Research on influencing factors of consummate performance behaviors: a configurational approach

ABSTRACT

Contractors’ behavior for consummate performance (BCP) has an important impact on the value added of the project. Therefore, this paper first objective is based on theory of planned behavior (TPB) to understand contractors’ BCP. In addition, this study investigates which combination of influencing factors influence BCP by qualitative comparative analysis (QCA). A questionnaire survey including 195 respondents was used to collect the data. The results indicate that the BCP is mainly influenced by the perception of profit/cooperation/value attitudes, subjective norm, perceived behavioral control and the behavioral intention. The analyses also reveal two equifinal combinations of influencing factors that result in BCP. This study identified the factors leading to Chinese contractor BCP by extending the TPB model to the context of construction. The extended TPB model may serve as a theoretical framework and basis for Chinese contractors to perform BCP. Furthermore, this study contributes to the contractors’ behavioral literature by empirically presenting a configurational solution.

KEYWORDS:

• Behavior for consummate performance (BCP)
• configuration analysis
• theory of planned behavior (TPB)

1. Introduction

Diverse ways of contract fulfillment would result in different performance (Hart and Moore 2008). The purpose of rational organizational behavior and contract compliance in engineering projects is merely to improve efficiency (Anvuur and Kumaraswamy 2007). The key to solving the strategic issues of innovation, conflict, etc. in a project is the contractors’ behavior for consummate performance (BCP). BCP is defined by the contract reference point theory (Hart and Moore 2008). In this theory, a contract provides a reference point for the parties’ trading relationship through the feelings of entitlement (Hart and Moore 2008). A party’s ex-post performance depends on whether the party gets what he is entitled to relative to the outcomes specified by the contract (Hart and Moore 2008). Hart and Moore (2008) distinguished between perfunctory performance and consummate performance. Perfunctory performance can be judicially enforced and performance within the letter of the contract, whereas consummate performance cannot judicially be enforced, which performs within the spirit of the contract. The contractor would provide consummate performance if he feels that he is getting what he is entitled to but will withhold some part of consummate performance if he is shortchanged. An important assumption is that a party’s sense of entitlement is determined by the contract he has written, which is referred to as a reference point.

Due to the complexity of construction projects, and the incompleteness of contracts, information asymmetry is easily generated. Based on transaction cost theory (TCE), contractors as agents are easy to use their own information and experience to choose opportunist behavior to obtain more extra profits. Based on the above logic, the theoretical and practical circles will focus on suppressing and reducing the opportunistic behavior of contractors. However, the management logic and cognition focusing on enhancing the control function of the contract and avoiding opportunism can only restrict the contractor to fulfill the contract obligations in accordance with the contract to realize the basic requirements of the project, but cannot meet the requirements of the employer to use the contractor’s technology and management experience to realize the value-added of the project. Obviously, the key to value-added projects is the contractors’ BCP which is beyond the contract agreement, although it is a rational choice to complete the construction tasks in accordance with the contract and maximize contractors’ respective interests (Yang et al. 2018). Therefore, it is meaningful to understand the formation of the contractor BCP. However, there is a lack of research framework for inducing factors of BCP. Hart and Moore (2008) studied CPB mainly from the psychological. However, it failed to include the factors such as ability, situation and intention into the analysis framework, thus lacking systematic research on BCP in construction projects. The theory of planned behavior (TPB), as a mature behavior study model, has provided us with inspiration for a comprehensive study of BCP. TPB includes not only psychological factors (attitude) but also external environment factors (subjective norms) and ability factors (perceptual behavior control).

Additional there is a lack of research on the combination of factors inducing the contractor BCP. Traditional empirical research is based on the regression analysis paradigm. The obtained results mainly answer whether a certain factor has a significant influence on the contractors’ behavior, but cannot answer which factors are needed for the occurrence of the contractors’ behavior. A limitation exists since the occurrence of specific behavior does not necessarily require all the influencing factors. The key is to find the combination of influencing factors that can produce effects. If the combined effects of different influencing factors are not clarified, the solution triggered by the contractors’ behavior cannot be obtained. Different from the traditional regression analysis paradigm, qualitative comparative analysis (QCA) focuses on the combination of variables. QCA can help to understand three major problems in causal inference: 1) Equivalence, that is, different combinations of various conditions can lead to the same result; 2) Non-linear problem, meaning that the effect of one condition may change due to the existence or non-existence of other conditions; and 3) Asymmetry, that is, the existence of a certain condition associated with a given result does not necessarily mean that the absence of the condition is not associated with the result. The focus of this study – the contractors’ intention to adopt BCP is often the result of a combination of several influencing factors. Therefore, using QCA to analyze the contractors’ intention to adopt BCP can further reveal the formation mechanism of BCP.

The purpose of this study is to establish an analytical framework of the factors influencing the contractors’ BCP based on the TPB. The partial least square method is relied upon to build a theoretical analysis framework of the contractors’ IBCP from the perspectives of psychology and the external environment. After that, the QCA method was employed to explore how the combination of different influencing factors impacts the contractor BCP. This paper is organized as follows. Section 2 introduces the theoretical foundation and the theoretical model. Section 3 presents the research method of a questionnaire survey, its results, and discussion. Section 4 reports fuzzy set Qualitative Comparative Analysis (fsQCA) in the configurational analysis, its results, and discussion. It is ended up with implications for theory and practice as well as concluding remarks.

