Complete Research Paper from attached feedback

Graver & Ramanlal/Six Myths of Information and Markets

Qarterly SIX MYTHS OF INFORMATION A N D MARKETS:

INFORMATION TECHNOLOGY NETWORKS,

ELECTRONIC COMMERCE, A N D THE BATTLE FOR

CONSUMER SURPLUS^

By: Varun Grover Management Science Department College of Business Administration University of South Carolina Columbia, SC 29208 U.S.A. vgrover@sc.edu

Pradipkumar Ramanlal Department of Finance College of Business Administration University of Central Florida Orlando, fL 32816 U.S.A. pradipkumar.ramanlal@bus.ucf.edu

Abstract The infusion of powerful information networks into business environments is beginning to have a profound impact on the nature of governance between buyers and sellers in the marketplace. Most articles in this area emphasize the benefits to the consumer side of the equation due to reduced coordination, search, and transactional costs. This article presents a broader view of information and markets by elucidating innova- tive ways that sellers can survive in intensely competitive markets. The article is framed in

'Allen Lee was the accepting senior editor for this paper.

terms of six myths and counter-myths of infor- mation technology and effective markets. The myths provide a conventional view of how increased customization and outsourcing, open architectures, a larger customer base, and low price guarantees will benefit the buyer. The counter-myths illustrate that it is altogether fea- sible for IT to enable supplier strategies that extract consumer surplus. For instance, suppliers could use IT to price discriminate by tailoring product offerings and charging buyers as much as they are willing to pay. They could also seg- ment markets making comparative shopping dif- ficult, thus avoiding the competitive equilibri- um. Also, suppliers could focus on the creation of networks that lock in customers or follow aggressive pricing strategies that deter price competition.

Both the myths and counter-myths are presented and examined in a polemical format using sim- ple, fundamental economic arguments. We hope to provide provocative new avenues for dis- course in this area by recognizing the complexi- ty of interactions between buyers and suppliers in a highly networked environment.

Keywords: Electronic markets, networked mar- kets, myths of markets, economic theory, sell- er strategies, buyer strategies, electronic com- merce

ISRL Categories: A M 0 2 , B A 0 1 , CAIO, G A O l , HAO7O2

MIS Quarterly Vol. 23 No. 4. pp. 465-495/December 1999 465

Grover & Ramanlal/Six Myths of Information and Markets

Introduction A classic example of information technology that created strategic advantage is the ASAP system at American Hospital Supply Corporation (AHSC). This interorganizational system allowed hospitals to order their supplies directly through personal computers or mainframes linked directly to AHSC's mainframes. Customized information flow through this proprietary network created switching costs for the buyer hospital, thereby providing AHSC with monopolistic power. In 1985, AHSC was purchased by Baxter Travenol, which partnered with General Electric Information Services to create a hospital supply marketplace. Successive generations of ASAP created a more distributed market-based mode! that today involves many hospital suppliers (competitors of the old AHSC) and buyers involved in an end-to-end electronic commerce infrastructure for the hospital industry. This evo- lution is not unlike changes in other industries. For instance, American Airline's SABRE system, once a source of monopolistic rents for the provider, is now an independent profit center that facilitates an unbiased electronic market (Applegate 1995).

The rationale behind these consistent patterns can be found in increased industry competition, reduced technology costs, and increased inter- connectivity of technology through open archi- tectures. If we complement these trends with the recent explosion of almost ubiquitous infrastruc- tures like the Internet, it is possible to argue for distributed market structures and consequently more effective markets. Indeed, such reasoning can be found through the lens of transaaion cost economics, where information technology decreases coordination costs between buyers and sellers,^ making market structures more viable (Malone et al. 1987). The power of Internet technologies to match buyers and sell- ers, share information, reduce search costs, and compare complex products has been argued to alleviate market imperfections, resulting in more effective markets (Bakos 1991a).

The terms "sellers" and "suppliers" are used inter- changeably in this document.

While hard evidence for such a definitive trend toward markets has been limited at best, some longitudinal observations indicate survival of monopolistic positions, despite significant infu- sion of information technology (Hess and Kemerer 1994; Lee 1998). However, no study to our knowledge has had the luxury of observing patterns of evolution on the Internet, a technolo- gy that involves a quantum leap in its reach, openness, and impact (Quelch and Klein 1996; Widdifield and Grover 1995). Morover, the ges- tation period for evolution of pure market struc- tures is unknown, wbile the arguments for tbese structures remain vociferous and reasonable.

In tbis paper we seek to provide a broader view of information and markets. We do so by fram- ing conventional arguments on how the free flow of information on Internet-like infrastruc- tures should favor buyers, thereby improving market effectiveness. We then present plausible counter-arguments to illustrate possible supplier strategies tbat are now viable. With tbe new economics of information, these strategies may indeed impede market effectiveness. We expand current thinking by providing arguments for and against market effectiveness, presented in the form of "myths" and "counter-myths." They draw extensively from fundamental economic principles. Economic examples are also provid- ed for illustrative purposes and not as a substi- tute for empirical research. It is important to note that these alternative arguments and exam- ples do not follow deterministically as the only possible deductions and conclusions from exist- ing economic theory but represent altogether feasible strategies tbat economic actors can enact. The intent here is simple: we wish to be provocative in impressing upon readers the complexity of interactions in networked con- texts, and the need to broaden the scope of investigation when examining market effective- ness on tbe Internet.

The following sections provide background into fundamental economic theory of effective mar- kets, the role of information technology, and gen- eral assumptions that are made in subsequent discussion. Examples and economic arguments for each myth and counter-myth of information and markets follow this. The final two sections discuss the implications and conclusions.

466 MIS Quarterly Vol. 23 No. 4/DBcember 1999

Grover & Ramanlal/Six Myths of Informatior) and Markets

Microeconomic Theory of Effective Markets ^ i ^ ^ A fundamental, comprehensive representation of a market is the aggregate demand curve (see Figure 1). It specifies the consumers' reservation price, i.e., the maximum amount they are willing to pay for each successive unit of the product. In practice, consumers pay only the market price, which depends on the degree of competition between suppliers and their ability to price dis- criminate between consumers. The difference between the reservation price and the market price (summed across all customers) is the con- sumer surplus. Market effectiveness may be char- acterized in terms of the division of this surplus between consumers and suppliers, where more effective markets are characterized as those that benefit the consumer (see Varian 1984).

In markets v̂ fhere suppliers face perfect competi- tion, the equilibrium price will be the minimum a supplier is willing to accept, equal to the mar- ginal cost of producing the product. Pricing above the marginal cost will drive a consumer from one supplier to another who offers a lower price. Under perfect competition, consumer sur- plus takes on its maximum possible value. The entire surplus accrues to consumers (given by the area under the demand curve but above the equilibrium price PO in Figure 1). Thus, markets are most effective when perfect competition holds.

In contrast, markets are least effective under per- fect price discrimination. If suppliers can assess their customer's reservation price and are able to segment markets, each customer could then face a different market price equal to the reservation price. By charging the maximum amount that each customer is willing to pay along the entire demand curve, suppliers extract total consumer surplus, thereby making markets least effective.

This perspective of markets describes the two extremes of a spectrum: perfect price discrimina- tion, where suppliers extract the entire consumer surplus and markets are least effective; and per- fect competition, where markets are most effec- tive because the surplus in its entirety accrues to consumers.

Given the reality of opposing motives of suppli- ers and consumers, however, most markets fall between these two extremes. The conflicting forces at play will determine the eventual market structure. For example, commoditization enhances product substitutibility and lower search costs permit comparative shopping, which results in more effective markets. Conversely, information asymmetries between suppliers and customers and cost differentials between them for acquiring information about products and prices result in less effective mar- kets. In the latter case, some of the consumer sur- plus will be expropriated from customers as sup- pliers set prices based on decisions made within the corporate hierarchy rather than as a conse-

Value to Consumer {P1)

P r i c e

equilibrium price (PO)

In perfect competition P0 = marginal cost ot product

QO Quantity

Figure 1. Demand Curve and Consumer Surplus

MIS Quarterly Vol. 23 No. 4/December 1999 467

Grover & Ramanlal/Six Myths of Information and Marked

quence of market forces. This situation sustains itself when there are inhibitors to pure competi- tion, which presupposes that buyers can cost- lessly search for alternative suppliers and com- pare product offerings.

Our premise of the market's effectiveness reflects this continuum where effective markets, which work in favor of the consumer, are contrasted with ineffective markets, which work in favor of the supplier. The remainder of this paper will draw upon these concepts, which relate market effectiveness to the distribution of consumer sur- plus in discussing the myths of the impact of information technology on market structures.

IT and Effective Markets The worldwide web of computer networks and electronic information services allow a large, growing number of suppliers to interact with a large, growing number of buyers. Transaction revenue on the Internet is estimated at over $2 billion, and it is just beginning to enter a phase of rapid grov^h. Forrester Research, a Cambridge, Massachusetts, based research organization, forecasts annual online sales in 2000 will approach $7 billion. The U.S. telecommunica- tions deregulation bill, the growth in multimedia services, the availability of intelligent search agents, and the investments in widening band- width are just some of the trends that bode well for electronic commerce.

Conventional thinking would suggest that tbese trends point us in the direction of greater com- petitiveness and more effective markets, where the free flow of information permits buyers to search competing sellers, thereby forcing the competitive equilibrium. Presumably, sellers are less able to sustain monopolistic positions. Four aspects of this argument that suggest movement toward price competition are:

1. IT reduces transaction costs: If IT reduces the buyers' costs associated with conducting mar- ketplace transactions, namely, searching for suppliers, information seeking, and negotiat- ing contracts, then comparative shopping is feasible and the supplier's ability to generate monopolistic rents is reduced. Conversely, if

coordination costs are high, buyers may not incur the search costs to uncover competing suppliers, allowing suppliers to increase prices recognizing that buyers would rather pay the higher price than incur an expensive search. In this case, lowering prices will not increase the supplier's profitability because additional buyers would not be attracted due to restrictive search costs. Therefore, prices charged would be higher than those under pure competition, resulting in less effective markets. As these costs decrease through IT, these excess profits are eventually competed away (Bakos 1991a, 1991b; Curbaxani and Whang 1991).

2. IT reduces perceived complexity of products: If IT allows buyers to search and compare complex products by providing information in a manner that is easy to interpret, the buyer perceives less complexity and can change suppliers based on this information. This results in more effeaive markets (Malone et al., 1987). In other words, information reduces uncertainty for the buyer regarding the value of complex products, thereby reduc- ing the suppliers' ability to capitalize on this information asymmetry and generate higher rents.

3. IT reduces asset specificity: If IT allows prod- ucts to be recast into alternative uses (i.e., reducing assets specificity) by changing some of their information attributes at minimal cost, then a broader range of consumers will receive more "customized" products {Malone et al. 1987). However, now consumers cannot be charged a premium for customization, because alternative suppliers could gather information and do the same customization at virtually no cost. In other words, customiza- tion becomes a commodity service that is sub- ject to market forces, resulting in better value for the consumer and more effective markets {Bakos 1991b).

4. IT increases free information flow. If IT allows buyers to become more informed about suppliers in the marketplace and their product offerings, and suppliers can access information on the needs of buyers, the basis for generation of excess profits is reduced. It is

468 MIS Quarterly Vol. 23 No. 4/December 1999

Grover & Ramanlal/Six Myths of Information and Markets

a well-known principle of economic theory that in competitive markets with efficient, free, and complete flow of information, sellers receive no excess return beyond the cost of capital employed (Varian 1984). The buyer and seller both understand the value of the product as perceived by the other party, there- by reducing opportunistic behavior. Simple microeconomic models of perfect competi- tion assume that buyers can costlessly acquire information about prices resulting in a single market price. While unrealistic at one time, today's electronic markets are making this assumption increasingly realistic.

In summary, conventional wisdom and econom- ic theory would suggest that IT reduces market imperfections and allows more players to com- pete in cyberspace, thereby resulting in more effective markets. While it may be premature to observe these effects in every industry, given that the gestation period might not yet be completed, anecdotal evidence exists of a growing number of Internet-based market mechanisms that have resulted in lower prices. Based on the arguments presented above, and support from a number of academic and industry observers, we can expect this trend to continue. Therefore, the core propo- sition that follows from these arguments can be stated as follows:

Increasing pervasiveness of information technol- ogy and information on ubiquitous global net- works (the Internet) will continue to increase the effectiveness of markets.

