Social Capital and Transformational Leadership in Building the Resilience of Local Governance Networks

Abstract

Resilience has become more popular among researchers and practitioners of public governance. Previous studies indicate the importance of social capital and leadership in this research field, but mainly theoretically and rarely are considered together. Therefore, this article aims to analyze the impact of social capital and transformational leadership on the resilience of local governance networks based on a questionnaire conducted among 199 local governments in Poland. The variance-based structural equation modelling (SEM) based on the partial least squares path modelling method (PLS) has been used to analyze the data collected. The obtained results indicate that social capital, primarily relational and cognitive, directly and significantly affects resilience, but transformational leadership impacts indirectly. In addition, the resilience of local governance networks is associated with bouncing back and change. These findings add value to the emerging theory of resilience in public governance and can also be helpful for public professionals.

1. Introduction

Governance networks are essential in creating and delivering high-quality public services [1]. These networks can respond quickly and directly to society’s needs by joining the resources of many organizations. However, they struggle with many problems resulting from jointly carrying out activities by autonomous actors, such as limited transparency, division of power, conflicts and rivalry, distrust, and opportunism [2,3,4]. Problems in governance networks may also arise from many external challenges such as environmental turbulences, financial crises, pandemics, and the complexity of social issues [5]. These conditions make resilience increasingly important in governance networks [6,7].

Resilience is an emerging research area in public governance. It indicates how to cope with turbulence and complexity for sustainable development. It also refers to using the emerging challenges to transform and implement innovative solutions [3,8,9]. Research on resilience in public governance began after the 2001 attacks on the World Trade Center and initially grew up in emergency management [10]. However, more and more researchers perceive this theory as an opportunity to face the complexity of different public service delivery processes [3,4,6,11]. Nowadays, resilience is an important capability in times of many different economic, social, technological, or political threats, but research in this area is still in the infancy stage [6,7,8].

Stephanie Duchek [12], Julia Hillmann [13], and Julia Hillmann and Edeltraud Guenther [8] believe that this is still an umbrella concept in its excitement phase and that in-depth study and explanation are needed to be able to assess and develop resilience. The concept of resilience in governance networks is emerging, and its exploration may bring new prospects for developing these networks [10,14].

Researchers highlight the influence of social capital and leadership on resilience [14]. The role of social capital is researched mainly in emergency management [15,16,17], tourism and hospitality [18], sociology [19], and supply chain management [20]. In turn, leadership is an essential factor that encourages creativity and improvisation [21,22,23] that can also turn crises into development opportunities [24]. In addition, it affects social capital and builds organizational resources by developing a common platform for exchanging knowledge and experience in network relations [25,26]. Transformational leadership is particularly interesting because it is the dominant leadership style [27,28], forms the basis of a relational approach to leadership [29,30], and applies to governance networks [27,29]. However, many researchers believe these relationships are insufficiently studied, and scientists have not understood them well enough to use them effectively [18,30]. There is also a research gap concerning the analysis conducted in networks. For the above reasons, this article aims to verify the relations between resilience, social capital, and transformational leadership in governance networks.

Although current research explores the relations between social capital and resilience [15] or transformational leadership and resilience [23,24], they are studied relatively rarely and not together. This article examines them both, and the impact of transformational leadership on social capital is also analyzed. Moreover, research on the relationship between social capital, transformational leadership and resilience is conducted primarily from the organization’s perspective. This article is part of the underexplored research area related to the network level.

The findings indicate that contemporary local governance networks understand their resilience as an opportunity for development, not only for maintaining stability, which is an added value to the current scientific achievements [6,7]. They confirm the last research results [12,23] that social capital significantly impacts building resilience. However, the most important are relational and cognitive capital. The findings also indicate that transformational leadership has an indirect influence on resilience. This result aligns with [29] because transformational leaders mainly shape their followers’ values, beliefs, and behaviors. In general, this article contributes to developing the theory of resilience in governance networks.

This article is organized as follows. First, the theoretical background creates the foundations for analyzing the issues of the three examined constructs: resilience in local governance networks, social capital, and transformational leadership. Then, the methodology section presents the research approach and measures used in the research for each construct individually. It also explains the used research method—Computer-Assisted Web Interview—and utilized method of results analysis—Partial Least Squares Structural Equation Modeling. Finally, the discussion section includes verifying the obtained results regarding the current scientific achievements presented in the literature. This research contributes to the development of governance theory and practice included in the conclusions section.

2. Theoretical Background and Hypothesis Development

2.1. Resilience in Governance Networks

Governance networks consist of at least three goal-oriented and interdependent entities that engage in various relationships (e.g., collaboration, cooperation, coopetition) to develop a public value (tangible or intangible) and introduce innovations [31]. They are defined as “more or less stable patterns of social relations between mutually dependent actors, which cluster around a policy problem, a policy programme, and/or a set of resources and which emerge, are sustained, and are changed through a series of interactions” [1] (p. 11). Governance networks relate primarily to using and developing ties binding various entities to achieve common goals. Implementing activities in such networks seems beneficial, as combining the knowledge, competencies, and resources of many organizations from different sectors should achieve more than acting alone [31,32,33,34].

