Social capital and community representation: How multiform networks promote local democracy in Los Angeles

Abstract

Chinese

This paper examines the variant roles that specific forms of networked-based social capital play in supporting the democratic functions of a neighbourhood governance network in Los Angeles. A significant body of empirical work has demonstrated the positive role that social capital plays in the functioning of civil society, but there has been less attention to the manner in which different types of network-based social capital promote support efficacy of multifaceted civic organisations. This paper utilises network measures from a survey of the members of a neighbourhood governance network in Los Angeles to explore the associations between types of network ties – within group; among groups; to different external stakeholders – and perceived self-efficacy of the member associations. We find that internal cohesion, or bonding social capital, promotes both advisement of city officials and promotion of local participation, suggesting that this network-based resource is fungible. Other network structures appear to have value in different contexts, in that stakeholder connections promote participation, while bridging social capital is associated with perceived success in advising city officials. The findings suggest that architects of participatory reform should be attentive to system goals in establishing supports for varying forms of system relationships.

Keywords

collective action community community representation governance local democracy neighbourhood participation social capital

This paper considers the variant role that social capital network structures play in promoting the efficacy of a participatory neighbourhood council system in Los Angeles. In many cities in the USA and internationally, such participatory networks attempt to serve as mediating institutions that connect community stakeholders to processes of urban policy formulation and service delivery (Berry et al., 1993; Musso et al., 2011; Sirianni and Friedland, 2001; Thomson 2001). These organisations are held to play important affective and instrumental roles. Associational involvement is argued to promote local democracy by providing opportunities for political interchange; developing civic skills; and nurturing attitudes of interpersonal trust and self-efficacy (Cohen and Rogers, 1995; Morales and Geurts 2007; Putnam, 2000). Community councils may partner with city entities to produce local public goods and services through processes of co-production and voluntary assistance. More profoundly, engagement in local governance may improve representation of localised preferences by mobilising citizens to express their concerns and interests and thus to promote citizen influence over local government activities (Cunningham and Kotler, 1983). Based on case comparison of cities with such participatory systems, Berry et al. (1993) argue that such neighbourhood involvement in governance improves the quality and efficacy of citizen participation; reduces alienation and strengthens trust in government; increases tolerance and reduces destructive conflict; cultivates a sense of community; and enhances government responsiveness to its citizens.

The City of Los Angeles adopted a charter reform to create such a neighbourhood governance system in 1999, implementing a new Department of Neighborhood Empowerment and charging it with creation of a system that would ‘promote more citizen participation in government and make government more responsive to local needs’ (Los Angeles Charter Section IX). The political motivations for charter reform were manifold, involving a desire by then Mayor Richard Riordan to strengthen mayoral powers, and concern about secession threats by several areas of the City that sought greater influence in downtown decision making (Cooper and Musso, 1999). Proponents of neighbourhood councils included several City Council members who had sponsored district-level community advisory systems, but had not been able to muster sufficient support to create a system legislatively. During charter deliberations, public interest in neighbourhood governance grew, and the framework for a participatory neighbourhood council system was drawn in broad strokes, leaving much of the systemic detail to a planning and implementation process that spanned several years.

The charter reform established the neighbourhood councils as advisory, and identified the broad range of goals that they might pursue, directing the system to promote participation, represent the diverse interests within each community, and make government more responsive to local needs. The Los Angeles councils are recognised and certified by the city to represent specific geographical areas that were defined by the council during a process of self-organisation that involved stakeholder mobilisation, boundary designation, development of by-laws, and community-level election of governing board members. As such, the Los Angeles neighbourhood council system is heavily grass roots in character. At the writing of this paper, the system includes 93 certified neighbourhood councils, on average governed by boards of 21 community stakeholders, and representing an average population of about 38,000 residents.

The city provided modest resources from the outset, most importantly a small staff of community organisers that help with organisational issues and a US$50,000 yearly grant to each council for organisational operations and community projects (in recent years this grant has been cut significantly because of city fiscal constraints). While the councils vary in their relative task orientation, they typically undertake some combination of community-based activities (community beautification; engagement of local problems such as homelessness or street congestion; community festivals) and advisory activities in which they provide input to City Council deliberations, the Mayor’s budget process, and local land-use decision making.

The neighbourhood governance system in Los Angeles has experienced mixed success. An evaluation of system performance after seven years of operation found that while a city-wide system had emerged in the face of widespread tokenism on the part of the City, nevertheless many councils still struggled with operational challenges and internal strife (Musso et al., 2007). The elected boards under-represent Latino residents relative to population, and have experienced some difficulty establishing their legitimacy with city officials (Musso et al., 2007). Also, early broad support for councils by residents weakened after they were in operation for a number of years. Nonetheless the system has spurred the development of political networks that connect councils to city policy bodies and to each other. The Los Angeles neighbourhood councils have advocated on an array of issues that include not only local community land-use concerns but also city-wide issues as diverse as City Council term limits, Department of Water and Power fees, and development of a city-wide solar power policy. Many neighbourhood councils support local schools, sponsor community events, and mobilise volunteers for street beautification or graffiti removal. One indication of their legitimisation as neighbourhood representatives is that council members are frequently referenced by local newspapers are spokespeople on community issues (Musso et al., 2007).

This paper considers how distinct types of network-based social capital provide Los Angeles community stakeholders with access to varying instrumental and affective resources, in turn supporting a diverse set of collective activities. In Los Angeles, neighbourhood councils provide advice on service delivery, constitute forums for deliberation and participation in democratic activities, mobilise resources to address community problems and advocate for neighbourhood policy interests in a manner that counters central-city elite influence (Berry et al., 1993; Ferman, 1996; Fung, 2004). Such diverse activities in turn require access to differentiated network-based resources and organisational capacities. For example, community-based networks connecting local associations and stakeholders are likely to promote organisational cohesion, localised collective action and play a positive role in recruiting and maintaining community participation. Nevertheless, such community-based networks are less likely to afford resources necessary for councils to advocate successfully on behalf of the community to city-level decision makers (Granovetter, 1973; Hill and Matsubayashi, 2005). Rather, direct networks with City decision makers and bridging networks connecting organisations to other City activists are more likely to enable councils play a more important role in marshalling information and communicating locale preferences to decision makers.

