Citizens in the City-Regions

Introduction

The role of the citizens in metropolitan governance has received insufficient attention, despite the fact that its political institutions ought to be democratically legitimate. Studies of metropolitan governance have tended to focus on the efficiency, effectiveness and competiveness of governing institutions, while the question of their democratic quality has been largely neglected (Kübler and Heinelt 2005). Because a growing number of people in all countries live in cities and surrounding areas, city-regions are becoming increasingly important as centers of social and economic activity (Rodriguez-Pose 2008) and also as political entities (Wood 1958; Herrschel and Newman 2002). As polities, city-regions might be organized in different ways for the purpose of handling collective matters. Independent of such variation, we need to improve our understanding of the role of citizens in city-regions.

In Western countries, governance in metropolitan areas is usually highly fragmented, with a large number of separate municipalities and weak or even nonexistent governing institutions for the city-region as a whole (Heinelt and Kübler 2005; Hoffmann-Martinot and Sellers 2007). However, citizens’ political orientations may vary. If citizens mainly have a parochial focus, that is, are concerned only with matters in their own local jurisdiction, then this is not likely to foster an understanding of matters that involve several municipalities or even the city-region as a whole. If metropolitan governance is to be relevant for those living there, then citizens should be concerned with conditions in the larger territory.

This citizen’s dimension of metropolitan governance has been only partly investigated. In a Swiss context, Kübler (2005) has detected an interest in politics across municipal boundaries in city-regions, although the nature and extent of interest varies among metropolitan areas. In previous studies, I have suggested that there is an intermunicipal political orientation in Swedish city-regions, but how it is manifested and what explains it is only partly understood (Lidström 2006, 2010). Many questions remain unanswered, in particular whether it is possible to identify different types of intermunicipal political orientations and, if so, whether these have distinct explanations.

The purpose of this article is to map out and analyze patterns of intermunicipal political orientations among citizens in two Swedish city-regions: Göteborg and Umeå. The main questions to be addressed are (1) in what ways and to what extent are citizens politically oriented toward other municipalities in the city-region, and (2) how can variation in such orientations be understood?

In the next section, the theoretical basis for the analysis is developed. This is followed by an empirical test of the expectations generated by the theory. Citizen surveys carried out during 2009-2010 in the city-regions of Göteborg and Umeå are used in this analysis. Finally, the main conclusions are summarized, and the implications for metropolitan democracy and governance are discussed.

The City-Region as a Demos

The concept of the demos usually refers to “the appropriate political community for democracy” (Saward 1998). It identifies those who constitute “the people” in a given setting, in contrast to those that are not included. In a multilevel democracy, several layers of demos correspond to different scales of collective needs, including local, regional, national, and international. According to a participatory conception of democracy, citizenship includes being politically oriented toward the demos, for example, by taking an interest in politics and public policies, participating in debates, and influencing public decision making (Pateman 1970; Lowndes 1995; Norris 2002).

There is no generally accepted principle for how the borders of a demos should be drawn. Indeed, democratic theory does not provide any guidelines about how to solve “the boundary problem,” that is, to decide which citizens to include in a given demos (Dahl 1970, 1989; Dahl and Tufte 1973). However, scholars have suggested a number of possible normative principles. Communitarians argue that each demos should correspond to a community in which people share common cultural traits, interact with each other, and have a collective identity. Sometimes boundaries themselves may create such communities (Briffault 1996). Liberals emphasize the rights of the individual and regard the demos as the territory in which one can exercise these rights (Lowndes 1995). This might include voting rights, but also the right to choose public services, for example, schools for one’s children or a residence facility for one’s elderly parents. Another principle is that a demos should be composed of all persons who are affected by particular collective decisions (Arrhenius 2005).

To what extent would city-regions qualify as demos? In most metropolitan areas in the democratic world, the basic unit for democratic decision making is the municipality. Most local citizenship rights, such as the right to vote and to use local public services, as well as duties like the obligation to pay local taxes, are linked to the municipality. Yet many of the collective needs in city-regions extend beyond municipal borders or are most efficiently dealt with on a larger scale than the municipality. These may include transportation needs, collection of refuse, sewage treatment, and a healthy environment. Indeed, because of the functional specialization that is characteristic of many city-regions and the separation of workplaces and residences, many citizens spend most of their day outside of their home municipality.

