Internet Technology Use Across the Food Industry Supply Chain.

Abstract

An investigation is conducted of Internet usage across the food industry supply chain to determine whether distributors are currently using or are expected to use Internet technology at the same rate as retailers and manufacturers. Our findings show that there are no significant

differences in Internet usage between distributors and retailers or between distributors and manufacturers, after controlling for food industry sector and firm size. No significant differences were found in anticipated Internet use, either. We did find, however, significant differences in Internet usage according to firm size. To the extent that distributors are, on average, smaller than manufacturers, as was the case with our sample, they may still be at some risk in falling behind (at least manufacturers) in Internet technology usage to support supply chain activities.

The consensus that emerges across a wide swath of industries is that "pure middlemen" -- companies that act as little more than go-betweens and order-takers -- are in serious trouble as a result of the Internet. But distributors and resellers that add product value and customer service along the way can use the Web to enhance their positions and improve their businesses. Many are doing just that -- along with a crop of new players becoming business intermediaries in cyberspace (Wilder, 1997).

There is little doubt that the Internet has changed and will continue to change the way business is conducted. The Internet has allowed the creation of new, direct, business-to-consumer (B2C) channels that have made possible the direct marketing of products and services from suppliers to end users. It has also provided new means for businesses to interact with other businesses (i.e., B2B channels), for example by disseminating product information through on-line, electronic catalogues. Although the ultimate size of the Internet economy is unknown, according to a number of estimates, by 2004 the Internet could account for close to $2 trillion in economic activity in the United States--roughly one-quarter of the national economy (Fry, 1999).

A firm's ability to compete in the future may depend on its willingness and ability to use Internet technology. According to the results of a survey of international freight forwarders by Murphy and Daley (2000), Internet technology provides a number of important benefits to adopting firms. These benefits include quick access to information; improved tracking and expediting; improved communications with customers; better customer service; and reduced costs. Conversely, firms that fail to use Internet technology, or lag behind their competitors in Internet technology adoption, may find themselves either playing "catch-up" or out of business.

The business risks arising from the failure to use Internet technology may be especially large for intermediaries in the supply chain. Since the Internet facilitates the direct marketing of products from suppliers to end users, intermediaries risk being cut out of the supply chain, or "disintermediated" (Griffith and Palmer, 1999). This phenomenon has had wide coverage in the trade literature. [1] According to industry sources, the best defense against disintermediation by supply chain intermediaries may be to invest in Internet technology; to become the prime provider of value-added web-based information and other services, thereby "re-intermediating" the supply chain (Wilder 1997; information Week 1998; Madsen 1999; Skillen 2000).

This article examines current and expected use of Internet technology across the supply chain in the food industry. The food industry is chosen for study because of its well established supply chain and its leadership in inter-organizational supply chain initiatives, such as efficient consumer response (Fisher et al., 1994; Barrett and Konsynski, 1982). We compare the use of Internet technology among manufacturers, distributors, and retailers, to see whether the distributors, prime intermediaries in the food industry supply chain, use or anticipate to use Internet technology at the same rate as retailers and manufacturers.[2] If use of Internet technology by distributors is found to lag behind adoption by retailers and manufacturers, then the food industry distributors may be under threat of disintermediation, or at least replacement by more web-savvy intermediaries. The results provide significant information to both academics and practitioners interested in studying how Internet technology is affecting or m ay affect supply chain structures.

The rest of the article is structured as follows: The next section reviews relevant literature on technology adoption and use. Section three describes the research methodology used. Section four discusses our regression models and the results from the estimations. Finally, section five presents conclusions, discusses the managerial implications of the results, and identifies the potential for future research.

LITERATURE REVIEW

A considerable body of published research has been conducted on the adoption and use of technology by businesses. Much of the transportation, logistics, and supply chain management literature relevant to this article relates to the adoption and use of electronic data interchange (EDI) and other inter-organizational systems, designed to facilitate communication and document exchange among supply chain members. The papers revolve around a number of themes, including the use of EDI, the benefits to be derived from EDI, and the motivations and conditions promoting the use of EDI. First, with respect to EDI usage, Crum and Allen (1997) and Crum, Johnson, and Allen (1998) assessed the growth in EDI use in the U.S. motor carrier industry between 1990 and 1996. Using results from a survey of motor carriers, the authors found that the percentage of respondents using EDI increased from 29.3 in 1990 to 37.4 in 1996 (Crum and Allen, 1997). The number of firms employing EDT increased in large part because of the benefits derived from EDI usage.