2. Literature review and conceptual framework

2.1. The contractors’ BCP

The traditional research on contractor’s behavior has two perspectives. One perspective is based on transaction cost theory (TCE), focusing on opportunistic behavior and cooperative behavior of contractors. Researchers have investigated formal governance and informal governance and revealed their inhibitory effect on opportunistic behavior (Lu et al. 2015a; Ahimbisibwe 2014). Existing studies point out that integrity of the contract (Lu, Zhang, and Zhang 2016), clarity of the contract terms (Ning 2018), trust (Lu et al. 2015b), uncertainty (You et al. 2018; Lu et al. 2015b; Wang et al. 2013), complexity (Lu et al. 2015b), asset specificity (Shi et al. 2018; Wang et al. 2013) and risk perception (Zhang and Qian 2017) have an impact on the opportunistic behavior of contractors. In addition, some studies also prove that the control, adaptation and coordination functions of the contract (Quanji, Zhang, and Wang 2016; Wang et al. 2017), Contract Flexibility (Song et al. 2018), risk sharing (Zhang et al. 2016; Xu et al. 2018), trust (Kadefors 2004; Xu et al. 2018) and fairness (Lu, Li, and Wang 2017) have positive effects on the contractor’s cooperative behavior. However, under the TCE theory, research on the contractors’ behavior focuses on the influence of governance, which is at the project level (Li et al. 2019). However, the role of factors at other levels (such as the individual level and the organizational level) is ignored. In addition, whether it is opportunistic behavior or cooperative behavior, scholars are more concerned about the contractor’s in-contract behavior but pay less attention to the extra contract behavior that cannot be agreed in the contract and has a positive effect on the project. The second perspective is to construct a theoretical framework of engineering organizational behavior (Li et al. 2019) including individual, team, project and organizational field level from the perspective of organizational behavior theory, focusing on the organizational citizenship behavior of the project. It is pointed out that the behavior of construction projects is a voluntary individual behavior, including innovative behavior, voice behavior, responsible behavior, interest defense behavior and relationship maintenance behavior (He et al. 2015). Yang et al. (2018) identified five types of behavior phenomena in major engineering organizations through quantitative cross-case studies: harmonious relationship maintenance, contingency coordination, obedience, initiative, and responsibility. Wang et al. (2018) and Lim and Loosemore (2017) have proved that institutional pressure and interactional justice are the key factors for organizational citizenship behavior in construction projects. The project organizational citizenship behavior has a positive effect on the project from the perspective of organizational behavior theory, but belonging to the category of individual behavior and represents the behavior at the individual level, not organizational level. In a word, the existing research on contractor’s behavior focuses on opportunistic behavior, cooperative behavior, and project organization citizenship behavior, but pays less attention to the contractors’ behavior having positive effects on the project beyond the contract, which represents the behavior at the organizational level.

Different from the TCE and organizational behavior, according to the contract reference point theory established by Hart and Moore (2008), this study regards the contract as a reference point that accords with the expectations of both parties to the transaction. The anchoring effect produced by the reference point makes the parties to the transaction only willing to complete the perfunctory performance according to the contract literal, and not necessarily to do their best to achieve consummate performance according to the spirit of the contract. On the other hand, if the reference point produces a positive reference point effect, the transaction subject will choose to achieve consummate performance, and the behavior associated with consummate performance is the consummate performance behavior referred to in this study. Therefore, this study defines the contractor’s consummate performance behavior as the inter-organizational performance behavior (Wang et al. 2017) that the contractor is willing to make extra efforts to implement the project in the spirit of the contract and has nothing to do with specific tasks after the contract is concluded, which is characterized by voluntary, initiative and altruism.

2.2. The driving factors of BCP

Theory of Planned Behavior (TPB) is widely used to study interdisciplinary behavioral intentions, including technical acceptance (Chen and Huang 2012), blood donation (Masser et al. 2009), online shopping (Hansen 2008), and health-related issues (Ajzen and Manstead 2007). Although TPB was originally used to study individual behavior, there are numerous empirical studies at the organizational level (Dodor and Rana 2009; Koropp et al. 2014; Grandon and Mykytyn 2004; Cheng 2016; Zheng et al. 2017; Liu, Lin, and Feng 2018). For example, Aibinu and Al-Lawati (2010) developed a theoretical model based on TPB to determine the key factors that determine the intention of construction organizations to participate in electronic bidding. Lee, Yiu, and Cheung (2016) suggests using TPB to summarize the influencing factors of alternative dispute resolution among project organizations. Cheng (2016) used TPB to study the intention to form project partnership in Hong Kong’s construction industry, expanding the application of TPB at the organizational level. Based on TPB, Zheng et al. (2017) explains the influencing paths of inter-organizational relational behavior, showing that Benefit perception attitude, subjective norm and perceived behavior control have significant driving effects on relationship behavior intention. Liu, Lin, and Feng (2018) reveals the mechanism of Chinese engineering contractors’ intention to purchase engineering insurance through an extended TPB model. The above study on TPB’s behavior intention at the organizational level provides a reference for this study on the IBCP. Therefore, this study is based on the existing research (Cheng 2016; Liu, Lin, and Feng 2018), based on the theory of planned behavior, constructs a research framework of BCP.

The TPB includes five elements: behavior attitude, subjective norm (SN), perceived behavior control (PBC), behavior intention and behavior (Ajzen 1991). Behavior attitudes reflect the degree of the actor’s love or approval for the implementation of a particular behavior and his expected comprehensive evaluation of the behavior as a whole (Van Kenhove, De Wulf, and Steenhaut 2003). SN refers to the social pressure that the actor feels when deciding whether to execute a specific behavior. It reflects the influence of important others or groups on the behavior decision of the behavior subject (Rivis and Sheeran 2003). PBC refers to the degree to which the actor perceives it easy or difficult to execute a specific behavior. It reflects the actor’s perception of factors that promote or hinder the execution of the behavior. According to TPB, the behavior intention is influenced by behavior attitudes, SN and PBC.

We used semi-structured interviews to interview 26 experienced contractors. On this basis, the grounded theory is used to analyze the interview data: 1) Open coding. Through the analysis of the interview content, we find the content that is relevant to the research topic, and get 69 tags in total; 2) Axis coding. By examining the results of open coding and combining with the background of construction projects, three main categories of factors affecting the contractors’ IBCP in the construction project context are formed, as shown in Table 1.

Table 1. The driving factors of IBCP

The main categories	Definition	The sub-categories	Conceptual category
Behavior attitude	Reflect the contractor’s psychological evaluation of the degree of acceptance of BCP.	Perception of value attitude	Social responsibility; Enterprise idea
		Perception of cooperation attitude	Trust of owners; Long-term cooperation opportunities
		Perception of profit attitude	Expected profit of the project; Payment of contract price
Subjective norm	The contractor’s adoption of BCP is perceived pressure from inside and outside the enterprise organization, which reflects the influence of important third parties on its behavior decision-making.	Directive specification	Enterprise requirements; Leadership advocacy
		Exemplary specifications	Peer competition; Owner evaluation; Project supervision
Perceived behavior control	The Contractor’s Perception of Difficulty in Performing BCP	Self-efficacy	Engineering and technical capability; Project management capability; Past experience
		Availability of resources	Sufficient manpower and material resources

2.3. Behavioral attitude and intention of behavior for consummate performance (IBCP)

Behavioral attitude is a description of behavioral results and value expectations. Specifically, the behavioral attitude of the contractors’ BCP is an evaluation of the degree of recognition of BCP. The traditional principal-agent theory holds that the contractor is self-interested, that is, the contractors’ behavior and attitude are directly related to the material benefits obtained. However, the social preference characteristics of the contractor have been largely confirmed in reality. Social preferences such as altruism, reciprocity, and fairness stimulate the contractor to care about the interests of others. They wish to achieve perfect project performance and value added through BCP in order to obtain long-term cooperation with the owner (Lu and Hao 2013). It can be seen that the internal dynamic mechanism of the evolution of the contractors’ attitude towards BCP is transformed from interest-driven to cooperation-driven and spirals up to value-driven. The contractor pursues not only project profits but also the trust of owners, social respect and realization of enterprise value. Therefore, as shown in Table 1, this study decomposes the contractors’ attitude towards BCP into three dimensions: perception of profit attitude (PPA), perception of cooperation attitude (PCA), and perception of value attitude (PVA).

