Chapter 2. Information and Uncertainty: A Conceptual Framework

Introduction

To answer the question Are there any circumstances under which an intermediary can succeed in the information economy?, the general function of an intermediary must be understood. Also, before examining the role that intermediaries play in reducing uncertainties in the digital age, it is necessary to lay out a conceptual framework, focusing on uncertainties of information in any economic system. Only by understanding the fundamental role that information plays in the economy can we examine the factors that give rise to the role of the intermediary. This conceptualization will serve as a basis for the following chapters, in which the intermediary is analyzed in both historical and present day markets. Furthermore, by understanding the general principles underlying the function of the intermediary, the viability of the "infomediary" model in the digital age can be analyzed.

To this end, the chapter first defines information and its role in the economy, focusing on the uncertainties to which information gives rise. Particular emphasis is given to information asymmetry and the commodification of information. Secondly, the chapter identifies how technologies serve as catalysts for changing uncertainties related to information. Considering all these factors, the next section presents two theories-agency theory and transaction cost theory-each of which seek to explain the conditions of uncertainty that are inherent in any economy. Next, the chapter defines the role of intermediaries, how they function, and how they reduce uncertainty and transaction costs. Based on this analysis, this chapter identifies the key conditions under which intermediaries can be viable economic actors.

Information and its role in the economy

i. Information defined

Information is a "fuzzy" concept, which has become the subject of economic analysis only in the past 100 years (Rose, 1999, 6). Economists differ in their definitions of information.3 However, given the focus of this thesis on the role of intermediaries, a definition that relates to uncertainty is most appropriate. Thus for the purpose of this thesis, the following definition will take precedence. Information is that which can be exploited to reduce uncertainty in decision-making (Rose, 1999, 10). Uncertainty is defined as "the dispersion of individuals' subjective probability (or belief) distributions over possible states of the world," (Hirshleifer, 1973, 31). Thus, information is a message that an individual receives that reduces uncertainty in his or her environment. In economics, this refers to the knowledge of prices, agents, and supply and demand of products.

In some cases, uncertainty also creates a market for information. When agents are uncertain about future events-the quality of products, or the honesty of others-they look to purchase information to reduce uncertainty (Kingma, 2001). In this sense, information decreases the risk involved in complex transactions. Risk is "the potential for loss when uncertain future events may cause economic harm," (Kingma, 2001, 89). To understand how agents can exploit information, it is first necessary to operationalize these uncertainties.

All information can be described by common properties. For one, information is an intangible resource and an immaterial good (Rose, 1999). Second, information is a highly fungible resource (Sampler, 1998). In this sense, it can be exchanged or interchanged with other information. Yet, information varies in its degree of fungibility or alternative uses(Sampler, 1998).4 Thus, the nature and value of information change according to its use. In addition, the production of information goods involves high fixed costs and low marginal costs (C. Shapiro & Varian, 1998). There are also high sunk costs in creating information, but it is relatively cheap to replicate. Thus, "information is costly to produce but cheap to reproduce," (C. Shapiro & Varian, 1998, 3). Unlike tangible goods, information is subject to perfectly increasing returns. Once the initial investment is made, information can be reused at no additional cost. Likewise, information can be replicated at zero cost and without limit, and it will never wear out or deteriorate, unlike tangible goods. However, information may become obsolete or untrue with time, and its value may decrease the more it is replicated (Evans & Wurster, 2000).

Information is a non-exclusive good, a feature that generates uncertainties in markets where information is sold (Kingma, 2001, 7). For example, when information transfers between a producer and user, usually only a copy of the information is sold. The producer keeps the actual information. Likewise, the seller retains the information even once it is sold. Thus that same piece of information can be enjoyed by more than one consumer simultaneously (Kingma, 2001, 7). The buyer can also copy and distribute the information without the seller's knowledge. Thus, the seller risks loss of compensation for the information's future uses. The non-exclusive nature of information also induces a disclosure problem for agents selling it, because its use by one agent may reveal the information to others. Moreover, when agents hesitate to reveal information additional market uncertainty is created. As a result, agents are likely to under-invest in the production of knowledge and information (Rose, 1999, 18).

