This study is premised on the collaboration theory, which postulates that collaboration is key to problem solving in every sector of the economy (Gray 1985:911). Collaboration relates to a joint decision-making strategy designed to collectively share resources and with the ultimate objective of solving problems and challenges that one cannot otherwise resolve (Selin & Chavez 1995:190). Effective collaboration is a source of competitive advantage that aims to improve customer service, profit generation, asset utilisation and cost reduction (Pooe & Mathu 2011:320). Characterising the collaboration theory are five elements, namely (1) the independence of stakeholders, (2) solutions to problems and other challenges are found through mutual understanding cooperation, (3) parties must contribute to decision-making, (4) parties must take responsibility on the development of the strategy and (5) parties must understand the salient point of the strategy plan as it is there to provide a foundation from which they can manage and deal with the uncertainty of the environment in which they are operating (Gray 1989:227). This theory was deemed appropriate for this study since it is about the effective and efficient interaction of parties through the mutual sharing of input resources. Furthermore, the article contends that collaboration underpins information sharing, supplier trust and supplier synergy.

There is now hardly any dispute about the role and contribution of SMEs in economies. The SME sector has been described as an essential factor in sustaining the economic growth and development of most economies due to their abilities to adjust to environmental changes and technological orientation (Adams et al. 2012:25). According to Abor and Quartey (2010:220), the SME sector has massively contributed to growth and development of the South African economy, especially in the area of employment creation, which is crucial in fostering and improving socio-economic development, alleviating poverty and reducing crime. Not surprisingly, the performance of SMEs has generated great attention from scholars and other decision-makers (Alpkan, Yilmaz & Kaya 2007:152; Sandada, Pooe & Dhurup 2014:662). In his study on SME performance, Sidik (2012:376) conceptualised SME performance in terms of its ability to increase or improve its overall outcomes in terms of cash flow, profitability, customer satisfaction, sales growth and employee growth. Since there is usually little separation between ownership and control of operational activities in SMEs, Ahmad, Halim and Zainal (2010:67) posit that the success of SMEs is related to the abilities of their owners to make the best use of their managerial skills, experience and expertise. Due to their socio-economic role and importance, it was found critical that the influence of information sharing, supplier trust and supplier synergy on supplier performance be investigated from the perspective of SMEs.

Information is vital for the effective functioning of any business. It has been described as the lifeblood of organisations. Information sharing refers to the extent to which a firm openly communicates important and sensitive information to its partners (Shou et al. 2012:2). The notion of information sharing has attracted significant attention from a number of scholars who have stressed its role and influence in the supply chain environment (e.g. Jraisat, Gotsi & Bourlakis 2013:323; Kembro & Naslund 2014:179; Leung, Choy & Kwong 2010:64). Information sharing has also been regarded as an effective predictor factor of a supply chain's effectiveness (Zhang & Chen 2013:181). In that regard, a study by Hsu et al. (2008:297) found that information sharing contributes largely to improved relationships between suppliers by facilitating efficient coordination and responsiveness as well as integration of partners’ information systems. Furthermore, effective sharing of information between supply chain members has also been determined to be a crucial antecedent aspect in mitigating the negative impact of the bullwhip effect (Kelepouris, Miliotis & Pramatari 2008:3657). Chinomona and Pooe (2013:8) found that timely and accurate sharing of strategic information can foster the reduction of unwarranted wastages and costs in a supply chain, thus leading to increased SME profitability. Since information sharing is considered so important for effective supply chains, it is a factor worthy of investigation in the case of supplier performance, especially in the context of SMEs.

