Policy mobilities as informal processes: evidence from “creative city” policy-making in Gdańsk and Stockholm

ABSTRACT This paper emphasizes the role of informal and ad hoc processes in policy-mobility by analyzing “creative city” policy in two relatively marginalized and neglected urban contexts– Gdańsk (Poland) and Stockholm (Sweden). The paper extends studies of “creative city” policy to diversify understandings of policy mobilities as a “social condition” in which territorial and relational aspects are combined as cities “arrive at” mobile policy, to contribute to provincializing urban theory and comparative urban analyses. Extending recent literature emphasizing the role of formal “informational infrastructures” in understanding policy mobilities this paper develops new insights into: the ways in which informal and ad hoc processes co-exist with and are important for the operation of formal processes; what the learning process in cities actually looks like in different contexts; and the role of individuals in policy-making as a social condition. The Conclusion draws out the wider implications of these points for understanding policy mobilities.


Introduction
This paper develops current understandings of policy-mobilities by analyzing them in two quite different and relatively marginalized urban contexts -Gdańsk (Poland) and Stockholm (Sweden). Specifically, the paper draws upon recent analyses of creative cities in the Global South (Cohen, 2015;Kong, Gibson, Khoo, & Semple, 2006;Luckman, Gibson, & Lea, 2009;Nkula-Wenz, 2018;Söderström & Geertman, 2013) to develop an emphasis on the more informal and ad hoc nature of policy-mobilities, and the central role played by individuals in these processes. In doing so we therefore also develop recent calls to extend the range of actors considered (Baker, McCann, & Temenos, 2019;Temenos, Baker & Cook, 2019;Ward, 2018) in analyses of how cities change themselves to "arrive at" globallymobile policy (Robinson, 2015), in order to extend understanding of the embodied social labor which makes policy mobilities happen Ward, 2018).
To do this the analysis thus extends understandings of the role of relatively neglected informal practices and processes and how they mutually support the more studied formal parts of the "informational infrastructures" (Andersson & Cook, 2019;McCann, 2008McCann, , 2011 underpinning policy mobilities. In drawing attention to the diversity of ways in which cities "learn" from other places and demonstrating the centrality of people as mobile agents mobilizing policy themselves, the paper shows that the less formal aspects of policy-mobilities are not simply characteristics of cities in the Global South, but play an important role in a range of different contexts and are demanding of further analysis. The paper begins by reviewing literature to develop a focus on how the informal and individual practices of a wider range of actors play a role in how policy-mobilities operate. After outlining the method the paper then analyzes this in three ways: how informal and ad hoc processes are important alongside formal processes, co-existing and supporting them; what the learning process in cities actually looks like in different contexts; and the role of individuals in policy-making as a "social condition" (Peck & Theodore, 2015). The Conclusion draws out the wider implications for understanding policy mobilities.

The informal and the personal in policy mobilities
The international spread of "creative city" policymaking is a key example of Peck and Theodore's (2015, p. 223) identification of a new policymaking condition -"fast policy . . . characterized by the intensified and instantaneous connectivity of sites, channels, arenas, and nodes of policy development, evolution, and reproduction" (Evans, 2009;Prince, 2012Prince, , 2014Rindzevičiūtėa, Svensson, & Tomson, 2016). This "fast policy" is a "social condition", a diverse policy-mobility process imbued with power and personal relations which shape the resulting policy (Peck & Theodore, 2015). Understanding policy mobilities thus rests on conceptualizing policy as "a socially structured and discursively constituted space, marked by institutional heterogeneity and contending forces" (Peck & Theodore, 2015, p. xxiv). The challenge is thus to "elucidate the various interconnections . . . among people, policy, and places that make policy-making a social and political practice" (Temenos & McCann, 2013, p. 352), in which different actors organize how cities chose how to "arrive at" mobile policies (Robinson, 2015). Temenos and McCann (2013, p. 347) therefore argue for detailed empirical analyses of the contexts and practices of policy mobilization or, in Ward's (2018, p. 279) words, more emphasis on understanding "the systems of comparing, borrowing, exchanging, imitating, learning, reinterpreting and translating"key processes which shape how cities are rendered comparable and how policy-makers learn. Literature has focused on the formal nature of these processes, particularly formal networks (Temenos & McCann, 2013), conferences, study tours (or "policy tourism" -González, 2011), meetings with mobile policy consultants (Prince, 2014), and award ceremonies as key sites in "informational infrastructures" (Andersson & Cook, 2019;Cook & Ward, 2012;McCann, 2008McCann, , 2011. Andersson and Cook (2019) make the point that the ways in which policy-makers learn about other places and mobilize these ideas into forms of mobile policy often occurs in these informational infrastructures, spaces in which, often through formal educational experiences (visits, seminars, expert meetings etc.), particular policy imaginaries are shaped. Thus these informational infrastructures are important spaces in which policymakers learn particular ways to "frame and package knowledge about best policy practices, successful cities and cutting-edge ideas and then present that information to specific audiences" (McCann, 2008, p. 12).
However, though such sites are conceptualized as more formal spaces of learning and exchange, Temenos and McCann (2013, p. 346) also acknowledge their role as "sites of encounter, persuasion, and motivation", in which "different notions of expertise and understanding [are] performed, making them active forces in the transforming of the policy-making environments of which they were part as opposed to simply reflecting them" (Temenos, 2016, p. 127). This suggests that even such formal spaces can be sites of less formal exchanges and learning, opening up the need for more consideration of the informal and personal.
