The domestic dimension of cross-border governance: Patterns of coordination and cooperation

Contemporary debates in border studies tend to see the national level as a rather residual category. There are, however, strong arguments that the national level still plays an important role for the governance of border regions. The question is what kind of formats are in place related to coordination and networking of border regions within national settings. The aim of this paper is to provide an exploratory typology of domestic governance settings. The underlying empirical study shows a large diversity of governance patterns in thirteen European countries. A synthetic typology reveals five approaches, namelycentralist,decentraland selectivecoordination as well as transnational and laisser-faire approaches. On this basis, the explanatory factors of the revealed governance patterns are discussed. The state structure (federal, centralist) plays an important role. Moreover, contingent political strategies, embedded in political culture and path dependency, help to understand the differences and the similarities between the analysed countries. The domestic governance dimension seems to bridge the frictions between the rather classical in-ner-state hierarchies and the soft networking tools of cross-border governance in the strict sense.


Introduction
The debate on cross-border governance is intense both in academic reflection (e.g. Coman 2019) and in political practice (e.g. Verschelde 2019). However, the focus is mostly on cross-border issues. The relations amongst actors from border regions within a given country tend to be a blind spot. The paper at hand contributes to this aspect by exploring intra-national coordination patterns, largely relying on network elements. The study investigates the relations between public institutions and actors mandated with border-related tasks within nation states. In contrast to existing studies of cross-border governance, relations across the border are not the object of analysis.
In recent years, the domestic and national dimension of border region development has hardly been addressed (Agnew 2008;Bürkner 2019). Nevertheless, the national level still plays a relevant role in various policy arenas with a potentially high cross-border dimension, including metropolitan policies (Sohn/Reitel 2016) or health policy (Svensson 2017). On the polity level, the debate on secondary foreign policy reflects the extent to which border region development is (an often implicit) part of national foreign policy (Klatt/Wassenberg 2017).
If we accept that multi-level governance still depends on domestic elements, the linkages amongst the domestic players have to be taken seriously, also concerning border issues. This is relevant for the analytical perspective, as the inclusion of domestic arrangements provides a more complete understanding of governance in border regions. Furthermore, the relevance is high from the normative perspective. Depending on the ways in which domestic coordination is organised, there is a large potential for learning processes amongst the border regions of a given state (Moyson/Scholten/Weible 2017). Programming procedures for funding programmes and addressing juridical misfits across borders are just two examples of daily challenges where exchange amongst border regions of a given national affiliation can be helpful. On the meta-level, the exchange between several domestic networks of border regions can also play an important role.
The recent crises have shown that governance patterns in border regions are still not very strong or resilient. The partial closure of borders due to the migration flows in 2015 and the Covid pandemic since 2020 led to enormous political complications, serious difficulties for the border regions' citizens and economic challenges (Opiłowska 2021). One of the reasons for these difficulties is the particular setting of European border regions, which are the "contact zones" of different national political and administrative systems. Even after half a century of European integration, the constellations tend to be complex and sometimes fragile.
A better understanding of the domestic governance settings fills a gap in border studies.
The aim of this paper is to provide an exploratory overview of national coordination and the networking patterns of border regions. The question is what role do coordination patterns and networks play on the domestic level. The empirical material is based on a research project on behalf of the German Ministry of the Interior (Chilla 2020). I analyse the modes of domestic coordination of border regions in Germany's neighbouring countries and three complementary countries. The analysis focuses on the predominant level of coordination and the intensity of coordination. The notion of coordination is understood in a broader sense, capturing different modes including hierarchic, formal mechanisms, and informal cooperation and networking processes. Mainly based on 39 expert interviews, this paper develops a typology of coordination patterns that shows the broad range of patterns in place. It concludes by reflecting on the results from the conceptual and normative perspective and with regard to further research needs.
