The flat glass industry in Taiwan

Taiwan's glass industry offers a broad and diverse range of products. It can be divided into flat glass and glass products based on their use, among which flat glass is considered to be mass produced and to be the most closely related to our lives. Hence, this study employs flat glass for exploration and investigation, as it includes float glass, patterned glass, lined-in and netted glass, heat reflective glass, curved glass, tempered glass, heat-strengthened glass, laminated glass, insulating glass, and burned porcelain paint glass.

In order to meet the needs of the international architecture, processing, and automobile industries, the flat glass industry focuses on continuous construction and equipment expansion; however, the profit has shrunk severely because supply exceeds demand. Thus, competition in the market depends on the quality of the product and the service. In recent years, glass manufacturers in Taiwan have spared no effort in working on research and development as well as improvement, placing importance on ‘green’ building by producing low-emissivity glass containing such characteristics as high perspective and high temperature insulation. This approach is meant to produce a power-saving, cost-saving, and eco-friendly product, based on the concepts of high-quality life and environmental protection, which have gradually gained affirmation from both the government and consumers. The glass industry also intends to move toward diversification of business, introducing such products as glass fiberboard, which is used in high-tech products and which has greatly added value and spatial margins to industry profitability. Especially after the signing of the ECFA, glass manufacturers need to continue to focus on research and development as well as innovation in response to the productivity of mainland China.

The production of flat glass in Taiwan is based solely on local glass and on the supply and demand in the domestic market. In addition, the price rests mostly in the hands of Taiwanese glass companies. The upstream and downstream vendors in the glass industry have been working with each other for a long time, resulting in a unique relationship that is both competitive and cooperative, and the glass industry in Taiwan has spent a long time cultivating these cordial relationships. Therefore, in terms of competing with China, it is very likely to maintain its advantages in the international market when faced with challenges after the signing of ECFA. The Taiwan Mirror Glass Enterprise Ltd., in particular, by means of promoting industry clustering and good management, has led the way in deepening the professional level and expanding the business scope to form a specialized economy and generate a high-quality, low-cost product with a high degree of flexibility. Even mainland Chinese vendors wish to imitate its success; however, it would take them a long time to catch up. In addition, since 2005 the quality of glass products manufactured in China has been suspected and questioned by many EU countries; thus, relatively speaking, Taiwanese products are highly regarded, and orders have increased significantly.

It is anticipated, however, that the signing of the ECFA will have several effects on the flat glass industry in Taiwan. There is one advantage in that glass products exported to mainland China from Taiwan may enjoy the advantage of zero tariffs, which will mean that the products can be competitive in the mainland Chinese market. However, more disadvantages are anticipated than advantages. First, the terms of the ECFA require that Taiwan will need to be open to many industrial products from mainland China that have been restricted from import into Taiwan (including glass products), and these products will enjoy zero tariffs as well; in this case, it is highly likely that the uneven and low-price glass products imported from China will erode the domestic market, which is not a propitious situation for domestic glass manufacturers. In order to maintain a competitive edge, domestic flat glass manufacturers must be able to meet the following requirements:

1. (1) They must enhance the quality of glass products, reduce production costs, create innovative research designs, produce glass products with high output values, and differentiate their goods from the cheap products made in the mainland China.

2. (2) They must establish their own brands and participate in international fairs to facilitate the promotion of the export market. The export market may be able to blunt the impact on the import market caused by the ECFA.

3. (3) They must enhance their local enterprise advantage and retain their existing domestic customers through better services, excellent quality, and rapid shipping.

With regard to the present study, most previous studies focused on network-ties-related issues, involving discussions on the transmission and sharing of knowledge and teamwork in knowledge management (Zhong, Huang, Davison, Yang, & Chen, Reference Zhong, Huang, Davison, Yang and Chen2012; Peng, Wan, & Woodlock, Reference Peng, Wan and Woodlock2013) or the strength of ties (Petróczi, Nepusz, & Bazsó, Reference Petróczi, Nepusz and Bazsó2007). Agility-related studies mostly focus on how agility in supply chain management affects efficiency, performance, or performance output, most of which directly involve the effect and correlation between ‘network ties and performance’ or ‘agility and performance’ (Sheng, Zhou, & Li, Reference Sheng, Zhou and Li2011; Gligor & Holcomb, Reference Gligor and Holcomb2012). However, in the dynamic environment, the network agility in the organizational structure has become increasingly important (Lin, Desouza, & Roy, Reference Lin, Desouza and Roy2010). There is scarce literature on discussions of the direct or indirect effects of organizational agility that may exist between the network ties and organizational performance. Therefore, the unique situation of the Taiwanese glass industry – with its already-established cooperative relationships, yet faced with a precarious situation that requires some organization changes – makes it a good candidate for the purpose of this study: to examine the impact of network ties on organizational performance. By means of a literature research and a questionnaire survey, this study explores network business application-related subjects, strategies, and procedures – from the establishment of network organizational structure and organizational agility, to organizational performance – in order to discuss strategies that can be adopted by the organization in the network system. The purpose is to develop the correlative subjects that can help enhance organizational agility from the network organizational environment, and further to analyze them accordingly. In the context of Taiwan's flat glass industry, this study will research the relationships between network ties and organizational agility, and organizational performance. This study will also establish a set of indexes or factors to measure network ties and agility among members in a network organization by means of the literature review. We will also discuss the strategies businesses use to enhance each other's agility among organizations by an actual survey of the industry. Moreover, based on the above modeling framework, the relationship of these aspects to organizational performance is studied so that this research may contribute to network cooperation among organizations.