Here we follow literature which emphasizes the importance of understanding how informal processes are a vital component of how international organizations and networks actually function (and cf. Cohen, 2015;Söderström & Geertman, 2013). A range of literature has now emphasized that to understand international forms of organization requires an understanding of diverse processes of "informal governance", defined by Stone (2013, p. 133) as "a systematic influence of unwritten rules, shared expectations or norms within . . . organizations that substantially modify or substitute for formal . . . provisions." In this vein recent work on policy mobilities has come to emphasize the prosaic  and everyday (Craggs & Neate, 2017) actions of policymakers which can form "an aggregate of practices, norms and exchanges that take place beyond formal/state regulation", but which can intertwine with policymaking mechanisms within the state (Urinboyev, Polese, Svensson, Adams, & Kerikmäe, 2018, p. 54).
Recent research on the spread of notions of creative cities in the "Global South" has demonstrated the importance of ad hoc networks and "less-than-neoliberal planning practices" (Cohen, 2015;Söderström & Geertman, 2013;Temenos et al., 2019). Cohen's (2015, p. 35) study of Bandung in Indonesia, for example, highlights the importance of identifying: . . . the specific, ad hoc networks that create new topological spaces and bring sites into connection in a power-laden manner . . . research on policy mobilities has a blind spot for these networks because of its focus on the popular sites from which policies originate . . . more research is needed on how policies arrive in sites outside of main policy flows . . . the power to create and structure such ad hoc networks is an important addition to . . . the policy mobilities concept. McGuirk (2016, p. 93) points out that Peck and Theodore (2015) emphasize that "processes of policy making and normative pre-filtering are replete with indeterminacy and unpredictability," but that: Without addressing these 'how' questions of practice . . . we risk falling back on assuming the pathways through which the constitutive relations of mobile policy making have their effect [and] risk asserting the powers of effectiveness (if not powers of determination or omnipotence) of influential actors . . . without fully unpacking how that power is achieved in practice (McGuirk, 2016, p. 94).
As Robinson (2015, p. 833) argues, "policy makers compose their ideas in the midst of a myriad influences from elsewhere", resulting in an "often messy and unmappable complexity" (Robinson, 2011, p. 26), including "more ephemeral spaces of interaction and communication" (Robinson, 2013, p. 10). Söderström and Geertman (2013, p. 258) research on Hanoi (Vietnam), for example, leads them to argue that "policy in the making" is characterized by "loose threads", ie. "virtual policies suggested by a set of different connections to different elsewheres" comprised of a repertoire of connections between locales.
Focusing on the formal thus may over-emphasize the importance of short-term and official contexts in which learning takes place. To complement and extend this understanding, therefore, it is also necessary to examine less formal aspects of relationalities that co-exist and potentially mutually support (or hinder) formal relationships. This potentially could also lead us to temper the emphasis in the notion of "fast policy" on the speed, intensity and instantaneous nature of engagement between the territorial and the relational (Peck & Theodore, 2015;cf. Dzudzek & Lindner, 2015), opening up instead coexisting realms in which things occur at different paces and intensities, and people manoeuver the new policy context in more individually motivated ways.
In the spirit of Söderström and Geertman (2013, p. 258) "repertoire of connections", we follow Temenos et al. (2019, p. 106) in emphasizing the role of "embodied social labour and the material and discursive practices that shape and facilitate the circulation of policy", highlighting the performative aspects of mobile people, knowledge, materialities and politics (and see McCann, 2011;Robinson, 2015), or, as Craggs and Neate (2017, p. 47) put it, "the everyday embodied and social nature of policymaking". Here we seek to advance the emphasis on the importance of individuals in the "various acts of interpretation and translation as actors of varying geographical reach seek to arrive at a particular 'local' urban policy" (Ward, 2018, p. 277;Craggs & Neate, 2017). Actors in place decide what mobile policy to adopt and prepare places to "arrive at" mobile policy. Policy ideas are made mobile by "embodied members of epistemic, expert and practice communities" (Peck & Theodore, 2010, p. 170;Larner & Laurie, 2010). People are "infrastructure in the arriving and making up of policies", and In the case of the rendering of policies mobile, the emphasis on the embodied and performative nature of the work done brings centre stage the importance of how people communicate and interact, of the various objects, spaces, technologies and times that facilitate these various intersections. (Ward, 2018, p. 278).
This points to the need to extend the range of actors considered Jakob & van Heur, 2015;Temenos & McCann, 2013;Temenos et al., 2019;Temenos & Baker, 2015;Ward, 2018). Baker et al. (2019) suggest that policy mobilities have largely been analyzed as the outcome of the efforts of a relatively small class of elite actors, neglecting the roles played by non-elite state and non-elite non-state actors. While mayors have received attention, various bureaucratic actors should be considered and Kuus (2011), Larner andLaurie (2010), and Baker et al. (2019) are among the few studies considering middling technocrats whose understandings of what is going on is crucial to what happens on the ground. Apart from consultants, the role of various non-state actors is neglected, including cultural institutions, cultural intermediaries and independent cultural producers, both in situ and through their own mobilities as "mobile actors" . Recognizing that those who make policy accumulate expertise and knowledge over the course of their working life (Craggs & Neate, 2017;Larner & Laurie, 2010) also avoids the presentism trap in analysis.