Public organizations play a crucial role in governance networks as they are responsible for actions undertaken in their administrative area and coordinate the functioning of other entities [32]. Although governance networks develop at all levels of the state organization, the local level contributes most to providing high-quality public services [35]. The proximity to society allows local governments to respond directly to emerging needs.

However, governance networks at all state levels struggle with many problems, such as limited transparency and autonomy or the division of power and responsibility [34]. In such circumstances, conflicts and rivalry are part of everyday life and come from distrust, opportunism, and conflict of interest [3,4,6]. Problems in governance networks may also arise from many contemporary challenges, e.g., the complexity of social problems, turbulences, joint decision-making, service integration, and resource constraints [5,36]. In addition, financial crises, disasters, climate change, political conflicts, terrorism, and pandemics such as COVID-19 cause new challenges with coordination, collaboration, and public service delivery. For this reason, resilience is becoming more and more critical in governance networks [6,7].

Resilience is an interdisciplinary concept, defined and operationalized in various ways, depending on the research area [10,13]. The first works on resilience were conducted in ecology in the 1970s [37]. In the social sciences, research on resilience began in 1988 with the work of Wildavsky, who defined resilience as the “capacity to cope with unanticipated dangers after they have become manifest, learning to bounce back” [38] (p. 85). In his view, resilience is a fundamental, dynamic and evolutionary capacity to deal with risk issues in society. Further development of research on resilience in the social sciences has been devoted to high-reliability organizations, where the ability to cope with threats is more important than operational efficiency [39,40,41].

Nowadays, resilience is understood as a multi-level dynamic capability to meet emerging challenges and implement functions and goals in turbulent conditions. This article uses the resilience definition of Williams et al. [3] (p. 742) according to which it is “the process by which an actor (i.e., individual, organization, or community) builds and uses its capability endowments to interact with the environment in a way that positively adjusts and maintains functioning prior to, during, and following adversity”. These capabilities allow actors to return to stability, adapt to new operating conditions, learn from experience, and introduce changes and innovations [8,9]. Thanks to them, actors can survive turbulence and threats, bounce off them, maintain or improve their development direction, and even improve their operations.

The contemporary approach to resilience also emphasizes its process nature. For example, Ma et al. [42] (p. 254) identified three dimensions of resilience: anticipate and plan, manage and survive, and learn and grow. Duchek et al. [14] and Duchek [12] used three dimensions of organizational resilience, which are: (1) anticipation includes observation, identification, and preparation; (2) coping which consists of accepting the situation, developing and implementing solutions; and (3) adaptation that is reflection, learning, and change. These process-based approaches to resilience enrich its understanding and emphasize the manifestations of its dynamics [3,8,12,23]. As a result, the following resilience dimensions are adapted in this paper:

• Coping—the capability to develop and implement solutions to a specific problem [14] to anticipate and prevent future challenges [23,38,40]. It is also the capability to deal with emerging challenges together. Coping can relate to the first stage of dealing with threats—planning and preparing activities.
• Adaptation—the capability to function in new operating conditions that strengthen the ability to react quickly and flexibly to threats; it is based on mutual learning processes and can lead to transformation [7,38]. It is an ability that occurs during and immediately after the threat.
• Transformation—the capability to implement changes practically, transform organizational structures, create new connections, develop and apply new operating strategies [14,39,40]. According to Folke [43], transformation is possible through interactions between different entities. It is a capability that manifests itself in the long term, resulting from the adaptation processes.

It is also noteworthy that resilience is cross-scale in nature. This nature shows that resilience is not permanent but changes over time [44]. The level of change results from adaptation cycles, the potential of actors, resilience factors, and internal and external relationships [13]. It is also crucial that the adaptation cycles “are nested within each other across space and time scales” [45] (p. 396). This cross-scale nature of resilience indicates that actors are interrelated through various relationships and interactions and emphasizes the need to build resilience at the organizational level and in inter-organizational settings. Linnenluecke has recognized this need and writes that “inter-organizational structures should be designed for resilience, recognizing that organizations are not entities operating in isolation” [10] (p. 25). This phenomenon is significant in local governance networks because changes in one actor may cascade onto other actors [45]. Resilience could help cope with such challenges, finding a new way to deliver public services. Moreover, it could affect the effectiveness and innovativeness of the everyday governance processes [6,7,11,17].

2.2. Social Capital in Building Resilience

Considering that in local governance networks, the development of value is based on inter-organizational relationships, social capital seems to be one of the fundamental determinants of their resilience. It allows us to achieve much more thanks to the combination and use of resources dispersed across many units, developing mutual relationships, fostering innovations, and building a common perception of the situation. It develops over time, builds structured relations between actors, increases the possibilities of acquiring, assimilating, and using knowledge in inter-organizational relationships [46], and intensifies learning processes [47] to foster innovation. It facilitates collaboration, joint problem-solving, and the achievement of common goals [16,48]. Straub et al. [30] compare social capital to pathways, which link individual and institutional actors and their resources, and push them into joint action. In this article, social capital is understood as “the shared knowledge, understandings, norms, rules, and expectations about patterns of interactions that groups of individuals bring to a recurrent activity” [49] (p. 176).

