1

TO BE OR NOT TO B2B? AN EVALUATIVE MODEL FOR E-PROCUREMENT CHANNEL ADOPTION

Qizhi Dai

Doctoral Program in Information and Decision Sciences

Robert J. Kauffman Associate Professor of Information and Decision Sciences

Carlson School of Management

University of Minnesota, Minneapolis, MN 55455 Email: qdai; rkauffman@csom.umn.edu

ABSTRACT

With the popularity of commercial use of the Internet and WWW, business-to-business (B2B) e-commerce and e-procurement are moving corporate purchasing to the Web. Basically, there are two types of B2B e-commerce business models. Extranets connect the buyer and its suppliers with a closed network. In contrast, electronic markets create open networks for buyer and supplier interactions. Extensive IS and economic research has studied the benefits and costs of interorganizational information systems (IOS), and has examined the changes in buyer-supplier relationships caused by implementation of IOS. For example, Bakos and Brynjolfsson (1993) have argued that IT will turn buyer-supplier relationships into partnerships. Yet recently, we have observed a new trend towards the use of electronic markets as an alternative channel for buyers to search for and exchange supplies with a large supplier base. What are the motivations for buyers to move from extranets to electronic markets? Why is this value-maximizing for the firms? This paper models this choice and indicates that the buyer’s decision about an e-procurement approach depends on: (1) desired gains from lower search costs and operation costs enabled by an electronic market; (2) the importance of information sharing between its suppliers; (3) the competition present in the supplier market; and (4) the desired levels of supplier relationship-specific investments. KEYWORDS: Business-to-business e-commerce, buyer-supplier relationships, e-business,

electronic markets, e-procurement, interorganizational information systems

ACKNOWLEDGEMENTS The authors would like to acknowledge the useful input of Alina Chircu, Baba Prasad and the participants of the "Information Systems, Electronic Commerce and Economics" doctoral seminar during Spring 2000 at the Carlson School of Management at the University of Minnesota.

2

INTRODUCTION Broadly speaking, business-to-business (B2B) e-commerce involves conducting business

with suppliers, customers and other companies through computer and telecommunication

technologies. The Internet provides new opportunities for developing electronic links among

businesses. Since it is a public network infrastructure, the costs of conducting e-business on the

Internet are greatly reduced. Connecting business partners via the Internet for information and

data exchange is growing rapidly. By 2004, B2B e-commerce is predicted to grow to more than

$7 trillion in the United States (Sapp, 2000). Supporting this prediction is another recent trend:

companies increasingly also are moving to purchase and procure supplies on the Internet (Ovans,

2000; Sawhney and Kaplan, 1999). As a result, new business models in supply chain

management are emerging that are based on the Internet and the World Wide Web, and that aim

to create new value within traditional supply chains.

The changes in the way that firms purchase goods and services occur across three

purchasing stages: information search, negotiation and settlement (Gebauer and Segev, 1998).

E-catalogs, for example, allow buyers to browse and search online, online auctions and bidding

support real-time negotiation, and electronic bill payment provides new mechanisms for financial

settlement. As a whole, these new electronic markets of the World Wide Web facilitate the basic

functions that markets are supposed to deliver: matching and aggregation. To aid managerial

decision makers in understanding how to decide between alternatives for e-procurement, in this

article we will derive a choice model that employs theories from information systems and

economics research, and that emphasizes the contrasting aspects of different e-procurement

approaches.

Extranets and Electronic Markets

3

B2B e-commerce is built upon interorganizational information systems (IOS) or

extranets, and electronic markets.

IOS and Extranets. Interorganziational systems (IOS) are automated information

systems that create, store and exchange information between business partners (Cash and

Konsynski, 1985). During the 1980s, organizations began to deploy information systems to

support information sharing and communications with suppliers, distribution channels and

customers. These IOS started the practice of B2B e-commerce and electronic procurement. The

functions of IOS range from simple order entry and invoicing, to product promotion, to

document and data sharing, to joint product development and knowledge transfer (Johnston and

Vitale, 1988; Riggins and Rhee, 1999; Chatfield and Yetton, 2000).

One important type of IOS is electronic data interchange (EDI), the business-to-business

exchange of electronic documents in a standard machine-processable format. Using EDI,

business partners are able to electronically exchange structured documents, such as purchase

orders and invoices, with little or no human intervention. Companies typically use EDI solutions

to achieve more efficient data and information management; they reduce processing time and

eliminate redundant data entry. The potential benefits of EDI are attractive, but the costs of

implementing and running such systems solutions are relatively high. A typical first-time

implementation of EDI for a small firm, including hardware, software, implementation and

training, is estimated to cost around $50,000 (Waltner, 1997). Besides the initial investment,

companies using EDI have to pay in excess of $100,000 per year in value-added-network (VAN)

charges. These large expenditures create barriers for small and medium-sized companies that

want to adopt EDI. This reduces the net benefits that such extranets can provide to the

companies.

4

The commercialization of the Internet has brought about Internet-based EDI or Web EDI,

by which firms transmit data via public network infrastructure instead of proprietary value-added

networks or VANs (Riggins and Mukhopadhyay, 1999). By Web EDI, small businesses are able

to send ordering or invoicing data using web browsers to business partners, especially their large

EDI-enabled customers. Although Web EDI systems aim to support whole business processes

involving manufacturing, billing, shipping, accounting and so on, these capabilities are not fully

implemented. Most current Web EDI solutions function as message translators between EDI

messages and data collected from Web forms in order to bridge the gap between small

businesses and their EDI-enabled big customers or suppliers. For example, 3Com built a Web

EDI system for its customers to enter orders using web forms which will be sent directly into

3Com’s database.

Another form of Internet-based B2B e-commerce is the extranet which is a secure,

private, Web-based network, providing suppliers, customers and other business partners the

access to the initiator’s corporate databases, or facilitating collaborative tasks among a group of

organizations (Riggins and Rhee, 1998). Via extranets, firms not only can order and purchase

from suppliers, but they also can share product and sales information with each other.

Although IOS like EDI and extranets utilize different technologies, they share some

important features with respect of business models. First of all, companies conduct transactions

electronically on these networks. More importantly, information sharing is common between the

initiator and its business partners via electronic document exchange and controlled data retrieval.

Another key feature is that they are proprietary closed networks in the sense that these networks

only open to pre-selected business partners. Considering these characteristics, we think of these

5

networks as one type of e-procurement channel, and use extranets or IOS interchangably to refer

to such IT-based procurement channels in the following discussion.

Electronic Markets for E-Procurement. According to Kaplan and Sawhney (2000),

there are four types of business models for electronic marketplaces, classified by the type of

product and the characteristics of transactions they support:

q MRO hubs (maintenance, repair, and operation) enable systematic purchasing for

operating supplies and they are horizontal markets.

q Yield managers are horizontal markets supporting spot purchasing for operating supplies.

q Electronic catalog hubs provide integrated product information in vertical markets and

are used for systematic purchasing for manufacturing inputs.

q Exchange hubs are vertical markets for spot purchasing of manufacturing inputs.

To participate in these markets, firms first must register on the markets as members.