Mackay and Rosier (1996) examined the benefits to be derived from EDI adoption and use. Using results from a survey of Australian automotive parts suppliers, the authors found that improved productivity, clerical staff and administrative cost savings, increased data accuracy, and customer service were most closely associated with the provision of EDI benefits. Droge and Germain (2000) surveyed U.S. members of the Council of Logistics Management (CLM) and found that EDI usage contributed to improved financial performance, but was also associated with higher inventory levels, rather than inventory stock reductions. Stank, Crum, and Arango (1999) examined the benefits to be derived from EDI usage and other forms of inter-organizational communication in the U.S. food industry. The authors found that EDI and other forms of inter-firm coordination contributed to decreased inventory levels (contradicting Droge and Germain [2000]), decreased order cycle time, and decreased order cycle variance. Finally, Walton and Ma rucheck (1997) surveyed company buyers and EDI managers to assess the relation ship between EDI use and supplier reliability. They found that one measure of supplier reliability, delivery performance, was positively associated with EDI investment and systems integration.

Williams (1994), Williams, Magee, and Suzuki (1998), Daugherty, Germain, and Droge (1995), and Walton (1994) examined the variables likely to lead to EDI adoption. Using surveys of CLM members, the four studies found that a number of organizational variables, such as firm size and structure, the decentralization of EDI adoption decisions, and formal benchmarking procedures influenced EDI adoption. As well, competitive variables, including the desire to stay competitive, reduce costs, influence channel members, and increase customer service, and external variables, such as industry competitiveness and demand uncertainty, were also found to affect adoption. Further, Williams (1994) concluded that the effects of these variables often varied significantly between shippers and carriers, and between suppliers and customers. Finally, Suzuki and Williams (1998) examined factors influencing the resistance of firms to adopting EDI. Again, using a survey of CLM members, the authors found that high levels of technologica l uncertainty, a low diffusion rate of EDI formats, and little improvement in processing time due to the use of EDI were factors that most increased the resistance of firms to EDI adoption.

Researchers in a number of disciplines outside of transportation, logistics, and supply chain management have also been active in investigating technology adoption and the use of EDI. The information systems area has, not surprisingly, been one of the most active research disciplines in this area. Information systems researchers have found that the decision to adopt EDI has been motivated by factors such as the ability to use technology as a substitute for other resources (Reinganum, 1981), sales volume (Barua and Lee, 1997), and competitive position and control of unique resources (Clemons and Row, 1989 & 1993). Benefits from EDI adoption and use have been found to include improvement in a firm's financial performance and the creation of competitive advantages (Cash and Konsynski, 1985; Clemons and McFarlan, 1986; Palmer and Markus, 2000; and Sethi, Hwang, and Pegels, 1993).

There has been little work conducted on the adoption and use of Internet technologies to facilitate communication across the supply chain. Exceptions include Murphy and Daley (2000), Mm and Galle (1999), and Yao, Palmer, and Dresner (2001). Murphy and Daley (2000) surveyed international freight forwarders on Internet usage. An important finding from their work was that the Internet is viewed as a complement to EDI usage, not as a replacement. Fully 78 percent of respondents to the authors' survey agreed with the complementary nature of Internet technology, compared to only 6 percent who disagreed. Min and Galle (1999) surveyed members of the National Association of Purchasing Managers. The authors examined firm size as a determinant of the use of electronic commerce in the supply chain. They found that larger firms were more likely to require that suppliers use electronic commerce than were their smaller counterparts. The reasons provided were that larger firms could reap greater rewards from an electronic commerce relationship with suppliers, and larger firms were in a stronger position to mandate electronic commerce usage. Yao, Palmer, and Dresner (2001) surveyed firms in the food industry to determine reasons for using Internet technologies in the supply chain. They found that a number of perceived benefits from Internet technology usage were important determinants in adopting the technology. These benefits included transaction cost reduction, improved inventory management, pricing optimization, improvements in order processing, and improved customer service.

In summary, there has been much work conducted on the motivations behind the adoption and use of EDI and the benefits derived from EDI, although little academic research has been published on Internet technology adoption and use. The general conclusions of the EDI research are that EDI adoption and use is not uniform; that is, that there are a number of organizational, competitive, and external factors that influence adoption and use. In general, the literature shows that EDI technology provides significant benefits to users. Our research adds to the literature in two important ways. First, it provides some preliminary results on the uses of Internet technology in an important industry: the food industry. Second, it provides some evidence as to current and expected use of Internet technology across the various levels of the supply chain.

METHODOLOGY AND DESCRIPTIVE STATISTICS

A survey methodology was used to assess current and expected use of Internet technology in the food industry. The survey instrument (Appendix A lists the survey questions used in the study) was developed based on variables used in previous studies (Palmer and Markus, 2000) and understanding gained from a series of interviews conducted with twelve executives in the food industry. The industry interviews encompassed executives at various levels of the supply chain, including technology providers, manufacturers, food distributors, and operators. Finally, copies of a draft questionnaire were sent to industry professionals to pre-test for the objectiveness and clarity of questions. Their insights were incorporated into the final version of the questionnaire.