Network ties

Sociology was the first to observe the network ties that cover the phenomena of linking, interaction, consultation, and information exchange among people. As sociologists began to study and analyze network ties, such fields as economics and organizational theory also began to notice and initiate research on the network phenomenon. The analysis of network ties is divided into three major types: the tie among individuals, the tie between individuals and organizations, and the tie among organizations.

In this study, the third level will be discussed, the network tie among organizations. As indicated by Johanson and Mattsson (Reference Johanson and Mattsson1987), based on the hypothesis that external corporate resources must be obtained through the industrial network, the network is regarded as the behavior of interactive relationships among firms. Concerning an individual firm's position and its strategic implications, this study proposes five attributes of networks as follows: (1) A network is formed by the relationship of a group of firms that are interdependent on one another. (2) The relationship implies specialization, which is coordinated through the interactive behavior in the network. (3) The interaction consists of two types: resource exchange and the strategic action of adaptation. (4) The function of exchange is to obtain corporate external resources, sell corporate products and services, and facilitate the technological knowledge and information diffusion, while adaptation is to deal with the inconsistency of network ties and to make them endure. (5) In the network, the relationships of complementation and competition exist simultaneously. Formal and informal coordinated mechanisms are also considered.

Having summarized the aforementioned concepts, this study attempts to measure the following: the ‘betweenness centrality’ proposed by Freeman (Reference Freeman1979): this is a measure of the shortest paths between node pairs that connect through a node representing the most power or importance. The ‘density,’ as introduced by Tichy, Tushman, and Fombrun (Reference Tichy, Tushman and Fombrun1979), is the ratio of the number of actual relational ties compared with the number of possible relational ties. ‘Intensity’ is the intensity of network ties or contact numbers within unit intervals of network members. Finally, ‘reciprocity’ relates to whether the intensities of the relationships are equal to one another among network members.

Organizational agility

Compared with the past, the strategy and procedure of a world-class firm must be fast, fluent, and flexible. Achieving these targets requires initiating the potentiality of all new technologies and new concepts by implementing innovative management philosophy and administration. The organization with agility is not only highly decentralized, but also more agile, dynamic, diversified, and thoughtful in structure (Barker, Reference Barker1995). Youssef (Reference Youssef1992) argued that the ability of an agile organization to respond quickly to competitive threats and market opportunities is related to time-based competitiveness. Gunasekaran (Reference Gunasekaran1999), however, proposed a composite argument from the aforesaid two scholars and defined agility as a firm's capability to respond rapidly and effectively to the market and to make a profit in a competitive environment that is ever-changing and unpredictable. It can also be considered as a firm's survivability. Chengalur-Smith, Duchessi, and Gil-Garcia (Reference Chengalur-Smith, Duchessi and Gil-Garcia2012) proposed that when a large company initiates information sharing, other companies can follow the policy. As a result, rapid response is the key to becoming an agile organization in an environment that requires flexibility and speed, as well as being able to provide new customer services promptly. Hence, it becomes an index whereby to judge an organization's agility in terms of its capability to grasp the sudden opportunity. Therefore, agility will influence the organization's operating strategy and the way its business is operated (Corstjens & Lal, Reference Corstjens and Lal2000). Agility refers to quick responses to customers, minimization of manufacturing lead time, and speed of delivery (Jacobs, Droge, Vickery, & Calantone, Reference Jacobs, Droge, Vickery and Calantone2011). In order to achieve true agility, supplies partners must make good use of existing equipment and promote efficiency in order to act faster (Pearson, Masson, & Swain, Reference Pearson, Masson and Swain2010; Costantino, Dotoli, Falagario, Fanti, & Mangini, Reference Costantino, Dotoli, Falagario, Fanti and Mangini2012; Mirzapour Al-e-Hashem, Malekly, & Aryanezhad, Reference Mirzapour Al-e-Hashem, Malekly and Aryanezhad2012). Numerous existing literatures advocate that the achievement and maintenance of agility are the key elements to sustain corporate competitiveness in the dynamic and uncertain environment (Yusuf, Musa, Dauda, El-Berishy, Kovvuri, & Abubakar, Reference Yusuf, Musa, Dauda, El-Berishy, Kovvuri and Abubakar2013).