Supplier trust is another key variable in this study. Supplier trust has been defined by Cavusgil, Deligonul and Zhang (2004:7) as the confidence that trading partners have about each other's reliability and integrity. Trust has been defined to be an important contributor and facilitator to any transaction that calls for a mutual involvement and collaboration of diverse parties (Cheung et al. 2011:184). Supplier trust has also received attention from a number of scholars who have extensively investigated its impact and relevance in buyer-supplier relationship (Bonte 2008:855; Gao, Sirgy & Bird 2005:397; Kwon & Suh 2005:26; Wasti & Wasti 2008:118). Trust between suppliers and the buying firm has been viewed to be an important predictor factor impacting on suppliers’ ability to integrate their supply chain activities (Yeung et al. 2009:66). According to Panayides and Lun (2009:35), supplier trust plays a vital role in enabling and enhancing a firms’ innovativeness and consequently its supply chain performance outcomes. In light of the fact that it is so vital for SMEs to be integrated in supply chains, it is imperative that supplier trust be well understood.

Supplier synergy is yet another important aspect deserving of a clearer understanding. Supplier synergy can be defined as the coordination and complementarity of similar activities performed by two or more firms, groups or individuals in a business relationship to produce superior mutual outcomes (Osarenkhoe 2010:201). According to Hoegl and Wagner (2005:530), supplier synergy has a positive effect on the organisation's ability to provide a quality product to their customers. This reflects the importance that coordinated work between suppliers’ parties has on the competitiveness of a firm's supply chain. Wu et al. (2010:117) believe that supplier synergy allows firms to eliminate or lessen the recurrent technical and quality problems in production. Besides, effective synergistic collaboration also leads to the sharing of strategic and key explicit and tactic knowledge. In their study, Yeung et al. (2013:546) postulate that sound synergy enables businesses to obtain great benefits, such as improved quality of products and flexibility of operations. Thus, supplier synergy reduces unnecessary duplications responsible for possible inefficiencies within the supply chain. Therefore, supplier synergy becomes ever so critical for SMEs characterised by flexibility of operations and whose quality of products is constantly under scrutiny.

Perhaps the most important yardstick to determine the success of suppliers is their performance. Wu et al. (2010:115) define supplier performance as how well a supplier supplies the required products to the buyer as reflected through operational outcomes such as quality, delivery, responsiveness, cost and technical support. Huang, Yen and Liu (2014:64) posit that a firm's effort to adequately integrate its supply chain activities has a massive impact on their supplier performance. In addition, supplier performance has been described as a major predictor of reseller satisfaction (Yilmaz, Sezen & Kabadayi 2004:854). Thus, well-performing suppliers represent a key factor that businesses should strive to develop and maintain in their long-term sustainability and profit's aspiration (Sanchez-Rodriguez, Hemsworth & Martinez-Lorente 2005:289). Krause et al. (2007:528) further suggest that a supplier performing at optimal level is crucial in enabling the buying firm to reach its performance outcomes in terms of serving its customers more efficiently. A well-performing supplier is likely to remain in the supply chain and grow its relationship with the buying firms – something that all SMEs should aspire to.

Information sharing has always been highly regarded as an effective facilitator and a key enabler of collaboration between supply chain members (Cheng & Wu 2005:1159), which suggests the vital role that information sharing has in contributing to the development and improvement of activities performed by firms’ supplier partners. This view has been further stretched by Gosh and Fedorowicz (2008:453) who posit that sound and accurate exchange of sensitive and strategic information amongst supplier is important for the establishment of strong and long-lasting trusting relationship between the supply chain partners. Furthermore, concise sharing of information has been described as a prerequisite of strengthening buyer-supplier trust in that it enables them to collectively share the same objectives and develop mutual problem resolutions (Kwon & Suh 2005; Nyaga, Whipple & Lynch 2010:101). Moreover, Doney and Cannon (1997:41) assert that the ability and willingness to openly share confidential information portrays their trustworthiness to engage in a sound and collaborative type of relationship. Based on the aforementioned, the following hypothesis is postulated:

