How Do Local Policy Makers Learn about Climate Change Adaptation Policies? Examining Study Visits as an Instrument of Policy Learning in the European Union

Abstract

Study visits are an underresearched phenomenon, particularly in the field of climate change adaptation. Drawing on interviews with key local stakeholders, this article investigates study visits organized by European municipal climate networks. The results of this exploratory research show that study visits about adaptation policies can (1) particularly stimulate conceptual learning, (2) increase the credibility of policies within municipal administrations, (3) be used as strategic instruments by mentor cities, (4) be more successful if the peer-cities are not too different (in terms of size, institutional context), and (5) under certain conditions, lead to policy adoption in a learning city. Future research needs to critically discuss the mass suitability of learning from frontrunner cities. Furthermore, a call is raised for more research and practical action on how to initiate and improve learning exchanges beyond the strict division between mentors and learners. Instead, the focus needs to be on mutual learning exchanges.

Keywords

climate change adaptation knowledge brokers study visits policy learning urban policy mobilities

Introduction

Climate change is putting increasing pressure on cities. The instability is manifested in a higher incidence of flooding, heat waves, water scarcity, and so on. Responding to these challenges places new and often complex demands on local policy makers. Local governments need to accelerate the learning process on how to respond to these challenges, by looking for solutions that have worked elsewhere, for instance, and by learning from those experiences. City networks are increasingly offering tools to disseminate these solutions and to facilitate mutual learning processes among local policy makers (see Bansard, Pattberg, & Widerberg 2017; Haupt et al. 2019). These tools include conferences, workshops, webinars, and study visits (Haupt and Coppola 2019). This article focuses on the latter. More precisely, the focus is on study visits through which representatives of one or more cities learn from a host city that has faced or faces similar adaptation challenges.

Referring to the in-person exchanges of local policy makers, the most commonly used terms in the literature are study visits (Ma 2017; Sheldrick, Evans, and Schliwa 2016), study tours (e.g., Montero 2017; Wood 2016), site visits (Ma 2017), and policy tourism (Cook and Ward 2011, 2015, González 2011). This article investigates study visits, which are mainly organized and managed by the two collaborating European city networks: EUROCITIES and the Covenant of Mayors (CoM, henceforth). Through the so-called CoM Twinning Program (referred to as CoM Twinning Programme in the European Union), these two organizations connect European cities that face similar adaptation challenges to one another and co-organize and facilitate multiday study visits between or among them. In the literature, facilitators of knowledge transfers, such as EUROCITIES and the CoM, are also referred to as knowledge brokers (see Dotti and Spithoven 2017) or policy brokers (Howlett, Mukherjee, and Koppenjan 2017).

Despite being a commonly used tool in urban governance, we still know little about the importance and outcomes of study visits (Cook and Ward 2011). More generally, learning through international cooperation is, despite growing importance, still an underresearched phenomenon (Vinke-de Kruijf and Pahl-Wostl 2016). The study visits covered by academic research mostly focused on topics such as urban regeneration (González 2011), mega-events (Cook and Ward 2011), or transport and mobility (Ma 2017; Montero 2017; Sheldrick, Evans, and Schliwa 2016). Although for some of the transport-related research, climate change mitigation also played a certain role (e.g., cycling policies), study visits related to climate change adaptation or the integration of adaptation and mitigation have not yet been extensively studied. Indeed, also independently from study visits, previous research has pointed to a certain dichotomy between adaptation and mitigation (at the local scale) (Biesbroek, Swart, and Van der Knaap 2009) and highlighted potential conflicts between the two approaches, such as policies that have a positive effect on mitigation but a negative one on adaptation or vice versa (Grafakos et al. 2020). However, recently, the co-benefits and synergies between adaption and mitigation are increasingly being explored and highlighted (Grafakos et al. 2020; Pilato, Sallu, and Gaworek-Michalczenia 2018). Particularly due to the lack of investigation on adaptation policies, Yeganeh, McCoy, and Schenk (2019) suggested that future research on the adoption of climate policies, as promoted through formats such as study visits, should first and foremost focus on adaption rather than on mitigation. This would allow us to decipher how learning on adaptation may differ from learning on mitigation. This is especially interesting as adaptation and mitigation often involve the same municipal staff.

In particular, this article aims to contribute to a better understanding of learning exchanges among local policy makers: Indeed, how do local policy makers actually learn about climate policies through study visits (if they do)? This overall question is divided into five subquestions deriving from key research insights and gaps from the literature on policy mobilities and policy learning:

How did the knowledge brokers organize the study visits?

Which forms of learning took place (if any at all)?

What were the mentor’s motivations to host and organize study visits, and could they also learn something from the learners?

Did contextual differences affect the learning process and outcome, and if yes, how?

Could the study visits lead to policy adoption?

These questions were addressed through 14 expert interviews with key stakeholders attending or co-organizing the visits, most of them local policy makers. Four twinning visits that included 10 municipalities from nine European countries were examined.

The article is structured as follows: First, the policy mobilities and policy learning literature addressing study visits is reviewed and discussed. Following that is a presentation of the CoM Twinning Program. Thereafter, the methodology section outlines the case selections and explains the inquiry design. The findings are then presented, followed by empirical analysis and discussion. Finally, main conclusions are drawn.

Study Visits and How to Examine Them

This section introduces the key policy mobilities and policy learning literature covering study visits. At first, study visits are defined and explored. Thereafter, key insights from the literature are presented and grouped into the five themes that determine the research questions of this article.

Exploring Study Visits

“Urban politics and policy are never just local,” according to Baker and Tenemos (2015, p. 824). Indeed, the literature on policy mobilities “emphasizes the mobile and changeable character of policies” (Peck and Theodore 2010, p. 170). Therefore, researchers center their attention on “the processes, practices, and resources brought together to construct, mobilize and territorialize policy knowledge” (Baker and Tenemos 2015, p. 825), which is increasingly done through study visits (McFarlane 2011; Montero 2017).

Montero (2017, p. 336) defined study visits as “short visits in which a delegation of people travels to another place to experience something with potential to improve their organizations or places of origin.” McFarlane (2011) characterized them as coordination tools that allow the movement of knowledge. It is not particularly new for local policy makers to learn from their peers to transfer ideas from elsewhere back to their place. The first documented study visit between local governments probably took place around the early 1900s when the idea of garden cities was promoted throughout the United Kingdom (Montero 2017). However, in recent years, learning through study visits has become more common (González 2011). One main reason for this was its promotion, organization, and facilitation through national and international institutions such as Interreg or EUROCITIES (González 2011).

González (2011) and Cook and Ward (2011, 2015) embedded study visits in a phenomenon they called policy tourism. González associated classic leisure tourism with the travels of “urban policy tourists” who go to other places to learn about innovative policies. Similar to leisure tourism, “policy tourism is also wrapped up with myths about policies in ‘other’ places” (González 2011, p. 1400). However, a study visit is not a cruise but is a key approach to learn about a policy and disseminate knowledge (Ma 2017). The findings of Montero (2017) and González (2011) suggest that study visits are not just “neutral learning devices” (Montero 2017, p. 346). Indeed, they can also help create legitimacy (González 2011) to initiate policy change (Montero 2017). Furthermore, the participants of the visits are often selected based on “their capacity to influence policy back home” (Montero 2017, p. 337). In this context, Ma (2017) showed that study visits facilitate the transfer and diffusion of policies from “innovation stars.” However, despite enjoying growing popularity, the actual learning possibilities enabled by study visits were contested by critical scholars. Indeed, as pointed out by Montero (2017, p. 338), there are concerns that these tours could be or become “lobbying instruments in the hands of elite local actors.”

Examining Study Visits

In the following sections, five themes that need to be considered for the examination of study visits are introduced and explained. The themes are derived from key findings and research gaps of the policy mobilities and policy learning literature. The themes determine the five subquestions this article addresses and closely relate to the article’s main research question: How do local policy makers learn through study visits (if they do)?

(1) The role of knowledge brokers.

What all visits examined in this article have in common is that they were facilitated and co-organized by a knowledge broker. Variously referred to as brokers, boundary organizations, and intermediaries, the literature confirms the emergence of new stakeholders acting between the producers and the users of knowledge (Dotti and Spithoven 2017). In the policy mobilities literature, these brokers are also referred to as transfer agents (see McCann 2011). Research on knowledge brokers is somewhat thin and in its early stages, and many aspects are barely explored. These include, among others, their influence on transnational partnerships and their overall strategies (Dotti and Spithoven 2017).

