The ethics of co-design

There is a lack of definition in policy of the term co-design, and yet local authorities and developers are increasingly using it. To avoid that this term becomes meaningless, it is essential to define how to run co-design processes ethically. Building on case studies, professional experience, collaborations with communities, and a Participatory Action Research approach, this paper defines a set of principles on how to run a co-design process ethically and genuinely including communities in decision-making. Departing from the legal Principles for Fair Consultation in England and Wales, the paper expands them and results into ten ethical principles for co-design.


Introduction
The last decade has witnessed the emergence of the term 'co-design' in urban regeneration processes, as well as other related terms such as 'co-production' or 'co-creation'. In England, as in many other parts of the world, local authorities as well as third-sector and private organizations are using the term co-design to brand consultation events where planners and architects ask local communities questions related to the planned transformations in the built environment and justify their proposals with some of the inputs received during that consultation. In most cases, these consultation processes do not involve any actual co-creation or co-design since they do not involve any collective reflection on how the future urban space in question could be. The overuse of this term has become very problematic since there is not a clear definition in policy of what codesign means and what should be expected from a process to be accurately named codesign. When every consultation event is branded as co-design, the term can become meaningless and generate an adverse reaction among communities, who get tired of participating, giving up their time in explaining their ideas, which are then not considered when making decisions. Other terms in urban planning, such as 'regeneration' or 'mixed communities' are already generating adverse reactions among communities because of their misuse. To avoid the same happens to co-design and to keep experimenting with radical forms of direct democracy and decision-making in urban design, it is crucial to define what we mean by co-design and how to carry it out responsibly and ethically as an urban design practitioner.
There is a lack of a clear definition in policy and, to a certain extent, in urban literature. There is plenty of literature on 'collaborative planning' (Healy 2006), 'communicative planning' (Innes 1995) and 'radical planning' (Beard 2003). There is literature on the principles and ethics of co-design in the arts (Kelly 2019). However, there is little literature on the professional ethics of facilitating a co-design process. Al Waer and Cooper (2019) have explored the role of facilitators in 'community-based', 'design-led' events. Community Led Housing London has published guidance on co-production, with a strong focus on governance, the delivery of a project, and the relationships created between different stakeholders (CLHL 2022). At an international level, the International Association for Public Participation (IAP2) has come out with seven 'Core Values for the Practice of Public Participation' (IAP2 2017), which aim to 'help make better decisions which reflect the interests and concerns of potentially affected people and entities' (IAP2 2017).
Co-design is not the same as public participation. One of the key differences between public participation and co-design is that the later implies collective design, as the term itself suggests -the prefix co-means together, 1 which leaves co-design as designing together. In order to create the conditions for co-design to take place, it is essential to recognize and address the existing power imbalances, and to put in place the necessary structures and partnerships to give power and skills for people to actively participate in the act of designing places together. Co-design involves teams of urban designers that have the necessary skills to facilitate these processes of designing collectively and contribute to enabling the conditions for it to happen.
In terms of policy, while England has specific regulations on how consultations should take place for statutory planning documents and planning applications, policies are extremely vague about how to involve communities in the planning process. For example, the National Planning Policy Framework (MHCLG 2021) explains that plan-making should 'be shaped by early, proportionate and effective engagement between plan-makers and communities, local organizations, businesses, infrastructure providers and operators and statutory consultees' (MHCLG 2021, 8). However, it does not detail what 'early, proportionate and effective' means and how it will be assessed. The Mayor of London (2018) guidance on estate regeneration provides some details on how engagement could be, but this is just guidance rather than statutory policy, and it does not provide mechanisms to establish minimum requirements nor to check whether local authorities and housing associations are carrying out a meaningful engagement. Furthermore, it is rare to find the term co-design in a statutory policy in England, yet local authorities use it continuously to brand their consultation events. Although policy is generally vague, there are legal precedents in England and Wales that establish the 'Principles for Fair Consultation' (Sackman 2020), which establish what local authorities need to do to ensure that a consultation process has been carried out lawfully. These legal principles share many aspects of IAP2ʹs core values, such as that people should be given sufficient information to consider proposals and that the consultation process should influence the decision. Although they are meant for consultation processes rather than for co-design processes, they can be helpful to understand what is legally considered fair. The Royal Town Planning Institute (RTPI) provides advice on ethics and professional standards for planners in the UK, which includes among their required ethical standards '(r)unning a consultation event, or responding to or taking account of community engagement/responses, in a proportionate and open manner' (RTPI 2017) and provides some general guidance on this. But it does not address how to involve the public in decisions on urban design. This lack of definition in policy of co-design results in a misuse of the term for naming what are actually consultation events. This generates discontent, situations of injustice, and imbalances of power between decision-makers and communities. The latter are asked for their time and local knowledge with nothing in return and their ideas not being considered. As urban design professionals, it is not ethical to pretend that one is facilitating a co-design process when there is no actual co-creation involved.
This paper aims to define the principles and ethics that a co-design process should follow in urban regeneration schemes. This paper does not focus on methods or techniques for facilitating the co-design process but on the ethics and principles. These ethics of co-design are based on evidence, grounded on literature and built on case studies and professional experience.

