Diversity in business-to-business information exchange: an empirical analysis of manufacturers...
Abstract
This article examines the performance implications of information exchange in industrial supply chains. While existing literature has addressed the critical role of information exchange in supply chain integration, existing studies fail to address the specific characteristics
**********
The exchange of information between firms within supply chains is of great interest to both researchers and practitioners. Supply chain literature recognizes the value of exchanging information to improve the supply chain performance in key functional areas such as logistics (Daugherty et al. 2002; Lieb and Butner 2007). The exchange of information is noted for its role in interfirm integration, sharing of performance data, and transparency (Cruijssen et al. 2007). Information exchange is also recognized for its positive performance effects when leveraged to dampen the bullwhip effect (Lee et al. 1997; Cachon and Fisher 2000; Machuca and Barajas 2004; Steckel et al. 2004). Strategically, the leveraging of information exchanged between firms is noted for its effects on competition (Sanders and Premus 2002) and specific areas of firm performance (Zsidisin et al. 2007). Complimenting the research on information exchange from a supply chain and logistics perspective is research on the use of information technology (IT) to span organizational boundaries. IT research recognizes that using IT to exchange information within the context of supply chains provides additional benefits to the participants (Bakos and Brynjolfsson 1993; Mukhopadhyay et al. 1995; Mukhopadhyay and Kekre 2002). The benefits to firms include decreased inventory investment (Mukhopadhyay et al. 1995), improved customer service (Allen et al. 1992), and reduced shipment errors (Srinivasan et al. 1994).
While firms may choose to exchange information with their trading partners in order to improve supply chain performance, they must balance the risks associated with providing information that can be used against them. Sharing forecast information with a supplier may allow the supplier to efficiently schedule production or it may alert the supplier to an opportunity to re-negotiate pricing. A supplier that uses information for opportunistic gain may impair the performance of its trading partner. Information exchange then becomes a double-edged sword where it is a source of efficiency in coordinating firm resources across the supply chain but can allow firms to act selfishly.
Prior research on information exchange has been limited to studies using perceived measures of information exchange collected through surveys (Whipple et al. 2002) and modeling studies that have simulated information exchange (Cachon and Fisher 2000; Cachon and Lariviere 2001; Angulo et al. 2004; Gaur et al. 2005b). Even studies using objective measures of information exchange have been limited to the study of a single buyer to its multiple suppliers (Mukhopadhyay et al. 1995). This study takes a unique approach by using archival data of actual electronic information exchanges from multiple firms within an electronic exchange network.
By observing information exchange occurring though an IT-enabled channel, this study utilizes unique measures of information exchange characteristics. This study uses specific IT-based measures to capture information exchange volumes and information exchange diversity.
HYPOTHESES
To be effective in dynamic markets, firms integrate externally with their trading partners (Rozenzweig et al. 2003; Vickery et al. 2003). Foundational to this integration is the exchange of information that will support the coordination of supply chain participants (Porter and Millar 1985; Cooper et al. 1997: Moberg et al. 2002). Existing literature supports the positive effects of integration on firm competitive performance (Daugherty et al. 2002; Whipple et al. 2002). As firms interact with their trading partners, they have the opportunity to minimize the cost of exchanging information by leveraging technology. Electronic data interchange (EDI) is a specific technology that uses standardized formats to electronically exchange business documents within and between organizations. In an EDI-enabled environment, firms may exchange large volumes of information with their trading partners at a minimal cost. Once the initial cost of formatting the information and establishing the communication link is made, the incremental cost of each additional document is minimal. Although the upfront cost of creating an EDI relationship has been noted as a deterrent to EDI implementation (Iacovou et al. 1995; Crum et al. 1998), these costs become sunk costs once the firm implements the technology. Once the electronic channel is established, firms can lower overall transaction costs by increasing the volume of information exchanged through the channel or by including more trading partners in the network (Crum et al. 1998). From an ordering cost perspective, inventory turnover can be improved by placing multiple orders for smaller quantities. Additional information including forecasts, production schedules, point-of-sale demand data, and inventory positions can all be exchanged electronically to improve the coordination of interfirm processes. Thus,
HI: Information exchange volume is positively associated with firm performance.
Information diversity refers to the number of unique types of information exchanged by a firm. Firms can choose how much information they exchange with their trading partners and how they exchange that information. EDI is used by firms to efficiently exchange business-to-business (B2B) information on a timely and cost-effective basis. A base level of information exchange must occur in order to do business in the supply chain. At a minimum, the buyer identifies what they want to purchase and when they require delivery. The supplier then confirms the pricing and availability back to the buyer. When the transfer of the physical product is complete, invoice and payment documents are exchanged. Each of these foundational information exchanges can be made electronically using the standard EDI requisition, purchase order, invoice, and remittance documents. But beyond basic transactions, firms can create unique competitive advantages by exchanging additional information (Dyer and Singh 1998). Empirical research has found that in an EDI-enabled relationship there is a positive relationship between production schedule sharing and supplier performance (Walton and Marucheck 1997). In a case-based research example, a retailer that provides point-of-sale data to its manufacturer can create processes whereby the manufacturer monitors demand and automatically replenishes the retailer stock (Lee et al. 1999). The retailer can improve performance in the form of lower stockouts and higher inventory turnover. The manufacturer can better plan production and balance safety stock across the supply chain. The exchange of additional information can support the development of processes that increase the efficiency of the supply chain. Thus,
H2: Information exchange diversity is positively associated with firm performance.