Then they must be verified by the market makers. With a verified membership, participating

firms then can post product catalogs and requests for proposals, and participate in auctions. Some

B2B e-markets even provide the capability of streamlining procurement processes via software

systems that integrate with corporate enterprise systems and organizational intranets. In order to

transact on these online markets, firms either pay license fees, or transaction-based fees, or

monthly fees, or combinations of quote fees and transaction fees. For example,

ChemConnect.com (www.chemconnect.com), for each transaction, charges the buyer and the seller

either 0.1% of the transaction sales value or $250, whichever is higher. Compared with the costs

of setting up and maintaining EDI systems, e-markets are more cost-efficient.

Along with electronic catalogs, electronic auctions and other capabilities, electronic

markets aggregate product and price information, match supply and demand, and facilitate

6

transactions. Through these online markets, buyers can do one-stop, comparison shopping for

thousands of suppliers and select the best source in real time. They also can negotiate with

suppliers, place orders, make payments and receive invoices. Fundamental features of these

procurement mechanisms distinguish them from extranets are. They are open networks with

potentially larger pools of business partners for their member firms. And they give firms more

flexibility and opportunities of searching for and selecting suppliers or customers. On the other

hand, these networks are inferior to extranets in supporting information sharing and collaboration

between suppliers and buyers.

Theorizing About Channel Adoption. Buying firms are faced with choosing between

two adoption alternatives: extranets and the new electronic procurement markets of the Internet.

Theories about electronic markets argue that information technologies (ITs) facilitate the market

functions of matching and aggregating by reducing search costs and coordination costs (Malone,

Yates and Benjamin, 1987). Lower search costs, in turn, enable buyers to search more and thus

obtain better offers (Bakos, 1991). Electronic markets also bring a larger potential supplier

(customer) base to buyers (suppliers) due to their abilities to bridge both spatial and temporal

distances. And, as a larger business network attracts adopting firms (Kauffman, McAndrews and

Wang, 2000), electronic markets become an even more advantageous procurement channel for

buyers.

Contrary to the positive network externalities exhibited in electronic markets, negative

network effects are found in suppliers’ adoption of IOS (Riggins, Kriebel and Mukhopadhyay,

1994), and as a result, buyers usually induce suppliers to join the EDI systems by subsidy or

punishment (Barua and Lee, 1997). A commonly observed result of adopting IOS is a reduced

number of suppliers transacting with a particular buyer. This is because the buyer wants to give

7

its suppliers incentives to make non-contractible relationship-specific investments (Bakos and

Brynjolfsson, 1993). Another reason is that an IOS links the buyer and its suppliers more closely

through electronically sharing data and information. This reduces the buyer’s bargaining power

and then encourages the buyer to develop closer ties with a small number of suppliers (Clemons,

Row and Reddi, 1994). But, empirical research has found that firms can obtain more business

value by exchanging strategic data and information with business partners via IOS (Chatfield and

Yetton, 2000). The interorganizational coupling enabled by the IOS actually improves the firm

performance based on enhanced interdependence and coordination between buyers and suppliers

(Clark and Lee, 2000).

Starting from these theories and prior research, we will answer the basic (Shakespearian)

question posed in the title of this paper: “To be or not to B2B?” What e-procurement channel

should be selected, depending on the competitive conditions, firm characteristics, and the various

important qualities of the systems solutions? An extranet? Or an electronic market? And how

important are network externalities? To answer, we will develop and analyze a formal model of

buyers’ choices in adopting the extranet and electronic market e-procurement channels.

BENEFITS AND COSTS OF E-PROCUREMENT EXTRANETS AND E-MARKETS

We next consider the benefits and costs of two potential e-procurement channel choices:

extranets and electronic markets. Our assessment provides the basis for formal modeling of the

benefits and costs, in a manner that reflects what we know about the referent literature and

current industry practice.

Extranet Benefits and Costs

Research on IOS, including traditional EDI systems, provides useful insights for

examining the benefits and costs of adopting extranets for e-procurement. Traditional IOS

8

require a lot of expenses for setting up the systems. These costs include expenses for computer

systems, system supporting personnel and training. Both buyers who initiate these systems and

the suppliers who participate in the systems have to bear such costs. For a large volume buyer,

these costs may be compensated by benefits. But for a small supplier, these costs may prohibit

adoption.

When it comes to the benefits of IOS, Mukhopadhyay, Kekre and Kalathur (1995), and

Riggins and Mukhopadhyay (1994) found that the implementation of an IOS improved the

operational efficiency of the buyers who initiated the systems. Using electronic means for

purchasing, buying companies can reduce order entry costs and inventory management costs, and

at the same time get quicker responses from their suppliers.

For suppliers participating in the systems, timely information about customer demand

may help them to forecast demand and schedule their production more efficiently. Suppliers can

also save expenses on order entry and at the same time improve customer services. Such

benefits deliver strategic value to the suppliers, which cannot be obtained without joining the

systems.

Moreover, the participating suppliers obtain a competitive advantage relative to their

rivals who do not participate:

q First, the participating supplier will be able to increase its sales because the buyer may

reduce its orders to those suppliers who do not participate and increase orders to those

who participate (Barua and Lee, 1997).

q Second, with access to the buyer’s inventory, sales and product information, the

participating supplier will be able to accumulate expertise about market demand and

9

product features, which usually will lead to product innovation and market expansion

(Subramani, 1999; Riggins and Rhee, 1999).

Another important benefit of extranets is that they give the suppliers access to the buyers’

business information, including inventory management data, and product design and sales data.

Once suppliers have such information, buyers will find it difficult to control the use of these

resources and inevitably become more vulnerable to supplier opportunism (Clemons, Reddi and

Row, 1993). As a result, suppliers will gain bargaining advantage relative to their buyers for

pricing and more beneficial contract terms. From the viewpoint of buyers, this is a cost due to

sharing business information with the suppliers.

In other research on buyer-supplier relationships, Bakos and Brynjolfsson (1993) found

that there are non-contractible investments from the side of the suppliers that benefit the buyer in

terms of responsiveness, higher quality, innovation and technology adoption. To protect

suppliers’ incentives to make these non-contractible and relationship-specific investments, the

buyer typically will keep its supplier network small.

Electronic Markets Benefits and Costs

Similar to extranets and EDI systems, e-procurement via electronic markets also

improves purchasing efficiency for buying companies. At the same time, e-procurement

applications can cost as much as other enterprise information systems. For example, the

licensing fees for these systems are reported to range from $1,000 to $4 million (Waltner, 1999).

Thus, the system setup costs are comparable for adopting either extranets or electronic markets.

However, the operational costs associated with the adoption of an electronic market approach are

lower than with an extranet approach. The reason is that the buyer needs only one link via an

10

electronic market -- instead of several links with an extranet mechanism -- to be electronically

connected with its suppliers.

In contrast to closed extranets, electronic markets are open networks where buyers have

access to more product and price information. In recent studies of automatic teller machine

(ATMs) adoption in consumer banking, Kauffman and Wang (1999), Kauffman, McAndrews

and Wang (2000) argued that banks that expect to obtain a larger customer base will adopt ATM

networks earlier and will achieve higher profits. By the same token, we argue that the potential

access to a larger supplier base will attract buyers to join electronic markets, by virtue of the

positive network externalities that are created on the buyers’ side. At the same time, search costs

and coordination costs are reduced in electronic markets (Bakos, 1991), and buyers are more

likely to receive better offers than they can get with extranets.