This study reviews and summarizes the literature related to agility and further classifies agility into three types of correlative capabilities as follows: customer agility, partnership agility, and competitor agility (Treacy & Wiersema, Reference Treacy and Wiersema1993; Cronin, Reference Cronin2000; Tapscott, Lowy, & Ticoll, Reference Tapscott, Lowy and Ticoll2000). Customer agility means including customers in the activities of exploration, innovation, and competitive opportunity. It represents a firm's capability of making choices between obtaining market intelligence from customer feedback and detecting opportunity in the competitive activity. Partnership agility, through alliance, partnership, and joint venture, means balancing the assets, knowledge, and capabilities of suppliers, distributors, contracted manufacturers, and logistics. Competitor agility is a firm's capability of understanding and reacting to information from competitors. It enables the firm to use opportunities in competitive activity more rapidly and precisely, and to achieve cost-efficiency. The above three dimensions together interpret agility.

Organizational performance

Day (Reference Day1984) argued that organizational capability must include the ability to learn and understand customers’ requirements and needs, to create new products and services for meeting their requirements, and to further enhance the organizational performance. Provan and Sebastian (Reference Provan and Sebastian1998) summarized the past research on performance and suggested that performance is the most commonly used standard for evaluating an organization. In the field of organizational theory, performance broadly implies the result of three aspects: efficiency, effectiveness, and efficacy. Performance evaluation not only enables a firm to review its resources and see if they are utilized effectively, but also becomes a valuable reference for a firm's decision making. Venkatraman and Ramanujam (Reference Venkatraman and Ramanujam1986), however, argued that performance consists of three dimensions: (1) ‘financial performance,’ which uses the sales growth rate and profitability as a measurement index; (2) ‘business performance,’ including market share and product quality in addition to financial performance, and (3) ‘organizational effectiveness,’ which includes employee morale and customer satisfaction.

Yusuf, Gunasekaran, Adeleye, and Sivayoganathan (Reference Yusuf, Gunasekaran, Adeleye and Sivayoganathan2004) argued that agile manufacturing has an impact on manufacturing performance, including great flexibility; low cost; and fast, reliable, and consistent quality. In particular, the authors indicated that the approach of agile manufacturing has an apparent impact on flexibility. In order to measure performance effectively, there must be a standard of measuring it. Although performance indexes differ according to organizational purpose and needs, they must meet requirements that are reliable, comprehensible, and usable.

Farrell and Buckley (Reference Farrell and Buckley1999) stated in research on the overall performance model of network organization that there are five criteria for measuring the overall performance of network organization: (1) the capability of accomplishing tasks with limited time and resources, (2) greater efficiency than that in traditional organizations, (3) finished products of higher quality than the ones manufactured by traditional organizations, (4) adaptation of sensitivity and flexibility to environmental changes, and (5) achievement of the target of overall development and the demand on resources in the network. The five measured items, as stated above, imply respectively that such organizations have the capability of rapid response, task accomplishment, quality assurance, active reaction and adaptation, and profit growth and market share.

This study summarizes the views of the aforesaid scholars and further proposes the indexes to measure, through agility activities, whether a firm can attain the capabilities of enhancing its response to market changes, manufacturing high-quality products, providing excellent customer services, responding to market changes, shortening the time of satisfying customer needs, and increasing corporate profits.

Network ties and organizational performance

The establishment of network ties will enable a firm to reduce trading costs, decrease environmental uncertainty, and obtain the operation resources needed, including manpower, capital, operation knowledge, and information technology. It will enhance a firm's relative competitiveness, as well as the organization's operational performance. Short and Venkatraman (Reference Short and Venkatraman1992) presented the concept of replacing business process reengineering by network tie redesign through which a firm can rebuild relationships among businesses and enhance overall organizational performance. Hagedoorn and Schakenraad (Reference Hagedoorn and Schakenraad1994) analyzed the impact of strategic alliance numbers and network interactive relationships on economic performance in IT, machinery, and petrochemical companies in Europe, the United States, and Japan. They found that building strategic networks can enhance economic performance in those organizations. Powell (Reference Powell1990) developed organizational learning through the concept of business networks and pointed out that alliance numbers, degree of network centralization, and network experience will enhance organizational learning speed. Blankenburg Holm, Eriksson, and Johanson (Reference Blankenburg Holm, Eriksson and Johanson1999) considered that establishing a firm's network will help to create value because of the mutual trust and commitments among members.