The survey sample was derived from the database of subscribers to Food Logistics magazine, a trade publication targeted at managers in the food industry. Food Logistics provided us with subscriber names, firm addresses, phone numbers, and job titles. The survey instrument was sent to managers across the entire food industry, to both managers working in firms whose primary business is grocery products and to managers working in firms whose primary business is food services. Respondents represented managers in top positions at target firms, with 129 (70.5 percent) indicating that they worked in corporate management, 32 (17.5 percent) stating that their position was in operating management, 14 (7.7 percent) who worked in logistics management, and 8 (4.4 percent) whose position was in purchasing management. It was thought that respondents in top management positions were likely to be knowledgeable of their firm's electronic supply chain strategy and practices.

A stratified random sampling procedure was used to ensure that all levels of the supply chain were sampled (i.e., manufacturers, distributors, and retailers), as well as managers representing both the food services and grocery products segments. Surveys were mailed to a total of 1,497 managers, with initial non-respondents receiving up to three rounds of mailings, with intervals of 3 weeks between each. Of the total sample, mailings to 109 addresses resulted in delivery problems, reducing the effective sample size to 1,388. A total of 216 responses were received from the three mailings for a response rate of 15.56 percent. Among the returned surveys, 26 were discarded since there were large percentages of missing answers (more that 20 percent), while 7 surveys were received from third-party logistics providers, who were not within the scope of this study. Finally, 183 usable surveys were left to analyze.

To address the potential for non-response bias, twenty addresses were randomly selected from the non-respondents. A set of key questions from the survey was sent to the twenty addresses by express mail, followed by a round of telephone calls. Seven of the surveys were returned due to either undeliverable addresses or because the contacts were no longer with the company. The remaining thirteen surveys were used to conduct MANOVA tests for non-response bias. There was no significant difference between the respondents and non-respondents at the 5 percent level. As an additional test, answers on a number of key questions from late respondents (surveys received after the third mailing) were compared to answers from earlier respondents to determine if there were any statistical differences. Prior research has found that the profile of late respondents may resemble those of non-respondents (Malhotra and Grover, 1998; Biemer, 1991). Again, using a MANOVA test, no significant differences (p=.7) were found between the early and late respondents, suggesting that non-response bias was not a problem.

Responses to the survey were received from all of the segments in the supply chain, with 16 percent of the respondents representing manufacturing firms, 42 percent distribution establishments, and 42 percent retailers. Forty-six percent of the respondents indicated that they worked primarily in the food services sector of the industry, while 54 percent worked primarily in the grocery products sector. Size of firms ranged from under $10 million in annual sales (54 percent of respondents) to over $1 billion in annual sales (7 percent of respondents), reflecting the general makeup of the food industry (Standard and Poor's, 2000). Table 1 presents a descriptive analysis of the respondents.

In order to obtain information on communication practices with both suppliers and customers, respondents were asked to indicate the percentage of orders received from customers or placed with suppliers using various methods of communication. The results are presented in Tables 2 and 3. The tables show, in general, that the food industry does not make much use of Internet or proprietary EDI technologies to communicate with suppliers or customers, indicating, perhaps, a low level of technical sophistication in the industry. As shown in the tables, only 2.4 percent of orders were placed with suppliers and 1.9 percent of orders were taken from customers over the web. The most prevalent modes of communicating with supply chain partners remain phone, fax, and personal contact. The tables do not reveal that the distributors are falling behind manufacturers or retailers in the use of web-based technology. The data show that distributors may be using web-based technologies at slightly higher rates than manufacturers o r retailers and that the food services sector may use the Internet to a greater extent than the grocery products sector.

Table 4 shows the predicted use of the web in terms of the percentage of web orders firms plan to place with suppliers or take from customers one year in the future. [3] The table shows that the food services sector of the food industry plans more active use of the web for placing and taking orders than the grocery products sector. As well, manufacturers appear to plan to take a larger percentage of orders over the web than do retailers or distributors. Finally, in comparing predicted versus current web usage, it is evident that firms in all categories surveyed plan to significantly expand the number of orders placed or taken over the web within a year.