Market uncertainties also arise because the true value of information cannot be predicted ex ante ((Rose, 1999, 18-19). More specifically, a consumer cannot know the value and worth of a piece of information before buying it. If the seller exhibits the information before selling it, it has in effect been given away. Thus, the buyer is at risk, never knowing the exact nature of the information being purchased. As in the case of non-exclusivities, these uncertainties may also lead to an underdeveloped market for information.

Given these unique characteristics, information does not easily fit in the traditional neoclassical economic model. As Boisot has pointed out, "Neoclassical economics has addressed the problem of information goods somewhat schizophrenically," (Boisot, 1998, 76). On one hand, economic theory treats information as a free good that is not subject to trading and is instantaneously available to all economic agents. In this respect, it functions as a support to exchange and "to foster the perfect foresight so essential to efficient markets," (Boisot, 1998, 76). Conversely, neoclassical economics also treats information as a good that, through artificial means, can be made subject to the same scrutiny of physical goods and thus also to the normal rules of trading. In the first case, information has utility, but because it is in abundance, it has no value. In the second case, it has both utility and scarcity, and therefore has value (Boisot, 1998).

ii. Information: an uncertain commodity?

Because information is different from tangible goods, questions arise as to whether it can function as a true commodity in the marketplace. An economic commodity is an item that can be traded for a price (Allen, 1990, 268). To be classified as a commodity, goods must directly or indirectly satisfy a human need, they must be limited, and their production must compete with other commodities for scarce resources (Rose, 1999, 36). Furthermore, buyers and sellers must be able to judge whether the properties of a good satisfies their needs. As well, they must be secure in the property rights of the commodity, so they can use the resource, retain the returns, alter the resource, and alienate the resource (Tietzel, 1981, as cited by Rose, 1999, 37). Lastly, the commodities must possess an exchange value (Rose, 1999, 36).

According to these criteria, information possesses some characteristics of a commodity. The fit, however, is problematic, which is the source of additional market uncertainties. As we have seen, information is non-rivalrous insofar as the use of information by one person does not exclude its use by others. Moreover, because of its low marginal costs, it does not compete for scarce resources, and it is easily reproduced. Equally problematic, maintaining property rights in intangible resources, such as information, is very difficult because such commodities are non-exclusive. Thus, the question as to who owns the information arises. Both the buyer and the seller can simultaneously possess the information, and use it as they chose.

iii. Information asymmetry

The problematic nature of information as a commodity exacerbates the problems of information asymmetry and imperfect information, which are features of any economy. Asymmetries are associated with the uneven distribution of information in the marketplace (Phlips, 1988, 4). When information is commoditized and exchanged for value in the marketplace, certain agents will invariably have more information than other agents. For example, a buyer may have more complete information than the seller, and visa versa.

Information asymmetries may occur for a number of reasons. For example, in some markets, information is available but producers are not willing to reveal it to consumers. A producer will get a better price if he can hide information from consumers. These uniformed or misinformed consumers make poor decisions about a product's value. Thus, they are reluctant to buy goods and services, thereby reducing the number of transactions in a market (Kingma, 2001).

Information asymmetry also causes information impactedness. Defined by Oliver Williamson, information impactedness is the condition that occurs when the buyer and the seller have knowledge of different and essentially private information when they take part in complex contracting (Williamson, 1985, 51). This condition is costly to overcome and gives rise to a trading hazard, occasionally resulting in market failure (Akerlof, 1970, cited by Williamson, 1985, 212).

Asymmetrical information negatively affects market transactions. At a minimum, asymmetries make the market operate less efficiently insofar as prices will not represent real costs and benefits (Kingma, 2001, 92). In the worse case, if parties are discouraged from transacting, asymmetries can undermine the market all together. Economist George Akerlof (1970) provides the classic example of such a case in his article, "The Market for Lemons." As he describes, the seller has an incentive to misrepresent the car's quality, and the buyer has little incentive to believe what the seller says. This uncertainty prevents transactions between high quality buyers and sellers (Akerlof, 1970).