This paper theorises that specific forms of social capital are required for three distinct participatory activities: promoting local democratic participation, solving neighbourhood problems, and city-level advocacy. Bonding social capital supports organisational capacity to perform a variety of forms of collective action. In contrast, ties to neighbourhood stakeholders, City decision makers, and the broader network of neighbourhood councils create more specific forms of social capital that support some civic activities better than others. The theory is tested empirically through an analysis of network survey data from the governing board members involved in the neighbourhood councils created by the Los Angeles charter reform. The empirical findings suggest that differing forms of network-based social capital are associated with the promotion of local participation and advisement on city policy making. In particular, general forms of social capital that promote civic attitudes and develop civic and organisational capacity are associated with both self-evaluated success at promoting participation and advising the city. In contrast, more specific forms of social capital have specialised impacts wherein bonding type community-based ties promote participation while bridging types of social capital appear more closely related to perceived success in advising the City.

Social capital and democratic efficacy

As a system of mediating local neighbourhood council organisations, LA neighbourhood councils connect in complex ways internally, to stakeholders, city institutions, and to one another, creating social and political relationships that support distinct types of civic activities. Because the range of activities undertaken is quite diverse, efficacy requires access to a disparate set of resources within local communities as well as the development of relationships that support interaction and engagement in an array of domains. Some forms of social capital, such as intra-organisational ties that develop and maintain group norms, are likely to be important in supporting an array of group activities (Adler and Kwon, 2002; Baldassarri and Diani, 2007; Coleman, 1990; Portes, 1998). Other forms of network-based social capital provide access to more specialised sets of resources and are consequently less fungible (Adler and Kwon, 2002; Newton, 2001). For example, access to city officials is important for activities that involve political participation and representational activities but arguably less so for community-based problem solving, which depends more on mobilising diverse community resources (Laumann and Pappi, 1976; Lin, 2008; Son and Lin, 2008; Verba et al., 1995).

A related conceptualisation of social capital emphasises the importance of network-based resources that are accessed through diverse network ties, those that link an individual or organisation to others with varied characteristics or network positions (Bourdieu, 1986; Lin, 2001, 2008). Lin argues that social capital increases as a function of the diversity of individual social and economic positions within one’s network. For example, more diverse networks facilitate an individual’s search for employment and increase the frequency of volunteering because diversity increases the probability that network members have information on jobs and volunteering opportunities (Lin, 2008; Son and Lin, 2008). Similarly, heterogeneity advances collective action seeking provision of public goods by increasing the probability that a critical mass of interested supporters can emerge (Heckathorn, 1993; Oliver et al., 1985).

This essay argues that the relative importance of bonding, bridging, and other network-based social capital is context dependent, and that differing network structures support distinct facets of civic engagement. The complex multi-sectoral character of the Los Angeles neighbourhood council system suggests that while associational and bonding forms of social capital will be important across the board, other bridging and network forms of social capital are in effect task-specific. The relationship between specific types of network-based social capital and arenas of collective action is summarised in Table 1, and the remainder of this section discusses these relationships in more detail.

Table 1.

Theorised role of various forms of social capital in arenas of collective action.

Form of social capital	Arena of collection action
	Promote political participation	Advise city on city-wide policies	Address neighbourhood problems
Cohesive intra-group networks	+	+	+
Associational social capital	+	+	+
Number of ties to city officials	+	+
Diverse ties to city officials	+	+
Number of ties to stakeholders	+		+
Diverse ties to stakeholders			+
Brokerage among councils	+	+	+
Diverse ties among councils			+
High SES
Dissension within councils

‘Fungible’ social capital

Certain forms of social capital, in particular the associational social capital that is the focus of Putnam’s (2000) seminal analysis, are general in nature. Civic involvement entails a variety of activities that develop norms of generalised reciprocity and general civic attitudes. This type of social capital, like other forms of capital, can be considered ‘fungible’ to the extent that it supports the capacity of the organisation to accomplish a range of differing activities, an important asset in local associations that are so heavily reliant on volunteers to effect collective action (Adler and Kwon, 2002). In particular, social capital theory suggests that two forms of social capital, membership and internal bonding, are important to a wide array of organisational functions. The first of these, associational social capital, is a central requisite of participatory behaviour and civic orientation (Morales and Geurts, 2007; Putnam, 2000). Repeated engagement with individuals in an organisational context fosters political interchange and is theorised to have politicising influences, promoting attitudes of individual and collective political efficacy that are critical to group-level commitment, democratic discourse and representation, and social problem solving (Morales and Geurts, 2007; Stolle, 1998).

A second form of social capital that applies broadly is the internal bonding capital that arises from dense intra-group networks. To achieve collective goals the councils must recruit volunteers to engage in the basic organisational maintenance tasks and develop the capacity to address internal conflicts that can threaten group stability. Threatened by the liability of newness, they must also attain some level of internal coordination to gain the external legitimacy necessary for coordinating collective action (Putnam and Feldstein, 2003; Singh et al., 1986). Dense intra-group networks promote the development of group norms, facilitate common goals, and inhibit shirking and counterproductive behaviours (Coleman, 1990; Putnam, 2000). The goodwill and trust that arise from dense ties are requisites to use of other social capital assets (Galaskiewicz, 2007; Krackhardt, 1992; Schalk and Torenvlied et al., 2010).