In this article, the broker is a consortium of city networks. As Borrás and Radaelli (2011) outlined, international organizations engaging in the transfer of knowledge often have clear normative ideas of how policy makers should learn. To put those ideas in practice, they often provide a variety of instruments. These instruments include facilitated coordination tools such as study visits (Borrás and Radaelli 2011).

Knowledge brokers’ key function is to link up groups that otherwise would remain disconnected (Howlett, Mukherjee, and Koppenjan 2017). Knockaert and Spithoven (2014) called them crucial actors as “not all knowledge can be ‘traded’ as a commodity because it requires adaptation and translation” (as cited in Dotti and Spithoven 2017, p. 1). Also, it is the broker’s job to find the “right” match and connect the most suitable stakeholders (Dotti and Spithoven 2017). This is precisely what the city networks mentioned above did for the CoM Twinning Program.

(2) Different forms of learning.

Despite the existence of several theories and definitions of policy learning, there is still little knowledge on how policy makers actually learn (Gilardi and Radaelli 2012). What complicates the exploration of learning exchanges among policy makers is that many of these are undertaken through informal interactions (Bell and Park 2006; Haupt et al. 2019). However, what can be done is to systematically analyze the local policy maker’s learning experiences. In this context, Kemp and Weehuizen (2005) suggest distinguishing three different forms of learning: technical, conceptual, and social learning.

Technical learning is learning about certain instruments, and how those could be improved to achieve a certain goal (Kemp and Weehuizen 2005). More precisely, it is about the search for appropriate policy instruments to achieve previously set goals (Kulsum and Sánchez-Triana 2008). Rather than contesting these goals, technical learning “just” focuses on how to implement them (Kulsum and Sánchez-Triana 2008).

Conceptual learning aims to see the bigger picture or look at an issue from a new perspective (e.g., developing strategies or overall goals) (Kemp and Weehuizen 2005). It includes debating appropriate strategies and goals or reformulating them (Kulsum and Sánchez-Triana 2008).

Social learning means learning about values, norms, or responsibilities (Kemp and Weehuizen 2005). Of particular importance are “multi-stakeholder views and information for improving both technical and conceptual learning” (Kulsum and Sánchez-Triana 2008, p. 161). Given the evident overlaps between conceptual and social learning, a clear distinction is not always possible (Kemp and Weehuizen 2005).

(3) Motivation to host and organize study visits.

Cities that have and want to maintain an image of being progressive in a particular field often organize study visits and invite visitors to learn from their models. In this context, the branding of the study destination (Andersson 2015) and the strategic self-promotion of their place by local policy makers play a key role (Fisher 2014). Two popular and well-studied examples are Bilbao and Barcelona, which regularly welcome policy makers from abroad to learn more about (allegedly) successful urban regeneration (see González 2011). Hjerpe, Storbjörk, and Alberth (2015) demonstrated that an increasingly important role for local politicians is also in adaptation, attracting attention, and marketing their places as leading cities.

Besides the motivation to use study visits as strategic instruments for place branding, the question about the host city’s learning opportunities emerges. While there is evidence that learning from supposedly leading cities can help a learning city (Haupt et al. 2019), little attention has been paid to the potential learning opportunities for the host city of the study visit. Indeed, Kern (2019) found that there are successful learning dynamics in European local climate governance between leading, following, or laggard cities and that “upscaling local experiments helps to close the gap between these actors” (p. 125). Then again, it is unclear if and how a supposedly leading city learns through these exchanges.

(4) Contextual differences among the learning partners.

Generally, little is known about the transfer process of urban climate policies “in bilateral cooperation that embeds knowledge gaps and different political context” (Shefer 2019, p. 61) and “how they are reinterpreted in each context” (Fisher 2014, p. 154). However, transnational learning, as increasingly done through study visits, also requires taking into account the role of different institutional or cultural contexts and how these may influence the learning outcomes or the adoption of policies. In this regard, Stead and Pojani (2018), who focused on the international circulation of Dutch transport and urban planning policies, found that “in many cases, contextual differences (e.g., cultural and social norms, language, planning legislation and financial resources available to planning) limit the extent to which Dutch planning approaches can be employed elsewhere” (p. 59). Fisher (2014, p. 171), who studied the movement of urban climate policies in India, also emphasized the key role of “the particular cultural context” and stressed that “the spaces for policy mobility need to be catered” to it.

(5) Learning and policy adoption.

A possible and desired outcome of a twinning could be the adoption of a policy by a participating city as a result of the learning exchange. This raises the current question about the relationship between learning and policy adoption. In this regard, Kemp and Weehuizen (2005) suggested exploring what role learning has played for policy change or adoption.

It is important to emphasize that policy adoption is a complex process that starts with recognizing the issue and often requires the involvement of several government levels (Prater and Lindell 2000). Furthermore, it usually takes time (Jaffe, Newell, and Stavins 2005; Prater and Lindell 2000). Consequently, there needs to be sufficient time between the study visit and the examination of possible policy adoption. Indeed, for the examination of policy learning, Kulsum and Sánchez-Triana (2008, p. 161) suggested choosing “long time frames and to have conservative expectations about the potential for actual learning.”

Adoption usually means translating a policy into the specific local context while mere replication is rarely considered to be useful (Stead and Pojani 2018). This process of translation and how policies are reinterpreted in different contexts is not well understood (Fisher 2014; Jokinen et al. 2018).

Kern, Koll, and Schophaus (2007) found that the size of a city often plays a key role in the adoption of climate or sustainability policies. Indeed, larger cities have greater institutional capacities (e.g., staff) to implement policies and to position themselves as frontrunners (Kern, Koll, and Schophaus 2007; Wang 2012). Also, Hjerpe, Storbjörk, and Alberth (2015) found that “size matters.” Apart from having more staff, larger cities usually also have more specialized staff. In turn, a lower rate of policy adoption can often be explained by capacity constraints (Homsy and Warner 2015).

Moreover, it needs to be highlighted that even if no policy adoption has taken place, study visits can be helpful for local policy makers. In this context, the question arises whether the occurrence of policy adoption is absolutely necessary to label a study visit, or more specifically, a study visit’s outcome, as positive. Indeed, the “simple inspiration and openness to knowledge” can help “practitioners, politicians or academics to understand their ‘home’ situation better and to consider their own practices and knowledge in a different light or with a more critical eye” (Stead and Pojani 2018, p. 64). Consequently, it is equally conceivable that study visits fail to reach the initial goal of triggering the adoption of a certain policy but succeed in drawing attention to certain policies or general approaches on the visiting policy makers’ side. From the knowledge gained through the study visit, the learning policy makers might draw more fitting conclusions for their city. Indeed, they know their “home” situation better than the policy makers that hosted the study visit and presented the policies that worked in their place.

The Covenant of Mayors Twinning Program

This section introduces the CoM Twinning Program. The descriptive background information and data provided in this section can be found on the websites of the CoM and EUROCITIES and in reports and academic work on the CoM initiative (see Haupt 2018; Kona et al. 2015). First, the emergence history of the eponymous CoM, its goals, and its embeddedness in the European Union (EU) policy-making framework are described. Thereafter, a closer look is taken at the actual Twinning Program, including how it was conceived, how it functions, and how it evolved over the years.

The program was established in 2015 under the name Mayors Adapt Twinning Program. Mayors Adapt was an adaptation initiative launched by the European Commission (EC) in 2014. In 2015, Mayors Adapt merged with the European mitigation initiative CoM that was established in 2008. The new initiative was named the Covenant of Mayors for Climate and Energy and focuses on mitigation and adaptation at the local level. With the ending of Mayors Adapt, the program was renamed as CoM Twinning Program.

The EC established the CoM in 2008 shortly after the European Parliament had adopted the 2020 climate and energy package. The package included goals for the reduction of greenhouse gas emissions, an increase of renewable energies, and an increase in energy efficiency. Initially, these were goals for the EU member states, but, through the CoM, the local level was included, and local governments were explicitly encouraged to surpass their own country’s ambitions. A city that joined the CoM committed to the reduction of greenhouse gas emissions by at least 20% on its territory, a 2020 goal, or by 40%, which refers to the 2030 goals, and to develop a joint approach for tackling mitigation and adaptation. Therefore, every member is asked to develop and implement a climate and energy action specifying how the targets set are planned to be achieved. Until July 2019, 9,664 local authorities had joined the CoM.