2 Conceptual frame: the domestic dimension in studying border regions 2.1 Border regions in a multi-scalar context Border regions are rather "new regions in a changed international environment" (Plangger 2019: 158). They are positioned at the interface of manifold scalar politics (Johnson/Jones/ Paasi et al. 2011;MacKinnon 2011). The scaling debate stresses that the dominant role of the national state, as it was known in the post-war decades, has been proliferated and relativised due to processes of globalisation and internationalisation (Harrison 2013: 58). The multi-level governance approach is an important concept in this context. It reflects on the "changing relationships between the supranational, the national and the subnational level and between the public and the private sector" (Plangger 2019: 158). The rising EU regional policy was one of the main inspirations in this discourse (Hooghe/Marks 2001), and border regions played an important role in this debate (Perkmann 2007). New cooperation programmes and institutional routines offered an important opportunity for border regions to directly get in contact with the supranational level and to implement a range of funded projects (e.g. Harguindéguy/Bray 2009). A series of cooperation formats was established. The "classical" format of Euregios (or Euroregions) exist in parallel to further cooperation formats (Evrard/Engl 2018;Guillermo Ramírez 2018). In most border regions, different formats of cooperation overlap, in the territorial and the institutional sense. These soft spaces show a high variability as they can harden and weaken again, and they can disappear (Allmendinger/Chilla/Sielker 2014;Pupier 2020;Kaucic/Sohn 2021). A prominent geography of border regions is the European funding programme of INTERREG A (i.e. the cross-border cooperation strand of territorial cooperation). This programme does not only reflect the difference in national approaches to eligibility areas (from rather large-scale approaches in Scandinavia to a much smaller focus in countries like Germany). The programme also reflects territorial dynamics as the funding landscape may change from one funding period to the next.
Despite these considerable dynamics, the cross-border institutionalisations and their political mandates remain rather soft. Many political issues like health policy or spatial planning (Fricke 2015) remain predominantly in the hands of national authorities. On the regional level, the political, administrative and diplomatic frictions have to bridge across nation-state borders for which no simple solution applies. Adequate cooperation formats have to be developed for each institutional and actor setting, helping to bridge the "multilevel mismatch" along the borders (Chilla/Evrard/Schulz 2012).
In parallel, border regions are still an important object of national political agendas. The "secondary foreign policy" is an important example that reflects on border regions as objects and tools of national geopolitical strategies (Klatt/ Wassenberg 2017). The establishment of cooperation formats in border areas can be associated with concrete political interests over time. Thus after World War 2 -especially on Germany's western borders -the issue of peacekeeping and reconciliation was essential. After the opening of the "Iron Curtain", the stabilisation of the new geopolitical setting and cohesion processes played an important role. More recently, the crises connected to the migration flows of 2015 and the pandemic since 2020 have shown that issues of national security manifest at European internal borders. These examples show that the national level obviously matters for border regions. However, academic reflection tends to see the national level as a rather residual player in a complex and dynamic setting and, thus, underestimates its role.
From a normative and practical point of view, this aspect is of high relevance. Cross-border cooperation has to carefully develop specific solutions. The key question is: "How to find the right degree of institutionalisation and the appropriate legal form for different cross-border tasks by developing a good balance between open network and classical organizational approaches whenever structuring the crossborder working context; how to avoid both institutional sclerosis and informal/individual arbitrariness?" (Beck 2019: 18). This perspective necessarily has to also consider the national dimension.

Vertical and horizontal governance elements
The underlying dilemma of political coordination in border regions is the following: Whereas domestic institutions of a given territory, in principle, can be coordinated in a classical "top-down" way, the "foreign" actors of a border region are beyond the scope of hierarchical coordination. This situation leads to a multiplicity of governance arrangements. The debate on cross-border governance captures governance as a form of political management alongside hierarchy, primarily focusing on networks (Nienaber/Wille 2020). This network concept refers to an open setting of actors that can include a number of institutions and representatives in addition to those "actually responsible". The territorial dimension of networks is not dominant at first glance as the networks "are based on vertical, horizontal, as well as lateral networks" (Zumbusch/Scherer 2015: 502). However, the co-operative arrangements tend to refer to the specific and "well-defined spatial reference framework" (Zumbusch/ Scherer 2015: 502).
The diversity of structures and the limited governance capacity across the border means that actors from very different contexts have to face multiple challenges. In the earlier years of cooperation programmes, it could be said that one "national authority governs the development on one side of the border, another the other side and the general framework is set by the EU" (Jauhiainen 2002: 163). More recently, the linkages between networks and questions of power and mandates have been put at the forefront. Networks might even come along with a certain transfer of power away from the national level (Plangger 2019: 159) and might challenge national representation and political coordination (Frątczak-Müller/Mielczarek-Żejmo 2020: 10). Cross-border networks are considered to be "territorial development assets" (Berzi 2017(Berzi : 1584 and network capacity is considered to be crucial for cross-border functioning and its diversities (Nienaber/Wille 2020: 3). They are often perceived as social networks that allow personal contacts and trust to be built beyond the border and, at the same time, that play relevant roles for institutions of representative democracies (Frątczak-Müller/Mielczarek-Żejmo 2020: 9).