REGRESSION MODELS AND RESULTS

The purpose of this article is to examine current and expected Internet technology use across the food industry supply chain, and especially to compare use of the technology by distributors to usage rates by retailers and manufacturers. In this section, we use regression models to more formally assess usage differences among supply chain levels. In order to assess whether there are statistical differences in usage rates across the supply chain, a multiple regression was estimated as follows:

Web Orders = [[beta].sub.0] + [[beta].sub.1] Retailer + [[beta].sub.2] Manufacturer + [[beta].sub.3] Food Service + [[beta].sub.4] Firm Size (1)

where:

* Web Orders is the average of the percentage of orders a firm stated that it placed and accepted over the web;

* Retailer is a dummy variable to indicate if the firm is a retailer and allows the comparison of web adoption rates between retailers and distributors;

* Manufacturer is a dummy variable to indicate if the firm is a manufacturer and allows the comparison of web adoption rates between manufacturers and distributors;

* Food Service is a dummy variable to indicate if the firm is in the food service segment of the food industy, rather than the grocery products sector. This variable is included to account for potential competitiveness, uncertainty, or other differences between the two food industry sectors that could influence use of the technology (Willams, 1994; Willams, Magee, and Suzuki, 1998);

* Firm Size is a measure of the firm's annual revenues. [4] Firm size is included because size has been found in past studies to influence technology (i.e., ED1) adoption and use (e.g., Williams, 1994; Williams, Magee, and Suzuki, 1998).

A TOBIT procedure was used to estimate the models. [5] TOBIT regressions are recommended, rather than ordinary least squares, when the distribution of the dependent variable is truncated, as is the case with our data (Greene, 1997). [6] The results of the regression are shown in the first data column in Table 5. Neither the dummy variable for manufacturers, nor for retailers, is significant at the 5 percent level, indicating that there is no significant difference in web usage between manufacturers or retailers and distributors, after controlling for firm size and industry sector. The food service dummy coefficient is positive and significant at the 5 percent level, indicating that web usage is more prevalent in the food services sector than in the grocery products sector of the food industry. Finally, firm size has a positive and significant coefficient, indicating that web usage increases with firm size.

The results, as described above, do not indicate that distributors have fallen behind retailers or manufacturers in the use of web technology. In order to see whether these results are likely to continue to hold, we estimated the same regression except for the dependent variable. In this case, the dependent variable was the average of the orders (in percentage terms) a firm stated that it planned to place and accept over the web one year in the future. As noted in Table 4, the anticipated levels of web use, one year in the future, are considerably higher than current web use, indicating plans to increasingly make use of the Internet. The results of the anticipated use regression are presented in the second data column of Table 5. Again, there are no significant differences in anticipated web usage between either retailers or manufacturers and distributors, after controlling for industry segment and firm size. Other results show that the food services segment anticipates greater web use than the grocery prod ucts segment of the food industry, and that anticipated web use increases with firm size.

One additional comparison was conducted across the supply chain levels to assess Internet usage. Firms were asked about specific activities that they conducted over the web with their customers and suppliers (see Table 6). These activities included the following: sharing inventory information, obtaining or providing product information, obtaining or providing pricing information, conducting demand forecasting, obtaining or providing order status information, tracking orders, providing or receiving payments, and engaging in Internet trading exchanges. Respondents were asked, in each case, whether they engaged in light, moderate, or heavy activity over the web. When provided with this wider array of Internet activities, fully 50 percent of respondents indicated that they used the Internet for at least some activity with suppliers or customers. As indicated in Table 6, the activities performed most often over the web include obtaining pricing and product information from suppliers and providing pricing and pro duct information to customers.

In order to compare Internet use across the supply chain, an overall Internet usage measure was first calculated for each respondent. A score of 1 was awarded for a light activity, 2 for a moderate activity, and 3 for a heavy activity. A respondent's Internet usage measure was the sum of scores over all of the activities listed in Appendix A. The Internet usage measure was then regressed on the independent variables using a TOBIT procedure, as was the case with the previous regressions.

Results are presented in the third data column of Table 5. The results are similar to those described above. Both the manufacturer and retailer dummy variables are not significant, indicating no difference in Internet use between distributors and either retailers or manufacturers. Firm size is significant at the 5 percent level, again indicating that larger firms are more likely to use the Internet. One difference, however, was that in this estimation, there is no significant difference in Internet usage between the two industry sectors.

CONCLUSIONS, MANAGERIAL IMPLICATIONS, AND FUTURE RESEARCH

The purpose of this article is to investigate current and expected Internet usage across the food industry supply chain to determine whether distributors are using or anticipate to use Internet technology at the same rate as retailers and manufacturers. It has been suggested in the trade literature that since the Internet facilitates direct sales to end users, distributors could be disintermediated, and that the best defense against disintermediation may be to invest in Internet technology (Wilder, 1997; Information Week, 1998; Madsen, 1999; Skillen, 2000).