With extranets, buyers have long-term contracts with suppliers, and prices are preset even

when market prices change. In electronic markets, however, the buyer will do more searches, or

participate in more auctions to hunt for information and get better offers (Barua, Ravindran and

Whinston, 1997). In addition, the level of information sharing between buyers and suppliers will

be lower than in extranets. Suppliers may obtain ordering information but not inventory

information. As a result, the suppliers’ bargaining advantage relative to the buyer and

competitive advantage relative to its rivals will be less compared to what can be achieved with

extranets.

Since the supplier knows that there is a large number of potential suppliers in the

marketplaces whose product and price information is available at the mouse click of the buyer,

the supplier will feel the threat that the buyer may switch to its rivals easily. Using Shapley

value analysis from game theory, Raupp and Schober (2000) argued that when the number of

11

suppliers increases, the value that one supplier can obtain from the buyer-supplier relationship

decreases. This lower ex post gain will reduce the supplier incentive to make non-contractible

investments. Therefore, in electronic markets, suppliers will make fewer such investments than

in extranets.

MODEL AND ANALYSIS

Seidmann and Sundararajan (1997) proposed a general framework for describing the

contract games that occur between one buyer and two suppliers in choosing levels of information

sharing in supply chain management. In this paper, we extend their model to understand

decision making regarding the adoption of extranets versus electronic markets for e-procurement.

(We will refer to the choice of an extranet as E, and the choice of an electronic market as M in

the remainder of the paper.)

Modeling Preliminaries

Let us assume that there is one big buyer and two suppliers. The buyer is going to make

a choice between an extranet and an e-market mechanism for procurement, and the suppliers will

decide whether to join the buyer network. This is a simplification of the real world cases where

big buyers initiate electronic procurement channels and request suppliers to participate. For

example, Chrysler launched its EDI program in 1984 and almost all its suppliers adopted this

network by 1990 (Mukhopadhyay, kekre and Kalathur, 1995). More recently, when big buyers

start to move to online markets, they also request suppliers to join the e-markets, just like

Schlumberger, which asked its suppliers to participate in CommerceOne’s MarketSite when it

chose MarketSite for purchasing office supplies (Ovens, 2000). We also assume that the two

suppliers are symmetric: that is, the benefits and costs are the same for the suppliers. As we

discussed in previous sections, the buyer will obtain savings i from reduced inventory and

12

ordering costs. (See Table 1 for an overview of all the variables that we will use in our model.)

No matter what type of information systems are in use, either an extranet, E, or an electronic

market, M, mechanism, the buyer is able to obtain the same cost reductions that derive from

improved internal efficiencies. Considering the fact that the system deployment costs incurred

by the buyer firms are comparable for adopting either extranets or electronic markets, we do not

include these setup costs in our model.

In adopting the extranet approach, the supplier who is participating in the system will

have access to the buyer’s inventory information and will know how its product is sold, and thus

the information asymmetry will be reduced (Seidmann and Sundararajan, 1997). Benefiting

from this, the supplier will obtain the benefits from greater relative bargaining power, b,

compared to the buyer. The buyer will think of this as receiving a payment for achieving

efficiency. However, when the buyer-supplier relationship occurs in an electronic market, this

value will be discounted by d, with 0 < d < 1. This discount is due to the fact that in the market

the supplier will have only limited information about how the buyer is consuming its products.

The information available to the supplier is provided by the buyer, and it may be incomplete or

may not be what the supplier desires.

13

Table 1. Definitions of Variables

Variable Name

Definition Comments

B Supplier benefits from bargaining power over the buyer

Suggested by Clemons, Reddi and Row (1993)

C

Supplier competitive advantage (disadvantage) when only one supplier is doing business electronically with the buyer

Derived from Seidmann and Sundararajan (1997), Riggins and Rhee (1999), and Subramani (1999)

D

Discount rate, 0 < d < 1, on competitive advantage (disadvantage) from information sharing in electronic markets relative to the extranet channel

I Buyer benefits from improved efficiency

Suggested by Mukhopadhyay, Kekre, and Kalathur (1995)

K Supplier’s system setup costs in adopting an extranet approach

N Buyer’s benefits from adopting an electronic market approach Based on Bakos (1991).

Q Competitiveness in supplier market, with 0 < q <1

S Supplier’s benefits from strategic value

Also motivated by Seidmann and Sundararajan (1997)

Sj Supplier j, where j ∈ {1, 2} For analytical tractability, we limit our analysis to just two supplier firms, as did Seidmann and Sundararajan (1997)

vE, vM

Supplier relationship-specific investment value with an extranet approach, E, or an electronic markets approach, M

vE1 and vE2 (vM1 and vM2 ) represent how much the buyer requests of a supplier for relationship-specific investment in its strategy to induce one supplier (either 1 or 2) to join its extranet (electronic market).

wE, wM

Buyer’s net welfare or benefits with an extranet (electronic market) approach

wE1 (wM1) is the buyer’s net benefits when only one supplier is joining the extranet, and wE2 (wM2) is the benefit when both suppliers are using extranets.

Being electronically connected with the buyer increases the supplier’s capability to better

forecast product demand and to schedule its production more efficiently. These strategic benefits

for the supplier, s, occur when only one supplier is electronically connected with the buyer.

When both suppliers do business electronically with the buyer, this advantage is reduced by a

14

factor of q, with 0 < q < 1 representing the competitiveness among suppliers. When q → 1, the

competition is very low; in contrast, as q → 0, competitive pressures become very high. If only

one supplier, S1, is using an extranet, then the supplier S2, who is not participating, will suffer a

loss in competitive position because S1 will accumulate the knowledge and expertise for product

innovation and market expansion based on its exclusive access to the buyer’s proprietary

information. We express the amount of this loss as c for S2. At the same time, S1 will obtain a

benefit c. Similarly, in electronic markets, S2 may also incur some loss because it does not have

the knowledge about the buyer’s demand that S1 has. However, the amount of this loss is

reduced in proportion to the degree of reduced information sharing between the buyer and S1,

i.e., S2 will lose dc and S1 will gain dc.

In addition, in electronic markets the buyer will enjoy low search costs and an expanded

supplier base. Moreover, the buyer will incur lower operating costs with an electronic market

solution than with an extranet. We consider this as another of the benefits that is derived from

adopting an electronic market mechanism. In our model, such benefits are expressed as n, which

are generated only in electronic markets.

In buyer-supplier relationships, the non-contractible value from the suppliers’

relationship-specific investments is represented as vE for extranets and vM for electronic markets.

We assume that vE and vM are paid by the supplier to the buyer. When they become negative, a

subsidy is required and the buyer has to pay for the supplier to participate in the systems. When

extranets are adopted, the suppliers have to incur setup costs k, including the expenses for

computer systems to communicate directly with the buyer.

Finally, our model enforces the logical constraint that the buyer will adopt either an

extranet or an electronic market approach, but cannot adopt both at the same time. As a result,

15

there will be no case where one supplier is using an extranet while the other supplier is using an

electronic market. Therefore, there are five possible situations requiring further analysis. (See

Table 2.)

Table 2. Non-Electronic and Electronic Procurement Cases

Case Name Description 1 Non-Electronic No electronic approach is adopted.

