SPECIAL SECTION ‘E-BUSINESS IMPACTS REVISITED’

DOI: 10.1080/10196780500208756 Copyright ß 2005 Electronic Markets Volume 15 (3): 213–224. www.electronicmarkets.org

Downloaded By: [Schmelich, Volker] At: 20:44 23 March 2010

A

b

s

t

r

a

c

t

This paper examines how World Wide Web (WWW) usage for procurement purposes affects corporate buyers’ search costs and buying organizations’ coordination costs and how it ultimately affects the efficient boundaries of buying organizations. The analysis is based on a model that integrates Transaction Costs Economics (TCE) (Williamson 1975, 1985) and forces of electronic interconnections (Malone et al. 1987). Data collected from 110 corporate buyers from over 100 organizations doing business in a wide range of industries were used to test the validity of the proposed model. Findings show that WWW usage is associated with a reduction of buyers’ search costs, which can give buying organizations the flexibility to increase their base of suppliers when such an increase fits their interest. No clear relationship was found between WWW usage and the reduction of coordination costs buying organizations face to integrate their value chain activities with those of their suppliers. Findings show however that a WWW-based reduction in coordination costs can give buying organizations the flexibility to outsource certain activities when such a move fits their interest. The TCE framework was used to identify the transactions most likely affected by these changes in buying organizations’ boundaries. Keywords: e-markets, search costs, coordination costs, transaction costs

A

u

t

h

o

r

s

Younes Benslimane ([email protected]) is an Assistant Professor of Information Technology at York University, Toronto, Canada. His research interests include the economics of IS and E-commerce. Michel Plaisent ([email protected]) is a Professor of MIS at the University of Quebec in Montreal, Canada. His research interests include e-commerce and e-learning. Prosper Bernard ([email protected]) is a Professor of Strategic Management at the University of Quebec in Montreal, Canada. His research interests include strategic IS and e-commerce.

Investigating Search Costs and Coordination Costs in Electronic Markets: A Transaction Costs Economics Perspective YOUNES BENSLIMANE, MICHEL PLAISENT AND PROSPER BERNARD

INTRODUCTION Electronic commerce (EC) has received much attention from Information Systems (IS) researchers and practitioners because it can improve the performance of individual users (Bakos 1997; Barua et al. 1997) and of organizations (DeLone and McLean 1992, 2003; Jones and Beatty 1998; Mukhopadhyay et al. 1995). From buyers’ perspective, EC is the use of telecommunication networks for the identification of potential suppliers, the selection of business partners, and the execution of transactions (Choudhury et al. 1998). A narrow definition of EC suggests that all three activities be conducted online, whereas a wider one, the one adopted in this research, suggests that at least one activity be supported electronically (Reimers 1996). Traditionally, EC has been supported by complex and expensive interorganizational systems (IOS) and as a result, only a small fraction of business transactions has been conducted electronically (Barua and Lee 1997; Scala and McGrath 1993). In the 1990s, the World Wide Web (WWW) emerged as a new and valuable tool for EC and thousands

of vendors have set up commercial websites to do business over the Internet. These commercial websites differ in terms of the functionalities they provide: The simpler ones are informational and the more complex ones are transactional (Lightner 2004; Liu et al. 1997). Depending on their functionalities, commercial websites support, in whole or in part, electronic transactions. Transactional websites typically include electronic catalogues, shopping carts, payment systems and order tracking systems that offer buyers a convenient and cost-effective way to support the procurement process as they allow for the identification and the selection of suppliers and the execution of business transactions. Commercial websites can support both electronic hierarchies – by linking an organization to its sole supplier of a given input – and electronic markets – by linking multiple vendors and buyers. Conducted from the perspective of corporate buyers and buying organizations and based on a model that integrates forces of electronic interconnections (Malone et al. 1987) and Transaction Cost Economics (TCE) (Williamson 1975, 1985), this study analyses the relationships between WWW usage for procurement

214 Younes Benslimane, Michel Plaisent and Prosper Bernard & Investigating Search Costs and Coordination Costs in Electronic Markets

and the productivity of corporate buyers and the efficient boundaries of buying organizations.

LITERATURE REVIEW

Downloaded By: [Schmelich, Volker] At: 20:44 23 March 2010

The evolution of the Transaction Costs Economics framework The Transaction Costs Economics (TCE) framework has been used predominantly to analyse the efficient boundaries of buying organizations (Bakos and Kemerer 1992). It first originated in Institutional Economics (Coase 1937; Williamson 1975, 1985) and has later been extended and applied in the field of IS to analyse the consequences of IOS usage on firms’ governance structure (Bakos 1997; Bakos and Brynjolfsson 1997; Choudhury et al. 1998; Ciborra 1993; Gurbaxani and Whang 1991; Malone et al. 1987). The original framework views the governance structure within a dichotomy between markets, which refer to the outsourcing of the activity, and hierarchies, which refer to the internalization of an activity. The extended framework introduces another dichotomist view of the governance structure for an outsourced activity. In this new context, the governance structure is viewed within a dichotomy between markets, which refer to relying on a pool of suppliers for a given input and hierarchies, which refer to exclusive relationships between an organization and its sole supplier for a given input. Both frameworks are based on the assumptions of human opportunism and bounded rationality, and aim at describing the preferred governance structure for an organization. With regards to the ‘make-or-buy’ decision, the framework maintains that the choice for a given input is based on a comparison of total costs of the alternative options and that organizations tend to select the governance structure that minimizes the total cost. Two different views of such total cost can be found in the literature, each one focusing on a particular aspect of the cost of using the market. The first one, in accordance with Williamson (1975, 1985), considers the total cost as the sum of the production cost and the contractual cost for producing or procuring the given input. The second one, pioneered by Malone et al. (1987) and further refined by Gurbaxani and Whang (1991), sees the total cost as the sum of the production cost and the ‘operational’ cost for producing or procuring the given input. An organization that relies on the market for a given input tends to incur both operational and contractual costs. Operational costs include search costs, transportation costs, inventory holding costs and communication costs whereas contractual costs are the costs of writing, monitoring and enforcing the contract related to a given transaction. Both views of the total cost suggest that production cost for a given input tends to be lower in markets than in