Even if cross-border cooperation, strongly based on INTERREG funding, is often seen as a success story of Europeanisation, the relevance of national systems must not be overlooked. Many border-related problems are rooted in differing national legislations, and problem-solving capacity is also linked to the national level (cp. Medeiros 2018). The implementation of cooperation programmes shows con-trasting implementation patterns depending on a number of factors, including the state systems involved. A series of studies shows the limited impact of cooperation dynamics, also due to long-standing historical settings, including the national policy frameworks (Colomb 2018;Noferini/Berzi/ Camonita et al. 2020).
A range of arguments from classical multi-level governance debates are of great relevance in this context. Governance throughout the multi-level system comes along with "a shift in the mode of interaction from power and control to information exchange, communication and persuasion" (Benz/Eberlein 1999: 343). The intensive processes of communication and bargaining can be seen as "'embedded games', in which policy-making in one arena sets the context for negotiations in other arenas" (Benz/Eberlein 1999: 343). Empirical studies have shown that subnational authorities engage in dynamics of Europeanisation due to rational strategic reflections, also in fields beyond the explicit political mandates (e.g. Huggins 2018). A proactive position in governance dynamics tends to be part of a strategy that aims at maximising opportunities. A multitude of factors influences the success of multi-level engagements. These are not predominantly related to structural characteristics, but are linked to political context: "many domestic factors, such as embeddedness, party politics and interest compatibility

DomesƟc governance of border regions
Municipal level Sub-region (e.g. canton) Region (e.g. federal state) NaƟon state European Union naƟonal border are crucial to understanding the strategies and channels employed by territorial actors" (Callanan/Tatham 2014: 206). These factors result in stronger and weaker players concerning financial and regulatory ambitions. However, these new regional configurations are "not serving to replace inherited landscapes of territorially embedded state scalar organization but emerge alongside" (Harrison/Growe 2014: 37).

The domestic side of cross-border dynamics
The literature review shows a rather dense discourse on spatialities, governance and networks in the cross-border dimension. However, the national, domestic dimension of the dynamics is addressed rather as a residual category. The national level is often seen as the defender of hierarchical, territorial authority: a rather traditional power visà-vis a multiplicity of multifaceted networks, and often as a black box. By focusing on domestic governance, the present study adopts a complementary perspective (Fig. 1). This study focuses on linkages between public institutions and actors mandated with border-related tasks within nation states, excluding relations across the border are not the object of the analysis. The relevance of the issue is obvious. Firstly, the analytical relevance lies in a better understanding of governance processes and patterns along and across borders. If we accept that the national domestic mandate remains crucial and if we consider the high relevance of governance approaches for border regions, then a better understanding of the "domestic side" of the process is important.
Secondly, from the normative perspective, the domestic dimension is likely to bear potentials for learning processes (within and across countries) for improved governance arrangements. The potentials of effective policy coordination to strengthen cohesion processes are more than plausiblein particular, as border regions are often rather weak in socio-economic terms.
Against this background, the study at hand provides an exploratory overview of intranational coordination and networks with regard to border-related governance. The wording has to remain flexible: top-down approaches can best be captured as "coordinating" whereas bottom-up approaches can rather be described as networks or cooperative settings. The aim of the study is to reveal the national patterns, the relevance and the explanatory factors for the settings.

Methodology
Due to the novel character of the topic, the study is of exploratory character. A qualitative approach ensures a multifaceted picture and aims to reveal the relevant patterns and processes. The study focuses on the domestic coordination patterns of thirteen countries, concretely of Germany (DE), its nine neighbouring countries (AT, BE, CH, CZ, DK, FR, LU, NL, PL) and three complementary countries (HU, IT, SE). The focus on Germany and its neighbouring countries is rooted in the underlying research project, commissioned by the German Ministry of the Interior. The three additional countries were involved in order to cover a broad range of policy styles and planning cultures across Europe, including Eastern European, Germanic, Napoleonic and Scandinavian cases (Knieling/Othengrafen 2009).
The data gathering is based on desktop research and expert interviewing. The desktop research explored the relevant information in scientific and grey literature. The empirical core of the study consists of expert interviews. For each country, at least two experts were interviewed, and a total of 39 expert interviews were conducted in autumn 2020, covering 13 countries and complemented by the EU Commission's perspective. The expertise was linked to the institutional position and personal experience, i.e. in most cases to affiliations to national or regional ministries and/or cross-border cooperation formats. The expertise of the involved persons is based on high-level, strategic knowledge from longstanding operative involvement in the interactions of border regions in a given state. None of the experts is a political representative, but most are civil servants or public employees. The identification of the experts was based on visibility in relevant publications (brochures and websites of national border-related politics) and on snowball elements throughout the interview campaign. A few of these experts are assigned to two countries due to their high familiarity with more than one national system. The interviews were conducted in the form of phone or video calls, mostly in the English language (some in German and French).