DATA COLLECTION AND RESEARCH METHODOLOGY
This study tests the effects of information exchange characteristics on firm performance. The data used in this study were gathered as part of a larger study of how information exchange affects supply chain relationships. Data for the measurement of B2B information exchange is gathered from an electronically mediated industrial exchange network. This proprietary data has been made available by one of the world's largest providers of B2B integration services. Additional data on firm characteristics and performance are gathered from Standard and Poor's Compustat database.
Since these data are collected through an electronic exchange network that uses EDI standards for the coding of the transactions, distinct information exchange characteristics can be captured at the firm level. There are two EDI standards used to format data for exchange. The EDI integrator supplying data for this study supports both ANSI and EDIFACT formatted EDI messages. Each of these two standards is accepted in practice; however, firms that use EDI can choose which format to implement or may implement a combination of both standards. EDI documents can be exchanged through proprietary telecommunications networks or through existing Internet connects. The provider of these data supports both a proprietary network and AS2-enabled Internet transactions.
By using EDI transactions as a measure of firm information exchange, this study captures specific measures of information exchange characteristics. First, information exchange volume (INFO_VOLUME) is measured based on the number of electronic business documents sent and received by the firm. The EDI integrator that manages the electronic exchange tracks the number of transactions processed on a monthly, quarterly, and yearly basis. The quarterly measures are used to match with the quarterly firm data gathered from Compustat.
Firms must choose not only how much information is exchanged with their trading partners but also what information is exchanged. At a base level, firms may simply send electronic purchase orders to their suppliers and receive back an electronic invoice. While this rudimentary application of EDI improves the cost efficiency of the purchasing cycle, far more information can be exchanged through EDI and used to support additional supply chain initiatives. A study of Campbell Soup Company identified that additional information was exchanged that allows Campbell to monitor end customer demand such that the company centrally manages the replenishment of its retailers (Lee et al. 1999). Information diversity in an EDI environment is a measure of how many different types of information are exchanged by the firm (Massetti and Zmud 1996). Hundreds of business documents are included in the EDI document formats within the ANSI and EDIFACT standards and each is identified by a unique transaction code. Some examples of the transaction codes and their descriptions are provided in Table 1. The information diversity measure (INFO_DIVERSITY) is operationalized by counting the unique transaction codes sent or received by a firm during the quarter. The quarterly firm diversity measures represent average values for the firm across the two-year study period. The averaging technique alleviates variances created by seasonality in the transaction characteristics.
Additional variables are collected from Standard and Poor's Compustat database. The dependent variable in this study is inventory turnover (INVENTORY_TURNOVER). Inventory turnover is often used in supply chain management empirical research due to its proximity to the actual physical goods that move in the supply chain (Kalwani and Narayandas 1995; Droge and Germain 2000; Gaur et al. 2005a). High-level firm performance measures such as ROI, ROA, and stock price are often too far removed from the operations of the firm to measure the effectiveness of supply chain strategies without the noise of other firm activities.
Studies have recognized that larger firms experience economies of scale in their inventory turnover such that there is a positive correlation between inventory turnover and firm size (Gaur et al. 2005a). Controlling for firm size can minimize these confounding effects. Similar studies have measured firm size using total assets, sales, and the number of employees (Zhu and Kraemer 2002). For the purposes of the study, the measure of firm sales reported quarterly in the Compustat database is used as the measure of firm size (FIRM_SIZE). The variables used in the model are summarized in Table 2.
Research Methodology
The goal of this study is to expand the understanding of the relationship between interfirm inventory exchange characteristics and firm performance. This is achieved by focusing on the two hypothesized aspects of information exchange and testing the hypotheses using a large sample of manufacturing firms. An ordinary least squares (OLS) regression is used to estimate the relationship between the firm information exchange characteristics (volume and diversity) and firm performance (inventory turnover). The model is expressed in Equation 1.
[INVENTORY_TURNOVER.sub.i] = [[beta].sub.0] + [[beta].sub.1][INFO_VOLUME.sub.i] + [[beta].sub.2][INFO_DIVERSITY.sub.i] + [[beta].sub.3][FIRM_SIZE.sub.i] (1)
where i is the firm level observation
Sample
The exchange network used to support this research provides EDI services for over 100 technology champion firms. A technology champion firm in this context is a business that has formed a contractual relationship with the network service provider to electronically route its EDI transactions. As part of the agreement, the network service provider establishes relationships with each of the technology champion firm's trading partners to format their business documents using EDI. Trading partners of the technology champion firm include both customers and suppliers of the focal firm. Firms were identified for inclusion in the study based on their level of participation in the exchange network. Since there is competition in the network service provider market, some technology champion firms split their electronic traffic among multiple providers. Firms that were known to not have 100 percent of their electronic information exchange through the network were eliminated from the study. The study is also restricted to publicly traded firms so that reliable performance measures and firm characteristics could be collected from Compustat. Additionally, a manufacturing focus was adopted to minimize the variance created by including multiple echelons of the supply chain. The resulting twenty-three firms are included in the sample.