hierarchies. Such a difference is due to the fact that markets benefit from specialization effect and economies of scale. Therefore, it is more cost-effective for an organization that does not specialize in the production of a given input to get that input from a supplier that specializes in the production of such a component. However, using the market can be expensive because it makes the buying organization incur costs that would not apply if it had produced the given input itself. A first component of the market-related costs is the search cost, which refers to the time and effort spent identifying potential suppliers and selecting a vendor (Bakos 1997; Ciborra 1993; Gurbaxani and Whang 1991; Malone et al. 1987). A second component is the coordination costs, which are the costs an organization incurs in order to manage the flow of goods and services and synchronize its activities with those of its supplier (Ciborra 1993; Clemons et al. 1993; Gurbaxani and Whang 1991; Malone et al. 1987). The remaining component of such market-related costs are the contractual costs, which refer to the costs of writing, monitoring and enforcing a contract in order to protect a buying organization against the risks of information asymmetries and of loss of bargaining power it may face when dealing with a supplier (Ciborra 1993; Clemons et al. 1993; Gurbaxani and Whang 1991; Williamson 1985). Malone et al. (1987) summarized the trade-off between markets and hierarchies in terms of relative production costs and relative coordination costs arguing that relative production costs are higher in hierarchies and that relative coordination costs are higher in markets. The extended framework has focused on transactionrelated factors that help determine the optimal number of suppliers for a given input. Three important factors have been discussed: they are complexity of product description, non-contractible investments and price variability. Malone et al. (1987) have defined the complexity of product description as the amount of information needed to specify characteristics of a given input. A higher level of complexity of product description increases the cost of searching the market prior to each transaction, which makes hierarchy the preferred option. Drawing upon the theory of incomplete contracts, Bakos and Brynjolfsson (1997) have focused on the investments in non-contractible resources a transaction can require. Such non-contractible investments are by definition difficult to specify in a contract and therefore require a mechanism to protect the buyer. For such transactions, hierarchy is the preferred coordination mechanism as it helps build long-term relationships that prevent suppliers from behaving opportunistically. Clemons et al. (1993) as well have discussed the protection that buyers benefit from when building exclusive relationships with their suppliers. Grossman and Hart (1986) and Hart and Moore (1990) have suggested that vertical integration (i.e., asset ownership) and the authority that it confers can

Electronic Markets Vol. 15 No 3

Downloaded By: [Schmelich, Volker] At: 20:44 23 March 2010

help prevent opportunistic behaviour under incomplete contracting. Finally, Lacity and Willcocks (1995, 1998) have shown the limitations of outsourcing decisions that were based on Transaction Costs Theory and have presented selective outsourcing, experience and effective contractual arrangements as better strategies for risk mitigation. Finally, price variability in commodity markets has also been considered relevant to decisions related to the number of suppliers for a given input (Bakos 1991a, 1991b, 1997; Choudhury et al. 1998; Rosenthal et al. 1993). Researchers have argued that a higher level of price variability in commodity market makes the market the preferred option. In commodity markets, a given input tends to be identical across all vendors, which reduces both search costs and vulnerability to non-contractible investments. Moreover, price variability provides buyers with the incentive to alternate suppliers in order each time to get the needed input from the seller with the lowest cost.

The role of forces of electronic interconnections Malone et al. (1987) have been the first to link IOS usage to firms’ efficient boundaries arguing that IOS have electronic communication, brokerage and integration effects that help a buying organization reduce some of the costs it incurs when using the market: the electronic communication effect allows buyers to get more and/or faster information from potential suppliers, hence reducing the costs of communicating with potential suppliers; the electronic brokerage effect allows buyers to consider a larger pool of potential suppliers, which permits them to select better the supplier of a given input; the electronic integration effect allows buyers to reduce the uncertainty related to crossorganizational flow of activities, consequently insuring a tighter coupling of activities across adjacent value chains. These effects can affect the governance structure of organizations that have adopted such IOS. From this perspective, two basic moves have been considered: the first one relates to the decision to outsource activities (Brynjolfsson et al. 1994; Ciborra 1993; Gurbaxani and Whang 1991; Kambil 1991) and the second one relates to the decision to increase the number of suppliers for a given input (Bakos 1991a, 1991b; Bakos and Brynjolfsson 1997; Barua et al. 1997; Choudhury et al. 1998; Clemons et al. 1993; Malone et al. 1987). Regarding the first move, researchers have argued that forces of electronic interconnections can reduce specific components of the market-related costs, making outsourcing certain activities become more cost-effective. The electronic communication and brokerage effects of IOS can help reduce buyers’ search costs whereas the electronic integration can help reduce the coordination costs a buying organization incurs in order to integrate its value chain with that of its trading partner. Drawing