2 Focal Extranet The buyer implements an extranet involving just one supplier S1

3 Spanning Extranet The buyer uses an extranet to connect to both suppliers, S1

and S2.

4 Focal Market The buyer purchases from one supplier S1 using an electronic market mechanism.

5 Spanning Market The buyer purchases from both suppliers S1 and S2 using an electronic market mechanism.

The benefits and costs that materialize for each party in each situation are summarized in Table

3. The sums of the net benefits of the buyer and the two suppliers are displayed as Total in Table

3 to show the social welfare in the above five cases.

Table 3. Net Benefits of the Buyer and Suppliers

Case Name Buyer Supplier 1 Supplier 2 Total

1 Non- Electronic

0 0 0 0

2 Focal Extranet i -b+ vE b + s + c - vE - k -c i + s – k

3 Spanning Extranet

2i – 2b + 2vE b + qs –vE – k b + qs – vE- k 2i + 2qs - 2k

4 Focal Market i – db + vM + n db + s + dc - vM - dc i + s + n

5 Spanning Market

2I – 2db + 2vM + n db + q s – vM db +qs – vM 2i + 2qs + n

Adopting an Extranet (E) Approach

Let us examine Cases 2 and 3 where the buyer intends to use an extranet approach. Since

the buyer has two suppliers and there is a loss of competitive advantage for a supplier who does

16

not join the extranet if the other supplier does, the buyer can design a strategy to induce suppliers

to participate.

Proposition 1 – The Supplier-Specific Extranet Investment Proposition: The supplier relationship-specific investment will be less than the sum of the gain from bargaining power, the gain from strategic value, and twice the gain from competitive value, less the implementation costs. This is given by vE < b + s + 2c – k.

Proof. The highest possible gain for a supplier to agree to do business with the buyer via an

extranet is b + s + c – vE – k. The lowest possible gain is – c, when one supplier does not do

business this way and the other supplier does. Suppliers are rational, however, and will select an

extranet when they will be better off by so doing. That is, b + s + c – vE – k > - c. Thus, we get

vE < b + s + 2c – k. '

From the Supplier-Specific Extranet Investment Proposition, we see that the possible

relationship-specific investment the buyer can obtain from one supplier is capped at b + s + 2c –

k. Thus, a supplier will not be willing to make effort or investment costing more than this

amount that will be specific to the buyer.

Under what circumstances will both suppliers adopt an extranet approach? Our next

proposition analyzes this question in terms of the extent of the relationship-specific investments

that these firms must make:

Proposition 2 – The Supplier Extranet Adoption Proposition: If the relationship-specific investment vE satisfies vE < b + qs + c – k, then both suppliers will adopt an extranet with the buyer.

Proof. Assume that supplier S1 first joins the extranet. In this case, if supplier S2 also decides to

join the extranet, it will get b + qs – vE - k > – c. The latter value is what supplier S2 will get if it

does not join under the condition of vE < b + qs + c – k. Therefore, S2 will join the extranet. If

S1 does not join, however, then the net benefit S2 will obtain from joining the extranet will be b

+ s + c – vE – k, which is greater than zero with vE < b + qs + c – k. As a result, S2 will accept,

17

and thus S1 will end up with – c. But, with vE < b + qs + c – k, S1 will prefer to join the extranet

to get net benefit b + qs – vE – k, which is greater than – c. So, both suppliers will join the

buyer’s extranet. '

The Supplier Extranet Adoption Proposition indicates that if the buyer wants two

suppliers to both join the extranet, it will not get a relationship-specific investment from each

supplier that exceeds the sum of each supplier’s gain in bargaining power, the reduced gain from

strategic value and the gain from competitive value, less the implementation costs. If the buyer

wants more relationship-specific investment, it has to use other strategies to obtain the funding.

Now to obtain at least one supplier to participate in the extranet, the buyer may first try to

set up a connection with one supplier, say S1 (since the suppliers are symmetric, this arrangement

will not lose generality) requiring relationship-specific investment vE1. If S1 accepts, then the

buyer may stop. If S1 rejects, then the buyer may invite S2 to join with a relationship-specific

investment of vE2. The result is as follows:

Proposition 3 – The Focal Extranet Proposition: If vE1 < b + s +2c – k and b + qs + c – k < vE2 < b + s + c – k, then there will be two equivalent outcomes:

(3A). The buyer invites S1 to join the extranet with vE1 and S1 accepts.

(3B). The buyer invites S2 to join the extranet with vE1 and S2 accepts.

Proof. If S1 rejects when the buyer first invites it to join, then the buyer will invite S2 with

investment vE2 < b + s + c – k. In this case the net benefit S2 will obtain is b + s + c – k - vE2 >

0, and thus S2 will agree to join the buyer extranet. As a result, S1 will lose c. Using backward

induction, it can be shown that S1 will accept when the buyer initially invites it with vE1. Since

vE2 > b + qs + c – k, S2 will not join after S1 has already joined. S1 and S2 will not join together.

Therefore, the outcome will be that the buyer will work with either S1 or S2, and obtain supplier

relationship-specific investment vE1 < b + s +2c – k. '

18

The buyer’s net benefit will be different in Cases 2 and 3. Our examination of the

conditions that will occur when the buyer makes a choice yields the following result:

Proposition 4 – The Buyer’s Supplier-to-Extranet Connection Preference Proposition: The buyer will select the appropriate supplier-to-extranet connections based on the following preference conditions in terms of q, a measure of the firm’s competitiveness:

(4A). When q < (k + s - i)/2s, the buyer will prefer an extranet with one supplier, and its net benefit will not exceed i + s +2c – k.

(4B). When q > (k + s – i)/2s, the buyer will prefer an extranet with both suppliers, and its net benefit will not exceed 2i + 2qs +2c –2k.

(4C). When q = (k + s – i)/2s, the buyer will be indifferent about the number of supplier-to-extranet connections, and its net benefit will not exceed i + s +2c – k.

Proof. The Supplier Extranet Adoption Proposition shows that the highest possible relationship-

specific investment the buyer can obtain is b + qs + c – k. In this case, the buyer’s maximum net

benefit or welfare for an extranet will be:

wE2 = 2i – 2b + 2(b + qs + c - k) = 2i + 2qs + 2c – 2k

The Focal Extranet Proposition indicates that the buyer will get relationship specific investment

no more than b + s + 2c – k. This will give the buyer maximal net benefit:

wE1 = i – b + (b + s + 2c - k) = i + s + 2c – k

If q < (k + s – i) /2s, then

wE2 = 2i + 2qs + 2c – 2k < 2i + 2s*(k + s – i) /2s + 2c – 2k = i + s + 2c – k = wE1

As a value-maximizing firm, the buyer will choose suppliers based Max (wE2, wE1). In this case,

the buyer firm will invite only one supplier to join the extranet.

Similarly, if q > (k + s – i) /2s, then wE2 > wE1, and as a result, the buyer will prefer two

suppliers joining the extranet. If q = (k + s – i) /2s, then wE1 = wE2, and the buyer will be

indifferent. '

19

The Buyer’s Supplier-to-Extranet Connection Preference Proposition implies that when

the level of competition in the supplier market is high, the buyer will prefer to invite one supplier

to join its extranet. And when the competition is reduced to lower levels, the buyer will ask both

suppliers to join the extranet. Also note that when the buyer efficiency benefits i and supplier

strategic benefits s increase, it is more likely that both suppliers will participate in the extranet.