We now turn to the counter-arguments. Interestingly, the enactment of such counter- arguments falls within the "feasible space" of the same microeconomic theory that allows us to make the case for effective markets. In framing these alternative views as "myths," we hopefully open up a more encompassing view of informa- tion and markets that recognizes the opposing motives of buyers and sellers and the potential impact of these motives on the nature of market structures that result.

The following six general tenets hold for all myths in the subsequent analysis:

1. The unit cost of IT-based processing, storage, and communication is becoming infinitesi-

maliy small, although not necessarily at the same rate for buyers and sellers.

2. Information, selectively applied, can repre- sent two sides of the same coin, i.e., it can either clarify or distort perceptions. The latter is often not considered.

3. IT capitalization is asymmetrical among sup- pliers and individual consumers, thereby resulting in information asymmetry (which exists despite declines in IT costs). In general, supplier firms are still better capitalized than individual consumers, which means that we might be able to distinguish between the qual- ity of information obtained by each party.

4. All parties work in their own respective self- interests, which may not be consistent with creating an effective market. In particular, suppliers might be more interested in using IT to sustain monopolistic positions.

5. Different consumers could value the same information differently. This creates a problem in allowing market forces to assess the "true" value of information.

6. The market structure evolves toward an equi- librium state rather than being one that is specified a priori. However, the market struc- ture is a variable influenced by IT. These two parts of the tenet are intricately related. Specifically, the state of equilibrium is dynam- ic; and this dynamic is influenced by IT.

The key question is: How do sellers survive in intensely competitive markets on IT networks with low coordination costs that seem to favor buyers? Low costs make comparative shopping simple and force sellers to compete based on price. Sellers dread price competition, however, because the resulting equilibrium leads to their making small or zero profits. The trick for sellers to survive in such a "hostile" environment is to use the very resource that gave rise to their prob- lems, namely, contemporary and powerful low- cost IT.

The myths below illustrate that it is altogether feasible for IT to enable supplier strategies that extract consumer surplus. For instance, suppliers could use IT to price discriminate by tailoring produa offerings and charging buyers as nnuch

MIS Quarterly Vol. 23 No. 4/December 1999 469

Grover & Ramanlal/Six Myths of Information and Markets

as they are willing to pay. They could also seg- ment markets making comparative shopping dif- ficult, thus avoiding the competitive equilibri- um. Also, suppliers could focus on the creation of networks that lock in customers or follow aggressive pricing strategies that deter price competition.

These supplier strategies are presented below in the form of myths and counter-myths. Note that: (1) while we devote more space to discussion of the non-conventional counter-myths, it reflects our desire to address their limited exposure in the literature rather than their imminence, and (2) the economic examples used are for illustra- tive purposes only, to express potential devia- tions from effective markets in tangible mone- tary terms.

Six Myths of Information and Markets ^^^^^m

Myth # 1: Product customization, enabled by IT networks, would benefit buyers

Counter-myth #1: Product customization, enabled by IT networks, could allow sellers to exploit buyers

A General Perspective on Myth and Counter-myth #1 Arguments for the myth follow fairly conven- tional thinking. Many suppliers operating in net- worked environments attempt to capture infor- mation on buyers. Buyers often provide their information in the hope that they will benefit by getting a product tailored to their needs. ShopKo maintains such information in a 400 gigabyte data warehouse and makes this information available for potential target marketing via browsers on the Internet. Similarly, House of Seagram uses a database on 10 million adults who consume spirits to build brand loyalty as well as target consumers of rival products, It can be reasonably expected that in a highly compet- itive environment, this customization, if valu- able to the customer and if based on informa- tion technology, can and will be replicated by a

number of suppliers. This would keep the price of such a service down. We can observe this outcome on the Web where hundreds of finan- cial service sites exist (e.g., Stockpoint) that allow any client to enter their personal stock portfolio, track it in real time, and use various options for organizing the data. Today, this ser- vice is often given away free. Buyers seem to be getting customized service for an extremely low price.

However, we propose there is an alternative per- spective (counter-myth). Suppliers using power- ful computers and comprehensive databases could exploit smaller and smaller market niches, by making inferences about an individual buyer's flexibility on product and price, propensity to shop around, etc. In doing so, they could inhibit competition and charge higher prices in these market niches.

An emerging tool to facilitate this exploitation is IBM's net. Mining solution, which uses sophisti- cated computer algorithms to transform site-visil information into detailed customer reports (McCune 1997}. Netscape's cookie technology is remarkable in that it stores information on transaaions and web pages visited by users on their own machines, providing an integrated profile of the customer for suppliers (Hagel and Rayport 1997). Companies like the Gartner Group and Lexis-Nexis can sell customized reports or information to different customers, even charging different prices to different cus- tomers if they can infer the customer's willing- ness to pay. Similarly, Kurzweil Applied Intelligence uses a versioning strategy by selling seven different versions of its voice recognition software, each with different features turned off and on and priced based on how different cus- tomers value the products (Shapiro and Varian 1998). With such superior information obtain- able by suppliers, buyers may end up paying more for customized products that may cost vir- tually nothing to create. Buyers may be willing to pay more to forego the difficulty of shopping around for highly customized products. This is particularly true for individual consumers that rely on generic search engines and relatively low IT sophistication.

Growth in database marketing, relationship mar- keting, data warehousing, and data mining epit-

470 MiS Quarterly Vol. 23 No. 4/December 1999

Grover & Ramanlal/Six Myths of Information and Markets

omize this trend. In fact, the Meta Group esti- mates that the data warehousing market grew as much as 40% in 1998 with 85% of manufactur- ers and retailers believing that they will have large integrated data analysis capabilities in place by tbe year 2000 {Business Weefe 1997).

An Economic Perspective on Myth and Counter-myth #7 Myth: The myth presumes that if sellers use IT to gather information on buyers, they can "narrow- cast" or tailor offerings to different buyers or buyer segments. Tbe customized product increases the buyer's reservation price. The low cost of entry into cyberspace, however, permits multiple suppliers to collect and process buyer information, to customize tbeir offerings. Buyers can search out these alternative suppliers and choose tbe supplier that provides the best value/price tradeoff. Thus, while the reservation price increases, tbe competitive equilibrium holds, ensuring the increased consumer surplus accrues to the buyer. This market works in favor of the buyer and is therefore more effective.

Counter-myth: Interestingly, the enactment of an alternative scenario can also be examined through tbe economic lens. Tbe above argument is sound, provided that tbe cost of information access for both the buyer and seller decreases proportionately. If so, the competitive equilibri- um may be sustainable. The equilibrium may be unstable, however, because the slightest cost dif- ferential between buyers and sellers may be leveraged to a big advantage given that the costs of information gathering are minimal. Suppliers (i.e., firms) in general are in a better position to leverage information asymmetry, as they are larg- er and have economies of scale in gathering and analyzing buyer information. Buyers (i.e., indi- viduals), on the other hand, must resort to public search engines and tbeir own processing infra- structure to locate and compare suppliers.

Consider a supplier who collects information about customers over the Internet. The supplier would like to use this information to extract any consumer surplus. A self-interested customer is aware of this. The customer is nevertheless willing todivulgeinformation with the promise of obtain- ing a customized product. The supplier in turn uses this Information to make inferences about the cus-

tomer's reservation price. By customizing prod- ucts for customers with different reservation prices, the supplier can effectively extract surplus- es along the entire demand curve, as long as the priceexceedsthe supplier's marginal cost. This tar- get marketing reverses the traditional approach where suppliers divulge information about them- selves (specifically, price and product information) and customers comparative shop to obtain the sur- plus. Competition in the market for customized products that presumably could shift the surplus from sellers to buyers is inhibited for two reasons: (1) sellers segment markets through product cus- tomization and (2) buyers are limited in tbeir abil- ity to assess price/value tradeoffs across market segments due to differential IT capitalization between buyers and sellers. In other words, while sellers with very low costs of customization could presumably compete, individual buyers face pro- portionately higher search costs, thereby reducing the incentive to search out alternative sel lers. Thus, lower IT costs for sellers relative to buyers, even as botb sets of costs approach zero, provide an effec- tive means by which suppliers can expropriate sur- plus from customers.

A positive externality of the supplier's profit max- imization objective is tbe increased market size. With information about customers, suppliers can broaden the market by providing a customized product to consumers, albeit at prices close to the maximum they are willing to pay.

Why is this market not to tbe benefit of the con- sumer? The customer gives up information expecting a return from the supplier in the form of better service or customization. While this out- come may indeed occur, the supplier is now in a better position to use the information to extract surplus from to charge every customer in every segment beyond that which was possible under perfect competition. This follows because of the seller's ability to segment markets through cus- tomization, thereby impeding competitive forces. Tbe maximum amount that consumers will pay for customization corresponds to tbe limiting case of perfect price discrimination, where consumers are charged the maximum price they are willing to pay for a product, in which case all surplus accrues to the supplier. Tbis is not an effective market because value derived from the efficient use of IT accrues to suppliers (see Figure 2).

MIS Quarterly Vol. 23 No. 4/December 1999 471

Grover & Ramanlal/Six Myths of Information and Markets

Suppliers

Suppliers ase IT to cotlect and process inrtrmialion on buyers and inrer reservation price.

Suppliers customize informabon intensive products and scgmmt the market inhibiting product comparison

and extracting surplus.

Buyers

Figure 2. The Myth of Product Customization (Myth 1)

Economic Example

Consider the way Intuit versions its popular financial software product Quicken. If Intuit chose to market a single version of its product to the largest number of consumers, it may consid- er one with only the most popular features like tracking persona! expenses, online bill payments, and links to financial websites. Less sophisticated consumers who wish to pay bills the traditional way, via mail, will be unwilling to pay for the online billing feature and therefore may forego purchasing the product. More sophisticated con- sumers who want complex and detailed invest- ment advice, which the product does not offer, may also forego purchasing it. The problem with offering a single version of the product is that it fails to discriminate between customers. Offering multiple versions overcomes this problem and permits Intuit to price discriminate between con- sumers and extract surplus in the process as the following numerical example illustrates.

Suppose the optimal single version of the finan- cial software product has five features and is priced at $10 per feature for a total of $50. And suppose it costs $8 per feature on average to pro- duce the software for a total of $40. Thus a sur- plus of $2 per feature accrues to the supplier.

Potential customers are of two types: those who prefer more features whom we refer to as the

"high" type, and those who prefer fewer features whom we refer to as the "low" type. We assume the high type is not only more affluent (i.e., able to pay more) but also has a higher reservation price (i.e., willing to pay more) and desire a more sophisticated product (i.e., with a larger number of features). Conversely, the low type is less afflu- ent, has a lower reservation price, and prefers fewer features.

Suppose the high type's reservation price is $11 per feature with a desire for seven features. And suppose the low type's reservation price is $9 per feature with a preference for three features. The supplier has no knowledge of these reservation prices and customization preferences. Thus, the original product with five features was intended to strike a balance. The high type likes the price but not the customization; the low type likes nei- ther. Thus, only the high type will potentially be attracted to the offer with probability less than 1.

Now suppose the supplier uses data mining soft- ware on the Internet to obtain information on consumers' income levels and personal prefer- ences and infers the reservation prices and cus- tomization preferences of the two types. As a result, two versions are marketed: a base version with three features priced at $27 ($9 per feature) and an advanced version with seven features priced at %77 ($11 per feature). Indeed these prices are comparable to Intuit's base and top-

472 MIS Quarterly Vol. 23 No. 4/December 1999

Grover & Ramanlal/Six Myths of lnfom)ation and Markets

line versions of Quicken. Now both types will likely be drawn to Intuit's offerings.

Customization serves not only to identify those consumers with higher reservation prices from whom surpluses can be expropriated, but also facilitates market segmentation necessary to exploit the surpluses along the entire demand curve. This follows because the high and low types reveal their identities by the choices they make. Knowing that only the high type will choose the advanced version permits the suppli- er to extract the consumer surplus of $3 per fea- ture from that customer. Correspondingly, the surplus of $1 per feature is obtained from the customer that chooses the base version. This per- sists because the two markets for the high and low types are segmented. Lack of features keeps the high type from choosing the base version, and the smaller reservation price keeps the low type from choosing the advanced version.