Research on the effects of social capital on resilience is just emerging. They are carried out in particular in the area of emergency management [16,50,51,52], community resilience [15,53], and tourism management [18]. Aldrich and Meyer [15] proved that social capital plays a more significant role in building resilience than physical infrastructure. Chan et al. [52] found that this capital is the basis for collaboration, innovation, and partnership, and it mobilizes joint action and deepens the ties connecting individual entities. On the other hand, Lee [53] pointed to the need to deepen research on the influence of social capital on resilience because the relationships between these constructs are complex.

One of the most popular classifications of social capital that could be useful in resilience research divides this capital into structural, cognitive, and relational dimensions [46,54]. Structural social capital covers the number of nodes, their arrangement in the network, and the connections between them. Cognitive social capital refers to a common language, codes, and narratives used in communication, principles of joint activities, vision, and values. Relational social capital describes the closeness of relations between nodes in networks, including the level of trust and mutual respect. Based on this classification, Johnson and Elliott [51] analyzed resilience sources within a public–private partnership and found that primarily structural linkages enhanced by relational factors allow a creative response. In turn, the analysis by Jia et al. [16] related to the functioning of private companies after an emergency indicates that structural capital is of greater proactive importance, relational capital only has a reactive effect, and cognitive capital has no significant impact on both reactive and proactive resilience. Based on the above, it is reasonable to suppose that dimensions of social capital can also stimulate resilience in local governance networks. In this regard, it was assumed that:

Hypothesis 1.

Social capital of members in local governance networks positively and significantly influences these networks’ resilience.

2.3. Influence of Transformational Leadership on Resilience

After the Van Wart [55] appeal, research on public leadership is highly intense. This great interest comes from the opportunities that leadership creates in building the effectiveness and performance of public organizations. Undoubtedly, leadership in local governance networks is also essential to ensuring that actors engage in cooperative activities and help each other [56,57].

Over the years of analysis, researchers have identified many leadership styles, e.g., transactional, transformational, charismatic, shared, integrative. This article focuses on transformational leadership because it is the dominant style of leadership [27,28], constitute a basis of a relational approach to leadership [29], largely coincides with integrative and servant leadership [27,58], and applies to public organizations [27,29]. Transformational leadership is the first positive leadership style to replace the classic approach in this research field. It was introduced by Burns, who defined it as “a process where leaders and their followers raise one another to higher levels of morality and motivation” [59] (p. 20). This approach was further developed by Bass (1999), proving that transformational leadership is a way to encourage employees to change attitudes and values, develop and engage in activities.

Research on leadership in the public sector considers both the perspective of general leadership theory [60] and its specificity in public governance [56]. Scholars use, among others, the scale developed by Carless et al. [61], based on Podsakoff et al. [62]. It consists of seven dimensions: vision, staff development, supportive leadership, empowerment, innovativeness, leading by example, and charisma. Similar approaches are in the studies of Mary [63], Harland et al. [24], and Ng [28]. Empirical research confirms the strong impact of transformational leadership on the attitudes and commitment of employees, their job satisfaction, and the results achieved by organizations [27,63,64]. These positive effects are since transformational leaders translate the needs of others above their own, dedicate themselves to the benefit of the organization and followers, and act as integrators connecting the actions of individuals to achieve a common goal.

By influencing social capital, transformational leadership builds organizational resources [25,26]. As role models, transformational leaders formulate a clear agenda for change, shape motivating future visions, stimulate exchange of information, and create knowledge [26]. As a result, they develop a common platform for exchanging knowledge and experiences between people from different organizations, strengthening ties between them, and encouraging them to develop skills and work together [65]. It leads to the assumption that:

Hypothesis 2.

Transformational leadership positively and significantly influence social capital in local governance networks.

Sutcliffe and Vogus [23] emphasize that leadership is an essential factor influencing resilience. Harland et al. [24] explored the relationship between transformational leadership and subordinate resilience based on a questionnaire with 150 part-time MBA students. They stated that this type of leadership is a driving force that takes advantage of crises and can turn them into development opportunities [24]. Their results show a direct and positive impact of transformational leadership on overcoming difficulties by employees. Transformation leaders also have the opportunity to build resilience in inter-organizational relations by supporting joint activities, searching for elements that bind organizations together, improvising, and encouraging creativity [21,22,66,67]. These rationales allow us to assume that:

Hypothesis 3.

Transformational leadership positively and significantly influences local governance networks’ resilience.

The summary of key findings on resilience theory from the literature review is in

Table 1. The summary of key findings on resilience.