We may infer from the research summarized above that network ties should affect organizational performance; thus, Hypothesis 1 may be established, as follows:

Network organization and organizational agility

The reason network organization emerges is that it possesses the attributes of flexibility and agility in response to rapid, dynamic, and diverse market needs. Networks are formed for numerous reasons, and diverse arguments have been proposed by many scholars. Miles and Snow (Reference Miles and Snow1995) suggested that networks evolve from organizations and present a cooperative relationship according to environmental change. Through cooperation with one another, organizational members can reduce trading costs, attain dependence and reciprocity on resources, decrease environmental uncertainty, and further enhance their own competitive advantages. Information technology usually accompanies network organization in which interactive and complementary relationships exist. To deal with changeable and dynamic operations and competitive patterns, more and more firms have begun to utilize the information technology for integrating the resources of material purchase, production, logistics, delivery, and sales. Thus, it can generate the effects of responding rapidly to market needs, reducing operational cost, and enhancing operational efficiency.

Organizations that collaborate through network ties can achieve common interests through agility, cooperation, and knowledge sharing, which is a must, considering today's uncertain environment (Lin, Desouza, & Roy, Reference Lin, Desouza and Roy2010). With regard to communication between enterprises, ties provide advents such as learning, sharing, and transmission of internal information and engaging in resource allocation in an unfamiliar market (Li & Zhou, Reference Li and Zhou2010). Chung (Reference Chung2012) found that as the level of network ties between enterprises increases, the effectiveness of aspects such as market intelligence and responsiveness is enhanced. Earlier studies also pointed out that enterprises’ networks are important indicators that determine their success (Boso, Story, & Cadogan, Reference Boso, Story and Cadogan2013), as these networks provide the following benefits: increased sharing of resources and market intelligence, improved cooperation in logistical efforts, reduced transaction costs, and opportunistic behaviors of partners (Park & Luo, Reference Park and Luo2001; Luo, Hsu, & Liu, Reference Luo, Hsu and Liu2008; Boso, Story, & Cadogan, Reference Boso, Story and Cadogan2013). Yusuf et al. (Reference Yusuf, Musa, Dauda, El-Berishy, Kovvuri and Abubakar2013) reported that clustering-oriented organizations can enhance the overall organizational agility and promote the implementation of agility in their respective clusters.

Having summarized the above scholars’ research, we may infer that network ties should have an impact on organizational performance. Thus, by inference, Hypothesis 2 can be established as follows:

Organizational agility and organizational performance

What agility emphasizes is responding to uncertain environmental changes rapidly, agilely, and precisely so that an organization is able actively to adapt itself to the environment. The agility concept nowadays is broadly utilized in business. A firm should realize that agility is the basis of maintaining survival and competitiveness (Danuta & Swierczek, Reference Danuta and Swierczek2009). Therefore, a firm must acquire a deeper understanding of the elements, attributes, capabilities, and implementation stages that comprise agility. Goldman, Nagel, and Preiss (Reference Goldman, Nagel and Preiss1995) indicated that the elements of an agile organization must include customer support and dialogue, continuous improvements and changes, employee support, and the flexibility for a rapid-response process. Sharifi and Zhang (Reference Sharifi and Zhang1999) stated that the implementation pattern of agility consists of three diverse elements, as follows: (1) Agility drivers – that is, a firm must seek new approaches to maintain corporate competitive advantage in response to environmental changes or survival pressure. (2) Agility capabilities – the required abilities possessed by a firm for maintaining profitability. (3) Agility providers – the places in which agile capabilities exist or the providers offering agile capabilities. For a firm, agility is a required ability for earning profits in business operation under the competitive environment with sustainability, uncertainty, and continuous change in terms of customer opportunity (Goldman, Nagel, & Preiss, Reference Goldman, Nagel and Preiss1995). According to Sambamurthy, Bharadwaj, and Grover (Reference Sambamurthy, Bharadwaj and Grover2003), agility can contribute to organizational performance in regard to three aspects. First, a firm with agility can enhance customer satisfaction and loyalty through rapid response to changes in customer needs. In addition, agility can be utilized through the company's embedded influence on customers and the turning points generated by frequently grasping market opportunities (Kidd, Reference Kidd.1994; Goldman, Nagel, & Preiss, Reference Goldman, Nagel and Preiss1995). Second, through establishing an extended partnership of networks based on impact, it can make use of the assets, knowledge, and capabilities of suppliers, agents, contract manufacturers, and logistics in enhancing organizational performance (Venkatraman & Henderson, Reference Venkatraman and Henderson1998). Lastly, the impact of agility on organizational performance is also implemented quickly in redesigning and simplifying the organizational process using speed, preciseness, and economic cost (Teece, Pisano, & Shuen, Reference Teece, Pisano and Shuen1997).

Having summarized the above scholars’ research, we may infer that network ties and organizational agility should have an impact on organizational performance. From this inference, then, we may establish Hypotheses 3, as follows:

Intermediary effects in organizational agility

Based on the results of Yusuf et al. (Reference Yusuf, Musa, Dauda, El-Berishy, Kovvuri and Abubakar2013), it was found that the stronger and more obvious the industrial cluster, the better is the agility of external environment changes reflected and customer satisfaction achieved, which in turn will affect the overall competitiveness and organizational performance. In the competitive industry, senior generals establish effective ties to obtain benefits from internal information exchanges in larger industries with market intelligence and effectively lock on market demands and objectives, all of which will contribute to the success of the enterprise (Boso, Story, & Cadogan, Reference Boso, Story and Cadogan2013). Due to the geographical locations and industrial characteristics of Taiwan's floating glass industry, industrial clusters of floating glass manufacturers are available in Zhangbing Industrial One, and this feature has contributed to the special network ties and collaborative relationships.