With the CoM, the EC has established an organization to implement EU policies on the local level. Beyond that, the CoM exemplifies the increased importance of the local level within the EU policy-making framework and demonstrates how the EU has identified networking as a strategy for policy dissemination and implementation. The networking approach—more precisely, the networking among cities and among many different initiatives or institutions as achieved through formats like the Twinning Program—is also reflected in the interrelationships between the organizations managing the program. As the name indicates, the Twinning Program is formally run by the CoM. However, the CoM office is managed and supported by several further related organizations, such as Energy Cities, the Climate Alliance, The Council of European Municipalities and Regions, Local Governments for Sustainability, and EUROCITIES. Overall, the CoM has 19 employees, of which nine are part of the Capacity Building and knowledge-sharing unit, which primarily focuses on the twinning program. Among the consortium of organizations supporting the program, EUROCITIES is the organization dedicating most of the human resources. The contact person for general inquiries works for EUROCITIES, and most of the persons co-organizing and attending the twinning in person are employees of EUROCITIES. Through the program, municipalities or provinces are connected to collaborate and share experience and knowledge on one or more topics related to adaptation or mitigation. Past twinnings covered, for example, cloudburst and water management, stakeholder involvement and awareness, or heat wave and urban heat island management. Successful applicants for participation received funding to cover the travel costs of the visit. In preparation of the visits (main topics, goals, schedule), the partners were assisted by a twinning facilitator, mostly from EUROCTIES. In most cases, the twinning facilitator also attended the visit. In the first two twinning rounds¹, seven facilitators were from EUROCITIES, three from Mayors Adapt (now the CoM), and one from Ecofys² and the Climate Alliance. All communication—orally during the visit and in writing for the prior and post exchange of emails—was done in the English language.

As illustrated in Figure 1, since 2016, 43 municipalities and two provinces have participated in the program, and 20 twinnings were realized. The main stakeholders attending twinning visits and being present throughout the entire program were the local policy makers representing their cities (visitors or hosts). However, depending on the twinning, several other stakeholders also took part in the visits. These included but were not exclusive to, municipal institutions (e.g., utility companies), researchers, consultants, or representatives of local businesses. Generally, the number of participants varied widely among the twinnings. For instance, the twinning between Cascais and Ilion was attended by five stakeholders, whereas 18 were involved in the twinning between Bologna, Lleida, and Terra di Leuca. However, in this context, it needs to be stressed that usually only very few persons of each twinning partner (one to three local policy makers or technicians) attended the entire course of the visit and participated in the organizational work prior to the visits. Additional participants attended selected parts of the agenda (e.g., presenting a specific measure to the visitors). The funding provided by the CoM did not allow “bigger” exchanges: Host cities received 500 Euros to cover their costs (e.g., local transport, catering, booking rooms). Visiting cities received up to 710 Euros (traveling, accommodation).

Figure 1.

Twinnings on adaptation and mitigation realized between 2016 and 2018.

Since its establishment in 2015, the program underwent changes every year. The first twinning round consisted exclusively of so-called mentoring exchanges. For a mentoring exchange, a mentor hosted one or more learners for a two- or three-day study visit. In the second year, the program was complemented by peer-learning exchanges. Other than mentoring exchanges, peer-learning exchanges are between partners that are supposed to be on the approximate same level in the respective twinning topic. In the third program year, provinces were also allowed to participate and visit a host local government, while a return visit to the twinned local government was added as a compulsory element of the program.

Methodology

The methodology section is organized as follows: first, the selection of the examined twinnings and surveyed cities are outlined, followed by the explanation of the inquiry design.

Explaining the Case Selection

Overall, four twinnings were examined: three mentoring and one peer-learning exchange. These included the mentoring exchanges between Copenhagen and Antwerp; Cascais and Ilion; and Bologna, Lleida, and Terra di Leuca; and the peer-learning exchange between Bratislava, Bremen, and Arnhem.

Three main criteria determined the choice of cases. First, it was important to select twinnings that had already occurred in the past. As outlined above, studying and assessing policy learning also requires looking at its impact on policy change or adoption (Kemp and Weehuizen 2005). Then again, policy adoption takes time (Jaffe, Newell, and Stavins 2005; Prater and Lindell 2000). The assumption was that selecting cases from the earlier twinning rounds would increase the probability of study exchanges that resulted in policy adoption.

Second, both types of twinnings (mentoring and peer-learning exchanges) should be represented. With the first criteria in mind, only mentoring exchanges from the first twinning round (2016) and peer-learning exchanges from the second round (2017) were selected (there were no peer-learning exchanges in the first round). By applying the described exclusion criteria, 11 mentoring and two peer-learning exchanges were shortlisted.

The third criterion concerned the similarities and differences between the twinning partners. The twinnings included exchanges with similar but also quite different partners. For instance, there were differences, or similarities, in their size, stage of work, and adaptation challenges (see Table 1). Those features were also reflected in the choice of examined cases. Copenhagen and Antwerp represent a twinning between cities from Northern Europe; Cascais and Ilion, and Bologna, Lleida, and Terra di Leuca were South-to-South twinnings; whereas Bratislava, Arnhem, and Bremen represent a twinning between two Northern European and a post-socialist, East-Central European city.

Table 1.

Examined Twinnings.

Twinning 1 (T1): Copenhagen (host, M) with Antwerp (L)		Twinning 2 (T2): Cascais (host, M) with Ilion (L)
• April 25–28, 2016, first twinning round • Shared climate hazards: Extreme temperatures, flooding sea-level rise • Twinning topics: Cloudburst management, urban water management, green climate adaptation		• March 3–4, 2016, first twinning round • Shared climate hazards: Extreme temperatures, flooding sea-level rise • Twinning topics: Urban green parks and green corridors, water management, strategic adaptation planning
Copenhagen (Denmark)	Antwerp (Belgium)	Cascais (Portugal)	Ilion (Greece)
• Capital of Denmark • Population: ~623,000Size: 86.20 km²	• Capital of the province Flanders • Population: ~523,000Size: 204.51 km²	• Municipality in the metropolitan area of Lisbon • Population: ~205,000Size: 97.39 km²	• Municipality in the metropolitan area of Athens • Population: ~85,000Size: 9.453 km²
Previous climate action		Previous climate action
• 2009: Joined the CoM • 2011: Adaptation plan2012: cloudburst management plan	• 2009: Joined the CoM • 2015: Adaptation plan	• 2008: Joined the CoM • 2010: Strategic climate plan	• 2010: Joined the CoM • No adaptation plan/strategy
Twinning participants		Twinning participants
• 10 participants from Copenhagen (local policy makers, technicians, energy policy experts, urban designer, & architects) • 2 participants from Antwerp (local policy makers)1 twinning facilitator (EUROCITIES)		• 5 participants from Cascais (local policy makers, technicians) • 1 representative from Ilion (local policy maker) • 1 twinning facilitator (EUROCITIES)
Twinning 3 (T3): Bologna (host, M) with Lleida (L) and Terra di Leuca (L) • April 18–19, 2016, first twinning round • Shared climate hazards: Extreme temperatures, droughts, storms, water scarcityTwinning topics: EU funded projects (LIFE+, Central Europe, IEE, Comenius), water scarcity, agricultural needs, heat waves, extreme rain events
Bologna (Italy)	Lleida (Spain)	Terra di Leuca (Italy)
• Capital of the region Emilia-Romagna (northern Italy) • Population: ~389,000Size: 140 km²	• Capital of the province Lleida in the autonomous region Catalonia (Northern Spain) • Population: ~138,000Size: 211.7 km²	• Union of municipalities in the region Puglia (Southern Italy) • Population: ~30,000Size: 134.92 km²
Bologna (Italy)	Lleida (Spain)	Terra di Leuca (Italy)
Previous climate action
• 2008: Joined the CoM • 2013: Adaptation strategy	• 2009: Joined the CoM • 2015: Adaptation plan	• Did not join the CoM • No adaptation plan/strategy
Twinning participants
• 12 participants from Bologna (local policy makers and regional policy makers, technicians, and energy and climate experts) • 2 participants from Lleida (local policy makers, technicians) • 3 participants from Terra di Leuca (politicians, local policy makers)1 twinning facilitator (Mayors Adapt)
Twinning 4 (T4): Bratislava (host, PL) with Lleida (PL) and Terra di Leuca (PL) • January 16–17, 2017, second twinning round • Shared climate hazards: Extreme temperatures, floodingTwinning topics: Urban heat islands, intelligent and cost-effective water management, green infrastructure
Bratislava (Slovakia)	Bremen (Germany)	Arnhem (The Netherlands)
• Capital of Slovakia • Population: ~433,000Size: 367.661 km²	• City state (Northern Germany) • Population: ~569,000Size: 325.56 km²	• Capital of the province Gelderland (The Netherlands) • Population: ~159,000Size: 101.54 km²
Previous climate action
• 2012: Joined the CoM • 2014: Adaptation strategy	• 2008: Joined the CoM • 2012: Adaptation strategy	• 2012: Adaptation strategy • 2014: Joined the CoM
Twinning participants
• 12 participants from Bratislava (local policy makers, technicians, researcher) • 2 participants from Arnhem (local policy makers) • 1 participant from Bremen (local policy maker) • 3 external experts (Ecofys, University of Ústí nad Labem)1 twinning facilitator (Mayors Adapt)

Source. Covenant of Mayors (n.d.); EUROCITIES (n.d.).