Method
Recent literature has called for consideration of a greater range of cities, decentering the privileging of cities in the "Global North-West" as suitable for theory-generation (Ferenčuhová & Gentile, 2016;Gentile, 2018;Robinson, 2016aRobinson, , 2016b. Studying how cities in the Global South "arrive at" mobile creative city policy has been instrumental in opening up an understanding of the different processes at play (eg. Cohen, 2015;Nkula-Wenz, 2018;Söderström and Geertman, 2013). We address these issues through the study of how cities arrive at mobile "creative city" policy in two different and relatively marginalized urban contexts -Gdańsk and Stockholm.
Studies of creative city policy have considered an extended range of contexts, including the Global South. However, research lacunae still exist, and considerations of even capital cities in Scandinavia, and non-capital cities in post-socialist, new European Union (EU) accession countries represent understudied areas. In addition, very little literature on creative city policy-making has explicitly considered the relationships between the urban and supra-national scales, specifically with the EU. Stockholm and Gdańsk thus represent cities in these analytical categories. Both cities are also analytically relevant as they have developed extensive strategies and institutional infrastructures focused on culture and creativity. Stockholm is recognized in the Scandinavian context as a cultural center with one of the larger cultural and creative infrastructures, with a significant concentration of creative and knowledge-intensive industries. Gdańsk is a regional center in northern Poland. It has less of an international cultural profile, though it has some international festivals and relatively high-profile cultural institutions. However, Gdansk is regarded in Poland as an open and tolerant city with a lively cultural and creative scene. Both cities have, albeit in different waysand with a longer history in Stockholmresponded to the general trend in urban development to push cultural and creativity policy agendas, including various ways of engaging internationally in this field. Stockholm has a longer tradition of incorporating culture into policy, while for Gdańsk it dates from the end of the 1990s and was intensified during the EU-accession process and then membership in 2004. Gdańsk has a more intense relationship with the EU than Stockholm.
The choice of cities therefore follows recent arguments in comparative urbanism by adopting an approach which involves "starting from anywhere" (Robinson, 2016b) to consider what cities in different contexts can say about policy mobilities (cf. Cohen, 2015;Söderström & Geertman, 2013). The aim is not to "territorially" compare the cities (with each other or other cities), but to draw upon Robinson's (2016a) notion of the "comparative imagination" when "thinking with elsewhere" to analyze "the informality of governance arrangements . . . and their external partnerships" (2016a, p. 194) with the aim of "building comparisons through putting case studies into wider [conceptual] conversations" (2016a, p. 195). This follows a similar approach to bringing disparate cities into comparative focus developed by Tuvikene, Neves Alves, and Hilbrandt (2017) in order to develop understanding of common processes. The focus of the comparative analysis is thus people's embodied performance in relation to the different forms of mobile ideas and policy that they engage with.
The empirical underpinning of the paper is based on 30 semi-structured expert interviews in 2016-18 with a range of actors in the fields of urban and cultural policy (14 in Stockholm, 16 in Gdansk). The interviews were mainly conducted in English, or carried out in Swedish (by author Borén) or in Polish (through a translator) and subsequently transcribed into English. The sample was derivedfollowing McCann and Ward (2012)as a way to "study through" the creative policy scenes of each city.
While key policy-makers and influencers in the respective city councils formed the core of the interviewsparticularly higher officials (CEOs and vice-mayors) and their cultural departmentsthe sample was extended to include other institutions and both elite and non-elite actors  which play a key role in culture and creativity, such as regional authorities, key cultural intermediaries of different kinds who are highly connected with the city authorities, and NGOs and individual cultural producers. The idea was to sample the key institutions and individuals making up the local "policy ecology" responsible for shaping and influencing (directly and indirectly) policy-making.
Understanding the social condition of policy mobilities: formal and informal processes, learning and people as mobile agents In order to develop these points further, in this analysis section we explore policy-making as a social practice (Temenos & McCann, 2013) by addressing three relatively neglected but interlocking aspects of policy mobilities: the ways in which informal and ad hoc processes are important alongside formal processes, co-existing and supporting them; what the learning process in cities actually looks like in different contexts and how this is changing; and the role of individuals in policy-making as a "social condition" (Peck & Theodore, 2015, p. 1).

1) Temporality, formal and informal processes and the long-term processes of policy-making
Engagement with formal "informational infrastructures" (Andersson & Cook, 2019) has played an important role in policy mobilities in both cities. Parts of the urban administration in Stockholm have been members of international networks like Eurocities from the mid-1990s. As one example, a member of the Culture Administration in Stockholm talked about how their membership of the World Cities Culture Forum has shaped how they observe and learn from other cities internationally: . . . every year they issue a culture report from around 40 cities and there you can see statistical figures . . . how much money they put on different things, and how many visitors there are, and so, yeah, you can measure and compare . . . you can read data about the cities, you can read about cultural strategies for world cities . . . And we thought it was interesting to see how are they workingwhat results do they have, what are their strategies?
In Gdańsk, this kind of interaction with formal networks developed later, as one cultural office within the city administration described: I suppose the city really started thinking about it when they started (2006) to apply for the title of the European Capital of Culture 2016. As a city we lost . . . the title, but the whole process didn't stop, and the City Culture Institute in City Hall was formed . . . And during that process, of course, they establish many links with other European cities. And then I suppose it was the moment when they entered this Creative Cities Network as well. So, it all added upon each other and of course many of the employees started to go to many seminars and conferences . . .