2.3.1. PPA and IBCP

The contractors’ PPA refers to the idea that the contractor will perform BCP due to external profit incentives. Timely payment and contract compensation will also affect the contractors’ enthusiasm to perform BCP (Akintoye and Main 2007) and contribute to improving the project efficiency and stakeholder satisfaction (Chan, Chan, and Ho 2003). Generally speaking, when the contractors’ perceived profit is close to or higher than the positive psychological threshold, it is possible for the contractor to make extra efforts as a return to the owner because of the effect of reciprocity (Fehr, Hart, and Zehnder 2015). The contractors’ perception of profit is mainly affected by the final total contract income. If the contractor believes that BCP has a positive effect on timely payment and sharing of added value, the contractor will perform BCP. Besides, the contractors’ initiative to make up for the loopholes in the contract not only brings benefits to the project but also can indirectly improve the income level of the contractor itself by reducing the expenses for rework and maintenance in the later period. All these economic incentives make the contractor aware of obvious or potential benefits and drive the generation of the intention to perform BCP. Thus, we develop the following hypothesis:

H1a: The contractors’ PPA is positively related to the contractors’ IBCP.

2.3.2. PCA and IBCP

During the process of project contract execution, all parties involved not only pursue their own interests but also have social preferences such as reciprocity. The contractor should not only focus on the benefits achieved from the current project but also consider the benefits in the long run brought by long-term cooperation. The cooperative relationship between the two sides is not only conducive to the smooth progress of the current project but also lays a solid foundation for their long-term and amicable cooperation. Therefore, if the contractor believes that performing the contract well can promote cooperation between the two parties, it is likely to agree with this behavioral selection. The contractors’ perception of cooperation means that it is willing to perform well in order to pursue long-term cooperation with the owner and maintain friendly relations. A cooperative relationship can also effectively reduce the contractors’ anger, resentment and other bad emotions during the contract execution, and stimulate the contractors’ enthusiasm to perform well. The positive and mutually beneficial cooperation paradigm is a win-win choice for both the owner and the contractor. The transition from a hostile relationship to a cooperative relationship opens up a way for the contractor to perform well. Thus, we develop the following hypothesis:

H1b: The contractors’ PCA is positively related to the contractors’ IBCP.

2.3.3. PVA and IBCP

The contractors’ PVA is influenced by the enterprise’s responsibility for the project and the mission to pursue project performance. After completing the project tasks, the contractor is required to finish “extra value”. In addition, the contractor adheres to the concept of being responsible for the project and providing high-quality services to the owner in order to win the trust of the owner and enhance the reputation of the enterprise itself. This is similar to the fact in the field of enterprise management that some employees are willing to innovate with the aim of being appreciated by leaders and obtaining expected rewards other than salaries. It can be seen that in the Chinese social context, the contractors’ pursuit of corporate reputation, social status, and the sense of achievement, etc. greatly affect the preference for the perfect consummate behavior that can create added value to the project. Thus, we develop the following hypothesis:

H1c: The contractors’ PVA is positively related to the contractors’ IBCP.

2.4. The subjective norm (SN) and IBCP

The SN for the contractor to perform BCP refers to the pressure from inside and outside the enterprise. After the construction contract is signed, the contractor lies in a specific project network. Due to that, the contractor has to abide by various relationship norms and industrial practices formed by the project participants through corresponding social exchanges. The first manifestation of the contractors’ SN is the mandatory norm which is mainly originated from the requirements of the enterprise and the advocacy from the leaders. It is a kind of latent norms and culture within the enterprise, affecting the formation of the contractors’ willingness to perform well (Conner et al. 2007). The second manifestation is the demonstration norm. It is the pressure and demonstration effect brought to the contractor by other stakeholders outside the contractors’ enterprise, including peer competition, the owner’s evaluation and project supervision (Lee, Yiu, and Cheung 2016). Previous studies have shown that these social exchange channels embedded in the construction project organization tend to determine the behavior of the contracting parties and the development of the industry. Therefore, the contractors’ IBCP will also be influenced by the social norms existing in the social network environment. In the TPB, SN refers to the actor’s social pressure when performing a specific behavior. Such pressure is caused by the norm formed by important people or organizations around the actor, which reflects the influence of important people or groups on behavioral decision-making. Research in the field of organizational performance has confirmed that there is a significant correlation between SN and behavioral intentions. During the contract execution, the contractors’ stakeholders include not only government departments and the owner, but also design units, subcontractors, and suppliers. If the contractor realizes that the enterprise strategy, the project stakeholders, and the relevant government departments all agree on the behavior of performing the contract perfectly, it will inevitably generate the contractors’ motivation to keep in line with these norms and stimulate the willingness to perform the contract perfectly.

First, the contractor will be affected by competitors, subcontractors, and suppliers, and their active activities will also have an exemplary effect on the contractor. It implies that the contractor will be under competitive pressure although it does not have IBCP. Second, existing studies have pointed out that the contractor can also receive the guidance from the government’s industrial policies and the project owner’s recognition, which is likely to promote the contractor to take positive actions. Finally, the contractors’ IBCP is affected by the group company’s strategy, development pressure, and corporate culture. The parent company often places a high emphasis on the enterprise’s enterprising spirit of innovation, the quality consciousness of creating high-quality products and providing high-quality projects. This requires the contractor to take the initiative to do some altruistic work in addition to completing the basic contractual obligations during the contract implementation. Thus, we develop the following hypothesis:

H2: The SN for the contractor to perform BCP is positively related to the contractors’ IBCP.