H_1A: Councils whose members have more and more committed associational memberships are better able to achieve goals across a range of activities.

H_1B: Councils with more cohesive internal networks are better able to achieve goals across a range of activities.

Promoting local participation

Now turning to specific organisational functions, a critical function of the neighbourhood council system is the promotion of local democratic participation through engagement of community stakeholders – residents; business owners; organisational members – in more qualitatively intensive forms of involvement. In essense this is the distinction between associational memberships versus voting and checkbook politics (Berry et al., 1993; Skocpol, 2013; Skocpol and Fiorina, 2004). Strong social and friendship ties are important for recruiting individuals to these more intensive forms of engagement (Diani and McAdam, 2003). More extensive ties between council members and community members will increase the likelihood that recruitment occurs, and promote participation through social pressure and the development of group consciousness (Austin et al., 2012; Finkel and Smith, 2011; Gerber et al., 2008; Klofstad, 2007; Quintelier et al., 2012). Importantly, all of these mechanisms require high levels of direct contact between community members.

The councils can also facilitate community participation to the degree that they maintain active relationships with city officials, thereby mediating between the community and political elites and increase the likely efficacy of participation (Cohen and Rogers, 1995; Warren, 2001). Diverse ties to city officials are important because they facilitate access to a wider range of resources, and reduce the risk of being perceived as too cozy with political elites. In terms of hypotheses:

H_2A: Councils that maintain more ties to neighbourhood stakeholders are better able to promote participation.

H_2B: Councils that maintain more ties to city officials are better able to promote participation.

H_2C: Councils that maintain more diverse ties to city officials are better able to promote participation.

Advising city policy makers

One of the critical roles of the neighbourhood council system is to serve as a system of mediating organisations that voice community interests to city policy makers. This was a particular impetus for the reform in Los Angeles, where 15 City Council members each represent districts as large in population and geographic scale as a mid-sized American city. The literature on community participation suggests that advocacy and exertion of power on policy issues requires the ability to monitor policy making and to lobby decision makers, activities that in turn necessitate connections to informational and political resources and access to decision makers (Burt, 2000; Hill and Matsubayashi, 2005; Scholz et al., 2008). Three particular network structures are particularly important in this regard. First, entrepreneurial activities require strategic access to information and advantageous positions vis-à-vis other key actors. Actors who are able to bridge structural holes, areas within a social structure that are disconnected, can access novel information and negotiate from a position of advantage (Burt 2000, 2009). Second, developing more linkages with city officials promotes political efficacy by promoting access to city networks that can be utilised to gather information and exert political influence. Third, connecting to a diverse set of city officials connects councils to broader sources of novel information and promotes influence on complex city issues that are likely to involve a multiplicity of city actors in different agencies (Lin, 2001, 2008). Stated as hypotheses:

H_3A: Councils that broker structural holes within the network are better able to represent community interests to policy makers.

H_3B: Councils that maintain more ties with city decision makers are better able to represent community interests to policy makers.

H_3C: Councils that maintain more diverse ties with city decision makers are better able to represent community interests to policy makers.

Addressing neighbourhood problems

The neighbourhood council system also plays an important co-productive role, engaging volunteers in activities that address issues of local concern. To produce community-based goods and services the local councils need to promote neighbourhood level collective action and leverage a variety of resources, a form of community capacity involving broad inter-organisational networks (Chaskin, 2002: Chaskin et al., 2001). This importantly includes the creation of ties between community organisations, and the mobilisation of an array of resources and expertise, such as media, organising and legal skills that may derive from different sources within the community. Access to such resources is improved within diverse networks (Lin, 2008). Expressed as hypotheses:

H_4A: Councils that maintain more ties to neighbourhood stakeholders are better able to address neighbourhood problems.

H_4B: Councils that maintain more diverse ties to neighbourhood stakeholders are better able to address neighbourhood problems.

H_4C: Councils that maintain more diverse ties to other councils are better able to address neighbourhood problems.

In sum, the forms of network-based social capital that are theorised to be general prerequisites for effective civic action include bonding social capital based on intra-group cohesion that promote organisational capacity and general associational memberships that provide general training in civic skills and of access to resources. Other forms of network-based social capital in contrast provide access to more specialised sets of resources, and consequently they are less fungible (Adler and Kwon, 2002). Access to city officials is important for activities that involve political participation and representational activities but less so for community-based problem solving, which depends more on mobilising diverse community resources. Ties to community stakeholders are important for political recruitment and problem solving, but may detract from the ability to engage policy makers. Similarly, having efficient access to information on city-wide matters by means of brokerage across structural holes provides advantages when lobbying the city but is less important for purely community-based activities or for promoting participation.

An important constraint on the productive use of network-based social capital is the degree to which neighbourhood councils are capable of effectively employing these assets. As Galaskiewicz (2007) has argued social capital may be best conceptualised as a latent resource that needs to be matched with organisational purpose and capacity to be transformed into tangible civic results. For voluntary civic groups, such requisite organisational capacity can be fleeting. In their study of Portland neighbourhood councils Putnam and Feldstein (2003) find that only about one-third of the councils were effectively active at any particular time. Two main factors are likely to impede organisational capacity. The first is internal dissension amongst members, which Crenson (1974) has found to be is a major impediment to community-level collective action. A related factor is the degree to which members have the basic civic skills required to run a small organisation. There skills are strongly correlated with socio-economic status (Verba et al., 1995). Consequently, in a city such as Los Angeles that is characterised by extreme levels of economic and ethnic diversity, it is likely that not all councils are managed by with members with high levels of civic skills. This leads to our final two hypotheses:

H_5A: Councils that experience internal dissention are less likely to achieve goals.

H_5B: Councils with members with higher levels of civic skills are more likely to achieve goals.