Note. CoM = Covenant of Mayors; EU = European Union; IEE = Intelligent Energy Europe; M = Mentor; L = Learner; PL = Peer Learning.

The reason why two mentoring exchanges between somewhat similar partners (Copenhagen and Antwerp, and Cascais and Ilion) were selected was mainly because, in one case, the mentor was a very experienced host of international delegations (Copenhagen), and in the other case, the mentor was doing this for the first time (Cascais). These two twinnings quasi-represent the most different cases (experienced vs. inexperienced mentor) among similar cases (mentors and learners with similar city profiles). Indeed, selecting the most different cases can increase the “understanding of complex spatial variables” (Sovacool, Axsen, and Sorrell 2018).

While the conducted research certainly has comparative aspects, its level of generalizability should not be overstated. The exploratory nature of this research is mainly due to the early stage of extant scholarly work on study visits in the field of adaptation. Indeed, the Twinning Program has not yet been the object of any investigation. The chosen twinnings represent cases that are different from one another. However, a certain degree of homogeneity is guaranteed through participation in the program (e.g., complying with the same guidelines) and belonging to the EU (e.g., joint governance framework). Nevertheless, besides the mentioned advantages of choosing somewhat different cases, this variation within cases also bears the danger of reducing the generalizability of the findings (Sovacool, Axsen, and Sorrell 2018).

Explaining the Inquiry

Interviews with 14 twinning participants were carried out between March 2017 and July 2018 (see Table 2). The interviews were semistructured and open-ended and were conducted on site (two), via Skype or phone (11), or email (one). The questions asked of the interviewees were oriented on five themes derived from key insights and research gaps of the policy mobilities and policy learning literature as outlined in the literature review:

How did the knowledge brokers organize the study visits?

Which forms of learning took place (if any at all)?

What were the mentor’s motivations to host and organize study visits and could they also learn something from the learners?

Did contextual differences affect the learning process and outcome, and if yes, how?

Could the study visits lead to policy adoption?

Table 2.

Interviewed Stakeholder.

Twinning	Interviewees’ Affiliation	Interview Date
T1: Antwerp–Copenhagen	Environmental Office—Energy and Environment, Antwerp municipality	March 24, 2017
T1: Antwerp–Copenhagen	Environmental Office—Climate Adaptation, Antwerp municipality	Dec. 6, 2017
T1: Antwerp–Copenhagen	Technical and Environmental Department—Climate Unit, Copenhagen municipality	April 4, 2019
T1: Antwerp–Copenhagen	Saint Kjelds Neighbourhood, Copenhagen municipality	July 13, 2018
T1: Antwerp–Copenhagen	HOFOR, Greater Copenhagen utility company	April 13, 2018
T2: Cascais–Ilion	Environmental Department, Cascais municipality	April 3, 2018
T2: Cascais–Ilion	Technical Department, Ilion municipality	April 18, 2018
T3: Bologna–Lleida–Terra di Leuca	Environmental Department—Sustainability Office, Bologna municipality	May 23, 2018
T3: Bologna–Lleida–Terra di Leuca	Environmental Department, Lleida municipality	May 25, 2018
T3: Bologna–Lleida–Terra di Leuca	Environmental Department, Lleida municipality	May 25, 2018
T4: Bratislava–Arnhem–Bremen	Public Space, Water and Ecology, Arnhem municipality	Nov. 29, 2017
T4: Bratislava–Arnhem–Bremen	The Senator for Environment, Urban Development and Mobility, Bremen municipality	Dec. 7, 2017
T4: Bratislava–Arnhem–Bremen	Comenius University, Bratislava	April 30, 2018
No specific twinning	EUROCITIES, Brussels	May 16, 2018

Source. Own table.

Could the study visits lead to policy adoption? The intention was to ask the target group of the inquiry quite specifically, How do local policy makers learn through study visits (if they do)? Consequently, it was mainly local policy makers who were actively involved in the twinning visit (as organizers or as visitors) that were chosen as interviewees. Nevertheless, also other stakeholders that attended the twinnings or contributed to their organization (e.g., private partners, other local actors) were interviewed (see Table 2). Contacts (names and positions) of participants of the visits could be found in the reports of each twinning. In some cases, additional interviewees from the same twinning arrived through recommendations from past interviewees (snowball sampling). Their names could be found in the corresponding twinning report where all attendees are listed.

The analysis of the interview data mostly oriented on the presented a priori themes that also mainly determined the guideline questions asked of the interviewees. The interviews were recorded and fully transcribed shortly afterward, on the same day or the day after. After transcription, the focus was on the sequencing of the interviews according to thematic units (paraphrasing). The paraphrasing followed the course of the conversation, taking into account the overall messages of the interviewees and not just extracts of them. This ensured that their messages were not misunderstood and reproduced in a shortened or even incorrect way (see Meuser and Nagel 2009). Therefore, at first, all interviews were separately analyzed in their entirety. In the following step, the interview data were coded using emergent codes. The final analytical step, the thematic comparison, went beyond individual text units and focused on identifying and clustering comparable text units from the interviews.

Undoubtedly, focusing on local policy makers comes with the danger of relying too much on the perceptions of only a few stakeholders. Therefore, other sources and necessary information to verify the impression from the twinnings interviews were used and analyzed to double-check the given answers. The most relevant documents were the twinning reports, documents of usually around 15 pages. They contain, among other details, the agendas and the main topics of the visits.

Findings

The following section summarizes the main results of the interviews and is divided into five subsections that outline key research insights and gaps of the policy mobilities and policy learning literature.

How Did the Knowledge Broker Organize the Study Visits?

To facilitate and organize the twinnings, EUROCITIES had several duties. These included setting up and advancing the program, reviewing the applications, selecting the twinning partners, and accompanying, and attending the twinning visits. Over the years, the program was changed, some elements were replaced or complemented by other elements. Drawing on interviews with several local policy makers and a representative of EUROCITIES, the following describes and discusses this in more detail.

As co-managers of the program, EUROCITIES examined the applications and matched the twinning partners. The main criteria to twin the “right” local governments were based on several matchmaking criteria, such as the applicant’s preferred exchange form (mentoring or peer-learning), which was usually strongly related to their current stage of the mitigation and adaptation work (e.g., implementation, design phase). Apart from these criteria, the brokers tried to match cities that share sufficient similarities in the following areas: nature of climate hazards faced, interests in sectors/areas of intervention, preferred time slot to organize twinning visit, and their socioeconomic and institutional context (e.g., population size, institutional capacities). The interview with a representative of EUROCITIES revealed that the latter criterion was added in a later phase. After the first twinning round, the program designers had the impression that twinnings between cities that were very different with regard to their socioeconomic and institutional context were not as fruitful as twinnings between more similar partners. For instance, during the first twinning round, a twinning was realized between the metropolis Barcelona (approximately 1,620,000 inhabitants) and the Crete town Chania (approximately 108,000 inhabitants).

The further development of the program is strongly based on the feedback and the experiences provided by the initial participants. When asked about possible improvements, Ilion and Bremen remarked that a two- or three-day twinning was not enough to understand and discuss all issues at stake. In addition, Bologna and Bremen suggested that a second meeting could have improved the twinning outcomes. Bremen specified that, for a second meeting, which could take place half a year after the first visit, every partner could reflect on the lessons learned and prepare further questions. It was remarked that, during one visit, it is hardly possible to fully understand a new measure or policy.

It was because of these previous twinning participants’ remarks that EUROCITIES decided to make return visits a compulsory program component after the second twinning round. Moreover, during the first two rounds, EUROCITIES did not demonstrate efforts to keep the collaboration alive. Indeed, all interviewees confirmed that, after completing the final twinning report, EUROCITES did not contact them again. However, for the third round, EUROCITIES expected and requested the partners to stay in close contact to organize and prepare the return visit one year after the first visit. Helping to maintain contact could be seen as a task for a knowledge broker. Indeed, the interviews revealed that only very few of the twinning partners were still in contact after the end of the formal program. In some cases, the collaboration ended immediately after finalizing the twinning report. In other cases, it continued for some months in the form of post-twinning information exchanges.

