On the nature of information and knowledge and the interpretation in the economic sciences.

THE NATURE AND CONCEPTUALIZATION OF INFORMATION

During the last forty years or so, the term "information" succeeded in pervading science, technology, as well as everyday life. In this regard, "information" is not unlike the term "energy" which--in spite of its usage since the time of Aristotle--attained widespread scientific and popular use during the nineteenth century.(2) Indeed, there may exist other parallels between these two concepts and perhaps even a more intimate common ground. Above all, it seems likely that information requires a conceptual process no less torturous than the one that purified the notion of energy during the last 200 years. Although this article is primarily concerned with economists' conceptions of information and knowledge, the first part may help to illuminate these crucial notions from a more general point of view.

Energy as Information and Information through Energy

The thought that energy and information might be two facets of one and the same concept is still too remote for most scholars to contemplate. Yet, in physics, energy is defined as the potential to do work, which means it has the potential to change things (or prevent change where it would occur without a particular source of energy). Surprisingly, the major feature of information too is the potential to induce change (see the definition in the section "Information: Different Meanings and Common Ground"). On a physical level, energy and information seem to be so alike that one hardly needs the term "information"; the single expression "energy" usually suffices. Two electrons, for example, repulse each other because of the negative electrical charge of each of them, but, in this process, the electrons exchange a photon. Is it not this photon that "informs" each particle of the charge of the other? How else would those particles get information about the other's charge? How could a force that does not inform (in the most basic sense of the word) trigger such a repulsion or any other action?

On higher levels of reality, increasing differentiation between the two notions of energy and information seems to occur, but, even there, information without the presence or consumption of energy is just as impossible as is the transfer of energy without information (in the physical sense as mentioned earlier). Furthermore, the question lingers on as to whether energy without information could be the ultimate impetus for moving this world and holding everything together, be it the tiniest subatomic particle or the entire cosmos. How could all those entities be bound to each other without some flow of information between them? Even physicists admit that the four forces of nature (interactions) are negotiated by the exchange of a particle that acts as a kind of information.

Mind and Matter versus Information and Energy?

Perhaps the philosophers' long-standing dispute about the duality of mind and matter could find a resolution in the duality or even quasi-identity of information and energy. Such an idea could be traced to Schopenhauer's (1958) work The World as Will and Representation. This philosophy already emphasized the energy-enforcing as well as the information-representational aspects of our universe.

But, speculation aside, the fact is that our present ideas about "information"--and the latter's relationship to such similar expressions such as "data" "message," "knowledge," etc.--are not yet sufficiently clear. In spite of Machlup's (1980, pp. 56-58) significant endeavor, this issue will require further examination that is not likely to be completed overnight. After all, it took physicists centuries to recognize the essence of energy and its relationship to such notions as force, work, heat, and many other variables.

Assuming a holistic point of view and tracing information back to the physical level, one may conclude, as a first item, that every manifestation of active or potential energy is connected to some kind of information, just as the transmission of any kind of information requires active or potential energy. Under the admission that, on one side, the forces of nature constitute a kind of enforcing (or compelling) information and, on the other side, that no information transmission can take place without material things (i.e., potential or active energy such as light, air, books, etc.), the above proposition is difficult to refute.

Second, the physical interactions (and their corresponding particles) seem to be the most basic medium for transmitting information as well as energy--though physicists and philosophers still speculate about the possibility of superluminal or instantaneous information. Since photons and gravitons belong to these elementary particles, light and other electromagnetic waves as well as gravity must be counted to such basic information media.

Third, the ultimate source and very essence of energy still lacks scientific explanation, and something similar is true with regard to information. A recent insight of physics holds that the total energy in the cosmos is zero. Stephen Hawking (1988), one of the most reputed experts, points out that matter in this universe is regarded as positive energy because the gravitational field that causes all pieces of matter to be drawn to each other is a kind of negative energy: "two pieces of matter that are close to each other have less energy than the same two pieces a long way apart, because you have to expend energy to separate them against the gravitational force that is pulling them together" (p. 129). He also indicates that this negative energy has been demonstrated to be exactly equal in amount to the total positive energy of the universe.

