Academic journal article Journal of Managerial Issues

Social Networks and Corporate Performance: The Moderating Role of Technical Uncertainty

Academic journal article Journal of Managerial Issues

Social Networks and Corporate Performance: The Moderating Role of Technical Uncertainty

Article excerpt

Between the rise of globalization and the development of information technology, the complexity of the operating environment in industry increases daily, perhaps making intangible assets the key to competitive advantage (Norton and Kaplan, 1996, 2003). Social networks, one such significant intangible asset, have gained prominence in recent years (e.g., Gulati et al., 2000; Koka and Prescott, 2002). The analyses of social networks are tied to multiple levels: individual, firm, and regional. This paper focuses primarily on social networks at the firm level. That is, firm network is being referenced. In terms of firms, social networks refer to a set of firms that are connected by a specific type of social relationship. Prior research has suggested that interfirm networks facilitate the transmission of information and the access to unique resources and thus can be thought of as an inimitable and non-substitutable asset (Gulati, 1999; Palmer et al., 1995). A number of studies have also suggested that social networks provide several benefits for firms, such as learning benefits (e.g., Powell et al., 1996), coordination benefits (e.g., Coleman, 1990; Uzzi, 1997), performance benefits (e.g., Maurer and Ebers, 2006), and increasing legitimacy (e.g., Higgins and Gulati, 2003). Therefore, whether and how social networks improve corporate performance has become a dominant theme in strategy and organizational research (see, e.g., Carpenter and Westphal, 2001; Goerzen, 2007; Zollo et al., 2002).

However, extant research examining the performance effects of social networks reports mixed results. Some research has found a positive impact: Carpenter and Westphal (2001) found that social networks formed by members of the boards of directors helped to obtain useful corporate strategy information and increase corporate performance; Inkpen and Tsang (2005) found that social networks not only promoted knowledge transfer among members of a network but also between union firms, thus promoting performance and innovation; Sorenson et al. (2009) found a positive impact of network ties in family businesses on corporate performance. Other research has found a negative impact: Goerzen and Beamish (2005) found that international companies with more diverse networks had, on average, lower economic performance than international companies with less diverse networks; Goerzen (2007) pointed out that social networks had a negative impact on corporate performance when there was a greater overlap of organizational partners. These conflicting findings reveal two possibilities. One is that the relationship between social networks and corporate performance might be nonlinear. Another is that the value of social networks might depend on a number of moderators and contingencies (e.g., Gulati and Higgins, 2003; Gupta et al., 2011; Hansen et al., 2001; Rowley et al., 2000).

To clarify the relationship between social networks and corporate performance, this research extends previous research about the performance impacts of social networks (e.g., Goerzen, 2007; Goerzen and Beamish, 2005) by exploring whether the relationship between social networks and corporate performance is a non-linear inverted U-shape curve. The authors particularly focus on examining network centrality, which signifies a firm's network position and determines the set of resources and information the firm could mobilize through social networks (Davis, 1991). When an organization has a higher network centrality, it can receive and transfer a greater amount of information, and therefore has an advantage in controlling information transfer and can benefit more from social networks. Despite that several other network properties have been studied in prior research, such as network structure and network composition, a firm's network position plays an important role in securing timely access to novel information and benefiting from social networks. Obtaining access to a greater number of more diverse information through a central position also comes at a price. It thus provides a good opportunity to understand the nonlinear relationship between social networks and corporate performance. Therefore, this study uses network centrality to examine the performance effects of social networks. Furthermore, the authors examine if the relationship between social networks and corporate performance is moderated by the level of technical uncertainty.

Technical uncertainty refers to the extent to which the complexity of the product or process technologies changes (Fynes et al., 2004; Ragatz et al., 2002). Burt (1992) believed that the uncertainty of production process and technical characteristics affect the firms' investment in social networks. Several papers point out that companies are more likely to make up for their inadequacies through organizational cooperation when they are faced with technical uncertainties (Robertson and Gatignon, 1998; Teece, 1992). In particular, companies rely more on interfirm networks to access different resources and information in an uncertain environment (Sparrowe et al., 2001; Westphal et al., 2006). Ragatz et al. (2002) indicated that technical uncertainty was an important factor for firms to consider when they select partners. As technical uncertainty is the major challenge faced by high-tech companies, this study explores how it affects the social network-performance relationship.

This study defines interfirm relationships by the interlocks which are created by either of the following two conditions: one is that a firm's CEO, senior management, or directors sit on the boards of other firms; another is that any board member who is primarily affiliated with another firm as the CEO, senior manager, or director. Existing literature generally agrees that interlocks are created for the firm's needs rather than for the individual's benefits (Mizruchi, 1996). Thus, the interlocks have been widely used in prior studies to measure interfirm networks (Grandori and Soda, 1995; Koenig and Gogel, 1981). Using interfirm network data collected in 2007 from high-tech companies in Taiwan, this research found the relationship between social networks and corporate performance, as measured by sales growth, is inverted U-shape. Specifically, when network centrality begins to increase, sales growth increases until it reaches equilibrium after which it begins to decrease with the increase of network centrality. This shows too much network centrality has a negative impact on corporate financial performance, while moderate network centrality can lead to the greatest performance in a company. Also, a higher level of technical uncertainty leads to a greater positive impact of social networks on corporate performance. Therefore, the higher the level of technical uncertainty companies face, the more companies may rely on building social networks to gain necessary resources and promote business development.

This study extends the literature in the following ways: first, previous studies examining the performance effects of social networks have not reached a consensus (e.g., Carpenter and Westphal, 2001; Goerzen, 2007; Inkpen and Tsang, 2005). This research shows that the relationship between social networks and corporate performance is not a linear monotonically increasing relationship but rather a nonlinear inverted U-shape curve. It provides one explanation for the mixed results. Specifically, most of prior research consider social networks as valuable assets and consistently infer that social networks positively affect the economic outcomes of firms. However, this study suggests the possibility to benefit from social networks is importantly conditioned on a firm's network position. Though central firms benefit from having timely access to novel information, they also bear higher costs to absorb and integrate a greater number of more diverse information. Hence, understanding the value of social networks requires a dual emphasis on the benefits obtained and the costs incurred. Second, most prior studies about the value of social networks focus on investigating the direct impact of social networks on economic performance (e.g., Goerzen and Beamish, 2005; Kim, 2005). This study suggests that the benefits of social networks vary based on the characteristics of the technical environment, indicating that the value of social networks is dependent on the external environment. Moreover, this study contributes to the current literature by identifying an important moderator, technical uncertainty.