215

upon these three effects, IS researchers have argued that more activities once internalized will be outsourced, leading to a vertical disintegration of firms’ production operations. Most of the IS research in this area has been theoretical (Bakos and Brynjolfsson 1997; Clemons et al. 1993; Clemons and Row 1992, 1993; Gurbaxani and Whang 1991). Very few studies have empirically tested the direct relationship between IOS usage and the move to more outsourcing, and those who did, have relied on secondary data from selected US industries: Kambil (1991) found a positive correlation between information technology (IT) investments and firms’ vertical disintegration whereas Brynjolfsson et al. (1994) found a positive correlation between IT investments and the decline in firm size. Other scholars have investigated the relationship between IT investments and coordination costs without referring to the move to more outsourcing. Instead, they focused on the increased coordination requirements in contexts of vertical integration and of diversification and found that firms that expand their operations to other lines of business tend to invest more in IT, providing support for the idea that IT can reduce coordination costs (Dewan et al. 1998; Hitt 1999). Some case studies analysed the effect of IOS usage on coordination costs and their results were consistent with the theoretical work linking IT usage to reduction in coordination costs (Clemons and Row 1993; Iacovou et al. 1995; Jones and Beatty 1998; Kraut et al. 1999; Mukhopadhyay et al. 1995). As for the second move, researchers who have agreed on the ‘firm disintegration’ hypothesis have referred to two different yet complementary views of the activities most likely involved in such disintegration. The first one is known as the ‘move to the market’ hypothesis and was pioneered by Malone et al. (1987). It suggests that technological change causes transactions to be less assetspecific and to involve inputs that are easier to describe. This will lead organizations using electronic markets to broaden their base of suppliers for a given input because the electronic communication and brokerage effects of such systems help reduce the search costs buyers incur. Such IOS make it easier for buyers to search for a better deal each time they need to order an input. Bakos (1991a, 1991b, 1997) has argued that this move to ‘more market’ is likely to prevail in electronic markets trading commodity products where price-based competition is high. The second perspective proposed by Clemons et al. (1993) and Bakos and Brynjolfsson (1997), while taking into consideration the IOS-based reduction of search and coordination costs, insists on the necessity for firms to choose long term cooperative relationships with their suppliers in order to protect themselves against the risks of opportunistic behaviour. Based on Bakos and Brynjolfsson (1997), such option is likely to prevail in electronic markets trading noncommodity products that require more non-contractible resources. Empirical evidence on this topic is very

216 Younes Benslimane, Michel Plaisent and Prosper Bernard & Investigating Search Costs and Coordination Costs in Electronic Markets

limited. Rosenthal et al. (1993) who investigated buyers’ intentions in a hypothetical electronic market in the chemical industry found that the use of electronic hierarchies would remain the preferred coordination mechanism. Choudhury et al. (1998) analysed the consequences of an electronic market in the aircraft parts industry but did not directly test the effect of the IOS usage on the size of the supply base of a given input. They concluded however that by reducing buyers’ search costs, IOS usage help lower prices and broker usage, which may provide some support for the move to ‘more market’. Finally, using a Delphi analysis of possible consequences of Internet usage, Croom (2000) reported that Internet usage is likely to lead to outsourcing of maintenance, repair and operating (MRO) items to a reduced number of suppliers.

Downloaded By: [Schmelich, Volker] At: 20:44 23 March 2010

RESEARCH MODEL AND HYPOTHESES Because research on the TCE-based effects of IOS usage is mostly theoretical, many researchers have stressed the need for a rigorous operationalization of relevant concepts and testing of hypothesis (Bakos 1991b, 1997; Brynjolfsson et al. 1994; Clemons et al. 1993; Clemons and Row 1992, 1993; Kraut et al. 1999; Malone et al. 1987). The analysis of WWW usage for business-to-business (B2B) transactions, which are estimated to represent about 80% of the value and the volume of all transactions conducted over the Internet (Perkins 2001; Rosen and Howard 2000), gives the opportunity to formally validate previous theoretical work. The objective of this research is to analyse TCEbased effects of WWW usage on corporate buyers and buying firms. It attempts to validate possible relationships between electronic markets usage, search costs, coordination costs and moves to ‘more outsourcing’ and to ‘more market’ for specific inputs. Figure 1 presents the model that summarizes the hypothesized relationships. Each variable and relationship is discussed below.

Figure 1. Research model

WWW usage The analysis of WWW usage for procurement purposes should take into consideration the tasks and the types of purchase corporate buyers execute. Basically, the procurement process includes three activities: the identification of potential suppliers, the selection of a business partner and the execution of the transaction (Choudhury et al. 1998; Novak and Simco 1991). The WWW helps buyers identify a set of potential suppliers of a given input as thousands of vendors have set up their websites in order to do business over the Internet (Ghosh 1998; O’Connor and O’Keefe 1997; Poon and Swatman 1999). Also, when pricing and other relevant information are made available on commercial websites, the WWW allows buyers to select a business partner for the pondered transaction (Bakos 1997; Barua et al. 1997). Finally, when vendors have set up transactional websites, buyers can execute their transactions online by placing orders, making payments and checking the status of their orders online. Based on supply chain management literature, straight re-buys (hereafter called routine purchases) must be distinguished from modified re-buy or new buy (hereafter called non-routine purchases) as each type of purchase may require a different usage of the WWW. Typically, a routine purchase requires little or no additional information to be completed, whereas a non-routine purchase involves some complexity and therefore requires that corporate buyers first search for information about the input, about its price and about potential suppliers (Novack and Simco 1991).

Reduced search costs Search costs refer to the time and effort a buyer spends to access the information needed to identify potential suppliers of a given input and to select a vendor (Bakos 1991a, 1991b, 1997; Barua et al. 1997; Choudhury et al. 1998; Ciborra 1993). The WWW has an electronic communication effect and an electronic brokerage effect

Electronic Markets Vol. 15 No 3

that allow buyers to get more and/or faster information from potential suppliers and to better select a supplier respectively (Malone et al. 1987). Hypothesis 1: WWW usage will be positively related to a reduction of search costs.