Within the manufacturing segment there is variation in the information exchange attributes and firm performance. Table 3 provides a stratification of the key measures by their two-digit SIC codes. The number of firms representing each segment is noted in column two. Some manufacturing segments are represented by data from multiple individual firms. The twenty-three firms in this sample represent nine distinct manufacturing segments. The average inventory turnover is approximately twelve but ranges from four for the Chemicals and Allied Products firms to forty-three for firms in Industrial and Commercial Machinery.
Descriptive Statistics of the Data
Descriptive statistics for each variable are included in Table 4. The firm performance variable (INVENTORY_TURNOVER) has a relatively large standard deviation given the mean of the sample. Even with a focus on manufacturing firms, there is large variance among firms. Firms vary in the volume of information exchanged, as shown by the large variance in the average quarterly volume of information exchanged (INFO_VOLUME). The mean of information exchange volume indicates that, on average, firms in the sample exchange 0.38 million documents per quarter. Quarterly information exchange volume ranges from twenty-six thousand business documents to over one million. Similarly, the firms in this study vary in the diversity of information exchanged (INFO_DIVERSITY). The average number of document types exchanged by firms in the sample is twenty-two but ranges from a low of nine to high of thirty-seven. These values represent an average for the firm across two years of data so that variations based on seasonality would not affect the analysis.
Due to a violation of the assumption of normality in the raw data, the inventory turnover variable (INVENTORY_TURNOVER) was modified using a natural log transformation. The results of the OLS regression are reported using the logged dependent variable. Additionally. a pair-wise correlation table is provided to evaluate the relationships between the explanatory variables. Table 5 provides the specific correlations and their statistical significance. The only statistically significant correlation is the positive relationship (0.3983) between the size of the firm (FIRM_SIZE) and the volume of information exchanged (INFO_VOLUME). Researchers suggest that correlations between explanatory variables will not bias the coefficient estimates if all pair-wise correlations are below 0.50 (Dielman 2005). It is not surprising that larger firms would exchange more information than smaller firms. The measure of firm size is included in the regression model to control for the effects of firm size on firm performance. By including firm size as a control variable, the remaining effects of information exchange characteristics on firm performance can be specifically addressed by the model results. While the relationship between firm size and information exchange volume is statistically significant, it is not so large that it will cause a bias in the regression results.
RESULTS
The OLS model results show a good statistical fit for the data based on the statistically significant results of the F-test. Additionally, the explanatory power of the model is high based on an R-square of 0.5459. This suggests that over 50 percent of the variance in average inventory turnover is explained by the model. Complete model fit statistics are provided in Table 6. Additional details on the development and interpretation of the OLS regression are provided in Appendix A.
The coefficient for the measure of firm quarterly information exchange volume (INFO_VOLUME) is statistically significant and positive. This result provides support for Hypothesis 1, which states that the volume of information exchange is positively associated with firm performance. The coefficient for the information diversity measure (INFO_DIVERSITY) is negative and statistically significant. This result does not support Hypothesis 2, which states a positive relationship between information diversity and firm performance. The surprising result for Hypothesis 2 will be addressed further in the discussion and conclusion section. Full results from the OLS regression are provided in Table 6.
DISCUSSION AND CONCLUSION
This study uses unique quantitative measures of information exchange characteristics and finds an important result showing a strong relationship between the volume of information exchange and firm performance. Prior literature has attempted to address this issue using modeling, perceived measures, or single firm analyses. This is the first study to address the issue using quantitative archival data from a large number of firms in one sector of the economy.
The positive result for Hypothesis 1 supports the results of prior literature on the relationship between information exchange volume and firm performance. Although not surprising, it is useful to confirm the positive effects of information exchange within the context of an electronically-mediated network using archival information exchange data. Managerially, this result is interesting since it supports current practices of expanding the electronic gathering and exchanging of information across the supply chain. At an extreme level, Wal-Mart integrates all suppliers by requiring a mix of RFID, EDI, and Web-based technologies (Retail Link[R]) to mandate electronic information exchange.
The surprising result for Hypothesis 2 requires further analysis. While the hypothesized positive relationship between information diversity and firm performance was not supported, the negative and statistically significant result is important. Firms in this sample that exchanged more types of information with trading partners experienced lower inventory turnover. Since this sample includes only firms that are identified as manufacturers, these results may indicate that increased information diversity in manufacturing supply chains allows trading partners to act opportunistically. In a study of grocery retailers, it was shown that suppliers were more likely to act opportunistically when supply chain dependence or power imbalance exist within a relationship (Morgan et al. 2007). As more information is available to trading partners, they may choose to take actions that are not in the best interest of the manufacturer. This situation may be affected by the level of asset specificity of products transacted with manufacturers. If a manufacturer requires a very specific input and a supplier has additional information, the supplier may be able to raise prices or hold up supply in order to extract additional concessions.