By offeri ng two versions, not only does the market size increase, but the supplier is able to draw more of the surplus on a per-feature basis. The market size increases because both the high and low types are satisfied when previously neither was. The sur- plus on a per-feature basis that the supplier extracts is $2.40 {$3 per feature for the advanced version with seven features and $1 per feature for the base version with three features for a weighted average of $2.40 per feature, up from $2.00 per feature for the single-version case). Using buyer information, the supplier has successfully price-discriminated along the demand curve. While both customer types are satisfied and the market is expanded, sur- plus is moved away from consumers, rendering markets less effective.

Myth #2: Increased outsourcing, enabled by IT networks, would lower prices and benefit buyers

Counter-myth #2: Increased outsourcing, enabled by IT networks, could reinforce the seller's monopoly by sustaining higher prices

A General Perspective on Myth and Counter-myth #2

The myth is based on the argument that we would expect outsourcing, coordinated by IT

networks, to reduce the supplier's production cost. These lower costs for suppliers should trans- late into lower prices for consumers if competi- tion prevails. Thus, outsourcing product compo- nents to specialized manufacturers could poten- tially result in more effective markets. Conversely, failure to outsource may result in less effective markets. For example, if CM manufac- tured its own tires, we would likely have more expensive cars.

In discussing the counter-myth, however, we should recognize that complex products or spe- cialized components that are outsourced are unlikely to be produced by a large number of suppliers. Thus, the market for outsourced prod- ucts may indeed be less competitive as firms use a common supplier or set of suppliers. One example is Massachusetts General Hospital's use of "teleradiology" as a means to remotely access the diagnostic skills of medical experts, tn essence, traditional radiology is reengineered into two parts: an in-house task of a radiology technician coupled with the remote service of a radiology expert. As hospitals increasingly com- pete {e.g., with HMOs), the remote service, which requires complex and specialized skills, are shared by multiple hospitals, reducing com- petitiveness along this dimension (Apte and Mason 1995). Similarly, UPS integrates its deliv- ery service as part of product sales by its corpo- rate clients, providing a single sale-and-delivery package. As competing corporate clients increasingly adopt similar arrangements with UPS, competition among these corporations with respect to the product component of the package persists, but competition with respect to shipping services is reduced giving UPS market power along this dimension {Computerworld 1997).

The noncompetitive impact of outsourced prod- ucts is apparent in the computer industry as well. Computer assemblers like Dell compete aggres- sively on all components and services except Intel's chips, which hold a monopolistic power in the industry. While Intel decreases prices regu- larly for chips with aging technology, its cutting- edge chips command a premium suggesting mar- ket power. In fact, Intel's key strategy has been to commoditize complementary products like chipsets and motherboards to stimulate demand

MIS Quarterly Vol. 23 No. 4/December 1999 473

Grover & Ramar^lal/Six Myths of Information and Markets

for its core produa, the microprocessor (Shapiro and Varian 1999). Yet another interesting exam- ple is the airline industry. What appears to be fierce competition among large numbers of online travel agents essentially boils down to competition among a few large airlines. Most of these airlines control the value chain and are making money at the cost of on-line agencies (Davis 1999).

We emphasize that increased outsourcing, enabled by information technology, could rein- force the seller's monopoly. The ability of monopolies to sustain higher prices is well known. The focus here is to illustrate the ability of IT to accentuate this effect. Monopolies with strong pricing power but low market penetration can be less profitable than monopolies with less pricing power but greater market penetration facilitated by IT. And while vertical integration can increase monopolistic power, the outcome could be less desirable if it comes as the expense of market penetration.

Summarizing, even though IT reduces coordina- tion costs between firms giving rise to the poten- tial benefits of outsourcing, the reduced costs could also make monopolistic supply chains and higher prices easier to sustain. This is particular- ly true in the case of complex and specialized components that are common among the prod- ucts of competing firms.

An Economic Perspective on Myth and Counter-myth #2 Myth: Transaaion Cost Economics is an impor- tant theoretical framework that has been used to study the decision to outsource. The firm's deci- sion to outsource is usually thought to depend on the total cost of producing as opposed to procur- ing a given component. It is often true that the given component can be obtained at a lower cost from independent outside suppliers who enjoy economies of scale and scope in specialized pro- duction. However, the resulting market gover- nance calls for locating and monitoring a reliable supplier, which could involve significant transac- tion costs. Alternatively, the firm could produce the component in-house through an organiza- tional hierarchy, thereby saving on transaction costs but at the expense of higher production costs. IT can alter the balance of these tradeoffs

by lowering transaction costs, thereby favoring markets (outsourcing) to hierarchies (Williamson 1975, 1979). As large global networks make out- sourcing increasingly viable, market governance in lieu of hierarchical organizations will result in lower overall costs (Evans and Wurster 1997). With increased competition among suppliers, the benefit of lower costs will be passed on to the consumer.

Counter-myth: However, economic arguments offer another possibility. While it may be cost effective for firms to rely on forms of market gov- ernance for product components, a limit to the number of suppliers that the industry can sustain may exist. Each marginal supplier who enters the industry incurs an expense in real assets as well as in transactional relationships. Moreover, the increase in suppliers reduces the volume of trans- actions per relationship, thereby increasing the cost per transaction. As a result, successive sup- pliers contemplating entry Into the industry face increasingly larger relationship hurdles. These hurdles are reduced to the extent that the indus- try expands and can therefore support additional suppliers.

In contrast, fewer relationships permit the firm and the supplier to invest in asset-specific resources, i.e., resources allocated to the rela- tionship. Suppliers benefit in terms of being assured of a market for the components they provide. Firms benefit in terms of reduced mon- itoring costs. But fewer suppliers of a compo- nent reduce competition, which increases their ability to extract consumer surplus via the firm from the end customer. The surplus is then divided between the supplier and the firm according to their relative market powers, there- by making markets less effective. Suppliers will enter the industry only if the derived consumer surplus exceeds the hurdle to overcome rela- tionship costs. This has been referred to as the "move-to-the-middle" hypotheses (demons 1993).

As IT costs approach zero, and outsourcing becomes prevalent, it is reasonable to expect that specialization would result in a core set of sup- pliers, particularly for complex idiosyncratic products. Within this core set, a single supplier could well exist for each component that the firm outsources. The firm would package the various

474 MIS Quarterly Vol. 23 No. 4/December 1999

Grover & Ramanlal/Six Myths of Information and Markets

components into a product for sale to the cus- tomer. While several firms could be competing amongst themselves for the customer, this would exert no competitive force on the few individual suppliers to yield any consumer surplus that they could derive from the individual components they provide.

In summary, the core suppliers could derive supranormal rents from their monopolistic posi- tion on each component provided. Thus, as IT costs approach zero, the market governance (outsourcing) structure that results for complex idiosyncratic components permits suppliers to retain excessive consumer surplus, thereby mak- ing markets less effective.

As can be seen in Figure 3, the market structure is determined by the complexity of product com- ponents. This contrasts to the market structure in Figure 2, which is driven by customization.

Economic Example

Here it is interesting to consider the role of Intel in the computer hardware industry. If computer manufacturers such as Dell and Compaq, who compete intensely in the market for personal computers, were to develop their own micro- processor units, it would probably result in more expensive computers. Thus, the microprocessor, which is a complex and idiosyncratic compo- nent, is outsourced to Intel. While competition between Dell and Compaq has resulted in lower

prices and lower margins for computers, Intel has sustained higher prices and margins for its micro- processors. And while consumers are pleased with the lower prices for computers, it is often overlooked that these prices could be lower still if not for Intel's market power as the following example illustrates.

The two computer manufacturers, labeled A and B, sell computers composed of two types of com- ponents: generic disk drives, memory, modems, and multimedia peripherals (labeled x), and the unique microprocessor (labeled y). Multiple manufacturers enjoy economies of scale for x because the components are simple. In contrast, y is complex and idiosyncratic and relationships cost are high, and therefore in equilibrium it is cost effective for only one supplier, labeled C, to produce this component.

Suppose the marginal costs of the components x and y are $900 and $100, respectively. Then the marginal cost of the computer, that firms A and B sell, is $1,000. At that price, surpluses accrue to consumers. Now suppose firms A and B as well as supplier C attempt to obtain some of the sur- plus by raising prices. Supplier C raises the price of component y to $600, to obtain a surplus of $500 (given its marginal cost of $100). Firms A and B price the computer at $2,000, also for a surplus of $500 (given their marginal costs are $1,500: $900 for x and $600 for y). The question arises: are these prices sustainable in equilibri-

S u p p l i e r s

y c o m p e l i l l v e m a r k v l p l i i c c . C a m p e l i l i u n i s a D d i m I D S ions o f p ro d uc I e i r e p( com p un en t s p ro in I h i m o n o p o l i s t i c s u p p l y c h a l n .

C ustom e rs

Figure 3. The Myth of Outsourcing (Myth 2)

MIS Quarterly Vol. 23 No. 4/December 1999 475

Grover & Ramanlal/Six Myths of Information and Markets

um, and what economic forces are at play that will influence them?

Firms A and B will be unable to maintain the computer's price at $2,000 because of competi- tive forces. Price competition will force each firm to lower price successively until price equals marginal cost at $1,500, reducing their surpluses to zero. On tbe other hand, there are no compet- itive forces that will cause supplier C to reduce the price it charges firms A and B for component y, thereby keeping its surplus of $500. Neither A nor B can credibly pressure C to lower its price (say, by threatening to not produce tbe product composed of x and y) because if B withdrew from the market, its market share would be cap- tured by A.

In this example, IT might keep transaction costs related to outsourcing down. However, the ben- efits that accrue to consumers are reduced owing to tbe surplus captured by the monopolist in tbe supply chain. This capture follows because component y is not competitive in tbe final product. By raising price, supplier C affects the cost structure of both firms A and B equally. Unless tbey bave tbe latitude to absorb it into tbe cost structure of other components, both firms have to pass the cost down to tbe con- sumer. In other words, supplier C is absorbing consumer surplus from the end customers. Firms A and B, however, are competing on all otber dimensions of tbe product except component y (where they have no differentiation). Ironically, casual observers of the end produa see tbe mar- ket as highly competitive.

Myth #3: Open IT network architectures lower prices and benefit buyers as dependence on supplier hierarchies is reduced

Counter-myth #3: Open IT network architectures could be exploited by suppliers to create captive buyer networks that can sustain higher prices

A General Perspective on Myfh and Counter-myth #3

The myth implies that as IT architectures become more open ala Internet and TCP/IP

standards, companies have greater difficulty sustaining hierarchical positions with respect to their customers based on technology protocols and interfaces. This is unlike many widely reported strategic systems of the 1980s, where proprietary links (e.g., American Hospital Supply's ASAP System) were established (Johnston and Vitale 1988; Porter and Millar 1985) to increase switching costs for buyers and force dependence relationships. Open architectures make such dependence relation- ships difficult to sustain because a buyer can easily search and switch to an alternative sup- plier. We can therefore expect more compari- son shopping by buyers ultimately forcing sup- pliers to lower prices.

However, the counter-myth is based on argu- ments contrary to conventional wisdom. As networks become IT intensive and both buyers and sellers gain better access to information, we can expect differential IT capitalization between buyers and sellers to tilt tbe balance of power. Larger suppliers are likely to be in a better position to package information products as well as information about products and ser- vices in a manner that makes it difficult for consumers to assess their value for compara- tive purposes. Difficulty rn assessing tradeoffs between the product's price and the value it affords would inevitably impede competition. Tbis is abundantly apparent on tbe Web, where suppliers have successfully created sub- nets or private access areas in which "special" consumers can obtain membership by paying a fee or providing personal information. Once membership is obtained, the subnet affords consumers certain privileges sucb as special product discounts and improved customer ser- vice. For instance, ESPN's Sports Zone allows members to obtain inside information and reports on their favorite teams that are not available outside the subnet. In fact, the con- sumer is part of a captive buyer network because there is a time, information, and pos- sibly financial cost to exiting the subnet and registering for another one to compare prod- ucts. Moreover, assessing the subnet's value prior to entry is often difficult.