Hence, for many of its residents, the city-region might be a more relevant demos than the municipality, regardless of whether we use a communitarian conceptualization of demos, a liberal one, or the criterion that those affected by collective decisions should be part of the demos. In addition, there may still be a sense of identity connected to the region as a whole (compare Lowndes 1995; Kübler 2005; Lidström, Eklund, and Westin 2010). We would therefore expect that many metropolitan citizens have a territorial political orientation that extends beyond their own municipal borders.

Intermunicipal political orientations among citizens are likely to be beneficial for governance of the metropolis. However, the reasons for this may vary, depending on which normative model of metropolitan governance is favored. In the literature on metropolitan governance, it is common to distinguish among a consolidated, governance, and fragmented model (compare Kübler and Heinelt 2005; Hoffmann-Martinot and Sellers 2007; Savitch and Vogel 2009). The consolidated model favors a directly elected level of government for the city-region as a whole, either through amalgamations of municipalities or by establishing a second tier of regional government. Under this model, it is advantageous if citizens have an understanding of matters outside their own community, because this establishes a popular basis for prioritizing among different parts of the area, something that metropolitan-wide governments necessarily must do. To be legitimate in the eyes of the citizens, priorities need to be seen as fair and reasonable. Because metropolitan decision makers are directly elected and can be replaced by the citizens, the consolidated model has an in-built mechanism for making them democratically accountable.

The governance model (sometimes called the new regionalism model) is also a form of city-region-wide government. In this case, cooperation takes place between and among the municipalities rather than through directly elected politicians. The relationship between citizens and metropolitan governance is therefore more indirect. As a result, it should be expected that citizens’ orientations will focus largely on matters in their home municipalities. Of course, it is still advantageous if citizens have an understanding of the problems and policies that concern other municipalities and the metropolitan area as a whole, as this could increase the legitimacy of the cooperative arrangements. However, the model has no direct mechanism for making the metropolitan decision makers accountable to the citizens. Indeed, this is a general criticism of the governance model (Franzke et al. 2007).

The fragmented model favors competition among the municipalities in the metropolitan area. Each municipality sets its own priorities as regards type and quality of services and level of taxes. Citizens are expected to choose to live in a municipality that corresponds to their preferred mix of taxes and services. Although citizens might choose to actively participate in politics in order to influence decisions in their municipality, their main method of influence is to vote with their feet. This puts pressure on the municipalities and makes them sensitive to the preferences of their citizens. Proponents of this model regard it as the most efficient way of providing the municipal services that most closely correspond to the preferences of each individual citizen.

In the fragmented model, elected governments with metropolitan-wide functions are regarded with suspicion because they disturb competition between the municipalities and reduce the impact of mobility. In a fragmented metropolitan region, citizens are expected to be aware of conditions in other municipalities, although not because this facilitates taking part in developing or approving metropolitan-wide priorities. Rather, such knowledge serves as a basis for decisions about whether to abandon one’s own municipality in favor of a neighboring one that offers a mix of taxes and services that more closely correspond to the citizen’s own priorities.

Levels of Activism

Because all three models of governance benefit from a citizenry that has an intermunicipal political orientation, it is important to understand what explains variation in this orientation. Such knowledge might increase our understanding of the prerequisites for metropolitan governance as well as generate knowledge that can be used to develop policies aimed at enhancing citizen involvement in city-regional matters.

It has already been emphasized that political orientations in the city-region vary with regard to their scope. They may be parochial and focus on conditions in one’s own municipality, but they might also extend beyond municipal borders. We will be concerned mainly with activism directed toward matters in other municipalities in the city-region. Such an orientation is understood as being intermunicipal.

One can also conceive of an orientation that focuses on city-regional matters that are not linked to specific municipalities. Such matters may include overarching planning priorities, environmental problems, and the running of airports and ports. City-regionalism may also concern a general sense of belonging to the city-region or a sense of identity. The focus of this article is explicitly on intermunicipal political orientations. However, we would assume that citizens with an intermunicipal political orientation are also more likely to be city-regionally oriented in this more general sense.