To overcome the problems of imperfect information and information asymmetry, a third party can mediate between the buyer and the seller. One solution is the creation of an agent who will act in both the user's and producer's interest (Steinmueller, 1992). "An agent's incentives to compromise the producer's information by revealing it to [the] user can be offset by contracts and the incentives to continue to gain economic returns through repeat dealing with producers," (Steinmueller, 1992, 191). Repeat dealing and reputation effects would discourage misrepresentation of information to the buyer. This solution aligns incentives between producers and users (Steinmueller, 1992, 191). Thus, a third party -an intermediary-reduces information asymmetry.

Technology and information

Technology is intrinsically tied to information in an economy because technology advances affect how information is produced and transmitted. In the past, major technological advances improved communication techniques, which affected both the use of information and physical transport. "Information technologies related to communication and transports have a common property, namely to be very general (generic) in application, and contributing to the coordination and filtering processes of the economy they exercise leverage effects on the entire economic system," (Gunnar, Folster, Lindberg, Pousette, & Taymaz, 1990, 18).

Examples of technologies that have changed the distribution and control of information include the printed word, the steam engine, electricity, standardization, automobile transport, financial institutions, electronics based information technology, and general education. The printed word was a "path-breaking production technology (that) made it possible to pass on large volumes of knowledge in the abstract form of written information, a technology in itself," (Gunnar et al., 1990, 19). With the advent of the printing press, barriers to the spread of information and knowledge were removed. The advances in transportation systems (notably steam engines and automobiles) also changed communication methods, increased the transport of goods, and thus increased the flow and exchange of information (Gunnar et al., 1990).

As technology advances, certain uncertainties are resolved. But, at the same time, new uncertainties emerge as relationships are restructured (Garcia, 2002, 51). "Although technologies can reduce transaction costs and improve economic performance, their usage will generate a whole range of new uncertainties and transaction costs,"(Garcia, 2001, 3). To contend with these new uncertainties, new means of control must be established. "For, while reducing information-related costs in one part of the transaction chain, these solutions serve simultaneously to generate new transaction costs elsewhere in the chain, thereby creating anew the need for organizational and technological innovations,"(Garcia, 2002, 47). Information and communications technologies serve to control economic activities of production, distribution and consumption of goods and services (Beniger, 1986, 16). "A society's ability to maintain control-at all levels from interpersonal to international relations-will be directly proportional to the development of its information technologies,"(Beniger, 1986, 8-9).

Technology also affects the commoditization of information in the economy. Because technologies offer increased scope and reach in communication across the globe, information has a greater ability to be valued and exchanged in the marketplace. "The increased availability and capacity of communication networks stresses the basic characteristics of interdependence, inappropriablilty and externality of information as a commodity. At the same time it makes information an evergrowing strategic input in decision making, in production and consumption," (Stiglitz, 1989, 198). Thus, technology increases both the flow and value of information in an economic system.

Contending with uncertainty

Neoclassical economics fails to take into consideration the uncertainty of information, requiring alternate theories to explain the organization of markets. In this section, two of these theories are examined: agency theory and transaction cost theory. Both offer explanations of and solutions for dealing with uncertainty, and thus help to explain the role of the intermediary.

i. Agency theory

One relatively new theory that accounts for uncertainty is agency theory. This theory, first developed in 1976, examines the contractual relationships in an organization (Karake-Shalhoub, 2002, 103). Agency theory assumes that human beings are rational, risk-averse, and motivated by self interest (Karake-Shalhoub, 2002, 10). The theory aims to develop efficient organizational and contractual structures that can overcome problems of uncertainty arising in economic relationships. The main sources of these problems are the asymmetric distribution of information between actors, and uncertainty about the future states of the environment and actions taken by the agents (Rose, 1999, 27).