Despite the described changes that were made to the program over the years, there has never been a systematic evaluation of the twinning outcomes (e.g., on policy adoption) from the side of EUROCITIES or the other involved organizations. This is astonishing because the interviewee representing EUROCITIES also emphasized that the EC, the funder of the twinnings, specifically asked the program designers to focus on increasing the applicability potential of the measures and policies presented during a study visit. There had been attempts to do so but it proved to be too difficult to find appropriate indicators to systematically assess the learning outcomes of a twinning.

Which Forms of Learning Took Place (if Any at All)?

In the following paragraphs, the different forms of learning that took place among the policy makers or that were intended to be performed are presented and explained.

The main motivations for the different cities to participate in the twinning also afford a look at their learning objectives and whether they were met. These main motivations are summarized in Table 3. Antwerp’s and Bremen’s main motivation to participate in the twinning was that the cities had identified a window of opportunity. In Antwerp, major infrastructural works were taking place, which opened up new possibilities to co-design urban development in a more climate-adaptive way. At the same time, the city was in the development phase of its water plan. Bremen felt that the twinning opportunity came at the right moment because they were in the midst of adaptation strategy development and looking for further inspirations and experiences. Both cities were in a strategy development phase (seeing the bigger picture, looking at an issue from a new perspective), which represents conceptual learning (see Kemp and Weehuizen 2005; Kulsum and Sánchez-Triana 2008). The learning experiences of Bremen and Antwerp are explained in more detail below.

Table 3.

Motivations to Participate in the Twinnings.

Clear Learning Objective	Other Objectives
• Antwerp (L, T1), Bremen (PL, T4): Seeking inspiration and ideas for strategy development (conceptual learning). • Ilion (L, T2): Seeking inspiration and ideas to start the adaptation work from scratch (conceptual learning).Arnhem (PL, T4): Learning about stakeholder involvement strategies (technical learning).	• Bologna (M, T3), Cascais (M, T2), Ilion (L, T2), Lleida (L, T3): Generating new EU-level partnerships for future projects. • Cascais (M, T2), Copenhagen (M, T1): Receiving a critical review of their work.Copenhagen (M, T1): City branding.

Source. Own table.

Note. EU = European Union; M = Mentor; L = Learner; PL = Peer Learning.

In contrast, Bremen’s twinning partner, Arnhem, already had a strategy that had been approved by the city council. Their main motivation to participate was to find answers to a specific question: How can a local government successfully involve stakeholders to put the approved strategy into practice? Arnhem hoped to learn more about this from its partners, Bremen and Bratislava. Arnhem intended to perform technical learning, which is defined as learning about certain instruments, and how those could be improved to achieve a previously set goal (see Kemp and Weehuizen 2005; Kulsum and Sánchez-Triana 2008). However, Arnhem’s clearly defined learning objective was not met through the twinning visit. Arnhem’s representative explained this was due to the partners’ different stage in strategy development. While Bratislava was at the stage of climate mapping and Bremen was in the midst of its adaptation strategy development, Arnhem had already an approved strategy in place.

In a sense, among the learners, Ilion represents the counter example to Arnhem. Ilion had no previous experience in adaptive planning (and, thus, no very specific questions) but started the twinning with Cascais. In 2015, heavy rainfall in the Athens region caused 24 deaths and revealed Ilion’s vulnerability. Starting from scratch, Ilion’s motivation was to learn as much as possible (conceptual learning) from a comparable city that had already successfully implemented actions under similar conditions (climate, institutional context, little financial means).

Even though Cascais’s community gardens, which were a key component of the visit, proved to be nontransferable for Ilion due to too much air pollution, Ilion described this particular item on the agenda as the most helpful one of the entire exchange. This was mainly because the Cascais hosts explained in depth how to effectively involve citizens in urban policy-making processes. Ilion highlighted that this approach could also work in several other contexts of urban policy making and was the key lesson learned through the twinning.

Contrary to the other learners, Lleida’s decision to apply was not driven by specific learning objectives. Instead, their main goal was to establish new partnerships with local government practitioners within the EU. This motivation was due to positive past experiences with other EU projects that led to long-term relationships with other European cities. Then again, it turned out that through the visit to Bologna, Lleida started a conceptual learning exchange that, in the end, even led to policy adoption in the city. This exchange is portrayed in more detail below.

What Were the Mentor’s Motivations to Host and Organize Study Visits and Could They Also Learn Something from the Learners?

Before participating in the surveyed twinnings, neither of the two mentors defined specific learning objectives (see Table 3). Indeed, the question arises as to why a mentor city organizes a multiday visit to provide its solutions to a learning city for “free?” Some might claim that this could also be done through a paid consultancy.

Apart from the predictable answer of sharing visions to help others, the interviewees from the mentors Copenhagen and Cascais also explained that a key motivation was to receive a critical review. A representative of an autonomous municipal water utility company in Copenhagen pointed out that those critical comments are preferred in the course of an in-depth study visit with peers doing the same job elsewhere than, for example, at a conference.

Bologna and Cascais stated that the main motivation to participate was to generate new European-level partnerships for future projects. Cascais specified that this could, for instance, be joint grant applications for EU projects. Copenhagen remarked that hosting guests for a study visit is part of the city branding and helps the city to become known for being progressive and innovative.

With regard to the own learning experiences, it was remarked by the mentors Bologna, Cascais, and Copenhagen that a learning exchange is never completely one-sided, with one partner only “teaching” and another only “learning.” However, when the mentors were asked to name concrete own learning outcomes, they could give no examples. Instead, the lessons drawn involved realizing that the learners were facing the same challenges and difficulties as they did when they were in earlier stages of their adaptation work. This, however, hardly qualifies as learning.

There were also cases when cities decided against mentoring. Arnhem explained that they were initially supposed to be twinned for a mentoring exchange. However, they had the impression that they would be mainly helping another local government without learning a lot themselves. Instead, Arnhem decided to participate in the peer-learning exchange with Bratislava and Bremen. Indeed, EUROCITIES stated that, in the previous application rounds, they had trouble finding mentors. After reviewing the first twinning rounds, EUROCITIES had the impression that the responsibility to organize the visits was distributed unevenly, to the disadvantage of the mentors. As an example, it was the mentor’s duty to set up the visits onsite (e.g., contacting, inviting, and instructing all participating stakeholders, or checking the availability of and if necessary, booking the sites to be visited).

Did Contextual Differences Affect the Learning Process and Outcome, and if Yes, How?

Many cities collaborating through the twinning were quite different from each other in terms of size, institutional traditions, procedures, and capacities. Almost all exchanges were between partners from different countries. But how did those differences affect the learning processes and outcomes? Or how did similarities affect them?

Several interviewees, particularly the nonmentors, reported that they were explicitly looking for similar partners. Specifically, they wanted to work with cities that faced similar adaptation challenges and grappled with similar questions.

Of the sample studied, Bologna (mentor), Lleida (learner), and Terra di Leuca (learner) formed the twinning with the most heterogeneous partners. Lleida stated that, while the learning exchange with Bologna was fruitful for them, there was no learning or even knowledge exchange between them and Terra di Leuca. Lleida argued that this was mainly because they could not find any common ground. Lleida is a small city in the foothills of the Pyrenees while Terra di Leuca is a union of very small towns and villages located in a rural and touristic coastal area. Conversely, Lleida felt that they could learn a lot from Bologna, a partner they felt was more similar (as noted previously). As a municipality, Bologna is just as different from Terra di Leuca as Lleida. However, since having to comply with the same national (administrative) rules, their exchange was described as helpful for both sides.

Bremen highlighted that it is generally important to understand the context of the partner (e.g., institutional settings, decision-making processes, financial situation, funding landscape). However, especially at the international level, this is quite time-consuming. Moreover, the greater the contextual difference, the longer it takes to understand the other city’s situation. EUROCITIES pointed out that adding the return visit to the program was also done to give the partners more time to understand each other’s context and to reflect on it between visits.

A look at the geographic distribution of the partners (see Figure 1) reveals that the vast majority of the twinnings were either between cities from Northern/Western Europe or Southern Europe. There was not one single twinning between partners from the North and South. In that context, EUROCITIES remarked that it was not their particular goal not to organize North–South twinnings but it was the result of the evaluation of the received applications. For instance, the adaptation challenges proved to be similar among the Southern cities on one hand and the Northern cities on the other hand. However, EUROCITIES also reported that they were aware of some major institutional and legislative differences between Northern and Southern European countries that worked against those twinnings.

Could the Study Visits Lead to Policy Adoption?

