Re-examining the link between relational governance and project management performance: an integrated framework

ABSTRACT

Relational governance has showed great potential in addressing many of the construction industry’s problems, such as opportunism and agency problems. While multiple research efforts have been directed at examining different roles of relational governance elements, few studies paid attention to show the link between different relational governance elements and project management performance. So, this study aims to address such gap by fully considering relational governance elements affecting project management performance. To this end, this paper (1) divided relational governance elements into relational situation, relational norm, and relational behavior for the first time, deepening our understanding of relational governance elements in the construction industry; (2) determined the impact of these relational governance elements on project management performance. The findings contribute to the understanding of relational governance elements in the construction industry, how relational governance and project management performance are really linked, and what the fundamental paths are in the chain by specifying the elements of relational governance in a more detailed way. These findings also give owners and contractors insight into improving project management performance.

KEYWORDS:

• Project management performance
• relational situation
• relational norm
• relational behavior

1. Introduction

The continuous improvement of construction project management performance has always been the goal pursued by project management researchers and practitioners (Li, Yin, and Zhang 2022). Here, projects are not the core business of owners and projects just extend owners’ core business according to Winch’s three domains model (Winch 2014). Since the 21st century, the world construction industry is moving towards a new stage, which makes construction projects larger in scale, more intelligent, and more diversified in attribute. Compared with other periods, the goal of continuous improvement in project management performance has become more urgent. Researchers have witnessed transformed research paradigms focusing on the hot topic of “how to improve construction project management performance”. During the research paradigms transforming, four research paradigms appeared in turn, namely, technical paradigms, organizational paradigm, contractual paradigm, and relational paradigm (Li, Yin, and Zhang 2022). With the wide application in project management practice, formal contract has shown some defects in project management practice, such as lack of flexibility (Camén et al. 2020), which possibly leading low project management performance and even causing project failure. Thus, relational governance mechanism has attracted increased attention.

As one of the frontiers in construction project management research field, relational governance depends on the informal structure and self-enforcement mechanism of partners (Abdi and Aulakh 2017; Chi et al. 2017; Li et al. 2021; Mirkovski, Lowry, and Feng 2016) and is regarded as an effective mechanism to the contractual hazards or opportunism (Macneil 1977). Relational governance emphasizes transacting in markets through the mechanism of mutual trust and relational norms (Benítez-Ávila et al. 2018; Li et al. 2021). Given that relational governance in most eastern countries, such as Japan, China, and India, has been deeply embedded in their cultures (Lin et al. 2021), guanxi is one of the major dynamics and is the delicate fibers woven into every individual’s social life in these countries (Dong, Ma, and Zhou 2017). Many researchers agree that relational governance plays an active role in improving project management performance in these Confucianism-based societies (eg, Jiang, Zhao, and Zuo 2017; Li, Yin, and Zhang 2022). However, existing research is controversial about the content of relational governance, and the types of relational governance elements are still unclear. Much research pointed out that relational norms such as trust, loyalty, and shared values are important parts of relational governance (ie, Benítez-Ávila et al. 2018; Jeong, Ha, and Lee 2021; Vitasek, Winn, and Nickel 2020). Dekker (2016) also pointed out that prior research on inter-organizational relationships has predominantly focused on control choices and relational controls at the firm level, largely ignoring the effects of boundary spanners’ relational behavior. Here, relational behavior refers to the desired actions involved in the exchanges that promote the development of a collaborative relationship (Zheng, Lu, and Chang 2019). Much research also pointed those relational behaviors such as flexibility, information exchange, and solidarity are also important in project management practice. For example, Zheng et al. (2018) pointed out that inter-organizational behavior can determine the effective adoption of relational mechanisms in the context of construction projects. However, there are no clear studies examining the link between relational behaviors and project management performance. Finally, Gillath and Shaver (2007), and Funder (2006) pointed that behavior is often conceptualized as a joint function of personality and situation, but the principles by which traits and situations jointly shape behavior are still unclear. So, it is necessary to explore how project partners evaluate and select relational behaviors under different situations. This is an important issue for relational governance researchers, because partners in cooperation relationships are often reacting to changing situations, commitment cues in their partners’ behaviors, and comments on the relationship. However, current research has not studied the situation factors rooted in transaction relationships between contracting parties (Lu et al. 2019). Inter-organizational project differed from other organizations for its temporary characteristic (Winch 2014) and requires a cooperation atmosphere during project implementation (Kosmol, Reimann, and Kaufmann 2018). The situation complexity makes it impossible for contracting parties to integrate all relationships with a same way. Recently, Cantù et al. (2021) divided relational governance elements into relational norms and relational activities. Moreover, they pointed out that relational atmosphere can influence joint actions (ie, relational activities). However, there are no clear studies showing the link between these relational governance elements and project management performance, causing the insufficient revealing to the relational governance effect.

Consequently, this study aims to address these gaps and contributes to relational governance research. Specially, the purpose of this paper is to show the link between different relational governance elements and project management performance. Drawing on the research of Cantù et al. (2021) and Yang et al. (2021) who have divided relational governance elements into relational norm, relational activities, and relational atmosphere, we classify the elements of relational governance into three categories: relational situation, relational norm, and relational behavior. Then, this study constructs an integrated framework including these elements and re-exams the impact of relational governance elements on project management performance. By this way, we can detailly see the relationships between relational governance and project management performance.

The rest of the present study is organized as follows. The next section reviews the theoretical background related to this study. Then, the following section establishes the research model and proposes several hypotheses. Section 4 introduces the research method of a questionnaire survey and shows the results, while Section 5 gives the discussion. This study is ended up with Section 6, consisting of result summary, limitations, and recommendations.

2. Theoretical background

2.1. Relational situation: asymmetric dependence

Resource dependence theory (RDT) points out that interorganizational cooperation strategies can reduce environment uncertainty and resource dependence, which matters for organizational survival and performance (Pfeffer and Salancik 2003). As a situation factor (Wang, Li, and Zhang 2019; Kim and Park 2015), asymmetric dependence is the divergence between firm A’s dependence on firm B and the latter’s dependence on the former (Lawler and Bacharach 1987). In a construction project context, the owner dependents the contractor to finalize construction projects, while the latter dependents the former to provide incomes (Jiang, Lu, and Le 2016). In such situation, the contractor is more dependent on the owner for resources. So, the owner is usually the strong party. The contractor, as the weak party, usually needs to follow the intention of the owner to execute specific work. There is a prominent phenomenon of asymmetric dependence in the dyadic owner–contractor relationship. Because their demand on resources and ability to manage resources cannot be changed dramatically during project implementation, asymmetric dependence will continuously affect project management performance (Brinkhoff, Özer, and Sargut 2015).

2.2. Relational norm: trust

From economics perspective, trust is an expectation: a psychological state comprising the intention to accept vulnerability based upon positive expectations of the intentions or behavior to the another (Jiang and Zhao 2019; Rousseau et al. 1998). In other words, trust is a potential psychological state caused by interactive behaviors or choices, and then affects the subsequent interactive behaviors or choices. Among relational governance elements, trust is an important factor (Yan and Zhang 2020) which affects contractors’ behavior and is regarded as the core element of relational governance (Ke et al. 2015; Li, Yin, and Zhang 2022; Yang, Cheng, and Zhang 2021). This is because trust can effectively reduce trade barriers and ideological confrontation between contractors and owners, making up for lack of formal contract, and thus improving project management performance. As a result, the present study regards trust as an important variable (relational norm) in the integrated framework.