This gives rise to the inference that the ultimate explanation of existence may not be found in energy but in a kind of "fluctuating nothingness" (e.g., a gigantic quantum fluctuation) that--under the possible impact of some information--is capable of splitting into positive energy (matter) and negative energy (gravitational attraction).

Fourth, in examining different levels of reality--cf. the "onion model of reality" suggested in Mattessich (1991a)--one discovers emergent properties that give rise to different kinds of information on each level. On the purely physical level, there always seems to be enforcing information, yet this information is not necessarily deterministic; occasionally it is probabilistic. If the information conveyed by a photon between two electrons compels the electrons to repulse each other, one may speak of enforcing (or compelling) information as well as deterministic information; whereas, the information "controlling" the decay of a radium atom, though no less inevitable, does not affect a uniquely determined [alpha]-particle for expulsion from the nucleus but randomly compels one among many. Climbing the next major step in this hierarchy to the level of biological reality, probabilistic information seems to become more frequent (e.g., the transmission of hereditary information), while further steps upward, on the levels of psychological and social reality, one encounters nonenforcing (voluntaristic or quasi-voluntaristic) information--the response to which is subject to the "will" of a human being or animal of more or less higher order. This nonenforcing information is the one most frequently identified with the commonplace notion of information.

Information: Different Meanings and Common Ground

A major feature of nonenforcing information (in contrast to enforcing information) is its influence on the intentions and expectations of the individual, be it a human being or an animal. A major concern of information economics, for example, is the effect which information has upon a person's expectations and the resulting decisions she or he may contemplate or enact. Most economists are even prone to regard only those data as genuine information that are capable of changing a person's expectation about a certain event. Since these expectations are best quantified by means of probabilities assigned to future events (states of the world, including nature, actions of competitors, etc.), the relationship between information signals and changing probabilities becomes crucial.

However, alongside the information for predictive purposes there exists information for contractual purposes as well as for retrospective uses (such as historical and learning purposes). Butterworth et al. (1982) suggested a total of seven even more refined categories of uses for information.

In genetic biology, however, the emphasis is on the transmission of genetic information, on the "syntactical" as well as on the "semantic" sequence of the nucleic acids, and, above all, on the consequences or "actions" that result from a specific information transfer (e. g., a genetic defect of the individual concerned). In electrical engineering, "information theory" (more appropriately called "communication theory")(3) also has a semantic overtone. But, apart from measuring the information content (in bits(4)), the concern is on the redundancy of information and on the ability to decode a string of information distorted by noise. Ecology and the systems sciences concentrate more on the cybernetic or feedback mechanism in which information plays an indispensible role. In library science, the emphasis shifts from information to knowledge, its preservation and obsolescence, as well as its manifestations in different media.

Today, many scientific disciplines use the term "information" in one way or another. The term is now commonplace within a wide academic spectrum that includes astronomy, electrical engineering, biochemistry, medicine and biology, psychology and the behavior sciences, the economic sciences, as well as the humanities. And, although there are differences in meaning between the application of this expression to each discipline, in order to understand the very nature of the concept of information, it is advisable to explore the similarities of those meanings or, at least, the interrelations (in Wittgenstein's sense of "family resemblances") between those meanings (cf. Wittgenstein, 1953; Mattessich, 1978, p. 96).

Bunge and Ardila (1987), for example, distinguish between the following seven different ways in which this term is used:

[information.sub.1] = meaning (semantic information)

[information.sub.2] = structure of genetic material (genetic "information")

[information.sub.3] = signal

[information.sub.4] = message carried by a pulse-coded signal

[information.sub.5] = quantity of information carried by a signal in a system

[information.sub.6] = knowledge

[information.sub.7] = communication of information6 (knowledge) by social

behavior (e.g., speech) involving a signal ([information.sub.3]). (p. 106)

But the confusion behind this array of terms can easily be eliminated by two steps: (1) by qualifying each term, and (2) by showing how the concepts behind those terms are interrelated.

