1. Introduction
With the expansion of the digital economy and the maturity of globalization, the breadth and depth of enterprises’ demand for resources have increased considerably. Given the increasing difficulty of meeting development needs, enterprises have pursued more heterogeneous resources through external cooperation [
1,
2]. Increasingly more companies join the supply chain management, seeking their own long-term development with the overall operation ability of the supply chain. The competition among enterprises has gradually evolved into competition among supply chains. Improving supply chain management has been recognized as a crucial strategy for many businesses seeking to achieve sustainable development [
3,
4,
5]. However, because of uncertainties in the external environment and the limited rationality of the internal organization, cooperation and competition in the supply chain often occur simultaneously among member enterprises. Constant opportunism, short-term gaming, moral damages, and other problems result in high transaction costs, which erode profits and benefits brought about by supply chain management [
6,
7]. This issue has prompted many scholars to ask: What kind and extent of cooperative relationship should be maintained between companies within the supply chain?
Resource dependency theory suggests that an organization requires multiple resources to survive and must interact with other organizations in its environment to obtain much-needed resources [
8,
9]. Transaction cost theory suggests that good cooperative relationships help reduce transaction costs. However, maintaining relationships costs money, and different relationships have different prices [
10,
11,
12,
13]. Social network theory posits that resources can be obtained not only through possession but also through network relationships, which can be divided into “strong relationships” and “weak relationships” [
14]. Some scholars believe that strong relationships are based on emotion and trust, which make it easier for firms to obtain high-quality information and resources and improve the efficiency of information and resource transfer [
15]. In contrast, some scholars suggest that close relationships deliver mostly redundant resources and require high costs for relationship maintenance [
16]. Some argue that weak relationships can provide firms with more valuable heterogeneous resources and do not require higher maintenance costs [
17], while others believe that the loose structure of weak relationships is not conducive to information dissemination and knowledge sharing among firms, and it is difficult to meet firms’ innovation needs [
1]. Thus, existing studies have unanimously recognized that “partnership is an important way for enterprises to obtain external resources” but have different understandings of the question of what kind of relationship between enterprises can achieve higher performance [
18]. Is it “distance or intimacy”? This seems to be a dilemma. Thus, when this dilemma encounters Eastern culture, what is the answer?
For Western developed countries, the national economic development level is relatively high, the market mechanism is relatively perfect, and the transactional contract has a strong binding force. However, for Eastern developing countries represented by China, the national economic development level is low, the market mechanism is not perfect, and the transactional contract is weak [
19]. In addition, under the profound influence of Eastern culture such as “collectivist values” and “seeking common thinking”, the establishment of close strategic partnership often becomes the priority of enterprises in Eastern developing countries [
3,
14,
20,
21,
22]. They hope to bind interests and share risks with upstream and downstream enterprises through the dual constraints of transactional and relationship contracts to fill the “institutional hole” and “market gap” existing in the entrepreneurial environment to a certain extent [
23]. However, existing studies on strategic partnership mostly analyze enterprises in Western developed countries, while there is still a lack of sufficient discussion on enterprises in Eastern developing countries [
23]. The impact of strategic partnership on enterprise performance in Eastern developing countries requires further study.
While the majority of existing research indicates a positive correlation between strategic partnership and enterprise performance [
24], some studies indicate a negative [
2] or nonexistent relationship [
6]. It can be seen that there may be complex mechanisms between strategic partnership and enterprise performance that need to be further explored. Close strategic partnerships foster mutual trust and commitment, which improves the quality and level of information sharing among firms [
25,
26,
27,
28]. Sufficient information sharing enhances supply chain flexibility, allowing firms to profit by responding quickly to market changes [
14,
29,
30]. This implies that information sharing and supply chain flexibility may play critical roles in the relationship between strategic partnerships and enterprise performance [
31,
32]. Clarifying this issue would help reveal the “black box” between strategic partnership and enterprise performance.
The core question in this study is: In the era of the digital economy, how does strategic partnership affect enterprise performance in developing countries with Eastern culture? To answer this question, based on the theory analysis on resource dependence theory, transaction cost theory, and social network theory, this study constructed a chain mediation model analyzing Chinese enterprises’ supply chain management practices. In the model, a strategic partnership was used as the independent variable, information sharing, and supply chain flexibility as the intermediary variables, and enterprise performance as the dependent variable. Data from 243 Chinese enterprises were used in carrying out the empirical tests. Our research’s first objective is to determine whether close strategic relationships increase corporate performance in developing Eastern countries, in order to contribute to the contextual analysis of cooperative relationship research. Second, by examining the chain mediating role of information sharing and supply chain flexibility, we can better understand the effects and pathways of strategic partnership on business performance. Finally, the findings are discussed in order to provide advice and suggestions on how to apply supply chain management, assisting firms in developing countries of Eastern culture in achieving sustainable development in the digital economy era.