The research theories proposed by the abovementioned scholars were summarized and inferences were derived from them. A certain kind of relationship was inferred to exist between organizational agility and network ties and organizational performance. This inference led to the establishment of Hypothesis 4, as follows:

Research framework

This study will examine organizational agility and network ties among organizations in relation to the organizational performance in the network system. In an attempt to develop correlative question items, we analyze how to further enhance a firm's operational performance through maintaining proper network climate and reinforcing relationships among suppliers. Hence, based on the research purpose and motivation along with the theoretical literature, the study is approached from the view of network ties: one is the direct influence of the three aspects of network ties on organizational agility, of network ties on organizational performance, and of organizational agility on organizational performance. Moreover, we construct a linear structural equation model using network ties as the antecedent variable, organizational agility as the intervening variable, and organizational performance as the outcome variable to examine the cause and effect of these important variables. The research framework is shown in Figure 1.

Figure 1 Research framework

Questionnaire design and sampling method

The design of the items in the questionnaire is divided into four parts: (1) basic information concerning the subjects’ companies, including such continuous and discontinuous variables as company types, capital of the establishment, years of establishment, numbers of employees, average yearly sales turnover, etc.; (2) scale of network ties; (3) scale of organizational agility; and (4) scale of organizational performance. The last three parts consist of 55 questions with a 5-point Likert scale as the measurement. The subjects all are the owners or general managers of companies in the flat glass industry as registered at the Industrial Development Bureau, Ministry of Economic Affairs, chosen in order to ensure that the subjects have sufficient experience and vision to answer the organizational level questions in this study.

Methods of sampling and questionnaire collection are as follows: (1) Paper questionnaire: the researcher personally visited the subjects’ companies and had them fill out the questionnaires, of which 136 were retrieved; (2) Internet online questionnaire: had the subjects recommend other subjects in the same industry, or the researcher contacted the companies registered at the Industrial Development Bureau of Ministry of Economic Affairs and sent out the questionnaires to subjects after phone confirmation; of these, 180 were retrieved. The total questionnaires retrieved were 316, and from these, 250 valid questionnaires emerged after invalid samples with such problems as incomplete answers and omitted information were discarded. The valid sample rate is 80%.

Method of information analysis

For this study, a scale questionnaire survey was adopted, and SPSS18 used for the reliability analysis, correlation analysis, and descriptive statistics. Further, a structural equation analysis measurement model and theoretical framework were constructed using AMOS18.0. In the information analysis procedure, the suggestion of Anderson and Gerbing (Reference Anderson and Gerbing1988) was followed in the adoption of a two-phase analysis. In the first phase, the measurement model of variables was evaluated for reliability analysis, validity analysis, and confirmatory factor analysis to understand the various dimensions of reliability and validity (convergent and discriminant). In the second phase, a theoretical framework was evaluated, and each hypothesis was validated.

Descriptive statistics analysis

The valid questionnaires of this study number 250. The company types, establishment capital, years of establishment, number of employees, and sales turnover are shown in Table 1.

Table 1 Demographics of sample company

Structural equation modeling

Validity analysis

For validity assessment, the suitability of each construct measurement model was assessed using the maximum likelihood estimation of confirmatory factor analysis in order to evaluate if each construct has sufficient convergent and discriminant validity.

Convergent validity analysis: According to the convergent validity analysis criteria suggested by Anderson and Gerbing (Reference Anderson and Gerbing1988), the evaluation criteria of confirmatory factor analysis proposed by Bagozzi and Yi (Reference Bagozzi and Yi1988) and the goodness of fit index presented by Gefen, Straub, and Boudreau (Reference Gefen, Straub and Boudreau2000), an assessment was conducted with the suggested data with seven resulting evaluation criteria, as follows: (1) goodness of fit index (GFI), normed fit index (NFI), and comparative fit index (CFI) are >0.9, while the root mean square residual is <0.05; (2) the factor loadings of each index are significant; (3) the composite reliability of each construct is >0.7; and (4) the average variance extracted is >0.5. Table 2 shows the results of the convergent validity analysis for (a) the scale of network ties, (b) the scale of organizational agility, and (c) the scale of organizational performance. The analyzed data reveal that the convergent validity for all of these scales is within the acceptable range.

Table 2 Assessment of convergent validity

Note. AVE = average variance extracted.