The different types of activism may take different forms. In the literature on political participation, it is common to think in terms of a ladder of participation, where each step represents a gradually deeper form of political involvement (Arnstein 1969; Norris 2002). In a corresponding way, in my analysis of political activity I conceptualize different activities as steps on a ladder of activism. The steps range from softer to harder forms of activism:

If citizens are likely to vary in the extent to which they have an intermunicipal political orientation, the next question is how this variation can be understood. We will test three theories based on different ideas about what promotes citizen involvement in matters outside their own municipality. However, as the metaphor of a ladder implies that softer forms are expected to be used more frequently than harder ones, we will also discuss why citizens are more likely to choose certain forms rather than others.

The first set of assumptions about why intermunicipal political orientation varies suggests that this is simply a reflection of established patterns of public participation. Those who are likely to take part in political matters in general are also expected to do so on an intermunicipal scale. Theories of public participation usually identify specific conditions that facilitate or hinder citizen participation. The Civic Voluntarism Model has been particularly influential, not least because its predictions have been supported by numerous empirical studies (Seyd, Whiteley, and Pattie 2001; Oliver 2001; Rose 2002; Pattie, Seyd, and Whiteley 2003; Barkan 2004; Lidström 2006, 2010). Earlier theories suggested that differences in participation could be explained by socioeconomic conditions, but Verba, Schlozman, and Brady (1995) argued that such explanations were too limited. They claimed that inactivity in political life was attributable to a lack of resources (citizens being unable), to issues of engagement (citizens being uninterested) or a failure in recruitment (citizens not being asked to get involved). Therefore, resources, engagement, and recruitment were seen as factors that contribute to making citizens politically active. The question here is whether these factors are also relevant for explaining why citizens have different forms of intermunicipal orientations. The potential explanatory power of the Civic Voluntarism Model for this particular political behavior has not been studied previously.

The second theory emphasizes citizens’ integration into the city region and suggests that those who are highly integrated are also more likely to be intermunicipally oriented (Lidström 2006, 2010). Integration extends the territorial field of experience of the citizen and broadens one’s awareness of problems and possible solutions (Walks 2006). Such a position is enhanced by various forms of mobility such as commuting to shop and spending leisure time in other municipalities in the city-region. Indeed, such interaction is a key defining element of the city-region (Barr 2005). In addition, there may be personal interests at stake, for example, if one owns property or is a member of a local association in a different municipality. Commuting across municipal borders may also be regarded as creating a stakeholder position, as the workplace or place of study is located in a different municipality. City-regional integration may also derive from previous experiences from other parts of the city-region, for example, having been a resident of a different municipality.

Finally, the location of one’s home municipality within the city-region may affect citizens’ intermunicipal political orientations. Previous studies have emphasized the importance of context within metropolitan regions for public participation (Kelleher and Lowery 2004; Gainsborough 2005; Walks 2006). In polycentric as well as monocentric city-regions, the core(s) typically function as a center for economic and cultural activities (Barr 2005). Particularly in a Scandinavian setting, where the core city is not an area of deprivation, the city could be even more attractive (Lidström and Sellers 2011). The sheer importance of city politics and how it is covered by the media are likely to attract interest and activism among those living in the suburbs (compare Williamson 2002). Hence, we would expect that it is more likely that politics in dominant cities attracts the attention of citizens in the other municipalities than vice versa. This would be particularly significant in small municipalities in city-regions characterized by large differences in size between the smallest suburban municipalities and the core city. Furthermore, we would expect that intermunicipal political orientations are more extensive in larger city-regions. Apart from the effects of more extensive interaction, which was discussed above, we assume that the more dominant the role that a large city-region has in the lives of its citizens, the more likely it is that citizens will orient themselves to matters in other municipalities.

Although I have discussed these theories separately, there is no reason to believe that they are mutually exclusive. On the contrary, each may contribute to a fuller understanding of how intermunicipal orientations vary. Nevertheless, some of the explanatory factors might be more relevant than others.