Downloaded By: [Schmelich, Volker] At: 20:44 23 March 2010

Reduced coordination costs Coordination costs refer to the efforts deployed in order to manage interdependencies among activities. According to Clemons and Row (1993: 74), ‘problems in coordination stem from uncertainty due to insufficient information or information processing capacity … across firm boundaries.’ The coordination costs are the costs buying organizations face to reduce that uncertainty, leading to an integration of their value chain activities with those of their suppliers (Ciborra 1993; Clemons and Row 1993; Gurbaxani and Whang 1991; Malone et al. 1987). Transactional websites provide additional information and processing capacity that can allow for a reduction of these coordination costs. Hypothesis 2: WWW usage will be positively related to a reduction of coordination costs.

217

researchers have argued that IOS help reduce the external coordination costs, making it more efficient to externalize certain activities (Bakos and Brynjolfsson 1997; Brynjolfsson et al. 1994; Ciborra 1993; Clemons et al. 1993; Gurbaxani and Whang 1991). TCE literature suggests that transactions occurring at a lower frequency and involving standard inputs are likely to be affected by such a move because (1) low-frequency activities do not benefit from economies of scale and specialization effects and outsourcing them should be more cost-effective and (2) such activities involve standard inputs, which reduces the information, contracting and monitoring costs buying firms would incur. Suppliers’ transactional websites can help decrease coordination costs, encouraging organizations to outsource some activities in order to benefit from a lower total cost. Hypothesis 4: Reduced coordination costs will be positively related to the move to ‘more outsourcing’ for low-frequency transactions involving standard inputs.

METHODOLOGY Research design and source for data

More market The move to ‘more market’ refers to an increase in the pool of suppliers for a given input following the reduction in search costs associated with the electronic communication and brokerage effects of the WWW. IS researchers have argued that, just because they are conducted in electronic marketplaces, some transactions turn out to be executed among a larger number of suppliers (Bakos 1991a, 1991b, 1997; Bakos and Brynjolfsson 1997; Barua et al. 1997; Choudhury et al. 1998; Malone et al. 1987). TCE literature suggests that transactions occurring at a lower frequency and involving standard inputs are likely to be affected by such a move: for such transactions, buying firms using electronic markets incur no switching costs and benefit from an increased competition among suppliers. Therefore, they tend to move to ‘more market’ as it becomes easier for their buyers to search the market for the ‘best’ deal each time they have to purchase the given inputs. Hypothesis 3: Reduced search costs will be positively related to the move to ‘more market’ for low-frequency transactions involving standard inputs.

More outsourcing The move to ‘more outsourcing’ can be a consequence of the electronic integration effect of the WWW. IS

A survey was used to collect data on the evolution of search costs and coordination costs in WWW-based electronic marketplaces. In Spring 2001, a questionnaire was sent to each buyer member of the ‘Corporation des Approvisionneurs’ in the province of Quebec, Canada, an organization affiliated to the Purchasing Management Association of Canada. Within six weeks, of the 988 questionnaires sent, 110 usable ones were completed and returned for a response rate of approximately 11.2%. The sample spanned corporate buyers from over 100 organizations doing business in a wide range of industries.

Operationalization of constructs Except for WWW usage, no pre-existing instruments could be located. Therefore, new scales based on relevant literature were developed and tested for reliability and validity. Following Churchill’s (1979) guidelines, multiple item-scales measuring the investigated constructs were developed. In accordance with Nunnally (1988), seven-point Likert scales were used to ensure statistical variability among survey responses. The instrument was pilot-tested with five corporate buyers, one EC consultant and four IS researchers and their suggestions helped develop the final questionnaire. The section of questionnaire used for this research is presented in Appendix A.

Downloaded By: [Schmelich, Volker] At: 20:44 23 March 2010

218 Younes Benslimane, Michel Plaisent and Prosper Bernard & Investigating Search Costs and Coordination Costs in Electronic Markets

The instrument developed by Massetti and Zmud (1996) was adapted to measure WWW usage (USE) for both routine and non-routine purchases. Three items for each type of purchase were used: they are the importance of the number of purchases, of the dollar amount of purchases and of the percentage of suppliers dealt with using the WWW. The Reduced Search Costs (RSC) construct was measured in terms of increased information obtained from alternate vendors, of ease of locating potential suppliers, and of improved selection process for the product or service to be bought (Bakos 1991a, 1991b, 1997; Barua et al. 1997; Choudhury et al. 1998; Ciborra 1993; Malone et al. 1987). The Reduced Coordination Costs (RCC) construct was measured in terms of reduced paperwork, reduced clerical errors, increased data transmission speed, reduced lead times, reduced inventory costs and reduced stockout problems (Clemons and Row 1993; DeLone and McLean 1992, 2003; Iacovou et al. 1995; Jones and Beatty 1998; Reekers and Smithson 1996; Scala and McGrath 1993). Finally, two new three-item scales based on relevant TCE literature were developed to assess the move to More Outsourcing (MO) (Gurbaxani and Whang 1991; Williamson 1975, 1985) and the move to More Market (MM) (Bakos and Brynjolfsson 1997; Choudhury et al.

1998; Clemons et al. 1993; Malone et al. 1987) and to identify the activities and inputs most likely to be affected by such moves. The reliability of the scale was assessed using the Cronbach alpha. No item deflated alpha, so all were kept. The range for all alphas varied from .82 to .95, which indicates a high degree of consistency between the multiple items measuring each construct. The validity of the scale was assessed using a factor analysis. The principal components method with Varimax rotation was used to extract the five factors included in the analysis. A factor loading greater than .50 with the theoretically correct sign was required for the assignment of an item to a factor. The results of the factor analysis confirmed the validity of the scale. The score for each retained factor equalled the mean score of its retained items. Table 1 summarizes the results for the reliability and validity analyses.