In the Gdańsk case, interaction with EU programmes and other European organizations has been an important example of engaging with formal "informational infrastructures", sometimes involving a very direct form of learning. As one Vice-Mayor put it: Actually we're copying. We go somewhere . . . I think this is the biggest part of our . . . of all of the innovations. You go somewhere, and you copy. Of course, you adapt some solutions . . . this is very important. There are no inventions nowadays like starting something from the beginning. It is always that you look on someone, if somewhere something works, and then you try to adapt it to the city.
In Gdańsk, given its relatively recent membership of the EU and the importance of funding from there, Europe plays an important role in these processes as one independent cultural institution observed: Involvement in formal networks and events is thus important for both cities, and adds emphasis to the importance of studying these kinds of formal learning spaces which make up "informational infrastructures" (Andersson & Cook, 2019).
However, what also emerged from both cities was a sense of the temporality of these engagements with more formal contexts. This raises a question about the sustainability of such contacts in policy-mobilities and emphasizes their often ephemeral character and the difficulties in tracing what effects they actually have on policy-making. As a representative of the Culture Administration in Stockholm noted "We used to be involved in Eurocities but now our focus is more put on another global organisation, the World Cities Culture Forum." Thus formal associations can shift and change, and it is interesting here that in this case there is a progression of engagement, from a formal organization at a European scale to one at a global scale, which implies a development of the scale and reach of co-operation.
However, at the same time both cities provide examples of co-operation which did not last. The temporalitythe short-term nature or even "failure" (Lovell, 2019;cf. McLean & Borén, 2015;Stein, Michel, Glasze, & Pütz, 2017)of such policy-mobilities is not often discussed in the literature. Informational infrastructures, with their focus on "best practice" and "expert knowledge", often do little to encourage discussion among policymakers of things that didn't work, and it is important to explore how these discourses circulate in less formal channels. The organization Stockholm Business Region, for example, talked about the experience of being part of the Baltic Metropoles Network, a forum for capitals and large metropolitan cities around the Baltic focusing on innovation and competitiveness: . . . they had a broad agendasustainability, creativity and . . . innovation, and we did a lot of projects and events and I was involved in the creative programme. And we co-operated with Riga and some other cities, I think all eleven were active . . . We did a benchmark and it was exciting, but you know, after the project, nobody would carry it around, it just sinks and nobody remembers.
Similarly, a cultural office in the Gdańsk city administration described their experience with EU-funded projects around creative cities: No, no, I am so sorry to say that, but I mostly think that they are wrongly managed, poorly co-ordinated. The money's just going between your fingers. If I can see that amount of money and I could use it for something that we don't have money to do . . . I want to cry alone, really. It's poorly managed and totally useless. Those projects are meant to create some tools, instruments that could be somehow replicated, but really, they don't have that outcome.
This points to both the possibility of the failure of, and therefore a temporality of, international connections and learning from globally-mobile ideas, further developing Robinson's (2013, p. 10) argument about the need to focus on "more ephemeral spaces of interaction and communication". Such interaction can drop off, be limited to the lifespan of specific projects or funding opportunities, and specific connections and networks can cease to function. Some long-term development of knowledge and practices may result, and longer-term capacity building developed, but such a process can be patchy and perhaps even incoherent. It suggests that the notion of "informational infrastructure" needs to be more nuanced and recognize that learning in formal spaces is an unpredictable process, the longer-term effects of which are difficult to trace. This selectivity and temporality of learning from formal international connections is well illustrated in Stockholm: Well, you shouldn't exaggerate the importance of Richard Florida [an American academic and consultant famous for the world-wide promotion of his 'creative class' thesis], it's just trend and fashion and following the crowd. Before, looking at other cities and international cooperation was more visible. We were in touch with leading cities. We were really studying the efforts by the UK government into creative industries . . . And tried to follow. Now, I don't know where we go and how we see and how we value the international co-operation. I should say that we are more inward, more happy, more satisfied with Stockholm being the "best in the Nordic class", and not so much aware any longer about the competition . . . I've been following it over 20 years, so I see that we were the best at a certain time, and we are not the best one any longer. We are in competition with many other cities performing at least equally . . . I think that we should win on being in more lively relationships with some of the best performers, but I can't say that we have such co-operation.
This suggests that the role of some international co-operation and learning has decreased over time, and that globally-mobile ideas are having less of an influence, partly due to a certain satisfaction with homegrown policy and partly to a decline in practices which involve looking internationally. Here, as McGuirk (2016, p. 94) argues, focusing on the "'how' questions of practice" opens up understandings of policy-making which go beyond solely focusing on its mobile aspects.
The nature of international links has also changed, with city administrations acting in a more self-organized and less formal fashion. For example, the office responsible for producing the Stockholm Regional Development Plan (the Tillväxt-och regionplaneförvaltningen, Stockholms läns landsting) talked about networks which are: . . . more self-organised. I was in a meeting last week in Oslo and that was totally selforganised, because they have a regional plan that was finished about a year ago and they wanted to talk with us about how we work with ours. They invited us to come to a meeting they had with their municipalities in Oslo, and also to talk about how they should work with following up their regional plan . . . I'd say it's very much of an exchange. This is thus less about membership of formal networks than reciprocal relationships arranged on a more personal and less formal basis. They can be one-off meetings around a specific set of goals, rather than project outcomes, and a less formal kind of "exchange" or co-learning by comparing experiences, rather than a mobility of knowledge or "best practice".