If one defines information as "the configurative, pictorial, or conceptual representation(5) of an empirical phenomenon possessing the potential of changing the action, intention or expectation of an entity in such a way that without this information the entity would act, intend or expect differently,"(6) then the following relations and notions emerge: 1. Semantic information turns out to be information that states or

clarifies the relation between either a term or a concept and the

corresponding empirical phenomenon. This may be regarded as

useful or efficient information (see note 6) if some person receiving

it would act, intend, or expect differently (than without receiving

it) as far as her or his semantic attitude is concerned.(7) 2. Genetic information is a specific sequence of nucleic acids in the

DNA of a virus, plant, or animal which (in a sequence of further

steps) ultimately determines (or codetermines) the production of

certain enzymes, proteins, and so on that could not be produced

without this information. 3. A signal or data (datum) might be regarded as a medium possibly

carrying some information. To confuse the signal with the

information is similar to confusing the box containing a present

with the present itself. Depending on convention and other factors,

different signals may, for example, convey the same information,

or the same signal may, in different situations, convey different

information.(8) 4. A message might best be regarded as data, information, or

knowledge in the process of communication. 5. The quantity as well as the quality of an information has to be

distinguished from the information itself. To identify either

quantity or quality (including value) with information proper

would be similar to confusing a parcel's weight or value,

respectively, with the content of the parcel. 6. Most human knowledge, whether pragmatic or scientific, is ultimately

hypothetical (certainly as far as it is contained in universal

propositions). And a hypothesis is the attempt to represent some

empirical phenomenon; it is not the phenomenon itself. The relation

between information and knowledge may be compared to

that between raw material or component on one side and a larger

system containing this component on the other.

For example, the imprint on a sandy beach, which Robinson

Crusoe noticed, was a signal carrying the information that this

is a footprint. But the resulting thought that this forlorn island

must recently have been visited by another person constitutes the

knowledge induced and supported by the earlier mentioned

information.

Another illustration of the relation between information and

knowledge was offered in Mattessich (1974, p. 783; 1978, p. 231),

where the plotting of different data in a two-dimensional diagram

(cost/production) constitutes information, while the process of

connecting these data (e.g., curve fitting by statistical or

econometric means) creates knowledge.

In Mattessich (1978), scientific knowledge is characterized as law-like

statements sufficiently supported by appropriate evidence,

acceptable as (provisional) truth by a certain branch of science (p.

231). Thus, information should not be equated with knowledge.

To do so would be like mistaking the ruler with the people who

bestow the power upon him or her. Though knowledge may depend

on relevant information, the former contains a creative" element

which information lacks. Figure 1 illustrates this: a handful of data

of operating costs under varying production volumes (left-hand

side) offers nothing but information (in the form of data). Only

after the creative act of fitting a curve to this information does

one get a hypothesis (right-hand side). This hypothesis can be

expressed in a formula or sentence and becomes scientific

knowledge as soon as the experts agree that, first, enough data

have been collected and, second, that the cost curve is the best

among all feasible alternatives.

But a piece of information may itself be of a hypothetical nature

and may thus assume characteristics of knowledge similar to a

vassal who plays the ruler in his or her own backyard. Thus, it

may depend on the context whether something should be regarded

as information rather than knowledge. A related idea is expressed

by Machlup (1980, pp. 56-58). He too recognizes a clear and

meaningful difference between information and knowledge but

admits that sometimes "the contents of the information received

may be the same as the contents of what is known as a result"

(p. 57). He continues to suggest that, as regards content, all

information is knowledge while not all knowledge is information. 7. Communication is the transmission of either signal, information,

or knowledge from one person or place to another Its metaphor

is that of transportation--e.g., the shipping of the components

of a bicycle by truck from one place to another. Thereby the wheels

and other individual parts compare to information, the wrappings

of those components compare to the signals, the entire bicycle

compares to knowledge, and the box in which all this is shipped

would be the metaphor for the message.