The adoption of measures or policies studied during the visits could be seen as one, probably the key factor to determine the “success” of a twinning. Indeed, EUROCITIES states that policy adoption is a key program goal. EUROCITIES reported that the EC specifically requested that the program should be designed to focus on increasing the likelihood that the studied measures or policies could be applied reciprocally to learning and peer-learning cities. As described earlier, return visits were added as a compulsory part of the program. Through this, EUROCITIES hopes to trigger policy adoption.

For all cases where policy adoption happened or is in the discussion phase, the interviewees confirmed that referring to the visits helped a lot to increase the credibility of the project within their municipal administrations. In this context, Bremen invoked the adage of the “prophet” who was ignored in his own country. Consequently, demonstrating that one has looked outside of the box could help to persuade the municipal administration.

Bratislava, Bremen, and Ilion remarked that they took back several suggestions and ideas that were discussed and evaluated after the visit. For instance, Bratislava’s adaptation action plan was inspired and enriched by the common discussion with Arnhem and Bremen. However, there were also two very concrete cases of policy adoption in Lleida and Antwerp that are presented in more detail below.

The interviewees from Lleida reported that the urban forest project they were currently implementing was directly inspired by the visit to Bologna. Through a visit to the Urban Center Bologna, Lleida’s delegation was introduced to Bologna’s social urban gardening project and its citizen involvement approach. They picked up the idea and adapted it to their local conditions (conceptual and technical learning). Lleida is currently redesigning municipal green spaces to establish urban forests together with residents who are planting trees and who will be involved in the future preservation of the forest. To kick off the project, Lleida’s concern was to find and motivate the “right” people to go ahead. Therefore, after the twinning, Lleida requested specific information, mainly on the instrument of stakeholder engagement (technical learning and social learning), from the Urban Center Bologna. The project was discussed with the responsible deputy mayor for environmental affairs and won approval from the city council. Ultimately, Lleida’s environmental department’s visit to Bologna led to the realization of a project that included other municipal departments and municipal employees (e.g., engineers and architects).

The second example of post–twinning policy adoption was in Antwerp. The current development of the municipal water plan and upcoming major infrastructural works allowed Antwerp’s climate adaptation unit to actively co-design a large-scale urban redevelopment project. To better adapt to extreme cloudburst events, they saw that they needed to increase the sewage capacity, and decided to do this mainly above ground.

Copenhagen’s internationally renowned cloudburst management concept, part of its adaptation strategy, served as the model for Antwerp. This concept was totally new in Belgium. In collaboration with numerous stakeholders (e.g., municipal companies, architecture, and landscape architecture firms), Copenhagen explained and illustrated the cloudburst plan to the visitors from Antwerp (conceptual and technical learning). Of particular interest for Antwerp were emergency water exits and buffers. For flood protection reasons, emergency buffers were of great importance for Antwerp’s ringway area, one of the city’s lowest areas. The plan was to make the buffers greener and design them as a large emergency pool that increases the water absorption capacity to prevent or reduce damage in the inhabited areas of the city.

Antwerp stated that it was Copenhagen’s idea that was being adopted rather than specific technical aspects of it. However, to find the best solution for Antwerp, Copenhagen convinced them to use the same (quite expensive) runoff model. The model was the basis for Antwerp’s cost-benefit analysis and the risk analysis. To achieve this, a hydrologist from Antwerp was connected with the responsible architects in Copenhagen after the twinning to exchange knowledge regarding the modeling.

In summary, among the cases studied, policy adoption only happened when the adopting city had specific questions in mind, and when it was matched with a peer-city that could contribute to answering them, and which drew on its own experiences with already implemented similar measures. The case of Arnhem showed that having clear learning objectives and questions were not enough if the city was not matched with the “right” partners. In both cases where policy adoption happened, a mix of conceptual and technical learning—and in Lleida also social learning—could be observed. While sole conceptual learning, as in Ilion and Bremen, could provide valuable input, for example, for the strategy development process, it did not lead to immediate policy adoption.

Discussion

By interviewing participants of the Twinning Program that was mainly organized by the European city networks EUROCITIES and the CoM, this article examined how local policy makers learn in the field of climate change adaption. Drawing on insights from the policy mobilities and policy learning literature, the main results of the interviews are discussed below.

Knowledge Brokerage and Learning Brokerage: Two Very Different Animals?

Borrás and Radaelli (2011) noted that international organizations engaging in the transfer of knowledge often have clear normative ideas of how policy makers should learn. To a certain extent, this can be confirmed for EUROCITIES and the CoM and how they designed the Twinning Program at the outset. Indeed, in the first twinning round, there was the strong assumption that a more advanced mentor should “teach” a learning city with less experience how to successfully adapt to climate change. However, later development of the program also revealed that the knowledge broker EUROCITIES did not adhere to assumptions in a dogmatic manner. The program was modified when there were new insights or evidence that some elements did not work as desired. This general observation was confirmed by the many and quite substantial changes made to the initial program within only two twinning rounds (e.g., shifting from mentoring to peer-learning exchanges, matching more similar partners, adding a compulsory return visit). It can, therefore, be noted that, as a whole, experimentation and trial and error played a much bigger role in the design and development of the program than normative a priori assumptions.

Whereas EUROCOTIES itself was in a permanent process of learning to modify and improve the program, little attention was paid to the learning processes among the twinning partners. Neither was there an a posteriori assessment of the possible learning processes, nor did EUROCITIES follow up on the post-twinning activities of the cities (e.g., if policy adoption had happened or not). Moreover, the program organizers did not do a deep dive into the very nature of learning. For instance, they did not distinguish different forms of learning, assess, or theorize how these might lead to different outcomes, or whether different forms of learning were suitable for certain kinds of cities or not. It became apparent that the focus of EUROCITIES was on matching partners that had sufficient common ground to collaborate. In line with Howlett, Mukherjee, and Koppenjan’s (2017) understanding of knowledge brokers, EUROCITIES linked groups that otherwise would have most probably remained disconnected. They provided the basis and helped organize the visits but the actual policy learning process was left to the local policy makers. The question arises whether EUROCITIES should have also acted as a learning broker. Alternatively, it could have called in an external learning broker to assist and consult the twinning partners. As it stands now, the Twinning Program focuses on matchmaking and exchanging policy (ideas) but not on organizing systematic policy learning exchanges.

How Should Learning Be Studied and Performed in the Future?

As stated beforehand, learning was neither specifically defined and intentionally triggered, nor was it systematically assessed by the program organizers. Furthermore, the literature on policy learning and much more, on policy mobilities provide relatively little guidance to researchers on how to study and assess policy learning exchanges. The suggestions of Kemp and Weehuizen (2005) to carefully look at the relationship between policy learning and policy change or adoption and to distinguish different forms of learning (conceptual, social, and technical) go in the right direction. Indeed, they help to systemize and order the interview data and give guidance on what to center the attention. However, this approach needs to be further specified. In particular, it remains unclear which variables should be examined when exploring the relations between policy learning and policy change or adoption.

The aim cannot be to develop a panacea framework providing universal guidance on how to undertake and study learning exchanges among local policy makers. The described significance of contextual particularities (see Fisher 2014; Stead and Pojani 2018) for successful learning exchanges strongly suggests that a silver bullet framework for learning among local policy makers cannot be developed. However, given the described urgency of cities to learn from each other to adapt to climate change, we need to gain a better understanding of how policy learning should be performed in the “best” way possible. Indeed, the high significance and potential of policy learning for practitioners cannot be overstated. For instance, what is urgently needed is more in-depth and consolidated evidence on the key obstacles and enhancers for learning exchanges among local policy makers.

Learning from Frontrunners, a Few People’s Story?

This article confirms Andersson’s (2015) and Fisher’s (2014) findings that progressive or frontrunner cities also organize study visits for place promotion. Policy tourism (see González, 2011) to frontrunners to learn about innovative policies received increasing attention from academics. However, in the field of adaptation, examinations of policy tourism are still missing. This article makes the first exploratory attempt to help close this research gap.

González’s (2011) and Montero’s (2017) observation that study visits can help create the legitimacy to initiate policy change can be confirmed as well. Furthermore, it was shown that study visits to a frontrunner could be successful, assuming that policy adoption is an indicator of success. However, for the cases examined, this can only be confirmed for visiting cities that were not too different from the host city. What is urgently needed to enhance the generalizability of the findings are more in-depth case studies covering cities that regularly organize study visits in the field of adaptation.