DATA ANALYSIS Profile of respondents Table 2 presents the descriptive statistics for our sample. By job title, respondents were directors (26.9%),

Table 1. Validity, reliability and descriptive statistics for constructs

Mean St. Dev. Alpha RUSE1 RUSE2 RUSE3 NRUSE1 NRUSE2 NRUSE3 RSC1 RSC2 RSC3 RCC1 RCC2 RCC3 RCC4 RCC5 RCC6 MO1 MO2 MO3 MM1 MM2 MM3

WWW usage

Reduced search costs

Reduced coordination costs

More outsourcing

More market

2.59 1.29 .91

5.32 1.37 .83

3.30 1.47 .92

2.35 1.15 .95

4.22 1.55 .83

.899 .868 .665 .899 .903 .657 .798 .868 .685 .730 .810 .798 .826 .886 .879

Rotated component matrix showing coefficients with values equal to or greater than .5

.898 .932 .901 .783 .835 .800

Electronic Markets Vol. 15 No 3

219

Table 2. Respondents’ profile

Downloaded By: [Schmelich, Volker] At: 20:44 23 March 2010

Respondents’ profile

Frequency (%)

Industry Manufacturing Retail Service EDI usage Yes No Hierarchy level Director Manager Senior Buyer Buyer Early respondent Yes No Age 20–29 30–39 40–49 50–59 Annual volume of transactions ($) Less than 5 millions 5 millions - 10 millions 10 millions - 20 millions Over 20 millions Task Supported by the WWW (Average on a 7-point Likert scale) Identification Selection Execution

53.8 38.7 7.5 31.8 68.2 26.9 9.3 30.6 33.2 61.4 38.6 3.4 36.8 32.2 27.6 26.4 18.1 19.4 36.1

5.7 4.1 2.5

managers (9.3%), senior buyers (30.6%) and buyers (33.2%). The most common age group of respondents was 30–39 (37%), followed by 40–49 (34.3%), 50–59 (25%) and 20–29 (3.7%). The majority of the respondents worked in the manufacturing sector (53.8%), followed by those in the retail industry (38.7%) and those in the service sector (7.5%) and 31.8% of them worked in organizations that used electronic data interchange (EDI) systems to support their procurement process. In terms of annual volume of transactions, the respondents had purchased on average over $24 millions worth of input. Results based the seven-point scale show that corporate buyers use the WWW essentially to identify potential suppliers (5.7) or to select a supplier (4.1) and less to execute a transaction (2.5) and that on average the level of WWW usage for procurement purposes remains low (2.6). An additional screening assessed the non-response bias by comparing early respondents (i.e., those who had sent a completed questionnaire within the first three weeks) to late respondents (i.e., those who had sent a completed questionnaire after the third week) on the basis of the constructs under study. The rational for this screening is that late respondents are likely to have similar characteristics to non-respondents (Armstrong and Overton 1977). None of the five tests (a55%) indicated significant differences between the two groups, which suggests that there was no non-response bias. Finally, with regards to the relationships between respondents’ profile and the constructs under study, the only significant difference was for the level of WWW usage among buyers in the retail industry (Mean52.98) and those in the manufacturing sector (Mean52.33). Table 3 summarizes the results of these tests.

Table 3. Respondents’ profile and non-response tests Effect on Constructs under Study Profile factors Industry EDI usage Hierarchy level Early respondent Age Volume of transactions

WWW usage F5.3.495 Sig.5.034 t5.825 Sig.5.441 F5.824 Sig.5.442 t5.628 Sig.5.534 F5.663 Sig.5.577 F5.595 Sig.5.694

Reduced search costs Reduced coordination costs F51.563 Sig.5.214 t5.988 Sig.5.342 F5.082 Sig.5.922 t51.590 Sig.5.114 F5.657 Sig.5.592 F5.818 Sig.5.517

F5.394 Sig.5.675 t5.604 Sig.5.551 F51.000 Sig.5.371 t51.466 Sig.5.197 F5.447 Sig.5.720 F5.582 Sig.5.677

More outsourcing F51.148 Sig.5.321 t52.015 Sig.5.988 F51.594 Sig.5.208 t51.428 Sig.5.560 F5.087 Sig.5.967 F5.443 Sig.5.776

More market F5.233 Sig.5.792 t5.791 Sig.5.431 F51.631 Sig.5.077 t51.300 Sig.5.218 F5.802 Sig.5.495 F5.362 Sig.5.835

220 Younes Benslimane, Michel Plaisent and Prosper Bernard & Investigating Search Costs and Coordination Costs in Electronic Markets

Downloaded By: [Schmelich, Volker] At: 20:44 23 March 2010

Method of analysis

rho was .346 (Sig.5.000) and the computed t value was 23.142 (Sig.5.003).

Two statistical techniques were used to test the research hypotheses on WWW usage for procurement purposes. First, because all the variables under study were metric, the Pearson correlation was used to test the hypothesized relationships. Second, WWW usage, reduced search costs and reduced coordination costs were later recoded into variables distinguishing, for the first variable, the group of respondents with a high level of WWW usage (i.e., topthird of users) from the group with a low level of WWW usage (i.e., bottom-third of users); for the second variable, the group of respondent who experienced an important reduction of search costs from the group with the lowest reduction of search costs; and for the third variable, the group of respondents who experienced an important reduction of coordination costs from the group with the lowest reduction of coordination costs. In all three cases, the group in the middle was discarded. For each recoded variable, the two groups were of comparable size and t tests were used to assess the stability of the results from the Pearson correlation. A 5% level of significance was used for all the statistical tests. Findings are summarized in Table 4.