Some further reflection may be required as to the connection between different information concepts on different reality levels. Above all, the question arises whether the natural sciences' meaning of information, which is to be empirical, can be reasonably related to the social sciences' and humanities' notion of information, which is to be conceptual. Compare genetic information, which is undeniably something physical (a DNA molecule or part of it), with economic information (e.g., the quotation of a negotiated price) which obviously is conceptual. The above-stated definition stipulated information to be "the configurative, pictorial, or conceptual representation of an empirical phenomenon...." That means it is not supposed to be the empirical phenomenon itself but its representation. How can this chasm between information as a real thing (in the natural sciences) and as some representation" (in the social sciences and humanities) be reconciled?(9)

To resolve this dilemma, one has to consider that every representation, even a conceptual one, can only be achieved by means of material things. Thus, the first point is to bear in mind that for any kind of representation (whether for scientific or everyday activity), one needs physical things, such as air and sound waves; or ink and paper; magnetized tapes and decoding machines; electromagnetic waves and television sets; and above all, neurons, electro-chemical reactions, and neural transmitters, and so on. In other words, one is always representing (i.e., approximating) reality, whether it is physical, psychic, or social, by means of physical reality. That is to say, representation is relative to the level from which one looks at it. Assume an object is represented on a television screen; such a television picture then becomes a (pictorial) representation, but somebody may describe this picture verbally, which then becomes itself the object of further representation--though, in this case, a conceptual one.

The second point is that nature itself has various means of representing itself. The cloning of a paramecium, for example, is not only a self-replication but also a self-representation--this may be particularly obvious in the case of miosis, when the chromosomes and their genes are being copied, hence also represented configuratively. And our definition of information deliberately included not only conceptual but also configurative as well as pictorial representation as possible candidates for information. Thus, conceptual representation (and the information conveyed by it) can be viewed as a natural extension of configurative and pictorial representation. Of course, conceptual representation is undoubtedly the form most important for the social level of reality. Where would we be without it?(10)

THE ECONOMICS OF INFORMATION, KNOWLEDGE, AND EDUCATION

The economics of information, knowledge, and education area consists of two major parts, "information economics" (IE) and "the economics of knowledge and education" (EKE), each of which may further be subdivided.(11) The two pioneers of IE are Jacob Marschak (e.g., 1954, 1964, 1974; also his work co-authored with Miyasawa, 1968; and, with Radner, 1972), who made decisive contributions to what he called the "economics of information," and George Stigler (1961, 1962), who received the 1982 Nobel Memorial Prize for his seminal work in the "economic theory of information" (together with his theory of public regulation). Another pioneer, Fritz Machlup (e.g., 1962, 1980, 1982, 1984), laid the groundwork and elaborated on what has become the "economics of the production and distribution of knowledge" (here addressed as the "economics of knowledge and education").

In the wake of these pioneers and their publications followed a host of other authors who greatly expanded each of these fields or branched out into new related ones. However, today the pioneering efforts of Stigler as well as Marschak have influenced the central core of standard economics to a greater extent than the work of Machlup. Although such a trend may reverse itself in the future, this article will have to consider the evolution of IE (and its fusion with agency-contract theory, as well as subsequent developments, such as the "economics of imperfect information," including the "theory of asymmetric information"), in spite of the fact that EKE might have a closer affinity to library science.

Information Economics as an Extension of Decision Theory

In the following discussion, the attempt is made to sketch, in rough strokes, the development of information economics and economics of knowledge and education in a way comprehensible to noneconomists. IE analyzes the economic consequences of, as well as the demand for, alternative information systems. As Feltham (1984) points out:

the development of models of rational choice under uncertainty by such

pioneers as von Neumann and Morgenstern ... can be viewed as the starting

point of information economics. They demonstrated that if an

individual's choice behavior satisfies a few rather basic "consistency"

axioms, then his behavior can be represented as the maximization of

his expected utility for the consequences of the actions available to him.