2.3. Relational behaviors: communication and coordination

2.3.1. Communication

According to Anderson and Narus (1990), communication refers to the timely and appropriate formal/informal information spreading among communicators. As a relational behavior (Leuthesser 1997; Tsaur, Yung, and Lin 2006), communication often serves as a bridge between enterprises and external environment, and the main aim of using communication instrument is to influence individuals’ perceptions toward companies (Hooghiemstra 2000). According to previous researches, improving in communication can directly and indirectly promote project success in construction project research field (Adenfelt 2010; Kivrak et al. 2008; Li, Yin, and Zhang 2022). Conversely, the difficulties in communication among project partners are an important factor that hindering project management performance improvement (Adenfelt 2010; Hsu et al. 2012). So, we include relational behavior communication into the integrated framework.

2.3.2. Coordination

Scholars focusing on organizational theory pointed out that coordination means integrating or linking together different parts of an organization to accomplish a collective set of tasks (Van de Ven, Delbecq, and Koenig 1976). The foundation of organizational theory is the premise that all organizations need coordination. In construction project management field, coordination refers to the specific activities related to the management of resources with organized manners so that efficient construction and orderly operation in construction projects can be achieved (Hossain 2009). Although the very importance of coordination to project cost, schedule, and quality has attracted much attention from scholars (Chiocchio 2007), related research on inter-organizational coordination among project stakeholders is extremely limited (Liberatore and Luo 2009). Scholars discussed coordination as an important relational behavior related to the project implementation process (Benítez-Ávila et al. 2018; Li, Yin, and Zhang 2022, 2022; Yang, Cheng, and Zhang 2021). So, relational behavior coordination is brought into this study.

2.4. Project management performance

Project management performance and project success are related concepts. Project management performance can be described as the achievement degree of project management goals and objectives from the literature on project management (Bjorvatn and Wald 2018; Demirkesen and Ozorhon 2017; Mathur, Jugdev, and Fung 2013; Pinto et al. 2022). The project management literature also argues that there is a positive relationship between project management performance and project success (Jugdev and Muller 2005; Pinto et al. 2022). Mir and Pinnington (2014) argued that project success and project management performance are distinct yet inter-related concepts and a positive relationship between them is sought. From empirical evidence of the relationship between project management performance and project success, they explain how factors of project management performance can enhance the project success rate (ie, project management performance explained 44.9% of the variance in project success, with a very significant relationship explained by statistical analysis results). So, project success is the goal of project management performance improvement, and project management performance is the indicator of project success. Besides, Mir and Pinnington (2014) pointed out that project management KPIs focus on indicating results achieved in relation to meeting the requirements of project stakeholders. It is important for organizations to have a system that ensures that KPIs are developed from the perspective of all stakeholders and encompass not only short-term benefits (for example, meeting cost, time, and quality objectives of current projects) but also the long-term benefits for the organization (Mir and Pinnington 2014). So, this research proposes survey items about stakeholders when designing the questionnaire.

3. Model development and hypotheses

3.1. Relational situation and project management performance

Scholars point out that high degree of asymmetric dependence (relational situation) will lead to poor cooperation performance. For instance, Villena and Craighead (2017)found that asymmetric dependence will result an unanticipated increase in uncertainty, creating a disproportionate level of vulnerability whereby the larger company may use its bargaining power to gain control over its smaller partner (Crook and Combs 2007). So, the stronger partners (ie, the larger companies) can generally get their way using various means including opportunistic behaviors. In contrast, their counterparts are liable to experience frustration and tension, given a less-than-satisfactory outcome to a conflict episode. As Anderson and Weitz (1989) pointed out, asymmetric relationships are less stable due to lower trust and confidence in the future, which may result in feelings of dissatisfaction and the perception that conflict exists by more dependent parties. As a result, asymmetric dependence will increase conflicts among cooperative members and reduce their performance perception (Gundlach and Cadotte 1994).

In construction projects, when there is a highly asymmetric dependence and interest divergence between owners and contractors, both of who may concentrate on finding contract loopholes and then profiting from them (Lu et al. 2016). Asymmetric dependence will reduce mutual trust among project partners, leading to short-term behaviors such as mistrust and interest grabbing between owners and contractors (Li, Yin, and Zhang 2022). All of these behaviors will then damage project management performance. So, we propose the hypothesis H1 as follows:

H1:

Asymmetric dependence between project partners has a negative influence on project management performance.

3.2. Relational norm and project management performance

According to the transaction cost theory (TCT), a trust relationship between transaction partners can decrease transaction costs through increasing the flexibility in dealing with uncertain events, reducing conflict during project implementation, and lowing the waste in time and energy (Yang, Ho, and Chang 2012). Trust makes interactions between transaction partners more efficient (Brinkhoff, Özer, and Sargut 2015). In construction projects, trust is especially significant for project success for the reason that trust can save time and cost during implementation phase by substituting unnecessary control mechanisms (Vosselman and van der Meer-Kooistra 2006). In addition, empirical researches have confirmed that trust is critical to ensure project management performance. For example, Jiang et al. (2017) explored the relationship between owner-contractor trust and project success by using the data collected from owners and found that trust positively influences project success. A distrust relationship, on the contrary, may result in negative effects for construction projects and even lead to complete project failure (Jiang, Lu, and Le 2016). Therefore, we believe relational norm trust plays a positive role in improving project management performance. So, this study proposes the hypothesis H2 as follows:

H2:

Trust between project partners has a positive effect on project management performance.

3.3. Relational situation and relational norm

According to resource dependence theory, the dominant partner can weaken or overcome resource vulnerability in their business environment, comprising the possibility of resorting to opportunistic behaviors (Crosno and Dahlstrom 2008), which usually reduces mutual trust between project partners. Given asymmetric dependence, the built and development of trust between project partners will be significantly affected (Jambulingam, Kathuria, and Nevin 2011). In addition, asymmetric dependence usually leads to more conflict between transaction parties (Cheng 2011; Scheer, Miao, and Palmatier 2015), which can then result in higher transaction costs and agency costs. Both costs are harmful for trust level in transaction relationship. As a result, asymmetric dependence negatively affects the mutual trust between project partners (Caniëls et al. 2018; Jambulingam, Kathuria, and Nevin 2011; Scheer, Miao, and Palmatier 2015). So, this study proposes the hypothesis H3 as follows:

H3:

Asymmetric dependence has a negative effect on trust between project partners.

3.4. Relational behaviors and project management performance

Scholars believe that communication is an important pre-factor for improving project management performance (Forcada et al. 2017; Kozhakhmetova et al. 2019; Lindhard and Larsen 2016). Communication has positive effects in reducing information asymmetry, enhancing understanding and cooperation, and reducing differences and conflicts among project participants during cooperation period. In addition, Wood et al. (2002) found that communication is extremely important for building trust between project teams, while open and smooth communication channels are even more important. Pinto et al. (2009) has pointed out that effective communication between project partners plays an extremely important role in project success, and this role is more prominent when the trading environment is highly uncertain. Conversely, Khalfan et al. (2007) explained the role of communication in construction projects from the opposite side of efficient communication. Through interviews, etc., they finally found the important factor causing project conflicts is just inefficient inter-organizational communication. So, this study proposes the hypothesis H4 as follows:

H4:

Communication between project partners has a positive effect on project management performance.