Methodology: participatory action research as a co-design approach
This paper builds on the results of various knowledge exchange projects in partnership with community groups to co-produce knowledge. These projects have used a Participatory Action Research methodological approach, which according to Fals-Borda is 'a research methodology that combines theory, action and participation (. . .) committed to further the interests of exploited groups and classes' (Fals-Borda 1987, 329). These projects have established partnerships with community groups, treating them as 'full partners' in the co-production of knowledge as argued by Fals-Borda (1995). The ethical principles presented in this paper emerge from the experience of these Participatory Action Research projects and the interaction with the projects' partners.
The partners of these projects are communities living in social housing facing the demolition of their homes. The projects have involved co-producing a social impact assessment of the planned scheme on their neighbourhood (Colombo et al. 2021) -looking at the potential impact it will have in their lives, social relationships and care infrastructure -and co-designing alternative schemes with communities, which they then present to their local authority as their vision. These are not statutory plans developed by the local authority, but alternative visions created by residents. This paper uses two of these projects as case studies to discuss the ethics of co-design: the Community Plan of William Dunbar and William Saville houses in South Kilburn, London Borough of Brent (co-produced with WDWS Residents' Association and Granville Community Kitchen), the People's Plan of Alton Estate in the London Borough of Wandsworth (co-produced with the local group Alton Action and Just Space). All these projects have included a diversity of co-design workshops and activities with residents, in which they have actively participated in the co-production of evidence and proposals. Researchers, with the support from community partners, facilitated these co-design activities. Some co-design workshops were run in person in a community space within the neighbourhood, while others -during the Covid-19 pandemic -were done online using Zoom and the platform Miro.com to co-produce the evidence and proposals. In addition to the co-design activities, the team held regular meetings with community partners to track the project's progress and carried out surveys as well as multiple conversations with residents and stakeholders that informed the evidence collection. This paper does not have the scope and space to elaborate on the context of the case studies.
However, the reports of the projects are available online . In addition to these two case studies, the paper also uses other collaborations with community groups to illustrate some of the arguments. These collaborations, which have taken place between 2015 and 2022, have consisted on advising communities on planning and urban design issues, as well as providing them with skills to participate in decision-making.
While these projects have involved exploring co-design methods, this is not the paper's focus. Instead, this paper focuses on the ethics of co-design, which has been deeply experimented with in these projects. This experimentation has explored how to create equal partnerships with communities and address the power imbalances between professionals and communities. They have also consisted of looking at the time, resources and activities needed to carry out a genuine co-design process. The projects also explored how to involve communities at all stages, from creating the brief to co-designing the process and the outcomes, as well as looking at aspects of the implementation stage such as the viability studies. This equips communities with skills to have more power in participating in decision-making processes about planning and urban design. The projects have also explored how to co-produce schemes that generate a sense of ownership and authorship in the process. For example, in the case of the People's Plan for the Alton Estate, the local campaign Alton Action showcases the People's Plan on their website as their plan and has collected a significant number of signatures from residents to support it. After the plan was cocreated, Alton Action kept working on it by developing their infographics that explain the plan to other residents.
Through these case studies, the paper argues that the ethical principles of the Participatory Action Research methods can inform the ethics of co-design processes run by professional planners and urban designers, since the imbalances of power between researchers and subjects are very similar to those between professionals and participants in co-design activities: Do not monopolise your knowledge nor impose arrogantly your techniques but respect and combine your skills with the knowledge of the researched or grassroots communities, taking them as full partners and co-researchers. That is, fill in the distance between subject and object. (Fals-Borda 1995) This argument is informed by both the two mentioned knowledge exchange projects, as well as professional practice, including the experience of acting as consultant in various co-design processes for public and third-sector organizations, in which these same ethical principles have been applied. While this paper does not use these professional projects as case studies, this practice does inform the ethical principles presented here.
The paper first explores how to go from a consultation to a co-design process, and the implications. After that, it takes the abovementioned legal Principles for Fair Consultation that apply in England and Wales and explores how to expand these same principles to take them further into a co-design process that involves creating equal partnerships with communities and collective thinking.