In agency theory, there is a defined relationship between the "principal" and the "agent." The agent (provider of the service) acts on behalf of the principal (user of the service). Information asymmetry affects the principal-agent relationship. Because both the principal and the agent are individuals who act in their own self-interest, a conflict can occur when both parties attempt to maximize their individual self-interest (Karake-Shalhoub, 2002, 105). Uncertainty occurs when the agent's action is not directly observable by the principal (Rose, 1999, 27). Without the ability to monitor the agent, the principal may assume that the agent is hiding information, which is a problem of moral hazard.

Agency costs incur from uncertainty in principal-agent relationships. Agency costs are the losses resulting from imperfect information, and they can result directly from the conflict between agent and principal. Moreover, agency costs can reduce the value of an organization. Thus, agency theory sets out to explain how to best coordinate relationships to reduce these costs (Karake-Shalhoub, 2002, 109).

Agency theory offers solutions to help economic actors reduce agency costs and mitigate problems of uncertainty, moral hazard and adverse selection in the market. For example, signaling, which reveals private information to the buyer, prevents adverse selection. Risk sharing--wherein the principal bears a portion of the risk to minimize opportunism--can overcome the problem of moral hazard. Furthermore, uncertainty can be diminished by using incentive schemes and control mechanisms in contracts (Karake-Shalhoub, 2002).

ii. Transaction cost theory

Another valuable theory for contending with uncertainty is transaction cost theory. Transaction cost theory originated in the early twentieth century, and, like agency theory, it only recently came into focus in economics. Transaction cost theory is concerned with how firms might best be organized to reduce information related costs. Transaction costs, as originally defined by Ronald Coase in 1937 in his article "The Nature of the Firm," add inefficiencies to the market (Coase, 1937, 390). Both Kenneth Arrow and Oliver Williamson played a seminal role in expanding on Coase's postulations in the 1970's and 1980's. Williamson defines transaction costs as the "costs of running the economic system,"(Williamson, 1985, 18).

Transaction costs add to the price of a good or service, and involve search costs, information costs, bargaining costs, decision costs, policing costs and enforcement costs (Downes & Mui, 1998). In business transactions, transaction costs also include the negotiating, monitoring and enforcement costs associated with the transfer of goods and services between the firm and the customer (G. R. Jones & Butler, 1988, 204). Transaction costs can apply to both the firm and the consumer, yet it is often the consumer that faces higher transaction costs, leading to information asymmetry in the market. Economic agents seek to reduce transaction costs, making the market more efficient (Downes and Mui, 1998, 38).

Transaction cost economics characterizes economic agents as having several determining features: bounded rationality, opportunism, and asset specificity (Williamson, 1985, 30). Agents are subject to bounded rationality when making decisions, "whence behavior is intendedly rational, but only limitedly so," (Simon, 1961, as cited by Williamson, 1985, 30). In this respect, transaction cost economics differs from agency theory, which assumes that economic agents are rational. In transaction cost economics, environmental and social factors limit an agent's rationality.

Opportunism often involves blatant and subtle forms of deceit (Williamson, 1985, 47). This refers to the incomplete or distorted disclosure of information, especially the effort to mislead, distort, disguise, obfuscate, or otherwise confuse other agents (Williamson, 1985, 47). Opportunism is responsible for conditions of information asymmetry and uncertainty, which, according to Williamson, "vastly complicates problems of economic organization," (Williamson, 1985, 48).

The most critical element in transaction cost economics is asset specificity (Williamson, 1985, 30). Transactions that are supported by investments in transaction-specific assets experience "lock-in" effects (Williamson, 1985, 53). Assets that are specific to the transaction include cost advantage, such as a unique location or learning, or task-specific labor skills (Williamson, 1985, 54). There are four types of asset specificity: site, physical, human asset, and dedicated assets(Williamson, 1985, 55).5

Transaction cost economics thus seeks to explain how uncertainties in the market might be reduced. Williamson proposes that an agent should "organize transactions so as to economize on bounded rationality while simultaneously safeguarding them against the hazards of opportunism,"(1985, 32). Williamson also argues that uncertainty would disappear if individuals were either fully open and honest in their efforts to realize individual advantage, or if full subordination, self-denial, and obedience could be presumed in the transaction (Williamson, 1985, 49).