3. Research Method
3.1. Questionnaire Design
The scale design is the core part of the questionnaire design, and the scales used in this study are all maturity scales. The reliability and validity of maturity scales are highly recognized in literature and are constantly improved through repeated use, validation, and refinement [
95]. Mature scales have certain limitations, such as cultural and linguistic differences that may lead to understanding bias, and the timeliness of the study may be compromised over time. To mitigate these limitations, the scales used in the study were applied and reported in articles published in recent years in authoritative journals with high citation rates. All scales were empirically validated by a sample of Chinese companies to confirm their reliability and fit. Furthermore, back translation, expert judgment, and pretesting were used to revise and improve the questionnaire.
(1) Back translation. First, two researchers with good English competency translated the scale into Chinese. The translated versions were combined and were discussed with the research team. The combined Chinese scale was then translated into English by an English teacher and compared with the original version to check the accuracy of the translation [
96].
(2) Expert evaluation. We invited three recognized experts to evaluate the questionnaire: a professor from Sun Yat-sen University in the field of supply chain management research, a professor from Shandong Normal University in business administration, and a corporate executive with practical experience in supply chain management from China’s second largest e-commerce company, jd.com. Based on their professional knowledge and work experience, these experts were asked to comment on the language (i.e., if there were unclear semantics and multiple meanings), privacy issues, logic of the questions, difficulty of the question, overall length of the questionnaire, etc.
(3) Pretest. For research cost and convenience, the research team pretested the questionnaire on two EMBA classes in a Chinese university. A total of 80 questionnaires were distributed in the pretest, and 50 valid questionnaires were finally obtained after questionnaire screening. Using preliminary statistical analysis, the scientificity and rationality of the questionnaires were evaluated to avoid any significant deviation.
The formal questionnaire (“Questionnaire on Supply Chain Management Practices of Chinese Enterprises”) was finalized and included the following: The first part is the introduction, which provides the purpose of the questionnaire, the commitment to respondents, contact information of the researchers, and other relevant information. The next part is the text content, which elicits information on the company (i.e., nature of the company, industry type, number of employees, corporate management culture, fixed assets, and turnover) and the supply chain management practices (i.e., strategic partnership, information sharing, supply chain flexibility, and corporate performance).
Strategic partnership (SP) refers to the research of Li et al. [
39] and comprises six items, including “solving problems with cooperative enterprises”. Information sharing (IS) refers to the research scale of Li et al. [
39], comprising six items, such as “will not inform the cooperative enterprise in advance when the demand changes (reverse item)”. Supply chain flexibility (SCF) refers to the research scale of Chuu [
97], with eight items such as “supply network flexibility”. Enterprise performance (EP) refers to the research scale of Liu et al. [
98] and is composed of six items, including “the company’s sales profit has increased over the past three years”. Measurements were made using a 7-point Likert scale. See
Appendix A for details.
3.2. Data Collection
We conducted the following procedural controls to maximize the recovery rate of the questionnaire, ensure the representativeness of the sample, improve the quality of the collected data, and reduce the common method bias:
(1) The questionnaire was given in three formats (i.e., paper, online, and electronic) to facilitate respondents to choose a convenient way to answer the questions.
(2) The study conducted sample collection by various means, such as checking the yellow pages of enterprises, contacting enterprises cooperating with schools and enterprises, interviewing participants of executive training courses in universities, and distributing questionnaires in the field. Multiple ways were used in parallel to improve the speed of questionnaire collection.
(3) The questionnaire was anonymously answered by respondents who were required to be managers at supervisory level or above and had worked in the company for more than two years to ensure that the respondents have more objective and comprehensive knowledge of the company.
(4) Considering the variety in the economic development levels of different cities, the study used Guangzhou, Jinan, and Jishou as the first-, second-, and third-tier representative cities, respectively. The questionnaire surveys were initially conducted in these areas and radiated to the surrounding cities.
(5) Industry feature was not the focus of the study, so the enterprises selected for the survey came from different industries, such as manufacturing, medical and pharmaceutical, electronic communication, and culture and sports, to improve the representativeness of the sample.
Data collection was conducted from March 2020 to August 2020, with a total of 500 questionnaires distributed and 298 finally recovered.
3.3. Sample Screening
To ensure data correctness and completeness, we screened the recovered 298 questionnaires by applying the following principles:
(1) The questionnaires whose missing values exceeded one-fifth of the total data were rejected. For questionnaires with less than one-fifth missing data, we first tried to supplement missing information by telephone or email follow-up. For questions that could not be remedied but were not critical, we used the mean replacement method to fill in the data. If the questions were critical, the questionnaire was discarded.
(2) The questionnaires that showed significant regularity in the answers were discarded. This included questionnaires with the same response for all questions and those who chose all “neutral” answers.
(3) The questionnaires with blurred handwriting were confirmed by phone/email or re-collected.
(4) The electronic questionnaires with garbled codes were re-collected.
(5) For multiple questionnaires from the same company, the mean value was taken and recorded as one questionnaire. Because the study conducted sample collection through multiple channels, there were 16 questionnaires from 6 companies, and the data from the same company were the mean values and recorded as 6 questionnaires in total.
(6) Questionnaires with outliers, noticeable logical errors, or inconsistent answers were re-collected or discarded.