Besides, inter-organizational coordination is an important challenge for project partners (Bubshait, Siddiqui, and Al-Buali 2015). This is because plenty of work in construction projects is inter-organizational, which requires joint efforts of different participants to achieve the set goals. At present, a significant number of projects are pretty complex in the design of electrical and mechanical installations. So, effective coordination between project partners is required (Alaloul, Liew, and Zawawi 2016). In addition, Iyer and Jha (2005) investigated the factors affecting project cost in India, finding that coordination of project participants is the most important one. Since project cost is one of the main dimensions evaluating project management performance, it demonstrates that coordination plays an important role in construction project management performance. Relevant scholars focusing on external coordination within project organization also pointed out that a good coordination mechanism could make team members full of confidence in finishing their goals (Ma et al. 2021), to maximize the personal efficacy into team effectiveness, and ultimately positively affect project management performance. Therefore, this study proposes the hypothesis H5 as follows:

H5:

Coordination between project partners has a positive effect on project management performance.

3.5. Mediating effects

3.5.1. Relational norm-relational behavior and project management performance

Even though project management performance improvement usually bases on the foundation of trust between project partners, trust does not necessarily lead to improvement in project management performance. As Schoorman et al. (2007) had already pointed out, trust only provided confidence guarantee of cooperation for project partners, but there must be some mediating mechanisms for its impact on project success. Based on the integrated framework, we assume that the influence of relational norm (trust) on project management performance will also be indirectly achieved via relational behaviors.

As discussed previously, trust can promote communication (Dalmolen and Sikkel 2015; Yamin and Sim 2016) because trust plays an active role in restraining opportunistic behavior, thus paving the way for frank communication between the project participants. In addition, when owners and contractors cooperate in a tense and inefficient cooperative atmosphere, it is easy to appear failures in controlling schedule, cost, and quality (Wu et al. 2017). So, trust is usually considered as the premise of effective communication between project partners. Besides, researches have shown that mutual trust among project team members will affect project performance by enhancing communication (Coovert, Miller, and Bennett 2017). Similarly, Cheung et al. (2013) found that trust can ultimately affect project performance through communication in cross-organizational cooperation. That is to say, the effect of relational norm trust on project management performance can be indirectly achieved via relational behavior communication.

Besides, trust can promote cooperation and coordination in exchange relationship. According to relational governance theory, relational governance and formal governance are complementary in improving cooperation performance (Huang, Cheng, and Tseng 2014; Poppo and Zenger 2002; Roehrich et al. 2014). Trust is the core element of relational governance, and formal contracts can stipulate the coordination mechanism of trading partners. Therefore, a good trading relationship requires good cooperation between trust and coordination. In addition, scholars paying attention to the influence of inter-organizational trust have found that inter-organizational trust can reduce transaction cost and promote inter-organizational coordination (Hatak and Roessl 2011), which is of great benefit to project management performance and project success (Lindhard and Larsen 2016). In a word, trust can indirectly affect project management performance by strengthening coordination among project partners. So, we expect relational behaviors (communication and coordination) to play mediating roles:

H6:

Relational behaviors mediate the effect of relational norm on project management performance.

H6a:

Communication mediates the effect of trust on project management performance.

H6b:

Coordination mediates the effect of trust on project management performance.

3.5.2. Relational situation-relational behaviors and project management performance

Inter-organizational dependence can influence cooperation performance by shaping partners’ behavior patterns (Cooke-Davies 2002). The joint dependence dimension of inter-organizational dependence has a positive effect on the coordination in dyadic relationships, while the asymmetric dependence dimension of inter-organizational dependence will damage the possibility of relationship continuity and reduce partners’ cooperative behaviors (Griffith et al. 2017). All of these are extremely detrimental to project management performance improvement. Besides, scholars point out that when asymmetric dependence in cooperative relationships is high, communication between cooperative partners will decrease. In such situation, unilateral and formal communication will be more common (Maskin, Qian, and Xu 2000), which has a greater adverse impact on communication efficiency. Also, empirical researches on inter-organizational relationship have found that communication played a mediating effect in the relationship between asymmetric dependence and project success (Brinkhoff, Özer, and Sargut 2015; Li, Yin, and Zhang 2022). So, relational behavior communication may be a mediator between relational situation asymmetric dependence and project management performance.

Additionally, related researches focusing on technological innovation network also show that joint dependence positively affects partners’ cooperative behaviors, while asymmetric dependence is negatively correlated with cooperative behaviors (Brinkhoff, Özer, and Sargut 2015). When joint dependence between partners is high, they will conduct more frequent communication and coordination for the purpose of pursuing common benefits. On the contrary, when asymmetric dependence between partners is high, more interest differences and conflicts occur (Kumar, Scheer, and Steenkamp 1995). Consequently, communication and coordination between partners will decrease. What is more,Lusch and Brown (1996) regarded asymmetric dependence as a context factor and discussed its influence on relational behavior. Therefore, relational situation asymmetric dependence can also have an indirect effect on project management performance through relational behavior coordination. Therefore, we expect relational behaviors (communication and coordination) to play mediating roles:

H7:

Relational behaviors mediate the effect of relational situation on project management performance.

H7a:

Communication mediates the effect of asymmetric dependence on project management performance.

H7b:

Coordination mediates the effect of asymmetric dependence on project management performance.

3.5.3. Relational situation-relational norm and project management performance

Many research has pointed out that asymmetric dependence destroys trust (Emerson 1962; Ganesan 1994). On this basis, asymmetric dependence can affect transaction performance by influencing the trust level of project partners (Ganesan 1994; Zaheer, McEvily, and Perrone 1998). That is to say, the higher degree of asymmetric dependence, the lower trust level between project partners, and then the lower the cooperation performance occurs. From distrust perspective, asymmetric dependence between partners will increase their distrust level, and then negatively affect cooperation efficiency and performance. Scholars have also pointed out that only when dependence and trust level match, can their combination have a positive effect on transaction performance (Wicks, Berman, and Jones 1999). In other words, the combination of joint dependence dimension of inter-organizational dependence and trust is helpful to improve transaction performance, while the combination of asymmetric dependence dimension of inter-organizational dependence and trust will have a negative effect on transaction performance. Moreover, Gulati and Sytch (2007) found that trust partially mediated the relationship between asymmetric dependence and project management performance in automobile manufacturing industry. So, we expect relational norm trust to play mediating roles:

H8:

Trust mediates the effect of asymmetric dependence on project management performance.

3.5.4. Relational situation-relational norm-relational behaviors and project management performance

Based on mixed data collected from automobile and construction industries, Brinkhoff et al. (2015) executed an empirical study on cross-organizational supply chain projects success, and finally found that the negative effects of asymmetric dependence on project success could be realized through the following multiple serials path: asymmetric dependence → trust → employee commitment → project success. Here, commitment is also regarded as a relational behavior. What is more, Li et al. (2022) have proposed multiple serials mediating models in construction project management field. Similarly, this study will explore the multiple serials effects of relational norm trust, relational behaviors (coordination and communication) in the relationship between asymmetric dependence and project management performance. So, we proposed the following multiple serials mediating models:

H9:

Relational norm and relational behavior mediate the effect of relational situation on project management performance.

H9a:

Trust and communication mediate the effect of asymmetric dependence on project management performance.

H9b:

Trust and coordination mediate the effect of asymmetric dependence on project management performance.