This research provides several managerial implications as well: first, gaining more social networks may be detrimental to corporate growth, meaning that companies should find a balance between the cost and benefit of building social networks to find the optimal level of investment; second, the optimal level of investment in social networks varies depending on the environment that companies face, meaning that companies should build social relationships based on their needs rather than blindly pursuing connections to other firms; third, when a company faces greater technical uncertainty, it may rely more on inter-organizational networks to decrease R&D risks and increase company effectiveness.

In the remainder of this paper, the authors will briefly review the literature and develop two testable hypotheses. The authors then describe the research methodology. Subsequently, the implications of empirical findings will be discussed. Finally, this study raises some concluding remarks with implications for both managers and scholars.

THEORETICAL DEVELOPMENT

Definition of Social Networks

A social network is broadly defined as a set of nodes that are connected by a specific type of social relationship (Laumann, 1978). Here, nodes may refer to individuals, events, organizations, or firms. That is, social networks could refer to the specific relational connections between a group of people, individual firms, or specific events, etc. (Knoke and Kuklinski, 1982). Generally, social network constructs encompass individual, interfirm, and intrafirm networks. In terms of interfirm networks, the relationships can be various kinds, such as supplier-buyer, firm-customer, and regulator-firm, etc. This study defines the interfirm relationships by the interlocks, which is the most widely employed measure of interfirm networks (Mizruchi, 1996). In general, actors within a network may repeatedly and consistently interact with each other (Podolny and Page, 1998) to obtain resources, technical, or emotional support (BarNir and Smith, 2002).

As argued in prior literature, social networks are highly associated with social capital (Lin, 2001). Social capital encompasses all types of relationships and all levels, such as individual and firm, and can be analyzed from different perspectives. Based on the network perspective, social capital bears potential value as it provides an opportunity for actors to access information and resources in their social networks (Maurer and Ebers, 2006). This concept of social capital has been discussed extensively since the mid-1980s (e.g., Bourdieu, 1980; Coleman, 1990; Flap, 1988, 1991, 1994). The core intuition behind these definitions offered is that social capital signifies resources that are embedded in social relationships and in the social structure that can be mobilized by the firm to increase the success rate for certain actions (Lin, 2001). In other words, social capital is an investment in social relationships by an organization expecting returns from the market (Biggart and Castanias, 2001; Burt, 1992; Coleman, 1988, 1990). Therefore, this study employs the network perspective to examine whether social networks are valuable and particularly focuses on interfirm networks.

Social Networks and Corporate Performance

From an organizational sociology viewpoint, connections between two or more "actors" bring out specific characteristics and interactions, and "relationships" are produced through the interactions of the actors. There are four theories in the literature based on inter-organizational relations: resource dependence theory, transaction cost theory, network theory, and social capital theory. The first three theories emphasize observation of the position of each separate organization within a network and how they "build relationships" with other organizations. Social capital theory attempts to understand why the network relationships are formed within a network.

According to resource dependence theory, firms are dependent on their external environment for resources that are essential for their functioning (Goes and Park, 1997). Specifically, since firms cannot generate all necessary inputs, they exchange with other firms to obtain them (Cummings, 1984; Pfeffer and Salancik, 1978). Transaction cost theory states mutual benefit and standards between organizations within a network minimize information asymmetry, which then decreases the chance of opportunism (Kogut, 1988, 1989), and decreases the cost of consultation, contracting, and supervision, thus adding creative value for the organization (Pisano et al., 1988). Also, organizational networks provide a channel to relay information at a far faster rate that greatly exceeds the amount that an organization could find within itself, making it more valuable (Burt, 1992). Network theory suggests that different companies have varying amounts of resources and capabilities, and therefore working through connections with other companies helps integrate diverse bases of knowledge of technology or the market, thus creating synergy (Butt, 1980; Granovetter, 1973; Lin, 1998).

Interlocks are considered one type of interfirm relationship (Grandori and Soda, 1995; Mizruchi, 1996). A network of interlocks may bring benefits for firms. For example, interlock networks provide firms with the opportunity to exchange firm-specific information and general industry information, which allow firms to be more responsive to their environments (Allen, 1974; Burt, 1979). Interlock ties also allow firms to improve inter-organizational coordination and reduce transaction costs (Palmer, 1983; Pfeffer and Salancik, 1978). Furthermore, prior studies suggest that a firm's network position is an important aspect of social structure that can enhance the firm's ability to benefit from network ties and to achieve economic goals (e.g., Gilsing et al., 2008; Tsai and Ghoshal, 1998). Specifically, an interlock network provides channels for firms to exchange their resources and information. However, such resources and information are usually distributed unevenly within a network. Different network positions represent different opportunities for a firm to access required resources and information.

By occupying a central network position, firms can obtain a greater amount of information through the networks (Sparrowe et al., 2001). Specifically, because central firms have greater connections to others, they have more relationships to draw from in obtaining information (Cook and Emerson, 1978). When an organization receives more diverse information, senior management is more likely to use that information to identify market opportunities and improve corporate performance. Also, a central position is associated with greater power and control (Cook and Whitmeyer, 1992). With higher network centrality, firms have greater control over resources because they can choose from a greater number of alternative firms in sharing beneficial resources. In addition, central firms are more likely to exert supervision and sanctions to decrease the probability of members behaving opportunistically (Antia and Frazier, 2001; Coleman, 1988). Furthermore, as several scholars have argued, network centrality can enhance a firm's ability to create value (e.g., Coleman, 1990; Tsai and Ghoshal, 1998), which will improve the firm's corporate financial performance.

However, building social networks is not costless. The differences in business operation methods within the network may lead to a greater chance that members lack a common target and therefore the motivation for cooperation (Goerzen and Beamish, 2005; Rindfleisch and Moorman, 2001). Also, a company occupying a more central position in the network must invest large amounts of capital to hold together its many network relationships. Besides, with higher network centrality, firms have to deal with a higher volume of more diverse information (Gnyawali and Madhavan, 2001). This consumes time and resources that cannot be allocated to integrating the new insights (Gilsing et al., 2008). Also, overly high centrality might impede a firm's capability to absorb and digest information it obtained (Ahuja and Katila, 2004), which has a negative impact on corporate performance. When potential losses incurred by inter-organizational network ties outweigh the potential benefits, a social network may not only have no benefits to corporate performance but may instead be detrimental to the company.