Reference	Purpose	Method	Main Findings
Duit, 2016 [6]	clarification of how social-ecological resilience research can influence studies on resilience in public administration	literature review	identification of key shortcomings of social–ecological resilience thinking
Linkov and Trump, 2019 [7]	unpacking what resilience governance entails and how it has been understood in various governance’ paradigms and practices	literature review	indicate the future development directions of resilience governance
Hillmann and Guenther, 2021 [8]	pointing out conceptual problems of resilience for its future development and conceptual clarity	systematic literature review	discussing reasons for resilience criticizing and identifying the lens of its development and taxonomies
Popp and Nowack, 2020 [9]	analyzing the capacity of a dairy system in Northwest Germany to resist, adapt or transform in response to external challenges	in-depth case study	an indication that insurances and savings are relevant for robustness against short-term shocks; resilience-enhancing attributes such as cooperation and knowledge transfer beyond boundaries contribute to the long-term improvement of resilience
Linnenluecke, 2017 [10]	identification of the resilience theory development and gaps in this research area in knowledge in business and management	systematic literature review	identification of five research streams in resilience research, discussion of similarities and differences between these streams, and ways of resilience operalization
Barata-Salgueiro and Guimarães, 2020 [11]	uncovering the path of coping with retail challenges in cities for increasing cities’ resilience and sustainability, including the role of public policy	case study	identification of a close link between public initiatives, private entrepreneurship, and urban resilience
Duchek, 2020 [12]	“deepening the understanding of the complex and embedded construct of organizational resilience”’ (p. 215)	literature review	conceptualization of resilience as a meta-capability and its decomposition into parts and stages: anticipation, coping, and adaptation; overviewing resilience capabilities and interactions in each of these stages
Hillmann, 2021 [13]	investigation of causes for resilience ambiguity and disciplines that shaped this concept	systematic literature review	identification of five disciplinary perspectives on resilience research, their ontologies, research tools and methods, and highlighting the perspectives of future research
Duchek et al., 2020 [14]	analyzing the role of diversity in the development of organizational resilience and explaining the mining of resilience-enhancing diversity	literature review	developing a theoretical framework of resilience-enhancing diversity, formulating propositions, and discussing implications for further research
Jia et al., 2020 [16]	examination of the influence of social capital on organizational resilience	case study	finding that “structural capital improves proactive organizational resilience, relational capital only improves reactive organizational resilience” (p. 1).
Zeng et al., 2022 [17]	clarification of the urban resilience concept and developing its key indications	systematic literature review	identifying “key indicators of urban resilience under three major components like adaptive capacity (education, health, food, and water), absorptive capacity (community support, urban green space, protective infrastructure, access to transport), and transformative capacity (communication technology, collaboration of multi-stakeholders, emergency services of government, community-oriented urban planning) [...] [and] several indicators under major dimensions (social, economic, and environmental) of urban sustainability” (p. 1)
Chowdhury et al., 2019 [18]	evaluating “the relationship between social capital (structural, relational and cognitive) and organizational resilience as predictors of business performance” (p. 1209)	postal survey	finding that structural capital has a positive influence on cognitive and relational capital that affects adaptive resilience
Valero et al., 2105 [21]	examination of the influence of transformational leadership on organizational resilience	survey	finding that transformational leadership positive and significant influences organizational resilience
Sutcliffe and Vogus, 2003 [23]	examination of “the roots, the mechanisms, and the future of the study of resilience as an emerging integrative concept” (p. 95)	literature review	finding “that resilience emerges from ordinary factors that manifest in nontraditional ways that promote competence, restore efficacy, and encourage growth” (p. 110)
Harland et a., 2005 [24]	evaluating “the relationship between leader behaviors and subordinate resilience” (p. 2)	questionnaire	determination that “participants who mentioned their leaders as a positive factor in dealing with the situation exhibited greater resilience than participants who did not” (p. 2)

.

3. Materials and Methods

The research on the resilience of local governance networks is based on the questionnaire carried out in Poland at the local level in December 2019 and January 2020 among random selected Polish counties and cities with county rights. There are 314 land counties in Poland, covering several to several dozen neighboring communes and 66 cities with county rights (magistrate counties) [68]. Polish counties make decisions and provide public services in partnership and collaboration within governance networks [69,70,71]. However, collaboration in such networks is extremely challenging, and many problems appear in this area, e.g., the dominant discourse, difficulties in finding unambiguous solutions, and lack of inter-organizational trust [72,73]. Moreover, counties focus on implementing supra-communal issues. In the opinion of the co-founder of the Public Administration Reform in Poland in 1990 and 1998, this level of state organization appears to be the most appropriate for researching governance issues [74] (p. 354). Therefore, Polish counties have experience in conducting activities within governance networks and are organized in a way that enables effective public service delivery. For this reason, they are a case that allows the identification of significant dependencies supporting the development of governance networks.

The data collection was based on the CAWI method (Computer-Assisted Web Interview) in this research. This method is an information-gathering technique in quantitative research in which the respondents complete an electronic questionnaire. The questionnaire consisted of statements assessed on a 5-point Likert scale. The statements adapted in the questionnaire are based on the literature review presented in the theoretical background section. Definitions of each variable used in the empirical analysis are in

Table 2. Definitions of variables used in the empirical research.