Based on hypotheses H1~H9, we draw the theoretical model, as shown in Figure 1.

Figure 1. Research model Based on the integrated framework.

4. Research methodologies

4.1. Questionnaire design

We adopt a questionnaire survey to collect data for this study and adopt the measurement items from published literatures for the variables in the integrated framework. After designing the questionnaire, we invite experienced practitioners and professionals to give their optimization suggestions on the initial questionnaire. Here, the experienced practitioners gave feedback on (1) whether the length of the survey is appropriate and (2) the clarity and understandability of the items of the questionnaire. We made minor modifications to the initial questionnaire based on the received feedback. In addition, a protest on the questionnaire was carried out before the formal questionnaire survey. After distributing 100 questionnaires to owners in Tianjin and Zhengzhou of China, we recover 72 questionnaires, and obtain 56 valid ones. Based on data gathered from these questionnaires, a small sample protest was undertaken, and none of the items of the questionnaire shall be deleted. After previous steps have been met, the final questionnaire is obtained whose items are shown in the Appendix. Each measurement criterion adopted a 5-point Likert scale. The reason for using the 5 Likert Scale is to cover the lacks of 4 Likert scale that has no neutral choice. It is believed that the respondent has understood and will not use neutral choice as the main alternative choice if he does not understand the content and purpose of the question 5 of these Likert scales (Adelson and McCoach 2010). Besides, it overcomes the defect of too many options of 7 Likert scale and is more helpful for respondents to make their chooses.

4.2. Sampling and procedure

Owners are the overall integrators and organizers in construction projects, and they have authorities to exclude anybody from working during project implementation when dispute events occur (Besley and Ghatak 2001). Moreover, plenty of research focusing on project management performance has collected empirical data just from owners (ie, Brinkhoff, Özer, and Sargut 2015; Ke et al. 2015; Li, Yin, and Zhang 2022). Thus, the present research takes owners as the targeted participants. Besides, this research focuses on overcoming the difference between practice and theory within the Chinese context for the following reason: Firstly, as the largest developing country in the world, China has a large number of construction projects under planning and implementation, which provides convenience for obtaining the research data. Secondly, China is influenced by its unique guanxi culture and Confucian cultural, which will then have a significant impact on the relationship between construction project participants. Finally, as a developing country, China’s laws and regulations system is not so sound, so the cost for contractors to safeguard their own rights and interests is still relatively high. Finally, China’s legal system is still not very sound (Zhang, Skitmore, and Peng 2014), which provides soil for testing the related problems of relational governance. So, evaluating relational governance theory on the Chinese context has a natural advantage and is urgently needed. The scope of owners was limited to Jiangsu, Zhejiang, Hubei, Guangdong, Sichuan, and Shandong provinces of China. These provinces are the representatives in the Chinese construction industry because they completed 38.9% of the country’s total output value of construction in 2020.

In addition, the form of electronic survey was adopted in this study for the following reasons: First of all, electronic survey holds the advantages of cost reduction (Fernando and Chukai 2018), and the speed of data collection by electronic survey can be dramatically increased compared with paper survey (Zhuang et al. 2021). Besides, electronic survey affords respondents great flexibility in offering their responses at their convenience in time and place (Adom, Osei, and Adu-Agyem 2020). What is more, the absence of the researcher’s presence aids in reducing all forms of biases in responses, ensuring high objectivity (Minto et al. 2017). The main form of the electronic survey used in this study is electronic questionnaire. To be specific, we inquired about the key construction projects in 2021 in the above six provinces, and obtained the owners and their contact information of some construction projects. Here, key construction projects refer to the ones that have a significant impact on national economic and social development of China. Previous research has collected data from provincial key construction projects of China (eg, Wang, Li, and Zhang 2019) because these projects are standardized and representative. We use E-mail and chatting software such as WeChat to distribute questionnaires to these owner companies in the six provinces and then recover the distributed questionnaires. In particular, to improve the recovery rate of the questionnaires and help respondents understand our intention and the meaning of the questionnaire item, we contact the special respondents who are in charge of the survey once a week, and the survey lasted for 6 months from March 2021 to August 2021. We distribute a total of 580 electronic questionnaires, and then recover 412 ones, with an effective recovery rate of 71.03%. By eliminating invalid questionnaires, we finally obtain 282 valid questionnaires (>200), which meeting the needs of empirical analysis proposed by scholars (Boomsma and Hoogland 2001; Kline 2015).

In addition, this study adopts the form of electronic questionnaire for convenience reason, and the survey lasted for 6 months from March 2021 to August 2021. We distribute a total of 580 electronic questionnaires, and then recover 412 ones, with an effective recovery rate of 71.03%. By eliminating invalid questionnaires, we finally obtain 282 valid questionnaires (>200), which meeting the needs of empirical analysis proposed by scholars (Boomsma and Hoogland 2001; Kline 2015).

4.3. Common method bias

The data collected from respondents were self-reported, so, it is imperative to prevent the issue of common method bias. Out of this reason, we adopt two methods, namely ex-ante control and ex-post control. In ex-ante control, a brief explanation consisting of investigation purpose and confidentiality measures dispels respondents’ doubts on the survey. Besides, different data sources were adopted to avoid the occurrence of homologous bias. About ex-post control, we adopted Harman’s single-factor test and confirmatory factor analysis. We conducted an exploratory factor analysis to extract common factors from all items (KMO = 0.853; Bartlett spherical test is significant at the level of 0.001). By this way, we got five factors, which was in line with the questionnaire structure. The variance accounted by the first factor before rotation is less than 40%. Moreover, confirmatory factor analysis is conducted to load all the items on a common factor. The results indicate that the model’s fit index is worse than that of the initial model. Consequently, it can be believed that there is no common method bias in this study.

4.4. Data analysis

The model in this study is done through a Partial Least Squares approach by means of Smart-PLS 3.0 software. Smart-PLS 3.0 software has many obvious advantages. For instance, neither large samples nor the normally distributed data is required in PLS analysis (Hair et al. 2017). That is to say, PLS has a higher statistical power (Hair et al. 2019). The research model was tested via a two-step process. Firstly, we estimated the measurement model to identify constructs’ reliability and validity. Then, we evaluated the structural model to identify the significance and the strength of the relationships between variables in the model.

4.5. Descriptive analysis

Characteristics of respondents and projects are presented in Table 1. As can be seen in Table 1, college degree or above account for 81.56% of all the respondents, and 76.24% the respondents have worked in their positions for 3 years or more. With respect to the current position, project manager or above account for 43.26% of respondents. All of these ensure that the respondents were familiar with construction projects and can understand the survey items.

Table 1. Characteristics of respondents and projects.

Characteristic of respondents	Frequency	Characteristic of projects	Frequency
Education		Total investments
Doctoral degree	12(4.26%)	＞500 million RMB ($72.1 million)	116(41.13%)
Master’s degree	46(16.31%)	100–500 million RMB ($14.4–72.1 million)	112(39.72%)
Bachelor’s degree	172(60.99%)	50–100 million RMB ($7.2–14.4 million)	28(9.93%)
Others	52(18.44%)	<50 million RMB ($7.2 million)	26(9.22%)
Work experience		Project type
＞10 years	105(37.23%)	Office and residential construction	126(44.68%)
6–10 years	63(22.34%)	Industrial construction	11(3.90%)
3–5 years	47(16.67%)	Public construction	34(12.06%)
< 3years	67(23.76%)	Infrastructural construction	44(15.60%)
Job position		Commercial construction	54(19.15%)
Manager at the headquarters	67(23.76%)	Others	13(4.61%)
Project manager	55(19.50%)	Project delivery system
GMTP Others	144(51.06%)	DBB and DB	141(50%)
	16(5.67%)	EPC and PPP	141(50%)

GMTP = General management/technical personnel.