In addition to these theories, we also need to understand why citizens choose different forms of activism. Why do some limit their engagement to taking an interest in news or politics in neighboring municipalities, whereas others attempt to influence opinions and policies? There are no previous studies that can be used as a starting point for developing hypotheses. However, we will assume that attempts to achieve change are linked to both a strong sense of personal engagement and material self-interests related to matters in other municipalities in the city-region. Such factors are already included in the other three theories under investigation. Hence, the engagement component in the Civic Voluntarism Model and integrating factors that emphasize a stakeholder role, for example, having a workplace or place of study in a different municipality, owning property, or being member of an association in another municipality are expected to be the major explanations for selecting harder forms of intermunicipal activism.

Method, Material, and Operationalization

City-regions may be conceptualized as one or a few major cities with a surrounding hinterland that is functionally linked to the cities through interconnected markets for labor, housing, education, shopping and leisure. As these city-regions are not formal jurisdictions, they lack fixed borders. Hence, there is no obvious basis on which to select city-regions for empirical studies.

A common way of identifying the limits of city-regions is through the extent of daily commuting that takes place between the municipalities (OECD 2002; Karlsson and Olsson 2006). However, there are no commonly agreed criteria about what level of commuting between municipalities is required for a certain area to be considered a city-region. In addition, commuting patterns vary over time and larger commuting areas tend to expand (Sancton 2008). A less fluid alternative is to define the city-region as the area around the city in which the municipalities regularly cooperate with each other about city-regional functions. These areas tend to correspond fairly well to commuting areas, but they are more stable.

The data used in this study were collected from two cases of city-regions of different sizes. They are both monocentric and consist of a dominant city surrounded by municipalities of various sizes. The larger of these is the Göteborg city-region on Sweden’s west coast. It is the country’s second largest city-region and includes 13 municipalities with a total of 920,000 inhabitants. The municipalities cooperate within the framework of the Göteborg Region. The other is Umeå city-region, the 11th largest in Sweden, located on the country’s northern east coast. It consists of six municipalities that cooperate in the Umeå Region and has a total population of 140,000. The locations of these city-regions and the municipalities that they include are shown on the maps in Appendix A.

In each city-region, samples of citizens between the ages of 18 and 84 were randomly selected and mailed a survey with questions covering various issues about citizenship in city-regions. The questionnaire was designed to take into account the specific conditions in the two city-regions, and different versions were sent to respondents in the core cities and in the surrounding municipalities. Of the 6,000 citizens surveyed, a total of 3,200 persons (60%) returned a completed questionnaire. Of the respondents, 55% are women, and on average they are 50 years of age. Furthermore, 76% come from the Göteborg region and 36 from the Umeå region. In the analysis, these have been given weights to correspond to the population sizes of the two regions. The analysis of nonrespondents indicates that women, the middle aged, and people who vote in general elections are slightly over-represented.

Intermunicipal and Parochial Activism

If intermunicipal political orientation is important for democratic governance in the city-regions, then it should not be a marginal phenomenon. We will start by exploring how common this orientation is in the two cases, using the previously mentioned ladder of activism. Figure 1 summarizes the overall extent of intermunicipal activism and compares it to parochial activism in the two cases.

OPEN IN VIEWER

Figure 1.

Overall extent of parochial and intermunicipal activism in the two cases (in percentages).

The figure clearly illustrates that although it is more common that people are concerned with matters in their own municipality, various forms of intermunicipal political activism are highly relevant. On average, almost 80% are interested in news about at least one other municipality in the city-region. A majority is interested in politics in one or many another municipalities. This seems to indicate fairly high levels of involvement across municipal borders, even if we take into account that citizen questionnaires of this kind tend to exaggerate levels of activism.

It is also noteworthy that the different forms of intermunicipal and parochial activism follow the expected ladder-like pattern. It is more common to be interested in news than in politics, and this is more common than conversations with others about politics, and in particular more common than attempts to influence other citizens or political or administrative decisions. Thus, also empirically, these various forms represent different degrees of activism.

Table 1 compares the occurrence of intermunicipally oriented activism in the two cases and between center and periphery.

OPEN IN VIEWER

Table 1.