From consultation to co-design
"Co-creation is not gathering ideas from a bunch of people, then figuring out what you do with them later. That's a consultation. One that is at best poorly planned, and at worst highly insulting." (Lee 2020) Going from a consultation process to a co-design process is a great challenge for architects, planners, local authorities, developers and residents and communities participating in the process. This implies moving from just collecting people's ideas and producing design proposals that respond to these inputs, to a genuine process of co-creation, where different people come together to collectively think about the future of the area, where meaningful partnerships are created, and where power imbalances between different stakeholders are addressed.
These three points -collective thinking, creating partnerships, and addressing power imbalances -are essential for this transition from consultation to co-design. First, a codesign process needs to achieve moments of what Di Siena (2019) defines as 'collective intelligence', which implies that people think collectively rather than as individuals. The term 'collective intelligence' has been used by several authors in diverse disciplines. Smith (1994, referencing Malone & Bernstein 2015 defines 'collective intelligence' as the moment in which a collective of humans think as one entity, rather than as a collection of individuals (Sendra and Di Siena Forthcoming). Achieving moments of collective intelligence requires facilitation, as well as the time and resources to carry out this facilitation. It also requires local authorities, developers, architects, and planners to acknowledge that there will be conflicts between different stakeholders, which will need to be addressed during this process. Local authorities, planners, architects and even communities are afraid of conflicts that might come up during these processes (Sendra and Sennett 2020). Second, a co-design process implies the creation of partnerships between diverse stakeholders, which need to include residents, community groups, local businesses, the local authority and/or other public, private or third sector organizations involved in the development. Designing the process to create these partnerships is an essential job of the professionals facilitating the co-design process since they create the basis for a collaboration that gives residents and communities equal decision-making power. The third point is related to the second one, and it is about addressing power imbalances. A co-design process cannot start from the premise that everyone holds the same power to participate in making decisions. As Lee (2014Lee ( , 2015Lee ( , 2021 explains, codesign processes must address historical injustices, structural inequalities, and power imbalances. Ignoring them would mean reproducing them in the process and giving power to those that traditionally hold it. The guidance published by CLHL make emphasis on 're-shaping power dynamics' (CLHL, 2022, 8) and practical actions to address this. This is also present in the work of Fals Borda (1995), who, when discussing Participatory Action Research, also proposes to address power imbalances in the co-production of knowledge. Arnstein's (1969) Ladder of Participation illustrates this transition from consultation to co-design. Arnstein's seminal work represents the different levels of participation in a ladder with eight rungs. The upper the steps, the higher the degree of participation. The two bottom ones, grouped as 'nonparticipation', are 'manipulation' and 'therapy'. The three middle ones, grouped as 'degrees of tokenism', are 'informing', 'consultation', and 'placation'. The three top ones, grouped as 'degrees of citizen power', are 'partnership', 'delegated power' and 'citizen control'. The transition from consultation to co-design is a transition from 'degrees of tokenism' to 'degrees of citizen power' -more specifically, from 'consultation' to 'partnership'. Since 1969, there have been multiple interpretations or critiques of Arnstein's Ladder of Participation (Connor 1988;Hulbert & Gubta 2015;Natarajan 2019;Ciaffi 2019). Based on the case studies presented in this paper as well as on professional experience, co-design processes are too complex and expanded in time to remain continuously in one single rung of the Ladder. A co-design process starts when (co)defining the scope of the project and continues through the design and construction of the project (see Al Waer and Cooper 2019 on 'post-event stages'), as well as in shared forms of governance and maintenance once the project is built. Through a project's life, the Ladder's different rungs can vary from 'informing' to 'citizen control'. In many situations going down the Ladder might not mean a lack of democracy or a loss of the codesign process. However, despite the difficulty of remaining permanently on the partnership rung, the Ladder of Participation provides a clear way for urban designers, local authorities, developers, and communities to reflect on which rung of the Ladder they are in each moment and what they can do to move towards a partnership or other degrees of citizen power. The Ladder is an easy-to-understand tool for the public, free of jargon and which people can associate with a particular situation they have lived in during a participatory process (see Westway23 2019: a reflection by a community group on the Ladder in their relation to an institution).
Achieving degrees of citizen power requires experimentation on expanding current practices of participation and consultation into genuine co-creation. Since there are already legal precedents in England and Wales on what is a fair consultation, this paper departs from these legal principles to expand them into ethical principles for facilitating co-design processes.