But when the supplier implementation costs k are high, this will increase the likelihood that only

one supplier will want to join the network. The buyer’s preference is illustrated in Figure 1.

(See Figure 1.)

Figure 1. The Buyer’s Supplier-to-Extranet Connection Preference

vE

wE wE2

wE1

2i - 2b

i - b

s + b + 2c - k b - i qs + b + c - k

q > (k+s-i)/2s

q < (k+s -i)/2s

A

B

C

Figure 1 depicts the buyer’s welfare (wE) as a function of the supplier’s relationship-

specific investment (vE) with an extranet approach. Line wE1 represents the possible buyer

welfare when one supplier joins the extranet, and point A (Va, Wa) shows the buyer’s maximum

possible welfare in this case. If both suppliers join the extranet, the buyer’s welfare will fall on

line wE2 and the maximum value is at point C (Vc, Wc). We are going to show how the change in

q changes the maximum possible benefits the buyer can obtain with such a mechanism, with all

20

other things being equal. This means that point C is fixed, while point A will move as q changes.

When q = (k + s – i)/2s, A and C have the same value along the y-axis which represents the

buyer’s welfare, i.e., Wa = Wc, indicating that the buyer is indifferent about the number of

connections with the suppliers. When q increases, line AB moves to the right. That is, the

buyer’s welfare with two suppliers joining its extranet increases to surpass the buyer welfare

when only one supplier joins, and thus A moves above C, resulting in Wa > Wc. This is the case

when q > (k + s – i)/2s, which is represented as the area on the right of line AB in Figure 1. In

this case, the buyer will prefer a spanning extranet strategy. On the other hand, when q < (k + s

– i)/2s, line AB moves to the left, causing A to be lower than C, or Wa < Wc. That is, the

buyer’s maximum possible welfare is less when both suppliers join in the extranet than if only

one supplier joins the extranet. This is shown by the area on the left of line AB. In this case, the

buyer will prefer the focal extranet strategy. Figure 1 also shows that when the supplier

relationship-specific investment that the buyer can obtain is less than b – i, the buyer’s welfare

will become negative, and thus it will not initiate the extranet. At the other end, the maximum

supplier relationship-specific investment is s + b +2c – k, beyond which neither supplier will

agree to join the buyer extranet.

Adopting an Electronic Market (M) Approach

Now, let us examine Cases 4 and 5 where the buyer decides to use an electronic market.

Proposition 5 -- Supplier-Specific Electronic Market Investment Proposition: The supplier relationship-specific investment will be less than the sum of the gain from reduced bargaining power, gain from strategic value and twice the gain from discounted competitive value. That is, vM < db + s + 2dc.

(Our proof follows the same logic as that for the Supplier-Specific Extranet Investment

Proposition. See the Appendix.)

21

The Supplier-Specific Electronic Market Investment Proposition implies that, in using an

electronic market, the buyer can expect the possible relationship-specific investment from one

supplier to be no more than db + s + 2dc. This is due to the fact that the supplier will not obtain

positive net benefits if it has to make an investment exceeding this amount. Next, let us examine

the case where both suppliers are invited to participate in an electronic market.

Proposition 6 -- Supplier Electronic Market Adoption Proposition: If both suppliers participate in the electronic market, the supplier’s relationship-specific investment will be less than the sum of its gain from its discounted bargaining power, its gain from strategic value under competition and its gain from discounted competitive value. That is, vM < db + qs + dc.

(Our proof is similar to that for the Supplier Extranet Adoption Proposition and also is given in

the Appendix.)

Suppose the buyer is satisfied with the case where only one supplier participates in the

electronic market. It may follow a strategy that is similar to the one it used to induce one supplier

to join the extranet. That is, the buyer may first try to set up a connection with one supplier, say

S1. Since the suppliers are symmetric, this arrangement will not lose generality. This will

require supplier relationship-specific investment vM1. If S1 accepts, then the buyer may stop. If

S1 rejects, then as before, the buyer may invite S2 to join with vM2 relationship-specific

investment. The result is as follows:

Proposition 7 -- The Focal Electronic Market Proposition: If vM1 < db + s + 2dc and db + qs + dc < vM2 < db + s + dc, then there will be two equivalent outcomes:

(7A). The buyer invites S1 to join the electronic market with vM1 and S1 accepts.

(7B). The buyer invites S2 to join the electronic market with vM1 and S2 accepts.

(The proof is similar to that for the Focal Extranet Proposition and also is given in the

Appendix.)

22

The buyer will compare the maximum possible net benefits it can obtain in the two

electronic market cases, and make a choice as suggested by the following proposition.

Proposition 8 -- Buyer’s Supplier-to-Market Connection Preference Proposition: The buyer selected the appropriate supplier-to-market connections based on the following preference conditions:

(8A). When q < ( s – i)/ 2s, the buyer prefers to transact with one supplier in the electronic market, and its net benefit will not exceed i + s + 2dc+ n.

(8B). When q > ( s – i)/ 2s, the buyer prefers to have both suppliers in the electronic market, and its net benefit will not exceed 2i + 2qs +2dc+ n.

(8C). When q = ( s – i)/ 2s, the buyer is indifferent about the number of supplier-to-market connections, and its net benefit will not exceed i + s +2dc + n.

(The proof for this proposition is similar to that for the Buyer’s Supplier-to-Extranet Connection

Preference Proposition and is included in the Appendix.)

Based on our analysis of Cases 4 and 5, we display the buyer firm’s preferences in Figure

2. (See Figure 2.)

Figure 2. The Buyer’s Supplier-to-Market Connection Preference

vM

wM wM 2

wM1

2i - 2db + n

i - db + n

s + db + 2dc qs + db + dc

q > (s-i)/2s

q < (s-i) /2s

A

B

C

-n - (i - db)

Figure 2 depicts how the buyer’s welfare changes with supplier relationship-specific

investment in an electronic market. As in Figure 1, the buyer’s possible welfare, when both

23

suppliers are in the electronic market, is described by line wM2, and the maximum value is

indicated by point A (Va, Wa). When only one supplier is in the electronic market, the buyer’s

possible welfare is represented by line wM1, and point C (Vc, Wc) captures the buyer’s maximum

welfare. Supposing that only q changes, we view point C as fixed and point A as changing.

When q = (s – i)/2s, Wa = Wc, which means that the buyer will obtain the same maximal welfare

no matter how many suppliers join the network. In this case, the buyer is indifferent about the

number of connections with the suppliers via an electronic market mechanism. When

q > (s – i)/2s, wM2 increases and leads A to be higher than C, i.e., Wa > Wc. In this case, the

buyer will prefer a spanning electronic market with more than one supplier participant. On the

other hand, when q < (s – i)/2s, wM2 decreases and thus A moves below C, i.e. Wa < Wc. In this

case, the buyer will prefer a focal electronic market strategy. Figure 2 also shows that when the

supplier relationship-specific investment that the buyer can obtain becomes less than – n – (i –

db), the buyer’s welfare will become negative, and thus it will not initiate the electronic market.

At the other end of this value spectrum, the maximum supplier relationship-specific investment

is s + db +2dc, beyond which neither supplier will agree to join the buyer electronic market.