The main purpose of the interviews was to understand the formats and actor constellations of domestic coordination patterns and involved networks. This approach can be categorised as a "systematizing expert interview" with exploratory objectives (Döringer 2021: 265). The interviews followed a rather flexible outline, mostly starting with the question "When do you meet representatives from other border regions of your country, and how well do you know them? Who invites?". The interviews as such operationalised networking/coordination patterns (including formal networks and procedures, informal routines, parliamentarian networks, programming procedures of funding programmes), resources, understanding of border regions and practical issues. The temporal focus was on the contemporary situation, even if historical arguments were sometimes of relevance for understanding the formation of the coordination patterns. The pandemic situation was not a focus of the interviews.
The data interpretation followed a qualitative approach, starting with the categorisation of the information and further interpretation in order to reveal the relevant patterns and processes (qualitative content analysis; cf. Mayring 2014). More concretely speaking, the experts are key agents for one (or sometimes two) national contexts (Döringer 2021: 270). Their explicit information allowed a structured picture on networking and coordination of border regions with the respective national frameworks. The formal outcomes of this step were country profiles. In some cases, these country profiles underwent feedback loops with the experts.
In a second step, the comparative interpretation allowed the thirteen cases to be positioned in relation to each other and finally to be summarised in country groups. The comparative qualitative interpretation calibrated the information in the form of a two-dimensional graphical heuristic (cp. Rubinson 2019; Fig. 2). The two dimensions capture the following characteristics: The "level of coordination and networking" addresses the question of which level is the most influential one in the networking and coordination dynamics. The axis in the graphic goes from "low" for the local level over regional and national to "very high" for the transnational level. The positioning of each case on this axis reflects the arguments like organisational mandates (e.g. who invites and sets the agenda) and who has resources (personnel, budgets for the activities). The "intensity of coordination and networking" addresses the interactions involving several domestic border-related actors. More concretely, it reflects the frequency of the communications and meetings, and the political range and relevance of the addressed topics.
The position of each country results from the relative values on each axis, in particular relative to the countries adjacent in the graphic. Several rounds of comparing the above-mentioned indicators led to the positioning of the countries as "higher" or "lower" with regard to other countries. The country groups bring together countries that have similar values on both axes. By visualising the results in a condensed manner, the interpretation applies the "institutional mapping" perspective (cp. Chilla/Lambracht 2022: 4).
Even if some interviewees commented on what seems to function well or not, the ambition was not to evaluate the coordination patterns. The following section presents five country groups that position all countries of the sample visualised in Fig. 2.  Fig. 2 visualises the results in a condensed manner. The countries are categorised in the five groups described above, based on the most important similarities of domestic coordination and networking. The x-axis captures the intensity of intra-state interaction, reflecting the frequency of meetings and number of bodies, strategic orientation, resourcing, etc. The y-axis illustrates the level at which networking and coordination is primarily anchored. This reflects the relationship between the more nationally coordinated networks and the more regional, bottom-up networks.
There is a certain correlation between the intensity of networking and anchoring at a central level: the stronger the national role, the higher the intensity of networking tends to be, even if there is no strict proportional link. For example, in Switzerland and in Austria, networking is more intensive than in a number of non-federal states.
The size of the squares indicates the number of inhabitants for each country. One might assume that increasing country size might be accompanied by a higher number of border-region actors, calling for more coordination. However, the displayed pattern does not show a relationship between country size and coordination patterns. Other explanatory factors apparently play a stronger role. Moreover, differences between Eastern and Western European countries can hardly be identified. Even if central forms of organisation can be found in the three Eastern European states of the sample, their positioning differs to such an extent that a categorisation according to East and West is inappropriate.
The following sections go into more detail with regard to the country groups. The category "centralist coordination" comprises two national settings where a central state structure translates into border region coordination (France and Hungary). The similarity between the two systems is also because the key Hungarian institution was established by explicitly following the French model (Interview 4).