Hypothesis testing Hypothesis 1. Results of both the Pearson correlation and the independent-sample t test show that WWW usage is significantly associated with a reduction in corporate buyers’ search costs. The computed Pearson’s

Hypothesis 2. Findings from the two tests show conflicting results. Whereas the Pearson correlation (rho5.231 and Sig.5.015) shows that WWW usage is significantly associated with a reduction in coordination costs, the t test shows no support for the hypothesized relationship (computed t value521.573 and Sig.5.121).

Hypothesis 3. Findings from both tests show support for the positive relationship between reduced search costs and the move to ‘more market’. The Pearson’s rho and the computed t were .501 (Sig.5.000) and 23.993 (Sig.5.000) respectively.

Hypothesis 4. Results of both statistical tests support the expected relationship between reduced coordination costs and the move to ‘more outsourcing’. The Pearson’s rho was .442 (Sig.5.000) and the computed t was 24.803 (Sig.5.000).

DISCUSSION The WWW is an electronic marketplace that links users acting as potential buyers to thousands of vendors that have set up commercial websites. Corporate buyers using the WWW to carry out their procurement tasks can expect some benefits. These benefits would depend on the commercial websites’ functionalities buyers use during the procurement process.

Table 4. Hypotheses testing

Hyp.1 RSC Hyp.2 RCC

Hyp.3 MM

Hyp.4 MO

High-level users (N534)

Low-level users (N532)

Mean

Mean

St. Dev.

4.48 1.65 2.87 1.46 Users with low RSC (N533)

Mean

Mean

23.142 21.573 (NS) Computed t (t test)

.346** .231* Pearson’s rho (N5110)

23.993 Computed t (t test)

.501** Pearson’s rho (N5110)

24.803

.442**

St. Dev.

4.99 1.57 Users with high RCC (N533)

3.46 1.56 Users with low RCC (N534)

Mean

St. Dev.

Mean

St. Dev.

3.02

1.44

1.70

0.66

** Significant at the .01 level * Significant at the .05 level (NS) Not significant at the .05 level

Pearson’s rho (N5110)

St. Dev.

5.64 1.32 3.45 1.54 Users with high RSC (N534) St. Dev.

Computed t (t test)

Downloaded By: [Schmelich, Volker] At: 20:44 23 March 2010

Electronic Markets Vol. 15 No 3

When using the WWW to support their identification and selection tasks, corporate buyers can expect to reduce their search costs. Our findings show that WWW usage is associated with a reduction in search costs, which may suggest that WWW electronic communication and brokerage effects help corporate buyers to quickly and easily identify potential suppliers and, when the relevant information is available on the websites, to select their suppliers better. When using the WWW to place orders, make payments or check the status of their purchases, corporate buyers can benefit from its electronic integration effect, that is, the additional information and/or additional information processing capacity that help buying organizations reduce their external coordination costs. Results of this survey do not clearly support the hypothesized relationship between WWW usage and reduced coordination costs: only the correlation analysis shows support for the relationship and that relationship is not as strong as the one between WWW usage and reduced search costs. Such finding may appear surprising when an abundant IS literature presents the decrease of coordination costs as a consequence of IOS usage. A possible explanation is that IS literature has investigated IOS such as Electronic Data Interchange, Value Added Networks and other proprietary electronic marketplaces used extensively for the online execution of transactions and that tend to be more integrated to other IS in buying organizations. In this context, benefiting from WWW-based reduced coordination costs would imply first that corporate buyers are extensively using transactional websites and second that suppliers’ transactional websites are – to some extent – integrated to buying organizations’ IS. Descriptive statistics for our data set show that respondents use the WWW essentially to identify and select suppliers and marginally to execute transactions. As a result, WWW usage should be more associated with a reduction in search costs than with a reduction in coordination costs. As for the possible integration of a buying organization’s IS to a supplier’s commercial website, although not measured in this research, one may expect it to be low as the power of the WWW stems from the fact that it is an open electronic marketplace that requires no particular IT investment from buying organizations. Another possible explanation is that this ‘free for all IOS’ does not help test for trust and cooperation between business partners. Trust and cooperation have been found to influence IOS adoption, as these IOS often require that partners invest in specific assets that would have a much lower value in an alternative use should the original business relationship be prematurely terminated (Williamson 1985). From this perspective, one can argue that the uncertainty responsible for the coordination problem is not reduced only by additional information and information processing capacity but also by trust and cooperation between business partners (Iacovou et al. 1995) and by personal

221

and social relationships that may not be easily developed through electronic interfaces (Kraut et al. 1999; Uzzi 1997). WWW-based reductions in search costs and in coordination costs are themselves associated with changes in organizations’ boundaries. In accordance with past theoretical research, our model suggests that transactions involving standard inputs and occurring at a lower frequency are likely to change governance mechanisms as a result of such reductions in search costs and coordination costs. For those transactions, the reduction in coordination costs is associated with a move to ‘more outsourcing’, whereas the reduction in search costs is associated with a move to ‘more market’, implying that buyers will deal with new and more suppliers in order to reduce the total cost of inputs. Our data show evidence for both moves. First, respondents who report important reductions in coordination costs tend to also report a distinctive shift to ‘more outsourcing’. Based on the TCE framework, low-frequency activities do not benefit from economies of scales or specialization effect and should have a lower production cost in the market. The WWW-based reduced coordination costs and the lower production cost in the market make outsourcing these activities a logical solution. This shift to outsourcing is even more likely to occur when the activity to be externalized involves a standard input, which is normally widely available on the market. Second, respondents who report important reductions in search costs tend to also report an important move to ‘more market’. As an electronic marketplace, the WWW can reduce buyers’ search costs and increase price-based competition among vendors, making it easy and valuable for users to search the market for a better deal before purchasing inputs. In accordance with TCE-based literature, results from this survey show that transactions occurring at a low frequency and involving standard inputs are more likely to be subject to that shift because, normally, they do not benefit from price discounts and are considered low risk. Because of the low risk associated with standard inputs and of the absence of price discounts associated with infrequent purchases, buyers face no switching costs. This absence of switching costs, the reduced search costs and the increased pricebased competition in such electronic marketplaces make the shift from electronic hierarchies to electronic markets a logical solution. It should, however, be noted that, based on descriptive statistics from Table 4, the move to ‘more market’ is more distinctive than the move to ‘more outsourcing’. This suggests that WWW usage is more often linked to organizations dealing with new suppliers for inputs they were already buying than for inputs they were producing themselves. This limited move to ‘more outsourcing’ may be due to the less important reduction in coordination costs previously discussed. It may also be due to a restricted number of activities eligible to such a