Trust plays an important role in transaction cost economics. Karake-Shalhoub defines trust as "a social lubricant that allows consumers to transact with merchants who are not part of their immediate network," (2002, 38). By placing trust in a merchant, the consumer alleviates the perception of risk in a transaction. High levels of perceived risk correspond with a high level of trust needed to execute a transaction. Thus, "when risk is present, trust is needed to make transactions possible," (Karake-Shalhoub, 2002, 38).

Distrust is an uncertainty that increases transaction costs. When opportunism is present, trustworthiness is rarely transparent (Williamson, 1985). Relationships are easily invaded and exploited by agents who do not possess trustworthiness. In other words, a lack of trust on the part of some actors decreases trustworthiness in all relationships, because an agent cannot be sure whom to trust. If one party trusts in another, it is making itself vulnerable to the other party's behavior (Karake-Shalhoub, 2002). To overcome this, agents must make concessions to the effects of opportunism (Williamson, 1985).

There are several ways to build trust in economic relationships. First, competence can be a source of trust in asymmetric relationships. Competence means that parties display a level of professionalism, that they have the capability to follow through on promises, that they can realistically judge a situation, and they have strong interpersonal skills (Karake-Shalhoub, 2002, 31). Other factors for establishing consumer trust in a seller organization are reputation and size (Karake-Shalhoub, 2002, 39). Reputation is the extent to which buyers believe that the selling organization is honest and concerned about its customers. "The better the seller's reputation, the more the seller has presumably committed resources to build that reputation, the higher the penalty from violating the consumer's trust, and hence the more trustworthy the seller is perceived to be," (Karake-Shalhoub, 2002, 39). Size matters because a large organization signals to the consumer that the business has invested significant resources in the organization. Thus, the organization stands to lose a great deal if it acts in an untrustworthy way. Thus, the larger the firm, the greater the perception that it is acting in the customer's best interest (Karake-Shalhoub, 2002).

Again, George Akerloff's market for lemons exemplifies the importance of trust in transactions (1970). Trust plays a large role in the market for lemons. The presence of dishonest agents in the market who offer inferior goods tend to drive the market out of existence, because inferior products drive out the legitimate business (Akerlof, 1970). "The cost of dishonesty lies not only in the amount by which the purchaser is cheated; the cost also must include the loss incurred from driving legitimate business out of existence," (Akerlof, 1970, 495).

Intermediation

Having presented two theories to explain uncertainty and information asymmetry in a market, several questions are raised: what role do intermediaries play in this framework? Can intermediaries step in to reduce uncertainties? Intermediation is a possible solution to the problems of uncertainty in the economy. An intermediary steps in to overcome information asymmetry, information impactedness, distrust, and high transaction costs associated with information (Rose, 1999). An intermediary functions by seeking out suppliers, finding and encouraging buyers, selecting the buy and sell prices, defining the terms of transactions, managing the payments and keeping records of transactions, and holding inventories to provide liquidity or availability of goods and services (Spulber, 1996). Intermediaries provide utility by increasing the chances of a successful match between buyers and sellers, thus the need for an intermediary will come about because of frictions in the market (Cosimano, 1996).

Intermediaries relieve both agency costs and transaction costs. Intermediaries emerge in markets where transaction costs are high. As Coase and William purport, firms are organized to minimize the role of transaction costs (Coase, 1937; Williamson, 1985). Thus, the intermediary can be interpreted as a firm that acts to reduce transaction costs (Cosimano, 1996). The intermediary can also be interpreted as a mechanism for controlling uncertainty and sharing risk to reduce agency costs.