(pp. 181-82) Thus, IE is an extension of statistical decision theory(12) which usually begins with a finite number of strategies or possible actions to be chosen by the decision maker (e.g., alternative crops to be planted), as well as a number of alternative states of the world (e.g., different weather situations) which are beyond the control of the decision maker. To each of these combinations (of crops and weather alternatives), an estimated payoff value (e.g., a dollar-profit) is assigned. If each state of nature has a known (or estimated) probability of occurrence (determined, for example, by long-term weather forecasts), it is simple to calculate not only the expected payoff (i.e., average value) of each combination, but also of each strategy, covering all alternative states of weather. Finally, among these alternatives, the strategy with the highest expected payoff is chosen.

Information economics, the most important theorem of which was proven by Blackwell (1951, 1953),(13) introduces explicitly to this basic statistical decision model the notion of information. It asks how can each strategy (e.g., of crop planting) be improved by an information system (e.g., subscribing to a long-term weather forecasting information service), provided the probabilities of a successful forecast of each state can be estimated (e.g., from the track record of this service). Again, the expected value of each strategy, now in light of the weather forecast, can be calculated and compared with the expected value from the earlier-mentioned decision model (i.e., without benefit of a long-term weather forecasting service). The difference in each strategy is supposed to indicate the gross benefit of using the information system (i.e., subscribing to the forecasting service). The net benefit of each strategy results from deducting the information cost (i.e., subscription fee). But, most importantly, with the use of such an information system, the optimal strategy need no longer be the same as the one recommended by the simple decision model. Furthermore, similar analyses can be employed for competing information systems in order to choose the optimal system among them by determining not only the value of information but also the value of each of those systems. Thus, decision making, in the long-run, could be improved by using information which, (1) costs less than its net value, and (2) is cheaper than information supplied by another available information system.

It is not difficult to construct a simple information-economic illustration relating to library science. Assume, for example, a profit-oriented library (common in Europe not too long ago) wants to maximize profit by allocating its acquisition budget to attain an optimal combination of books from different categories (such as classical fictions, mystery stories, do-it-yourself books, popular science books, and so on). Different combinations become different strategies; varying future trends in readers' tastes become different states; and a research service, investigating and predicting changing readers' tastes, becomes the information system. In the case of a public library (without profit motive), the goal of profit maximization would have to be substituted by some kind of utility optimization which, however, would create further difficulties of measurement and conceptualization.

Information and Team Theory

Marschak's seminal (1954) paper explored the finding of what is currently called the optimal "information structure"(14) within a firm. In other words, which information and communication scheme between individual entities or persons (e.g., departments or their heads as a "team") is most desirable for attaining the goal of the entire enterprise? Or who (in a firm or other entity, perhaps even in a library) has to know what, and who has to report or communicate to whom? This analysis was later greatly elaborated on in Marschak and Radner (1972). By its very nature, this research focussed on internal relationships between persons of supposedly common interest. This stimulated much interest in circles of game and decision theory, organization theory, as well as academic accounting. Indeed, there exists considerable literature--particularly in the application of the team-theoretical aspects of information economics to accounting theory--pioneered by Butterworth (1968, 1972), Feltham (1967, 1968, 1972), Mock (1969,1971), Demski (1970, 1972/80), Feltham and Demski (1970), Demski and Feltham (1972, 1976), and others.

Information and the Market

However, external relationships were neglected by this kind of information research, and it is in the area of market information where another seminal paper, that by Stigler (1961), filled in a crucial gap. He was led, as admitted in his memoirs (cf. Stigler, 1988, pp. 79-80), to the problem of information by the obvious fact that, when shopping around long enough, one can of ten find a lower price for a (homogenous) commodity than originally encountered--this is contrary to the teachings of traditional economics of perfect competition. He also noticed that it costs time and of ten money to search for a better price, and stipulated that the major obstacle to a complete search in finding the best price are the information costs. But the latter are a type of "transaction costs" which around the same time were exposed by Coase (1960), another Nobel laureate, as crucial in impeding the workings of perfect competition according to standard economic theory.