222 Younes Benslimane, Michel Plaisent and Prosper Bernard & Investigating Search Costs and Coordination Costs in Electronic Markets

move. Ideally, future research should assess how many activities eligible to such a move were carried out internally prior to WWW usage and then assess how many were outsourced as a result of WWW usage. Such a review could determine whether or not the limited move to ‘more outsourcing’ is simply due to the fact that organizations are already focusing on their core activities.

Downloaded By: [Schmelich, Volker] At: 20:44 23 March 2010

CONCLUSIONS Findings from this study indicate that the WWW can be used to support the procurement process. Its electronic communication and brokerage effects are found to be associated with a reduction of corporate buyers’ search costs, which in turn, is associated with an increase in the base of suppliers for low-frequency transactions involving standard inputs. Results of this study show however no clear support for the electronic integration effect of the WWW. Nonetheless, they indicate that the WWWbased reduction in coordination costs is associated with a shift to outsourcing. In accordance with TCE literature, low-frequency activities involving standard inputs are more likely to be outsourced. The contribution of this research is two-fold. The first contribution is to research. Many IS researchers have stressed the need for studies that formally validate previous theoretical and anecdotal work on the TCEbased effects of electronic markets usage. This study, which involved 110 corporate buyers from over 100 organizations operating in multiple industries, fills that gap and aims to advance our understanding of the possible consequences of using electronic marketplaces for procurement purposes. However, similar studies in different settings are necessary to verify the generalizability of our results. Further research should also take into account additional factors that can explain the difference in reduction between search costs and coordination costs and the difference in strength between the move to ‘more market’ and the move to ‘more outsourcing’. The second contribution is to practice, as this research gives buyers, vendors and intermediaries insights on WWW usage for EC purposes. This study demonstrates that, by supporting the procurement process, the WWW can improve the performance of both buyers and buying organizations. Second, it shows that, if a web presence can be rewarding, vendors in electronic markets concerned with profitability must be ready to face the increased competition. This research finally indicates that the WWW can affect intermediaries by either threatening their survival as it helps reduce buyers’ search costs or by offering them new opportunities as it creates new needs for buyers and vendors in electronic marketplaces.

References Armstrong, S. and Overton, T. (1977) ‘Estimating NonResponse Bias in Mail Surveys’, Journal of Marketing Research 14: 396–402. Bakos, Y. (1991a) ‘A Strategic Analysis of Electronic Marketplaces’, MIS Quarterly 15: 295–310. Bakos, Y. (1991b) ‘Information Links and Electronic Marketplaces: The Role of Interorganizational Information Systems in Vertical Markets’, Journal of Management Information Systems 8: 31–52. Bakos, Y. (1997) ‘Reducing Buyer Search Costs: Implications for Electronic Marketplaces’, Management Science 43: 1676–92. Bakos, Y. and Brynjolfsson, E. (1997) ‘Information Technology, Incentives, and the Optimal Number of Suppliers’, Journal of Management Information Systems 10: 37–53. Bakos, Y. and Kemerer, C. (1992) ‘Recent Applications of Economic Theory in Information Technology Research’, Decision Support Systems 8: 365–88. Barua, A. and Lee, B. (1997) ‘An Economic Analysis of the Introduction of an EDI System’, Information Systems Research 8: 398–422. Barua, A., Ravindran, S. and Whinston, A. (1997) ‘Efficient Selection of Suppliers over the Internet’, Journal of Management Information Systems 13: 117–37. Brynjolfsson, E., Malone, T., Gurbaxani, V. and Kambil, A. (1994) ‘Does Information Technology Lead to Smaller Firms?’, Management Science 40: 1628–45. Choudhury, V., Hartzel, K. and Konsynski, B. (1998) ‘Uses and Consequences of Electronic Markets: An Empirical Investigation in the Aircraft Parts Industry’, MIS Quarterly 22: 471–507. Churchill, G. (1979) ‘A Paradigm for Developing Better Measures of Marketing Constructs’, Journal of Marketing Research 16: 64–73. Ciborra, C. (1993) Teams, Markets and Systems, Business Innovation and Information Technology, Cambridge: Cambridge University Press. Clemons, E. and Row, M. (1992) ‘Information Technology and the Industrial Cooperation: The Changing Economics of Coordination and Ownership’, Journal of Management Information Systems 9: 9–28. Clemons, E. and Row, M. (1993) ‘Limits to Interfirm Coordination through Information Technology: Results of a Field Study in Consumer Packaged Goods Distribution’, Journal of Management Information Systems 10: 73–96. Clemons, E., Reddi, S. and Row, M. (1993) ‘The Impact of Information Technology on the Organization of Economic Activity: The ‘‘Move to the Middle’’ Hypothesis’, Journal of Management Information Systems 10: 9–35. Coase. R. (1993) [1937] The Nature of the Firm, O. Williamson and S. Winter (eds), Oxford: Oxford University Press.