Degree and territorial orientation of intermunicipal political activism among citizens in the Göteborg and Umeå city-regions (percentages)

	Umeå city-region		Göteborg city-region
Intermunicipal political activism:	City of Umeå	Other municipalities	City of Göteborg	Other municipalities	In total, both city-regions
Interested in news	50	88	72	93	78
Interested in politics	27	57	48	68	53
Have talked to other citizens about political matters	29	44	37	50	41
Have attempted to influence other citizens’ political views	10	15	13	20	15
Have attempted to influence political or administrative decisions	5	12	8	11	9

Note: For information about the different forms of political activism, see note under Figure 1.

Intermunicipal activism occurs more often in the Göteborg city-region. It is more common that residents of outer municipalities are politically oriented across municipal borders, and this is true for all forms of intermunicipal activism in both city-regions. Hence, where a person lives in the city-region is important for whether he or she has an intermunicipal political orientation. The main direction of these orientations is from the suburbs to the core city rather than vice versa. Whether this pattern holds once individual-level variables have been taken into account will be discussed below.

Testing the Theories

Why are some citizens more intermunicipally active than others? Is it merely a reflection of traditional patterns of public participation or are there mechanisms specific to the city-regions that enhance the activism of citizens across municipal borders? How important are differences in local contexts? Are there different explanations for different forms of intermunicipal activism?

These questions are addressed with the help of a set of logistic regression analyses. Because one of the aims is to investigate whether there are different patterns of explanations for each type of activism, the different forms of intermunicipal activism are used as separate dependent variables. The forms are empirically interrelated, but not to the extent that it is justifiable to combine them into a joint index. A principal components analysis indicates that there are two dimensions, which together account for 68% of the variation. The correlation coefficients (Pearson r _xy ) between the dependent variables vary from 0.099 (interest in news and attempts to influence political or administrative decisions) to 0.503 (interest in news and politics). The full correlation matrix is presented in Appendix B.

As outlined in the theory section, three sets of assumptions are used as broad guidelines to analyze variation. Public participation theory will be operationalized with the components emphasized by the Civic Voluntarism Model, that is, individual resources, engagement, and recruitment. City-regional integration will be represented by eight measures of different forms of integration. The position of a municipality in the city-regional context is captured by classifying it according to the size of the municipality (with the core municipality being the largest) as a core or peripheral municipality and by the size of the city-region (i.e., using a dummy variable distinguishing Göteborg from Umeå). Information about the construction of all variables can be found in Appendix C.

The results of the analyses are summarized in Table 2. Several variables representing both the Civic Voluntarism Model and city-regional integration theory are significantly related to each form of intermunicipal activism, which suggests that both of these theories are relevant. Among the civic voluntarism variables, two are consistently related to all forms of activism—interest in politics in one’s own municipality and number of organizational memberships. This suggests that there is no trade-off between an interest in local political matters and a city-regional political orientation. Furthermore, membership in voluntary organizations enhances all forms of intermunicipal activism, which emphasizes how crucial civil society involvement is for broadening one’s political mind. Many of the other engagement variables are also important.

OPEN IN VIEWER

Table 2.

Explaining variation in intermunicipal activism (logistic regression analyses)

	Level of intermunicipal activism
Independent variables:	Interested in news	Interested in politics	Have talked to other citizens about political matters	Have attempted to influence citizens’ political views	Have attempted to influence policy or administrative decision
Civic Voluntarism Model
Resources
Middle-aged	–0.057	0.151	–0.050	–0.249	–0.076
Level of education	0.225**	–0.088	0.316***	0.203*	0.229
Nonimmigrant	–0.102	–0.075	–0.418**	–0.215	0.202
Employed	0.212	–0.122	0.272*	0.088	–0.393*
Engagement
Satisfied with democracy	0.044	0.018	–0.230***	–0.670***	–0.219
Party identification	–0.016	0.059*	0.036	0.146***	0.161***
Interested in politics generally	–0.070	0.503***	0.831***	0.936***	0.562***
Interested in politics in own municipality	0.753***	1.701***	0.460***	0.335***	0.351**
Recruitment
Organisational membership	0.131***	0.133***	0.066*	0.099**	0.219***
City-regional integration
Working/studying	0.042	0.057	0.277***	0.231**	0.681***
Shopping	0.273*	0.370***	0.175*	0.469***	0.313*
Spending leisure time	0.331**	0.192	0.150	–0.002	0.151
Previously resided	0.516***	0.532***	0.214*	0.206	0.044
Have family/relatives	0.359**	0.013	0.194*	–0.058	0.316
Have close friends	0.351**	0.306**	0.733***	0.597***	0.294
Own property	0.520*	0.660***	0.274	0.384*	1.113***
Member of organization	–0.219	0.356	0.164	0.342*	0.486**
City-regional context
Size of municipality
Small outer municipality	1.120***	0.723***	0.265	0.206	0.050
Large outer municipality	1.417***	0.874***	0.259*	0.006	–0.159
Large city-region	0.782***	0.747***	–0.040	0.089	–0.140
Constant	–4.537***	–9.512***	–5.610***	–6.397***	–7.287***
Cox & Snell R2	0.225	0.414	0.256	0.169	0.117
Nagelkerke R2	0.352	0.553	0.344	0.290	0.260
N	2,739	2,755	2,792	2,792	2,792