An alternative explanation of this surprising result has been identified in related research. It has been suggested that as firms take advantage of the technology that allows them to exchange more types of business documents, there is no guarantee that the firms are able to integrate that information into their internal or external processes (Clarke 1992; Massetti and Zmud 1996). This situation may be explained by the relationship between integration and communication (Stank et al. 2005). This research suggests that in order for information to add value, it must be exchanged in an environment of integrated processes such that the information can be leveraged for the benefit of the firm. If the data are exchanged but not actually used in the decision making process, the data do not provide value for the firm.
[FIGURE 1 OMITTED]
Closely related to research on information integration is the study of information overload. Researchers using a simulation methodology have noted that information overload can adversely affect firm performance (Steckel et al. 2004). This suggests that the additional information types exchanged are not only failing to contribute to performance but may actually be detrimental to performance by causing noise in the channel.
Managerially, the negative relationship between information diversity and inventory turnover is interesting. The results and existing research would suggest that managers use caution pertaining to decisions of what information to exchange, with whom to exchange information, and how the information is integrated into business processes.
Alternate Non-linear Model Specification
The potential exists for a non-linear relationship between the explanatory variables and firm performance. A non-linear relationship would occur if, for example, in the case of information exchange volume, firm performance was high for both low levels of information exchange and high levels of information exchange but low for intermediate volumes of information exchange (see Figure 1). This situation is appropriately described as a u-shaped relationship. Similarly, an inverted u-shaped relationship would occur when firm performance was low for both low and high volumes of information exchange but high for intermediate volumes of information exchange.
The u-shaped relationship is easily accommodated in regression analysis by including the square of the variable as part of the model. Alternate Model A includes the squared variable for information exchange volume (INFO_VOLUME_SQ). Similarly, Alternate Model B includes the squared variable for information diversity (INFO_DIVERSITY_SQ). The results for the alternative non-linear models are provided in Table 7.
The statistical fit of the alternate models to the data is acceptable based on the F-statistic. The R-square measures indicate that both models explain a large portion of the variation in firm performance. However, the coefficient estimates are not statistically significant for the squared variables. Additionally, the Adjusted R-Square indicates that including the squared variables in the two alternate models does not improve the explanatory power of the alternate models over the original model. These results suggest that the original linear model is better than the non-linear specification for understanding the relationship between information exchange characteristics and firm performance.
LIMITATIONS AND FUTURE RESEARCH
This article fills a gap in the literature by specifically using archival data and moving away from the often-used perceived measures of information exchange; however, the dataset is limited in its available measures. As noted in the discussion, the data provide visibility into the specific information exchange characteristics but do not capture how the information is being integrated into the business processes. Additional research is needed to simultaneously consider the information exchange characteristics and the level of interfirm process integration.
Further investigation is needed to evaluate the unexpected negative relationship of information diversity and inventory turnover. First, this model may be tested in other sectors of the supply chain to clarify whether the results are unique to the manufacturing sector. Second, additional qualitative data could be collected through a survey instrument to capture performance, integration, and information exchange characteristics variables.
The performance variable in this article, inventory turnover, captures one dimension of supply chain performance. Additional supply chain performance measures such as responsiveness and innovation may extend the understanding of how information exchange characteristics affect performance.
This article makes a significant contribution to the research on information exchange in supply chain relationships. Using a unique archival dataset, the positive effects of information exchange volume are validated. The article then opens additional discussions by providing new insights into the existence of negative outcomes related to information diversity.
APPENDIX
Regression Analysis Overview
This study uses regression analysis to test the relationship between firm performance and specific measures of information exchange. Regression analysis is a statistical technique used to model the relationship between a dependent variable (firm performance) and a series of independent variables (information exchange and firm characteristics). The independent variables in a regression analysis are often called the "explanatory" variables because they are used to explain the variation in the dependent variable. In the case of this study, the regression analysis is evaluating the variation in firm performance by considering the variation in information exchange characteristics. In other words, the researcher is statistically testing whether there is a relationship between the variation in firm performance and the variation in how firms exchange information with their trading partners.