Downloaded By: [Schmelich, Volker] At: 20:44 23 March 2010

Electronic Markets Vol. 15 No 3 Croom, S. (2000) ‘The Impact of Web-Based Procurement on the Management of Operating Resources Supply’, Journal of Supply Chain Management 36: 4–13. DeLone, W. and McLean, E. (1992) ‘Information Systems Success: The Quest for the Dependent Variable’, Information Systems Research 3: 60–95. DeLone, W. and McLean, E. (2003) ‘The DeLone and McLean Model of Information Systems Success: A TenYear Update’, Journal of Management Information System 19: 9–30. Dewan, S., Michael, S. and Min, C.-K. (1998) ‘Firms Characteristics and Investments in Information Technology: Scale and Scope Effects’, Information Systems Research 9: 219–32. Ghosh, S. (1998) ‘Making Business Sense of the Internet’, Harvard Business Review 76: 126–35. Grossman, S. and Hart, O. (1986) ‘The Costs and Benefits of Ownerships: A Theory of Vertical and Lateral Integration’, The Journal of Political Economy 94: 691–719. Gurbaxani, V. and Whang, S. (1991) ‘The Impact of Information Systems on Organizations and Markets’, Communications of The ACM 34: 59–73. Hart, O. and Moore, J. (1990) ‘Property Rights and the Nature of the Firm’, The Journal of Political Economy 98: 1119–58. Hitt, L. (1999) ‘Information Technology and Firm Boundaries: Evidence from Panel Data’, Information Systems Research 10: 134–49. Iacovou, C., Benbasat, I. and Dexter, A. (1995) ‘Electronic Data Interchange and Small Organizations: Adoption and Impact of Technology’, MIS Quarterly 19: 465–85. Jones, M. and Beatty, R. (1998) ‘Towards the Development of Measures of Perceived Benefits and Compatibility of EDI: A Comparative Assessment of Competing First Order Factor Models’, European Journal of Information Systems 7: 210–20. Kambil, A. (1991) ‘Information Technology and Vertical Integration: Evidence from the Manufacturing Sector’, in M. Guerin-Calvert and S. Wildman (eds) Electronic Services Networks, New York: Praeger, pp. 22–38. Kraut, R., Steinfield, C., Chan, A., Butler, B. and Hoag, A. (1999) ‘Coordination and Virtualization: The Role of Electronic Networks and Personal Relationships’, Organization Science 10: 722–40. Lacity, M. and Willcocks, L. (1995) ‘Interpreting Information Technology Sourcing Decisions from a Transaction Cost Perspective: Findings and Critique’, Accounting, Management and Information Technology 5: 203–44. Lacity, M. and Willcocks, L. (1998) ‘An Empirical Investigation of Information Technology Sourcing Practices: Lessons from Experience’, MIS Quarterly 22: 363–408. Lightner, N. (2004) ‘Evaluating E-Commerce Functionality with a Focus on Customer Service’, Communications of the ACM 47: 88–92.

223 Liu, C., Arnett, K., Capella, L. and Beatty, R. (1997) ‘Web Sites of the Fortune 500 Companies: Facing Customers through Home Pages’, Information and Management 31: 335–45. Malone, T., Yates, J. A. and Benjamin, R. (1987) ‘Electronic Markets and Electronic Hierarchies’, Communications of the ACM 30: 484–97. Massetti, B. and Zmud, R. (1996) ‘Measuring the Extent of EDI Usage in Complex Organizations: Strategies and Illustrative Examples’, MIS Quarterly 20: 331–45. Mukhopadhyay, T., Kekre, S. and Kalathur, S. (1995) ‘Business Value of Information Technology: A Study of Electronic Data Interchange’, MIS Quarterly 19: 137–57. Novack, R. and Simco, S. (1991) ‘The Industrial Procurement Process’, Journal of Business Logistics 12: 145–65. Nunnally, J. (1988) Psychometric Theory, Englewood Cliffs, NJ: McGraw-Hill. O’Connor, G. and O’Keefe, B. (1997) ‘Viewing the Web as a Marketplace: The case of Small Companies’, Decision Support Systems 21: 171–83. Perkins, A. (2001) ‘Life after the Great Internet Company Shake Out’, online at: http://www.gsm.ucdavis.edu/ innovator/winter2001/ShakeOut.pdf [accessed 22 October 2004]. Poon, S. and Swatman, P. (1999) ‘An Exploratory Study of Small Business Internet Commerce Issues’, Information and Management 35: 227–44. Reekers, N. and Smithson, S. (1996) ‘The Role of EDI in Interorganizational Coordination in the European Automotive Industry’, European Journal of Information Systems 5: 120–30. Reimers, K. (1996) ‘The Non-Market Preconditions of Electronic Markets: Implications for their Evolution and Applicability’, European Journal of Information Systems 5: 75–83. Rosen, K. and Howard, A. (2000) ‘E-Retail: Gold Rush or Fool’s Gold?’, California Management Review 42: 72–100. Rosenthal, D., Shah, S. and Xiao, B. (1993) ‘The Impact of Purchasing Policy on Electronic Markets and Electronic Hierarchies’, Information and Management 25: 105–17. Scala, S. and McGrath, R. (1993) ‘Advantages and Disadvantages of Electronic Data Interchange’, Information and Management 25: 85–91. Uzzi, B. (1997) ‘Social Structure and Competition in Interfirm Networks: The Paradox of Embeddedness’, Administrative Science Quarterly 42: 35–67. Williamson, O. (1975) Markets and Hierarchies, New York: The Free Press. Williamson, O. (1985) The Economic Institutions of Capitalism, London: The Free Press.

224 Younes Benslimane, Michel Plaisent and Prosper Bernard & Investigating Search Costs and Coordination Costs in Electronic Markets

Downloaded By: [Schmelich, Volker] At: 20:44 23 March 2010

Appendix A: Sections of the questionnaire used in this research