In summary, with increasing network centrality firms can obtain more diverse information and have greater control over resources. Also, opportunistic behavior may be minimized. However, overly high network centrality may increase the cost of maintaining the network, and increase the cost of absorbing and integrating the acquired information, which has a negative impact on corporate growth. Therefore, this study predicts an inverted U-shaped relationship between social networks and corporate performance, and the following hypothesis is developed:

H1: Social networks have an inverted U-shaped relationship with corporate performance.

The Moderating Effect of Technical Uncertainty

Technical uncertainty is an important aspect of external environments. Scholars have debated the effect of technical uncertainty on corporate performance, particularly with regard to high-tech industries. In an environment of high technical uncertainty, current technology may soon become outdated and firms need to become involved in different fields of technology to expand and extend their knowledge base (Cohen and Levinthal, 1989; March, 1991). Increased complexity and interrelatedness leads to higher cost and lower benefits of R&D, which increases the variation in R&D effectiveness and further affects the variation in corporate R&D investments. The coefficient of variation for R&D intensity, therefore, could reflect the level of technical uncertainty firms are facing.

Based on resource dependence theory, firms could reduce the negative impact of environmental uncertainty through cooperative relationships with other firms (Butt, 1983; Pfeffer and Salancik, 1978; Palmer et al., 1995). In other words, a firm is more likely to rely on social relationships to meet its information needs and further improve its corporate performance in a technically uncertain environment. Through corporative relationships, firms can shorten the time involved in obtaining technology and the capital invested for research and development when technical uncertainty is high (Robertson and Gatignon, 1998).

Based on contingency theory, the degree of certainty of a firm's external environment is an important moderator affecting a firm's network structure and its performance (Gargiulo and Rus, 2002; Gulati and Gargiulo, 1999; Shaner and Maznevski, 2011). Several studies also indicate that the technological environment has a great impact on inter-organizational networks (e.g., Madhavan et al., 1998). More specifically, companies often look to gain key technical information through connecting with or establishing specific networks in a technically uncertain environment. Through cooperation, firms can make up for each other's technical weaknesses and acquire critical knowledge to improve their technological capabilities at less cost (Hagedoorn, 1993, 1995). In particular, with higher network centrality, firms are more likely to have timely access to important and novel information through social networks. Firms thus are able to increase the possibility of achieving economic outcomes when facing technical uncertainty.

In contrast, in more certain environments, fewer interconnections are more than enough for firms to obtain information and resources necessary for their growth (Shaner and Maznevski, 2011). When technical uncertainty is higher, firms are more likely to gain new information and technology through these corporate networks. Increasing technical uncertainty increases a firm's reliance on the information and knowledge from social networks. Being open to external sources enables firms to draw in ideas from outsiders to deepen the pool of technological opportunities available to them and thus improve their performance (Laursen and Salter, 2006). Therefore, it is expected that technical uncertainty will moderate the effect of social networks on corporate performance in such a fashion that increasing technical uncertainty will increase the positive effect of social networks on corporate performance, and increase the amplitude of the effects of social networks. The second hypothesis is as follows:

H2: The level of technical uncertainty moderates the relationship between social networks and corporate performance in such a fashion that increasing technical uncertainty will: (a) increase the positive effect of social networks on corporate performance; and (b) increase the amplitude of the effects of social networks.

METHODS

Sample and Data Collection

The research subjects are firms in the electronics industry publicly listed in Taiwan in 2007. The authors initially identified 365 electronics firms in this industry. After excluding those companies missing financial data, research and development costs, annual reports, and other related information, this research finally included 353 electronic firms in the sample. Electronics firms were chosen based on the Taiwan Economic Journal (TEJ) industrial classification of electronics firms, which includes those involving semiconductors, computer and peripheral products, optoelectronics, communications networks, electronic components, data services, and other electronic services. All data were found in the Taiwan Economic Journal (TEJ) Database.

Variables Measurement

Dependent Variables. Following prior studies (e.g., Maurer and Ebers, 2006), this research measures corporate performance through the firm's sales growth rate. In the electronics industry, an increase in sales revenue indicates successful research and development of industrial products. Besides, the sales growth rate is a variable that is viewed as less noisy than other accounting metrics (Kor and Sundaramurthy, 2009), and is likely to be closely linked to network structure, since the connections in a network may contribute to the variables related to sales. Therefore, this study uses sales growth rate to measure corporate performance (PF). Sales growth rate is calculated by first subtracting sales revenue of year 2007 from sales revenue of year 2008 to produce the change in sales revenue and then dividing the change in sales revenue by sales revenue of year 2007.

Independent Variables. In terms of measuring corporate social networks, this study based its analysis on Nicholson et al. (2004). First, a two-way matrix is formed based on whether or not the chief executive officer (CEO), senior management, or directors of the company sit on the boards of another company; or if any board member of the company is primarily affiliated with another company as the CEO, senior manager, or director. If either of the above two conditions were met, then these two companies were found to be related and were marked as 1 in the matrix. If they were not related, they were marked as 0. Prior studies indicate that a firm can access resources or information from other firms by having ties to their boards (Mizruchi, 1996, 2004; Scott, 1991).

After establishing a two-way matrix of whether a company has relationships with another company, this study employs the eigenvector centrality measure proposed by Bonacich's (1987) in the evaluation of social networks before finally using UCINET network analysis software to make the final calculations. Network centrality refers to the position in a network, with a higher network centrality indicating a more central position, therefore making it closer to the core of the network. Eigenvector centrality can be seen as a weighted sum of both direct connections and indirect connections of every length. A higher eigenvector centrality score represents that the paths connecting the focal firm's position to other positions are the highly weighted short paths (Bonacich, 1987). As indicated in prior literature, this measure has advantages over other centrality measures (Bonacich, 2007). Different from degree-based centrality, which gives every contact an equal weight, the eigenvector centrality weights contacts according to their centralities. Thus, it takes into account the entire pattern in the network while other centrality measures do not.

As for the measurement of technical uncertainty (TU), this study first divided annual research and development expenditures by the year's sales revenue to find the R&D intensity for each year from 2003-2007, then obtained the standard deviation of R&D intensity and the mean of R&D intensity during the five-year period. This study further divided the standard deviation of R&D intensity rate by mean R&D intensity to obtain the coefficient of variation for R&D intensity, which is used to measure the technical uncertainty.