Construct	Dimension	Definition
Resilience	Coping	the capability to develop and implement solutions to a specific problem [14] to anticipate and prevent future challenges [23,38,40]
	Adaptation	the capability to function in new operating conditions that strengthen the ability to react quickly and flexibly to threats based on mutual learning [7,38]
	Transformation	the capability to implement changes practically, transform organizational structures, create new connections, develop and apply new operating strategies [14,39,40]
Social capital	Network ties	social relations forming information channels and providing access to resources [54] (p. 252)
	Network configuration	configuration of network ties developing intellectual capital [54] (p. 252)
	Appropriable organization	an exemplary form of organization that can be adapted in other social settings and is a source of valuable resources [54] (p. 253)
	Shared language and codes	“group-specific communication codes” [54] (p. 254)
	Shared narratives	myths, stories, and metaphors leading to enriching sets of meanings [54] (p. 254)
	Trust	willingness to be vulnerable to another parties’ behavior resulting from belief in these parties’ good intent, their competence and capability, reliability, and perceived openness [54] (p. 254)
	Norms	“the socially defined right to control an action is held not by the actor but by others” [54] (p. 255)
	Obligations and expectations	“a commitment or duty to undertake some activity in the future” [54] (p. 255)
	Identification	identifying with a group of people or another person [54] (p. 256)
Transformational leadership	Vision	“communicates a clear and positive vision of the future” [61] (p. 396)
	Staff development	supporting and encouraging staff [61] (p. 396)
	Supportive leadership	valuing and appreciating staff [61] (p. 396)
	Empowerment	creating working conditions based on trust, commitment and cooperation [61] (p. 396)
	Innovativeness	encouraging staff to solve problems in a new way and to seek novelty [61] (p. 396)
	Lead by example	proving recognized values through practice [61] (p. 396)
	Charisma	commanding respect and inspiring to action [61] (p. 396)

.

The statements in questionnaire are concerned opinions and experiences of public servants on implementing activities in networks. They are presented in

Table 3. The statements in the questionnaire.

Construct	Dimension	Statement in the Questionnaire	Sources
Resilience	Coping	Organizations in our network can cope with emerging challenges together	[10,14,23,38,40]
	Adaptation	Organizations in our network adapt to new operating conditions quickly and respond to threats flexible	[7,38,42,43]
	Transformation	Organizations in our network transform network structure and strategies, leading to its development	[14,39,40,42,43]
Social capital	Network ties	Our network ties provide access to resources (e.g., information, knowledge)
	Network configuration	Configuration of our network facilitates information and knowledge transfer and adaptation to circumstances
	Appropriable organization	The organization of our network activities ensures access to resources, exchange, and flexible everyday activities
	Shared language and codes	Organizations in our network come to an understanding and develop thanks to shared language and codes	[54]
	Shared narratives	Shared history and experience of the organizations in our network affect mutual openness and the development of practices
	Trust	Organizations in our network are trustworthy
	Norms	Existing social norms in our network increase the reliability of the activities’ implementation
	Obligations and expectations	Organizations in our network engage in joint activities and provide mutual assistance
	Identification	Organizations in our network form a community, identify with this network and collaborate
Transformational leadership	Vision	Leaders in my organization indicate perspectives that we can achieve through collaboration with other organizations
	Staff development	Leaders in my organization facilitate and encourage personal development based on relationships with other organizations
	Supportive leadership	Leaders in my organization recognize individual staff’s achievements in collaboration with other organizations
	Empowerment	Leaders in my organization involve team members in decision making related to collaboration with other organizations	[61]
	Innovativeness	Leaders in my organization have many ideas on how to increase the effectiveness of collaboration with other organizations
	Lead by example	Leaders in my organization can combine internal tasks with tasks resulting from collaboration with other organizations
	Charisma	Leaders in my organization meet the standards and norms of conduct they set for their staff

Source: own elaboration.

.

The research was conducted among public servants at the local level who are responsible for maintaining relationships across organizational boundaries. The research was preceded by a cover letter explaining its purpose and governance network concept, and a request was made to assess the inter-organizational relations between a given entity and other public organizations. The request to complete the questionnaire was directed to county chairpersons, who assigned it to competent persons in their office responsible for maintaining relationships with other organizations. The questionnaire was completed by specialists (30.2%), chiefs/directors/heads of department (29.1%), inspectors (28.1%), city/county secretaries (10.1%), deputy mayors/city presidents (2.5%). Their seniority in public administration was 2–3 years (5.5%), 4–10 years (14.1%), 11–20 years (51.3%), and over 20 years (29.1%). These were people with secondary education (3%), Bachelor/Engineer degrees (6%), Master’s degrees (89%), and PhD degrees (2%). As a result, 199 correctly completed questionnaires were obtained, which, with a materiality level of α = 0.05 and a permissible error of e = 5%, is a representative research sample. The sample included 43.6% counties and 33.3% cities with county rights. The analysis of differences in responses between counties and cities with county rights showed insignificant differences. For this reason, the analyses of the obtained results were carried out jointly for both groups.

The applied approach is two-stage and includes measurement and structural model assessments performed using the SmartPLS software [75,76] It is software for variance-based structural equation modelling (SEM) based on the partial least squares path modelling method (PLS). The Smart PLS could analyze standard and additional assessment criteria (e.g., the HTMT criterion, bootstrapping, the goodness of fit, multigroup analysis, confirmatory analysis, mediation). Thanks to its application, it was possible to test the complex relations between resilience, social capital, and transformational leadership in local governance networks.

4. Results

4.1. Measurement Model Assessment

The path analysis starts by verifying the construct’s consistency, the adopted scale’s reliability, and its adequacy to the tested model. For this purpose, it was checked whether the outer weights for a given factor are the highest in the case of the appropriate construct (

Table 4. Cross loadings.