Expanding the range of actors in the analysis further opens up the importance of this point. In Gdańsk, the NGO sector is becoming increasingly important as carriers of mobile policy into and out of the city, as one NGO involved in urban development described: We attended a huge "learning conference" in Oslo. It was very interesting for us . . . in Norway they have very well-developed NGOs and process of co-operation between local governments and NGOs. Because they funded these [Norwegian Grants] they were inviting us to tell them what problems we had and so on, and they were learning from us. And I think for them it's very, very good because they have a huge "Bank of Ideas", and then they can use them.
Again this reinforces a picture of temporal change in a co-existence of formal programmes and funding schemes with more informal informational infrastructures which mutually interact to support processes of co-learning. This was evident in the process of self-organization in the Gdańsk NGO sector, which developed into cooperation across the Tri-City region and with Europe, as another NGO working on improving urban public space reported: we are in the middle of organizing an event . . . it is not about working with NGOs from different countries but NGOs here, within the Metropolitan Area. It will be an opportunity of NGOs to discuss issues connected with their cities and about the development of the whole metropolitan area and I think it is the first such kind of event. And in my NGO this year we will be able to fill the application to Visegrad Funds and to Europe For Citizens grant, and it is obligatory there to have partners from different cities. So, as I said, previously we didn't have an opportunity to co-operate with different countries, but this year I think one ways of our institutional development is to start co-operating also with NGOs with different countries.
So what is significant here is that the range of actors involved in policy mobilities is expanding Ward, 2018), and the analysis here points to the need to do more to study non-state actors and the growth in self-organizing strategies. The activities of NGOs in Gdańsk and the wider region evidence their role as local actors, but also as learners from international contacts, and agents in and of themselves who mobilize ideas into and out of the city-region. They are also increasingly acting as a nonstate sector which will link the city into international forms of co-operation, acting as mobile agents who will carry ideas and put them into practice. Since the city administration interacts with NGOs in Gdańsk this will also influence policy-making and practice.
Overall this suggests that the formal events and spaces that are important parts of the "informational infrastructures" that support policy mobilities have beenand continue to beimportant in both cities. However, this analysis also reinforces Temenos and McCann's (2013, p. 352) argument about avoiding the trap of presentism by studying "the histories, presents, and outcomes of policy implementation [by avoiding] a narrow focus on current successful policies, without regard for what has come before, for what was perhaps unsuccessful, or for alternative policy narratives." In Gdańsk, such formal engagement has been an important part of the EU-accession process and EU-membership from 2004. EU membership has been a major influence on the adoption of culture and creativity in the city's urban policy, particularly since Gdańsk has attracted the highest rate of EU cultural (and overall) funding per capita in Poland. Stockholm has been involved in some EU and/or European projects, but the EU has been much less significant there.
However, in both cities these formal engagements with mobile policy wax and wane, are ephemeral, and sometimes fail to lead to longer-term learning or activities. Engagement with the formal aspects of "informational infrastructures" thus varies over time and in its intensity and impact. Furthermore, that picture is further complicated by various self-organized and/or informal types of contact, engagement and co-operation, something which has received relatively little attention in the policy-mobilities literature. A range of different actors are also important. These further complicate the types of learning which takes place as is developed further in the next sections.

2) What does the learning process in cities actually look like in different contexts and how is this changing?
This complex co-existence of formal, self-organized and informal relationships between a range of actors raises the question of what kind of learning actually takes place within that complexity. While some of the examples above demonstrate that, more so in the case of Gdańsk, there is some degree of direct learning or even "copying" from other cities, even in those cases the idea of learning from existing policy-models and practice was tempered by a realization that direct copying is impossible, and ideas have to be adapted to the locality. So, as one cultural intermediary in Gdańsk discussed: Of course, we follow, we read. So, while developing our own strategies, we also study other studies, reports . . . So, we adopt . . . we studied other research studies . . . we refer to them, programs what have been already achieved, 'cause it would be silly not to use in your own work something that have already been done with success.
While existing literature has explored the ways in which in recent years "the global has become more knowable by placing the experiences and performances of others into quantitatively and qualitatively encoded proximity" (Larner & Le Heron, 2002, p. 417) we suggest a shift in practice in which "learning from other cities" has developed into a more sophisticated process. In part this is about various forms of benchmarking. For example, the office responsible for producing the Stockholm Regional Development Plan stated that they use the UN Sustainable Development Goals, which include culture as an aspect of sustainability, to benchmark, but not to simply copy what is in the Goals but to check if they are missing anything significant in their strategy. However, what is evident is that relatively less formal and self-organized activities and relationships are important in these benchmarking activities which operate more through personal contacts, as different parts of Stockholm's urban administration discuss: we did a peer review study in this Autumn about the Regional Development Plan and we invited Oslo and Helsinki and Amsterdam and Vienna to talk about different kinds of regional plans . . . we are benchmarking with other cities on a yearly basis to check on how we are doing compared to other cities. (Office for Stockholm Regional Development Plan) Yeah, I think it's interesting to have this knowledge about what's happening internationally because you can check "are we very strange?" or "are we mainstream?" and so . . . it could be a reason to think over "are we doing the right thing?" (Culture Administration) Thus these kinds of encounters with globally-mobile policy ideas are no longer about learning and copying, or using "off-the-shelf" models, or more quantitative methods of international benchmarking, but instead are organized around more embedded (often self-organized) relationships with other urban authorities involving reflection and peerand self-evaluation. "Placing the experiences and performances of others into . . . proximity" (Larner & Le Heron, 2002, p. 417) has developed from a process of "learning from" to one of more mutually-supportive, co-learning and co-production of knowledge. For example, the office responsible for producing the Stockholm Regional Development Plan stated that membership of formal networks is valuable to: . . . learn from other cities how they are working, looking at what they're doing . . . get their perspective on how things work . . . We will also have a meeting in Stockholm because we are part of METREX [Network of European Metropolitan Regions and Areas] . . . we will host the spring conference this year, and then we will have this peer review as one part of the conference. So they will come back and reflect also on what issues we have in the regional development plan . . .