Ordinary Least Squares Regression
The ordinary least squares (OLS) regression is a basic form of regression analysis. When there are two or more explanatory variables, the relationship cannot be simply drawn on a two-dimensional graph but it can still be modeled as a linear equation that best "fits" the data. The equation that best describes the relationship between the dependent variable (INVENTORY_TURNOVER) and the explanatory variables (INFO_VOLUME, INFO_DIVERSITY, and FIRM_SIZE) is written as shown in Equation 1. The statistical software then evaluates values for the coefficients ([[beta].sub.0], [[beta].sub.1], [[beta].sub.2], [[beta].sub.3]) so that the squared differences between the estimated equation and the actual data are minimized (Dielman 2005). The regression routines available in software packages such as EXCEL, SPSS, STATA, and SAS estimate specific values for each of the coefficients in the equation based on the dataset provided. The sign and magnitude of the estimated coefficients can then be interpreted by the researcher to understand the relationship between the dependent variable and each of the explanatory variables. For this study, the coefficient estimates are provided in column 2 of Table 6. Along with the individual coefficient estimates, the statistical software provides additional measures to test how well the estimated model fits the data. Two of these measures are the F-statistic and the R-square value, which are provided in Table 6. The F-statistic is a comparative measure of model fit. By comparing the calculated F-statistic with the critical value on an F table (provided in most statistics textbooks), the researcher can verify that the estimated model is useful in explaining the variation in the dependent variable at a 95 percent level of confidence. The R-square value provides a measure of how much of the variation in the dependent variable is explained by the model. In this study, the proposed model explains 55 percent ([R.sup.2]=0.5459) of the variation in the dependent variable (INVENTORY_TURNOVER).
Dataset
The dataset for this study is well suited for regression analysis using OLS. Each of the twenty-three individual observations represents information about one firm in the study. The observation includes the inventory turnover for the firm and its information exchange characteristics for the same time period. Inventory turnover is the measure of firm performance that serves as the dependent variable in this study. The information exchange characteristics for the firm during that same period are included in the observation as the explanatory variables. It is expected that there are factors beyond the scope of this study that affect the differences between firms' inventory turnovers. If the researcher can identify measures of those factors, the measures can be included as control variables in the model. Including additional control variables can decrease the unexplained variation in the dependent variable. In this study, firm size was recognized as a factor that influences inventory turnover and is included as a control variable in the model.
When using an OLS regression for the statistical analysis of data, there are conditions and assumptions that must be met in order for the output to be valid for hypothesis testing. The following discussion provides details of the specific validation procedures used in this study.
For regression models with multiple explanatory variables, it is expected that the explanatory variables are related to the dependent variable. In statistics, this relationship between variables is referred to as correlation and can vary in direction (positive or negative) and magnitude (between -1 and 1). If, however, the explanatory variables are strongly related to each other, the regression model will fail to provide accurate coefficient estimates. This correlation within the explanatory variables is referred to as multicollinearity. Table 5 provides measures of the correlations between pairs of explanatory variables. For the data used in this study, the pairwise correlations are relatively small (0.3983 correlation between FIRM_SIZE and INFO_VOLUME). A rule of thumb used by researchers is that multicollinearity is not a serious problem when all pairwise correlations are below 50 percent (Dielman 2005).
As noted earlier, the variation in the dependent variable not explained by the model is referred to as error. The error can be specifically identified for each observation (firm) by calculating the difference between the actual firm performance and the firm performance estimated by the regression coefficients. It is expected that across all twenty-three observations, the error is normally distributed and the variance in error is constant across all firms. The distribution of error is easily tested by plotting the standardized error versus the standardized predicted values. For data that are normally distributed, 95 percent of the plotted values should be between -2 and +2 (that is, within two standard deviations of the mean). The data in this study did not produce normally distributed values because the dependent variable (INVENTORY_TURNOVER) was not normally distributed. This situation was corrected by taking the natural log of the dependent variable and rerunning the regression (Dielman 2005).
The other key assumption is that the error variance is equal across all observations. In statistics, this condition is referred to as heterogeneity. Heterogeneity can be verified by using the Breusch-Pagan/Cook-Weisburg test. This test compares the error variance for each observation and its predicted value based on a chi-square test statistic. Again, this test statistic can be compared with the critical value from a chi-square table to verify that the range of variance is within a 95 percent tolerance. If the error variance was not equal across the observations, then the estimated coefficients would not be useful for explaining the relationship between the dependent and explanatory variables at all levels of the dependent variable.
Additional information about creating and interpreting the output from an OLS regression can be found in most introductory business statistics texts, such as Applied Regression Analysis (Dielman 2005), which has been referenced in this article.
REFERENCES
Allen, Benjamin J., Michael R. Cram, and Charles D Brauschweig (1992), "The U.S. Motor Carrier Industry: The Extent and Nature of EDI Use," International Journal of Physical Distribution and Logistics Management, Vol. 22, No. 8, pp. 27-34.
Angulo, Andres, Heather Nachtmann, and Matthew A. Waller (2004), "Supply Chain Information Sharing in a Vendor Managed Inventory Partnership." Journal of Business Logistics, Vol. 25, No. 1, pp. 101-120.
Bakos, Yannis, and Erik Brynjolfsson (1993), "From Vendors to Partners: Information Technologies and Incomplete Contracts in Buyer Seller Relationships," Journal of Organizational Computing, Vol. 3. No. 3. pp. 301-329.