Otber examples include traditional businesses linking up to Internet Malls, which include a

476 MIS Quarterly Vol. 23 No. 4/December 1999

Grover & Ramanlal/Six Myths of Information and Markets

colleaion of online storefronts, each offering different products, accessible through a single Internet site. Many products are a result of partnerships between firms. Consumers could be hard pressed to make price comparisons of individual products due to bundling across products and even companies within the mall site (Wreden 1997b). Similarly, Reuters bundles commodity style news items and data services to meet the specific needs of clients in ways that allow charging higher prices. News col- lection by Reuters from multiple sources is facilitated by low-cost and open IT network architectures. However, the products are bun- dled in ways that make it difficult to compare prices and features across comparative offer- ings by competitors. Thus, suppliers have effertively created captive buyer networks or hierarchies and can consequently charge high- er prices.

An Economic Perspective on Mytti and Counter-myth #3 Myth: With open standards and declining search costs, customers need not depend on a single source of supply. This is particularly true for information-intensive products and services, which can be delivered through information net- works, as customers can easily search and acquire competitive products, tt therefore becomes difficult for the supplier to generate monopolistic rents and extract consumer surplus based on a relationship involving switching costs.

Counter-myth: While it is true that companies would find it difficult to sustain hierarchies based on idiosyncratic protocols, higher IT cap- italization could allow supplier firms to deliber- ately provide selective product information to targeted customers in ways that can inhibit product comparison. If product comparison is obscured, consumers have no frame of reference to assess the competitive price. They will tend to resort to their reservation price for making deci- sions. Information-intensive products in particu- lar allow the ease of "bundling" components, making it difficult, if not impossible, for con- sumers to compare products and make choices based on price. Thus, firms avoid the competi- tive outcome.

One way firms can extract consumer surplus along the demand curve is by successively adding components (e.g., upgrades) to the base product and correspondingly increasing its price by amounts that exceed the marginal cost of add- ons. This strategy targets jointly those customers who have the resources to pay the higher price as well as those who have a higher reservation price. Both are necessary for suppliers to extract surplus successfully. Consumers with higher reservation prices are willing to pay proportionately more for the upgrades because the surpluses they derive from the base product are sufficient to offset the premium they pay for the extras. In this manner, suppliers can extract surplus based on the maxi- mum price a customer is willing to pay rather than the minimum price a supplier is willing to accept. This is in contrast to the competitive outcome, where products are comparable and suppliers will lower prices (i.e., give away consumer sur- plus) to capture a larger consumer base. Here, consumers cannot compare individual products or prices directly because they are bundled, and they must assess price/value tradeoffs based on their reservation price.

As IT costs become arbitrarily small, firms are not limited to successive upgrades alone, instead, bundles of every perceiveci combination of com- ponent products can be offered. Comparative shopping across bundles could then becomes impossible, and the reservation price is the only basis on which choices can be made. In short, information selectively applied can impede com- parative shopping by consumers, and suppliers have a vested interest in using omniscient IT for this purpose. Indeed, some of the suppliers and bundles may altogether ignore consumers with smaller reservation prices in an effort to obtain the surplus from those who are willing to pay more. In contrast to Figure 4, traditional eco- nomic theory suggests that suppliers compete based on price, thereby promoting product com- parison across suppliers. The outcome that results in which surpluses accrue to consumers is clearly undesirable to suppliers. Indeed, from the supplier's vantage, the use of IT to avoid the competitive outcome is key.

Economic Example

To see how open architectures can sustain captive networks, consider how information

MIS Quarterly Vol. 23 No. 4/December 1999 ATI

Grover & Ramanlal/Six Myths of Information and Markets

S u p p l i e r s

B u y e r s

Figure 4. The Myth of Open Architectures (Myth 3)

services providers like Bloomberg and Zacks operate. Both service providers couple public information available on open architectures with proprietary information on their networks to create information products that customers value. For example, the public information corresponds to accounting data of publicly traded firms, and the proprietary information corresponds to the synthesis of analyst recom- mendations for stock performance. The two types of information are then combined in a convenient format and made available via the Web on a subscription basis ranging from sev- eral hundred to several thousand dollars per year. Other examples include Visual Numerics' use of the open-architecture Fortran and C++ programming languages together with its pro- prietary graphics and user interface software to create packaged products for programmers that command premium prices. The following example illustrates how these types of informa- tion may be combined to extract surplus from consumers.

Suppose there are two suppliers, labeled A and B, of information products and each supplier has a propriety component, labeled x and z, respec- tively. Supplier A's propriety information x can be thought of as Zacks' recommendation for stock performance, and supplier B's propriety informa- tion can be thought of as nonpublic information of security prices available on Blumberg's infor- mation-services network. Both these types of pro- priety information are potentially valuable to traders of financial securities. A third information component, labeled y, is available to both sup- pliers via open architectures. This common infor-

mation includes the accounting data of publicly traded firms. Thus, supplier A can bundle com- ponents X and y, while supplier B can bundle components y and z. The bundled information products are then made available to two cus- tomers, labeled 1 and 2.

Assume consumer Ts reservation prices for the three components, x, y, and z, are $5, $5, and zero, respectively; and consumers 2's reservation prices are zero, $5 and $5, respectively. In other words, consumer 1 values Zacks' proprietary information and the common accounting data, and consumer 2 values Blumberg's proprietary information and the common accounting data. These reservation prices may represent the monthly subscription rates that customers are willing to pay for on-line access. Assume that the marginal cost to the information-service providers of each component is $2.

Supplier A bundles x and y, while supplier B bun- dles y and z, the marginal cost of each bundle being $4. Both products are priced at $10. Due to the bundling, customers 1 and 2 will find it dif- ficult to compare the products that suppliers A and B provide, thereby excluding comparative shopping based on price (see Table 1).

Under this scenario, customer 1 purchases from supplier A and customer 2 from supplier B because in each case market price does not exceed the reservation price. In contrast, 1 will not purchase from B and neither will 2 from A, because in these cases market prices exceed reservation prices. Each supplier obtains con- sumer surplus equal to $6 (market price of $10 less the marginal cost of $4).

478 MIS Quarterly Vol. 23 No. 4/December 1999

Grover & Ramanlal/Six Myths of Information and Markets

Table 1. Product Bundling Scenario

Customer

Reservation Prices

Market Price

Marginal Cost

Supplier A (Bundles x and y)

1

$10

$10

$4

2

$5

1

$5

Supplier B (Bundles y andz)

2

$10

$10

$4

For the competitive equilibrium to hold (at least for component y, which is generic), suppliers must have incentives to lower prices with the potential to increase sales and profits. However, neither supplier has an incentive to lower the price below $10 because neither can draw the other's customer until the price hits $5. If one supplier lowers the price to $5, while both cus- tomers would be attracted, the consumer surplus which that supplier is able to extract would be reduced from $6 to $2 (market price, $5, less the marginal cost, $4, times the number of cus- tomers, 2). Thus, price competition is futile to suppliers.

Unbundling the products proves fruitless to sup- pliers as well. Suppliers A and B can charge a maximum of $5 for components x and z (equal to the reservation prices), respectively, while the market price for component y is $2 (equal to marginal cost) given direct competition between A and B in that market. Thus, unbundling reduces each supplier's surplus from $6 to $3. It is evident from this example that suppliers can use the pricing power of the unique components (x and z) to extract from consumer's surplus on the generic component (y).

The competitive equilibrium is also inhibited in the individual markets that suppliers A and B cre- ate. This follows because low IT costs encourage proliferation of new markets (i.e., structuring of new bundles) rather than entry into old markets (i.e., competition with suppliers of existing bun- dles). The former permits extracting surplus from consumers. Conversely, consumers derive sur- plus in the latter.

It is apparent that the market is being segment- ed where supplier A captures customer 1 only and supplier B customer 2 only, Each supplier

finds it unprofitable to capture the other suppli- er's customer. In turn, customers 1 and 2 are paying the maximum price they are willing to pay. In a competitive situation in the absence of bundling, at least component y would be com- parable directly across suppliers, allowing con- sumers to make decisions based on price. Notice that it is not the uniqueness of compo- nent y but rather its bundling with components X and z that permits suppliers to extract surplus from consumers who have a high reservation price for an otherwise generic product. Thus highly networked environments can reduce mar- ket effectiveness.

Myth #4: Linking multiple market centers using IT networks would result in consolidated markets that benefit the buyer

Counter-myth #4: Linking multiple market centers using IT networks could result in fragmented markets that benefit the supplier

A General Perspective on Myth and Counter-myth #4 The unprecedented growth of today's informa- tion networks is allowing the interconnection of diverse markets. For instance, what were once proprietary networks like American Airline's SABRE or EDI industry platforms are now becoming a part of the integrated network milieu. The myth is based on the argument that integrating market centers of buyers and suppli- ers with IT links can increase the size of the overall market, thereby improving the choices and prices for customers.

MIS Quarterly Vol. 23 No. 4/December 1999 479

Grover & Ramanlal/Six Myths of Information and Markets

The counter-myth however, implies that the IT connections themselves may not improve the nature of markets. Suppliers, driven by profit motives, may not be fully open in disclosing valuable information on the network when it may be used instead to gain strategic advan- tage. An example is the recently implemented linkages between the New York Stock Exchange and the Mexican Bolsa facilitated by American Depository Receipts (ADR) for cross-listed secu- rities. Mexican firms are permitted to list their stocks on the NYSE. Since trading of these stocks occurs on both exchanges, there is infor- mation flow between the markets. The objec- tives are for the NYSE to generate trading com- missions and for the Mexican Bolsa to become a more liquid market. While some improve- ment results from IT linkages, the markets remain fragmented in many respects as seen in trading cost and trading volume differences across the two trading centers (Domowitz et al. 1996). In short, IT linkages do not fully inte- grate the markets.

Similarly, the electronic integration of major security trading centers as mandated by Congress in 1975 has yet to be successful. Larger exchanges demonstrate unwillingness to fully reveal information in the expectation that they may profit from withholding it (Blume and Goldstein 1997; Grover etal. 1999). In the Fall of 1998, the Pacific Stock Exchange implemented an intranet-based trading system called Optimark that lets investors buy or sell in large volumes without telegraphing their moves to other traders, which helps keep the market from moving against the investors trading these large quanti- ties. These examples illustrate that network link- ages alone are not sufficient to render benefits to consumers if the supplier's incentive to profit from private information precludes making it freely available on the network.

While the previous examples pertain to securi- ties markets, important inferences can be drawn for information and product markets. Until recently, the concept of consolidated markets applied mainly to securities exchanges for the following reasons: financial securities are stan- dardized commodities and information sharing across financial markets was mandated by Congress and facilitated by electronic networks established by stock exchanges. With the advent

of the Internet, the trading environment for information and products now more closely resembles the trading environment for financial securities. The question is, what have we learned from securities markets that will tell us how networks of product and information mar- kets will evolve?

Despite the passage of congressional mandates for securities markets as far back as 1975, the impact of electronic linkages on the market's structure has only recently come to light. We believe a wait-and-see approach to the evolution of information markets on the Internet could prove costly since we believe it can significantly influence the balance of power between suppli- ers and consumers. Intuition suggests that on the Web, larger shopping sites (markets) could be reluctant to reveal full information because smaller sites could pre-empt their pricing. Therefore, despite the network linkages, the incentive to profit from information as opposed to freely revealing it may inhibit customers from getting all the benefits oî integrated markets.

An Economic Perspective on Myth and Counter-myth #4

Myth: If multiple markets are linked through IT networks so that buyers and sellers at different trading locations can access product and price information across markets, then the broader mar- ket would be more effective, benefiting the con- sumer. This follows because (1) the increased size of the combined market and (2) the efficiency of information flows across trading centers would alleviate potential inefficiencies in the individual trading locales, permitting comparative shopping, thereby making it more difficult for suppliers to extract consumer surplus. In short, a broader base of buyers and a broader base of sellers would result in more effective markets.