Our findings show that there are no significant differences in Internet usage between distributors and retailers or between distributors and manufacturers, after controlling for food industry sector and firm size. No significant differences were found in anticipated Internet use, either. We did find, however, significant differences in Internet usage according to firm size. To the extent that distributors are, on average, smaller than manufacturers, as is the case with our sample, they may still be at some risk in falling behind (at least manufacturers) in Internet technology use.

Our survey also showed that there is relatively little use of web-based technology in the food industry supply chain. Most transactions with suppliers and customers are still conducted by telephone, fax, or by personal contact. In addition, EDI accounts for a larger percentage of supply chain transactions than does the web. This latter result may reflect the high initial costs of establishing EDI connections and the reluctance of firms to switch from legacy systems to the Internet. The finding also lends support to the earlier work of Murphy and Daley (2000), suggesting that the Internet is used by firms as a complement to EDI use, not as a replacement.

There are two managerial implications that arise from our research. First, given the large number of benefits that have been attributed to Internet usage in the supply chain (Murphy and Daley, 2000), and the low level of Internet usage in the food industry, there appears still to be potential first mover advantages. Firms that can quickly establish web-based systems for transacting and sharing information across the supply chain may be able to achieve a competitive advantage over their rivals. For example, a food distributor that is first able to use information technology to link with a restaurant chain may be difficult to displace because of switching costs associated with changing to a different distributor and technology application. [7] Second, since larger firms are using Internet technology at greater rates than smaller firms in the food industry, smaller distributors may be under greater threat of disintermediation than larger distributors. The smaller distributors may have to quicken their pace of In ternet technology adoption and use or, if resources are an issue, merge with other firms to increase their size and resource availability.

A limitation of this research effort is the reliance on a sample of existing firms through the use of a cross-sectional methodology. A longitudinal study would be better able to capture the relationship between Internet investments and business success, since success and investments could be tracked over time. Other areas for future research include examining whether our results hold in other industries. It may be that characteristics of the food industry, including perishable and time-sensitive products, affect the level of Internet investment across the supply chain. As well, researchers could attempt to test directly for disintermediation. In this study, we tested for Internet technology use as a potential indicator of disintermediation. Further research could test directly for changes in goods flows through the supply chain to examine whether the role of distributors has been reduced as a result of Internet technology.

Mr. Dresner, EM-AST&L, is associate professor of logistics and transportation, R.H. Smith School of Business, University of Maryland, College Park, Maryland 20742; e-mail mdresner@rhsmith.umd.edu. Mr. Yao is a Ph.D. candidate in logistics and transportation, and Mr. Palmer is assistant professor of decision and information technology, R. H. Smith School of Business, University of Maryland.

The authors are grateful to both the Robert H. Smith School of Business and to Kevin Francella at Food Logistics for supporting this research effort.

ENDNOTES

(1.) See, for example, Wilder (1997), Information Week (1998), Madsen (1999), and Skillen (2000).

(2.) The term "distributors" is used to refer to both wholesalers and distributors.

(3.) The survey was conducted in the spring of 2000.

(4.) As indicated in Appendix A, respondents were asked to choose a revenue category for their firms. Firm size used in the regression was the midpoint of the category. There were four firms in the top category, greater than $5 billion in sales, and these were excluded from the estimation since a category midpoint could not be computed.

(5.) Following the suggestions of a reviewer, each of the models was also estimated using ANOVA. In all three models, similar results or the TOBIT estimation were obtained.

(6.) TOBIT is a maximum likelihood procedure designed to estimate models with truncated distributions of dependent variables. Whereas ordinary least squares yields biased parameter estimates under these conditions, TOBIT yields consistent coefficients; that is, unbiased estimates, as the sample size approaches infinity.

(7.) Although Internet applications use standard, rather than proprietary, technology to link firms, a major cost of implementation can still be connecting the technology to existing systems (e.g., inventory management, purchasing, accounting, etc.). Therefore, switching costs still exist.

REFERENCES

S. Barrett and B. Konsynski (1982), "Inter-organization Information Sharing Systems," MIS Quarterly, special issue.

A. Barua and B. Lee (1997), "An Economic Analysis of the Introduction of an Electronic Data Interchange System," Information Systems Research, Vol. 8(4), pp. 398-422.

P. P. Biemer (1991), Measurement Errors in Surveys, Wiley, New York, NY.

J. R. Brown, R. F. Lusch, and C. Y. Nicholson (1995), "Power and Relationship Commitment: Their Impact on Marketing Channel Member Performance," Journal of Retailing, Vol. 71(4), pp. 363-392.

J.I. Cash and B.R. Konsynski (1985), "Is Redraws Competitive Boundaries," Harvard Business Review, Vol. 63(2), pp. 61-68.

E.K. Clemons and F.W. McFarlan (1986). "Telecom: Hook Up or Lose Out." Harvard Business Review. Vol. 64(4), pp. 91-97.