Data and methods

The network data analysed in this study come from a 2006 survey of the board members of Los Angeles Neighbourhood councils. The survey was a major component of a multi-year evaluation that examined the degree to which this new system promoted the development of social capital and improved democratic governance. It included items concerning board members’ networking activities, political attitudes, political activities and demographic information. Because of the wide reach of the system, the survey included both roster-based items designed to collect whole network data and items akin to position generators that captured the range of contacts that neighbourhood councils maintained with outside organisations. At the time of the survey, there were 85 certified neighbourhood councils, though two were inactive and a third provided insufficient responses, leaving 82 councils for this analysis.

Project staff visited all neighbourhood councils to inform them of the upcoming survey, explain its purpose and answer any questions. Each board member received a personal email including a link to the web-based survey. Two email reminders were sent out, and then board members were contacted by phone and given an opportunity to respond verbally. To accommodate the large number of new immigrants in Los Angeles, respondents were able to take the survey in Spanish and Korean. There were a total of 1499 board members at the time and 702 completed the survey, for a response rate of 47%. Missing values for certain variables reduce the sample size in the analyses presented here.¹

Primary network measures

The network variables employed in the model are conceptualised as council-level features based on five sets of survey items. Individual-level survey responses were aggregated to the council-level, a method that has construct validity and is commonly employed to address issues with incomplete responses rates (Henry et al., 2011). The first item captured whole network data of intra-council communications. Each respondent was presented with a complete list of the members of their neighbourhood council and asked with whom they were in contact in the two weeks prior to their last board meeting ‘to discuss matters concerning politics, government or neighborhood issues’.² To capture the degree of bonding social capital within each council, the study measures Internal Cohesion as the mean out-degree for board members. This measure is preferred to the more common measure of network density (e.g. the proportion of possible pair of actors in a network that share a link) because it is robust against differences in the size of boards and the differing response rates of boards.³

Two other survey items collected information on the range of city and neighbourhood groups with which the member was in contact. The first included lists of city offices (e.g. mayor, city council, city departments) and the second listed stakeholder groups (e.g. homeowners, social service agencies, businesses), and respondents were asked if they had been in contact with a member of each office or group in the same two week period. To capture the aggregate amount of board communications with community stakeholders and city offices, we calculated the average number of neighbourhood stakeholder groups (Mean Stakeholder Contacts) or City Offices (Mean City Contacts) contacted by survey respondents from each neighbourhood council. The diversity of contacts with stakeholders and city offices are measured by Total Differing Stakeholder Contacts and Total Differing City Contacts, which are the total number of different stakeholder group or offices named by survey respondents on a particular council.⁴

A fourth item captured networking activity with other neighbourhood councils. Each survey respondent was allowed to provide up to six other neighbourhood councils with which they had been in contact in that two week period.⁵ Following other work on organisational networks, a neighbourhood council network was then developed by aggregating these individual responses (Berardo and Scholz, 2010; Henry et al., 2011; Scholz et al., 2008). The presence of a directed tie between two boards is recorded when at least one member of board A indicated that they had been in contact with board B.

This network permits the calculation of two broader network measures. First, the diversity of network-based resources a neighbourhood council may access is captured by employing Lin’s (2008) concept of a position generator that measures diversity based on the range of alters’ occupations within an actor’s network. The logic is that networks with a greater range of occupational prestige will afford greater access to a variety of resources. While these data do not include information on occupations, Board Network Diversity is measured by the range of the mean family incomes of the neighbourhoods with which a neighbourhood council is in contact.

To determine the strategic position, or access to structural holes, of a board within the broader neighbourhood council network, we employed UCINET VI to calculate Interboard Constraint, a measure of the degree to which an actor is connected to redundant alters (Burt, 2000, 2009). Lower scores for this measure indicate that a neighbourhood council is situated in a context rich with structural holes and benefits from information access and control advantages.

Respondents were also presented with a battery of questions to evaluate their associational-based social capital. Based on a standard list of 13 types of civic organisations, respondents rated their level of involvement in each type on a five-point scale, ranging from not a member coded as 0 to a leader in the organisation which was coded as 4. An Associational Index was calculated by summing these 13 scores.⁶ Council-level measures of the network-based social capital were developed by averaging these individual-level indices.

Organisational capacity measures

Two additional measures of organisational capacity are included in the model. Internal Conflict is a measure of the degree to which contentious intra-organisational dynamics impede councils from leveraging latent social capital. This measure was calculated as the proportion of survey respondents reporting internal conflict as a major challenge facing their council. Mean Income is included as a proxy for level of civic skills available in a community based on the high correlation between skill and socio-economic status. This variable was calculated by employing GIS to aggregate census blocks to the boundaries of neighbourhood councils and then calculating the mean family income for that area.

Dependent variables

The dependent variables measuring efficacy of the organisation are taken from a self-evaluation included in the survey. Respondents were asked to rate the performance of their council on three dimensions: (1) promoting citizen participation in government, (2) advising the City on city-wide policies, and (3) working to solve problems in the neighbourhood. The ratings were on a four-point scale ranging from poor (1) to excellent (4). These ratings are averaged across all respondents from each neighbourhood council. While objective measures of council performance would be preferable, such measures are costly and likely to be unreliable for neighbourhood associations, in that activities require issue saliency that will vary over time and across organisations. There are several reasons to conclude that the survey items capture important dimensions of council performance. The board members are the best informed individuals concerning their activities, though their opinions frequently diverge. Additional cross-validation is provided by similar performance ratings by the community organisers that worked with the councils. Although information was available only for a subset of councils, two out of the three measures have a positive, and statistically significant correlation with the council self-reports with correlations of 0.40 and 0.50. In the case of promoting citizen participation the ratings are still positively correlated, but the correlation does not reach statistical significance at the 5% level.⁷

Table 2 displays the descriptive statistics for these variables including the survey response rates by board. Table 3 presents their correlations. On all measures there is a significant amount of variation over the 83 councils in our sample. On the average council members had about six contacts with other board members, but this ranged between only 1.29 to as many as 12.75 contacts. Similar variation is found with the average number of contacts with the city and with community stakeholders. The average board was in contact with 14.2 different city and stakeholder groups (out of a possible 19 groups listed in the position generator items), while the least connected council only was in contact with six different groups and the most connected council was in contact with all possible groups. The measure of Interboard Constraint ranges between 0.12 and 1.0. Because these organisations civically engaged individuals with high SES, the average income and index of associational memberships are both quite high. Nevertheless, there a wide amount of variation between council in terms of income and associational memberships. The three independent variables are only moderately correlated, suggesting that they capture distinct dimensions of organisational performance.

