in terms of circular economy. The industry already understands that this is where they need to get because of the signals they’ve gotten from the EU, but in the national political landscape. But there is no space to have that discussion openly in a manner in which there is also an actionable pathway forward.

So if there would be that pile of cash in terms of actually pushing forward institutional innovation, it’s in those areas in which there’s a gap terms of mandates and accountabilities, yet there’s really clear clarity in terms of the problem and demand. How the institutional innovation in those places happen is really interesting and, I think, a really good and specific problem for us moving forward.

Jessica: So I’d actually agree with both of you. It seems like the common thread here is that uncovering latent coalitions can be an important way to move forward on constructive, practical, institutional design. 

This is one of the things you can do with financial resources, but not necessarily political power and standing other than a reputation for trustworthiness and some degree of legitimacy to be speaking and convening with people. With resources, you can take the time to understand, matchmake, document, and bringing together a supply of possibilities with a convened coalition that is latent just because they haven’t discovered each other yet. 

One of the big task areas for institutional architecture is to be that intermediary between the repository of possibilities and the convener of the right group of people who haven’t quite discovered each other. In those moments, you move into a potential blueprint space. That is a critical part of institutional architecture that differs from physical architecture – you have to create your “client,” rather than just hang out a shingle.  

The second big thing came from the first part of the conversation, which is that the institutional architecture blueprint must be incomplete by design to enable ownership at a level that’s more than finishing and fittings. 

These are two things to explore in future conversations. And we should – we can –  do that with a larger group in the community. 

Any other parting thoughts? 

Geoff: You can never know the trajectory. One example, which I wrote a bit about in my most recent book on science is Vannevar Bush, who probably had more influence on the world of science and technology of the 20th century than anyone else. He ran the US science operation in the Second World War, and after the war proposed to Harry Truman to create a National Science Foundation to drive forward American technological expertise.

At first, he proposed a budget of $120 million, which was several billion in today’s money. He didn’t actually persuade Truman, who gave him three million instead. It was a complete failure for a moment and sort of dramatically smaller scale than Bush had hoped for, though he had built a coalition. But then the dynamics of the Cold War sort of came to his aid and he got a bigger budget than he’d asked for, because the need, the politics shifted. The owner of an institutional idea will never be able to own the context and the environment and the resources, but he was smart enough to be adaptable both to triumph and disaster and then triumph as it were, as the winds changed.

Jessica: But in his case, the incremental well-thought-through documentation of a series of idea “Easter eggs” was important. So that’s also an important point.  

Geoff: He wrote think pieces all the way along, which tried to encapsulate the idea.

Juha: As an institutional architect you need to be really able to react to those different types of dynamics almost on the spot if those happen. The point we’re hopefully providing for the community is that you don’t have to do it alone – you can pull in different people at the right moments in order to actually be able to reply and respond to some of those requests that emerge. So I completely agree with that point.

Jessica: Also, it’s quite a high leverage investment to create those initial “Easter Egg” documentations of ideas. For philanthropies out there who are seeking to change the world: the population of the sense of possibilities, as well as the record of ideas that exist when the moment of crisis arrives is a pretty well-trodden path to significant change.

Geoff: There’s absolutely no doubt from history that creating institutions has vastly more impact than just individual programs. If you create a successful one, your influence and impact in the world goes on for decades, not for years. So that’s why it’s worth trying to think really hard about how to do this well.

Jessica: Exactly. Well, I think that’s a good note to end the practice on. It’s hard to create a day -to -day picture of an incredibly varied profession, but I do think we’ve gotten to some at least useful structuring points about the formation of the design space, the incompleteness and nature of the design space relative to architecture, both of which we could go through many conversations with guests and others in this field on the finer points of each of these. But through the accumulation, the larger picture of possibilities and patterns will come.