E.K. Clemons and M. Row (1989), "Information Technology and Economic Reorganization," Proceedings of the Tenth International Conference on Information Systems, Boston, MA, pp. 341-351.

E.K. Clemons and M. Row (1993), "Limits to Interfirm Coordination Through Information Technology: Results of a Field Study in Consumer Goods Distribution," Journal of Management Information Systems. Vol. 10(1), pp. 73-95.

M.R. Crum and B.J. Allen (1997), "A Longitudinal Assessment of Motor Carrier-Shipper Relationship Trends," Transportation Journal, Fall, Vol. 37(1), pp. 5-17.

M.R. Crum, D.A. Johnson, and B.J. Allen (1998), "A Longitudinal Assessment of EDI Use in the U.S. Motor Carrier Industry," Transportation Journal, Fall, Vol. 38(1), pp. 15-28.

P.J. Daugherty. R. Germain, and C. Droge (1995), "Predicting EDI Technology Adoption in Logistics Management: The Influence of Context and Structure," Logistics and Transportation Review, Vol.31(4), pp. 309-324.

C. Droge and R. Germain (2000), "The Relationship of Electronic Data Interchange with Inventory and Financial Performance." Journal of Business Logistics, Vol. 2 1(2), pp. 209-230.

M.L. Fisher, J.H. Hammond, W.R. Obermeyer, and A. Raman (1994), "Making Supply Meet Demand in an Uncertain World," Harvard Business Review, Vol. 72(3), pp. 83-93.

J. Fry (1999), "What Hath E-Commerce Wrought?," Wall Street Journal, Eastern Edition, December 31, p. R38.

W.H. Greene (1997). Econometric Analysis, 3rd edition, New York: Prentice-Hall, pp. 962-974.

D. Griffith and J. Palmer (1999), "Leveraging the Web for Corporate Success," Business Horizons, January/February, Vol. 42(1), pp. 3-10.

Information Week (1998), "Distributors Must Add Value and Provide Customer Service to Avoid Extinction," No. 705, pp. SS8-SS 10.

D. Mackay and M. Rosier (1996), "Measuring Organizational Benefits of EDI Diffusion: A Case of the Australian Automotive Industry," International Journal of Physical Distribution and Logistics Management, Vol. 26(10), pp. 60-78.

H. Madsen (1999), "Rule 4: Lever the New Intermediaries," Management Today, August, p. 40.

M.K. Malhotra and V. Grover (1998), "An Assessment Of Survey Research In POM: From Constructs To Theory," Journal of Operations Management, Vol. 16, pp. 407-425.

H. Min and W.P. Galle (1999), "Electronic Commerce Usage in Business-to-Business Purchasing," International Journal of Operations & Production Management, Vol. 19(9), pp. 909-921.

P.R. Murphy and J.M. Daley (2000), "An Empirical Study of Internet Issues Among International Freight Forwarders," Transportation Journal, Summer, Vol. 39(4), pp. 5-13.

J. Palmer and M.L. Markus (2000). "The Performance Impact of Quick Response and Strategic Alignment in Specialty Retailing," Information Systems Research, September, Vol. 11(3), pp. 241-259.

J. Reinganum (1981), "Market Structure and the Diffusion of New Technology," Bell Journal of Economics, Vol. 12, Autumn, pp. 618-624.

V. Sethi, K.T. Hwang, and C. Pegels (1993), "Information Technology and Organizational Performance," Information and Management, Vol. 25, pp. 193-205.

G. Skillen (2000), "Breaking the Supply Chains," Inform, January/February, Vol. 14(1), pp. 18-20.

Standard and Poor's (2000). Industry Surveys.

T. Stank, M. Crum, and M. Arango (1999), "Benefits of Interfirm Coordination in Food Industry Supply Chains," Journal of Business Logistics, Vol. 20(2), pp. 21-41.

Y. Suzuki and L.R. Williams (1998), "Analysis of EDI Resistance Behavior," Transportation Journal, Summer, Vol. 37(4), pp. 36-44.

L.W. Walton (1994), "Electronic Data Interchange (EDI): A Study of Its Usage and Adoption Within Marketing and Logistics Channels," Transportation Journal, Winter, Vol. 34(2), pp. 37-45.

S.V. Walton and A.S. Marucheck (1997), "The Relationship Between EDI and Supplier Reliability," International Journal of Purchasing and Materials Management, Summer, Vol. 33, pp. 30-35.

C. Wilder (1997), "Middlemen Beware?," InformationWeek, No. 653, pp. 94-97.

L.R. Williams (1994), "Understanding Distribution Channels: An Interorganizational Study of EDI Adoption," Journal of Business Logistics, Vol. 15(2), pp. 173-203.