4.6. Assessment of the measurement model

As stated previously, the first step estimating constructs’ reliability and validity was adopted. First, the key indicators for constructs’ reliability are presented in Table 2. As shown in Table 2, all constructs’ factor loadings are statistically significant and exceed 0.70 threshold level (Hair, Ringle, and Sarstedt 2011). Besides, each of the constructs’ composite reliability and Cronbach’s alpha values exceed 0.70 threshold level, establishing satisfactory internal consistency. These results demonstrate that all reliabilities are within acceptable ranges (Nunnally 1978).

Table 2. Reliability and validity analysis.

Variable	Items	LOA	CR	α	AVE	Variable	Items	LOA	CR	α	AVE
AD	AD1	0.868	0.868	0.868	0.868	PMP	PMP5	0.767
	AD2	0.880				CO	CO1	0.773	0.860	0.797	0.554
	AD3	0.694					CO2	0.816
TR	TR1	0.816	0.884	0.836	0.605		CO3	0.693
	TR2	0.823					CO4	0.786
	TR3	0.761					CO5	0.694
	TR4	0.733				CN	CN1	0.687	0.885	0.845	0.563
	TR5	0.751					CN2	0.790
PMP	PMP1	0.637	0.859	0.795	0.551		CN3	0.734
	PMP2	0.675					CN4	0.800
	PMP3	0.826					CN5	0.745
	PMP4	0.790					CN6	0.742

LOA = factor loadings; CR = composite reliability; α = Cronbach’s alpha; AD = asymmetric dependence; TR = trust; PMP = project management performance; CO = communication CN = coordination.

Besides, we evaluate the constructs’ validity from convergent validity and discriminant validity. With regard to convergent validity, as shown in Table 2, all constructs’ AVE value exceeds 0.50 threshold level, indicating their convergent validities are adequate and satisfactory (Hair et al. 2017).

We evaluate the constructs’ discriminant validity with three approaches as follows:

First, we adopt cross-loadings of measurement items. As shown in Table 3, each item’s loading on its own construct (the bold values) exceeds the cross-loadings on the other constructs.

Table 3. Cross-loadings for measurement items.

	Asymmetric dependence	Coordination	Communication	Project management Performance	Trust
AD1	0.868	−0.145	−0.174	−0.141	−0.315
AD2	0.880	−0.149	−0.128	−0.174	−0.302
AD3	0.694	−0.103	−0.065	−0.064	−0.180
CN1	−0.033	0.687	0.403	0.348	0.341
CN2	−0.189	0.790	0.542	0.477	0.448
CN3	−0.141	0.734	0.424	0.348	0.394
CN4	−0.129	0.800	0.500	0.481	0.487
CN5	−0.131	0.745	0.481	0.397	0.422
CN6	−0.101	0.742	0.524	0.438	0.379
CO1	−0.119	0.488	0.773	0.436	0.422
CO2	−0.220	0.519	0.816	0.422	0.481
CO3	−0.031	0.347	0.639	0.309	0.282
CO4	−0.129	0.501	0.786	0.429	0.394
CO5	−0.063	0.508	0.694	0.398	0.417
PMP1	−0.185	0.283	0.341	0.637	0.330
PMP2	−0.122	0.317	0.324	0.675	0.313
PMP3	−0.093	0.482	0.461	0.826	0.421
PMP4	−0.075	0.472	0.386	0.790	0.380
PMP5	−0.155	0.470	0.465	0.767	0.472
TR1	−0.353	0.440	0.421	0.397	0.823
TR2	−0.266	0.420	0.447	0.389	0.761
TR3	−0.158	0.384	0.379	0.410	0.733
TR4	−0.270	0.379	0.417	0.417	0.751
TR5	−0.249	0.517	0.448	0.426	0.816

Second, we conducted the analysis with Fornell-Larcker criterion. As shown in Table 4, square roots of all constructs’ AVEs (the diagonal elements) exceed the correlation between the constructs, demonstrating adequate and appropriate discriminant validity (Fornell and Larcker 1981).

Table 4. Correlation matrix.

	AD	Coordination	Communication	PMP	Trust
AD	1
Coordination	−0.165	1
Communication	−0.159	0.641	1
PMP	−0.164	0.558	0.540	1
Trust	−0.336	0.553	0.544	0.524	1

AD = asymmetric dependence; PMP = project management performance.

Finally, the HTMT approach proposed by Henseler et al. (2015) was adopted. Results in Table 4 indicate that none of the HTMT ratios exceed 0.90 threshold level. The HTMT_inference criterion, which is evaluated by complete bootstrapping, demonstrates that the uppers of 97.5% confidence bounds are all less than 1. As a result, discriminant validity problems are not a concern in the present research.

Though evacuation on constructs’ reliability and validity, we conclude that the designed scales are both dependable and valid in measuring constructs included in the research model. So, they could be used to the subsequent analysis on structural model.

4.7. Assessment of the structural model

Several criteria were applied to evaluate the structural model. The standardized root mean square residual (SRMR) is the first one. SRMR was recommended as the approximate criterion of model fit proposed by Henseler et al. (2016). According to the instruction of Smart-PLS, a SRMR value less than 0.10 represents a good fit. The SRMR value of the structural model is 0.065, meeting the threshold level. In addition, the chi-square of the model is 680.065. Moreover, all R² values which were proposed by Falk and Miller (1992) exceed the 0.10 threshold level (R²_trust = 0.113; R²_{communication} = 0.297; R²_coordination = 0.307; R²_{project management performance} = 0.407). Besides, all the values of Stone-Geisser’s Q² exceed 0 (Q²_trust = 0.066; Q²_{communication} = 0.158; Q²_coordination = 0.164; Q²_{project management performance} = 0.209), suggesting the satisfactory predictive relevance of the research model (Hair et al. 2017). The path coefficients of the relationships proposed in the integrated framework are presented in Table 5.

Table 5. Estimates for the research model.

Hypothesis	β	t-value	p-value	Support	Hypothesis	β	t-value	p-value	Support
H1:AD→PMP	0.000	0.009	0.993	No	H4:CO→PMP	0.230	3.373	0.001	Yes
H2:TR→PMP	0.247	3.388	0.001	Yes	H5:CN→PMP	0.274	3.696	0.000	Yes
H3:AD→TR	−0.336	5.425	0.000	Yes

β = Path coefficient.

As proposed in H2, trust has a positively and significantly effect on project management performance (β = 0.247, p < 0.01). Similarly, as proposed in H3, asymmetric dependence has a significant negative effect on trust (β=-0.336, p < 0.001). As hypothesized in H4 and H5, both communication and coordination are positively associated with project management performance, providing supports for H4 (β = 0.230, p < 0.01) and H5 (β = 0.274, p < 0.001). However, asymmetric dependence has no significant impact on project management performance (β = 0.000, p > 0.05), demonstrating the results do not support H1.