The importance and impact of information sharing has been significantly investigated in a number of studies (e.g. Cheng 2011:374; Sheu, Yen & Chae 2006:24; Zhang & Chen 2013:178), stressing its vital role in buyer-supplier relationships. According to Barratt (2004:30), information sharing plays a pivotal role in assisting supply chain partners to collaboratively engage in mutual strategic activities and decision-making. This therefore may allow them to effectively and efficiently work together and foster the value creation of each supply chain unit in a more synergistic manner. In this regard, Prajogo and Olhager (2012:514) posit that where there is an adequate and interrupted flow of information between each unit of activity, supply chain networks can increase the suppliers’ capabilities to perform synergistically. In light of the foregoing, it is hypothesised that:

Trust has been described by Zhang, Cavusgil and Roath (2003:550) as a key determinant in strong buyer-supplier relationships, as it enables supply chain partners to conduct their transactions and operations without fear of vulnerability. According to Jain et al. (2014:315), a trusting relationship between a buyer and its suppliers allows both parties to mutually share critical resources as well as strategic information freely. This therefore may enable them to obtain significant output resources that may improve the operational performance of their suppliers in terms of responsiveness, reliability and effectiveness in strategic problem resolution (Handfield & Bechtel 2002:367; Panayides & Lun 2009:38), which may consequently lead to the enhanced performance of the buying firm (Corsten & Kumar 2005:80; Nielsen 2007:337). This shows the likely influence of supplier trust on supplier performance, leading to the following hypothesis:

Suppliers engaging in synergistic collaboration have viewed it as a major strategic approach designed to enhance their core performance (Duffy & Fearne 2004:57; Sheu et al. 2006:24). This shows the benefit that suppliers working synergistically may have in maximising their performance in achieving their goals and objectives. In this regard, Cao and Zhang (2011:168) argue that the higher the level of synergy within the supply chain, the more likely suppliers’ performance will be improved. Scannell, Vickery and Droge (2000:23) and Cannon and Homburg (2001:29) posit that adequate synergy within a firm's supply chain results in better coordination and collaboration as well as a decrease in procurement costs. In view of the foregoing, it would appear that suppliers’ willingness to work in a synergistic manner is crucial in enabling them to improve their performance, which subsequently may allow firms to attain and sustain their competitive advantage. Therefore, the following hypothesis is formulated:

The sample consisted of SME companies in the Vaal Triangle (Vanderbijlpark, Vereeniging and Sasolburg). These were selected using the simple random technique, which allows for statistical inferences to be made about the target population from which the sample is drawn (Malhotra & Birks 2007:1). Questionnaires were distributed in person by one of the researchers during the month of August 2014 with the assistance of a trained fieldworker in the distribution process to ensure a speedy and effective process. Respondents were given two weeks to complete the questionnaires. Out of the 500 questionnaires that were initially distributed, 350 were retrieved. Of these, only 309 were properly completed by respondents and therefor deemed usable for analysis, resulting in total of 41 questionnaires discarded because they were either incomplete or inaccurately completed. This yielded a response rate of approximately 61.8%. The sampling frame was made up of various lists that included a register from the Gauteng Enterprise Propeller (GEP), the Vaal Triangle Business Directory as well as SME databases from relevant Sedibeng district municipality. The ethical clearance issued by the Vaal University of Technology enabled the field workers to approach prospective respondents who were informed that their participation in the study was voluntary and that the confidentiality of the information provided by the respondents was guaranteed. The questionnaire was completed by owners and managers of the participating SMEs.

The profile of SMEs that participated in the study is reported in Table 1.

Table 1 shows that the majority of participating SMEs were registered as private companies (34.3%; n = 106), followed by sole proprietors (24.6%; n = 76). The next most common type of ownership was partnerships at 19.7% (n = 61). Nearly 39% (n = 120) of respondents operated in the retail environment whilst 14.2% (n = 44) were in manufacturing and 11% (n = 34) operated in the transport industry. A total of 57% (n = 175) of the participating SMEs employed fewer than 100 people, whilst 21% (n = 65) employed between 101 and 200 people and 22% (n = 69) employed between 201 and 500 people. This distribution shows that most jobs are created by SMEs rather than large businesses. It is also notable that a combined 52% (n = 161) of participating SMEs were fairly new businesses, having been in existence for five years or less. This also shows the kind of resilience some businesses still have despite the difficult economic times experienced in recent years.