From legal to ethical principles
In England, participation is a legal right set by precedents in law. Public authorities have an obligation to carry out a fair and lawful consultation before they make any decision. The courts have quashed decisions when the consultation process has not been run fairly. These have set up precedents in law through which individuals or groups could take public authorities to the courts. These precedents are popularly known as Principles of Fair Consultation (Table 1), Sedley Principles (Sackman 2020), or Gunning Principles (Just Space 2017) in reference to the case that set the precedents.
Although this paper argues that co-design is not consultation, as it goes far beyond it, these legal requirements are helpful to understand what will be legally expected when running a co-design process. Legal does not necessarily mean that the process is entirely just; it just means that it has been carried out lawfully. As Sackman (2020, 112) explains, 'Judicial review is concerned not with the merits of the decision (. . .), but only with whether the public body has acted lawfully'. For this reason, while unpacking the six Principles of Fair Consultation, the paper explores how to expand these principles to reflect on how any co-design process can be carried out ethically -in addition to lawfullyand ensure that no one is disadvantaged.
As Kelly (2019, 329) explains, 'ethics are intrinsically involved in engagement with the world, including participatory and collaborative design projects'. Since planners act in the 'public interest' (RTPI, 2017) and co-design processes have a strong engagement with participants, paying attention to the ethics -rather than just at the legal or statutory requirements -becomes essential. As Kelly (2019) explains, in collaborative design projects, it is necessary to pay attention to both the ethics of the outcome and the ethics of the process. She also explains the complexities of ethics in design disciplines, where there is not always a correct answer.
By unpacking the six legal principles and looking at them from an ethical rather than just a legal point of view -i.e., not just whether a consultation has been carried out lawfully, but whether it also has been carried out ethically -it is possible to extract six key lessons (one per principle) on how to run a co-design process (Table 1):

Co-design must start before any decision is made
It is quite frequent that a participatory process starts when a decision has already been made. This contradicts the first Principle for Fair Consultation, which explains that consultation should take place 'when proposals are still at a formative stage'. The Core Values of the International Association for Public Participation also say that communities affected by a development 'have a right to be involved in the decision-making process' (IAP2, 2017). To genuinely participate in a decision, it is essential to start the co-design process before making any decision.
In many cases of estate regeneration, local authorities started engaging with residents and local communities when there were already decisions made about densifying the estate to address housing targets, demolishing, and redeveloping it.
Alternatively, in the two knowledge exchange projects carried out with residents from William Dunbar and William Saville houses ) and the Alton Estate , which were exploring an alternative approach to that of the local Table 1. Ethical co-design emerging from legal principles for fair consultation 8 .

Principle of Fair Consultation
Ethical co-design process 'consultation must be at a time when proposals are still at a formative stage' Co-design must start before any decision is made 'the proposer must give sufficient reasons for any proposal to permit of intelligent consideration and response' Provide sufficient information and skills to codesign 'adequate time must be given for consideration and response' Communities must feel comfortable with the timing of the co-design process 'the product of consultation must be conscientiously taken into account in finalising any statutory proposals' Communities must have decision-making power 'the degree of specificity with which, in fairness, the public authority should conduct its consultation exercise may be influenced by the identity of those whom it is consulting' Consider local identities 'the demands of fairness are likely to be somewhat higher when an authority contemplates depriving someone of an existing benefit or advantage than when the claimant is a bare applicant for a future benefit'. 9

Collective benefits
Source: author's elaboration from principles for fair consultation, which come from two cases (cited in Just Space 2017 authorities, the first step was understanding people's relationship to their neighbourhood and their neighbours. Through this process, it was possible to understand the relationships of care and solidarity between their neighbours, the attachment to their homes and particular places, and the potential social impact that demolishing the homes could have on residents (Colombo et al. 2021). Once evidence on these issues and on their housing and social needs had been co-produced with residents, collective discussions on the kind of improvements and potential proposals for regeneration could start taking place. This approach tested by the two knowledge exchange projects does not just contribute to making the co-design process more democratic but also builds trust with communities since proposals develop from their needs rather than from predefined agendas. When asking for feedback from William Dunbar and William Saville residents, some participants stated that it was the first time they felt truly heard, and the co-design activities had regenerated a comfortable environment to express themselves.