Counter-myth: The above reasoning underesti- mates the self-motivated incentives of suppliers in the individual market centers. Suppliers wish to maximize the surplus they can obtain from consumers. They might do this by inhibiting valuable information flows from their market to the other. For example, obtaining information from other markets but not revealing information to those same markets creates the potential to generate larger profits (e.g., Madhavan 1995). If

480 MIS Quarterly Vol. 23 No. 4/December 1999

Grover & Ramanlal/Six Myths of Information and Markets

S u p p l i e r s

S u p p l i e r s IT Conneclions M arkel .SIruclure 2

Markcl S t r u d u r e 1

B u vers

M H T k i ' l s e g m e n l B t i o D i s s u s t a i n e d d e s p i l r I T c u n n e c l i o n s a c r u s s I w c i g J u t i H l T i i u r k c l < > d u e I n ( h e m a r k F t s t r u c t u r e f a v o r a b l e li> e a c h s e g m e n t

B uyers

Figure 5. The Myth of Linking Markets (Myth 4)

•suppliers in both markets pursue this strategy (and they have the incentive to do so), the mar- kets would remain fragmented because trading information (e.g., price and quantity) in one mar- ket, though relevant, is unavailable to the other. Suppliers in the individual markets face reduced competition. Therefore, they are in a better posi- tion to expropriate surplus from their customers. In other words, tbese information-inhibition effects reinforce the conduct of business between sellers and buyers in tbeir respective markets. If markets bave no incentive to share information, the IT network provides little or no help in mak- ing markets more effective. Market effectiveness improves only through creation of tbe incentive to sbare information and/or change the structure of the tv̂ 'o markets (see Figure 5).

Economic Example

For this case, it is useful to consider securities markets. We provide a numerical example illus-

trating how the incentive to collect but not divulge information can lead to market fragmen- tation, thereby creating the opportunity for sup- pliers to obtain surplus from consumers.

The effect of information sharing (or the lack thereoO across markets can be seen by examin- ing tbe interaction of larger trading centers (e.g., tbe New York Stock Exchange) with tbe smaller regional exchanges (e.g., Cincinnati) (see Blume and Goldstein 1997). The Securities Act Amendment of 1975 bas resulted in the imple- mentation of IT linkages across these markets with tbe goal of sharing trading information fe.g,, prices and quantities of securities traded) for tbe potential benefit of customers.

Consider two markets, A and B, connected by IT. (see Table 2).

The larger trading center, market A, benefits from economies of scale. Security dealers at

Table 2. Security Markets Scenario

Stock Value

Ask and Bid Prices

Transaction Prices

Surplus Extracted

Market A (larger)

$10 (Known to A)

$10 1/2-$9 1/2

$10 1/4-$9 3/4

$0.25 per trade

Market B (smaller)

Unknown to B

$10 1/2-$9 1/2

$10 1/2-$9 1/2

$0,50 per trade

MIS Quarterly Vol. 23 No. 4/December 1999 481

Grover & Ramanlal/Six Myths of Information and Markets

this location collect information about the stock's value ($10) when setting ask (price at which the dealer sells) and bid (price at which the dealer buys) quotes ($10 1/2 - $9 1/2). The network allows the smaller trading center, mar- ket B, to free ride on information collected in the larger market. While the stock's value is unknown to market B, it matches the quotes $10 1/2 - $9 1/2 using the available IT link- ages. Of course, A expects B to free ride and therefore sets wide bid-ask spreads in an attempt to obscure information about the secu- rity's true value (Blume and Goldstein 1997). The wider spread makes It more difficult to ascertain the true value of the security and at the same time offers inferior trading terms to customers (i.e., higher ask price, lower bid price). In response, B draws customers from A by paying brokerage firms a small commission for directing orders to B and promising them prices as good as those posted in the larger market so that conflict of interest issues do not arise (Easley et al. 1996). But are customers get- ting the best price given the dealer's strategy in market A? Clearly, hiding information about the true value of the security benefits dealers in both markets but customers in neither. Thus, the unwillingness to share information because of self-motivated incentives results in fragment- ed markets. As a consequence, more surplus can be extracted from consumers, thereby mak- ing markets less effective. This outcome holds despite the presence of low cost IT linkages. Clearly, individual incentives must be addressed before markets can be more effective.

Myth #5: Expanding the customer base for a product using IT networks would result in greater benefits to buyers

Counter-Myth #5: Expanding the Customer Base for a Product Using IT Networks Could Allow Suppliers to Exploit Buyers

A General Perspective on Myth and Counter-myth #5

As the consumer base of products expands, the myth suggests that each consumer derives inher-

ent benefits arising from the number of other consumers using the same product. For example, the benefit derived from using Microsoft's Office Suite is not just that it permits word processing and spreadsheet analysis, but that it also facili- tates interaction between users of the same prod- uct. The size of the benefit depends on the num- ber of users. The Internet allows software com- panies to quickly build a customer base by giv- ing software away free of cost. The more con- sumers using the software, the greater their com- patibility in exchanging information related to the software, and the more the related software support. For instance. Real Networks is building a large installed base for its streaming technolo- gy, with more than 26 million users as of July 1998, by distributing its software free over the Internet. With the size of this installed base increasing, video streaming on the Internet has become prevalent.

Another plausible consequence (counter-myth) of a large customer base, however, is the suppli- er's ability to leverage its market power by charg- ing higher prices. For example, AOL aggressively builds its customer bases by enticing potential customers with offers of "free" hours, with the expectation of charging higher prices for prod- ucts sold via its Internet site once customers are locked in. HP takes advantage of its installed base in inkjet printers by charging a premium for its cartridges (Games 1998). Microsoft leverages its existing customer base by marketing subscrip- tion-based software through the Web. Similarly, Windows '98 and MS Office'97 might well be priced higher simply because customers cannot switch easily to competing products for two rea- sons: the cost of learning to use an alternative product and forfeiture of the advantages associ- ated with a network of customers. In fact, Microsoft's recent attempt to build a customer base for its browser software by bundling it with the operating system has been challenged by the Justice Department, which has recognized the market power associated with an installed cus- tomer base.

On the Web, McGraw-Hill Professional Book Group has its catalog of over 9,000 business and technical books on-line. It is receiving over 200,000 hits/month with a 30% annual growth in this number. Experimenting with a try-before- you-buy technology that allows readers to down-

482 MIS Quarterly Vol. 23 No. 4/December 1999

Grover & Ramanlal/Six Myths of Information and Markets

load chapters/materials from books (based on IBM's Cryptolope technology for copyrighted material over the Internet) and exchange ideas, McGraw-Hill is successfully building a network of customers for its product (Wreden 1997b).

An Economic Perspective on Myth and Counter-myth #5 Myth: In the absence of market fragmentation (myth 4), a larger customer base for a specific product increases collective buying power. Moreover, low IT costs enhance customer access to this product. If a product has a broader market, it is more likely to be sold through a market struc- ture rather than a relationship hierarchy. Further, more customers translates into intangible bene- fits for the consumer such as service and support infrastructure, informal information exchanges on the product, third party training, and greater confidence in vendor. All suggest accrual of sur- plus to consumers.

Counter-myth: As noted, the argument presented above can be framed in terms of network exter- nality (Katz and Shapiro 1985). Simply stated, the utility that a consumer derives from a product depends on the number of other users that are in the same product network (Brynjolfsson and Kemerer 1996). IT facilitates network externality simply by enabling networks, particularly for products that are complex. Complex products require specialized service and service infra- structure, standardized features that enable wide usage, interconnection with other products and product modules, and other features that create the potential for a large customer base.

Companies often invest in establishing a network by standardizing a product and creating a large customer base for its use (e.g., by giving away free software). Such a strategy calls for early investments with no immediate payoffs. The benefit that consumers derive from being part of the network can be conceived as a source of con- sumer surplus. Moreover, the surplus increases with the network's size at little cost to the suppli- er. The key issue that arises is the division of this surplus between the consumer and supplier.

The division of the surplus is driven by two fac- tors, market power and exit costs, both of which work to the supplier's advantage. Market power arises because the network inherently raises bar-

riers to entry by competing suppliers. Exit costs arise for customers who wish to switch to a com- peting product and therefore incur loss of the product's use as well as loss of the accompany- ing externality. As a result, suppliers can charge a price above their marginal cost permitting the transfer of surplus from buyers to sellers.

Exit costs influence comparative shopping among alternative products. Suppose a customer using a software product contemplates not pur- chasing the upgrade in favor of a competing product. By switching from the current product to a competing one, the customer (1) saves an amount equal to the price of the upgrade; (2) loses the value of the corresponding network externality; (3) incurs search and start-up costs to benefit from the externality of the competing product; and (4) incurs the cost of purchasing the alternative product. Costs due to (2) and (3) rep- resent hurdles to price-based comparative shop- ping between the two competing products. Unless the alternative product is lower in price by an amount that is sufficient to compensate for the hurdles imposed by (2) and (3), the buyer wilt not switch.

Suppliers can therefore hold the consumer base captive, charging a price well above marginal cost by an amount equal to the reservation price of the network externality. In doing so, suppliers can extract significant amounts of surplus from consumers. While it could be argued that this is a return on the investment for building network externalities, note that the amount of investment has no bearing on the amount of surplus that can be extracted. In the absence of competitive forces, the latter amount can be significantly larg- er than the former, in which case markets are less effective (see Figure 6).

Economic Example

Consider two suppliers, A and B, who market competing products but where only supplier A benefits from network externalities (see Table 3).

Suppliers face equal marginal costs of $10. Therefore, in a purely competitive market, the equilibrium price should be $10, with all sur- plus accruing to customers. However, supplier A builds a large customer base with product sup- port infrastructure resulting in product confi- dence and user connectivity that customers value at $5, i.e., supplier A has established a

MIS Quarterly Vol. 23 No. 4/December 1999 483

Grover & Ramanlal/Six Myths of Information and Markets

Su pplier w ith low network externality. Su pplier tannnt extract custom er surplus haKed im xternallty

Suppliers

Su pplier with |]if>li externality created by IT, Can generute customer surplus as there is no competition on the externality itself.

Buyers

Figure 6. The Myth of Expanding Customer Base (Myth 5)

network externality worth $5 per customer. Of course, establishing this network is not without cost; the cost of financing this network is part of supplier A's marginal cost. Supplier A can charge as much as $15 (marginal cost plus net- work externality) without fear of losing cus- tomers to supplier B. Indeed, by charging $14, supplier A can run supplier B out of business because for the latter supplier to draw cus- tomers, he or she must lower price to $9, which is below the marginal cost.

Here, the $5 consumer surplus (i.e., the excess of market price above marginal cost) goes to supplier A! This outcome manifests price dis- crimination, where the supplier can charge the consumer what he is willing to pay and is indicative of an ineffective market. Notice that this result holds regardless of what supplier A may have invested in the network externality. In other words, there are not competitive forces that draw a relation between the cost of the externality and the consumer surplus that is derived.

Clearly, today's IT tools and distribution infra- structure make it easier to establish externality for products, particularly information-intensive and software-based products. The captive cus- tomer base that results allows suppliers to cap- ture monopolistic rents, thereby making It more difficult for competitors to compete with-

out the externality. The short-run implication for suppliers is clear. They should invest in building an externality to render the market ineffective.

Myth #6: A low-price guarantee by suppliers in environments enabled by IT networks would result in markets that benefit the buyer

Counter-myth #6: A low-price guarantee by suppliers in environments enabled by IT networks could result in price fixing and higher prices for the buyer

A General Perspective on Myth and Counter-myth #6 The myth follows conventional thinking and pre- sumes that in networked environments like the Internet, customers can easily gather copious information across products and make informed choices on price and quality issues. Suppliers can nevertheless retain customers in this highly competitive environment by offering low-price guarantees, as is typical among physical stores like WalMart and Sears. In essence, the supplier offers to match the lowest price in the market that the customer is able to identify. Customers com-

484 MIS Quarterly Vol. 23 No. 4/December 1999

Grover & Ramanlai/Six Myths of Information and Markets

Table 3. Network Externality Scenario

Marginal Cost

Reservation Price of Network Externality

Market Price

Supplier A

$10

$5

$15

Supplier B

$10

0

$10

pare prices by suppliers of comparable products i)n the network. They then buy the product at the lowest offered price from that supplier offering the low-price guarantee. Clearly, the matching of [irices by suppliers seems to encourage price competition to the benefit of customers.

Alternatively, the counter-myth is based on the realistic premise that highly networked environ- ments also allow suppliers to track their com- petitors' prices. Such tracking could cause reluc- tance on the part of suppliers to lower prices. The reason is because lower prices may not have the intended effect of attracting additional customers if competitors are known to offer the low-price guarantee. Indeed, the supplier that lowers prices could even lose customers to other suppliers pro- viding the guarantee. This could have the effect of price fixing, rather than competitive pricing.