Table 2.

Descriptive statistics of 82 boards.

	N	Mean	Min.	Max.	Std. dev.
Promote Participation	82	2.41	1.13	4.00	0.49
Advise City on City-wide Policy	82	2.45	1.00	3.75	0.47
Solve Community Problems	82	2.81	1.25	3.67	0.51
Internal Cohesion (Mean Out-degree)	82	5.94	1.29	12.75	2.29
Index of Associational Membership	82	14.06	5.50	24.25	3.27
Mean City Contacts	82	2.46	1.00	7.25	1.01
Total Differing City Contacts	82	7.85	3.00	11.00	2.05
Mean Stakeholder Contacts	82	2.06	0.25	4.33	0.84
Total Differing Stakeholder Contacts	82	6.32	1.00	8.00	1.68
Interboard Constraint	82	0.35	0.12	1.00	0.15
Board Network Diversity (1000s)	82	48.71	0.00	13.33	31.83
Mean Income (1000s)	82	59.87	25.96	213.66	28.37
Internal Conflict	82	0.36	0.00	1.00	0.29
Board-level Response Rates	82	48.1	14.29	92.86	16.05

Table 3.

Variable correlations.

	[1]	[2]	[3]	[4]	[5]	[6]	[7]	[8]	[9]	[10]	[11]	[12]
Promote Participation [1]	1.00
Advise City on City-wide Policy [2]	0.52	1.00
Solve Community Problems [3]	0.61	0.52	1.00
Internal Cohesion (Mean Out-degree) [4]	0.33	0.31	0.27	1.00
Index of Associational Membership [5]	0.25	0.15	0.08	0.17	1.00
Mean City Contacts [6]	0.06	0.12	0.07	0.34	0.30	1.00
Total Differing City Contacts [7]	0.10	0.18	0.22	0.43	0.20	0.63	1.00
Mean Stakeholder Contacts [8]	0.28	0.18	0.06	0.29	0.26	0.64	0.44	1.00
Total Differing Stakeholder Contacts [9]	0.22	0.27	0.19	0.39	0.03	0.40	0.51	0.72	1.00
Interboard Constraint [10]	0.05	−0.34	−0.15	0.01	0.06	−0.06	−0.22	−0.13	−0.30	1.00
Board Network Diversity [11]	−0.05	0.03	0.06	0.04	0.07	0.05	0.06	0.01	0.21	−0.30	1.00
Mean Income [12]	0.19	0.16	0.30	0.20	0.12	−0.08	0.14	−0.12	0.02	−0.14	0.27	1.00
Internal Conflict [13]	−0.33	−0.28	−0.42	−0.15	0.03	0.02	0.02	0.03	−0.09	0.16	0.03	−0.10

Findings

Because of the network relationships between the councils in these data, the assumption of the independence of observations is violated. Consequently we employ a linear network autocorrelation model that corrects for the fact that an observation’s error term may be influenced by its location within the network of neighbourhood councils.⁸ Based on the sociomatrix of the network relationships among the 83 councils, this procedure estimates Rho that parameterises the error in the dependent variable based on a council’s neighbours in the network. The results of the three models are presented in Table 4.

Table 4.

Linear network autocorrelation model results.

	Promote Participation	Advise City on City-wide Policy	Solve Community Problems
Intercept	1.641**(0.334)	2.268**(0.333)	2.317**(0.362)
Internal Cohesion (Mean Out-degree)	0.049*(0.023)	0.047*(0.023)	0.024(0.025)
Index of Associational Membership	0.024(0.015)	0.023(0.015)	0.004(0.016)
Mean City Contacts	−0.157*(0.072)	0.009(0.071)	−0.035(0.077)
Total Differing City Contacts	0.004(0.003)	−0.018(0.032)	0.045(0.036)
Mean Stakeholder Contacts	0.278**(0.101)	−0.030(0.100)	−0.009(0.108)
Total Differing Stakeholder Contacts	−0.024(0.049)	0.041(0.048)	0.017(0.052)
Interboard Constraint	0.387(0.332)	−1.034**(0.343)	−0.0520.363
Board Network Diversity (1000s)	0.001(0.002)	−0.003(0.002)	0.000(0.002)
Mean Income (1000s)	0.003(0.002)	0.001(0.002)	0.003(0.002)
Internal Conflict	−0.528**(0.163)	−0.2980.167	−0.676**(0.178)
Rho	−0.001(0.004)	0.058(0.034)	0.032(0.036)
Degrees of freedomAdjusted R²	700.241	700.158	700.166

Notes: Significance codes **p < 0.01 *p < 0.05 (standard errors in parentheses).

Neighbourhood council capacity does appear to be important. The coefficients for the mean family income in each community are positive in all models, though none of the estimates are statistically significant. At the council level, higher levels of Internal Conflict are associated with lower levels of performance on all measures. The coefficients are statistically significant in two models and marginally significant in the third providing support for hypothesis 5_B. The magnitude of these coefficients is quite large. For example, in terms of Model III, working to solve neighbourhood problems, a one standard deviation increase in the measure of internal conflict decreases the council performance by over one-half of a standard deviation.