Moreover, the article shows that learning from international partners is more time-consuming than learning with national partners because the different institutional context needs to be understood first before assumptions on the adoption potential for the learner city can be made (see Stead and Pojani 2018). In addition, international study visits are more expensive (travel costs), usually require advanced English language skills, and places in funded programs are limited. Every city needs to adapt to climate change and natural disasters, but not every local policy maker can travel abroad to learn the techniques.

These points, taken together, raise serious doubts on the large-scale potential of learning from frontrunners. First, there are only a few frontrunners out there. Second, what this article shows and what previous studies have already suggested is that learning exchanges tend to be more successful if the contexts (institutional, cultural, size) of the involved partners are more similar to each other (see Fisher 2014; Stead and Pojani 2018). This means that those who learn from frontrunners are most likely local governments that are somewhat similar to them and that have already made some progress in adaptation. Consequently, even if successful learning took place, it was still learning within a somewhat “exclusive club.” Indeed, as pointed out by Montero (2017, p. 338), there are concerns that those tours could be or become “lobbying instruments in the hands of elite local actors.” An additional element of exclusivity is the lingua franca English that was a mandatory requirement for twinning participation. It can be assumed that this is also the case for most international study visits. While the interviewed twinning participants did not report that a lack of English skills was an obstacle, it certainly represents a selective criterion that excludes all cities that do not have English-speaking staff or staff that do not feel comfortable in communicating in English.

Can Learning from an Equal Become the New Gold Standard?

The interviews revealed that, from a learning perspective, the mentoring exchanges mostly represent a one-way traffic flow of knowledge from the mentor to the learner. However, the studied mentoring exchanges also proved that these could lead to policy adoption (in the learning city).

Neither do the findings confirm that peer-learning exchanges can trigger policy adoption nor do they suggest they are currently more successful than mentoring exchanges. Nevertheless, even if in their current form, mentoring exchanges are more effective (for one party), they simply cannot represent a comprehensive model for the future. It must be noted that EUROCITIES currently exclusively organizes peer-learning exchanges, also because they were unable to find new local governments willing to act as mentors. Indeed, as mentioned previously, every city needs to increase its capacity to deal with climate change and natural disasters but there is only a limited number of frontrunners to learn from. Consequently, it must be explored further how mutual learning exchanges between “more ordinary” cities can be improved or rethought to improve the learning outcomes and ensure that all involved partners learn.

The peer-learning exchanges that are currently organized by EUROCITIES include a compulsory return visit half a year after the initial visits. Therefore, a further examination of the newer round of peer-learning exchanges might lead to different results. In addition, the studied twinnings represented different cases, which resulted in reduced generalizability of the exploratory findings provided by this research. Consequently, the findings need to be tested and confirmed by additional case studies. Moreover, future investigations of study visits should not be limited to the CoM Twinning Program. Indeed, there are other learning programs without a clear mentor/learner hierarchy that could be studied. For instance, the United Nations Office for Disaster Risk Reduction (UNISDR) manages the so-called City-to-City Sharing Initiative. Another promising program to study is URBACT, an EU exchange and learning program to promote sustainable urbanism. Generally, an examination of these learning exchange programs can help to further advance our understanding of the significance of different contexts for learning and the adoption of urban climate policies as called upon by Fisher (2014) and Shefer (2019).

Conclusion

Drawing on interviews with key local stakeholders, this exploratory study examined and discussed study visits focusing on climate change adaptation policies. The article provides first insights on how local policy makers learn through study visits and contributes to a better understanding of an increasingly popular but still largely underresearched policy dissemination instrument.

The results show that most of the exchanges among local policy makers distinctively focused on gaining new, broader perspectives, for example, to receive inspiration for their own strategy development (conceptual learning) while some also included learning exchanges on specific instruments (technical learning). The latter learning was only the case when the learning city addressed very specific questions. Moreover, it is found that participating in a study visit can increase the credibility of the presented idea or policy within the administration of the visiting city. In addition, the findings confirm that study visits are often used as strategic instruments by frontrunners, for example, for place branding. Also, it is demonstrated that, under certain conditions, study visits can initiate policy adoption in a learning city. However, for the examined cases, this occurred only when the peer-cities were not too different from each other, particularly in terms of size and institutional context. Despite these generally positive findings, it must be noted that the research can only confirm substantial learning results, particularly policy adoption, within mentoring but not peer-learning exchanges. Furthermore, among the examined mentoring exchanges, this applied only to the learning cities.

Notwithstanding the above and taking into account that all cities need to adapt to climate change, learning exchanges based on the logic of relatively similar cities learning and improving together appears to be the only suitable large-scale solution for the future. There are so many more “ordinary” towns and cities compared with frontrunners. Indeed, the program organizers and knowledge brokers of the CoM Twinning Program were unable to find cities willing to act as mentors after just two years. Therefore, this article raises a call for more research and practical action on how to improve or even rethink learning exchanges for all participating parties and to provide solutions for more than just a handful of cities. To maximize the learning possibilities and outcomes of study visits, the organizers of existing exchanges and the designers of future exchanges are of key importance. Indeed, there is a need for a shift from mere matchmaking and knowledge brokerage to more holistic approaches that also take into account the different dimensions and dynamics of policy learning.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Wolfgang Haupt

Notes

Author Biography

Wolfgang Haupt is a research associate at Leibniz-Institute for Research on Society and Space in Erkner (Germany). He holds a bachelor’s and a master’s degree in geography from Dresden University of Technology and a PhD in urban studies from Gran Sasso Science Institute L’Aquila and Sant’Anna School of Advanced Studies Pisa. His research focuses on local climate governance, transnational municipal climate networks, and city-to-city-learning.

References

Andersson

Ida

. 2015. “Geographies of Place Branding: Researching through Small and Medium Sized Cities.” Stockholm: Stockholm Univ.

Baker

Tom

Tenemos

Cristina

. 2015. “Urban Policy Mobilities Research: Introduction to a Debate.” International Journal of Urban and Regional Research 39:824–27. doi:10.1111/1468-2427.12252.

Bansard

Jennifer

Pattberg

Philipp

Widerberg

Oscar

. 2017. “Cities to the Rescue? Assessing the Performance of Transnational Municipal Networks in Global Climate Governance.” International Environmental Agreements: Politics, Law and Economics 17:229–46. doi:10.1007/s10784-016-9318-9.

Bell

Stephen

Park

Alex

. 2006. “The problematic meta governance of networks: water reform in New South Wales.” Journal of Public Policy 26:63–83. doi: 10.1017/S0143814X06000432.

Biesbroek

Robbert G.

Swart

Rob J.

van der Knaap

Wim G. M.

2009. “The Mitigation–Adaptation Dichotomy and the Role of Spatial Planning.” Habitat International, 33:230–37. doi:10.1016/j.habitatint.2008.10.001.

Borrás

Susana

Radaelli

Claudio M.

2011. “The Politics of Governance Architectures: Creation, Change and Effects of the EU Lisbon Strategy.” Journal of European Public Policy 18:463–84. doi:10.1080/13501763.2011.560069.

Cook

Ian R.

Ward

Kevin

. 2011. “Trans-urban Networks of Learning, Mega-events and Policy Tourism: The Case of Manchester’s Commonwealth and Olympic Games Projects.” Urban Studies 48:2519–35. doi:10.1177/0042098011411941.

Cook

Ian R.

Ward

Kevin

. 2015. “Post-war Planning and Policy Tourism: The International Study Tours of the Town and Country Planning Association 1947-1961.” Planning Theory and Practice 16:184–205. doi:10.1080/14649357.2015.1027948.

Covenant of Mayors.n.d. “Covenant of Mayors for Climate and Energy.” https://www.covenantofmayors.eu.

10.

Dotti

Nicola F.

Spithoven

André

. 2017. “Spatial Perspectives on Knowledge Brokers: Evidence from Brussels.” Environment and Planning A 49: 2203–22. doi:10.1177/0308518X17724442.

11.

EUROCITIES. n.d. “Eurocities.” http://wsdomino.eurocities.eu/eurocities/home.

12.

Fisher

Susannah

. 2014. “Exploring Nascent Climate Policies in Indian Cities: A Role for Policy Mobilities? ” International Journal of Urban Sustainable Development 6:154–73. doi:10.1080/19463138.2014.892006.

13.

Gilardi

Fabrizio

Radaelli

Claudio M.

2012. “Governance and Learning.” In The Oxford Handbook of Governance, edited by Levi-Faur

Dāwid

, 155–168. Oxford: Oxford Univ. Press.

14.

González

Sara

. 2011. “Bilbao and Barcelona ‘in Motion.’ How Urban Regeneration ‘Models’ Travel and Mutate in the Global Flows of Policy Tourism.” Urban Studies 48:1397–1418. doi:10.1177/0042098010374510.