Control Variables. In order to avoid interference from other factors in the evaluation of the relationship between social network and corporate performance, this study incorporated the research by Goerzen (2007) and Kim (2005), including industry profitability, capital structure, firm size, R&D intensity, the scale of assets, and company age as control variables. Firm size (SIZE) is measured by taking the natural log of the number of company employees for the year. Industry profitability (INP) is measured by the industrial average return of assets. Capital structure (LEV) is measured by the financial leverage. The financial leverage is calculated by dividing total debt by shareholders' equity. R&D intensity (RD) is calculated by dividing annual R&D expenditures by the year's sales revenue. The scale of assets (TA) is calculated by taking the natural logarithm of total assets. Previous literature suggests that the scale of total assets affects corporate performance and therefore it is used as a control variable. Company age (AGE) is calculated by subtracting the year the company was established from the year of the sample data.

Empirical Models

To test Hypothesis 1, this study uses Kim's (2005) non-linear regression model with both linear social network and squared social network measures in the regression model. That is, the model is designed as a quadratic function. If the coefficient for the linear social network variable is positive, and the coefficient for the squared social network variable is negative, then it is suggested that there is an inverted U-shaped curve. The regression model is as follows:

[F.sub.it] = [[alpha].sub.0] + [[beta].sub.1][SC.sub.it.sup.2] + [[beta].sub.2][SC.sub.it.sup.2]] + [[beta].sub.3] TU + [[beta].sub.4] TU + [[beta].sub.4][SIZE.sub.it] + [[beta].sub.5][INP.sub.it] + [[beta].sub.6][LEV.sub.t] + [[beta].sub.7][RD.sub.it] + [[beta].sub.8][TA.sub.it] + [[beta].sub.9][AGE.sub.it] + [[epsilon].sub.it] (M1)

where,

[PF.sub.it] = the sales growth of company i in year t.

[SC.sub.it] = the social network of company i in year t.

[SIZE.sub.it] = the natural log of the number of company employees for company i in year t.

[INP.sub.it] = net profits after taxes divided by total assets for company i in year t.

[LEV.sub.it] = debt equity ratio for company i in year t.

[RD.sub.it] = annual R&D expenditures divided by the year's sales revenue for company i in year t.

[TA.sub.it] = the natural log of total assets for company i in year t.

[AGE.sub.it] = number of years since company i has been in business.

To test Hypothesis 2, this study added two interaction terms into Model 1 and developed the regression model as follows.

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

where,

[TU.sub.it] = the coefficient of variation for R&D intensity of company i in year t.

The definitions of the other variables are the same as in the first model (M1).

ANALYSES AND RESULTS

Descriptive Statistics

There are 353 firms included in the sample of this study. As seen in Table 1, the lowest value of social network (SC) was 0, the highest was 73.73, and the average was 12.44. This shows that there is a great difference in social network between companies. The lowest performance (PF) was -78.99%, indicating the firm's sales revenue of the current year dropped 78.99% compared to sales revenue of the prior year. The highest performance was 1111.83%, the average was 25.77%, and the growth rate data were not skewed. As for technical uncertainty, the minimum value was 0, the maximum value was 2.24, and the average was 0.41. This shows that there is a wide range in the level of technical uncertainty that sample firms face.

Table 2 shows the correlation matrix for all variables. Spearman rank correlation coefficients are listed on the upper right hand side and Pearson correlation coefficients on the lower left hand side. As shown in Table 2, firm size (SIZE) is positively correlated with linear social network variable (SC). Firm age (AGE) is negatively correlated with linear social network variable (SC). However, linear social network variable (SC) is not positively correlated with corporate performance (PF), which needs to be further examined with regression analyses. Besides, all correlation coefficients other than the correlation between linear social network variable and squared social network variable are below 0.8, meaning that there is no multicollinearity existing. In regression analyses, this study will further use the variance inflation factor (VIF) to test for multicollinearity and correct the problem if necessary. Specifically, if the VIF does not exceed 10, then the regression model is not affected by multicollinearity (Hair et al., 1998).

Regression Results

Table 3 shows the results of the regression analysis on the relationship between social networks and corporate performance. The first column of Table 3 presents the results of the respective baseline model including all of the control variables as well as the moderating variable. As expected, the explanatory power of the baseline model (adj. [R.sup.2] = 0.108) is lower than the explanatory power of the model with social network measures (Model 1). In particular, the model with social network measures and the interaction terms (Model 2) has the highest explanatory power (adj. [R.sup.2] = 0.152). These results provide initial evidence that it is appropriate to look at technical uncertainty as the moderating variable in the relationship between social networks and corporate performance.

In Hypothesis 1, the relationship between a firm's social network and its performance is suggested to be an inverted U-shape, implying that the social network enhances a firm's sale growth. In Model 1 and 2, because the linear terms for social networks are positive and significant ([[beta].sub.1] = 1.534, t-value = 1.846 in Model 1, [[beta].sub.1] = 1.945; t-value = 2.369 in Model 2), whereas the squared terms for social networks are negative and significant ([[beta].sub.2] = -0.038, t-value = -2.161 in Model 1; [[beta].sub.2] = -0.049, t-value = -2.833). The first hypothesis is supported.

In Hypothesis 2, it is proposed that the level of technical uncertainty faced by a firm moderates the inverted U-shaped relationship between social networks and corporate performance so that the positive effect of social networks on corporate performance is stronger and the amplitude of the effects of social networks is greater when the firm faces higher levels of technical uncertainty. To test the hypothesis, this study inserted the interaction terms between the social network variables (linear and squared terms) and technical uncertainty in Model 2 when predicting corporate sales growth. Considering that the social network variables may be correlated with their respective interaction terms if the social network variables were multiplied by the technical uncertainty variable directly, this study first standardized social network variables and the technical uncertainty variable respectively. The standardization process is subtracting the mean of one variable from every value and then dividing by the standard deviation of that variable, which has been suggested to be an appropriate way for standardization in prior literature (e.g., Davis, 1986). Then, standardized social network variables (linear and squared terms) were multiplied by standardized technical uncertainty variable to produce the interaction terms. In Model 2, the results reveal that the interaction between the linear social network and technical uncertainty is positive and significant ([[gamma].sub.1] = 22.651, t-value = 3.423), whereas the interaction between the squared social network variable and technical uncertainty is negative and significant ([[gamma].sub.2] = -18.861, t-value = -4.062). These results imply that the relationship between social networks and corporate performance remains an inverted U-shape when assessing the moderating effect of technical uncertainty. As predicted, technical uncertainty moderates this relationship in such a fashion that the positive effect of social networks on corporate performance is stronger and the amplitude of the effects of social networks is greater when the firm faces higher levels of technical uncertainty. The results obtained support Hypothesis 2.