Through the sample screening, 243 valid questionnaires were finally obtained, with an effective recovery rate of 48.60%. The basic information of the sample enterprises follows: In terms of the ownership, there were 57 state-owned enterprises (23.46%), 89 private enterprises (36.63%), 34 joint-stock enterprises (13.99%), 42 joint-venture enterprises (17.28%), and 21 other enterprises (8.64%). In terms of employee count, 46 had less than 100 employees (18.93%), 79 had 100−499 employees (32.51%), 42 had 500−999 employees (17.28%), 37 had 1000−4999 employees (15.23%), and 39 had more than 5000 employees (16.05%). In terms of the enterprise culture, there were 145 companies from Mainland China (59.67%), 27 from Hong Kong, Macao, and Taiwan (11.11%), 36 from Europe and America (14.81%), 19 from Japan (7.82% of the total sample), and 16 from other cultures (6.58%). In general, the sample distribution conforms to reality and has certain representativeness.
5. Hypothesis Testing
Previous studies have confirmed that company size (with the number of employees as a proxy variable), enterprise nature, and industry type affect enterprise performance [
102]. To exclude the effects of these factors, they were taken as control variables and used as dummy variables for measurement.
The results of the fitting index of the research model show that the model has a good overall fit and that hypothesis testing can be carried out (χ
2/df = 3.091, RMSEA = 0.093, NFI = 0.812, RFI = 0.839, IFI = 0.865, TLI = 0.849, CFI = 0.864). The Bootstrap function in AMOS23.0 software was used, selecting the nonparametric percentile Bootstrap method for deviation correction and extracting 5000 times. The 95% confidence interval (the 2.5th percentile and 97.5th percentile) was then analyzed to determine if zero is included in the range. If zero is not contained, the effect is significant; otherwise, the effect is not significant. The results of hypothesis testing are summarized in
Table 5 and the model output in
Figure 2.
5.1. Direct Effect Testing
(1) For the influence path from strategic partnership to enterprise performance, the direct effect 95% confidence interval was [0.092, 0.570] excluding 0, while the standardized path coefficient β1 was 0.334 (p < 0.001). The results indicate that strategic partnership has a significant direct positive effect on enterprise performance and that Hypothesis 1 is verified.
(2) For the influence path from strategic partnership to information sharing, the direct effect 95% confidence interval was [0.430, 0.724] excluding 0, and the standardized path coefficient β2 was 0.579 (p < 0.001). The values suggest that strategic partnership has a significant direct positive effect on information sharing and that Hypothesis 2 is confirmed.
(3) For the influence path from information sharing to enterprise performance, the direct effect 95% confidence interval was [0.067, 0.447] excluding 0, and the standardized path coefficient β3 was 0.262 (p < 0.001). The findings indicate that information sharing has a significant direct positive effect on enterprise performance and that Hypothesis 3 is verified.
(4) For the influence path from strategic partnership to supply chain flexibility, the 95% confidence interval of direct effect was [0.315, 0.650] without 0, and the standardized path coefficient β5 was 0.496 (p < 0.001). This means that strategic partnership has a significant direct positive effect on supply chain flexibility. Hypothesis 5 is also verified.
(5) For the influence path from supply chain flexibility to enterprise performance, the 95% confidence interval of direct effect was [0.101, 0.467] excluding 0, while the standardized path coefficient β6 was 0.270 (p < 0.001). The results suggest that supply chain flexibility has a significant direct positive effect on enterprise performance and that Hypothesis 6 is verified.
(6) For the influence path from information sharing to supply chain flexibility, the direct effect 95% confidence interval was [0.067, 0.391] without 0, and the standardized path coefficient β8 was 0.230 (p = 0.002). The findings indicate that information sharing has a significant direct positive effect on supply chain flexibility. Hypothesis 8 is also verified.
5.2. Indirect Effect Testing
The test results showed that the standardized path coefficients for direct effects β1, β2, β3, β5, β6 and β8 were all significant. For the influence path from strategic partnership to enterprise performance, the 95% confidence interval of indirect effect was [0.166, 0.482] excluding 0, indicating that the indirect effect between strategic partnership and enterprise performance is significant. For the influence path from strategic partnership to supply chain flexibility, the 95% confidence interval of indirect effect was [0.010, 0.143], not including 0, indicating that the indirect effect between strategic partnership and supply chain flexibility is significant. In the influence path from information sharing to enterprise performance, the 95% confidence interval of indirect effect was [0.042, 0.235], excluding 0. This suggests that the indirect effect between information sharing and enterprise performance is significant.
Overall, the results indicate that information sharing and supply chain flexibility jointly play a chain mediating role between strategic partnership and enterprise performance. The total mediating effect of the two was 0.322 (β2β3 + β5β6 + β2β6β8), accounting for 49.09% of the total effect. The mediating effect of information sharing alone was 0.152 (β2β3), accounting for 23.12% of the total effect. Hypothesis 4 is verified. The mediating effect of supply chain flexibility alone was 0.134 (β5β6), accounting for 20.41% of the total effect. Hypothesis 7 is also verified. The chain mediating effect of information sharing and supply chain flexibility was 0.036 (β2β6β8), accounting for 5.48% of the total effect. Hypothesis 9 is verified.