Provide sufficient information and skills to co-design
Local authorities, in some cases, try to seek consent (see Architects for Social Housing 2018) to carry out their proposals instead of co-developing proposals with residents and exploring alternatives. The diverse alternatives are explored and discarded in closed rooms before any consultation occurs. By the time they start the consultation process, there is one single proposal on which they are consulting.
A case that illustrates this very well is the ballot that takes place before an estate regeneration that involves the demolition of social housing, which is a condition that the Mayor of London set up to award funding to build affordable housing to local authorities and housing associations. 2 In many cases, this takes place in social housing estates where there has been a lack of investment for years, where there are residents living in temporary accommodation, and where some buildings are dilapidated. Local authorities and housing associations offer one option to residents, which in many cases is a one-fitsall approach to demolish and redevelop the whole estate, regardless of whether some buildings are in a better state than others. On some occasions, they offer temporary tenants a permanent home only if the result of the ballot is yes. This means that residents are left with the options of: (a) vote yes for the single option of demolition and redevelopment; or (b) vote no and regeneration will not take place -which mean that the decay of the estate will continue -and temporary tenants will not get a home.
This situation is not unique to estate regeneration ballots but also takes place in other forms of consultation. In the case of Cressingham Gardens in the London Borough of Lambeth, the council removed the refurbishment options from the consultation, which led to a Judicial Review that quashed Lambeth council's decision to redevelop Cressingham Gardens on the basis that the consultation was unlawful because the council decided not to proceed with the refurbishment options (Sackman 2020; Sendra 2018; Sendra and Fitzpatrick 2020).
Alternatively, this paper proposes that local authorities and professionals facilitating the co-design process should provide communities with all the necessary information to co-produce proposals. This goes far beyond presenting and discussing predefined alternatives so communities can choose or vote for one. Instead, co-design means that communities must be involved in exploring and co-developing these alternative proposals.
The knowledge exchange project with residents from the Alton Estate explored this. The resident-led group Alton Action, which were concerned about the demolition and redevelopment scheme proposed by Wandsworth Council and private developers Redrow, 3 approached a group of researchers from UCL to co-produce an alternative plan, which later received the name of the 'People's Plan'. Through a partnership between UCL, Alton Action and the London-wide network of community groups Just Space, they got funding from Research England to develop this knowledge exchange project. As part of the process, the three partners organized sessions to equip communities with skills and knowledge to co-develop proposals for their neighbourhood. This included a session on community organizing, organized by Alton Action, a session on built and cultural heritage organized by one of the UCL researchers (Michael Short), sessions to discuss the environmental impact of different types of interventions to prepare for conducting a Whole Life-Cycle Carbon Assessment organized by one of the UCL researchers (Sahar Nava), sessions where neighbouring community gardening organizations came to advise on urban farming to Alton Estate residents, and a session conducted by Just Space on London policies for estate regeneration and environmental issues.
This illustrates how the second Principle for Fair Consultation can be stretched into providing communities with the information, evidence, tools and skills to co-develop proposals with those facilitating the co-design process. CLHL (2022) also points at the need at providing skills and support so they can participate in the co-production process.

Communities must feel comfortable with the timing of the co-design process
Co-design requires time so it can be carried out fairly. This is something that professionals working for local authorities or other organizations need to require. Otherwise, the effects of carrying out a co-design process can be more damaging than not carrying it at all. As Lee (2021) explains, 'do it responsible or don't do it at all'. The relevance of carrying out a fair co-design process with appropriate time is generally underestimated, particularly on the social, political, economic and design implications it can have.
To carry out a co-design process, it is necessary to build trust from communities towards the professionals facilitating the co-design and the institution initiating the process. Trust can only be achieved with time, positive interactions, and by carrying a codesign process that is then reflected in tangible proposals that benefit everyone.
First, it is essential to allocate time and resources to a campaign that advertises the codesign process and produces an inclusive and wide-ranging invitation to all actors to participate. As part of this campaign, it is vital to invite local community groups (interest groups, place-based groups, cultural groups, ethnic groups, faith groups) to participate in the process and partner up with them so they can contribute to disseminating the invitation among their networks. In the People's Plan for the Alton Estate and in the Community Plan of William Dunbar and William Saville houses, it was essential to partner up with local community groups to disseminate the invitation to the co-design workshops and events co-hosted with them. Working with local organizations in these two case studies provided a level of trust towards an external organization that otherwise would take a very long time to build. Part of this collaboration implied co-hosting the workshops with them and involving local groups in facilitating the workshops.
During the first event, it is essential to clearly explain the timeline of the process, how they can participate, what the expected outcomes are, and when they can expect these outcomes. This also allows discussing and receiving feedback on this timeline, as well as on the schedule of the events. It is crucial to allocate time for multiple events hosted at different times and days of the week so that they can include people with diverse commitments and schedules. In the case of William Dunbar and William Saville houses Community Plan, most workshops were in the evening during weekdays. When the workshops took place on Saturday mornings, the population coming to the workshop changed, and people with caring and parental responsibilities could attend.
Professionals appointed to facilitate a co-design process generally have a time-limited contract equivalent to the length of the co-design process. Therefore, professionals must ensure that -during their contract -there is time to have events dedicated to discussing and providing feedback on interim outcomes of the co-design process and to set up community-based structures to follow up the next steps of the co-design process, which will be discussed in the next section.