Sensitivity Analyses

There are several measures of network centrality proposed in prior studies. In addition to Bonacich's (1987) eigenvector centrality measure, the degree-based centrality measure proposed by Freeman (1979) is widely used (e.g., Tsai, 2001). To increase the robustness of empirical results, this study additionally used Freeman's (1979) concept in the evaluation of network centrality and then replaced eigenvector centrality measure with this degree-based centrality measure to reanalyze model (M1) and model (M2) as reported in Table 4. Specifically, this study employs a normalized in-degree centrality. In-degree centrality represents the total number of firms with which a focal firm has direct relationships. A normalized in-degree centrality is created by normalizing in-degree centrality into a value between zero and one. As Freeman (1979) indicated, this is a very suitable centrality measure for capturing an individual actor's information access. The mean and the standard deviation of this network centrality measure is 0.62 and 0.71, respectively. As shown in Table 4, the explanatory power of the baseline model (adj. [R.sup.2] = 0.108) is less than the explanatory power of Model 1 and 2 (adj. [R.sup.2] = 0.131), showing the incremental explanatory power of social network variables and the interaction terms. In Model 1 and 2, the linear social network variable is significantly positive ([[beta].sub.1] = 31.612, t-value = 1.907 in Model 1; [[beta].sup.1] = 34.064, t-value = 2.056 in Model 2), whereas the squared social network variable is significantly negative (f12 = -12.253, t-value = - 1.802 in Model 1; [[beta].sub.2] = -14.246, t-value = -2.046 in Model 2), indicating that the relationship between social networks and corporate performance is inverted U-shaped. These results are consistent with Table 3, supporting Hypothesis 1.

This study further examines the moderating effect of technical uncertainty. Empirical results are reported in the third column of Table 4. To address the multicollinearity issues, this study first standardized social network variables and the technical uncertainty variable and then produced the interaction terms. As shown in Table 4, the interaction between the linear social network variable and technical uncertainty is significantly positive ([[gamma].sub.1] = 19.278, t-value = 2.991) and the interaction between squared social network variable and technical uncertainty is significantly negative ([[gamma].sub.2] = -12.707, t-value = -2.270), indicating that the positive impact of social networks on corporate performance is stronger and the amplitude of the effects of social networks is greater when the firm faces higher levels of technical uncertainty. In summary, the empirical results support the predictions of Hypothesis 1 and 2.

CONCLUSION AND DISCUSSION

In recent years, social networks have become a popular topic of study in the management field. This study extends prior research by analyzing whether there is a non-linear relationship between social networks and corporate performance and how technical uncertainty moderates that relationship. Empirical findings show an inverted U-shaped relationship between social networks and corporate performance. That is, network centrality has a positive effect on corporate performance at first, but after it reaches a threshold level, higher centrality adversely affects corporate performance. This is contrary to many prior studies that assume the impact of social networks is linear. It is also found that the positive impact of social networks on corporate performance is stronger and the amplitude of the effects of social networks is greater when companies face higher levels of technical uncertainty. This means firms are more likely to rely on interfirm networks to obtain resources and knowledge while in more technically uncertain environments.

This research provides empirical support for the existence of both a positive and negative relationship between social networks and corporate performance. That is, once network centrality reaches some maximum level, the relationship between it and corporate performance becomes negative. The findings of this study have several important management implications: first, this research points out that the overuse of social relationships may actually have a negative impact and harm corporate performance, meaning that companies should seek the optimum level of investment of social networks. Second, this research shows that the optimal level of social networks for each company may vary based on the technical environment. So, companies should take environmental factors into consideration when building social relationships and developing social networks. Finally, this study points out that network centrality is an important determinant of social networks while too much centrality may be detrimental to a firm. Finns should find the balance between linkage cost and information benefits to maximize corporate performance.

Regarding research limitations, this research focuses mainly on interorganizational networks (i.e., external networks) but does not take internal networks into consideration. As social capital may arise from internal networks and external networks, future research may consider the interaction between internal networks and external networks to further understand how to integrate different types of network ties to reap the greatest benefit. Besides, it may be useful in future studies to consider the interaction between network centrality and other network properties, such as network density and network diversity to further understand the potential of a firm's social networks. Finally, this study employs cross-sectional analysis to examine the performance impacts of social networks. An analysis of panel data or longitudinal data will be required to study how network dynamics affect corporate performance.

References

Ahuja, G. and R. Katila. 2004. "Where Do Resources Come From? The Role of Idiosyncratic Situations." Strategic Management Journal 25(8-9): 887-907.

Allen, M. P. 1974. "The Structure of Interorganizational Elite Cooptation: Interlocking Corporate Directorates." American Sociological Review 39: 393-406.

Antia, K. D. and G. L. Frazier. 2001. "The Severity of Contract Enforcement in Interfirm Channel Relationships." Journal of Marketing 65:67-81.

BarNir, A. and K. A. Smith. 2002. "Interfirm Alliances in the Small Business: The Role of Social Networks." Strategic Management Journal 40:219-232.

Biggart, N. W. and R. P. Castanias. 2001. "Collateralized Social Relations: The Social in Economic Calculation." American Journal of Economics and Sociology 60:471-500.

Bonacich, P. 2007. "Some Unique Properties of Eigenvector Centrality." Social Networks 29: 555-564.

--. 1987. "Power and Centrality: A Family of Measures." American Journal of Sociology 92:1170-1182.

Bourdieu, P. 1980. "Le Capital Social: Notes Provisoires." Actes de la Recbercbe en Sciences Sociales 3: 2-3.

Burt, R. S. 1992. Structural Holes: The Social Structure of Competition. Cambridge, MA: Harvard University Press.

--. 1983. "Firms, Directors and Time in the Directorate Tie Market." Social Networks 5(1): 13-49.

--. 1980. "Autonomy in a Social Topology." American Journal of Sociology 85: 892-925.