In this study, a two-step process was followed, which entails conducting CFA prior to testing the hypotheses, as recommended by Anderson and Gerbing (1988:415).

The purpose of the CFA procedure was to examine scale accuracy in terms of internal consistency (reliability), various types of validity (construct, convergent and discriminant) and the model fit of the multiple-item construct measures used in the study (Bollen 1990:258). The results of the CFA are reported in Table 2.

As shown in Table 2, reliability was assessed through the Cronbach's alpha coefficient. The Cronbach's alpha column demonstrates that all the constructs are between 0.950 and 0.966. This suggests that all the constructs were reliable as their values were above the recommended threshold of 0.7 (Bagozzi & Yi 1988:74; Nunnally & Bernstein 1994:34). Furthermore, the internal reliability of each construct was also evaluated using the composite reliability (CR) index test. A CR index value equal to or higher than 0.7 illustrates the adequacy of internal consistency of the construct (Hair et al. 2006:25; Nunnally 1978:4). The analysis shown in Table 2 indicates that the CR indexes for all the constructs were well above the requisite 0.7, thus confirming satisfactory internal consistency of all the constructs.

Convergent validity was measured by assessing whether the individual item loading for each construct was above the recommended threshold of 0.5 (Anderson & Gerbing 1988:417). As revealed in Table 2, the measurement instruments assessing each construct were well above the recommended 0.5: the information sharing construct (IS) had loadings ranging from 0.81 to 0.88, supplier trust (ST) ranged from 0.9 to 0.93, supplier synergy (SSY) ranged from 0.94 to 0.93 and supplier performance (SP) ranged from 0.91 to 0.88. All of the loadings were beyond the required threshold of 0.5, which indicates that the respective constructs converge well and were therefore valid. Discriminant validity was ascertained using Pearson correlation coefficient and the results indicate positive correlations with values of r ranging between 0.102 (p < 0.01) and 0.860 (p < 0.01). Since these inter-factor correlation values for all paired latent variables were below the recommended maximum threshold of 1.0, there was satisfactory discriminant validity in the measurement scales.

The acceptability of the model fit was established by chi-square value over degree of freedom (χ²/df), whose value should be between 1 and 3 (Schreiber et al. 2006:330), whilst the values of goodness-of-fit index (GFI), comparative fit index (CFI), incremental fit index (IFI) and Tucker-Lewis index (TLI) should be equal to or higher than 0.90 (Chinomona 2012; Hu & Bentler 1995:76) and the root mean square error of approximation (RMSEA) value should be equal to or less than 0.08 (Browne & Cudeck 1993:137). The results of the model fit assessment showed that χ²/df was 2.864 (χ²/df = 670.126/234) and GFI, CFI, IFI, NFI and RMSEA were 0.932, 0.967, 0.967, 0.951 and 0.078 respectively. In view of the aforementioned results, it can be observed that all the indicators values meet the above stated thresholds. Therefore, the collected data were able to fit the model.

The measurement of model fit of this study was done using the following indices: χ²/df, GFI, CFI, IFI, NFI and RMSEA. With regard to the χ²/df, the value was below the required threshold of 3 (χ²/df = 600.210/234 = 2.565). Furthermore, GFI, CFI, IFI, NFI and RMSEA provided respective ratios of 0.91, 0.95, 0.94, 0.911 and 0.07. These statistics depict that all the indicators were above the acceptable threshold of 0.9 for GFI, CFI, IFI, NFI and below that of 0.08 for RMSEA. This implies that there was acceptable model fit in the study. The results of hypotheses tests are reported in Table 3.

As shown in Table 3, the coefficient values for all hypotheses were significant at a level of p less than 0.01, with the exception of H₅ which was insignificant. Thus three of the hypotheses (H₁, H₂, H₄) were accepted whilst one hypothesis (H₃) was rejected.