Dimension	Transformational Leadership	Social Capital	Resilience
Charisma	0.775	0.644	0.562
Empowerment	0.678	0.506	0.557
Innovative thinking	0.699	0.561	0.530
Lead by example	0.732	0.626	0.510
Staff development	0.526	0.405	0.418
Supportive leadership	0.603	0.449	0.496
Vision	0.732	0.585	0.557
Appropriable organization	0.501	0.534	0.374
Identification	0.601	0.782	0.682
Network configuration	0.562	0.625	0.462
Network ties	0.261	0.292	0.218
Norms	0.560	0.700	0.587
Obligations	0.413	0.604	0.577
Shared codes and language	0.557	0.728	0.635
Shared narratives	0.620	0.790	0.675
Trust	0.609	0.768	0.650
Adaptation	0.655	0.720	0.858
Coping	0.466	0.489	0.591
Transformation	0.597	0.679	0.800

Source: own elaboration based on [76].

). This condition has been met.

Then, the following measures were used to test the reliability of the analyzed model: Cronbach alpha, rho_A, composite reliability, and average variance extracted AVE. For the first three measures, the acceptable minimum is 0.7, although the lower reliability limit of 0.6 is also possible in exploratory research in the case of Cronbach alpha [77,78,79]. The AVE coefficient assesses the convergent validity [79]. This measure shows what a given construct accounts for the proportion of the own indicators variance.

The obtained AVE measures in transformational leadership and social capital constructs were not met and indicated outliers in the analyzed model (

Table 5. Outer loadings and construct reliability and validity.

		First Iteration					Final Iteration
		Loading	Cronbach’s Alpha	rho_A	Composite Reliability	Average Variance Extracted (AVE)	Loading	Cronbach’s Alpha	rho_A	Composite Reliability	Average Variance Extracted (AVE)
Transformative leadership	Charisma	0.775	0.852	0.864	0.858	0.466	0.804	0.872	0.876	0.872	0.534
	Empowerment	0.678					0.707
	Innovative thinking	0.699					0.726
	Lead by example	0.732					0.749
	Staff development	0.526					Omitted
	Supportive leadership	0.603					0.629
	Vision	0.732					0.757
Social capital	Appropriable organization	0.534	0.855	0.889	0.871	0.441	Omitted	0.782	0.817	0.798	0.573
	Identification	0.782					0.805
	Network configuration	0.625					0.619
	Network ties	0.292					Omitted
	Norms	0.700					0.722
	Obligations	0.604					0.634
	Shared codes and language	0.728					0.762
	Shared narratives	0.790					0.828
	Trust	0.768					0.789
Resilience	Adaptation	0.858	0.782	0.820	0.799	0.575	0.848	0.892	0.900	0.894	0.549
	Coping	0.591					0.598
	Transformation	0.800					0.801

Source: own elaboration based on [76].

-first iteration). Therefore, the outlier factors were removed based on the rules in the method used and Hair et al.’s [79] recommendations. Deletion is not advisable for dependent variables. In the construct of transformational leadership, staff development (0.526) was removed, and in social capital–appropriable organization (0.534) and network ties (0.292). Table 3 presents outer loadings after the first and final iterations. As a result, AVE for transformational leadership increased to 0.534, and for social capital to 0.549.

Additionally, the model’s discriminant validity was verified using the heterotrait–monotrait (HTMT) ratio of the correlations [79]. The obtained results were as follows: transformational leadership–social capital (0.741), transformational leadership–resilience (0.750), social capital–resilience (0.842). The above results confirm that the criteria for examining the measurement model are met.

4.2. Structural Model Assessment

The assessment of the structural model concerns the relationship between exogenous and endogenous variables. The criteria used to carry out this assessment are the coefficient of determination R2 and the level and significance of the path coefficients [78,79,80]. However, before using these criteria, scholars should test the collinearity statistics (VIF) to see if they affect the regression results [79].

Table 6. Collinearity statistics (VIF).

Transformational Leadership Factors	VIF	Social Capital Factors	VIF	Resilience Factors	VIF
Charisma	2.783	Identification	2.270	Adaptation	2.630
Empowerment	1.794	Network configuration	1.389	Coping	1.295
Innovative thinking	2.686	Norms	2.111	Transformation	2.414
Lead by example	1.543	Obligations	1.890
Supportive leadership	1.599	Shared codes and language	2.340
Vision	2.041	Shared narratives	3.012
		Trust	2.089

Source: own elaboration based on [76].

presents the obtained results.

After obtaining the correct values of the VIF coefficient, the next step is to estimate the determination coefficient R2, which explains the variance of endogenous latent variables. Its value at the level of 0.75 is significant, 0.5 is moderate, and in the case of 0.25 isweak [78]. The R2 value is 0.718 for resilience and 0.549 for social capital in the analyzed model. These are the results that prove the model’s in-sample explanatory power. Next, the estimation of the Q2 value is needed to confirm the model’s predictive accuracy. In the tested model, the obtained Q2 value was 0.371 for resilience and 0.253 for social capital, which allows us to conclude that the model has a high degree of predictive relevance about these endogenous factors.