Here the idea of peer reviewbased on hosting and collaborating with practitioners from other citiesis important. Rather than sharing "best practice" models the process is one of mutual reflection on how policies and strategies work. This less formal process, with less tangible outcomes, was also reported in Gdańsk, where an officer in the cultural administration described how membership of formal networks worked through much less formal ways of learning: [We host] the Secretariate of the Union of the Baltic Cities which has a very active Cultural Commission . . . we co-operate with Umeå from Sweden . . . with some Finnish cities, with Bremen [in Germany] . . . vital co-operation with the Baltic Sea corporations. I think it is very important for us, as well as for our civil servants working here in the City Hall, because we can "put somebody on the chair sitting next to the desk" . . . And what are the results? Cooperation. It's important to note that maybe not even some visible results, we did not build something or we hadn't some huge successful artistic endeavor, but creating these connections is important for later co-operation for people with institutions . . . there are some nontangible things . . . ideas. Idea of tolerance, of identitynot as "my little field here" but more like "I am European". It is also important [for developing] a new social consciousness.
So, here, it is important that what is mobile is not so much policy directly but broader ways of thinking, which are particularly important in the post-socialist context in creating an attitude and environment beneficial to new (in that context) forms of policymaking. As one Gdańsk vice-mayor explained: You cannot copy everything because there is different law, actually different culture of the people . . . but I think the culture of Gdańsk's citizens changed as well. At the beginning of the 1990s people didn't believe that they can do something together, and people didn't care about the public stuff, public infrastructure. Right now, people . . . firstly, they believe in themselves, but on the other hand they take care about what happens around them. So, these are very important changes in culture.
Here, then, within formal informational infrastructures the learning taking place is sometimes less tangible but has longer-term implications for policy-making. The value comes from the possibility that someone from elsewhere "sitting next to your desk" could impart something useful, and, importantly, this does not have to be a policy or a programme or event, but can be ideas about the city and its identity or an attitude toward what is possible, which in turn can shape policy (and in the case of Gdańsk the development of an identity based on ideas of "Europeanness", openness to difference, tolerance and solidarity are very strongly shared and linked to and articulated in policy). Active membership of formal networks and staff exchanges provides the chance to observe and learn from what other cities are doing, but also to engage in a pro-active way with other partner cities and key mobile personnel in forms of mutual peer review or intangible exchanges to co-constitute policy.
The emphasis on a process of exchange as a form of mobility of people and ideas is interesting to observe in Stockholm's engagement with Oslo: . . . and actually the whole reason we did this peer review study was that we had been invited to come to Oslo and talk about their regional plan when they . . . did their first regional plans. They invited us to discuss how we had worked ours. But then we got the idea of "this is a very good way to exchange knowledge", so that's why we invited them back . . . (Office for Stockholm Regional Development Plan) This interaction took place as part of the office's formal role, but operated much more at the personal level and was based on personal contacts and then thinking about the value of using peer review, mutual reflection and co-operation as the basis of policy development This idea of the importance of the role of informal contacts within formal networks in shaping ideas was expressed very strongly by the Culture Administration in Stockholm: . . . in the World Cities Culture Forum and in Eurocities maybe the most important part of it is the informal contacts you make. That depends very much on yourself . . . We had a very concrete gain from this co-operation because I contacted the representatives from Amsterdam, Barcelona, Copenhagen, Oslo and Helsinki . . . and asked them if they are interested to make a comparable statistical report. I think there were 5 or 6 questionslibraries, funding, tourism, film and one or two cultural areashow many visitors do we have, how much money do we give away? So that was a very practical resultwe could compare with each other . . . These kinds of networks operate in different ways and offer the scope to be more oriented to the nature of localities, rather than looking at standardized models. As the Culture Administration noted: There is a longer plan that they would like to make more research about city development issues. So it's more practical than the Eurocities meetings, because you could go anywhere in Europe to these Eurocities meetings, and sit and listen and go back again, and you had no real mission, no homework to do for the next meeting. But [in the World Cities Culture Forum] you have also always homework to do, and sometimes I feel it's too much, but I think it's good to have it because then you learn about your own city as well.
What is interesting here is that engagement with a formal network shapes policy not only because of mobile ideas in a formal informational infrastructure, but because it prompts key individuals to be actively learning and developing ideas outside of (but connected to) that formal setting. The comparison with "school homework" is an interesting one, because it suggests something extra that should be done which the participant is a bit reluctant to take on board, but they see the longer-term benefits in doing it and thus take personal responsibility for it, and learn something of value for their own city by doing it. The learning taking place here is not so much about "learning from elsewhere" but is about being inspired by the demands of working in a formal international network to do your own work on researching and reflecting on your own city.