One set of businesses that can collect competi- tors' price data is the airline industry. The Airline Tariff Publishing Company (ATPCO) sends infor- mation electronically about fare changes to its subscribers, which means that every airline gets a near realtime look at what its competitors are doing. Airlines are therefore in a position to guar- antee low fares, being fully cognizant of their competitors' pricing information (Wreden 1997a). Similarly, NASDAQ is a network of com- puters where multiple security dealers can set bid and ask quotes on a specific stock. Each dealer's quotes are instantaneously available to all other dealers. Moreover, it is common practice for dealers to offer the low-price guarantee. Customers can go to any dealer or broker (who has an agreement with a dealer for the guarantee) and purchase the stock at the best price the mar- ket offers. In both the airline case as well as NAS- DAQ, the evidence suggests that this practice has increased rather than decreased prices (Huang andStoll 1996).

An Economic Perspective on Myth and Counter-myth #6

Myth: If we assume that search costs approach zero with pervasive and powerful IT, then com- parative shopping for simple products should force the competitive equilibrium. This is espe- cially true if suppliers offer a "low price guaran- tee," i.e., offer to match the lowest price of any supplier in the industry. Consumers can then search easily for the lowest price in the industry and be assured of that price from their supplier. To attract customers, suppliers would have to compete vigorously based on price, moving price rapidly to equal marginal cost, which results in the competitive outcome where all sur- pluses accrue to the consumer.

Counter-myth: The above argument is intuitively appealing, and low price guarantees can influ- ence consumer behavior. In highly networked environments, however, the negligible search costs might work as well, if not better, in favor of suppliers. We can assume in networked environ- ments that customers and suppliers know the prices offered by all competing suppliers in the market. If the suppliers charge different prices, customers would migrate to the supplier with the lowest one. This creates the incentive for each supplier to lower price in an effort to attract cus- tomers. In this scenario, the competitive equilib- rium holds.

However, if all suppliers offer the low-price guarantee, customers will be indifferent to where they shop. They can go to any supplier and get the lowest price in the market. Competing based on price becomes ineffective because no additional customers can be drawn to a supplier if that supplier lowers price (unlike the competitive environment). This situ- ation holds because, in effect, all suppliers

MiS Quarterly Vol. 23 No. 4/December 1999 485

Grover & Ramanlal/Six Myths of Information and Markets

Suppliers networked with so Ibal all suppliers know tach olhers' prites. There is NO incentive Tur suppliers to lower price h«cause unother supplier would gel the sale. Can set monopolistic prices through price fixing and gain customer surplus

Suppliers

Kvery Supplier orfers buyers the lowest price guarantee. Buyers can go lo

any supplier to get lowest price

Buyers

Figure 7. The Myth of Low-Price Guarantees (Myth 6)

lower their prices simultaneously. This simulta- neous movement of prices can have a perverse impact on competition because it not only removes the incentive to lower prices, but it actually creates an incentive to raise them. For the same reason that no customers can be gained by lowering prices, raising them will not lose customers. Moreover, if the supplier with the lowest offer in the market raised his price, all suppliers would benefit as a result (see example below).

In effect, price matching enables price fixing, where the price at which trade occurs bears no relationship to costs. Lowering prices would yield no benefit, but it could invoke retribution from other suppliers because the price and identity of all suppliers is known, whereas rais- ing prices would benefit suppliers (see Figure 7).

Economic Example

Assume three suppliers. A, B, and C, sell an iden- tical product, which has a marginal cost of $1. Suppose they price the product at $7, $6, and $5, respectively. In a non-networked environment, they could sustain these prices and generate monopoly rents due to high search costs for the consumer. If they all offer a low-price guarantee, they could still obtain monopoly rents as it is costly for the consumer to search the lowest price.

Under highly networked environments, however, the consumer knows the price of each supplier, and suppliers know the prices of their competi- tors. Suppose initially that only B and C offer the low-price guarantee. There is no incentive for C to lower the price below his/her initial level of $5 because no additional customers can be drawn since B offers a low-price guarantee. In contrast, C has an incentive to raise his/her price to the level of $6 posted by B because doing so will increase per-unit profits without losing customers to B. In that case, both B and C will make large profits (trading with the same customer base at a higher price). Of course, no customer will trade with A at $7.

For A to attract customers, he/she must either lower price from $7 to $6 or offer the low-price guarantee. Suppose the latter is adopted. Both B and C will lose customers to A. While B and C cannot get these customers back by lowering price, they can nevertheless increase profits by raising price to the level of $7 posted by A (again, trading with the same customer base at a higher price).

This illustrates that with low price guarantees, suppliers have no incentive to lower prices. In contrast, they have an incentive to raise prices. Thus, consumer surplus is absorbed by the sup- plier, while consumers have the perception that they are getting the lowest price.

486 MIS Quarterly Vol. 23 No. 4/December 1999

Grover & Ramanlal/Six Myths of Information and Markets

This myth is interesting because the implicit price fixing is difficult to identify by simply observing the market. Clearly, price matching impedes competition and renders markets less effective. Only after several in-depth comparative studies of the bid-ask spread on the NYSE vs. NASDAQ was the above problem revealed (see, e.g., Christie and Schultz 1994; Godek 1996).

Now, NASDAQ is revising Its trading rules to per- mit individual investors (trading certain stocks only) to compete based on price with dealers via limit orders that are observable to the public (NASD Press Release 1995). In other words, while dealers may implicitly agree among them- selves to fix prices by matching to the detriment of individual investors, these investors could avoid the inferior prices offered by dealers if they could trade among themselves directly. This direct trading is facilitated by investors submit- ting limit orders that are observable to other investors. In this environment, dealers cannot afford to offer inferior terms without the potential for losing customers who trade among them- selves directly and therefore are forced to com- pete based on price. Thus, the ability of low- price guarantees to fix prices is reduced, although not eliminated.

Implications

The basic message of this paper is simple. We propose that with the pervasiveness of open architectures and electronic commerce, both consumers and suppliers can and will leverage IT to their advantage. While conventional wisdom tends to emphasize reduced coordination cost for consumers in the evolution toward market structures, suppliers can be equally aggressive in using the same IT to maintain monopolies. Unlike the strategic systems of the 1980s, these strategies need not be based on idiosyncratic protocols tying buyers to suppliers, but are possi- ble in an environment of high and open connec- tivity. Therefore, this paper hopefully alerts acad- emics and practitioners to the altogether feasible strategies that can be enacted by firms in the pre- sent and future. While we devote more space to the counter-myths enacted from the suppliers' vantagepoint rather than more conventional

thinking, this does not reflect their inevitability. Rather, the equilibrium outcome in any market would be the result of a complex interplay between the vested interests of participants.

Forces Working in Favor of Suppliers Implicit in our discussion of counter-myths are three factors motivating suppliers to enact these strategies:

1. Increases in the set of feasible strategies: While IT by itself is neutral, its ubiquity and flexibility allows suppliers to manipulate or withhold information in ways that increase the coordination costs of the usually less cap- italized consumer. The possibility to manipu- late information flow using IT dramatically increases the opportunity set of potential strategies that suppliers can adopt.

2. Unique cost structure of information prod- ucts: Typical information products require sig- nificant up-front capital outlays for informa- tion gathering (or software development). The marginal cost of distribution or replication of these digital assets is (almost) zero. Under the scenario of intensive price competition, the competitive equilibrium based on traditional economic theory would yield price equal to marginal cost at zero. Clearly this outcome is very undesirable for suppliers. Therefore, they have every incentive to engage in strategies that prevent it from occurring.

3. Difficulty in defining the long run: It can be reasonably argued that supplier strategies, such as those described in this paper might not be sustainable in the long run, given mar- ket forces. For instance, any limit on informa- tion available to customers might create opportunities for intermediaries that work toward correcting it. The problem is defining the long run in a rapidly changing technolog- ical environment. We would contend that before the gestation period would end, sup- pliers could alter the environment sufficiently to resist the competitive equilibrium, and sus- tain positions of market power.

Therefore, as we have described, suppliers with their larger IT capitalization can process large amounts of data to engage in market segmenta-

MIS Quarterly Vol. 23 No. 4/December 1999 487

Grover & Ramanlal/Six Myths of Information and Markets

tion, and discriminatory pricing. They can also use IT to confuse buyers. Rather than simplify complex information products, they can use ver- sioning strategies or package them in a manner that precludes comparison. Also, supplier monopolistic positions in the value chain can be as easily automated and rendered more efficient using information technologies. Even when mar- kets are connected using these technologies, sup- pliers can attempt to withhold information to keep the markets fragmented unless incentives are provided to share information openly and honestly. Suppliers can also build up network externality by giving early incentives to ensure adoption of their products. The installed cus- tomer base is then leveraged with upgrades and new versions. Even suppliers that offer low-price guarantees have no incentive to reduce price in a networked environment.

Forces Working in Favor of Buyers

It is important to note that there are powerful forces at play on the other side of the equation that would allow customers to retain much of the consumer surplus.

The "indisputable" conventional arguments: The inherent nature and evolution of IT facilitates the ability of consumers to search, compare, and buy. Better information for the consumer clearly increases their power. While suppliers can resist this by deploying the strategies we describe, much of their battle will be against the inherent nature of IT itself, which is changing in a direc- tion of increasing openness.

Breakdown of assumptions: We based our counter-myths on the basic tenets of differential IT capitalization and information asymmetry between buyers and sellers. However, with open architectures, this gives rise to opportunities for new middlemen who can serve to bridge the information and IT gap. These "infomediaries" can serve as trusted custodians, brokers, and agents of consumer information by marketing it to businesses in return for favorable prices. By collectively bargaining on behalf of a number of consumers, these infomediaries can increase consumer bargaining power. Similarly, inexpen- sive software agents can shop for the best deal around on behalf of consumers.

Regulatory intervention: The role of regulators is primarily to ensure fair competition. Intervention will usually work in favor of the consumer as it reduces the probability of the counter-myths occurring in a dominant manner. As regulators begin to understand the implications of the new environment of electronic commerce, they can take action where they find evidence of false advertising, erroneous information, price dis- crimination, domination by one supplier, and price collusion.

While consumers may not have direct control over the monopolistic strategies of suppliers, awareness is a prerequisite to any concerted action. An informed consumer (with the time and inclination to be informed) is a powerful catalyst for market forces. Consumers need to continual- ly question their costs of customization, be reluc- tant to give up personal information, attempt to segregate product bundles, be wary of "free" products, have a propensity to shop around, try to be alert to getting over-committed to a prod- uct, and carefully evaluate their cost/benefit tradeoffs in purchasing decisions.

So Who Will Win? We don't know. However, it is our opinion that suppliers are not going to stand by as forces such as IT and infomediaries force them into intense price competition. The cost structure, particular- ly for digital products, would not allow them to survive. We would argue that suppliers would use every means possible to use the strategies described here to generate monopolistic rents. Even the proliferation of infomediaries, with the power of customer information behind them, would have to walk the tight line of generating advertising or transactional revenue from suppli- ers while maintaining bonds of trust with the consumer. In doing so, they would be loyal to the constituency that pays them and could in fact compound the network externality problem by reducing suppliers' costs of building a customer base. Similarly, infomediaries could reinforce the customization problem by allowing suppliers access to highly targeted information and allow- ing them to segment markets and price discrimi- nate. Even the much touted customer software agents would be only as good as the information suppliers choose to provide to them.

488 MIS Quarterly Vol. 23 No. 4/December 1999

Grover & Ramanlal/Six Myths of Information and Markets

Regulatory Intervention

Monopolistic Concerns

BUYER ^ To Buyer

STRATEGES •Product Search •Product Comparison •Awaroieis and Wariness •Use of [nfoniediaries •Control of Personal lnfi)rmation •Ki-guiator,v Intervention

Battle For Surplus

ToSupplie^SUPPLIER STRATEGIES

•CustumizationlVersioning) •Leveraging Supply Chain Monopoly •Creating Subnets and Bundling •Restricting Information Access Across Markets •Levera^ng Network b^xto'naUtics •Low Price Guarantees and I m p U d l P r k e Fixing

Information Technology Facilitates

Figure 8. The Battle for Consumer Surplus

Oi course, competition is the great equalizer. None of the strategies proposed would work if every product was commoditized and there was true competition. We would argue that markets in this new era require a delicate bal- ance between buyers, suppliers, and regulators. Underneath the surface is a complex set of conflicting objectives and self-interested motives at work. Competitive advantage would be based not only on product characteristics and the ability to use IT and information to leverage the complementary assets of the firm, but also on the "information game playing" as we have described and believe will continue to occur.