2.5. The perceived behavior control (PBC) and IBCP

According to TPB, PBC refers to the degree of difficulty in perceiving the execution of specific behavior, which reflects an individual’s perception of promoting/hindering factors (Lu, Zhang, and Zhang 2016). The stronger the controllability of perceptual behavior and the more controllable factors of executing the behavior required to produce the outcomes, the greater the possibility of executive behavior (Kraft et al. 2005). PBC in this paper refers to the contractors’ perception of whether or not it has sufficient ability to perform BCP. It mainly derives from two aspects: self-efficacy and availability of resources such as the contractors’ confidence in performing the contract perfectly, project management ability, and project implementation experience, the contractors’ evaluation of the resources and conditions needed to complete the tasks. When the contractor highly evaluates its own ability and can allocate the manpower and material resources, the contractor is likely to take the initiative to do more work for project success.

Under the context of complex construction projects, contractors often cannot foresee all risks such as changes in construction conditions, price fluctuations, force majeure, etc. When deciding whether or not to take the initiative to control risks, cooperate with each other, and make up for the loopholes in the contract, the contractor needs to evaluate all aspects of constraints, its own managerial experience, and the difficulty and cost of taking such actions. If the contractor has enough experience and ability to resolve the project risks and actively optimize the construction, it will be relatively easy to perform as well as possible, and thus the contractor will have more enthusiasm to perform BCP. Thus, we develop the following hypothesis:

H3: The PBC is positively related to the contractors’ IBCP.

2.6. The perceived behavior control (PBC) and BCP

Under complex situations of construction projects, contractors often cannot foresee all risks in the construction process, such as changes in construction conditions, price fluctuations, force majeure, etc. Whether the contractor is willing to take the initiative to control risks, cooperate and help each other, and make up for the loopholes in the contract, it needs to evaluate all aspects of constraints, its own management experience, and the difficulty and cost of taking such actions. If the contractor has enough experience, ability and conditions to resolve the project implementation risks and actively optimize the construction, it is relatively easy to perform as well as possible, and he will have more enthusiasm to perform as well. In addition, the study also found that perceptual behavior control can directly act on actual behavior beyond behavioral intention, and affect actual behavior together with behavioral intention. That is to say, in addition to the behavior will have a direct effect on behavior, perceptual behavior control will also have a direct impact on behavior. The theory of planned behavior holds that the ability of the actor to take a certain behavior, the necessary opportunity to implement the behavior and the degree of possession of resources are important factors that affect the behavior. In the field of construction projects, contractors need to have the ability of design, project operation, storage and transportation of resources, etc. In addition, more and more literatures have found that the contractor’s network ability will greatly enhance the control of enterprises on external resources.

H4: The PBC is positively related to the contractor BCP.

2.7. The IPBC and BCP

In sociology, psychology and many other fields, the positive effect of IBCP on behavior has been conformed. Conner  et al.  (2007) verified the influence of behavioral intention on behavior through meta-analysis of 185 published literatures, and were confirmed in many different fields. The contractor IBCP indicates the extent to which the enterprise is trying to achieve cooperative innovation and the extent to which it is willing to put in efforts to achieve behavior, which is one of the best predictors of behavior. Only when the contractor recognizes from the heart the current benefits or long-term value brought about by the perfect performance of the contract will he be more willing to perform the contract and stimulate the implementation of the perfect performance. The more the contractor feels the normative pressure to perform the contract perfectly, the more he has sufficient control over the performance of the contract perfectly and is more willing to perform the act, so as to produce the performance of the contract perfectly. From the above research, we can find that behavior will directly determine behavior. Thus, we develop the following hypothesis:

H5: The IBCP is positively related to the contractor BCP.

To sum up, with the aid of the theoretical analysis framework of planned behavior, the theoretical model of this paper is proposed. Figure 1 illustrates the hypothesized model.

Figure 1. Hypothesized model

3. Method

3.1. Questionnaire design

Although the TPB theoretical framework provides principles for the measurement of each variable, it also points out that different behavioral research fields utilize diverse connotations and compositions in terms of the three elements that determine behavioral intention. Therefore, the authors first developed measurement items based on previous relevant questionnaires with modifications suggested by interviews. Then we discussed with several experts and scholars to further refine the measurement items. Specifically, the measurement items of PPA, PCA, PVA, and SN are developed through comprehensive literature analysis and interviews with project managers. The items of the intention to perform BCP was revised based on the questionnaires of Ajzen (1991) and Zheng et al. (2017). The questionnaire used a 5-point Likert scale, ranging from totally disagree to totally agree. The higher the score, the stronger the degree of agreement. The measurement items are shown in Table 2.

Table 2. Items of measurement and reliability and validity analysis

Variable	Code	Measurement items	Key source(s)	CR	Cronbach’s α	AVE
PPA	PPA1	The relevant agreements on price adjustment and payment in the contract are fair	Yan, Wang, and Deng (2018)	0.907	0.848	0.766
	PPA2	The owner can pay according to the contract (e.g. actively pay amounts certified, contract compensation, etc.)
	PPA3	The contractor wants to reduce the cost of rework and maintenance
PCA	PCA1	The contractor hopes to establish a good relationship with the owner	Glagola and Sheedy (2002) Ling et al. (2014)	0.904	0.840	0.758
	PCA2	The contractor hopes that disputes can be reduced
	PCA3	The contractor Considers the later cooperation with the owner
PVA	PVA1	The contractor believes that the contractor has the responsibility and mission to perform BCP	Dubois and Gadde (2002)	0.920	0.870	0.794
	PVA2	The contractor believes that the contractor should pay attention to the sense of achievement and honor
	PVA3	The contractor believes that the contractor should pay attention to the improvement of project management level
SN	SN1	The contractor believes that the contractor has been influenced by enterprise requirements	Zheng (2017)	0.883	0.803	0.715
	SN2	Other enterprises adopt BCP
	SN3	The contractor believes that the contractor has been affected by government and policy
PBC	PBC1	The contractor has sufficient technical capability	Ajzen (1991) Zheng (2017)	0.938	0.913	0.792
	PBC2	The contractor has sufficient management capability
	PBC3	The contractor has sufficient resources
	PBC4	The contractor has enough experience
IBCP	IBCP1	The contractor wants to perform BCP now	Ajzen (1991) Zheng (2017)	0.961	0.939	0.891
	IBCP2	The contractor wants to perform BCP in the future
	IBCP3	The contractor wants to always perform BCP
BCP	BCP1	The contractor voluntary makes extra effort for project	Yan, Wang, and Deng (2018) Liu et al. (2019)	0.928	0.903	0.720
	BCP2	The contractor takes the initiative to put forward reasonable suggestions for the employer to help improve the performance of the whole project
	BCP3	The contractor actively helps relevant participants who need to adapt to the environment on the construction site
	BCP4	The contractor takes the initiative to inform the employer of errors or omissions in the drawings or the contract
	BCP5	The contractor actively takes reasonable measures to control and internalize project risks, and prevent risks from flowing back to the employer