Discriminant validity analysis: This study was based on the two criteria of examining discriminant validity proposed by Gaski and Nevin (Reference Gaski and Nevin1985): (1) the correlation coefficient of two constructs is <1; and (2) the correlation coefficient of two constructs is less than the individual Cronbach's α reliability coefficient, implying that the two constructs have discriminant validity. Besides, based on the approach of Fornell and Larcker (Reference Fornell and Larcker1981), this study presents the third criterion of examining discriminant validity. (3) The correlation coefficient of two constructs is less than average variance extracted's square root, implying that the two constructs have discriminant validity. The analysis data summarized in Table 3 all meet the above three criteria for examining discriminant validity, revealing good discriminant validity in each construct.

Table 3 Assessment of discriminant validity

Note. The diagonal values are the square roots of the AVE for each construct.

ACO = agility-competitor; ACU = agility-customer; AS = agility-supplier; AVE = average variance extracted; BCNT = betweenness centrality of network ties; DNT = density of network ties; INT = intensity of network ties; OP = organizational performance; RNT = reciprocity of network ties.

Overall structural equation modeling analysis

Analysis results

The analysis results are presented in two parts, including the evaluation of the theoretical model and the verification of hypotheses relationships, which are discussed successively.

Evaluation of theoretical model: The evaluation of model fit, based on Bagozzi and Yi's (Reference Bagozzi and Yi1988) suggestion, is assessed from three perspectives: (1) preliminary fit criteria, (2) overall model fit, and (3) fit of internal structure of mode, as follows:

(a) Preliminary fit criteria: The error variances of the theoretical model do not show negative values, and the standardized factor loading is not <0.5 or >1; in addition, all reach a significant level. Nor is a higher standard error found. Thus, the preliminary fit criteria of this study model overall reach an acceptable level.

(b) Overall model fit: The overall model fit is used to evaluate the entire model and to observe the fit level of information. The suggestion of Hair, Black, Babin, Anderson, and Tatham (Reference Hair, Black, Babin, Anderson and Tatham2006) was adopted, and three indexes were chosen for evaluating the entire model and observing the fit level of information: (1) absolute fit measures, (2) incremental fit measures, and (3) parsimonious fit measures. The absolute fit measures of the overall theoretical model in this study are χ² = 330.603, df = 74, χ²/df = 4.468, GFI = 0.838, RMR = 0.062, RMSEA = 0.118, and AGFI = 0.770. Except for RMSEA, the other indexes are all within the standard. The incremental fit measures of the overall theoretical model in this study are NFI = 0.897 and CFI = 0.917, and both are within acceptable range. The parsimonious fit measures of the overall theoretical model in this study are PNFI = 0.729 and PGFI = 0.590, and both are within acceptable range (>0.500). As a whole, except χ², judging from each index, the overall model fit of the theoretical model in this study is passable.

(c) Fit of internal structure of mode: The fit of the internal structure of mode is to estimate the significant level of parameters and the reliability of each index and latent variable in the evaluation model. The composite reliability coefficients of each construct, ranging from 0.943 to 0.963, all exceed the acceptable level of 0.7. The average variance extracted the coefficients of each construct, which, ranging from 0.707 to 0.867, all exceed the acceptable level of 0.5. As a whole, the fit of the internal structure of mode in this study is good.

Verification of hypotheses relationship: After confirming that each construct has a certain degree of validity and reliability, the next step is to examine the research hypotheses established by the research model and to analyze how the variables of ‘network ties,’ ‘organizational agility,’ and others collectively affect ‘organizational performance.’ Based on Figure 2, the regression coefficient of overall path analysis is used to examine the hypotheses. The analysis results support the four hypotheses in this study, as p-values all reach the significant level of .05. The findings are as follows:

Figure 2 Diagram of the path analysis results of structural modeling

1. (1) Concerning hypotheses 1 and 2, the reason network ties affect organizational agility has to do with the attributes of flexibility and agility, as network ties help an organization adapt to rapid, dynamic, and diverse market needs. From the view of the aforementioned transaction cost, transaction behavior occurs along with, among other things, the costs of information search, condition negotiation, and transaction proceedings. The more difficult the transaction procedure, the higher the transaction cost. However, establishing efficient network ties (density, intensity, reciprocity, and betweenness centrality) can increase the return rate of assets, lower unit cost, reduce waste or lower unit fixed cost, enlarge the investment/production rate, improve the organization's internal efficiency, and further enhance organizational performance through organizational agility.

2. (2) The findings from hypothesis 3 show that agility emphasizes responding rapidly, agilely, and precisely to an uncertain environment. Further, an agile organization can proactively adapt itself to the environment. Under the competitive environment of sustainability, uncertainty, and continuous change in obtaining customer opportunity, agility is the required capability for sustaining business profits and enhancing operational performance.

3. (3) In Hypothesis 4, it was found that establishing network ties will enable a firm to reduce transaction costs, decrease environmental uncertainty, and obtain required operational resources – including manpower, capital, business knowledge, and information technology. Thus, not only is the firm's relative competitiveness reinforced, but also its operational performance.