The roles available to an intermediary are determined by the types of uncertainties present in the market (Spulber, 1996). The occurrence of uncertainty is a prerequisite for the existence and the justification of intermediaries (Gumbel, 1985, as cited by Rose, 1999, 58). In markets where uncertainty is present, intermediaries coordinate transactions with brokering activities. When opportunism is present, intermediaries create market information and provide guarantees for quality. Likewise, when it is costly to observe the actions of agents, intermediaries provide monitoring and contracting support.

There is something intrinsic in intermediation which solves the problems of moral hazard and appropriability which tend to inhibit the production of information (Leland and Pyle, 1977, as cited by T. S. Campbell & Kracaw, 1980, 863). To eliminate information asymmetries regarding the qualities of assets, the market must perceive information to be reliable. Reliability can only be achieved with the resolution of any uncertainty and moral hazards which inhibit the production of information or diminish its integrity (Leland and Pyle, 1977, as cited by T. S. Campbell & Kracaw, 1980, 879).

Intermediaries also emerge as information producers because the production of information, the protection of confidentiality, the provision of transactions services, as well as other intermediary services, are naturally complementary activities (T. S. Campbell & Kracaw, 1980). The moral hazard problem is resolved when the intermediary has a large enough position in the market to be trustworthy. Intermediaries are profitable when they can jointly produce information and other services valued by investors. Thus, intermediaries do not emerge only when there are difficulties created by asymmetric and costly information. They can profitably emerge when they can jointly produce information as well as other products or services valued by buyers and seller (T. S. Campbell & Kracaw, 1980).

The insertion of an intermediary in the value chain can have negative impacts in certain situations. "Intermediation introduces a further step of transaction in the value chain and therewith causes additional transaction costs. So intermediation can only be advantageous if the increase of value added through intermediation overcompensates the transaction costs additionally incurred," (Rose, 1999, 46). Thus, to be viable, intermediaries must add value to transactions.

Conclusion

An intermediary seeks to reduce uncertainties for agents in markets that manifest information asymmetry and rising transaction costs. But the specific nature of these uncertainties changes with each successive generation. The intermediary functions by producing information, decreasing information asymmetry and reducing transaction costs. Thus, at any particular point in time, the function of intermediaries will depend on information asymmetry, information impactedness, the commodification of information, and transaction costs. It is clear that for an intermediary to succeed, it must add value for both the buyer and the seller. In any economy-the age of mercantilism, the Industrial Revolution, or the Digital Age-intermediaries emerge when uncertainties are present in the market. As these uncertainties change, the role of the intermediary changes as well.

With this conceptual framework laid out, further consideration can be given to the role of intermediaries in history and in present markets. Having identified the uncertainties that give rise to intermediaries in an economy, the next step is to analyze how these factors affected the role of intermediaries in historical markets, from the traditional town square market to the Industrial Revolution. What uncertainties and transaction costs in these markets gave rise to the first intermediaries? What roles did intermediaries play in these markets?




3Some definitions of information include "knowledge is power," (Stigler, 1971, 61); "the negative measure of uncertainty and an economically interesting category of goods," (Arrow, 1984, 138); and "anything that can be digitized," (C. Shapiro & Varian, 1998, 3).

4Sampler gives an example of using customer demographic data to explain the varying fungibility of information: "customer demographic data may be highly applicable to a wide variety of consumer goods firms, but information regarding the quantity of a particular product would be less fungible, i.e., it would have fewer alternative uses or be applicable in fewer situations," (Sampler, 1998, 346).

5According to Williamson, "asset specificity refers to durable investments that are undertaken in support of particular transactions, the opportunity cost of which investments is much lower in best alternative uses or by alternative users should the original transaction be prematurely terminated, and the specific identity of the parties to a transaction plainly matters in these circumstances, which is to say that continuity of the relationship is valued, whence contractual and organization safeguards arise in support of transactions of this kind, which safeguards are unneeded (would be the source of avoidable costs) for transactions of the more familiar neo-classical (nonspecific) variety," (1985, 55).