Choosing Between an Extranet and an Electronic Market

Facing the two alternatives, the buyer will compare the maximum possible net benefits it

can obtain in the two arrangements and make a choice. The results are shown in the following

proposition.

Proposition 9 -- The Buyer’s Choice Proposition: If n > (2 – 2d)c – k, then the buyer will choose the electronic market approach.

Proof. Since k > 0, the following inequality is true: (k + s – i)/2s > (s – i)/2s. Now, let us

examine the following three situations that describe the extent of the suppliers’ competitiveness

q, via constraints on its values:

24

1. q ≥ (k + s –i)/2s. In this case, q also satisfies q > (s – i)/2s. Under this condition, the

maximum net benefits the buyer can obtain from adopting the extranet and the electronic

market approach are wE2 and wM2, respectively. Since n > (2 – 2d)c – k, we have:

wM2 = 2i + 2qs + 2dc + n

> 2i+ 2qs + 2dc + (2 – 2d)c – k = 2i+ 2qs + 2c – k

> 2i + 2qs + 2c – 2k = wE2

Therefore, the buyer will choose an electronic market.

2. q ≤ (s – i)/2s. This case also implies that q < (k + s –i)/2s. Under such condition, the

maximum net benefits the buyer can obtain from adopting the extranet and the electronic

market approach are wE1 and wM1 respectively. Since n > (2 – 2d)c – k, we get

wM1 = i + s + 2dc+ n > i+ s + 2dc + (2 – 2d)c – k = i+ s + 2c – k = wE1

Therefore, the buyer will choose electronic market.

3. (s – i)/2s < q < (k + s –i)/2s. In this case, the maximum net benefits the buyer can obtain

from adopting the extranet and the electronic market approach are wE1 and wM2

respectively. The difference ∆ between these two possible values is given by:

∆ = wM2 – wE1 = 2i + 2qs + 2dc + n – (i + s + 2d – k)

= i + (2q – 1)s + n – (2 – 2d)c + k

Plugging (s – i)/2s < q < (k + s –i)/2s, we get n – (2 – 2dc) + k < ∆ < n – (2 – 2dc) + 2k.

Thus, given that n > (2 – 2d)c – k, ∆ > 0, then wM2 > wE1. Therefore, the buyer will

choose an electronic market. '

So we see that the Buyer’s Choice Proposition suggests that no matter how competitive

the supplier market is, the buyer still will prefer an electronic market. This is predicated upon

whether the buyer’s electronic market gains from lower search costs, the expanded potential

25

supplier base and lower operation costs are greater than the suppliers’ loss in competitive value

less the suppliers’ savings on system implementation. Figure 3 depicts the buyer’s choice. (See

Figure 3.)

Figure 3. Buyer’s Choice Between the Extranet and the Electronic Market Approach

wE2

wM2

(2 - 2d)c - k

3(a). q > (k + s - i)/2s

n

w

2i + 2qs + 2c - 2k

2i + 2qs + 2dc

3(b). q < (s - i)/2s

n

i + s + 2dc

wE 1

wM1

(2 - 2d)c - k

w

i + s + 2c - k

i + s + 2dc

3(c). (s - i)/2s< q < (k+s-i)/2s

n (2 - 2d)c - k

i + s + 2c - k

2i + 2qs + 2dc

w wM2

wE 1

Figure 3 shows the buyer’s welfare w as a function of its gains derived from adopting an

electronic market approach in the three different supplier market competition situations 3(a), 3(b)

and 3(c). When its gains from the market exceed the adoption threshold, (2 – 2d)c – k, the buyer

will obtain higher net benefits by adopting an electronic market mechanism. Figure 3 also shows

that when d decreases, the adoption threshold moves to the right. This implies that the buyer

needs higher gains from adopting the electronic market mechanism when the reduction in

information sharing has a larger impact.

DISCUSSION

Prior research in IS and economics has studied the effects of ITs on buyer-supplier

relationships. This article applies these theories to examine the impact of emerging e-

procurement electronic markets and finds some interesting results. Our work also offers a

number of practical implications for buyers and suppliers who wish to make choices between

adopting extranets and electronic markets for their purchasing activities.

26

Interpreting E-Procurement Channel Adoption Decision Making

Buyer Decision Making. What do we learn about the buyer’s decision making process?

From the buyer’s point of view, the decision will involve two levels. At the first level, the buyer

makes a choice between an extranet and an electronic market approach. Our results indicate that

when the buyer moves from an extranet to an electronic market, it gains the value conferred by

lower search costs and lower operation costs. Such benefits vary across products and transaction

types, however. Specialized products and spot purchasing will benefit more from lower search

costs, while systematic purchasing benefits from lower operation costs. Therefore, if a

procurement system is mainly used for systematic purchasing, then the buyer will examine the

reduction in operation costs by adopting an electronic market. If a buyer expects to purchase

specialized products in small quantities, it will check the various search mechanisms supported

by an electronic market and the number of suppliers participating in the electronic market.

Information Sharing Effects. What are the impacts of reduced information sharing?

Besides the possible gains from adopting an electronic market, the effects of reduced information

sharing will also influence the buyer’s preference. When the information sharing discount rate d

decreases, more information sharing is lost by moving from extranets to electronic markets. This

means that the supplier will lose bargaining power relative to the buyer and thus will make fewer

non-contractible investments. Therefore, the buyer will desire a higher return n from using an

electronic market to offset benefits lost due to the reduced non-contractible gain.

Why Product Characteristics Matter. And are there implications that can be drawn

from the kinds of products that are exchanged? We believe that the characteristics of the product

purchased will affect d. If the product is of low strategic value to the buyer, then the access to

the information will not increase the supplier’s bargaining power in an extranet channel. And so,

27

when the supplier moves to an electronic market, reduced access to buyer information does not

reduce its benefits (in other words, d will be large). As a result, the supplier will not

significantly reduce its investments that complement those of the buyer. Therefore, the buyer

does not need a large gain from adopting an electronic market to compensate for its loss from

supplier investments. This helps explain the large amount of online purchasing for office and

operation supplies. These products are of low strategic significance to the buyer, and will cause

little impact on reduced information sharing between the buyer and its suppliers.

Implementation-Related Cost Savings. How do the supplier’s cost savings affect the

adoption decision? Our results suggest that the higher the supplier’s cost savings due to a shift

from an extranet to an electronic market, the more likely the electronic market will be adopted.

The increase in the implementation costs will lead the supplier to make fewer relationship-

specific investments for the buyer in order to obtain positive net benefits. As a result, the

buyer’s welfare will be reduced. Therefore, the buyer will be more likely to use an electronic

market when the suppliers’ cost savings are higher. This is particularly true for small suppliers.

The implementation cost savings for small suppliers will greatly affect their decisions about

participating in an electronic channel with the buyer.

Number of Suppliers. After deciding whether to adopt an electronic market or an

extranet approach, how will the buyer decide upon the desired number of suppliers? Our model

shows that, with either an extranet or an electronic market, individual supplier’s relationship-

specific investment for the buyer is higher when only one supplier joins in the electronic

connection. Therefore, a buyer who wants more non-contractible investments from a particular

supplier will choose to have a dedicated link with that supplier. However, the buyer’s net

benefits depend on the competitiveness of the supplier market. When the competition is intense

28

among suppliers, individual supplier’s benefits are reduced and thus it will make fewer

investments for the buyer. As a result, the buyer should take into account the supplier market

situations when it decides the number of suppliers.