But not even the work of a Nobel laureate is beyond criticism, and Phlips (1988, pp..26-27) discusses the limitations of Stigler's work, which later stimulated others to overhaul some of his ideas either partially or even fully. Phlips claims that, in reality, consumers are usually aware of which shops are expensive and which are inexpensive, but less so about the distribution of prices--which is opposite to the assumption of Stigler, who also fails to offer an analysis of price distribution. Furthermore, Stigler assumes that the number of price searches (and other search rules) is determined beforehand, while it would be more realistic to keep the number of searches open ended and take a certain learning process into consideration. Thus, Stigler's search rule is not optimal, while a sequential rule (after each price quotation the buyer decides whether to continue his or her search) is claimed to be optimal (cf. Rothschild, 1974).

Some publications by another Nobel laureate, Arrow (e.g., 1984a, 1984b, 1979), show the latter's long-standing interest in IE. In a way, it was Arrow who created the preconditions that tie IE to the rest of modern neo-classical theory. Other economists as well as accounting academics contributed to this area, particularly to the special problem of public information. Hirschleifer (1971) provided the original analysis of public information (under pure market conditions and other stringent assumptions), and others, like Hakansson et al. (1982), Kunkel (1982), and Ohlson (1988), extend this analysis (the latter two papers included production conditions). Ohlson (1988) also examined the social or welfare value of public information, attempting to evaluate under what conditions does public information have value at all.

If the consumer has all the competing prices at his disposal, one speaks of complete information. But, first of all, this is rarely the case in actual practice, and second, such a situation is analytically less interesting than cases of incomplete information. Thus, it is hardly surprising if in recent times one prefers to speak of the "economics of incomplete information."

A large amount of research has evolved in this area, well surveyed by Phlips (1988), Laffont (1989), and others. It ranges from an examination of information sequences (e.g., predecision versus post-decision information) to different types of auctions, price dispersions, predatory pricing, signals and "signaling theory," credit rationing, antitrust implications, different kinds of economic equilibria, contingent markets and constraining contract clauses, competition among agents, even to cheating and misinformation.

One of the most influential ideas in this area is the notion of informational asymmetry, which is characteristic for most situations of market uncertainty. This notion is best illustrated by Akerlof's (1970) widely known paper "The Market for |Lemons'," which uses the second-hand car market to analyze and demonstrate the informational advantage the dealer has over the prospective buyer (the dealer is more likely to know about the accidents and repairs of a specific second-hand car than is the buyer). This phenomenon is widespread wherever contracts are being entered into, be it in hiring a manager or other employee (where the person to be hired knows much better her or his qualifications, as well as shortcomings, than the prospective employer does) or a medical insurance contract (where the person to be insured is often much better aware of his ailments than the insurer) or many other contractual arrangements. Its major conclusion confirms the insight that the optimal policy of the seller is to abstain from revealing some information (e.g., the product's quality), if the latter cannot be readily verified. But if such verification is possible, it is economically optimal for the seller to grant a warranty to the buyer, which in turn induces the production of better qualities.

Information Economics and Agency Theory(15)

The notions of asymmetric information as well as those of moral hazard and adverse selection (both explained later) have helped to develop another subarea of economics closely related to IE, namely agency theory, which deals with employment contracts and similar contractual arrangements in which information is crucial.

The first publications systematically analyzing the problems of work and management contracts were those by Coase (1937) and Simon (1951), both of whom received, decades later, the Nobel Memorial Prize. Yet, these publications found little immediate acknowledgment, and it took approximately two decades until a more widely accepted version of the principal-agency relations evolved. In economics, it was the paper by Alchian and Demsetz (1972) and in business administration a successive paper by Jensen and Meckling (1976) which provided the actual launching basis. Shortly before, special aspects of similar contracts were analyzed in two fundamental papers by Mirrlees (1971, 1976) as well as by Spence and Zeckhauser (1971). The integration of all those and a considerable number of later research efforts led to what has become known as "agency theory." But there is an essential difference between this originally predominantly descriptive agency theory and the subsequent predominantly analytical agency theory which one may address as agency-information analysis (also called "agency-contracting theory").