Cachon, Gerard P., and Marshall Fisher (2000). "Supply Chain Inventory Management and the Value of Shared Information," Management Science, Vol. 46, No. 8, pp. 1032-1048.
Cachon. Gerard P., and Martin A. Lariviere (2001), "Contracting to Assure Supply: How to Share Demand Forecasts in a Supply Chain," Management Science, Vol. 47, No. 5, pp. 629-646.
Clarke, R. (1992), "A Contingency Model of EDI's Impact on Industry Sectors," Journal of Strategic Information Systems. Vol. 1, No. 3, pp. 143-151.
Cooper, Martha C., Lisa M. Ellram, John T. Gardner, and Albert M. Hanks (1997), "Meshing Multiple Alliances," Journal of Business Logistics, Vol. 18, No. 1, pp. 67-89.
Cruijssen, Frans, Wout Dullaert, and Hein Fleuren (2007). "Horizontal Cooperation in Transport and Logistics: A Literature Review," Transportation Journal, Vol. 46, No. 3, pp. 22-39.
Crum, Michael R., Deborah A. Johnson, and Benjamin J. Allen (1998), "A Longitudinal Assessment of EDI Use in the U.S. Motor Carrier Industry." Transportation Journal, Vol. 38, No. 1, pp. 15.
Daugherty, Patricia J., Matthew B. Myers, and R. Glenn Richey (2002), "Information Support for Reverse Logistics: The Influence of Relationship Commitment." Journal of Business Logistics. Vol. 23, No. 1, pp. 85-106.
Dielman, Terry E. (2005). Applied Regression Analysis. Bemont, CA: Brooks/Cole Thomson Learning.
Droge, Cornelia. and Richard Germain (2000), "The Relationship of Electronic Data Interchange with Inventory and Financial Performance," The Journal of Business Logistics, Vol. 21, No. 2. pp. 209-230.
Dyer, Jeffery H. and Harbir Singh (1998), "The Relational View: Cooperative Strategy and Sources of Interorganizational Competitive Advantage," Academy of Management Review, Vol. 23, No. 4, pp. 660-679.
Gaur, Vishal, Marshall L. Fisher, and Ananth Raman (2005a). "An Econometric Analysis of Inventory Turnover Performance in Retail Services." Management Science. Vol. 51, No. 2. pp. 181-194.
Gaur, Vishal, Avi Giloni, and Sridhar Seshadri (2005b), "Information Sharing in a Supply Chain under ARMA Demand," Management Science. Vol. 51, No. 6, pp. 961-969.
Iacovou, Charalambos L., Izak Benbasat, and Albert S. Dexter (1995), "Electronic Data Interchange and Small Organizations: Adoption and Impact of Technology." MIS Quarterly, Vol. 19, No. 4, pp. 465-485.
Kalwani, Manohar U., and Narakesari Narayandas (1995). "Long-term Manufacturer-supplier Relationships: Do They Pay Off for Supplier Firms?" Journal of Marketing, Vol. 59, No., pp. 1-16.
Lee, Hau L., V. Padmanabhan, and Seungjin Whang (1997), "Information Distortion in a Supply Chain: The Bullwhip Effect," Management Science, Vol. 43, No. 4, pp. 546-558.
Lee, Ho Geun, Theodore Clark, and Kar Yan Tam (1999), "Research Report. Can EDI Benefit Adopters?" Information Systems Research, Vol. 10, No. 2, pp. 186-195.
Lieb, Robert, and Karen Butner (2007), "The North American Third-party Logistics Industry in 2006: The Provider CEO Perspective," Transportation Journal, Vol. 46. No. 3. pp. 40-52.
Machuca, Jose A.D., and Rafael P. Barajas (2004). "The Impact of Electronic Data Interchange on Reducing Bullwhip Effect and Supply Chain Inventory Cost," Transportation Research Part E, Vol. 40, No. 3, pp. 209-228.
Massetti, Brenda, and Robert W. Zmud (1996), "Measuring the Extent of EDI Usage in Complex Organizations: Strategies and Illustrative Examples," MIS Quarterly, Vol. 20. No. 3, pp. 331-345.
Moberg, Christopher R., Bob D. Cutler, Andrew Gross, and Thomas W. Speh (2002), "Identifying Antecedents of Information Exchange within Supply Chains," International Journal of Physical Distribution & Logistics Management, Vol. 32. No. 9/10, pp. 755-770.
Morgan, Neil A., Anna Kaleka, and Richard A. Gooner (2007), "Focal Supplier Opportunism in Supermarket Retailer Category Management," Journal of Operations Management, Vol. 25, No. 2, pp. 512-527.
Mukhopadhyay, Tridas, and Sunder Kekre (2002), "Strategic and Operational Benefits of Electronic Integration in B2B Procurement Processes," Management Science, Vol. 48, No. 10, pp. 1301-1313.
Mukhopadhyay, Tridas, Sunder Kekre, and Suresh Kalathur (1995), "Business Value of Information Technology: A Study of Electronic Data Interchange," MIS Quarterly, Vol. 19, No. 2, pp. 137-156.