Conclusion of research hypotheses verification

This study has found that through the intermediary of organizational agility, a firm can enhance organizational performance. This finding also indicates that customer-orientation has been a trend in the market for years. Because organizational changes are not accomplished in a short time, traditional industries, as they transform gradually, also need to face the challenge of quick response to a dynamic environment for survival. The aforementioned results imply that although network ties and organizational agility are both able to enhance organizational performance and though a firm may have the ability to fulfill all specifications of customer orders, the repeat order rate from customers may also be relatively low because of price competitiveness. The supply chain partnership is highly influenced by price competitiveness. However, the firms’ agility toward suppliers reflects that firms find it relatively difficult to collect information from suppliers. Thus, Taiwan's flat glass manufacturers need not only tighter and more active shares in network ties, but also supply chain integration, which is inevitable if a firm is to achieve synergy.

Research contribution and management implications

In the face of global competition, turbulence, and uncertainty, today's managers need to deal with in increasingly intense, complex, and paradoxical competitive environment through dynamic thinking and decision making in order to maintain a competitive edge. Taiwan's industrial background started with a foundry base. At the time, whenever production capacity was inadequate, orders were first accepted and then distributed to other companies in the same industry to ensure sufficient financial resources for survival and to avoid the loss of orders, and thus a unique competitive and cooperative relationship was formed. However, as times changed, the various manufacturers developed a variety of competitive advantages, and firms underwent phase-out and integration.

Because of the rise of Mainland China in recent years, the disappearing subtly competitive relationship has become an issue in the maintenance of overall competitive advantage once again, thus affecting not only the industry but also the national competitiveness. This issue is most urgently faced in Taiwan, in which the industrial structure is based on small to medium enterprises. Through network ties, an industry's overall competitive advantage can be maintained, thereby enhancing organizational agility and performance within the industry.

This study makes both academic and practical contributions, the latter being related to managerial implications. In terms of the academic aspect, this study explored the network ties, organizational agility, organizational performance, and other theories and dimensions through the literature. In addition, from a network organization perspective, a framework was constructed for exploring the relationship between network ties and organizational performance, with organizational agility as the intervening variable. The abovementioned verification results can be understood through this research framework, and the results may serve as a reference for related follow-up research. The research results may also serve as a research direction and reference target for the government as they assess and analyze the impact of ECFA on Taiwan's industries.

Regarding the practical aspect, in order for the government or traditional industry's managers to improve organizational agility and enhance the referential value and adoption of organizational performance through Taiwan's network ties among the industries after signing ECFA, three important managerial implications may be derived from this study:

First, the cooperative relationship in a supply chain is similar to that of gangs. In the process of close integration, market price and the organization's ability to respond to the dynamic market are put to the severest test. According to the competitiveness analysis results of this study, Taiwan's flat glass industry has relatively high entry barriers, and competition among industries is fierce. Because of consumers’ diverse needs, buyers’ bargaining power is high. However, subject to the sources of glass, suppliers’ bargaining power is relatively lower, but glass has better quality and attributes than other alternatives. Hence, threats from alternatives were relatively low. As mentioned above, under the environmental characteristics of complexity, versatility, and uncertainty, every kind of resource had other uses in the huge business opportunities. Additionally, a network system among organizations can provide the resources to members in need through mutual reciprocity or resource interaction. This will not only resolve the problems of partners among organizations, but also will more effectively convert idle resources into what enterprises need, thus preventing resource idling or freewheeling. In this way, enterprise costs can be reduced for deriving a direct competitive advantage or indirectly enhancing the overall competitiveness of the system. The reactivity toward market changes will also be greatly improved through the network organization's provision of information, which will be more conducive to helping the members gain the advantage of quick response to the market.

Second, the signing of the ECFA between the two sides across the strait will exclude a variety of adverse obstacles to exports, coupled with China's non-nominal tax items (of which there are 27 kinds), the yearly rise in labor wages, the increasingly serious lack of manpower in coastal areas, and the difficulty involved in cost control. Taiwan's labor is high in quality and active. Under the cross-strait industrial division of labor, such high-priced products as fiber glass and glass fabric will be produced in Taiwan. Because in this study, agility has been shown to enhance organizational performance, Taiwan's flat glass industry should take Taiwan Glass Ind. Corp's ability to integrate and their R&D capabilities as a guide. In the face of the new situation after the signing of ECFA, the development focus of the flat glass industry should lie in aggressive investment and expansion plans, including the production of solar photovoltaic glass, Double Low-E coated glass, and the development of transparent conductive oxide of optical glass, so as to enhance organizational agility and meet customer demand, thereby improving the organizational performance.