--. 1979. "A Structural Theory of Interlocking Corporate Directorates." Social Networks 1(4): 415-435.

Carpenter, M. A. and J. D. Westphal. 2001. "The Strategic Context of External Network Ties: Examining the Impact of Director Appointments on Board Involvement in Strategic Decision Making." Academy of Management Journal 44: 639-660.

Cohen, W. M. and D. A. Levinthal. 1989. "Innovation and Learning: The Two Faces of R & D." The Economic Journal 99: 569-596.

Coleman, J. S. 1990. Foundations of Social Theory. Cambridge, MA: Harvard University Press.

--. 1988. "Social Capital in the Creation of Human Capital." American Journal of Sociology 94: 95-120.

Cook, K. S. and R. M. Emerson. 1978. "Power, Equity and Commitment in Exchange Networks." American Sociological Review 43:721-739.

-- and J. M. Whitmeyer. 1992. "Two Approaches to Social Structure: Exchange Theory and Network Analysis." Annual Reviews 18:109-127.

Cummings, T. G. 1984. "Transorganizational Development." In B. Staw and L. Cummings (eds.) Research in Organizational Behavior Vol. 6. Greenwich, CT: JAI Press. pp. 367-422.

Davis, G. F. 1991. "Agents without Principles? The Spread of the Poison Pill through the Intercorporate Network." Administrative Science Quarterly 36(4): 583-613.

Davis, J. C. 1986. Statistics and Data Analysis in Geology. Wiley Publisher.

Flap, H. D. 1994. "No Man Is an Island: The Research Program of a Social Capital Theory." Presented at the World Congress of Sociology. Bielefeld, Germany. July.

--. 1991. "Social Capital in the Reproduction of Inequality, A Review." In M. J. Pelaez (ed.) Comparative Sociology of Family, Health & Education: A Volume in Memory of Ferran Valls i Taberner / coord, pp. 6179-6202.

--. 1988. Conflict, Loyalty and Violence. Frankfurt am Main: Peter Lang International Academic Publisher.

Freeman, L. C. 1979. "Centrality in Social Networks Conceptual Clarification." Social Networks 1: 215-239.

Fynes, B., S. D. Burca, and D. Marshall. 2004. "Environmental Uncertainty, Supply Chain Relationship Quality and Performance." Journal of Purchasing & Supply Management 10: 179-190.

Gargiulo, M. and A. Rus. 2002. "Access and Mobilization: Social Capital and Top Management Response to Market Shocks." Working Paper, France: INSEAD.

Gilsing, V., B. Nooteboom, W. Vanhaverbeke, G. Duysters, and A. van den Oord. 2008. "Network Embeddedness and the Exploration of Novel Technologies: Technological Distance, Betweenness Centrality and Density." Research Policy 37(10): 1717-1731.

Gnyawali, D. R. and R. Madhavan. 2001. "Cooperative Networks and Competitive Dynamics: A Structural Embeddedness Perspective." Academy of Management Review 26(3): 431-445.

Goerzen, A. 2007. "Alliance Networks and Firm Performance: The Impact of Repeated Partnerships." Strategic Management Journal 28: 487-509.

-- and P. W. Beamish. 2005. "The Effect of Alliance Network Diversity on Multinational Enterprise Performance." Strategic Management Journal 26: 333-354.

Goes, J. B. and S. H. Park. 1997. "Interorganizational Links and Innovation: The Case of Hospital Services." Academy of Management Journal 40(3): 673-696.

Grandori, A. and G. Soda. 1995. "Inter-firm Networks: Antecedents, Mechanisms and Forms." Organization Studies 16(3): 183-214.

Granovetter, M. 1973. "The Strength of Weak Ties." American Journal of Sociology 78: 1360-1380.

Gulati, R. 1999. "Network Location and Learning: The Influence of Network Resources and Firm Capabilities on Alliance Formation." Strategic Management Journal 20(5): 397-420.

-- and M. Gargiulo. 1999. "Where Do Interorganizational Networks Come From?" American Journal of Sociology 104: 1439-1493.

-- and M. C. Higgins. 2003. "Which Ties Matter When? The Contingent Effects of Interorganizational Partnerships on IPO Success." Strategic Management Journal 24: 124-144.

--, N. Nohria, and A. Zaheer. 2000. "Guest Editors' Introduction to the Special Issue: Strategic Networks." Strategic Management Journal 21 (3): 199-201.

Gupta, V. K., R. Huang, and A. A. Yayla. 2011. "Social Capital, Collective Transformation Leadership, and Performance: A Resource-Based View of Self-Managed Teams." Journal of Managerial Issues 23(1): 31-45.

Hagedoorn, J. 1995. "A Note on International Market Leaders and Networks of Strategic Technology Partnering." Strategic Management Journal 16: 241-250.

--. 1993. "Strategic Technology Alliances and Modes of Cooperation in High-Technology Industries." In G. Grabher (ed.) The Embedded Firm: On the Socioeconomics of Industrial Networks. London: Routledge. pp. 116-137.

Hair, J., R. Anderson, R. Tatham, and W. Black. 1998. Multivariate Data Analysis. Prentice-Hall.

Hansen, M. T., J. M. Podolny, and J. Pfeffer. 2001. "So Many Ties, So Little Time: A Task Contingency Perspective on Corporate Social Capital in Organizations." Social Capital of Organizations 18:21-57.

Higgins, M. C. and R. Gulati. 2003. "Getting Off to a Good Start: The Effects of Upper Echelon Affiliations on Underwriter Prestige." Organization Science 14: 244-263.

Inkpen, A. and E. Tsang. 2005. "Social Capital, Networks and Knowledge Transfer." The Academy of Management Review 30: 146-165.

Kim, Y. 2005. "Board Network Characteristics and Firm Performance in Korea." Corporate Governance: An International Review 13: 800-808.

Knoke, D. and J. H. Kuklinski. 1982. Network Analysis. California: SAGE Publications, Inc.

Koenig, T. and R. Gogel. 1981. "Interlocking Corporate Directorships as a Social Network." American Journal of Economics and Sociology 40(1): 37-50.

Kogut, B. 1989. "The Stability of Joint Ventures: Reciprocity and Competitive Rivalry." The Journal of Industrial Economics 38: 183-198.

--. 1988. "Joint Ventures: Theoretical and Empirical Perspectives." Strategic Management Journal 9: 319-332.