3.2. Sampling and procedure

The target respondents include project directors/managers and contract managers from contractors. Due to the vast number of construction contractors in China, it is impractical to conduct nationwide sampling in the whole population of construction contractors. In this regard, the sampling was limited to three typical cities in China, including Beijing (the capital city of China), Shanghai (the largest city in China), and Tianjin (one of the four direct-controlled municipalities of China). These three cities were considered representative in the Chinese construction industry because 20 percent of the top 500 Chinese construction companies in 2018 were located in Beijing, Shanghai, and Tianjin (Chinese Construction Enterprises Management Association 2018). Snowball sampling (also referred to as chain sampling, chain-referral sampling, or respondent-driven sampling) was adopted in this study, considering the notoriously low response rates and lack of reliable data of the random sampling approach (Zhai, Liu, and Fellows 2013). The initial participants included 37 project managers and contract managers from different construction projects in Beijing, Shanghai, and Tianjin. Then the initial participants were asked to recommend other potential participants who work in Beijing, Shanghai or Tianjin.

Finally, from 3 November 2018, to 23 May 2019, a total of 247 questionnaires were distributed and 195 valid questionnaires were received, with an effective response rate of 78.9%. Characteristics of respondents and their projects are shown in Table 3.

Table 3. Characteristics of respondents and their projects

Background characteristic	Frequency	Percentage
Enterprise type
State-owned enterprise	79	40.50%
Private enterprise	101	51.80%
Others	15	7.70%
Education
Bachelor degree	114	58.50%
Master degree or above	43	22.10%
Others	38	19.40%
Work experience
<3 years	34	17.40%
3–5 years	113	57.90%
6–10 years	27	13.80%
>10 years	21	10.80%
Job position
Manager at the headquarters	38	19.50%
Project/department manager	58	29.70%
General management/technical personnel	99	50.80%
Project type
Parallel contracting	95	48.70%
Engineering Procurement Construction	79	40.50%
Public-Private Partnership	21	10.80%

3.3. Common method bias

Because each participant in the study answered all the questions in the questionnaire, it is necessary to control the common method bias. The control of common method bias mainly includes two methods: ex-ante control and ex-post control. In terms of ex-ante control, the participants were told that the research is only for academic research and there are no correct or wrong answers. The answers are kept strictly confidential. Second, different data sources were used to avoid the occurrence of homologous bias. Execution post control, Harman’s single-factor test and confirmatory factor analysis were used. First of all, this paper used exploratory factor analysis to extract common factors from all items (KMO = 0.910; Bartlett spherical test is significant at the level of 0.001). Six factors were obtained, which were consistent with the questionnaire structure. The variance accounted for by the first factor before rotation is less than 40%. Then, the confirmatory factor analysis was used to load all the items on a common factor. The results showed that the fit index of the model is worse than that of the measurement model. Therefore, it can be concluded that there is no common method bias in this research.

3.4. Construct reliability and validity

In the first stage, confirmatory factor analysis was conducted on the measurement model to test its reliability and validity. In this paper, PLS-SEM was used to complete the model test. The results showed that the loading of each factor is greater than 0.5, representing that further analysis can be carried out.

To test for the internal consistency of each latent variable, the values of the composite reliability and the Cronbach’s alpha of each measure were used as the judging criteria. These values should be higher than 0.70 (Hair, Ringle, and Sarstedt 2011). Results in Table 2 indicate that the measurement model has acceptable reliability. To examine convergent validity, the AVE value of each variable must exceed 0.50 (Hair et al. 2016). Results in Table 2 indicate that the AVE values of the latent variables are in the range of 0.715–0.891, representing acceptable convergent validity. In addition, the convergent validity was also evaluated by the factor loadings of each measurement item. As shown in Table 4, the factor loadings of each item on its respective construct are all greater than the 0.7 threshold and no cross-loading problem exists (Hair, Ringle, and Sarstedt 2011).

Table 4. Cross-loadings for measurement items

	PCA	PVA	PPA	IBCP	SN	PBC	BCP
PPA1	0.264	0.236	0.913	0.413	0.165	0.232	0.308
PPA2	0.200	0.17	0.834	0.306	0.22	0.24	0.241
PPA3	0.256	0.171	0.877	0.33	0.191	0.254	0.309
PCA1	0.836	0.356	0.221	0.417	0.279	0.185	0.356
PCA2	0.887	0.395	0.207	0.448	0.347	0.243	0.427
PCA3	0.888	0.417	0.288	0.497	0.37	0.368	0.424
PVA1	0.403	0.873	0.255	0.481	0.281	0.265	0.415
PVA2	0.476	0.913	0.152	0.558	0.308	0.335	0.431
PVA3	0.306	0.886	0.197	0.456	0.285	0.189	0.302
SN1	0.335	0.342	0.183	0.409	0.851	0.231	0.338
SN2	0.293	0.22	0.166	0.306	0.854	0.133	0.261
SN3	0.341	0.25	0.196	0.333	0.831	0.274	0.255
PBC1	0.32	0.285	0.261	0.456	0.271	0.888	0.471
PBC2	0.294	0.29	0.203	0.365	0.183	0.855	0.340
PBC3	0.21	0.199	0.245	0.35	0.204	0.887	0.396
PBC4	0.271	0.288	0.262	0.409	0.239	0.929	0.450
IBCP1	0.487	0.564	0.417	0.94	0.378	0.382	0.540
IBCP2	0.534	0.498	0.383	0.941	0.413	0.455	0.525
IBCP3	0.461	0.532	0.347	0.951	0.398	0.433	0.510
BCP1	0.358	0.291	0.364	0.517	0.229	0.393	0.849
BCP2	0.372	0.310	0.389	0.454	0.249	0.480	0.858
BCP3	0.426	0.288	0.344	0.513	0.321	0.369	0.854
BCP4	0.378	0.246	0.328	0.335	0.303	0.359	0.835
BCP5	0.433	0.251	0.411	0.502	0.354	0.389	0.845

Bold values represent standardized factor loadings of the items on their respective constructs.