Note: The coefficients are B values from the logistic regressions. The reference category for the two indicators of municipal size is belonging to the city municipality. Data has been weighted to compensate for differences in sample size between the city-regions and between the core cities and peripheral municipalities. Information about how the variables have been operationalized is summarized in Appendix C. Maximum intercorrelation (r _xy ) between the independent variables is 0.545 (interest in politics generally vs. in own municipality), which indicates that there are no multicolinearity problems.

*

p < 0.05; **p < 0.01; ***p < 0.001.

With the exception of level of education, most of the resource variables are not significant. This is interesting because the resources of citizens have usually been important in other studies of participation (Norris 2002; Oliver 2001). For example, in a study of political interest in neighboring municipalities in Swiss city-regions, age and level of education were significant (Kübler 2005). A remarkable result is that contrary to initial expectations, those who are dissatisfied with the system of democracy are more likely to talk to other citizens about political matters in other municipalities and to try to influence others’ views on such matters.

Among the different functions that contribute to city-regional integration, shopping in other municipalities is significantly related to all forms of intermunicipal activism. This is clearly surprising and could be interpreted as an unexpected spillover effect of shopping outside one’s own municipality. Perhaps less of a surprise, having close friends in another municipality also generally increases one’s political orientation toward other municipalities.

As suggested in the theory section, the variables representing engagement are significantly related to the harder forms of intermunicipal activism. In addition, commuters and members of organizations located outside one’s own municipality, but also to some extent those who own property in other municipalities, engage in harder forms, such as trying to influence citizens or political and administrative decision makers in other municipalities. Along with those already living in a particular municipality, these groups can be seen as the most important stakeholders in municipal politics. Those who have previously resided in, have family or relatives in, and/or spend leisure time in municipalities other than the one in which they live focus on softer forms—that is, interest in news and politics and discussions with other citizens about political matters in other municipalities. Thus, the findings suggest that the way in which citizens are integrated into the city-region has an impact on which form of intermunicipal political orientation they develop.

Finally, territorial context also contributes to explaining variation, but only with regard to softer forms of intermunicipal activism. Citizens who live outside the core city and in the larger city-region are more interested in news about neighboring municipalities and in politics in these municipalities. However, it is the outer position of the municipality, rather than its size, that influences whether or not citizens are likely to have an intermunicipal political orientation.

We have now concluded that to some extent all three theories are relevant as explanations for different forms of intermunicipal activism, although the city-regional context makes a difference only with regard to softer forms of orientations. To more explicitly assess the explanatory contribution of each theory, we can investigate how much of the total variability is accounted for by each theory’s variables. This is summarized in Table 3.

OPEN IN VIEWER

Table 3.

Accounting for variability

	Levels of intermunicipal activism
Theories	Interested in news	Interested in politics	Have talked to other citizens about political matters	Have attempted to influence citizens’ political views	Have attempted to influence pol or adm decision
The Civic Voluntarism Model
Cox & Snell R2	0.103	0.340	0.187	0.127	0.064
Nagelkerke R2	0.160	0.453	0.251	0.219	0.141
City-regional integration
Cox & Snell R2	0.123	0.101	0.094	0.052	0.058
Nagelkerke R2	0.191	0.135	0.127	0.090	0.128
City-regional context
Cox & Snell R2	0.108	0.067	0.022	0.010	0.005
Nagelkerke R2	0.166	0.090	0.030	0.017	0.011
The full model
Cox & Snell R2	0.225	0.414	0.256	0.169	0.117
Nagelkerke R2	0.352	0.553	0.344	0.290	0.260

Note: The numbers in the table are the values for Cox & Snell R² and Nagelkerke R² from logistic regression analyses performed for each of the three theories separately and for all three theories together. The variables included in each theory are presented in Table 2.