Figure 8 provides an overview of the major players and their battle for consumer surplus. IT can facilitate both buyer and supplier strate- gies, while regulators observe the battle, mon- itor conditions for effective markets, and inter- vene if necessary. Table 4 summarizes the issues facing buyers, suppliers, and regulators in the marketplace with respect to each of the six myths.

Conclusion Electronic commerce over large ubiquitous net- works will soon be conducted in routine fashion. While some may question the timeframe involved, few will question its imminence. In this transient phase of rapid technological change, it is difficult to see the real implications of these changes for both business and society. Recent writings have elaborated on the power of infor- mation technologies to reduce the costs of coor- dination and transactions and consequently to influence governance structures between buyers and sellers. Much of the popular press is also fair- ly aggressive in providing anecdotes of innova- tive companies that have leveraged Web-based technologies by expanding, improving, or modi- fying product and service offerings. A subliminal theme in all this hyperbole is the notion that these technologies are good and will provide the consumer with many more options, services and advantages. This article challenges this theme by presenting alternative scenarios in which these technologies do not necessarily work in the best interest of the consumer,

MIS Quarterly Vol. 23 No. 4/December 1999 489

Grover & Ramanlal/Six Myths of Information and Markets

O b

s e rv

a ti

o n

o f

R e g

u la

to rs

: C

o n

d it

io n

s f

o r

E ff

e c ti

v e

M a rk

e t

C e n

tr a l

Is s

u e

s f

o r

S u

p p

lie rs

C e n

tr a l

Is s u

e s f

o r

C o

n s u

m e rs

M y th

• If

s u

p p

lie rs

a ct

i n

t h

e i

n te

re st

of c

u st

o m

e rs

. •

If b

u ye

rs h

a ve

I T

t o o ls

t h a t

a llo

w th

e m

t o

d e

te c t

p ri

ce d is

cr im

in a tio

n ,

• If i

n fo

m e d ia

ri e s

e xi

st t

h a

t w o

rk in

f a

vo r

o f

th e

b u ye

r. •

If b

u ye

rs a

n d

s e

lle rs

d o

n o t

h a ve

a sy

m m

e tr

y in

I T

c a p ita

liz a tio

n .

• H

ow c

a n

1 c

a p

tu re

i n

fo rm

a tio

n a n d p

ro fil

e m

y cu

st o

m e

rs ?

• H

o w

c a n

1 d

e te

rm in

e t

h e ir

re se

rv a

tio n p

ri ce

? •

H ow

s h o u ld

1 r

e p a ck

a g e m

y p

ro d

u ct

s a ro

u n d t

h e

ir n

e e d s

a n d c

h a

rg e t

h e m

t h e

re se

rv a

tio n p

ri ce

? •

H o w

c a n

1 i

n h ib

it p ro

d u ct

co m

p a ri so

n ?

• A

re t

h e re

o th

e r

cu st

o m

e rs

i n

th e

m a

rk e

t th

a t

1 ca

n t

a ilo

r m

y p

ro d

u ct

s to

?

• A

m 1

p a yi

n g t

o o m

u ch

fo r c

u st

o m

iz a

ti o

n ?

• H

o w

d o

e s

m y

p ri

ce c

o m

p a re

t o

o th

e r

cu st

o m

e rs

? •

H ow

c a

n 1

c o m

p a re

p ro

d u

ct o

ff e

ri n

g s

fr o

m m

u lti

p le

s u

p p

lie rs

? •A

re t

h e

re i

n fo

m e d ia

ri e s

w h o c

a n

w o

rk i

n m

y in

te re

st t

o h

e lp

m e

d o t

h is

?

P ro

d u

c t

c u

s to

m iz

a ti

o n

, e n

a b

le d

b y I

T n

e tw

o rk

s ,

w o

u ld

b e n

e fi

t b

u y

e rs

• If s

u p p lie

rs a

ct i n

t h

e i

n te

re st

o f

cu st

o m

e rs

. •

If c

o m

p le

x p

ro d

u ct

s d o

n o t

e xi

st in

t h

e s

u p p ly

c h a in

. •

If I

T a

llo w

s b

u ye

rs t

o p

ri ce

in d

iv id

u a

l co

m p

o n

e n

ts e

a si

ly .

• If I

T f

a c ili

ta te

s co

n su

m e

ri sm

w h

e re

b u

ye rs

r e co

g n iz

e t

h e

m o n o p o lis

tic p

o w

e r

in t

h e s

u p p ly

ch a in

a n d

p re

ss u re

t h e

m o n o p o lis

tic s

u p

p lie

r.

• H

ow d

o 1

e xt

e n

d m

y m

o n o p o lis

tic p

o si

tio n b

y b u n d lin

g m

y p ro

d u ct

t h ro

u g h

a n

u m

b e r

o f

d is

tr ib

u tio

n ch

a n n e ls

? •

C a n

1 r

a is

e p

ri ce

s in

a w

a y

th a

t is

i n vi

si b le

t o e

n d b

u ye

rs ?

• H

o w

c a n

1 s

u st

a in

th e

co m

p le

xi ty

o f

th e p

ro d

u ct

s o

th a t

1 ca

n c

a p

ita liz

e o

n a

ll o

u ts

o u

rc in

g a

rr a n g e m

e n ts

?

• Is

t h e re

a m

o n o p o lis

tic p

o si

tio n i

n th

e s

u p p ly

c h a in

? •

Is t

h e

re a

c o m

p o n e n t

o f th

e p

ro d

u ct

t h a t

is s

im ila

r a

cr o

ss co

m p

e tit

iv e p

ro d u ct

s? •A

m I

b e in

g c

h a

rg e

d t

o o m

u ch

f o

r th

a t

co m

p o n e n t?

• Is

t h

e re

r e a lly

c o m

p e tit

io n a

t th

e co

n su

m e r

le ve

i o r

d o e s i

t ju

st a p p e a r

th a t w

a y?

In c re

a s e d

o u

ts o

u rc

in g

, e n

a b

le d

b y

I T

n e tw

o rk

s , w

o u

id l

o w

er p

ri c e s a

n d

b e n

e fi t

b u

y e rs

490 MIS Quarterly Vol. 23 No. 4/December 1999

Grover & Ramanlal/Six Myths of Information and Markets

(A o re 3

OC

o

ra ti

o n

c (A

o (A

p li e r

s s u

e s f

o r

S u

p ,

C e n

tr a l 1

in

iu m

e

c o o

3s f

o r

'a l

Is s u

i e n

ti

o

ly th

2

ir k e t

s o>_>

E ff

e i

o

io n

s I

"• cou

s to

f

o B "c CD ( -

ti n

ti

o CO

p lie

rs

a C/J

•

— 1

D ro

d i

n 1

b u n d le

m y f

• H

o w

c a

f O

e ri

n g

: ^t

o ff

p ro

d I

o m

p a re

î

•

2

1 IT

n e

o

•S J 9

L U

O

CJ

:u lt

d iff

ic a w

a y t

h a

t it

is in

s u

c h

%

jp p

li €

fe re

n t

s i

*-.

0)

o

01

te c

tu ri

rc h

i

re

CD

X I

3 O

^e rs

CO

•

jr s t

o c

o m

p a re

fo r

b u ye

t> 3

p ro

d

CO

1 m

u d

ly in

g to

e

ca

•

* ;

0)

0)

s a

n d

n ee

CD

a.

lU JO

O

TD C CO

o Q .

p ro

d u c t

th a

t 1

id u

ct E

w it h p

rc Id

le

13

a> o

•S

sa.

rs a

s c

> 3Q. J3

CD

CO

"o c

"O CO

se lle

r e

rs a

n

•

id d o i

n 1

l im

it p

ri c e s

l- •

H o w

c a

n e e

•5 c

jp p

li e r

c 0

o n .

rT\

IT c

a

c

T ie

tr y

E

CO

O)

Ji re

ct iy

r e ve

a lin

b y

n o t

c )l ie

r su

p p

to t

h i!

g ra

ti n

g

E

E •

• 0 0) 0 3

•D

in . _

rc h

ie s

ie ra

c

o rk

i

"S

•Jo

s e

x i

ne m

e d ia

•

p ri

c e

? p

ro d

u c t

o ff

e ri

T iil

a r

- —

0 c

e t

h e re

i :a

u s

CD

ly e r.

_ j

of t

h e

> D

d u c

n 1

a d d n

e w

p rc

• H

o w

c a

0

n

je tt

in

CO

ro

0

0

?

n

jn t

fn ;

th a t

a re

d if fe

re b

u n

d le

s •n

e t?

X I

h e s

u fe

e f

o r

t

S. Q. —

ip e ti to

rs '?

0

e d ia

r n fo

m

—

ia s e d

a n u

n b i

le re

CO

•

• D

C e ts

a n

1 c

re a

te s

u b n

• H

o w

c a

1 h e lp

?

0

n e t

• s u b

le m

t o e

n te

r th

e te

m p

t th

to

li n

g ;

ro d u

3, w

it h

o u

t re

v e

a fo

r a

f e

e t io

n a

b o u t

th e p

in fo

rm a

l

B •£

03

CJ CO

p lie

rs

Q.

( 0

•

is to

rr n

1 re

ta in

m y C

L •

H o w

c a

r m e .

>. 0

e n d e :

th e

re a

t

t o

J I

•

1 re

E _ai

.9" *;

ng m

u in

kJ

- 1

CO

it o

m e

r

0

" 0

a ti o n

' lim

it in

g i

n fo

rm b a s e b

v

CD C

0

g o

t

0

a d

e ,

t )n

m y t r;

• 0 03 to CO

n

til

%

n et

D) C

ir s u

s i

"c m u

d in

g

EO • ^

C CO

tu re

s

0

ke t

s tr

CO

•

it e rs

' O

S S m

a rk

e t

c e r

fl o

w a

c r

le r?

th e 0

a n d n

o t

03 - ^ CO

E

d r

e s u

'O U

li

CO

0

0

"a 01 c CO

a re

c h

d in

g n

1

le v e

ra g

e t

ra •

H o w

c a

0

CO

c

m a

ti (

—̂ 0

c

n 1

o b

ta

CO

0 X •

re

w

rk e l

re

"5

o li

d a

tc

en c 0 u

en

tr a

d in

to to 0 0

h a t

1 m

y m

a rk

e t

s o t

ru le

s i

n

*?

la rk

e

0 • Q

c

0

le p

ro d L

b CO

CD

01

X I

fi t

th e

0)

•°

p ro

v i

03 CO

CO

E

•

t e a s e o

f tr

a d

in c a n l

im i

d if fe

r re

t o

(D

o a

d h

1 h a v e t l

0 "a

•

io n .

cd

E 0

h a re

:iv e s tc

c

T ta

rk e

ts ?

a c ro

s s r

-c CO CD

u re

s

T3

D ro

ce tr

a d

in g [

to E

MIS Quarterly Vol. 23 No. 4/December 1999 491

Grover & Ramanlal/Six Myths of Information and Markets

CA k. n

jl a

t( )f

R e g

i ti

o n

CO

u CA.a O

p p

li e

rs

(0

le s

f o

r

tf) en

C e n

tr E

lA

E <n

C o

n

o *̂CA <D

(A

e n

tr

u

E

0) j £ CO

ti v

e r

E ff

e c

o CA

c

it io

c o o

"o to CD

0

0

.o t

in t

h

CO

lie rs

CL CL

CO

•

'e s t

o d e

i n

ce o

v i

Q .

• C

a n

1

tQ

g fo

i

t z • > .

-

O

i X •

1 y

0)

c

CD a. X UJ

er s

U JO

us t

o

le r

b a si

CO

b u ild

x : en

o £ r-

-e ve

i ro

d i

CL

0 [ -

i3 0

c

1

ro d

a. CD L .