L.R. Williams, G.D. Magee, and Y. Suzuki (1998), "A Multidimensional View of EDI: Testing the Value of EDI Participation to Firms," Journal of Business Logistics, Vol. 19(2), pp. 73-87.

Y. Yao, J. Palmer, and M. Dresner (2001), "Electronic Supply Chains in the Food Industry: Benefits, Use, Integration, and Chain Position," working paper, University of Maryland.

Table 1.

Descriptive Statistics for the
Respondents


                                      Number of   Percent of
                                     Respondents  Respondents

Industry Segment

            Primarily food services      78          46.4
            Primarily grocery            90          53.6
            products
            Total                       168 [*]     100.0

Position in Chain

            Manufacturer                 27          16.1
            Distributor                  71          42.3
            Retailer                     70          41.7
            Total                       168 [*]     100.0

Annual Sales

            $10 million or less          88          54.0
            $11-50 million               34          20.9
            $51-100 million              12           7.4
            $101 million-$1 billion      17          10.4
            $1-5 billion                  5           3.1
            Total                       163 [*]     100.0



(*)The discrepancies in numbers are due to missing values.
Table 2.

Communication Methods Used With Suppliers (Percent of Orders)


          Manufacturers  Distributors  Retailers  Primarily  Primarily
                                                    Food      Grocery
                                                  Services   Products

Phone         55.6           40.5        46.5       49.5       42.2
Fax           19.2           31.8        18.5       21.6       26.3
Personal      11.7           11.6        14.5       13.3       12.4
Contact
EDI            7.9            8.6        15.2        6.1       15.7
Email          1.6            2.9         3.9        4.0        2.4
Web            3.3            3.8         0.7        4.6        0.6
Mail           1.2            1.0         0.5        1.1        0.6



          Total



Phone     45.6
Fax       24.1
Personal  12.8
Contact
EDI       11.3
Email      3.1
Web        2.4
Mail       0.9
Table 3.

Communication Methods Used With Customers (Percent of Orders)


          Manufacturers  Distributors  Retailers  Primarily  Primarily
                                                    Food      Grocery
                                                  Services   Products

Personal      24.1           51.1        58.7       48.2       51.3
Contact
Phone         32.6           24.9        21.8       28.0       22.0
Fax           23.1           14.1        10.5       12.4       15.7
EDI            8.3            2.1         5.2        1.9        6.6
Web            1.5            2.5         1.4        3.9        0.0
Email          5.5            1.0         0.5        2.6        0.9
Mail           3.2            2.5         0.4        1.0        2.0



          Total



Personal  49.8
Contact
Phone     24.9
Fax       14.1
EDI        4.4
Web        1.9
Email      1.7
Mail       1.5
Table 4.

Predicted Use of Web One Year in the Future vs. Current Web Use
(Percent of Orders)


                         Orders Placed with Suppliers

                                Predicted Use          Current Use

Manufacturers                        14.6                  3.3
Retailers                            15.4                  0.7
Distributors                         11.5                  3.8
Food services sector                 19.1                  4.6
Grocery products sector               8.9                  0.6
All firms                            13.6                  2.4



                         Orders Taken from Customers

                                Predicted Use         Current Use

Manufacturers                       16.4                  1.5
Retailers                            9.3                  1.4
Distributors                         9.0                  2.5
Food services sector                13.1                  3.9
Grocery products sector              7.8                  0.0
All firms                           10.3                  1.9
Table 5.

TOBIT Results on Web Use in the Supply
Chain (Standard Errors in Parentheses)


                           Dependent Variables

                              Regression 1          Regression 2

                             Percentage of      Percentage of Orders
                            Orders Currently       Expected to be
                           Conducted Over the    Conducted Over the
                                  Web             Web in One Year

Constant                        -124.8 [**]             0.7
                                 (35.6)                (6.5)
Retailer dummy                   -38.2                  3.3
                                 (25.8)                (7.5)
Manufacturer dummy                -5.0                 11.9
                                 (29.3)               (10.2)
Food service sector dummy         49.4 [*]             16.0 [*]
                                 (24.1)                (7.0)
Firm size                          0.38 [*]             0.12 [*]
                                  (0.17)               (0.06)
Number of observations            147                   154






                            Regression 3

                           Internet Usage
                              Measure



Constant                        -3.5
                                (2.9)
Retailer dummy                  -1.0
                                (3.3)
Manufacturer dummy              -0.2
                                (4.5)
Food service sector dummy        1.9
                                (3.0)
Firm size                        0.06 [*]
                                 (.03)
Number of observations            145



(*)5 percent significance

(**)1 percent significance
Table 6.