Cox & Snell R-squared and Nagelkerke R-squared are frequently used in logistic regression analysis as a means of summarizing the total amount of variability that is accounted for by the independent variables in relation to a dependent variable. The data in Table 3 can thus be seen as a rough estimate of the relative importance of the three theories. It turns out that the Civic Voluntarism Model is clearly the most powerful one, with the exception of interest in news about other municipalities, where city-regional integration is more important. The data suggest that the theory of city-regional integration also contributes to an understanding of variation in the other forms of intermunicipal activism. As previously mentioned, the city-regional context is also important but only with regard to softer forms of intermunicipal activism. Clearly, traditional participation theory has a broader area of applicability than one’s own municipality, but additional explanations are necessary in order to achieve a fuller understanding of how political orientations across municipal borders vary among citizens.

To investigate whether there are different explanations for intermunicipal and parochial activism, corresponding regression analyses were carried out using each of the forms of parochial activism as a dependent variable. ¹ Perhaps not surprisingly, variation in these forms is explained by measures in the Civic Voluntary Model; different indicators of city-regional integration have only a very limited effect. Hence, different kinds of connections to other municipalities in the city-region do not, on the whole, either enhance or restrain participation in one’s own municipality.

With few exceptions, the different forms of intermunicipal and parochial activism are explained by the same factors in the Civic Voluntary Model. For example, in general, those who are highly educated, immigrants, employed, dissatisfied with democracy, and interested in politics are likely to talk to other citizens about political matters in both their home municipality and in other municipalities in the city-region. This underlines the importance of the Civic Voluntarism Model for understanding variation in different forms of public participation.

Conclusions

One purpose of this article has been to investigate how common it is that citizens in two city-regions have an intermunicipal political orientation and whether this orientation is different for different forms of intermunicipal activism. An underlying assumption has been that extensive intermunicipal political orientation among citizens facilitates the establishment of systems of governance for city-regions as a whole. The results of the analysis indicate that there is a significant amount of intermunicipal political orientation in the city-regions, but that there are different forms of activism and not all of them are equally common. Thus, interest in news and politics in other municipalities is more common than efforts to influence political or administrative decisions in municipalities elsewhere in the city-region.

Another purpose was to identify possible explanations for variation in the use of different forms of intermunicipal activism. The study indicates that variation is not only a function of factors that normally promote citizen participation, but that there are also important additional explanations connected to the extent to which citizens are integrated into the city-region, the size of the city-region, and whether citizens live in the core city or a surrounding municipality. Softer forms of activism, such as being interested in news or politics in neighboring municipalities, tend to be more common among those who previously resided in, have family or relatives in, and/or spend leisure time in another municipality in the city-region. These forms are also more common in the outer municipalities in the Göteborg region. Harder forms, that is, attempts to influence other citizens or public decision makers in other municipalities, tend to be more common among those who are more engaged and among stakeholders, such as commuters, those who own property, or individuals who are members of organizations in city-region municipalities other than the one in which they reside. It is hoped that the research reported here will inspire additional studies of city-regional citizenship in other countries. This research should focus not only on intermunicipal orientations but also on the extent to which citizens have a political orientation toward city-regional matters that is not directly linked to specific municipalities.

Although the empirical basis for this study is only two city-regions, the results suggest that political participation in city-regions is influenced by elements that are not important drivers of participation in one’s own municipality. Promoting participatory democracy in the city-region cannot be limited to traditional forms of involvement but may require specific means that also take into account the views of commuters, shoppers, property owners, and temporary visitors who live elsewhere in the city-region. Hence, in order for city-regional governance to be legitimate, it must address additional challenges compared to those faced by municipal governments.