To confirm discriminant validity, the average variance extracted (AVE) of each variable should be greater than the highest squared correlation (HSC) of that variable with other variables. As shown in Table 5, the AVE of each variable is higher than the HSC. Thus, the tests provide evidence for the variables’ discriminant validity.

Table 5. Correlation matrix and the square root of AVE of factors

Items	PPA	PCA	PVA	SN	PBC	IBCP	BCP
PPA	0.875
PCA	0.277	0.871
PVA	0.223	0.449	0.891
SN	0.215	0.384	0.328	0.846
PBC	0.275	0.311	0.301	0.255	0.890
IBCP	0.405	0.524	0.563	0.42	0.448	0.944
BCP	0.552	0.471	0.343	0.284	0.309	0.275	0.849

Bold values on the diagonal represent the square root of AVE.

3.5. Test of the structural model and hypotheses

The second stage is the test of the structural model and hypotheses. According to the calculation formula of the GoF index () proposed by Amato, Esposito Vinzi, and Tenenhaus (2004), the goodness–of–fit index of the model was tested. After calculation, GoF=0.583, which is greater than the maximum standard value of 0.36, indicating that the model has good goodness–of–fit index.

The SmartPLS 3.0 software was used to calculate the path coefficients and test their significance. In this study, the number of bootstrap samples is 5,000, as recommended by Hair, Ringle and Sarstedt (2011). The results of hypotheses testing are shown in Table 6, and Hypotheses H1a to H5 proposed in this paper have been supported.

Table 6. Results of hypotheses testing

Hypothesis	Path	Standard path coefficient	P Value	Inference
H1a	PPA -> IBCP	0.190	0.002**	Supported
H1b	PCA -> IBCP	0.211	0.000***	Supported
H1c	PVA -> IBCP	0.319	0.000***	Supported
H2	SN-> IBCP	0.143	0.040*	Supported
H3	PBC -> IBCP	0.198	0.000***	Supported
H4	PBC -> BCP	0.279	0.000***	Supported
H5	IBCP -> BCP	0.430	0.000***	Supported

* :p < 0.05, **:p < 0.01, ***:p< 0.001.

3.6. Results and discussion

3.6.1. Three behavioral attitudes promote IBCP

PVA has the greatest impact on the IBCP, followed by PCA and PPA. The path coefficient β of the impact of PVA on IBCP is 0.319. It is significant at the level of 0.001, supporting H1c. This result is consistent with Lu, Li, and Wang (2017), who demonstrated that if the contractor can perceive the value brought by the BCP (such as meeting the sense of responsibility, mission, etc.), the enthusiasm for BCP will be enhanced even more. It can be seen that the realization of enterprise value and social reputation, as well as the long-term benefits generated by BCP, are the important forces to promote the contractors’ intention to perform BCP.

The path coefficient β of PPA on the IBCP is 0.190, which is significant at the level of 0.01, representing H1a was supported. This result supports Chan, Chan, and Ho (2003), who believed that organizations with a strong perception of benefits are willing to cooperate with other stakeholders and will be proactive in solving emergencies to eliminate potential problems in the organizational interface. The reason is that the current income of projects such as project payment and contract compensation is a basic premise for the contractor to perform well. Although the PPA has a significant effect on the behavioral intention, the degree of impact is not strong. This result also confirms Xu et al. (2018), who believed that focusing on economic incentives has certain limitations in meeting the needs of contractors with social human attributes.

The path coefficient β of PCA on the IBCP is 0.211, which is significant at the level of 0.001, supporting H1b. This result supports Julin (2011). When an organization has a strong perception of cooperation, it can perceive the long-term benefits gained through cooperation. Such perception often enables them to effectively resolve disagreements in order to maintain mutually beneficial results. Many scholars have pointed out that if the relationship between the contractor and the owner is amicable, their intention to perform well will be greatly enhanced (Cheng 2016; Li et al. 2019). The positive and mutually beneficial cooperation paradigm is a win-win choice for both parties. If both parties have few contradictions and disputes during the project implementation process, the contractor relying on the positive emotion of win-win cooperation will be enthusiastic in performing the contract and willing to make much contributions to the project and the owner. In practice, contractors will make extra efforts in the construction process in order to obtain long-term cooperation or establish partnership with the owners to enter the cooperation list, such as relationship between China Construction Second Engineering Bureau LTD and Wanda Group.

3.6.2. SN has a significant but not strong influence on the IBCP

The results show that the path coefficient β of SN on the IBCP is 0.143, supporting the H2. This is similar to the results of Phua (2006), who believes that pressure from internal enterprises, peer competitors and policies would, to a certain extent, encourage the contractor to form the idea of performing BCP. The result also shows that the T value is 2.064, which is slightly greater than the critical value of 1.96. It is significant at the level of 0.05, but the influence on the IBCP is not strong. This conclusion is quite different from Li et al. (2019). He argues that SN is the strongest driver of behavior. According to the theory of planned behavior, pressure from internal enterprises, peer competitors, and industrial policies from the government can have a significant impact on the contractors’ intention to perform well. However, this paper through interviews found that the main motivation to motivate contractors to perform BCP in the Chinese construction market is mandatory norms, which is an endogenous institutional pressure from general contracting enterprises. Due to the lack of industrial policies from the government to guide the contractor to perform BCP and the negligence of the owner in promoting the demonstration effect of such behaviors, the important influencing factors for the formation of SN do not play an obvious role. In addition, the Chinese construction management system, culture, industrial practices, etc. are basically pro-owner, resulting in the construction market trading system supervision is more than enough, lack of incentives and attention, so the impact is relatively low.

3.6.3. PBC has a driving effect on the IBCP

The path coefficient β of the influence of PBC on the IBCP is 0.198, which is significant at the level of 0.001, supporting H3. According to the theory of planned behavior, PBC is divided into two dimensions, namely, self-efficacy and control. Self-efficacy reflects the behavioral subject’s confidence in its own ability and access to resources, while control is reflected its ability to access external resources. Due to the contractors’ weak position and insufficient network capacity, the control of external resources is not enough 。to influence the intention to perform IBCP. Therefore, the contractors’ PBC is mainly reflected in the control of the enterprise’s own capabilities and internal resources related to contract completion. Bandura and Wood (1989) pointed out that the judgment and understanding of the actors on their abilities play a major role in their performance behaviors. The stronger the sense of self-efficacy, the stronger their efforts will be.