The nature of the encounter with globally-circulating ideas and what is learnt from it, and how, has thus changed over time and is a complex mix of formal and informal, tangible and intangible, further demonstrating the need to focus on the informal aspects of policy mobilities which has become apparent in studies of other contexts (particularly in the Global South, see Cohen, 2015;Söderström & Geertman, 2013). International contacts are in a constant state of flux. There is no one model, but there seems to be an overall shift from "looking at what other cities are doing" to forms of co-working, peerreviewing, knowledge-sharing, mutual reflection and co-operation as ways of learning, in combination with and independently of formal informational infrastructures. The analysis above highlights that how this turns out in practice relies to a great extent on the actions, personal motivation and initiative of key individuals, an aspect which is developed in the next section.

3) "Mobile actors" as key parts of policy mobilities
A key point emerging from this analysis is the role of individuals in these complex forms of mobilizing policy and preparing places to "arrive at" globally-mobile policy (cf. Temenos et al., 2019;Ward, 2018). This involves a range of actors, and some policy developments would not have occurred without this personal motivation and initiative, or would have been shaped in a different way. This analysis section further considers other ways in which individuals are a significant part of how policy mobility actually operates.
Research has demonstrated the importance of individuals within urban administrations in driving a particular focus in urban development, particularly influential mayors (eg. Borraz & John, 2004;Jayne, 2012;McNeill, 2001). This has been an important factor in the case of Gdańsk, where the late mayor Paweł Adamowicz was highly influential in making culture and creativity central to urban development in the city. However, rather less attention has been paid to the role of "middling bureaucrats" Kuus, 2011;Larner & Laurie, 2010). Actors at different levels within the city administration have demonstrated considerable initiative in engaging with mobile policy. This is well illustrated by this description of their everyday activities around culture in urban policy by a member of the urban administration in Gdańsk: For me personally, I read a lot. I try to keep up with what's going on at the national level in the cultural field, what's going on abroad . . . We discuss a lot. We send each other things that we find interesting. We try to meet with people, who are doing interesting things, with organizations that are maybe somehow similar to us but doing things differently. So, we did it more on a personal level I would say . . . Yes, I would say it's mostly doing from the personal interest.
This, and the analysis sections above, make the point that beyond political leaders there is a need to also examine the role of people at different levels of the urban administration, who may be displaying considerable initiative and developing their own styles of learning in ways which combine formal and informal contexts and sources . In Stockholm, a now retired leading officer in the culture administration related that: I would say that it is on a very individual level. I mean, I tried to take in literature and ideas as much as I could, and tried to write about them. I made a lot of papers about it for the politicians and my colleagues. We could talk about it at coffee, in the corridors, not so structured. So we learned only on an individual level and being colleagues and being very active out in the real lives, so to speak, I mean at the culture programmes. We went to a lot of theatres, we talked to them, and we of course we had a very, very active dialogue with all our recipients of support [theatres, artists, etc.].
Sharing things that you have come across with colleaguesboth in your own city and in otherscan be an important way of learning and then shaping policy, but it can also be unsystematic and sometimes involves rather chance encounters with knowledge that then becomes "expert knowledge" in various ways, emphasizing the importance of trying to grasp the less easily traceable ways in which policy learning operates (Robinson, 2011(Robinson, , 2015. And importantly, while personnel within the urban administration do engage in the kind of "policy tourism" which has been identified in the literature (González, 2011) by attending international events or visiting other cities, significant elements of the "learning exchanges" actually occur during the more social and informal occasions at dinners and in bars organized in connection with the formal program (see Andér, 2016) further interweaving the informal and the formal.
Furthermore, from studying these two cities it is clear that a broader range of types of actors must be considered. As one NGO in Gdańsk commented when reflecting on the importance of mobility: I think the most important factor during the time that we are part of the EU . . . yes, we are number one using European funds per capita . . . But on the other hand for me it's more important that we can make more and more co-operation with different countries. So now we can discuss, not with the people who are not prepared to these discussions, but now we can discuss with the people who are travelling around the world, travelling around Europe, who knows something about how different cities can develop. So, from the cultural point of view it is also very important that they were somewhere, and they have some friends from the different parts of the world, and this is very important in our present debate. This stresses the significance of a wider range of people as "mobile actors" or carriers of policy, including personnel within urban administrations but also NGOs and individual cultural producers who spend some of their lives abroad and mobilize ideas through their own travels and initiatives . As one participant in an NGO focusing on the revitalization of a derelict area of Gdańsk put it: It is a typical way of culture movement in Poland, that the main big ideas are being transferred from the West. Whoever goes for a trip to Berlin and another countries and saw a cool thing there. And then he comes to Poland and do it . . . But . . . the process is interesting that he does this in a creative way. He does not only copy it, he makes it fit with the Polish culture, the city preferences and differences. I would say that the main spread are people who visited those spaces, they become inspired by something that works there. And they think like "I can do it here". This complex intersection of people's life-courses and experiences, personal mobilities and entrepreneurship, is evidenced by one cultural entrepreneur who opened a cultural space based on "container architecture" in Gdańsk and who has subsequently interacted with the city council: I used to live in Amsterdam . . . Barcelona . . . Oslo. My wife used to live in London. So, this was an obvious influence, the places that we saw that we were living in. Amsterdam had a huge underground alternative culture . . . the whole squatting scene, a lot of the warehouses that were being reclaimed by artists and so on. So, that was a big influence, but we actually made kind of a field trip to check the places. London was one of big influences, Pub Brixton especially, as a venue that we feel we were almost directly inspired by. We watch very closely what they do, and we are not ashamed to admit that. We actually tried to stay in touch with them. Another inspiration was Berlin Tempelhof Airportanother space that was reclaimed for agriculture and culture. There are very similar spaces in Lisbon, also based on the shipping containers . . . there are these kind of places in the US as well, we try to watch them and . . . steal what's good and leave out what we don't like . . .