Both cases are based on strong institutionalisation and the rather solid national funding of a central institution. The central institution in France is the Mission Opérationnelle Transfrontalière (MOT), which has extensive experience in the policymaking and monitoring of border issues that stretches back about 25 years. This institution was formally established in 1997 and its members are primarily political representatives from the municipal and regional levels. The MOT was established at an inter-ministerial position, and the French Caisse des Dépôts is important for its financing, in addition to the ministries (especially of the Interior and of Foreign Affairs). Currently, the MOT has state funding for several employees (Interview 4). Its activities include multiple formats of support for border areas, both for improving the knowledge base and in advocacy (Peyrony/Denert 2012: 230;MOT 2018). Together with the Hungarian Central European Service for Cross-border Initiatives (CESCI) and the Association of European Border Regions (AEBR), the MOT is one of the three coordinated institutions in Europe that focus primarily on border regions (Interview 34). These three institutions are closely linked to each other, as is currently shown by the joint adoption of the so-called European Cross-border Citizens' Alliance, which is also supported by the Committee of the Regions. 1 In Hungary, CESCI plays the key role. It was founded in 2009 on the model of the French MOT. As a non-profit society, CESCI primarily has as members the representatives of the Hungarian European Groupings of Territorial Coop-eration (EGTC) as well as some Hungarian border towns or regions (and additionally the MOT). Although CESCI is not a formal state institution, the connection is close, especially since a significant proportion of the staff costs is currently borne by the Hungarian Ministry of Foreign Affairs (Interview 22). CESCI takes a rather operative role by conducting studies and organising events on a regular basis. The so-called "EGTC Forum" is a key element, and was located at the Ministry of Justice from 2010 to 2014 and since then at the Ministry of Foreign Affairs. This exchange focuses on learning processes amongst the actors from border areas. Representatives from various ministries, academia, etc. are involved. These meetings take place one to four times a year and are in strong demand (Interview 23). As a rather soft institution and formally a non-governmental actor, CESCI can build networks and address current topics in a less formalised way than is usual for ministerial activities (Interviews 22, 23).
In France, the great relevance of border questions for the national level is due to the intensity of cross-border commuting: several French regions show highly negative commuter balances (cf. Belkacem/Pigeron-Piroth 2020). 2 Given the high relevance of the topic, the national ambitions to address the situation have led to strong engagement via the MOT institution.
The centralist pattern in Hungary is linked to the prominence of the EGTC as this format always involves the national level. The Hungarian government decided in 2010 to position this legal form as the preferred kind of institutionalisation (Interview 23). This has to be seen against the background of the great heterogeneity of the Hungarian border areas and the complex geopolitical setting, for which a uniform legal format is seen as being helpful (Balogh/Pete 2018; Interview 22).
In both cases of "centralist coordination", the central institution is an expression of rather "high politics", with a more geopolitical facet in the Hungarian case and a stronger component of national cohesion policy in the French case.

Decentral coordination (PL, CH)
The country group of "decentral coordination" brings together the two very different countries of Poland and Switzerland. There are certainly many differences between these countries in socio-economic terms and state structure. Also the reasons for the strong role of the regional level differ: in Switzerland, rather constitutional factors ("kleine Außenpolitik") are relevant; in Poland, a bottom-up pro-cess in the involved regions drives the process. However, there are similarities in terms of coordination level and networking patterns.
Poland is structured in a centralised way where crossborder issues are traditionally linked to national foreign policy (Opiłowska 2017: 290). The national level can be actively involved in operative cross-border processes like in the development of the "Common Future Vision for the German-Polish Interaction Area". 3 Despite this general setting, strong bottom-up networking has evolved at the regional level. Since 1995, the so-called Forum of Euroregions has existed at this level, which was originally based on an annual, rather informal meeting in which border area problems were discussed and coordination processes initiated. Out of this informal exchange, a federation with its own legal personality was founded in 2012. The intention was to bundle certain concerns and to present a united voice towards the central authorities (Interviews 15,37).
The networking of the Swiss actors involved in border issues is based, firstly, on Article 56 of the Constitution 4 , which establishes Switzerland's "kleine Außenpolitik" (small-scale foreign policy). In the Swiss federal state, the 26 cantons have a considerable degree of autonomy. Article 56 includes the right of the cantons to conclude crossborder agreements largely on their own. Each canton has at least one "foreign relations officer" (Interview 9). The socalled network of cantonal foreign relations and European delegates meets twice a year, with the federal level also sending two representatives. These meetings are rather a form of networking than a form of coordination. The managing directors or other representatives of cross-border cooperation formats can be invited as guests to the meetings (Interviews 9, 10).
The so-called New Regional Policy (NRP) is the second basis for inner-Swiss networking, and has been implemented since 2008 in the area of spatial development. This strategy explicitly targets the border regions, in addition to the mountain regions and rural areas. The implementation process of the New Regional Policy is also based on regular meetings of the relevant experts, namely the "Conference of Cantonal NRP and Interreg Offices". The meetings include the annual assembly, the representative committee and thematic workshops, and bring together representatives of the Confederation, the cantonal New Regional Policy and Interreg agencies as well as state representatives (Interview 10).