With respect to direct network measures, the results are mixed, though the overall pattern of results supports the notion that various forms of network-based social capital have differing effects on the ability of community representing organisations to engage in collective action. Starting with Hypotheses H_1A and H_1B, the results provide moderate support that generalised forms of social capital support multiple organisational activities. Internal Cohesion is positively related to group performance in all of the models, and is statistically significant in two. In contrast while the estimated coefficient for the Index of Associational Memberships is positive in all of the models, these estimates are not statistically significant. This result may be due to the fact that the boards in this sample all benefit from high level of previous connections to civic associations, with a minimum board associational index of 5.5, such that there may be little marginal benefit from increasing the level of prior associational experience within this range.

The results support one of the three hypotheses (H_2A) concerning the specific network ties that support the goal of promoting participation, in that ties to community stakeholders are associated with higher self-reports on promotion of participation. Contrary to expectations, councils with more and more diverse ties to city offices do not appear to have greater success in promoting participation (H_2B and H_2C). The coefficient for diverse ties is not statistically significant while the coefficient for the number of ties is statistically significant but negative, counter to the hypothesised direction of relationship. It is possible that councils with many ties to city officials focus on political advocacy to the exclusion of reaching out to stakeholders that they purport to represent.

The results for the model on representing community interests also provide partial support for the specialised nature of social capital in promoting efficacy. The results do support H_3A that brokering structural holes provides councils strategic advantages in terms of political influence. Nevertheless, as in the participation model the hypotheses concerning the benefits of more and more diverse connections to city government (H_3B and H_3C) are not supported by the results. Finally, the third model that seeks to explain the success of councils in solving neighbourhood problems performs less well. None of the forms of network-based social capital that were hypothesised to facilitate problem solving were statistically significant. Rather, the main finding of the model is that the coefficient for Internal Conflict is negative and statistically significant, suggesting that conflict is an impediment to successful action. It is possible that during the study periods the neighbourhood councils faced a varying number of local problems with differing levels of salience. In the absence of mobilising events, latent network-based social capital may not have been activated.

While these findings reinforce the importance of understanding the role of social capital in a nuanced manner, it is important to acknowledge that this type of cross-sectional analysis has inherent methodological and contextual limitations that raise unavoidable issues with internal and external validity. There is the risk that self-reported perspectives on group efficacy may lead to spurious correlations at the respondent level (Meier and O’Toole, 2013). As a cross-check, the analysis was replicated using the subset of councils with performance ratings by city organisers who worked with them. The findings were qualitatively similar to the ones presented here, though most of the results did not attain statistical significance because of the greatly diminished degrees of freedom.

We also acknowledge the limitations of the network measures. We capture short-run communications related to a broad variety of community and political activities. These measures attain face validity. The inherent difficulties of capturing multifaceted networks in the community context combined with the complex and uncertain relationship between short-run communication patterns and long-term stocks of social capital leave much room for further study. As in all cross-sectional work, causation is an issue. The correlations between network structures and specific organisational outcomes in this study could be explained by goal-driven behaviour in which neighbourhood councils wished to pursue particular goals and then engaged in the networking activities necessary to achieve them.

The degree to which these findings are generalisable are challenged by two factors. First, the research context has a number of distinct characteristics compared with other cities with active neighbourhood governance systems (Fung, 2004; Putnam and Feldstein, 2003). Important differences include that the LA neighbourhood councils are quite large and that their relationship with the city is more fraught, which certainly affects the manner in which they accumulate and apply social capital.⁹ The self-selection of individual neighbourhood council members also raises issues about extending these results to other types of organisations. In particular, civic engagement networks are likely to attract individuals who are more connected (as evidenced by the relatively high associational index in our sample). These voluntary members are likely to have more polarised political views, which in turn may colour their assessment of self- or collective-efficacy. These validity issues are difficult to avoid in research that seeks to understand the function of networks, which invariably are the product of self-selective processes. In sum, while this analysis adds to understanding of the multiple pathways by which network-based social capital contributes to collective action, further refinement of these ideas will be promoted by additional community-level research. Given the importance of contextual details, research on differing organisations in different geopolitical locales will add important insights into how and when context matters. Observing the consequences on collective action does pose methodological challenges, and in particular, there is a need for research in urban fields where more objective measures of organisational success such as the completion of a specific projects or attraction of resources can provide cross-validated outcome measures.

Discussion and implications

These results suggest a nuanced relationship between variant forms of social capital and the diverse goals of the neighbourhood councils that constituted the Los Angeles neighbourhood governance system. Consonant with research suggesting that bonding and bridging networks have differential impacts in terms of the representation of community political demands and for collective action (Berardo and Scholz, 2010; Hill and Matsubayashi, 2005), we find that bridging between structural holes in the city-wide networks appears to promote advice to the City, but not other functions of the council, while community-based contacts uniquely promote local democratic participation. Similarly, these results are broadly sympathetic with previous research finding that community-based versus cosmopolitan contacts have different effects on community cohesion and political activities (Crenson, 1978; Guest and Oropesa, 1986; Hipp and Perrin, 2006).