£ C) CA CD

n

o u

si n

g 1

to

m e

r

o to

•

0

e ve

ra g

CO

th a

t 1 c

t o

E

c u

s tc

t o

0

n o t

(0 0)

• o

"5 o CA

o

0) c -

iin st

)b y a

g e

1

th e

o

0 E

0 m

a ta

l;

—

i >—o

fr e

e "

: CD 0

O

g 0

£

•

o CA

a>

0) % 0)

0

ir p ri

a n

o v e

o

ili ty

(

C\3 C

xt e

s to

m e r

o

a g e

m v

0

• C

a n

1

—

-E .c 0 Eo to o

i in

t

0

u o

in g

1

en

(A

0) ><

g a

js ly

•n tin

u o

L

o

n

b a s e

o ..̂ 0 D) "c o

W it

t p

p lie

rs 3

Q3

Jl ti

f

E

•

p g

ra d

e

CT ĉ

Tl

p ro

v i 1

• . ^

ro d

u ct

s Q_

:in g

Q

o o

JC tS

?

ip o .

lt a ry

p r

0

co m

p

0

CO

"cD

?r b

a s e

d

^s to

ri o

le ir

o m

e r

u ra

g e

c

o o 0

•C a

n 1

.9 a

o f

o n e

c

n a

tic

E CD O

E 2 •

It a

d d

iv e

ra g

e

" CD

u ct

u re

s 1 c

a n

b

^ o

h ^

s u

p p

i v a lu

e

O

c o

]i s la

ru s t !e

c

c

0

'5 cr •^

r-

o fi t

fo r

p r

:i ti o n .

0 CL Eo o

'*-

o u t

O f

ti n g

to rs

'c u

t o m

p e ti

/ u n d e i

" -O

13 to *- 0

•C a

n 1

b u si

n

0 CT

1 m

y la

i a s e ?

th e

m ,

c u

s to

to f

to 0 0 C

0

C t

in t

h i

CD

ie rs

p p l

•

t o

;k p

ri c e

tr a

c c lo

s e

ly

- d CO

ti

• H

o w

1

CT C

oE

ne e

0 CD

3=

0 . o Q. CO

er e

£

•

n d) S c i _

CO

o

<

e rs

.

E o 3 O

0

1 p

ro vi

d

c CD

le ti to

rs

u .

o CJ

o f

m y

<D

£ CO

CtJ

ie rs

Q .

CO

ip le

E

a.

% CO

0

o ( / I

m e n

st o

•

it e

e g u a ra

r 0 CJ

a

a l

o w

0

CO

3? .^— if) Q)o

•a

c en

0)

E c 2 "> c

o CO

s )

u se

1 d

ia ri

E "c

_c

p ri

c e

s Q CO

'2 •

C a n

1

0 £

a n

g

CD ^ .

er s

CO

0

0

•

u lt

re s

• a

3 O

(A

o

flU c

t

CO 0o n

ie r

1 II

s u p p i

CO

0

CO

o

>,

1 w

it h

n "

tz

c o

n ju

C l -

a rk

e

0 JZ.

t o 0 o

Q.

% O

e fi

t

c

CA

0)

E

CO9 0

0

0 n

o t

a ie

rs

g-

•

s in

g

o

; w

it h

O L

•e ti i

c o m

p

c to 0CT c CO x: o

CO

o

>.

CO

er e

£

•

o rs

0 Q

E o o

n t

h e

ir i

o

ti o

n

B U

J

o • -

i c u s to

o- co

— D. Q. to

Q .

".̂ ^ 3 E O)

o E

.9 Q.

t E o

o

CO

m e rs

JS tO

h ib

it C

L

c

• n

• H

o w

f

0

to <—

Q c o

s e

lle rs

u CO

CO

ye r

3

•

3S

a n d

.9

•E

fr o

m t

c

tio

CD N

1 it a li

IT o

a p

> .

sy m

m i

CO

)r s?

st itc

' o o m

p i

th o

s e

o

c ^

CT P

•i x in

n tit

r • p

ri c e

1 ji ri

n g

a

O CT

,io n

( 13

3 —

8 ^ :!; 0 •

c

lit K

0

JU JO

fo rc

e c

o c o

la ti

_ 0

492 M/S Quarterly Vol. 23 No. 4/December 1999

Grover & Ramanlal/Six Mynis of Information and Markets

In conclusion we would like to reiterate that the objective of this article is to challenge and expand existing thinking about the efficacy of information networks. The economic examples presented are intended to provide a tangible and reasonable face to rational counter-arguments. The myths presented are not illustrative of com- panies doing anything wrong or illegal. Instead they merely provide a broader context for the interpretation of information and markets. Whether these "myths" can be offset in the long run by the very nuances of the technology that created them remains to be seen.

A ckno wiedgemen ts

We v -̂ould like to thank the Center for International Business Education Research (CIBER), College of Business Administration, University of South Carolina, for their funding of this projea. We are also highly appreciative of the time and effort of Allen Lee and three anony- mous reviewers for their insightful comments that greatly improved the paper.

References

Applegate, L. Managing in an Information Age, Harvard Business School Publishing, Boston, 1995.

Apte, U. M., and Mason, R. O. "Global Disaggregation of Information-intensive Services," Managemenf Sc/ence {41:7), 1995, pp. 1250-1262.

Bakos, J. Y. "A Strategic Analysis of Electronic Marketplaces," MIS Quarterly (15:3), 1991a, pp. 295-312.

Bakos, J. Y. "information Links and Elearonic Markets: The Role of Interorganizational Information Systems in Vertical Markets," Journal of MIS {8:2), 1991b, pp. 31-52.

Blume, M., and Goldstein, M. "Quotes, Order Flow, and Price Discovery," Journal of Finance (52:1), 1997, pp. 221-240.

Brynjoifsson, E. and Kemerer, C. F. "Network Externalities in Microcomputer Software: An Econometric Analysis of the Spreadsheet Market," Management Science (42:12), 1996, pp. 627-1647.

Business Week, "Database Marketing," September 5, 1997, pp. 56-62.

Christie, W. G., and P. H. Schultz, P. H. "Why Do NASDAQ Market Makers Avoid Odd-Eighth

Quotes," Journal of Finance (49), 1994, pp. 1813-1840.

demons, E. K. "Information Technology and the Organization of Economic Activity: The 'Move to the Middle' Hypothesis," Journal of MIS (10, 2), 1993, pp. 9-36.

Computerworld, "IntegratedTracking," November 24,1997,p.41.

Domowitz, I., Glen, J., and Madhavan, A. "International Cross-Listing, Ownership Rights and Order Flow Migration," Working Paper, Marshall School of Business, University of Southern California, 1996.

Davis, ). "Are You Next: 20 Industries That Must Change," Business 2.0, March 1999, pp. 44-54.

Easley, D., Kiefer, N., and O'Hara, M. "Creme- skimming or Profit-sharing? The Curious Role of Purchased Order Flow," Journal of Finance (51:3), 1996, pp. 811-833.

Evans, P. B., and Wurster, T. S., "Strategy and the New Economics of Information," Harvard Business Review (75:5), September-October 1997, pp. 71-82.

Games, L. "High-Tech Inks Make Green, But Formulas May East Metal," Wall Street journal, August 19, 1998, p. 16.

Godek, P. E. "Why NASDAQ Market Makers Avoid Odd-Eighth Quotes," journal of Financial Economics (41), 1996, pp. 465-474.

Grover, V., Ramanlal, P., and Segars, A. H. "Information Exchange in Electronic Markets: Implications for Market Structures," International Journal of Electronic Commerce (3:4), Summer 1999, pp. 89-102.

Gurbaxani, V., and Whang, S. "The Impact of Information Systems on Organizations and Markets," Commun/cat/ons o/'fhe/ACM (34:1), 1991, pp. 59-73.

Hagel, |., and Rayport, |. F. "The Coming Battle for Consumer Information," Harvard Business Review (75:1), )anuary-February 1997, pp. 53-65.

Hess, K. M., and Kemerer, C. F. "Computerized Loan Origination Systems: An Industry Case Study of the Electronics Markets Hypothesis," MIS Quarterly (18:3), September 1994, pp. 251-275.

Huang, R. D., and Stoll, H. R. "Dealer Versus Auction Markets: A Paired Comparison of Execution Costs on NASDAQ and the NYSE,"

MIS Quarterly Vol. 23 No. 4/December 1999 493

Grover & Ramanlal/Six Myths of Information and Markets

journal of Financial Economics (41), 1996, pp. 313-357.

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

Katz, M. L., and Shapiro, C. "Network Externalities, Competition, and Compatibility," The American Economic Review (75-3), 1985, pp. 424-440.

Lee, H. G. "Do Electronic Marketplaces Lower the Price of Goods," Communications of the ACM (41:1), 1998, pp. 73-80.

Madhavan, A., "Consolidation, Fragmentation, and the Disclosure of Trading Information," Review of Financial Studies (8:3), 1995, pp. 579-603.

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

McCune, J. C. "Website Oversight," Beyond Computing, April 1997, pp. 42-45.

NASD Press Release, "Individual Investor Orders Will be Displayed in Best Quote on NASDAQ Under New Proposal," September 25, 1995.

Porter, M. E., and Millar, V. E. "How Information Gives You Competitive Advantage," Harvard Business Review (6A:4), July-August 1985, pp. 1 49-1 60.

Quelch, J. A., and Klein, L. R. "The Internet and International Marketing," Sloan Management Review 07-3), Spring 1996, pp. 60-75.

Shapiro, C , and Varian, H. R. Information Rules, Harvard Business School Press, Boston, 1999.

Shapiro, C , and Varian, H. R. "Versioning: The Smart Way to Sell information," Harvard Business Review (76:6), November-December 1998, pp. 106-114.

Varian, H. Microeconomic Analysis, Norton, New York, 1984.

Widdifield, R., and Grover, V. "Internet and the Implications of the Information Superhighway for Business," Journal of Systems Management (43), May-June, 1995, pp. 16-24.

Wilder, C. "Reaching for the Web," Information- week. August 26, 1997, pp. 47-58.

Williamson, O. E. Markets and Hierarchies: Analysis and Antitrust Implications, Free Press, New York, 1975.

Williamson, O. E. "Transaction-cost Economics: The Governance of Contractual Relations.

Journal of Law and Economics (22:2), 1979, pp. 233-261.

Wreden, N. "Turning One Success," Beyond Computing, September 1997a, pp. 19-24.

Wreden, N. "Web Wares," Beyond Computing, January-February, 1997b, pp. 27-33.

About the Authors

Varun Grover is a professor of information sys- tems and Business Partnership Foundation Fellow at the Daria Moore School of Business, University of South Carolina. Dr. Grover has published extensively in the information systems field, with over 100 publications in refereed journals. He was recently recognized as the most productive researcher in the field from 1991-1997 based on publications in the top IS journals. Dr. Grover's current areas of interest are electronic commerce, business process change, and organizational and interorganiza- tional impacts of IT. His work has appeared in the MIS Quarterly, Information Systems Research, Journal of Management Information Systems, Comm-unications of the ACM, Decision Sciences, IEEE Transactions, and California Management Review, among others. He recently edited a second book entitled Process Think: Winning Perspectives for Business Change in the Information Age, and two special issues of the JMIS on the topic of business process change. Dr. Grover has also served as the special editor for issues of Database, which focused on IT futures, celebrat- ing the 50th anniversary of ACM, the International Journal of Electronic Commerce, and Decision Sciences. He is the recipient of the Outstanding Achievement Award from the Decision Sciences Institute, the Distinguished Researcher Award, and two-time winner of the Alfred G. Smith Award for Excellence in Teaching. Dr. Grover is currently serving on the Board of Editors/associate editor of five IS jour- nals and is an active member of INFORMS, ACM, DSI, and AIS.

Pradipkumar Ramanlat is associate professor of finance at the University of Central Florida. He received a B.Sc. in physics, a Ph.D. in theoretical physics, and a Ph.D. in finance at the University of Michigan, Ann Arbor. Dr. Ramanial has pub- lished in various finance journals including Financial Management, Journal of Financial

494 MIS Quarterly Vol. 23 No. 4/December 1999

Grover & Ramanlal/Six Myths of Information and Markets

Research, journal of Financial Intermediation, He is also working in the area of electronic mar- Journa! of Portfolio Management, Review of kets and recently published a paper on this topic Quantitative Finance and Accounting, Journal of in the International Journal of Electronic Financial Engineering, lournal of Financial Commerce. Dr. Ramanlal's work has played a 5emces/?esearcft, and R/S/C, as well as in various key role in policy decisions relating to trading physics journals including Physical Review rules on NASDAQ, and he is listed in Who's Letters, Physical Review, and journal of Physics. Who in Finance and Industry.

MIS Quarterly Vol. 23 No. 4/December 1999 495