Third, in terms of strategic alliance, the most successful example is Taiwan Mirror Glass Enterprise Ltd., which spent 10 years in gathering 15 glass companies by 2005 to set up plants in the Chang-Ping Industry Area, thus creating Taiwan's glass industry team – the Team Taiwan Glass brand. The companies jointly accepted orders and engaged in industrial division of labor. For 10 consecutive years, orders placed in Taiwan have made up 80% of IKEA's glass furniture inventory. Currently, more than 200 glass small to medium enterprises have joined the team, and for this reason, Taiwan Glass Ind. Corp., the main supplier of glass raw materials, returned to Taiwan to set up its factory after 15 years of industry in China. At the time of the ECFA signing, the Taiwan Mirror Glass Enterprise Ltd. persuaded IKEA in Sweden to set up an R&D center in Taiwan to seize the cutting-edge R&D technology. Concurrently, several world-renowned furniture brands – such as Wal-Mart, HOMEDECO, and Japan TOTO – have also indicated strong recognition of the Taiwan glass team's Team Taiwan Glass brand, so orders from these companies can be expected soon. Based on the abovementioned successful glass industry integration and cooperation cases, Taiwan's flat glass manufacturers ought to consider joining Team Taiwan Glass in order to establish a more powerful cooperative network and engage in industrial level and vertical integration. If Taiwan's flat glass industry can cooperate with the upstream raw material market (Taiwan Glass Ind. Corp.), there will be room for price negotiation with the upstream companies, while the team alliance can also jointly enhance the quality control and R&D capabilities. At the same time, in response to the rapid rise of China's automobile market, industry officials should consider letting Taiwan Glass Ind. Corp. and Taiwan Mirror Glass Enterprise Ltd. take the lead in gaining entry into the automobile glass industrial market through the cooperative relationship between Taiwan Glass Ind. Corp. and China, thereby expanding Taiwan's flat glass industry overseas. The main reasons that contribute to success are summarized into four phases.

1. (1) Establishment of a trusting relationship and selfless offer of funding assistance: Assistance such as funds, equipment, factory premises, and raw materials were offered to solid companies that give recognition to industrial clustering, but face operating and capital problems. This approach not only saved many floating glass processors on the verge of bankruptcy, but also reduced the start-up costs for a number of new entrants. The new and existing companies could accept purchase orders and produce high-quality products. After making profits, they gradually returned the funds and materials borrowed through amortization. The companies that benefited from this approach not only expressed heartfelt gratitude, but also undoubtedly developed a sense of trust and recognition for the ‘coexistence.’

2. (2) Maintaining and strengthening the relationship trust and sharing of orders: As the small processing plants lacked resources and manpower, Taiwan Mirror Glass Enterprise Ltd. integrated the related glass industries and accepted unified external orders to engage in professional labor of division. Through original equipment manufacturer, the purchase orders were distributed to other organizations in the same industry in order to integrate resources and distribute profits, thereby preventing larger companies from losing orders due to inadequate production capacity, or smaller companies from losing orders due to a scarcity of resources.

3. (3) Resource integration: Taiwan Mirror Glass Enterprise Ltd. continued to engage in self-upgrade and strengthening scale of the collaborating team within the industrial cluster. It also continued to join external bids on behalf of the industrial team. If a third party under the group won a bid, Taiwan Mirror Glass Enterprise Ltd. would provide raw materials needed for production. If Taiwan Mirror Glass Enterprise Ltd. won a bid, the third parties under the group would be commissioned for original equipment manufacturer. This practice prevented bargaining from competitors and maintained the market prices. Then, through strategic alliances, with the glass operators from the Chunghwa area as the core, Taiwan was divided into three scopes to share and integrate resources.

4. (4) Formation of the main cluster network: With the tacit cooperation experience within the cluster network, the third parties were divided into two main groups based on their credibility, machines and equipment, quality, and delivery dates. Based on their professional capacity and processing processes, orders accepted by Taiwan Mirror Glass Enterprise Ltd. were distributed to cope with the customers’ rapidly changing needs, stabilize product quality, and enhance the competitiveness of orders.

Currently, Team Taiwan Glass has extended this spirit to other industries, and the ‘Taiwan Glass Team and Related Industries Association’ was established, with hopes of engaging in marketing, product development, and development through a greater cluster effect and involving more relevant industries.

As the competitive environment of the market continues to change, the organizational structure is also subject to adjustment. In response to business environmental needs, companies are no longer confined to their own borders. Focused on the perspectives of resource dependence and specialization, more and more organizations emphasize the importance of establishing relationships with upstream and downstream manufacturers. In addition, buyers and sellers continue to strengthen their cooperative behaviors and maintain long-term partnerships (Kalwani & Narayandas, Reference Kalwani and Narayandas1995). From the angle of competitive advantage, in addition to the need for the flat glass industry to bid farewell to the era of fighting alone, other industries in Taiwan should also follow suit. The government and industries should engage in resource, talent, and market integration in order to prevent the signing of the ECFA from destroying Taiwan's flat glass industry and even other industries.

Research limitations and future direction for research

This study has three main limitations, outlined below.