In the case of Gdańsk there are still enough derelict spaces in need of revitalization that such initiatives can take root and flourish. They are not lead by or funded by the city administration but, in several cases, they subsequently become supported by the urban authorities, who observe and in some cases support such initiatives, and who are seen to create an atmosphere in which such initiatives are possible (further supporting Baker et al.'s (2019) arguments about non-elite and non-state actors co-operating and converging with state actors). Thus, considering NGOs and individual private cultural producers and entrepreneurs who actively carry and adapt forms of culturally-led urban redevelopment strategies which change the city fabric and influences urban policy formation and practice is an important way to extend analysis of policy-mobilities (cf. Baker et al., 2019). These initiatives often lie outside of, but interact with and mutually constitute, formal infrastructures of policy mobility and learning, but play a significant role in the development of policy (cf. Cohen, 2015;Söderström & Geertman, 2013).

Conclusion
This analysis has focused on two cities located in relatively marginalized contexts in terms of studying creative city policy and generating urban theory. Adopting a "starting from anywhere" (Robinson, 2016b) approach and interviewing a wider range of key actors has allowed a number of processes to emerge which develop understanding of policy mobilities.
What emerges firstly from the analysis of these contexts is the importance of the personal in policy mobilities. The embodied social labor which makes policy mobilities happen is underpinned in some important ways by mobile people (developing the focus suggested by Temenos et al., 2019;Ward, 2018), and as Prince (2012, p. 328) suggests, "By not considering how actors take shape as policy mobilizers, we risk their becoming just functionaries in relation to various topologies." Changing personal connections, both formal and informal, are central to these processes, and allow for both mobilities and immobilities.
Key actors are literally moving ideas and policy around with them as they move, but the analysis here shows that this can involve both state and non-state actors. Another important point about analyzing the role of people is thus to extend the range of types of individuals. Clearly local state actors remain important, but in some contexts a range of elite and non-elite, quasi-or non-state actors, cultural intermediaries and even individual cultural producers can be significant in shaping or directly influencing policy-making as their personal experiences and life trajectories bring ideas and initiatives into cities which are sometimes picked up on by policy-makers (cf. Baker et al., 2019).
Learning from elsewhere also involves learning about yourself, both in the direct sense of "homework" done on your own city in the context of formal informational infrastructures, but also in a more transferred senseby looking at others the familiar "home city" is juxtaposed, contrasted and eventually also "de-familiarized", thus opening up new imaginations and other routes of change. Thus this can be less about the mobility of policy and more about the importance of self-reflection, a point that was made in the various accounts of benchmarking against other cities and mutual engagement with peers. These processes may occur within the framework of formal governance structures within the city, but they are often the result of personal interactions, invites, meetings and less formal activities, such as reflection and discussion.
Here, then, we would seek to at least temper the characterizing of the "new policymaking condition" as involving "intensified and instantaneous connectivity" (Peck & Theodore, 2015, p. 223) or "intense mutual engagement" between relational and territorial aspects (Dzudzek & Lindner, 2015, p. 391). While there might be specific times in the policy-making histories of cities where this is true, this may rather underestimate the diverse spatialities and temporalities of policy-making that emphasizing a focus on the individual provokes. Many of the respondents discussed the personal and ad hoc nature of international contacts, how they developed and changed over time, how they were ephemeral and difficult to sustain and sometimes fizzled out or ended in "failure". Their accounts actually present a complex picture in which personal initiative plays a key role in determining the intensity of engagement with the mobile.
The significance of recognizing individual agency also opens up the perspective that "arriving at" the mobile can be a rather less organized and certainly less formal process than is often portrayed, for example by perspectives which emphasize more formal "informational infrastructures" (Andersson & Cook, 2019). Policy-making is actually increasingly being "stretched over" multiple, ephemeral, formal and informal/ad hoc networks and relationships and loose connections of differing intensities. Relationships are sometimes sustained and fix a particular iteration of policy-making between places, but they can also develop into connections with other places, or fail. Alongside formal interactions, some things, which can be quite significant, simply would not happen if individuals did not use their initiative and their personal contacts play a key role. This points to a need for policy-mobilities research to focus further on people as individuals maneuvering this new social policy-making condition in a range of global contexts, including the Global South, various parts of Europe, the "Global East" and articulate these experiences back to urban studies and urban theory.
Thus by "addressing these 'how' questions of practice" (McGuirk, 2016, p. 94) this analysis points strongly to the need to also pay attention to the operation of informal and self-organized processes and how they not only co-exist with, but are mutuallyconstitutive and reinforcing of, the formal as a part of the new policy condition. Policymakers and policy-making are affected by the norms inherent in formal "informational infrastructures", but this understanding needs to be extended to include a wide range of forms of informal international co-working and co-creation. While studies have shown that this is important in cities in the Global South (Cohen, 2015;Söderström & Geertman, 2013) we argue that this is not something which is simply a characteristic of those cities, but a set of processes that require consideration in a full range of contexts. Future research should address these issues in a much broader range of contexts, considering the different temporalities and spatialities involved and their differing scales of reach and intensity.

Disclosure Statement
No potential conflict of interest was reported by the authors.