15.

Grafakos

Stelios

Viero

Giulia

Reckien

Diana

Trigg

Katie

Viguie

Vincent

Sudmant

Andrew

Graves

Catherine

Dawson

Richard

. 2020. “Integration of Mitigation and Adaptation in Urban Climate Change Action Plans in Europe: A Systematic Assessment.” Renewable & Sustainable Energy Reviews 121. doi:10.1016/j.rser.2019.109623.

16.

Haupt

Wolfgang

. 2018. “European Municipalities Engaging in Climate Change Mitigation and Adaptation Networks: Examining the Case of the Covenant of Mayors.” In Resilience-Oriented Urban Planning: Theoretical and Empirical Insights, edited by Yamagata

Yoshiki

Sharifi

Ayyoob

, 93–110. Cham: Springer.

17.

Haupt

Wolfgang

Chelleri

Lorenzo

van Herk

Sebastiaan

Zevenbergen

Chris

. 2019. “City-to-City Learning within Climate City Networks: Definition, Significance, and Challenges from a Global Perspective.” International Journal of Urban Sustainable Development. Published electronically November 21. doi:10.1080/19463138.2019.1691007.

18.

Haupt

Wolfgang

Coppola

Alessandro

. 2019. “Climate Governance in Transnational Municipal Networks: Advancing a Potential Agenda for Analysis and Typology.” International Journal of Urban Sustainable Development 11:123–40. doi:10.1080/19463138.2019.1583235.

19.

Hjerpe

Mattias

Storbjörk

Sofie

Alberth

Johan

. 2015. “There Is Nothing Political in It”: Triggers of Local Political Leaders’ Engagement in Climate Adaptation.” Local Environment 20:855–73. doi:10.1080/13549839.2013.872092.

20.

Homsy

George C.

Warner

Mildred E.

2015. “Cities and Sustainability: Polycentric Action and Multilevel Governance.” Urban Affairs Review 51:46–73. doi:10.1177/1078087414530545.

21.

Howlett

Michael

Mukherjee

Ishani

Koppenjan

Joop

. 2017. “Policy Learning and Policy Networks in Theory and Practice: The Role of Policy Brokers in the Indonesian Biodiesel Policy Network.” Policy and Society 36:233–50. doi:10.1080/14494035.2017.1321230.

22.

Jaffe

Adam B.

Newell

Richard G.

Stavins

Robert N.

2005. “Systematizing Policy Learning: From Monolith to Dimensions.” Ecological Economics 54:164–74. doi:10.1016/j.ecolecon.2004.12.027.

23.

Jokinen

Ari

Leino

Helena

Bäcklund

Pia

Laine

Markus

. 2018. “Strategic Planning Harnessing Urban Policy Mobilities: The Gradual Development of Local Sustainability Fix.” Journal of Environmental Policy & Planning 20:551–63. doi:10.1080/1523908X.2018.1454828.

24.

Kemp

René

Weehuizen

Rifka

. 2005. Policy Learning, What Does It Mean and How Can We Study It? Oslo: Publin.

25.

Kern

Kristine

. 2019. “Cities as Leaders in EU Multilevel Climate Governance: Embedded Upscaling of Local Experiments in Europe.” Environmental Politics 28:125–45. doi:10.1080/09644016.2019.1521979.

26.

Kern

Kristine

Koll

Claudia

Schophaus

Malte

. 2007. “The Diffusion of Local Agenda 21 in Germany: Comparing the German Federal States.” Environmental Politics 16:604–24. doi:10.1080/09644010701419139.

27.

Knockaert

Mirjam

Spithoven

André

. 2014. “Under Which Conditions do Technology Intermediaries Enhance Firms’ Innovation Speed?” Regional Studies 48:1–13. doi:10.1080/00343404.2012.708405.

28.

Kona

Albana

Melica

Giulia

Calvete

Silvia Rivas

Zancanella

Paola

Iancu

Andreea

Gabrielaitiene

Irena

Saheb

Yamina

Janssens-Manhout

Greet

Bertoldi

Paolo

. 2015. The Covenant of Mayors in Figures and Performance Indicators: 6-Year Assessment. Luxembourg: Publications Office of the European Union.

29.

Kulsum

Ahmed

Sánchez-Triana

Ernesto

. 2008. Strategic Environmental Assessment for Policies: An Instrument for Good Governance. Washington, DC: World Bank.

30.

Liang

. 2017. “Site Visits, Policy Learning, and the Diffusion of Policy Innovation: Evidence from Public Bicycle Programs in China.” Journal of Chinese Political Science 22:581–99. doi:10.1007/s11366-017-9498-3.

31.

McCann

Eugene

. 2011. “Urban Policy Mobilities and Global Circuits of Knowledge: Toward a Research Agenda.” Annals of the Association of American Geographers 101:107–30. doi:10.1080/00045608.2010.520219.

32.

McFarlane

Colin

. 2011. “The City as a Machine for Learning: Translocalism, Assemblage and Knowledge.” Transactions of the Institute of British Geographers 36:360–76. doi:10.1111/j.1475-5661.2011.00430.x.

33.

Meuser

Michael

Nagel

Ulrike

. 2009. “Das Experteninterview – konzeptionelle Grundlagen und methodische Anlage [The expert interview – conceptual foundations and methodological approach].” In Methoden der vergleichenden Politik- und Sozialwissenschaft [Methods of comparative political and social sciences], edited by Pickel

Susanne

Pickel

Gert

Lauth

Hans-Joachim

Jah

Detlef

, 465–479. Wiesbaden: Verlag Für Sozialwissenschaften.

34.

Montero

Sergio

. 2017. “Study Tours and Inter-city Policy Learning: Mobilizing Bogotá’s Transportation Policies in Guadalajara.” Environment and Planning A 49:332–50. doi:10.1177/0308518X16669353.

35.

Peck

Jamie

Theodore

Nik

. 2010. “Mobilizing Policy: Models, Methods, and Mutations.” Geoforum 41:169–74. doi:10.1016/j.geoforum.2010.01.002.

36.

Pilato

Giovanni

Sallu

Susannah M.

Gaworek-Michalczenia

Marta

. 2018. “Assessing the Integration of Climate Change and Development Strategies at Local Levels: Insights from Muheza District, Tanzania.” Sustainability 10:174. doi:10.3390/su10010174.

37.

Prater

Carla

Lindell

Michael K.

2000. “Politics of Hazard Mitigation.” Natural Hazards Review 1:73–82. doi:10.1061/(ASCE)1527-6988(2000)1:2(73).

38.

Shefer

Ira

. 2019. “Policy Transfer in City-to-City Cooperation: Implications for Urban Climate Governance Learning.” Journal of Environmental Policy & Planning 21:61–75. doi:10.1080/1523908X.2018.1562668.

39.

Sheldrick

Alistair

Evans

James

Schliwa

Gabriele

. 2016. “Policy Learning and Sustainable Urban Transitions: Mobilising Berlin’s Cycling Renaissance.” Urban Studies 54:2739–62. doi:10.1177/0042098016653889.

40.

Sovacool

Benjamin K.

Axsen

Jonn

Sorrell

Steve

. 2018. “Promoting Novelty, Rigor, and Style in Energy Social Science: Towards Codes of Practice for Appropriate Methods and Research Design.” Energy Research & Social Science 45:12–42. doi:10.1016/j.erss.2018.07.007.

41.

Stead

Dominic

Pojani

Dorina

. 2018. “Dutch Experiences of Policy Mobility in Urban Planning.” In Knowledge, Policymaking and Learning for European Cities and Regions from Research to Practice, edited by Dotti

Nicola F.

, 123–34. Cheltenham: Edward Elgar.

42.

Vinke-de Kruijf

Joanne

Pahl-Wostl

Claudia

. 2016. “A Multi-level Perspective on Learning about Climate Change Adaptation through International Cooperation.” Environmental Science and Policy 66:242–49. doi:10.1016/j.envsci.2016.07.004.

43.

Wang

Riu

. 2012. “Adopting Local Climate Policies: What Have California Cities Done and Why?” Urban Affairs Review 49:593–613. doi:10.1177/107808741246934.8.

44.

Wood

Astrid

. 2016. “Tracing Policy Movements: Methods for Studying Learning and Policy Circulation.” Environment and Planning A 48:391–406. doi:10.1177/0308518X15605329.

45.

Yeganeh

Armin

McCoy

Andrew P.

Schenk

Todd

. 2019. “Determinants of Climate Change Policy Adoption: A Meta-analysis.” Urban Climate 31. doi:10.1016/j.uclim.2019.100547.