Limitations and Future Research

Our investigation of e-procurement channel adoption in this article has several

limitations. First, our model is based on a market structure of one buyer and two suppliers. This

helps us to focus on the buyer’s adoption decision. But in reality, one buyer can set up EDI links

with several suppliers, and electronic markets can support many-to-many transactions among

buyers and suppliers. Second, the assumption of supplier homogeneity in their adoption costs

and benefits requires closer scrutiny. Suppliers differ in their capabilities to build electronic

connections with the buyer and to integrate the external link with internal business processes.

Those suppliers who have more resources and can reengineer their internal and

interorganizational business processes to take advantage of the electronic connections with

buyers, can obtain higher benefits (Lee, Clark and Tam, 1999). Third, our model only compares

two forms of electronic channels for corporate supply procurement, and does not account for

other competing channels. In addition, our model does not cover buyers’ choices among similar

e-procurement electronic markets, as is increasingly the case on the Internet (MPR, 2000). As

there often are several online B2B markets operating in one particular industry, buyers who

consider adopting e-procurement systems need to evaluate the various electronic markets.

Finally, our model is static: that is, the adoption costs and benefits are assumed to be time

invariant in our model. This implies that the decision about e-procurement channel adoption is

the same at various points of time. Thus our model cannot be used to identify the optimal time to

adopt a given e-procurement channel.

29

In future research, we will extend our model to represent more realistic market structures.

In a business network with multiple buyers and multiple suppliers, network size is an important

factor in determining the benefits for individual participants. On the one hand, the number of

suppliers in a buyer extranet or an electronic market will affect each supplier’s expected value

for participating in the network. Accordingly, an individual supplier’s incentive to make

relationship-specific investments with the buyer will be reduced when the number of suppliers in

the transacting network increases. On the other hand, an individual buyer’s expected value from

a supplier network will depend on the number of participating buyers (Clemons and Kleindorfer,

1992). But in an electronic market, the more buyers there are in the market, the more attractive

the online market will be to suppliers -- and vice versa. Therefore, a more sophisticated model

with multiple players will enable us to examine the electronic market dynamics and the

equilibrium outcomes.

As an exploratory analysis, our model implicitly assumes that the benefits and losses are

additive, leaving out the possible interactions between these individual elements. But in reality,

these factors may influence each other. For instance, the supplier strategic benefits may

positively influence the competitive advantage, and such an effect needs to be accounted for in

the future. Moreover, our analysis focuses on the buyer’s decision to adopt e-procurement

channels, and does not address the case where a supplier initiates an e-procurement program. To

study how suppliers choose e-procurement mechanisms, we need to adjust the model and the

analysis to reflect the standpoint of a supplier.

Finally, firms have different absorptive capacities (Cohen and Levinthal, 1989) which

determine their benefits in adopting innovations (GÖtz, 2000). To account for these differences

among firms, we should relax one assumption in our model to permit the analysis of

30

heterogeneous suppliers. In addition, our model can be extended to examine the buyer’s choice

among competing e-procurement channels, or electronic markets. In our model, electronic

markets differ in their support for information sharing between buyers and suppliers (i.e., d) and

the benefits derived from lower search costs among a larger potential supplier base (i.e., n). To

extend our model to evaluate various electronic channels, we will elaborate on these terms and

provide guidelines for selecting among electronic market choices. Finally, as system

development costs can be expected to change as technologies advance over time, and the benefits

derived from adopting e-procurement channels will also change over time when more firms

participate in electronic markets, a dynamic model incorporating time-variant variables needs to

be developed in order to shed light on the question of optimal adoption timing.

Through the decades, ITs have greatly increased the options that firms have to do

business with their business partners. Traditional IOISs facilitate information sharing among

firms, and foster close partnership between buyers and suppliers. The emerging e-procurement

applications and electronic markets of the Internet provide another alternative channel for buyers

to purchase goods and services from their suppliers. Faced with these additional options, senior

managers at buyer firms need guidelines to make the appropriate choices. This article is an

initial step in that direction. Although the richness of our exploratory analytical model is limited,

it nonetheless provides some necessary understanding for comparing different adoption

strategies, and can serve as a starting point for future research. We expect that e-procurement

systems and electronic markets will become an important corporate purchasing channel. So it is

important for researchers and practitioners to develop a deeper understanding and a prescriptive

framework for evaluating the various e-procurement channel choices.

31

APPENDIX OF PROOFS

Proof for Proposition 5 – The Supplier-Specific Electronic Market Investment Proposition.

The highest possible gain for a supplier to accept electronic market is db + s + dc – vM, while the

lowest possible gain is – dc when it does not accept. Once again, suppliers will maximize value,

and thus will only join the electronic market if db + s + dc – vM > - dc. Rearranging terms, we

get vM < db + s + 2dc. This completes the proof. '

Proof for Proposition 6 – The Supplier Electronic Market Adoption Proposition. Assume

that S1 first agrees to participate in the electronic market. In this case, if S2 also decides to join,

it will get db + qs – vM, which is greater than – dc that it would get if it did not join under the

condition of vM < db + qs + dc. Therefore, S2 will join the electronic market. If S1 does not

join, then the net benefits S2 will obtain from participating will be db + s + dc – vM, which is

greater than zero with vM < db + qs +dc. As a result, S2 will join the electronic market, and thus

S1 will end up with – dc. But, with vM < db + qs + dc, S1 would rather join the electronic market

to get a net benefit of db + qs – vM, which is greater than –dc. So, both suppliers will join the

electronic market. '

Proof for Proposition 7 – The Focal Electronic Market Proposition. If S1 rejects when the

buyer first invites it to join, then the buyer will invite S2 at vM2 < db + s + dc. That gives S2 a

net benefit greater than zero, and as a rational firm, S2 will accept. As a result, S1 will lose dc.

Considering this, S1 will join the electronic market when the buyer initially invites it at vM1.

32

Since vM2 is greater than db + qs + dc, S2 would not join after S1 has already joined. That is, S1

and S2 will not join together. Therefore, the outcome will be that the buyer will do business via

electronic market mechanism with either S1 or S2, and obtain relationship-specific investment

vM1 < db + s +2dc. '

Proof for Proposition 8 -- The Buyer’s Supplier-to-Market Connection Preference

Proposition. The Supplier Electronic Market Adoption Proposition shows that the highest

possible relationship-specific investment the buyer can obtain is db + qs + dc. In this case, the

buyer’s maximum net benefit or welfare in adopting the electronic market approach will be:

wM2 = 2i – 2db + 2(db + qs + dc)+ n = 2i + 2qs + 2dc + n

The Focal Market Proposition indicates that the buyer will obtain relationship-specific

investment no more than db + s +2dc. This will give the buyer maximal net benefit:

wM1 = i – b + (db + s +2dc)+ n = i + s + 2dc + n

If q < (s – i)/2s, then

wM2 = 2i + 2qs + 2dc + n < 2i + 2s*(s – i)/2s + 2dc + n = i + s + 2dc + n = wM1

As a value-maximizing firm, the buyer will choose Max (wM1 , wM2). In this case, the buyer will

invite only one supplier to join the electronic market. Similarly, if q > (s – i)/2s, then wM2 >

wM1, and as a result, the buyer will prefer two suppliers joining the electronic market. If q = (s –

i)/2s, then wM2 = wM1, and the buyer will be indifferent. '

33

REFERENCES

Bakos, J. Y. "A Strategic Analysis of Electronic Marketplaces," MIS Quarterly (15:3), September 1991, pp. 295-310.