Koka, B. R. and J. E. Prescott. 2002. "Strategic Alliances as Social Capital: A Multidimensional View." Strategic Management Journal 23: 795-816.

Kor, Y. Y. and C. Sundaramurthy. 2009. "Experience-based Human Capital and Social Capital of Outside Directors." Journal of Management 35: 981-1006.

Laumann, E. O. 1978. "Community Structure as Interorganizational Linkages." Annual Review of Sociology 4: 455-84.

Laursen, K. and A. Salter. 2006. "Open for Innovation: The Role of Openness in Explaining Innovation Performance among U.K. Manufacturing Firms." Strategic Management Journal 27: 131-150.

Lin, N. 2001. Social Capital: A Theory of Social Structure and Action. Cambridge: Cambridge University Press.

--. 1998. Social Exchange: Its Rational Basis. Montreal, Canada: World Congress of Sociology.

Madhavan, R., B. R. Koka, and J. E. Prescott. 1998. "Networks in Transition: How Industry Events (Re)shape Interfirm Relationships." Strategic Management Journal 19: 439-459.

March, J. G. 1991. "Exploration and Exploitation in Organizational Learning." Organization Science 2: 71-87.

Maurer, I. and M. Ebers. 2006. "Dynamics of Social Capital and their Performance Implications: Lessons from Biotechnology Start-ups." Administrative Science Quarterly 51 (June): 262-291.

Mizruchi, M. S. 2004. "Berle and Means Revisited: The Governance and Power of Large U.S. Corporations." Theory and Society 33: 579-617.

--. 1996. "What Do Interlocks Do? An Analysis, Critique, and Assessment of Research on Interlocking Directorates." Annual Review of Sociology 22: 271-298.

Nicholson, G. J., M. Alexander, and G. C. Kiel. 2004. "Defining the Social Capital of the Board of Directors: An Exploratory Study." Journal of the Australian and New Zealand Academy of Management 10(1): 54-72.

Norton, D. P. and R. S. Kaplan. 2003. Strategy Maps: Converting Intangible Assets into Tangible Outcomes. Boston, MA: Harvard Business School Press.

-- and R. S. Kaplan. 1996. The Balanced Scorecard: Translating Strategy into Action. Boston, MA: Harvard Business School Press.

Palmer, D. 1983. "Broken Ties: Interlocking Directorates and Intercorporate Coordination." Administrative Science Quarterly 28(1): 40-55.

--, B. M. Barber, X. Zhou, and Y. Soysal. 1995. "The Friendly and Predatory Acquisition of Large U.S. Corporations in the 1960s: The Other Contested Terrain." American Sociological Review 60(4): 469-499.

Pfeffer, J. and G. Salancik. 1978. The External Control of Organizations. New York: Harper & Row.

Pisano, G. P., S. Weijian, and D. J. Teece. 1988. Joint Ventures and Collaboration in the Biotechnology Industry. Ballinger Publishing Company.

Podolny, J. M. and K. L. Page. 1998. "Network Forms of Organization" Annual Review of Sociology 24: 57-76.

Powell, W. W., K. W. Koput, and L. Smith-Doerr. 1996. "Interorganizational Collaboration and the Locus of Innovation: Networks of Learning in Biotechnology." Administrative Science Quarterly 41: 116-145.

Ragatz, G. L., R. B. Handfield, and K. J. Petersen. 2002. "Benefits Associated with Supplier Integration into New Product Development under Conditions of Technology Uncertainty." Journal of Business Research 55: 389-400.

Rindfleisch, A. and C. Moorman. 2001. "The Acquisition and Utilization of Information in New Product Alliances: A Strength-of-ties Perspective." Journal of Marketing 65: 1-18.

Robertson, T. S. and H. Gatignon. 1998. "Technology Development Mode: A Transaction Cost Conceptualization." Strategic Management Journal 19: 515-531.

Rowley, T., D. Behrens, and D. Krackhardt. 2000. "Redundant Governance Structures: An Analysis of Structural and Relational Embeddedness in the Steel and Semiconductor Industries." Strategic Management Journal 21: 369-386.

Scott, J. 1991. Social Network Analysis. A Handbook. London: Sage.

Shaner, J. and M. Maznevski. 2011. "The Relationship between Networks, Institutional Development, and Performance in Foreign Investments." Strategic Management Journal 32: 556-568.

Sorenson, R. L., K. E. Goodpaster, P. R. Hedberg, and A. Yu. 2009. "The Family Point of View, Family Social Capital, and Firm Performance." Family Business Review 22: 239-253.

Sparrowe, R. T., R. C. Liden, S. J. Wayne, and M. L. Kraimer. 2001. "Social Networks and the Performance of Individuals and Groups." Academy of Management Journal 44(2): 316-325.

Teece, D. J. 1992. "Competition, Cooperation, and Innovation Organizational Arrangements for Regimes of Rapid Technological Progress." Journal of Economic Behavior and Organization 18: 1-25.

Tsai, W. 2001. "Knowledge Transfer in Intraorganizational Networks: Effects of Network Position and Absorptive Capacity on Business Unit Innovation and Performance." Academy of Management Journal 44(5): 996-1004.

-- and S. Ghoshal. 1998. "Social Capital and Value Creation: The Role of Intrafirm Networks." Academy of Management Journal 41 (4): 464-476.

Uzzi, B. 1997. "Social Structure and Competition in Interfirm Networks: The Paradox of Embeddedness." Administrative Science Quarterly 42: 35-67.

Westphal, J. D., S. Boivie, and D. H. M. Chng. 2006. "The Strategic Impetus for Social Network Ties: Reconstituting Broken CEO Friendship Ties." Strategic Management Journal 27(5): 425-445.

Zollo, M., J. J. Reuer, and H. Singh. 2002. "Interorganizational Routines and Performance in Strategic Alliances." Organization Science 13(6): 701-713.

Sui-Hua Yu (1)

Department of Accounting

National Chung Hsing University, Taiwan

Wei-Ting Chiu

Deloitte Touche Tohmatsu Limited, Taiwan

(1) The authors would like to acknowledge National Science Council of the Republic of China, Taiwan, for financially supporting this research under Contract NSC 99-2410-H-005-020, and are grateful for the comments provided by the participants in the American Accounting Association 2011 Management Accounting Section Midyear Meeting to earlier drafts, which were helpful in revising this article.