4.8. Assessment of the mediating effects

The widely accepted method for evaluating mediating effect is the bootstrapping method (Ma et al. 2021). We also adopt SPSS software and the PROCESS SPSS macro developed by Hayes to evaluate mediating by much research (Prebensen and Xie 2017). Therefore, this study adopted both bias-corrected method and the percentile method to test the mediating effects, and the confidence level for the confidence intervals was 95%. Specially, we can conclude partially mediating effects when confidence intervals for both direct and indirect effects do not include zero. We can conclude a fully mediating effect when confidence interval of the indirect effect does not include zero, and the confidence interval of the direct effect does include zero. Table 6 shows the assessment results.

Table 6. Results of mediating models estimated by bootstrapping method.

				95% CI				95% CI
Model		IE	S.E.	BL	BU	DE	S.E.	BL	BU	Mediationexistence
H6a	TR-CO-PMP	0.129**	0.041	0.054	0.213	0.369**	0.064	0.000	0.232	Partial
H6b	TR-CN-PMP	0.155**	0.047	0.066	0.251					Partial
Total		0.284***	0.054	0.181	0.392
H7a	AD-CO-PMP	0.005	0.013	−0.021	0.032					Refused
H7b	AD-CN-PMP	0.006	0.017	−0.026	0.039					Refused
H8	AD-TR-PMP	−0.083**	0.030	−0.147	−0.032					Full
H9a	AD-TR-CO-PMP	−0.044*	0.017	−0.079	−0.015	−0.040	0.038	−0.127	0.045	Full
H9b	AD-TR-CN-PMP	−0.052**	0.017	−0.090	−0.021					Full
Total		−0.168***	0.045	−0.259	−0.083

IE = indirect effect; DE = direct effect; S.E. = standard error; 95% CI = 95% confidence interval; BL = boot lower; BU = boot upper; Total = total indirect effects.

As shown in Table 6, trust has a partial mediating effect on project management performance through communication (coefficient = 0.129, both the confidence intervals of the indirect and direct effects did not contain zero) and coordination (coefficient = 0.155, both the confidence intervals of the indirect and direct effects did not contain zero). Thus, the results support H6a and H6b. In addition, the results show that communication and coordination have no significant influences on the relationship between asymmetric dependence and project management performance, indicating that H7a and H7b are not supported. The results show that trust has a significant influence on the relationship between asymmetric dependence and project management performance, and the confidence intervals of the direct effect contains zero. Thus, trust acts as full mediating effect between asymmetric dependence and project management performance, demonstrating the results support H8.

In addition, we test the multiple serials mediations. The result shows that asymmetric dependence is negatively associated with trust, and then, trust is positively associated with communication, which then relates to project management performance (coefficient = −0.044, the confidence intervals of the direct effect contain zero) (H9a). Similarly, asymmetric dependence is negatively associated with trust, and then trust can positively affect coordination, which then relates to project management performance (coefficient = −0.052, the confidence intervals of the direct effect contain zero) (H9b). So, both multiple serials mediations (H9a and H9b) are supported.

5. Discussion

5.1. Findings

5.1.1. Influence of relational situation on project management performance

In the test about relational situation asymmetric dependence, we find the results do not support hypothesis H1 (asymmetric dependence-project management performance), that is, asymmetric dependence has no significant effect on project management performance. It is so surprising that relational governance can have a negative on project management performance, and it is different from most existing studies (eg, Gundlach and Cadotte 1994; Haq et al. 2018; Jiang, Zhao, and Zuo 2017; Villena and Craighead 2017; Yang, Cheng, and Zhang 2021). To explore the reason for this inconsistency, we conducted interviews with the owners and contractors to obtain feedbacks from owners and contractors:

Regarding asymmetric dependence, the contractors said that their dependence on owners is quite different between public investment projects and private investment projects in the Chinese context. As far as public investment projects are concerned, their funds come from financial allocation of public sectors. Under the condition of government credit endorsement, although there is still unequal status between them and owners, contractors’ enthusiasm and autonomy during project implementation are still high. It is not necessary for contractors to worry too much about the payment for progress payments, etc. More specifically, project partners usually set up a special “fund co-management account” in banks in advance to deposit project-related funds. Once contractors finish corresponding construction quantity and prepare complete payment application procedures, they can obtain payments in a timely manner. Under this situation, contractors still have high enthusiasm to complete public projects despite of high degree of asymmetric dependence between them and owners. So, the role of asymmetric dependence in hindering project management performance has decreased in such situation.

On the other hand, the owners said that their asymmetric dependence with contractors can also be different at different stages of project implementation. Still, taking public investment project as an example, in general, owners have greater autonomy before signing contracts with contractors, so they can independently select satisfactory contractors to cooperate. In this stage, asymmetric dependence shows that contractors rely more on owners. Once owners and contractors sign contracts and contractors began participant in these public investment projects, owners are under the pressure of schedule, and the dependence relationship is more reflected in their dependence on contractors. This phenomenon is more obvious when projects are the key projects and must be completed before a certain time. In this case, once there are disputes between owners and contractors, owners always try every means to retain contractors, hoping that contractors can continue to cooperate and complete projects within the specified time consistent with the signed contracts. It can be inferred that hypothesis H1 is not supported may just because these public investment project samples.

However, we can still find out the negative and indirect effects of asymmetric dependence on project management performance by testing the simple mediation and multiple serials meditations: asymmetric dependence→trust→project management performance (H8, p = 0.006 < 0.01); asymmetric dependence → trust → communication → project management performance (H9a, p = 0.010 < 0.05); asymmetric dependence → trust → coordination → project management performance (H9b, p = 0.004 < 0.01). So, although asymmetric dependence does not have a direct influence on project management performance, it will still affect the trust relationship between owners and contractors and their subsequent communication and coordination, thus indirectly affecting project management performance. It can be inferred that the difference between practice and relational governance theory may be caused by this factor to a large extent.

5.1.2. Influence of relational norm trust on project management performance

The empirical results of this study provide support for hypothesis H2 (trust-project management performance), and demonstrate that trust has a significant positive impact on project management performance, which is consistent with the conclusions of existing research (Jiang, Zhao, and Zuo 2017; Li, Yin, and Zhang 2022). Trust is so important that it is crucial to maintain business relationships based on common goals (Cantù et al. 2021). Besides, trust can effectively alleviate the clear barriers and thinking confrontation that are generally exhibited by both partners in the transaction and that compensate for the absence of a formal contract to restrict opportunism (Mathur, Jugdev, and Fung 2013; Yang, Cheng, and Zhang 2021), thereby improving project management performance. In addition, by confirming that trust can significantly improve project management performance, this study supplements the findings of Lu et al. (2015), that is, relational and contractual governances are both critical to improving the project performance and that contractual governance is more effective in improving performance while relational governance is more helpful in restricting opportunism. Although the present research does not examine the influence of relational governance and contractual governance on opportunistic behavior, it also provides direct support for the effect of relational governance on project performance, and it confirms the important position of trust in relational norm research once again. In China’s Confucian cultural tradition, trust and honesty are regarded as the base of a person, which shows that China is influenced by its unique guanxi culture. Related to this, guanxi culture is bound to have a significant impact on the relationship between construction project participants. Under the background of China’s special Confucian culture, project managers attach importance to the role of relationship ties, but often despise the constraints from formal contracts both partners signed. More specifically, as a developing country, the legal system of construction industry in China is not very sound in general (Zhang, Skitmore, and Peng 2014), deepening the importance of the relational norms trust. Guanxi, which differs from formal relationship in western society, bases on expanded social relations such as kinship, consanguinity, and geography. And it contains special relationships including social laws and ethical norms such as mutual trust, human feelings, morality, and reward. So, guanxi, represented by trust is the delicate fibers woven into every Chinese individual’s social life and thus into many aspects of the Chinese society (Dong, Ma, and Zhou 2017). Consequently, it is not surprising that trust promotes project management performance.