Additionally, the Conference of Cantonal Governments has a certain importance: due to the fact that 14 of the 26 Swiss cantons are border cantons, topics concerning border areas are regularly discussed here. The political level of the cantonal governments is also divided into six regional cantonal conferences, which also (can) address border issues. For example, in north-west Switzerland, the Intercantonal Coordination Office at the Regio Basiliensis (ICRB) provides services on behalf of the cantons of Basel-Stadt, Basel-Landschaft, Aargau, Solothurn and Jura as a joint field office for cross-border cooperation on the Upper Rhine (Interview 11).

Transnational concertation (SE)
The category of "transnational concertation" comprises only one country of the sample (Sweden), even if we can assume that further Scandinavian countries might belong to this group. The level of interaction is not excessive, but the institutional level of activities is decisive.
The networking of border regions in Sweden is strongly mobilised by the Nordic Council, which is located on a transnational level (Hörnström/Smed Olsen/van Well 2012;Hörnström/Tepecik Diş 2013). Sweden belongs to those countries where cooperation is based on a series of so-called border committees ("Nordiska gränskommittéer"). These are funded by the Nordic Council, which is an intergovernmental cooperation of Denmark, Iceland, Norway, Sweden, Finland and other associated members that has existed since the 1970s.
In parallel to the financial support, political objectives play an important role here: issues of cross-border functioning have been continuously addressed in the Freedom of Movement Council since 2009 ("Gränshinderrådet"). In preparation for these regular meetings, coordination takes place at the national level, i.e. with the relevant ministry (Interviews 30, 31). These procedures are complemented by the usual tasks in the INTERREG A process on the national level. This focus on the transnational level does not mean a bypassing of the national level, but a clear political alignment beyond domestic procedures, systematically involving regional and national actors (Interview 31).

Selective coordination (AT, LU, NL)
The common ground of the cases Austria, Luxembourg and the Netherlands is that the coordination of activities is limited in an institutional and/or thematic way. The Austrian case is a clearly federal example with a certain form of coordination at the federal level with the Austrian Conference on Spatial Planning (ÖROK). Since 2011, the ÖROK has established the permanent "Working Group on Crossborder Cooperation" (AG CBC) in its office. Participants in this working group are the representatives of the various programme areas. In regular meetings, the topics of the IN-TERREG A programme (programme content, evaluation, interfaces with other programmes) are on the agenda (Interviews 19,20). However, this form of institutionalisation is limited to INTERREG issues and only links the federal states, not the Euregio level (Interviews 17, 18, 19).
In the Netherlands, the intra-state networking of actors takes place on two levels. At the national level, "hub offices" are funded by the Ministry of the Interior and the provinces. They take care of the concerns of the border areas and the networking of the participating provinces. The focus here is on the exchange of information on various problems at monthly meetings (Interviews 33,38). At the level of the border areas, networking meetings of the mayors from the northern border municipalities take place twice a year. In the southern border areas, networking is located on the level of the provinces.
In the Grand Duchy of Luxembourg, the labour market is characterised by maximum levels of cross-border commuting in Europe and beyond, linking four nation states (Germany, Belgium, France, Luxembourg). This comes with a great number of cross-border cooperation formats. Whereas national engagement has a cross-border dimension in many fields, the coordination of border regions plays a rather implicit role. This is partly due to the small size of the country that allows for informal coordination. The House of the Greater Region is home to nine crossborder formats of cooperation. It provides infrastructure that primarily serves to network and promote cross-border cooperation in general, but at least as a side effect also facilitates the networking of the Luxembourg actors and institutions (Interview 2). Furthermore, the so-called Summit of the Greater Region plays a key role as a large-scale, intergovernmental organisation frame. It ensures a wide range of coordination also on the national level (Interview 1).

Laisser-faire approaches (BE, CZ, DE, DK, IT)
A number of countries show a rather low level of innerstate networks or coordination. In these cases, there is no systematic procedure or format in place for the coordination of border regions.
In the Czech Republic, the linkages between the border regions concentrate on the formal procedures of the ministerial mandates, and on bilateral exchanges. An "Association of Euroregions of the Czech Republic" was established in 2002 to connect the Euregios in the Czech-German and Czech-Austrian border area (Interviews 6, 7). For several years, regular meetings of representatives of the Euregios were organised. However, this platform has not been active since 2010 (Interview 6). Only recently (December 2021) has a new initiative in the form of a memorandum been established, aiming for stronger interaction of the Euregios.