The structural model assessment procedure also needs to verify the model’s predictive power. The PLSpredict is useful for making these calculations, and in this research, k = 10 was set [77]. Verifying the Q2predict indicator and examining the prediction statistics were the basis for analyzing the obtained results. This examination uses benchmarking of the root mean squared error (RMSE) for the PLS-SEM model and a linear regression model. The results are in

Table 7. Prediction summary.

	PLS-SEM Model		Linear Regression Model
	RMSE	Q²_Predict	RMSE	Q²_Predict
Coping	0.812	0.180	0.830	0.143
Adaptation	0.532	0.331	0.532	0.331
Transformation	0.589	0.272	0.595	0.256
Network configuration	0.604	0.205	0.618	0.169
Shared codes and language	0.597	0.262	0.608	0.234
Shared narratives	0.592	0.329	0.605	0.299
Trust	0.586	0.288	0.571	0.325
Norms	0.525	0.240	0.515	0.267
Obligations	0.793	0.133	0.809	0.099
Identification	0.578	0.281	0.585	0.266

Source: own elaboration based on [76].

.

The Q2 predict is between 0.133 and 0.331 in the PLS-SEM model and between 0.099 and 0.331 in the linear regression model. These results prove the medium and large predictive relevance of variables. On the other hand, the comparison of RMSE shows that only in two cases—trust and norms—the value of the analyzed indicators were higher in the PLS-SEM model than in the linear regression model, and the remaining values are lower. These results mean that the model’s predictive power is greater than medium, nearly high [79].

The next step in the analysis was bootstrapping needed to assess the significance of the path coefficients. In this research, bootstrapping utilizes 5.000 subsamples and a significance level of 0.05. The critical t-value for a two-tailed test is 1.96 at the established significance level. The results obtained from the conducted analyses are in

Table 8. Path coefficients.

	Original Sample (O)	Sample Mean (M)	Standard Deviation (STDEV)	T Statistics (|O/STDEV|)	p Values
Social Capital -> Resilience	0.632	0.635	0.113	5.599	0.000
Transformational leadership -> Resilience	0.266	0.260	0.122	2.173	0.030
Transformational leadership -> Social Capital	0.741	0.736	0.060	12.304	0.000

Source: own elaboration based on [76].

.

“Original Sample” and “Sample mean” in Table 6 evaluate the value of the path coefficients’ significance, which fall in the range of −1 and +1. The results show the great importance of the direct influence of transformational leadership on social capital and social capital on resilience. The direct influence of transformational leadership on resilience is relatively weak. However, the obtained results also indicate an indirect effect of influence between these constructs at 0.468. As a result, the total effect of this relation is 0.734. The critical t-value for each researched construct is at the proper level, and p-value is also appropriate. A graphic representation of the obtained results is shown in Figure 1.

Finally, the fit model was verified using the standardized root mean square residual (SRMR), which measures the approximate fit of the model. The SRMR value in the analyzed model was 0.05, proving the model’s good fit to the data.

5. Discussion

There are many reasons for the growing importance of resilience in governance networks. One can mention here the need to deal with the challenges of maintaining and developing inter-organizational collaboration, social problems complexity, turbulences, threats, and crises appearing on an increasing scale. For example, the past global financial crises and the COVID-19 pandemic have exposed the weaknesses of governments and limited the delivery of public services. There is, therefore, a need to pay more attention to building resilient governance networks.

However, as still an umbrella concept, resilience is only at the paradigm-shaping stage, and much research is needed to understand, evaluate and develop it [8,14]. Existing research on resilience in the public sector underlines that resilience is essential not only in emergencies but in the day-to-day operation of public organizations and networks [6,7].

The results presented in this article indicate that the responders view resilience in local governance networks as an opportunity for development, proactive thinking about the future, learning, and transforming the current action methods. It shows that respondents perceive resilience more as adaptation and transformation than coping. It is not that coping capabilities are unnecessary. A possible explanation for the low outer loading of coping is that actors in local governance networks understand the inevitability of changes and conform to these changes and evolution. This finding has positive connotations as it means that local governance networks are focused on development, not just return to stabilization.

Some scholars see social capital as one of the fundamental factors of resilience [12,23]. This research explores this impact at the network level. The results of outer loadings indicate that appropriable organization and network ties are not consistent with the construct. In addition, network configuration reached low but acceptable values of outer loadings (0.625). It allows us to conclude that in the analyzed model, the factors of structural social capital do not significantly impact the resilience of governance networks. These results differ from the findings of Jia et al. [16], who conducted research among companies affected by the Sichuan earthquake in 2008 and found that structural social capital contributes to strengthening proactive resilience. This difference may be due to the level of analysis. Jia et al. [16] carried out their research at the organizational level among private companies. In turn, this research was conducted at the network level. Therefore, there is a need to deepen the analysis of the relationship between different types of social capital and organizational resilience regarding the conditions, scope, and context of the research conducted. Another possible explanation comes from “the strength of weak ties” of Granovetter [81] and the “structural holes” of Burt [82]. These “weak ties” and “structural ties” connect different groups, enabling the flow of information between them and creating more opportunities to deal with difficult situations. Therefore, one can conclude that relational and cognitive social capital could enable the building of resilience by using and strengthening existing links, mutual adjustment, and the development of shared situational perception. The path coefficient analysis confirms it and indicates that the factors of social capital included in the research, after verifying their reliability, significantly impact the resilience of governance networks (0.632). This result is consistent with previous studies [12,15,23,48] and supports Hypothesis 1; this relationship is positive and significant.