5.1.3. Influence of relational behaviors on project management performance

The empirical results of this study provide support for hypothesis H4 (communication-project management performance), demonstrating that relational behavior communication has a significant positive impact on project management performance, which is consistent with previous research (Adenfelt 2010; Hsu et al. 2012). Several reasons can account for such phenomena: First, when communication is improved at the initial stage of a project, the identified design criteria and solutions can ensure fulfillment of the client’s requirements and priorities (Cheung, Yiu, and Lam 2013). Second, communication plays a pivotal role to bring members together such that information can be shared (Müller and Turner 2005), especially when most of the contracting parties meet each other for the first time (eg, in Chinese public projects). Finally, communication is also useful in problem-solving, decision-making, and enhancing the effectiveness of cooperation (Goldstein et al. 2019). The test results of hypothesis H6a (trust-communication-project management performance) support the partial mediating effect of relational behavior communication in the relationship between trust and project management performance. Therefore, relational behavior communication can not only directly and independently affect project management performance but also be the bridge connecting trust and project management performance.

In addition, the test result of Hypothesis H5 (coordination-project management performance) shows that coordination between owners and contractors is also an important factor affecting project management performance, which is consistent with existing studies (Adenfelt 2010; Alaloul, Liew, and Zawawi 2016; Hsu et al. 2012; Pinto, Slevin, and English 2009). Similarly, the empirical test results of H6b (trust-coordination-project management performance) support the partial mediating effect of relational behavior coordination in the relationship between trust and project management performance. Williamson (1993) has pointed out that trust is an implicit resource embedded in social network. So, there must be mediating mechanisms for its influence on project management performance. In this study, through empirical test of hypothesis H6b (and H6a), the conclusion related to “mediation mechanism” is supported, which contributes to further opening the “black box” of trust’s role in improving project management performance. At this point, this study has inherited the consistent efforts from scholars in the field of relational governance research.

5.2. Theoretical implications

This paper contributes to the relational governance literature by examining how it affects project management performance through a more detail perspective compared to extant studies. Although existing research focusing on project management performance to a greater degree, it ignores the complexity relationship of relational governance elements and their joint effects on project management performance (Benítez-Ávila et al. 2018; Cantù et al. 2021; Thuy and Quang 2005; Yang, Cheng, and Zhang 2021). Akin to Cantù et al. (2021), and Yang et al. (2021), this study classifies relational governance elements in a more detailed way. To be specific, this study divided relational governance elements into three categories, namely relational situation, relational norm, and relational behavior for the first time in construction project management field, clarifying the relationships between relational norm, relational behavior, and the external environment and deepening our understanding of relational governance elements in the construction industry. We then construct an integrated framework including these relational governance elements with project management performance. So, the present research uncovers how relational governance and project management performance are really linked and what the fundamental paths are in the chain by specifying the elements of relational governance in a more detailed way. For instance, this research finds that relational situation (ie, asymmetric dependence) has a negative impact on project management performance through the mediated effect of trust and communication (or coordination). It is so surprising that relational governance can have a negative on project management performance, and it is different from most existing studies (eg, Gundlach and Cadotte 1994; Haq et al. 2018; Jiang, Zhao, and Zuo 2017; Villena and Craighead 2017; Yang, Cheng, and Zhang 2021). So, this research enhances our understanding on the joint impact of relational governance elements on project management performance for both academics and practitioners. So, this provides a reference and basis for follow-up further researches.

In addition, the present study moves us beyond the assumption of the direct relationship between relational norm trust and project management performance by testing the mediating effects of relational behaviors – communication and coordination, which are relevant to project implementation process. For this purpose, we investigate the relationship between trust and project management performance in more detail in inter-organization relationship under China’s construction project context. By doing this, we can see a more detail relationship nor a direct relationship between trust and project management performance, which has been already challenged by researchers (ie, Schoorman, Mayer, and Davis 2007).

5.3. Practical implications

Both owners and contractors need to avoid establishing asymmetric dependence relationships. Asymmetric dependence can indirectly hinder project management performance, potential project participants should focus on the resource complementarity when choosing project partners, which is extremely important for these projects with complex technology and process. For the tenderee (the owner), choosing a bidder (the potential contractor) with the lowest bidding cost can indeed effectively save project cost in the short term, but the bidder with key resources will bring greater value to projects (Unsal and Taylor 2010). Therefore, owners and contractors should ally with ones they share mutual trust. In general, it is inevitable that there is a certain degree of asymmetric dependence between owners and contractors. Through preparatory work in advance, owners and contractors can still screen out potential partners who may put partnership at a disadvantage, and a symmetrical dependence relationship can also be created. For example, owners can evaluate contractors basing on contractors’ size, qualification, performance, and reputation through the prequalification work. After formal bidding, owners can adopt a variety of bid evaluation methods, such as comprehensive scoring method to comprehensively consider bidders’ business bidding, technical bidding, and credit bidding of bidding documents, so as to avoid them paying too much attention to bidders’ tender price and neglecting resources complementary situation with themselves. In a word, a detailed investigation on the issues of trust and asymmetric dependence of the potential relationships can give the choosers opportunities to decrease the risks impeding project management performance. Then, project partners can then focus on the process-related factors to promote project management performance. In addition, when asymmetric dependence naturally occurs for private investment projects, it is critical for dependent partners to clearly communicate the value they add to projects in an appropriate manner. By this way, the dominant partners could also remain motivated throughout the project implantation.

It is necessary to increase the trust level between owners and contractors. To promote project management performance, the core task for owners is to sign contracts with trustworthy contractors. Specifically, owners should attach great importance to the prequalification work in bidding stage, and try to select contractors with higher initial trust, so as to pave the way for project implementation. Specially, owners can collect information about contractors in management ability, social reputation, financial status, safety and environmental ability, technical ability, qualification status, and past performance, etc., and do verification of multi-party and cross verification, to ensure that trustworthy contractors are screened out. In addition, given that EPC and PPP delivery models are becoming increasingly important construction models in China, it is necessary to increase the trust level between owners and contractors in EPC and PPP projects, whose trust level is low (Li, Yin, and Zhang 2022). The following strategies should be available: The first is to adopt an irrevocable system of favorable comments and unfavorable comments. With this system, when all EPC projects are tendering in the tangible market, synchronous virtual transactions on the Internet platform are added, so that buyers (owners) can give sellers (contractors) unfavorable comments (Li, Yin, and Zhang 2022). The second is to adopt the shortlist system. In such system, the government investment departments invite public bidding for procurement and then contractors bid to obtain shortlisted qualifications. The former’s evaluation score is used as the condition for the latter to be shortlisted next time. The more times the latter is shortlisted, the higher trust level will be. The third is to create EPC contract texts corresponding to different trust levels. The higher the trust level of owners to contractors, the greater trust terms can be injected into contracts.