In Belgium, the number of cooperation formats is rather modest, despite the intensive functional integration across the border. The concerns of the border regions are mainly addressed at two levels, both formally coordinating the federal and regional actors: the Belgian concertation committee and the inter-ministerial level. Their practical relevance for border issues is rather low and they have a more reactive than strategic orientation (Interview 21). Apart from these formal possibilities of coordination, no continuous networking formats are in place. In practice, however, it should be mentioned that many cross-border issues come to the attention of the prominent representative of the Germanspeaking Community of Belgium, Karl-Heinz Lambertz (Interviews 3,21). He has continuously addressed cross-border issues in prominent political functions for decades (e.g. Lambertz 2010).
In the rather small state of Denmark, there are only three relevant cooperation formats established (Sønderjylland-Schleswig, Fehmarnbelt and Öresund). There is no formal structure for their networking, but a number of informal contacts and personal networks exist (Interviews 31, 16).
The situation in Germany is characterised by a strong federalism, which sees cross-border development as the object of regional mandates. The national role concentrates mainly on financing pilot projects (Interview 14) and short-term initiatives like ministerial conferences on specific topics (Interviews 13, 35), complemented by the extensive work of the Federal Institute for Research on Building, Urban Affairs and Spatial Development (BBSR). The Association of European Border Regions (AEBR) network of German border region representatives is an important bottom-up initiative. However, there is so far no systematic networking or coordination at the national level (Interviews 35,39).
In Italy, the coordination of regional issues in general is not strongly institutionalised, also due to the absence of a regions' chamber (Interviews 24,25). Moreover, this is also linked to the very diverse border areas (Alpine land borders, Mediterranean sea borders).
These settings all have in common the absence of a proactive, structured coordination or network of border regions. Individual networking activities in different constellations might play a certain role, but the domestic level as such remains a rather technical one.

The instrumental perspective
Technically speaking, coordination of and networking amongst border regions takes place in manifold ways. Fig. 3 categorises the relevant networking formats and instruments, illustrated with one or more examples in the second column. These examples are typical for the presented format and characteristic for the involved country. The rows are arranged from top to bottom, indicating the increasing degree of formalisation.
The figure mentions one coordination format that is not part of the presented country groups. In recent years, the European Commission has been considering the future of border region development with the so-called cross-border review process (Sielker 2018;Verschelde 2019). In this context, the idea of so-called cross-border coordination points in the EU states has emerged, and is still in the process of political coordination (European Commission 2018). The logic of this proposal is of general importance: the idea is to establish contact persons between the individual border regions and the European level, which could simplify processes in the sense of bundled communication. The interviewees had differing opinions on this proposal, even if the general idea of structured networking was appreciated (e.g. Interviews 34, 39).
Furthermore, all analysed states take part in programmes of European territorial cooperation, in particular the IN-TERREG A strand on the cross-border level. The obliga-

Informal networks AEBR German regions (DE)
Exchange formats (regular events) MeeƟng of mayors ( tory programming processes are an important basis for the networking of actors from the border areas. The form can be a purely administrative process across different levels, and it can also involve "hubs" such as the Austrian office of the cross-border working group. In all countries, there is more than one instrument established to support the coordination and networking of border regions. Informal networks, for example, play a certain role in all countries. However, in a "laisser-faire" country like Germany, they are key for normal functioning, due to a lack of more formalised coordination mechanisms (Interview 16). In contrast, institutions in a comprehensive sense are only present in Hungary and France.
The effort of formal institutionalisation appears worthwhile if it helps to pursue a specific purpose. This is particularly the case if access to resources is to be regulated or if a clear role in coordination processes is to be established. In the case of Poland, the initiative to establish a legal personality was decided after several years of informal networking.

Explanatory factors
The multi-faceted patterns raise the question of which factors can explain the differences and the similarities between countries. As indicated in Fig. 3, simple determinism does not provide an explanation. Obviously, structural elements are not dominant. Neither geographical location (Eastern vs. Western Europe) nor country size determines the policy options. State structure, however, seems to play a certain role.
The intensity of networking tends to be higher in centralist states, as the examples of France and Hungary show. However, the picture remains complex. Poland and the Czech Republic, both centralist countries, show large differences. The federal countries of Switzerland and Austria demonstrate higher networking dynamics than several non-federal states.
The country group of decentral coordination is particularly interesting, comprising states as different as Poland and Switzerland. The regional networking in the Polish case is the result of a bottom-up process of institutionalisation, whereas the Swiss case reflects a sophisticated system of federal multi-level balance. In both cases, the coordination and networking activities are intensive, and the regional level is important.