Porter, Michael E., and Victor E. Millar (1985), "How Information Gives You Competitive Advantage," Harvard Business Review, Vol. 63, No. 4, pp. 149-160.
Rozenzweig, Eve D., Aleda V. Roth, and James W Jr. Dean (2003), "The Influence of an integration Strategy on Competitive Capabilities and Business Performance: An Exploratory Study of Consumer Products Manufacturers," Journal of Operations Management, Vol. 21, No. 4, pp. 437-456.
Sanders, Nada R., and Robert Premus (2002), "IT Applications in Supply Chain Organizations: A Link between Competitive Priorities and Organizational Benefits," Journal of Business Logistics, Vol. 23, No. 1, pp. 65-83.
Srinivasan, Kannan, Sunder Kekre, and Tridas Mukhopadhyay (1994), "Impact of Electronic Data Interchange Technology on JIT Shipments," Management Science, Vol. 40, No. 10, pp. 1291-1304.
Stank, Theodore P., R. Beth Davis, and Brian S. Fugate (2005), "A Strategic Framework for Supply Chain Oriented Logistics," Journal of Business Logistics, Vol. 26, No. 2, pp. 27-46.
Steckel, Joel H., Sunil Gupta, and Anirvan Banerji (2004). "Supply Chain Decision Making: Will Shorter Cycle Times and Shared Point-of-Sale Information Necessarily Help?" Management Science, Vol. 50, No. 4, pp. 458-464.
Vickery, Shawnee K, Jayanth Jayaram, Cornelia Droge, and Roger Calantone (2003), "The Effects of an Integrative Supply Chain Strategy on Customer Service and Financial Performance: An Analysis of Direct versus Indirect Relationships," Journal of Operations Management, Vol. 21, No., pp. 523-539.
Walton, Steve V., and Ann S. Marucheck (1997), "The Relationship between EDI and Supplier Reliability," International Journal of Purchasing and Materials Management, Vol. 33, No. 3, pp. 30-35.
Whipple, Judith M., Robert Frankel, and Patricia J. Daugherty (2002), "Information Support for Alliances: Performance Implications," Journal of Business Logistics, Vol. 23, No. 2, pp. 67-82.
Zhu, Kevin, and Kenneth L. Kraemer (2002), "E-Commerce Metrics for Net-enhanced Organizations: Assessing the Value of E-commerce to Firm Performance in the Manufacturing Sector," Information Systems Research, Vol. 13, No. 3, pp. 275-295.
Zsidisin, George A., Douglas M. Voss, and Matt Schlosser (2007), "Shipper-carrier Relationships and their Effect on Carrier Performance," Transportation Journal, Vol. 46, No. 2, pp. 5-18.
Mr. Porterfield is assistant professor, Department of eBusiness and Technology Management, College of Business and Economics, Towson University, Towson, Maryland 21252; e-mail tporterfield@towson.edu. The author would like to thank Brian Lowell from the University of Maryland--University College for his assistance in data collection and validation as well as Joseph Bailey and Phil Evers from the University of Maryland--College Park for their guidance in the development of this research stream.
Table 1. Examples of EDI Transaction Types on the Network
EDI Transaction
Standard Code Description
ANSI 511 Requisition
ANSI 810 Invoice
ANSI 850 Purchase Order
ANSI 856 Shipping Notice
ANSI 888 Item Maintenance
EDIFACT INVOIC Invoice
EDIFACT REMADV Remittance Advice
EDIFACT ORDCHG Purchase Order Change
EDIFACT SLSFCT Sales Forecast
EDIFACT PRODAT Product Data Message
Table 2. Measures and Data
Variable Definition
Dependent Variable
[INVENTORY_ The ratio of a firm's cost of goods sold to
TURNOVER.sub.i] the firm's inventory value. The average is
calculated across two years of data.
Independent Variables
[INFO_VOLUME.sub.i] The average number of information exchanges
with trading partners through an electronic
intermediary. The average is based on
quarterly observations averaged across two
years of data.
[INFO_DIVERSITY.sub.t] The average number of unique information
types exchanged through an electronic
intermediary. The value is based on quarterly
observations averaged across two years of
data.
Control Variables
[FIRM_SIZE.sub.i] Firm average quarterly sales. The average is
calculated across two years of data.