The analysis highlights the importance of distinguishing general and fungible forms of social capital from more specific access to network-based resources. General forms that encompass norms of reciprocity, general civic and organisational competencies are broadly supportive of a range of collective action goals. In contrast, ties to the community or to other councils provide access to specific resources such as information and expertise, that are more likely to be productive in specific activities. The strength of these network results is highlighted when compared with the Mean Income of communities. Socio-economic status is commonly found to be one of the most important predictors of civic involvement (Valelly, 1996; Verba, 1967; Verba et al., 1995). Nevertheless, in these models the network assets of councils are found to be more influential than SES in facilitating collective action. In particular, the consistently positive impact of Internal Cohesion on organisational performance demonstrates the importance of bonding social capital or strong ties to enabling informal, voluntary organisations to work together productively. Network diversity is the one specific form of network-based social capital that did not find any support within this analysis. It is possible that the lack of empirical validation is due to multicollinearity with other network measures. As seen in Table 3, the measures of the average number of contacts with the City and with community stakeholders have correlations ranging between 0.4 and 0.72 with the measures of the diversity of contacts with such stakeholders To test this possibility, we reran the three models omitting the Mean City Contacts and Mean Stakeholder Contacts variables from the model. In this case, the measure of diversity of contacts with stakeholder groups did achieve marginal statistical significance in explaining promoting participation.¹⁰ In sum, connecting to a diverse set of stakeholder groups may also play a role in conjunction with the overall connections with the community in promoting participation. Nevertheless, the multicollinearity in these data precludes firm conclusions on the independent effects of these distinct forms of social capital.

On that note, this study highlights the importance of context on the ability of neighbourhood councils to leverage social capital assets. In particular, internal dissension had a powerfully negative influence on all collective action goals. This suggests that the council system in Los Angeles may possess latent social capital, but that low organisational capacity may impede the councils’ ability to leverage this capital for collective goals (Galaskiewicz, 2007). In Los Angeles, local councils represent quite large communities, an average of 38,000 people, and this size has impeded their ability connect with other civil society or social service organisations (Musso et al., 2007). This general detachment from other associations may account for the lack of an observed positive relationship between the range of relationships and the ability of councils to solve local problems. Also, councils in Los Angeles have experienced contested relations with the city, possibly due to a tendency for those with more strongly held policy positions to be attracted to such participatory venues (Fiorina, 1999). Council members have complained that they are not afforded the attention they deserve, and the city has questioned the legitimacy of councils to represent their neighbourhoods. These tensions may diminish the value of ties between councils and the city, explaining why connections to the city actually decrease the ability of councils to promote participation and to advise the city. This suggests that network relationships can have both positive and negative valence, a nuance not generally addressed in the literature.

While one must be cautious of generalising from a complex case, there are cautionary findings with respect to the design of participatory arenas within an urban setting. While it is widely understood that local participatory systems such as community councils can have collateral benefits for both individual civic development and broader democratic functionality, urban policy makers should be cognisant that the specific structure of social capital advanced by these organisations may promote particular tasks with consequential implications for local governance. For example, if the goal is to encourage community organisations to work at the ‘street level’ on co-production of public services, it may be less important to develop channels that connect to policy makers, and more important to develop connections with administrators and specific groups of service users who can contribute to public service production. Conversely, if the goal is to encourage democratic dialogue that promotes civic understanding, more attention would need to be paid to internal communication channels and to engaging a diverse and representative group of community members. As such, the network structure of community governance systems, deriving from city-level design and implementation choices, may have important implications for the manner in which they operate and their success in promoting the complex inter-sectoral nature of civic engagement.

Footnotes

Funding

The authors are grateful to the following for funding: James Irvine Foundation, John Randolph Haynes and Dora Haynes Foundation, National Science Foundation, Information Technology Research Grant #0112899.

Notes

References

Adler

Kwon

S-W

(2002) Social capital: Prospects for a new concept. Academy of Management Review 27(1): 17–40.

Austin

SDW

Middleton

. (2012) The effect of racial group consciousness on the political participation of African Americans and black ethnics in Miami-Dade County, Florida. Political Research Quarterly 65(3): 629–641.

Baldassarri

Diani

(2007) The integrative power of civic networks. American Journal of Sociology 113(3): 735–780.

Berardo

Scholz

(2010) Self-organizing policy networks: Risk, partner selection, and cooperation in estuaries. American Journal of Political Science 54(3): 632–649.

Berry

Portney

. (1993) The Rebirth of Urban Democracy. Washington, DC: The Brookings Institution.

Bourdieu

(1986) The forms of capital. In: Richardson

(ed.) Handbook of Theory and Research for the Sociology of Education. New York: Greenwood, pp. 241–258.

Burt

(2000) The network structure of social capital. Research in Organizational Behavior 22: 345–423.

Burt

(2009) Structural Holes: The Social Structure of Competition. Cambridge, MA: Harvard University Press.

Chaskin

(2002) Networking neighborhoods. Social Service Review 76(4): 695–698.

10.

Chaskin

Brown

. (eds) (2001) Building Community Capacity. Hawthorne: Aldine de Gruyter.

11.

Cohen

Rogers

(1995) Associations and Democracy. London: Verso.

12.

Coleman

(1990) Foundations of Social Theory. Cambridge, MA: Harvard University Press.

13.

Cooper

Musso

(1999). The potential for neighborhood council involvement in American metropolitan governance. International Journal of Organization Theory and Behavior 2(1): 199–232.

14.

Crenson

(1974) Organizational factors in citizen participation. The Journal of Politics 36(02): 356–378.

15.

Crenson

(1978) Social networks and political processes in urban neighborhoods. American Journal of Political Science 22(3): 578–594.

16.

Cunningham

Kotler

(1983) Building Neighborhood Organizations. Notre Dame, IN: University of Notre Dame Press.

17.

Diani

McAdam

(eds) (2003) Social Movements and Networks: Relational Approaches to Collective Action. New York: Oxford University Press.

18.

Ferman

(1996) Challenging the Growth Machine: Neighborhood Politics in Chicago and Pittsburgh. Lawrence, KS: University Press of Kansas.

19.

Finkel

Smith

(2011) Civic education, political discussion, and the social transmission of democratic knowledge and values in a new democracy: Kenya 2002. American Journal of Political Science 55(2): 417–435.

20.

Fiorina

(1999) Extreme voices: A dark side of civic engagement. In: Skocpol

Fiorina

(eds) Civic Engagement in American Democracy. New York: Russell Sage Foundation, pp. 405–413.