The central problem of the original agency theory lies in the costs incurred by the potential goal conflict between principal and agent (e.g., monitoring of an agent's activity, profit reduction due to goal conflicts between the two parties, foregoing actions preferred by the agent in consideration of the principal's different preferences--in the last case, for example, the agency costs are borne by the agent). Closely related to this problem is the search for a Pareto-optimal contract (i.e., no party is worse off than before "contracting," but at least one party may be better off afterward) that motivates the agent but hopefully also enables risk sharing with the principal. Thereby the (accounting) information system employed plays a vital role. In this way, the agent (whose activity cannot always be monitored) shall be motivated in such a way that his or her interest coincides with that of the principal (self-enforcing contract) such that "agency costs"--i.e., the costs caused by the goal conflict between the two parties--be reduced to a minimum. In the realm of finance, agency theory tried to analyze the motivations and relations caused by certain shifts between internal and external financing in order to search for an optimal financing ratio. Such a finance theory may be more realistic than the theory of the two Nobel laureates, Modigliani and Miller (1958), which does not seem to recognize this kind of optimization problem.

However, as Butterworth and Falk (1986) have pointed out, within the predominantly descriptive approach of Jensen and Meckling (1976), it was neither possible to examine the equilibrium conditions of such a contracting model, nor certain consequences (e.g., bonding and monitoring features voluntarily accepted by the agent) which possibly might arise from this theory. Watts and Zimmerman (1978, 1979) have even attempted to incorporate political costs (e.g., connected with the lobbying for accounting standards) into the agency model. And the publication by Holthausen and Leftwich (1983) supplied early empirical tests to the agency theory.

Another phase of what has been called "the stewardship tradition" of accounting (see Mattessich 1990) arose out of the combination of descriptive agency theory on one side and information economics on the other. For many years, there was not much contact between descriptive agency theory and information economics, but, with increasing formalization of the former and the realization of the importance of information in contracting relations, both camps became aware of the need for a close cooperation, or even an amalgamation, of those research areas. For this reason, it is well justified to speak of "agency-information analysis" when referring to the analytical approach of agency theory. Its core is also to be found in the contractual relations, in risk sharing between principal and agent, as well as in improving the motivation of the latter.

Depending on the type of employment contract, management's share of the total enterprise profit (before its remuneration) might span a wide spectrum limited by two extremes: (1) on the one side we find a fixed managerial salary (under full monitoring of the manager's activity by the principal), whereby the total remaining profit goes to the principal who bears all the risk (principal is risk neutral, agent is risk averse); and (2) whereas the other extreme is found in the renting of a business by the agent such that the principal receives a fixed rent, and the agent, who bears all the risk, pockets the remaining profit (principal is risk averse, agent is risk neutral). There exist many types of contracts in between these extremes. Some lead to a Pareto-optimal profit and risk sharing between the two parties in accordance with classical marginal economic theory (principal and agent are both either risk neutral or risk averse). All these so-called "first-best solutions" are of less practical interest than the so-called "second-best solutions," because only the latter offer the means to cope with two crucial issues:

1. The problem of moral hazard. This arises from the agent's or the

principal's temptation to act in one's own interest even if the

contractual interest of the other party is thereby shortchanged.

As the principal can usually not fully monitor the agent's activity--and

since agency-information theory assumes that each party

maximizes its own utility--the agent's optimal action may not

be optimal for the principal (unless special contractual

arrangements, as recommended by this theory, are being made).

2. The problem of adverse selection. In many situations there exists

an asymmetry of information between principal and agent (of ten,

but not always, in favor of the agent). So, for example, the manager

may have an information advantage over the principal due to the

former's better or more specialized training and experience (for

further illustrations of asymmetric information see the earlier

section on "Information and the Market"). If, therefore, one party

withholds some information, which otherwise would lead the other

party to choose a contract or action less favorable to the first but

more favorable to the second party, then this adverse selection

impedes a first best solution.