Communities must have decision-making power
One of the key frustrations of communities that take part in any participatory event is that they feel they have not been listened to, and the decisions made do not reflect what they said during the consultation events. As mentioned, the reason is that the agenda has already been set before the consultation. On other occasions, even when the outcomes of a participatory process reflect what the residents said, these decisions are later modified when the set agenda changes. This is the case of what happened in the case of William Dunbar and William Saville houses. In the early 2000s, the area was chosen as one of the New Deal for Communities programme, a scheme developed by Tony Blair's administration that targeted 39 deprived areas in England. As part of the New Deal for Communities, Brent council produced a masterplan that included residents' participation. Through the participation mechanisms carried out during the New Deal for Communities period, residents from William Dunbar and William Saville houses contributed to the decision that the two buildings should not be demolished, but refurbished, and that is what the masterplan reflected (see South Kilburn New Deal for Communities and Brent Council 2005). However, in 2016, Brent Council carried out a Masterplan Review (Brent Council 2016), which in 2017 became a Supplementary Planning Document, in which William Dunbar and William Saville were proposed to be demolished and replaced by a scheme with an 83% of private market homes in 2026-2029(Brent Council 2017. The brochure explaining the Masterplan Review explains that this was carried out through a consultation that included: 'pop-ups, exhibitions, local consultation stations, flyers, infoletters, e-newsletters, online surveys and feedback forms' (Brent Council 2016, 4). During the knowledge exchange project, some residents explained that they were shocked when they found out in late-2016 that their homes would be demolished, some of whom had been promised that William Dunbar and William Saville would never be demolished. This frustrates residents with their lack of power to influence decisions and results in distrust towards local authorities and anyone doing consultancy for them, as well as in consultation fatigue.
The alternative approach is to create partnerships with local community organizations to co-design the scheme with them. The Community Plan of William Dunbar and William Saville Houses explored this alternative approach by partnering up with the local organization Granville Community Kitchen and collaborating with the residents' association. The researchers co-hosted co-design workshops with residents, which involved many steps, including understanding people's attachment to the place, their social relationships and networks of care, understanding their needs, co-producing a social impact assessment with them (Colombo et al. 2021), co-designing proposals through multiple techniques, and having various sessions to discuss proposals and refine them through feedback. By the end of this process, residents felt identified with the proposals and considered them their proposals rather than proposals by someone else in which they were being consulted.
In addition to the abovementioned process, it is crucial to set up community-based structures that allow for the continuation of the co-production process once the concept designs have been agreed. These community-based structures can be community steering groups, regular public presentations to discuss the updates of the project, as well as other methods that ensure continuous participation. This is very important because concept designs can go through alterations, which are not properly explained to communities and that then come as a surprise to them, which makes them feel disempowered because their proposals have been disregarded. Therefore, local authorities and professionals must ensure that these mechanisms for continuous participation after the codesign process are in place.

Consider local identities
After the original four Principles of Fair Consultation, other two were added. These brought some details that are very important when carrying out a co-design process and which are normally overlooked. The first of these principles mentions the 'identity of those whom it is consulting'. One could think that by organizing a series of co-design events that foster an inclusive environment, which is advertised publicly, and which involves creative practices for co-creation, one would be achieving an inclusive codesign process. However, such a process could ignore historical and structural injustices in the area (Lee 2014(Lee , 2015(Lee , 2021. Therefore, before planning a co-design process, it is vital to understand the history of struggles in the area.
For example, a public space in London, in an area with a significant contrast between a low-income and a very affluent population, was going to go through a re-design process. The space was a result of struggles and campaigning decades ago. When designing spaces in which there is a significant socio-economic contrast, as it happens in various neighbourhoods in London, it is important to understand the needs of the population with lower income and how a design may affect them. For example, lower income families have less access to public and free amenities for their children, while higher income families may have access to other spaces. This means that co-design is not necessarily achieving a compromise between different views, as this can lead to injustices towards lower income populations. In contrast, co-design must consider current and historical injustices, as well as doing research on the historical struggles that the spaces about to transform have gone through.
The example above relates to structural injustices related to class. There are also others related to culture, language and religion. At the UCL Civic Design course, 4 in which students collaborate with a community group in London to co-produce a piece of work, students collaborated with various campaigners and community groups in Church Street, City of Westminster. The area is about to go through a major regeneration scheme, which will involve the demolition of many social housing and redevelopment. The area has a large population that do not have English as a first language and who do not speak English. When the planning application was in the period of consultation, the local authority distributed brochures with information about the development. While the brochures had a brief note (in English, Arabic and Bengali) at the end saying that a translation can be arranged on request, this made it very difficult for someone who did not speak English to engage. In general, all the publicity was in English. During the Civic Design course, students explored how to engage with local populations that had not been included in the consultation process. In collaboration with the Marylebone Bangladeshi Society, students organized a workshop in their building, in which simultaneous translation in Bengali was arranged, where students discussed with people from the Bangladeshi community what their priorities for regeneration were and how they would like to be involved. 5 During the workshop, students interacted with residents who had never heard about the regeneration scheme that was about to take place in their area, which is an indicator of the lack of engagement with non-English-speaking communities.
In addition to class, ethnicity, religion, and language, it is important to pay attention to other aspects related to identity, equality, and diversity. In England, the Equality Act 2010 includes nine protected characteristics, which are disability, gender reassignment, marriage and civil partnership, pregnancy and maternity, race, religion or belief, sex, and sexual orientation. In addition to these legally protected characteristics, it is essential to think about which other issues can exclude certain groups or create imbalances in power. These can be issues such as caring responsibilities.
In each co-design process, it is crucial to start understanding the history of the place, the different struggles that have taken place, and the local cultures and populations of the area.