3.6.4. PBC and IBCP can drive the BCP

The effect coefficient β of influence of PBC on the BCP is 0.279, which is significant at the level of 0.01, assuming H4 is supported. The results show that if the contractor evaluates that he has sufficient technical ability, management ability, experience and resources, it is easier to generate the willingness to perform the contract well and take the initiative to contribute to the project. Therefore, the contractor’s PBC has a dual driving effect. This is consistent with the research conclusion based on TPB in other fields.

The path coefficient β of the impact of IBCP on the BCP is 0.430, assuming that it is significant at the level of 0.001, the H5 is supported. In other words, if the contractor has the intention to perform BCP, it is very likely to implement active performance. This is also consistent with the research conclusions of Li et al (2019), who studied the organizational relationship behavior of large-scale projects. The results of the study point out that behavioral intention has a direct driving effect on behavior in the field of construction engineering. They believe that the intention can directly determine the act, and when the organization believes that BCP is beneficial to the organization and is willing to perform the behavior, it can stimulate the contractor to perform the act well.

4. fsQCA in configurational analysis

This study adopts fsQCA, which is one of the most popular techniques in the configurational analysis (Liu et al. 2017; Ragin 2009). fsQCA is a set-theoretic approach, which considers the relationship between attributes according to the relationship between sets (Ragin 2009). This is different from multivariate linear regression and SEM examining the net effect of independent variables on dependent variables (Bedford and Sandelin 2015). The reasons why fsQCA was selected in this study are as follows: First, fsQCA is able to examine necessity and sufficiency of combinations of factors affecting the contractors’ intention to perform IBCP (Misangyi et al. 2017; Ragin 2009). Necessity means that a condition must exist for an outcome to occur; and sufficiency means that a condition can bring about an outcome (Ragin 2009). Second, fsQCA presents equifinality results and generates multiple paths which are equally valid (Wagemann, Buche, and Siewert 2016). Third, fuzzy sets permit calibrating partial membership in sets using values ranging from 0 (non-membership) to 1 (full membership) (Ragin 2009).

4.1. Calibration

The requirement of fsQCA for processing data is a continuous set of 0 ~ 1, which is called the member ship score. Therefore, it is necessary to check the Likert measurement of variables in this study before processing the data and convert it into the degree index of full membership, full non-membership and half cross-over point. We first calculated the average value of all items measuring each construct. Second, variables were calibrated into set memberships in terms of three thresholds. Different threshold values correspond to three membership degrees of 0.95, 0.5 and 0.05 respectively (Ragin 2006). Since the Likert scale of 5 points was used in this research, with “5” representing the highest degree and “1” representing the lowest degree, the critical values of 0.95 and 0.05 were respectively selected as “5” and “1”. We referred to the practices of Loughran and Ritter (2004), Bell, Filatotchev, and Aguilera (2014), taking the average value as the critical value of the membership degree of 0.5.

4.2. Necessary conditions

This study used fsQCA 3.0 software to analyze the relationship between all variables and the contractors’ IBCP and identified the antecedent configuration that determines the contractors’ intention to perform BCP. First, the necessity of the IBCP in the interpretation of the five antecedents of PVA, PPA, PCA, SN, and PBC was analyzed. As shown in Table 7, the necessity of each antecedent condition does not exceed the consensus threshold of 0.9, and thus does not constitute or approximate the necessary condition (Ragin 2006). In view of this, we proceeded from the combination of multiple factors to further analyze the relationship between the factor configuration and perform IBCP.

Table 7. Analysis of necessary conditions

	Consistency	Coverage
PPA	0.729	0.731
PCA	0.804	0.761
PVA	0.789	0.753
SN	0.851	0.708
PBC	0.888	0.691
IBCP	0.861	0.734

4.3. Sufficiency conditions

In the last step, the sufficiency test was carried out. A truth table was initially built, listing all possible configurations for explaining the contractors’ IBCP. In this study, a truth table of 2⁶ rows was produced (five denotes the number of conditions). Each row presents a possible configuration. In this study, the consistency threshold was 0.85 (Tho and Trang 2015). In addition, in order to ensure that the configuration with sample frequency greater than or equal to the frequency threshold covers at least 80% of the samples (Ragin 2009), we set the sample frequency threshold to 10. Configurations with consistency values greater than 0.85 are considered sufficient to produce results. The fsQCA analysis generates three possible solutions, namely the complex, intermediate and parsimonious solutions. The intermediate solution is most reported (Rihoux and Ragin 2009). As shown in Table 8, the intermediate solution results show that two configurations were obtained, namely, PVA * PCA * SN * PBC * IBCP (Configuration 1); and PPA * PCA * SN * PBC * IBCP (configuration 2). The coverage values of the three configurations range from 0.576 to 0.521, and the overall coverage rate is 0.746. The consistency values range from 0.873 to 0.883, and the consistency of the overall scheme is 0.768, which is higher than the critical value of 0.75 (Ragin 2009). In summary, there are three equivalent paths that lead to the formation of the contractors’ intention to perform BCP.

Table 8. Combinations of sufficiency conditions

Configuration	Raw coverage	Unique coverage	Consistency
PVA * PCA * SN * PBC * IBCP	0.576	0.087	0.873
PPA * PCA * SN * PBC * IBCP	0.521	0.023	0.883
Solution coverage	0.746
Solution consistency	0.768

*denotes the logical “and”

4.4. Results and discussion

The results show that the combination of PVA, PCA, SN, PBC and IBCP (configuration 1), and the combination of PPA, PCA, SN, PBC and IBCP (configuration 2) are likely to trigger the contractor BCP. It can be seen that PCA, SN, PBC and IBCP are the core of BCP, which cannot be replaced. PBC represents the productive capacity of organization, while IBCP represents the behavioral intention of organization. Therefore, the capacity is the foundation, and intention is the key. Therefore, in reality, those who are willing to perform BCP are often contractors with super-strong production capacity, because they have the ability and will be affected by the PCA which will trigger behavioral intention. If the configuration 1 and the configuration 2 managerial, it is not difficult to find the difference between them lies in the PVA and the PPA, which indicates that the PVA and the PPA have a substitutive relationship in the configuration of BCP. In practice, if there is a lack of PVA or PPA, the remaining factor can be used instead. This is also the fundamental reason why there are three equivalent paths for the formation of the contractor BCP.

An interesting finding is that the results of PLS-SEM show that the PVA has the greatest influence on the contractors’ intention to perform BCP among the behavioral attitudes. However, the results of fsQCA show that PCA is the most important among the three behavioral attitudes and cannot be replaced. Similarly, SN exerts the smallest impact among the five factors that affect the contractors’ intention to perform BCP but is indispensable. This indicates that the traditional regression-based analysis paradigm can only answer the degree of influence of a variable, but cannot judge the degree of importance of the variable, which is also one of the reasons why configuration analysis is introduced in this study.