We can conclude that state structure makes certain coordination characteristics more probable, but does not determine them. The border-related patterns are characterised by contingency. Each setting aims to develop a best-fit approach, combining the most adequate tools and options. It obviously depends on the individual arrangements in each country and their political and administrative cultures. Over time, path dependencies develop.

The multi-scalar governance perspective
Border regions are not territorial blocks but they have to be understood in terms of socio-spatial dynamics defined by social practice and discourse (Paasi 1999). This study reveals a large variety of intra-national coordination formats that do not reveal "standard patterns" and for which no simple explanation applies. Border regions are positioned at the interface of manifold scalar politics (MacKinnon 2011) and our findings support the critiques of a one-side focus on levels beyond the national level (Harrison 2013: 57). National scale certainly matters and nation state structures provide relevant context, as we have seen with the example of federal elements, centralist embeddedness and transnational alignment. These contexts do not delimit contingency in general, but facilitate or hamper certain activities. The relevance of the national level also applies from the governance perspective. In the border regions' search for fitting institutionalisation (Beck 2019), the resources of the national level are clearly relevant. The fact that resources are rooted on the national level does not mean that they have to be "hard" juridical instruments, they also comprise soft tools. Some of the instruments (networks, conference formats) have softened or hardened over time (cp. Pupier 2020). Soft elements -be they cross-border elements or domestic ones -are located "in the shadow" of classical authorities.
From the network perspective, most coordination formats can be categorised as rather soft network formats that do not fundamentally question established structures. They supplement the functioning of the public institutions, and they are instrumental as they serve the implementation of the interests of classical institutions. The study sheds light on the evolutionary character of the patterns in place and on the relevance of path dependencies. This is well illustrated by the institutional hardening of the Polish network of border regions, the vanishing of the former Czech coordination initiative, and the prominent role of a key political person in Belgium. It remains an open question whether trends over time tend to lead to a higher degree of formalisation or whether there is rather a steady up and down of formalisation.
The domestic governance dimension is positioned between the rather classical inner-state hierarchies and the soft networking tools of cross-border governance. This sheds light on the challenges for border regions: they not only have to manage the misfits of administrative and political levels on either side of the border, they also have to bridge the frictions between clear domestic hierarchies and rather diffuse governance settings.
If cross-border networks are considered to be "territorial development assets" (Berzi 2017(Berzi : 1584, there is also potential on the domestic side. Further potentials can be seen in learning processes and improving communication towards the European level. The complexity of European cross-border dynamics throughout the multi-level governance system is enormous. A stronger focus on inner-state networks of border regions can help achieve a better-structured dialogue.
From a normative perspective the question arises as to what "good governance of border-region coordination" could look like. This study has neither evaluated the coordination formats, nor can we judge the impact on border region development. Nevertheless, we can derive at least some initial arguments. Central coordination and firm structures might support strategic impact by influencing communication on the European and high domestic level. These elements certainly can help improve knowledge on border-related topics for a broad audience -assuming that the available resources are appropriate. Laisser-faire and decentralism are a good basis for bottom-up approaches of high efficiency, assuming that the actor settings and the communication channels fit. If intra-national learning processes are the main objective, regular meetings of decentral or central coordination bodies are certainly a promising tool. As a result, the best coordination system depends on both the main objectives of the coordination and the setting in place.
These initial findings lead to a series of further research questions. Firstly, the exploratory character of the study might lead to a certain bias on "visible" interactions, i.e. focusing on official or institutionalised ways of networking and coordination. As governance settings are generally characterised by informal types of interaction, there might be other dynamics that have not been fully captured by this study. Scrutinising the informal elements in more depth is an important point for future analyses.
Secondly, the focus on the domestic side of the borderrelated governance has intentionally excluded "classical" cross-border governance as such. It seems important to analyse how the two sides of the coin -the domestic and the cross-border dimension -are related and intertwined.
Thirdly, and as to be expected for an exploratory study, there is scope for validation and detailing of the findings. This is true, just for example, for the more thorough inclusion of the coordination of sectoral networks. The study at hand focuses on general cooperation patterns and on some aspects of regional development and planning. It is obvious that these activities are complemented by a series of sectoral approaches: the EURES network addressing labour market issues for commuters is one of the more institutionalised examples. One might also mention networks for concrete transport policy objectives or for cross-border services of general interest (i.e. medical care, schools). These aspects have also to be left for further research.
Finally yet importantly, the impact on regional development along the borders is a key question that has to remain open. A systematic evaluation could explore which approaches help to reach what kind of objectives for the development of border regions.