Collective benefits
The sixth Principle for Fair Consultation says that the degree of fairness should be higher if a scheme deprives communities of existing benefits. When taking this principle from consultation to co-design, it does not make much sense, as it is hard to imagine that communities would co-produce a scheme that effectively deprives them of existing benefits. A scheme that is genuinely co-designed brings collective benefit -it improves existing facilities rather than eliminating them.
In the case studies of Alton Estate and William Dunbar and William Saville Houses in South Kilburn Estate, there are three main types of benefits that communities are deprived of during regeneration processes: home, social infrastructure, and green space. This is common to most estate regeneration schemes that lack genuine co-design.
In terms of housing, council tenants are usually rehoused within the local authority, and schemes tend to promise that they will retain the same type of tenancy. This is not always 100% true, as in some cases tenants are transferred into a housing association and this leads to higher overall monthly costs of rent and services. Nevertheless, in addition to this, residents are dispossessed of their homes and detached from their network of support. As the social impact assessment for William Dunbar and William Saville Houses demonstrates (Colombo et al. 2021), demolishing the buildings and rehousing residents would break existing links and networks of support among residents, which can have a negative impact on them. In some cases, like is the case in South Kilburn Estate, residents are rehoused within the estate through a phased demolition. However, in other cases, where local authorities claim that residents will be given a chance to return, this does not normally happen as residents are moved out from the estate for a long period of time. For leaseholders -in many cases, ex-council tenants that bought their homes through the Right to Buy 6 -the situation gets even more complicated. They are offered either a compensation that would not cover a new home in the area, which effectively displaces them very far away (London Tenants Federation et al. 2014), or a percentage of ownership in a home within the new scheme, known as 'shared equity', through which they no longer own the entirety of their home, but share the ownership with a local authority or housing association. Klinenberg (2018, 5) defines social infrastructure as 'the physical places and organizations that shape the way people interact'. He also explains that communities with more robust social infrastructure are more resilient to challenges such as climate change. Housing estates experience a loss of social infrastructure through processes of disinvestment/abandonment and processes of regeneration. Alton Estate residents have seen how their community facilities are closed down over the last few years. In the case of South Kilburn, the scheme is dramatically increasing the density of the area while not providing new community spaces and decreasing the existing ones. 7 This loss of space is effectively a loss of social infrastructure, which negatively impacts communities. The situation is very similar with green spaces. Redevelopment schemes result in a substantial loss of green spaces, transforming cities into spaces with little spaces to gather outdoors and with nonpermeable surfaces that are more vulnerable to flooding.
Alternately, when co-producing proposals with residents, those facilitating the co-design process can understand the connections between people and place -how people relate to their homes, their neighbours, their housing block, their community spaces, and their green spaces. When working with residents from both the Alton Estate and William Dunbar and William Saville houses on co-producing alternative proposals with their neighbourhoods, the first co-design workshops were about understanding what people value about the place they live in, what is their attachment to their home, how they relate to their neighbours, where is their social network of support, where do they gather, their daily life, and their use of public space. From the evidence co-produced with residents during these workshops -which used different techniques to create empathy and exchange knowledge -and through a survey, the researchers created a Social Impact Assessment (Colombo et al. 2021). This Social Impact Assessment explored the principles developed by Lipietz et al. (2018) that a Social Impact Assessment should be co-produced with residents and explore alternative approaches to regeneration with them. It also used a Participatory Action Research approach, in which the subjects researched are treated as full partners (Fals-Borda 1995).
When starting from what people value, it is less likely to deprive communities of existing benefits. Instead, local authorities, planners, and architects can identify an area's emergent social infrastructure and support it to benefit even more people and help it thrive (Sendra and Sennett 2020).