5. Conclusions and implications

This study used SEM and fsQCA to reveal the formation mechanism of the contractor BCP in construction projects with the aid of TPB analysis framework. The following main conclusions are drawn: 1) The results of SEM show that PVA, PCA, PPA, SN, PBC and IBCP have significant positive effects on the contractor BCP. 2) Two equivalent paths of the contractor BCP were identified, namely, PVA, PCA, SN, PBC, IBCP, and PPA, PCA, SN, PBC,IBCP. This result further indicates that the trigger of the contractor BCP does not need all influencing factors to exist simultaneously, as long as the requirement of one of the two equivalent paths is met. 3) PCA, SN, PBC and IBCP are the core of the contractors’ IBCP. This result shows that among various configurations of the contractor BCP, some influencing factors are indispensable to each configuration, and they are the key to the formation of the contractors’ willingness to perform BCP.

5.1. Theoretical implications

This paper contributes to the academia in three aspects. Foremost, the analysis framework of contractor CPB is constructed. The realization of the project target is closely related to the contractor’s behavior. Therefore, it is necessary to separate the contractor’s behavior for research. However, previous studies mainly focused on opportunistic behaviors or cooperative behavior but ignored incentive factors of contractors’ intention of BCP. The contract reference point theory defines the different performances under the influence of two different execution behaviors. Therefore, it is necessary to study the contractors’ BCP. Regarding voluntary behaviors, it is of high value to study one’s intention to perform. this paper not only conducts a comprehensive analysis on the internal and external factors that affect the contractor BCP but also offered additional insights. Through the configurational analysis, two equifinal combinations of incentive factors of contractor BCP are identified. This study not only extended contractors’ behavioral literature and identified the incentive factors of contractor BCP but also promoted the research on project performance evaluation to the behavioral level.

Second, this paper identifies the key to institution construction to promote the contractor BCP. Previous studies have mostly focused on the prevention of contractors’ opportunistic behaviors, and have also paid special attention to creating a competitive environment with clear rewards and punishments to urge contractors to implement BCP. This paper has examined the factors affecting the contractor BCP, and has identified the key in the construction of the project management institution, such as attitude toward cooperation and subjective norm (the core for BCP), attitude toward social value(the greatest impact on the intention of BCP). These factors mean construction of corporate social responsibility, the construction of flexible engineering culture, the mutual trust system between the owner and the contractor, and the cultivation of the contractor’s core competitiveness. This study is a new supplement to the previous contractor management institution, which mainly suppressed opportunism.

Third, from the perspective of methodology, this study adopts SEM and fsQCA synthetically, forming a complementary method. “The objective of building and examining configurations is not to explain variance but to describe and explain combinations of features which accurately indicate a high score in an outcome condition” (Woodside 2014). SEM is good at explaining which factors have significant effects on the contractor BCP. However, fsQCA provides configurational reasons. Results of SEM show symmetric relation and net effect, while fsQCA contains complex causal relation and can analyze the necessary and sufficient conditions of the contractors’ intention to perform BCP.

5.2. Managerial implications

From a practical point of view, this study reveals the complicated connection of antecedent of BCP. It is helpful for construction practitioners to develop fine-grained cognition of performance improvement from the perspective of behavior. More importantly, the survey results show that the triggering of BCP requires a combination of influencing factors rather than a single factor. Specifically, the three management impacts are summarized as follows.

First, optimization and perfection of project contracting policies. The results of SEM show that the SN has less significant influence compared with others, which means that contractors do not perceive enough BCP of demonstration norms. However, the results of fsQCA show that SN is the core of the contractors’ intention to perform BCP. Therefore, it is urgent to speed up the optimization and perfection of the project contracting policy. The policy should strengthen the return in terms of the contractors’ benefits and other non-material benefits after performing BCP. The government should strengthen supervision over the orderly competition, provide institutional guarantee for the formation of a good atmosphere, and encourage contractors to produce the spirit of contract.

Second, building a culture of mutual trust in the general contracting market. The cooperative relations and communication and coordination mechanisms between the contracting parties are very important to induce the contractors to perform BCP. The results show that PCA is the core of the IBCP. Therefore, in the construction market, the owner should commend the contractor for the perfect performance to establish a friendly and cooperative relationship and stimulate the contractors’ motivation to perform BCP. In addition, the most important feature of the general contracting is that the owner should give the general contractor full control over the project so that the general contractor can perceive sufficient control over the resources for performing BCP. However, the transfer of control rights cannot be realized without trust between the contractor and the owner. The construction of an organizational culture of mutual trust and mutual benefit between the contractor and the owner has become the focus of developing the general contracting system.

Third, The construction of general contracting enterprises’ responsibility and core competitiveness. Different from the traditional view, although self-interest preferences such as PPA can stimulate the BCP, it has the least incentive effect. Irrational preferences, such as PVA, have the greatest impact and are the factors with the highest efficiency that promote the contractors to perform BCP. Large contractors with social responsibility pay much attention to the realization of their own values. Therefore, general contracting enterprises should attach importance to the construction of the modern enterprise system and social responsibility, and maintain and enhance the social reputation of enterprises.

Forth, the results show that attitude toward social value is an efficient factor to promote contractors to implement CPB. Contractors with social responsibilities attach great importance to the realization of their own values. Therefore, a supportive environment should be created to enhance the social responsibility and honor of general contractors.

5.3. Limitations and future research

This paper makes an empirical analysis on the driving factors and formation mechanism of the contractors’ intention to perform BCP and provides reliable suggestions for inducing the contractor to perform BCP under the construction context. However, there are still some deficiencies in the research, which needs to be improved by the follow-up research.

First, future studies may further consider the factors that affect the contractors’ intention to perform BCP and strengthen the depth of the research on antecedents. Examples of potential factors are the contractors’ value orientation, project benefit, the owner’s trust, the formal institutional environment of general contracting projects, internal resources of general contracting enterprises, etc. Second, further research may consider the interaction between antecedents of IBCP such as PVA, PPA, and PBC, e.g. whether or not there is a certain relationship among them, and which relationship leads to their substitutive relationship. Finally, cross-level research may be introduced to consider the differences between different types of transaction subjects, further distinguish the transaction characteristics of the project and verify them respectively.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This work was supported by the National Natural Science Foundation of China [71572128, 71471132 and 71402119].