21.

Fung

(2004) Empowered Participation. Princeton, NJ: Princeton University Press.

22.

Galaskiewicz

(2007) Has a network theory of organizational behaviour lived up to its promises? Management and Organization Review 3(1): 1–18.

23.

Gerber

Green

Larimer

(2008) Social pressure and voter turnout: Evidence from a large-scale field experiment. The American Political Science Review 102(1): 33–48.

24.

Granovetter

(1973) The strength of weak ties. American Journal of Sociology 78(6): 1360–1380.

25.

Guest

Oropesa

(1986) Informal social ties and political activity in the metropolis. Urban Affairs Quarterly 21: 550–574.

26.

Heckathorn

(1993) Collective action and group heterogeneity. American Sociological Review 58(3): 329–350.

27.

Henry

Lubell

McCoy

(2011) Belief systems and social capital as drivers of policy network structure: The case of California regional planning. Journal of Public Administration Research and Theory 21(3): 419–444.

28.

Hill

Matsubayashi

(2005) Civic engagement and mass-elite policy agenda agreement in American communities. American Political Science Review 99(2): 215–224.

29.

Hipp

Perrin

(2006) Nested loyalties: Local networks’ effects on neighborhood and community cohesion. Urban Studies 43(13): 2503–2523.

30.

Klofstad

(2007) Talk leads to recruitment: How discussions about politics and current events increase civic participation. Political Research Quarterly 60(2): 180–191.

31.

Kossinets

(2006) Effects of missing data in social networks. Social Networks 28(3): 247–268.

32.

Krackhardt

(1992) The strength of strong ties: The importance of philos in organizations. In: Nohria

Eccles

(eds) Networks and Organizations: Structure, Form,and Action. Cambridge, MA: Harvard Business School Press, pp. 216–239.

33.

Laumann

Pappi

(1976) Networks of Collective Action: A Perspective on Community Influence Systems. New York: Academic Press.

34.

Lin

(2001) Social Capital: A Theory of Social Structure and Action. London; New York: Cambridge University Press.

35.

Lin

(2008) A network theory of social capital. In: Castiglione

van Deth

Wolleb

(eds) Handbook on Social Capital. New York: Oxford University Press pp. 50–69.

36.

Meier

O‘Toole

(2013) Subjective organizational performance and measurement error: Common source bias and spurious relationships. Journal of Public Administration Research and Theory 23(2): 429–456.

37.

Morales

Geurts

(2007) Associational involvement. In: Van Deth

Montero

Westholm

(eds) Citizenship and Involvement in European Democracies: A Comparative Analysis. London: Routledge, pp. 135–157.

38.

Musso

Weare

Bryer

. (2011) Toward “Strong Democracy” in global cities? Social capital building, theory-driven reform, and the Los Angeles neighborhood council experience. Public Administration Review 71(1): 102–111.

39.

Musso

Weare

Elliot

. (2007) Toward Community Engagement in City Governance: Evaluating Neighborhood Council Reform in Los Angeles. Los Angeles, Neighborhood Participation Project, School of Policy, Planning, and Development, University of Southern California: 1–40.

40.

Newton

(2001) Trust, social capital, civil society, and democracy. International Political Science Review 22(2): 201–214.

41.

Oliver

Marwell

. (1985) A theory of the critical mass. I. Interdependence, group heterogeneity, and the production of collective action. American Journal of Sociology 91(3): 522–556.

42.

Portes

(1998) Social capital: Its origins and applications in modern sociology. Annual Review of Sociology 24(1): 1–24.

43.

Putnam

(2000) Bowling Alone: The Collapse and Revival of American Community. New York: Simon & Schuster.

44.

Putnam

Feldstein

(2003) Better Together: Restoring the American Community. New York: Simon and Schuster.

45.

Quintelier

Stolle

Harell

(2012) Politics in peer groups: Exploring the causal relationship between network diversity and political participation. Political Research Quarterly 65(4): 868–881.

46.

Schalk

Torenvlied

Allen

(2010) Network embeddedness and public agency performance: The strength of strong ties in Dutch higher education. Journal of Public Administration Research and Theory 20(3): 629–653.

47.

Scholz

Berardo

Kile

(2008) Do networks solve collective action problems? Credibility, search, and collaboration. Journal of Politics 70(2): 393–406.

48.

Singh

Tucker

House

(1986) Organizational legitimacy and the liability of newness. Administrative Science Quarterly 31: 171–193.

49.

Sirianni

Friedland

(2001) Civic Innovation in America: Community Empowerment, Public Policy, and the Movement for Civic Renewal. Berkeley, CA: University of California Press.

50.

Skocpol

(2013) Diminished Democracy: From Membership to Management in American Civic Life. Oklahoma, OK: University of Oklahoma Press.

51.

Skocpol

Fiorina

(2004) Civic Engagement in American Democracy. Washington, DC: Brookings Institution Press.

52.

Son

Lin

(2008) Social capital and civic action: A network-based approach. Social Science Research 37(1): 330–349.

53.

Stolle

(1998) Bowling together, bowling alone: The development of generalized trust in voluntary associations. Political Psychology 19(3): 497.

54.

Thomson

(2001) From Neighborhood To Nation: The Democratic Foundations of Civil Society. Hanover, NH: University Press of New England.

55.

Valelly

(1996) Couch-potato democracy? The American Prospect 25: 25–26.

56.

Verba

(1967) Democratic participation. The Annals of the American Academy of Political and Social Science (Social Goals and Indicators for American Society) 2(373): 53–78.

57.

Verba

Schlozman

Brady

. (1995) Voice and Equality: Civic Voluntarism in American Politics. Cambridge, MA: Harward University Press.

58.

Warren

(2001) Democracy and Association. Princeton, NJ: Princeton University Press.