Where i is the firm level observation
Table 3. Summary of Participating Firms by 2-Digit SIC Code
Average
Quarterly
Firm Inventory
SIC-2 Counts Description Turnover
20 1 Food and Kindred Products 4.10
25 1 Furniture and Fixtures 13.29
26 1 Paper and Allied Products 4.66
28 7 Chemicals and Allied Products 3.79
34 1 Fabricated Metal Products 5.52
35 4 Industrial and Commercial Machinery 43.38
36 2 Electronics and Electrical Equipment 5.64
37 4 Transportation Equipment 9.37
38 2 Measuring, Analyzing and 4.63
Controlling Equipment
Average Average
Quarterly Quarterly Average
Information Information Net
SIC-2 Volume * Types Sales **
20 0.27 30.12 4,732.13
25 0.13 11.25 619.93
26 0.38 28.63 4,780.38
28 0.35 37.33 6,451.26
34 O.12 16.88 1,676.17
35 0.84 19.54 5,925.85
36 0.26 25.06 2,742.54
37 0.32 21.40 5,769.93
38 0.11 18.36 2,242.04
* average information volume is stated in millions of transactions
** average net sales are stated in millions of dollars
Table 4. Descriptive Statistics
Mean S.D. Min Max
INVENTORY_TURNOVER 12.42 231.89 1.28 85.09
INFO_VOLUME * 0.38 0.37 0.03 1.46
INFO_DIVERSITY 22.16 8.79 9.25 37.33
Control Variable
FIRM SIZE ** 4,944.34 3,696.31 619.93 12,839.13
23 firms are included in this dataset (n=23)
* information volume is expressed as millions of transactions per
quarter
** firm size is based on millions of dollars of net sales
Table 5. Summary of Participating Firms by 2-Digit SIC Code
INFO_VOLUME INFO_DIVERSITY FIRM_SIZE
INFO VOLUME 1.000
INFO-DIVERSITY 0.2077 1.000
0.3416
FIRM-SIZE 0.3983# 0.3339 1.000
0.0598# 0.1195
statistically significant pairwise correlations are highlighted in
bold and italics the first value in each cell is the correlation and
the second value is the p-value measure of statistical significance
Note: Significant pairwise correlations is indicated with #.
Table 6. OLS Regression Results
Model 1
(log)INVENTORY_TURNOVER Coef. P<|t| sig
(Std Err)
Explanatory Variables
INFO-VOLUME 2.06 0.000 ***
(0.4530)
INFO DIVERSITY -0.0326 0.094 *
(0.0185)
constant 1.9851 0.000 ***
(0.4236)
Control Variables
FIRM-SIZE -0.00004 0.424 ns
(0.00005)
F-Statistic 7.62
Probability>F 0.0015
R-square 0.5459
Observations 23
*** <.01 ** <.05 * <.1 significance level
Table 7. Alternate Model Specification Results
Altenate Model A
Coef.
(log)INVENTORY_TURNOVER (Std Err) P<|t| sig
Explanatory Variables
INFO_VOLUME -0.6208 0.729 ns
(1.7641)
INFO_DIVERSITY -0.0071 0.772 ns
(0.0241)
INFO_VOLUME_SQ 1.9583 0.134 ns
(1.2465)
INFO_DIVERSITY_SQ
constant 1.9133 0.000 ***
(0.4107)
Control Variables
FIRM_SIZE -0.00004 0.377 ns
(0.00004)
F-Statistic 6.77
Probability>F 0.002
R-square 0.601
Adjusted R-square 0.512
Observations 23
Altenate Model B
Coef.
(log)INVENTORY_TURNOVER (Std Err) P<|t| sig
Explanatory Variables
INFO_VOLUME 2.08 0.000 ***
(0.4665)
INFO_DIVERSITY -0.0754 0.565 ns
(0.1286)
INFO_VOLUME_SQ
INFO_DIVERSITY_SQ 0.0009 0.740 ns
(0.0027)
constant 2.4327 0.099 *
(1.3997)
Control Variables
FIRM_SIZE -0.00004 0.410 ns
(0.00005)
F-Statistic 5.47
Probability>F 0.0046
R-square 0.5488
Adjusted R-square 0.4485
Observations 23
*** <.01 **<.05 * <.1 significance level
In addition, make sure to read these articles:
Email
Digg It
del.icio.us
Related Resources
Press Releases
- FastParts.com Selects webMethods B2B to EnableImproved Supply Chain Management for Global ExchangeMembers.
- Pressman Toy Uses EXTOL Solution to Maximize Supply Chain Integration After Major Global Manufacturing and Distribution Changes.
- China Entercom and GXS Partner to Launch B2B e-Commerce Hub for China Automotive Industry; China e-Auto Hub to Accelerate Supply Chain Automation in China's Booming Automotive Sector.
Business Resources
Sponsored Links
Site Map | Contact Us | FAQs | About Us | Media Kit | Reprints | RSS Directory | Sign Up for Free Newsletters | Disclosure Policy
Copyright © 1999 - 2008 AllBusiness.com, Inc. All rights reserved.
No part of this content or the data or information included therein may be reproduced,
republished or redistributed without the prior written consent of AllBusiness.com.
Use of this site is governed by our Copyright and Intellectual Property Policy, Terms of Use Agreement and Privacy Policy.
© COPYRIGHT 2008 American Society of Transportation and Logistics, Inc.
You may not repost, republish, reproduce, package and/or redistribute the content of this page,
in whole or in part, without the written permission of the copyright holder.
Online Business Database | Online Business Information | Sales and Marketing Solutions | Business Mailing Lists
AllBusiness.com, its parent company D&B, and its affiliates.