Conclusions: a charter for co-design
Through exploring the case studies and examining how to expand the Principles of Fair Consultation from a legal to an ethical framework, this paper has argued how a co-design process can be run ethically. The paper has concentrated on defining what constitutes codesign so it can be incorporated into policy and practice. For outlining recommendations that contribute to both defining co-design and shaping policy and practice, this paper presents a Charter for Co-Design, which builds on the literature, evidence and experience presented in this paper. While the case studies used in this paper are from London and within the English planning system context, the principles outlined are sufficiently flexible to adapt to different contexts.
Charter for Co-Design: (1) Co-design needs to involve collective thinking: Co-design needs to involve activities that imply collective thinking rather than just individual inputs. This is one of the main differences from a consultation process, which considers people's preferences as individuals rather than as a collective. Thinking as a collective rather than as a set of individuals is defined in literature as 'collective intelligence' (Smith 1994;Malone and Bernstein 2015;Di Siena 2019;Sendra and Di Siena Forthcoming). Achieving moments of collective intelligence during the process requires facilitation skills that professionals running the processes should have.
(2) Design how partnerships are created: In Arnstein's (1969) Ladder of Participation, co-design is usually in the 'partnership' rung. To deliver a co-design process, it is essential to create partnerships with community groups, local interest groups, residents' associations, local businesses, third-sector organizations, local authorities, and other stakeholders. These partnerships enable communities to co-lead the process, blurring the line between the promoter of the process and the participant in the process, as Participatory Action Research does between the researcher and the subject being researched (Fals-Borda 1995). The process for creating these partnerships needs to be designed from the very beginning. Since the work of the professional facilitating the co-design process is time-limited, it is essential to set up community-based structures to lead and steer the process, so this can continue beyond that time-limited period. (3) Investigate and learn from existing social infrastructures: An essential part of designing the abovementioned partnerships is investigating the existing social infrastructures in the area -those spaces and organizations that enable gatherings, social interactions, and create networks of care (Klinenberg 2018). Including organizations and people using those spaces in the partnerships will ensure that social infrastructure is protected, that the co-design process can learn from those spaces and organizations that foster social interaction, and that the existing social infrastructure is supported so it can benefit even more people.
(4) Address power imbalances: when creating partnerships, when doing preliminary research, when running co-design events, when interacting with people, when coproducing evidence, and when co-creating proposals or making decisions, it is crucial to take into account historical injustices, structural inequalities, and power imbalances (Lee 2014(Lee , 2015(Lee , 2021CLHL 2022). To do so, it is essential to identify which communities are less likely to have a say in decision-making and make an extra effort to include those communities in the process. It is also essential to consider if decisions may result in excluding or bringing disadvantages to specific communities and avoid that by including them in the decision-making process. To achieve this, it is vital to be proactive and take the actions outlined below. (5) Provide skills to involve communities in decision-making: To address power imbalances, it is vital to provide the skills to empower people to participate in and colead the decision-making process. This means organizing activities where participants not just participate but also learn about planning and how to participate in decision-making effectively. In addition, the learning process should include providing communities with the information and skills to make informed decisions, considering social, cultural, economic, and environmental aspects when cocreating proposals for their neighbourhood. This does not need to be excessively technical but adapted to a language and a set of activities the public can understand. (6) Inclusive events and language: To include those communities less likely to be involved in decision-making in the process, professionals facilitating the codesign process, as well as the organization in charge of the development, need to make extra efforts. This involves providing different forms of participation, considering how diverse communities may feel comfortable participating, organizing events at different times of the day/week, considering those that might have caring or labour responsibilities, and targeting particular groups that are not likely to attend public events and organizing tailored events for them. In addition, co-design activities need to use an inclusive language adapted to the targeted population, which is connected to their daily experience of the built environment. In areas where there are migrant populations, it is important to consider organizing events with simultaneous translation and/or having people in the team that speak multiple languages. (7) Collective benefits: The co-design process must result in collective benefits for the communities affected. Through the 'collective intelligence' mentioned above, one can ensure that participants think as a collective rather than individuals. Furthermore, since the role of the facilitator is addressing power imbalances, it is their role to ensure that the result of the co-design process brings collective benefit and does not exclude nor disadvantage some of the communities. (8) Transparency and clarity: Explain clearly what the objectives of the co-design process are, which spaces the process is looking at, how the co-created proposals will be used, what they should expect from it -i.e., what will be built or improved, and when they should expect that to happen, providing a detailed timing of the project. When this is not clear, it can lead to disappointments, loss of trust, and participation fatigue.
(9) Timing and resources: Planning, organizing, and facilitating a co-design process with the abovementioned principles requires time and resources. Therefore, codesign needs to be built into the timing and budget of any project in the same way other requirements are. A genuine democratic process can only be achieved when the co-design process is done with sufficient time -and work from the team facilitating -so communities can get involved. (10) A process that starts before any decisions is made and goes beyond generating proposals: Co-design needs to be built into every stage of every project, although it will inevitably have different intensity at different stages. Communities should be involved from the formulation of the brief for the project. Once a concept design has been co-created, it is necessary to have the appropriate communitybased structures so the co-design process and partnerships can continue during construction and beyond when the project is built.

Funding
Both case studies were funded by the Knowledge Exchange and Innovation Fund. This award is funded by Research England's Higher Education Innovation Fund (HEIF), which is administered by UCL Innovation & Enterprise.

Ethics statement
The two case studies were given ethical approval by the UCL Ethics Committee. All participants are anonymized, and no one can be identified in the research.