Table 1
Descriptive Statistics

Variables     N     Min      Mean    Std. Dev     Max

SC           353     0       12.44    14.10       73.73
[SC.sup.2]   353     0      353.00   660.75     5435.97
TU           353     0        0.41     0.43        2.24
SIZE         353     1.10     6.12     1.31       10.11
INP (%)      353     6.74     8.94     1.65       11.59
RD (%)       353     0        3.64     4.46       31.21
LEV (%)      353     2.26    40.58    16.89       92.43
TA           353     5.56     6.90     0.60        8.94
AGE          353     1.00    19.26     9.64       57.00
PF (%)       353   -78.99    25.77    91.53     1111.83

Table 2
Correlation Matrix

Variables      SIZE        INP         LEV          RD          TA

SIZE         1           0.010       0.122 **    0.002       0.728 ***
INP          0.072       1          -0.243 ***   0.143 ***  -0.059
LEV          0.064      -0.264 ***   1          -0.481 ***   0.336 ***
RD          -0.068       0.238 ***  -0.428 ***   1          -0.224 ***
TA           0.750 ***  -0.037       0.274 ***  -0.189 ***   1
AGE          0.015      -0.023       0.151 ***  -0.168 ***   0.055
PF           0.035      -0.064      -0.003      -0.126 **   -0.023
SC           0.292 ***   0.000      -0.003       0.029       0.325 ***
[SC.sup.2]   0.288 ***  -0.022       0.015       0.010       0.314 ***
TU          -0.121 **    0.007       0.027      -0.065      -0.068

Variables      AGE          PF          SC         SC2          TU

SIZE         0.007       0.061       0.256 ***   0.256 ***  -0.087
INP         -0.012       0.018       0.001       0.001      -0.024
LEV          0.131 **    0.015      -0.008      -0.008       0.046
RD          -0.165 ***  -0.205 ***   0.074       0.074      -0.143 ***
TA           0.040       0.026       0.255 ***   0.255 ***  -0.012
AGE          1          -0.189 ***  -0.277 ***  -0.277 ***  -0.159 ***
PF          -0.243 ***   1           0.054       0.054       0.100 *
SC          -0.256 ***   0.047       1           1           0.020
[SC.sup.2]  -0.229 ***   0.002       0.917 ***   1           0.012
TU          -0.187 **    0.211 ***   0.020       0.012       1

(a.) Pearson coefficients appear in the lower triangle; Spearman
coefficients appear in the upper triangle.

(b.) ***: I% significant level; **: 5% significant level; *: 10%
significant level.

Table 3
Relationship between Social Networks and Corporate Performance
And the Moderating Effect of Technical Uncertainty

                                 Baseline
                                   Model         Model        Model 2

Independent          Expected   Coefficient   Coefficient   Coefficient
Variables              Sign      (t-value)     (t-value)     (t-value)

Constant                ?       185.815 ***   177.595 **    157.576 **
                                 (2.690)       (2.546)       (2.301)
SC                      +        --             1.534 *       1.945 **
                                               (1.846)       (2.369)
[SC.sup.2]              -        --            -0.038 **     -0.049 ***
                                              (-2.161)      (-2.833)
[SC.sub.*]TU            +        --            --            22.651 ***
                                                             (3.423)
[SC.sup.2.sub.*]TU      -        --            --           -18.861 ***
                                                            (-4.062)
TU                      +        35.507 ***    35.327 ***    76.105 ***
                                 (3.223)       (3.219)       (5.172)
SIZE                    +        11.487 **     12.051 **     12.686 **
                                 (2.081)       (2.186)       (2.350)
INP                     +        -3.046        -3.361        -4.390
                                (-1.033)      (-1.143)      (-1.518)
LEV                     -        -0.161        -0.151        -0.015
                                (-0.501)      (-0.471)      (-0.049)
RD                      +        -3.506 ***    -3.539 ***    -3.139 ***
                                (-2.987)      (-3.025)      (-2.714)
TA                      +       -22.608 *     -22.287 *     -22.752 *
                                (-1.824)      (-1.770)      (-1.845)
AGE                     ?        -2.194 ***    -2.224 ***    -2.042 ***
                                (-4.408)      (-4.320)      (-4.021)

N                               353           353           353
Adj [R.sup.2]                     0.108         0.115         0.152
F-value                           7.092 ***     6.084 ***     6.750 ***
(n-value)                        (0.000)       (0.000)       (0.000)

***: 1% significant level; **: 5% significant level; *: 10%
significant level.

Table 4
Sensitivity Analyses: Relationship between Social Networks
and Corporate Performance and the Moderating Effect
of Technical Uncertainty

                                 Baseline
                                   Model        Model l       Model 2

Independent          Expected   Coefficient   Coefficient   Coefficient
Variables              Sign      (t-value)     (t-value)     (t-value)

Constant                ?       185.815 ***   191.559 ***   184.564 ***
                                 (2.690)       (2.739)       (2.663)
SC                      +        --            31.612 *      34.064 **
                                               (1.907)       (2.056)
[SC.sup.2]              -        --           -12.253 *     -14.246 **
                                              (-1.802)      (-2.046)
[SC.sub.*]TU            +        --            --            19.278 ***
                                                             (2.991)
[SC.sup.2.sub.*]TU      -        --            --           -12.707 **
                                                            (-2.270)
TU                      +        35.507 ***    35.035 ***    59.790 ***
                                 (3.223)       (3.187)       (3.963)
SIZE                    +        11.487 **     11.527 **     11.784 **
                                 (2.081)       (2.086)       (2.155)
INP                     +        -3.046        -3.318        -4.165
                                (-1.033)      (-1.125)      (-1.421)
LEV                     -        -0.161        -0.147        -0.086
                                (-0.501)      (-0.459)      (-0.271)
RD                      +        -3.506 ***    -3.582 ***    -3.252 ***
                                (-2.987)      (-3.055)      (-2.776)
TA                      +       -22.608 *     -24.683 *     -25.390 **
                                (-1.824)      (-1.954)      (-2.031)
AGE                     ?        -2.194 ***    -2.098 ***    -1.912 ***
                                (-4.408)      (-4.101)      (-3.737)

N                               353           353           353
Adj [R.sup.2]                     0.108         0.112         0.131
F-value                           7.092 ***     5.951 ***     5.821 ***
(p-value)                        (0.000)       (0.000)       (0.000)

***: 1% significant level; **: 5% significant level; *: 10%
significant level.
Author Advanced search

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.