In transformational leadership, one factor—staff development—was inconsistent with the construct. It is a factor added by Carless et al. [61] to the scale of Podsakoff et al. [62]. It seems that staff development can be associated to a greater extent with servant leadership, which, according to Bass [83], has many similarities to transformational leadership. Servant leadership, however, is more oriented toward serving, supporting, and developing followers. Transformational leadership motivates followers to act [64]. The results obtained may suggest that the involvement of people in joint activities is more important in building network resilience than the development of their above-average skills. The outer loadings of other factors indicate these findings. Such factors of transformational leadership as charisma (0.775), lead by example (0.732), and vision (0.732) obtained the highest values. Thanks to such features, transformational leaders can change their followers’ basic values, beliefs, and attitudes, inspire them to identify with network activities, and achieve common goals. These results are consistent with the theories reported in the literature [29,62,64]. They also confirm the statement of Folke et al. [84] (p. 451) that resilience-oriented leadership is about “... building trust, making sense, managing conflict, linking actors, initiating partnership among actor groups, compiling and generating knowledge, and mobilizing broad support for change”.

However, the significance of the path coefficient indicates that transformational leadership has a weak impact on resilience (0.266). On the other hand, its indirect effect is at 0.468, resulting in a strong indirect impact on resilience at 0.734. These results complement the studies by Valero et al. [21], who found that transformational leadership has a positive and statistically significant effect on the resilience of public and non-profit organizations. This study clarifies that transformational leadership has a strong but indirect influence on the resilience of local governance networks. These results partially confirm Hypothesis 3, as the direct effect is positive and weak, but the indirect effect is positive and significant. The path coefficient also indicates that the relationship between transformational leadership and social capital is very high (0.741), confirming Hypothesis 2. These results are consistent with the previous research findings [25,26].

6. Implications

The findings presented in this article are valuable for several reasons. First, they enrich the outcomes of Weick et al. [40], pointing out that cognitive and relational social capital could create common-sense making and mindfulness. Transformational leadership may foster this process by linking organizations and developing relational and cognitive social capital. Second, the findings confirm the evolution of the understanding of resilience and that there is currently more emphasis on transformation and development than on bouncing back from unanticipated threats. The results show that the respondents understand resilience in local governance networks as a possibility of adaptation and transformation rather than coping. Finally, these findings contribute to the unexplored theory of resilience in public governance [6,7] by explaining how to use social capital and transformational leadership to strengthen the resilience of local governance networks.

The practical implications of the findings demonstrate that public leaders should focus primarily on the relational dimensions of social capital. They should bear in mind the structural and cognitive dimensions of social capital, which also affect the resilience of local governance networks. Building resilience in local governance networks depends on relational factors and appropriate organizational arrangements and big picture creation. Therefore, public leaders should take an inter-organizational approach to build resilience, emphasizing the development of relational factors.

In reference to the above findings, managerial implications indicate that building the resilience of local governance networks has a very complex nature. Social capital is one of the factors influencing resilience significantly, primarily relational and cognitive capital. Appropriate relations between actors and the alignment of inter-organizational dependencies should also be developed to build the resilience of local governance networks. Choosing the right leader could help in achieving these requirements. The role of leaders is to create collaborative circumstances, and they have an indirect impact on the resilience of local governance networks.

7. Limitations

The results obtained are not free from limitations. Primarily, the research was conducted only in Poland. It would be advisable to conduct research in other countries among local governance networks and compare the results. The questionnaire was also conducted only among public servants. It would be helpful to learn the views of other participants in public service delivery processes on the factors influencing resilience building.

Limitations also result from the research implementation process. First, this research covered only the county level. Due to the cross-scale nature of resilience, taking into account the commune, voivodeship, and central level could help establish additional dependencies between analyzed dimensions. Therefore, the research could cover the remaining levels of the state organization in the future. Second, the research was carried out at a distance, making it impossible to answer responders’ potential questions and clarify the statements in the questionnaire. Last but not least, the questionnaire was filled in by chairpersons and people in various positions, selected by chairpersons, who could answer the questions through the prism of their workplace. This approach could also limit the obtained results.

The assurance of endogeneity of the conducted studies was based on careful consideration and selection of research variables based on adapted scales to predict relationships between constructs researched [85,86]. However, it is possible that additional factors have not been included in this study. Moreover, reverse causality is also possible, according to which resistance could affect social capital, and strong social capital could stimulate transformational leadership development. This article does not explore these two directions of causality, but it does not rule them out. For this reason, in the future, there is a need for two-way verification of variables included in these studies, also in other contexts, or with the use of qualitative methods, e.g., case studies, in-depth interviews or focus groups.

However, this article is one of the few empirical studies on building resilience in local governance networks and could be useful in conducting future research and developing this prospecting research area.