Bakos, J.Y. and Brynjolfsson, E. "From Vendors to Partners: Information Technology and Incomplete Contracts in Buyer-Supplier Relationships," Journal of Organizational Computing (3:3), December 1993, pp. 301-328.

Bakos, J. Y. and Kemerer, C.F. "Recent Applications of Economic Theory in Information Technology," Decision Support Systems (8:5), December 1992, pp. 365-386.

Barua, A., Ravindran, S., and Whinston, A. B. “Efficient Selection of Suppliers Over the Internet,” Journal of Management Information Systems (13:4), Spring 1997, pp. 117-137.

Barua, A., Lee, B. “An Economic Analysis of the Introduction of an Electronic Data Interchange System,” Information System Research (8:4), December 1997, pp. 398-422.

Cash, J. I., Jr., and Konsynski, B. R. "IS Redraws Competitive Boundaries," Harvard Business Review (63:2), March-April 1985, pp. 134-142.

Chatfield, A. K., and Yetton, P. “Strategic Payoff from EDI as a Function of EDI Embeddedness,” Journal of Management Information Systems (16:4), Spring 2000, pp. 195-224.

Clark, T. H., and Lee, H. G. “Performance, Interdependence and Coordination in Business-to-Business Electronic Commerce and Supply Chain Management,” Information Technology and Management (1:1), 2000, pp. 85-105.

Clemons, E. K., and Kleindorfer, P. R. “An Economic Analysis of Interorganizational Information Technology,” Decision Support Systems (8:5), September 1992, pp. 431-446.

Clemons, E. K., Reddi, S. P., and Row, M. C. “The Impact of Information Technology on the Organization of Economic Activity: ‘The Move to The Middle’ Hypothesis,” Journal of Management Information Systems (10:2), Fall 1993, pp. 9–35.

Cohen, W. M., and Levinthal, D. A. “Innovation and Learning: The Two Facets of R&D,” The Economic Journal (99:397), September 1989, pp. 569-596.

Gebauer, J., and Segev, A. “Assessing Internet-Based Procurement to Support the Virtual Enterprise,” Virtual-Organization.net, Newsletter (2:3), September 1998. (available at http://www.haas.berkeley.edu/~citm/procurement/).

GÖtz, G. “Strategic Timing of Adoption of New Technologies under Uncertainty: A Note,” International Journal of Industrial Organization (18:2), February 2000, pp. 369-379.

Johnston, H.R., and Vitale, M. R. "Creating Competitive Advantage With Interorganizational Information Systems," MIS Quarterly (12:2), June 1988, pp. 153-165.

Kaplan, S., and Sawhney, M. “E-Hubs: The New B2B Marketplaces”, Harvard Business Review (78:3), May/June 2000, pp. 97- 104.

34

Kauffman, R. J., McAndrews, J. and Wang, Y.M. "Opening the ‘Black Box’ of Network Externalities in Network Adoption," Information Systems Research (11:1), March 2000, pp. 61-82.

Kauffman, R. J., and Wang, Y. M. “Network Externalities and the Determinants of Network Default,” presented at the 11th Workshop on Information Systems and Economics (WISE-99), Charlotte, NC, December 1999.

Lee, H. G., Clark, T., and Tam, K. Y. “Research Report. Can EDI Benefit Adopters?” Information Systems Research (10:2), June 1999, pp. 186-195.

Malone, T. W., Yates, J., and Benjamin, R. I. “Electronic Markets and Electronic Hierarchies,” Communications of the ACM (30:6), June 1987, pp. 484-497.

Mukhopadhyay, T., Kekre, S., Kalathur, S. “Business Value of Information Technology: A Study of Electronic Data Interchange,” MIS Quarterly (19:2), June 1995, pp. 137-156.

National Public Radio. “Report on B2B Procurement Exchanges,” aired on the All Things Considered program, National Public Radio, Washington, DC, May 23, 2000. (available via the Minnesota Public Radio website at http://news.mpr.org/programs/futuretense/.)

Ovans, A. “E-Procurement at Schlumberger,” Harvard Business Review (78:3), May/June 2000, pp. 21-22.

Raupp, M., and Schober, F., “Why Buyer-Supplier Chains Differ: A Strategic Framework for Electronic Network Organizations,” in R. Sprague (ed.), Proceedings of the 33rd Hawaii International Conference on System Sciences, Maui, HI, IEEE Computer Society Press, Los Alamitos, CA, January 2000.

Riggins, F. J., Kriebel, C.H. and Mukhopadhyay, T. "The Growth of Interorganizational Systems in the Presence of Network Externalities," Management Science (40:8), August 1994, pp. 984-998.

Riggins, F. J., and Mukhopadhyay, T. “ Interdependent Benefits from Interorganizational Systems: Opportunities for Business Partner Reengineering,” Journal of Management Information Systems (11:2), Fall 1994, pp. 37-57.

Riggins, F. J., and Rhee, H-S. “Toward a Unifiied View of Electronic Commerce,” Communications of the ACM (41:10), October 1998, pp. 88-95.

Riggins, F. J., and Rhee, H-S. “Devloping the Learning Netwrok Using Extranets,” International Journal of Electronic Commerce (4:1), Fall 1999, pp.65-84.

Riggins, F. J., and Mukhopadhyay, T. "Overcoming Adoption and Implementation Risks of EDI," International Journal of Electronic Commerce (4:1), Summer 1999, 103-115.

Sapp, G. “Vertical Markets Nearing Hypergrowth,” InfoWorld (22:6), February 7, 2000, p. 6.

Sawhney, M., and Kaplan, S. “Let’s Get Vertical,” Business2.0, September 1999. (available at http://www.business2.com/articles/1999/09/text/models.html)

Seidmann, A. and Sundararajan, A. "Building and Sustaining Interorganizational Information Sharing Relationships: The Competitive Impact of Interfacing Supply Chain Operations with Marketing Strategy," J. DeGross and K. Kumar (editors), Proceedings of the 18th International Conference on Information Systems (ICIS-97), Atlanta, GA, December 1997, pp. 205-222.

35

Subramani, M. R. “Linking IT Use to Benefits in Interorganizational Networks: The Mediating Role of Relationship-Specific Intangible Investments,” P. De and J. DeGross (editors), Proceedings of the 20th International Conference on Information Systems (ICIS-99), Charlotte, NC, December 1999, pp. 358-363.

Waltner, C. “EDI Travels The Web -- EDI Over the Web Offers Companies Cheaper E-Commerce and Smarter Messaging Via Browsers,” CommunicationsWeek (668), Manhasset, NY, June 16, 1997, p. 55.

Waltner, C. “Procurement Pays Off – Companies Are Finding That Procurement Software Comes with a Quick ROI,” InformationWeek (745), July 26, 1999. (available at http://www.informationweek.com/745/procure.htm)