The Level of Use of Activities over the  Web


                                          Use of the Web to Conduct
                                             Business Activities
                                          (Percent of Respondents)

                                                    Light
                                                  Activity


Obtain product information from                     13.8
suppliers
Obtain pricing information from                     18.0
suppliers
Provide product information to customers            11.7
Provide pricing information to customers            11.7
Obtain order status from suppliers                  12.9
Track orders from suppliers                         13.5
Track orders to customers                            5.9
Receive payments from customers                     10.0
Provide payments to suppliers                        8.2
Provide order status to customers                    7.0
Conduct demand forecasting with                      7.6
suppliers
Share inventory information with                     5.9
suppliers
Engage in Internet trading exchange                  8.1
Conduct demand forecasting with                      7.1
customers
Obtain inventory information from                    7.0
customers







                                          Moderate   Heavy      All
                                          Activity  Activity  Activity
                                                               Levels

Obtain product information from            19.0       6.9       39.7
suppliers
Obtain pricing information from            15.1       5.2       38.4
suppliers
Provide product information to customers   13.5       7.6       32.8
Provide pricing information to customers   11.1       7.6       30.4
Obtain order status from suppliers          8.8       6.4       28.1
Track orders from suppliers                 7.0       4.7       25.2
Track orders to customers                  10.7       3.6       20.2
Receive payments from customers             4.7       4.7       19.4
Provide payments to suppliers               4.7       5.3       18.2
Provide order status to customers           5.8       5.3       18.1
Conduct demand forecasting with             5.9       3.5       17.1
suppliers
Share inventory information with            8.3       2.4       16.6
suppliers
Engage in Internet trading exchange         4.7       3.5       16.3
Conduct demand forecasting with             5.3       1.8       14.2
customers
Obtain inventory information from           5.8       1.2       14.0
customers

Appendix A. Survey Questions Used in the Study

Placing orders with your suppliers:

Please estimate the percentage of orders your firm places with your suppliers using each of the following methods. The total percentages should add to 100.

Personally provided to supplier __%

Telephoned to supplier __%

Mailed to supplier __%

Faxed to supplier __%

Web-based order placed with supplier __%

Electronic Data Interchange (EDI) with supplier

Total 100 %

E-mailed to supplier __%

One year from now, please estimate the percentage of orders from your suppliers that you anticipate your company will make over the web. Check one line only.

0% __ 1-10%__ 11-20%__ 21-30%__ 31-40%__ 41-50%__

51-60% __ 61-70%__ 71-80%__ 81-90%__ 91-100%__

Taking orders from your customers:

Please estimate the percentage of orders your firm receives from your customers using each of the following methods. The total percentages should add to 100.

Faxed by customer __%

E-mailed by customer __%

Telephoned by customer __%

Mailed by customer __%

Web-based order placed by customer __%

Personally taken by our sales staff __%

Total 100%

One year from now, please estimate the percentage of orders from your customers that you anticipate your company will take over the web. Check one line only.

0% __ 1-10% __ 11-20% __ 21-30% __ 31-40% __ 41-50% __

51-60% __ 61-70% __ 71-80% __ 81-90% __ 91-100% __

Other business activities:

Please indicate whether you currently conduct any of the following activities over the web. Please circle the appropriate number in each row:

                                      Light    Moderate   Heavy
                                     activity  activity  activity

Share inventory information with        1         2         3
suppliers
Obtain product information from         1         2         3
suppliers
Obtain pricing information from         1         2         3
suppliers
Conduct demand forecasting with         1         2         3
suppliers
Obtain order status from suppliers      1         2         3
Track orders from suppliers             1         2         3
Provide payments to suppliers           1         2         3
Obtain inventory information from       1         2         3
customers
Provide product information to          1         2         3
customers
Provide pricing information to          1         2         3
customers
Conduct demand forecasting with         1         2         3
customers
Provide order status to customers       1         2         3
Track orders to customers               1         2         3
Receive payments from customers         1         2         3
Engage in Internet trading exchange     1         2         3

General Company Information:

Which area of the food industry contributes the most revenues to your operations (Circle one option only).

1. Food processing or manufacturing-grocery

2. Food processing or manufacturing-foodservice

3. Supermarket chain

4. Full-line grocery wholesale distribution

5. Foodservice distribution

6. Full-line convenience store wholesale distribution

7. Third-party warehouse/transportation/logistics provider

8. Restaurant/foodservice chain

Please indicate the total revenues for your company (all locations) in 1999 (or fiscal year 1999) by circling the appropriate number:

1. Less than $10 million

2. $11 million-$50 million

3. $51 million - $100 million

4. $101 million - $1 billion

5. Over $1 billion but less than $5 billion

6. $5 billion or above