Communication and coordination mechanism should be improved. As for improving communication mechanism, it is important for project participants to adopt BIM technology in construction project management to provide an efficient communication channel. For example, in terms of technical communication, the use of BIM-based digital and visual communication platform can reduce information asymmetry between partners and break information island. With the help of the platform’s visual information communication function and intelligent terminal, it can accelerate the transfer process of related problems among project participants during project implementation. Besides, owners and contractors should organize formal and informal communication activities. For example, owners and contractors can agree in the contract on normal communication methods when signing contracts, such as joint weekly and monthly meeting, to promote the timely transmission and exchange of information between partners. In addition, mobile phones, videos, internet, and other tools can also be utilized to smooth informal communication channels. As for improving coordination mechanism, a special position should be set to be responsible for the coordination work between organizations, so that owners and contractors can maintain cooperation enthusiasm in the whole process of project implementation. For this purpose, project coordinator role should be continuously improved, so that it can play an active role in coordinating the relationship between owners and contractors, convening joint regular meetings and supervising project rectifications. Additionally, to achieve higher project management performance, the traditional vertical command and control relationship (conclusion from the electronic questionnaires) of “owner-design-contractor-supplier” should be changed. With the help of information network platform, it can be transformed into a horizontal cooperation relationship among owners, designers, constructors, and suppliers, etc., so as to realize the synergy effect for project management and promote the solution of related problems during project implementation and finally promote project management performance.

In addition, given that relational governance elements including trust, communication, and coordination are all found to be positive to project management performance, it is necessary for both owners and contractors to adopt relational transaction practices in their projects, especially when projects are complex (eg, non-conventional systems such as EPC and PPP). To cope with project complicatedness, it is important for both owners and contractors to design effective mechanisms to facilitate the preservation of relationships, cooperation between contracting parties and flexibility rather than over-rely on pre-set contracts (Ning and Ling 2015). It is also suggested that contracting parties can manage complex projects by selecting an appropriate procurement route. For conventional delivery systems, for example, Design-Build procurement system may be better to integrate the design and construction teams. So, it is important for contractors to adopt joint venture with experienced firms because these projects need to adopt new or complex technologies and construction methods.

6. Conclusion

Relational governance and contractual governance are both effective mechanisms to improve project management performance. It should be noted that relational governance must be complemented with good contractual governance (and vice versa) as they can play a different role in ensuring the project’s success and improve project management performance. The present research aims to show the link between different relational governance elements and project management performance. So, this research divides relational governance elements into three categories: relational situation, relational norm, and relational behavior. Then, this study constructs an integrated framework of relational situation-relational norm-relational behavior-project management performance based on previous research and then investigates the impact of these relational governance elements on project management performance following an empirical analysis process. Overall, 9 of 12 hypotheses in the present research are supported. The test results of the research model statistically support the following viewpoints: project management performance is indirectly influenced by relational situation asymmetric dependence (ie, asymmetric dependence) between owners and contractors. The relational norms trust and relational behaviors (communication and coordination) also have significant influence on project management performance.

This study’s value lie in follows: Firstly, this study divides relational governance elements into three categories for the first time, and especially the introduction of relational situation means the research innovation to relational governance research, deepening our understanding of relational governance elements in the construction industry. Then, this study constructs an integrated framework based on the classification: relational situation-relational norm-relational behavior-project management performance, which expends the existing framework of “relational governance – project management performance” and represents an important innovation in the development of relational governance theory. Secondly, this study puts forward some suggestions to overcome the above difference and improve project management performance within the Chinese context.

This study has the following limitations: Firstly, the data adopted in this study was only gathered from China, causing its representativeness and generalizability may be limited for other cultural contexts. So, the research conclusion and practical implications might be limited, and further examining on this topic should be conducted. Additionally, this study constructs the integrated framework and build the links among relational situation, relational norm, relational behaviors, and project management performance. Despite considering many factors affecting project management performance, it is still limited to completely explain the link between relational governance and project management performance. As is known to all, project management system is so complex, involving many variables which affect each other. Therefore, follow-up researches can consider adding more suitable variables, so that the research results can explain the above link more deeply. Finally, workers’ work experience and job position have important influence on the cognitive of project governance. It is obvious that the worker with rich experience such as a manager at the headquarters can easily cognitive of the importance of project governance. So, the follow-up research can attach importance to their role.

Acknowledgements

The authors acknowledge with gratitude the MOE (Ministry of Education in China) Project of Humanities and Social Sciences (No. 19YJC630078), The National Social Science Fund of China (21BGL029), and the Key Science and Technology Projects of Henan Province (222102320174). This study would not have been possible without their financial support.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The work was supported by the Key Science and Technology Projects of Henan Province (No.222102320174); the Chunhui Program of Ministry of Education of China (No. 202200695); the National Social Science Foundation of China (No.21BGL029); and the National Natural Science Foundation of China (No.72271091; No.71974056).

Notes on contributors

Xiaolin Li

Xiaolin Li （李孝林）is a Master’s supervisor at the School of Water Conservancy, North China University of Water Resources and Electric Power. His research focuses on the project management performance, and relational governance, and water conservancy reform and development.

Appendix.

Table

Construct	Code	Items	Source
TR	TR1	We are certain that other party have the ability to perform their tasks.	Romahn and Hartman,1999; Lu et al. (2016)
	TR2	We believe that the other party could meet the requirements of the project in technology and management
	TR3	We believe the project engineers and other technical people are competent at what they are doing.
	TR4	We believe that the other party are able to fulfill contractual agreements.
	TR5	We believe that the other party can keep their word throughout the life of the project.
AD	OWN1	It is difficult for us to find another contractor to replace the contractor.	Kumar et al., (1995); Kim and Choi (2018)
	OWN2	Failure to cooperate with the contractor will bring great losses to us.
	OWN3	Th contractor’s resources and technology are very important to us.
	CON1	It is difficult for the contractor to find a more generous profit by us.
	CON2	The contractor will suffer great losses once they failure to cooperate with us.
	CON3	The contractor has difficulty in finding other owners to cooperate.
CO	CO1	We share sensitive information.	Paulraj et al. (2008)
	CO2	Our project partner is provided with information that might help them.
	CO3	Exchange of information takes place frequently, informally and/or in a timely manner.
	CO4	We keep each other informed about events or changes that may affect the other party.
	CO5	We have frequent face-to-face planning/communication.
CN	CN1	Our project team worked interactively with the other party on important aspects of the project.	Liberatore and Luo (2009)
	CN2	Giving instructions or directions is an important way we coordinate with our project partner.
	CN3	There were established rules and procedures for coordination between the partner organizations.
	CN4	Team members were engaged in coordinating with our project partner.
	CN5	I worked interactively with my counterpart from the partner organization on important aspects of the project.
	CN6	Our project team worked interactively with the other party on important aspects of the project.
PMP	PMP1	The project deliverable meets the client’s objectives.	Wu et al. (2017)
	PMP2	This project has qualified acceptance and successful delivery.
	PMP3	The project can solve most problems encountered during the project execution.
	PMP4	The project process is satisfactory.
	PMP5	This project creates positive impacts for end users.

OWN1~OWN3 are used to evaluate the dependence of owners on contractors; CON1~CON3 are used to evaluate the dependence of contractors on owners. AD is the absolute value of the difference between owner’s dependence on contractor (OWN) and contractor’s dependence on owner (CON), similarly with Kumar et al. (1995) and Kim and Choi (2018). Because this research collected empirical data just from owners, we invited owners to evaluate the dependence of their contractors on them, and then take the owners’ estimate on CON1~CON3 as the true estimate of CON1~CON3.