A major task of agency information analysis, therefore, is to find conditions under which a Pareto or quasi-pareto optimal contract between both parties can be obtained. That is to say, one searches for an incentive and risk-sharing scheme which is optimal for both parties. Such an analysis would take care of both, the problem of moral hazard as well as that of adverse selection. Here again the aim is the efficient contract that leads to a compromise between two opposing tendencies: on one side one has to find an efficient risksharing contract between principal and agent. On the other side, it is necessary to motivate the manager sufficiently to act in the interest of the principal--indeed, research has shown that contracts which are risk efficient are inefficient as regards motivation. Through elimination of more and more restrictions on one side and further enrichments of the model on the other, many variations of this basic agency-information model have been created under the application of a good deal of mathematics.

Such agency contracts should be capable of reducing the agency costs to a minimum and enable an optimal position for both parties. Yet the capital risk need not only rest on the shoulders of principal and agent, it may also rest on those of other investors and creditors. Thus, agency-information analysis has a wide spectrum of application, particulary in economics, modern finance, and accounting theory. The contracts with creditors (e.g., bond indentures) might possibly be incorporated into a sophisticated agency-information model under consideration of market equilibrium (i.e., the prices are such that all goods can be cleared). Credit contracts may be represented as contingent claims toward the assets of the borrowing person or firm. This is the starting point of contingent claims analysis in which the price of the assets is determined by such stochastic processes as logarithmic-normal distribution. Black and Scholes (1973) have pioneered an equilibrium model for stock options under a simple capital structure of the pertinent enterprise. This work was extended and further developed by Merton (1973a, 1973b, 1974, 1976), Brennan and Schwartz (1977, 1978, 1979), Cox and Ross (1976a, 1976b), and others and was also applied to other areas.

For a survey of contingent claims analysis, we refer to Hughes (1984); for the application of agency theory to the area of finance, we refer to Barnea, Haugen, and Senbet (1985); and for the application to accounting, we refer to the following papers in Mattessich (1984); Feltham (1984); Baiman (1982); and Butterworth, Gibbins, and King (1982) who offer an appropriate overview. In each of these areas, the agency-information analysis was instrumental in clarifying a series of problems. See also the following five anthologies: Pratt and Zeckhauser (1985), Feltham et al. (1988), Bamberg and Spremann (1989), Hahn (1989), and Laffont and Moreaux (1991). For further references on agency-information analysis, see Baiman (1990), Eisenhardt (1989), and Mattessich (1991c, pp. 25-29).

Economics of Knowledge and Education

As previously mentioned, economics of knowledge and education was pioneered as well as further elaborated on by Machlup (1962, 1980, 1982, 1984). In contrast to information economics, much of which is analytical and highly mathematical, EKE is predominantly empirical and often descriptive; above all, it explores the economic aspects of the production and distribution of knowledge rather than dealing "merely" with information.

Our century has created conditions under which the long neglected economic side of knowledge creation, as well as of education, finally had to be subjected to careful analysis. Already a quarter of a century ago, it was estimated by Young and Margerison (1969) that scientific information increases a hundred times faster now than it did around the turn of the century, and that the volume of research doubles every decade.

It was Machlup (1962) who first took up the challenge; the impact was immediate and led, still in the same decade, to a great number of publications, particularly in the subarea of education economics (see the later discussion in this section). One may even speak of a new industrial (information and knowledge) revolution as Miller (1983) did when he wrote Machlup's epitaph in the foreword to Machlup and Mansfield (1983). There Miller pointed out that, in 1959 and 1960, Fritz Machlup gave a series of invited lectures on this subject at Cornell and Fordham universities.

The response to these lectures was so favorable that Machlup decided to expand the lectures into the book published in 1962. The concluding chapter of this book contains the following prophetic passage: "If employment opportunities continue to improve for high-level-knowledge-producing labor and to worsen for unskilled manual labor, the danger of increasing unemployment among the latter becomes more serious" (p. 397). This anticipated a trend that has not only continued but today constitutes one of the gravest threats to the social and economic health of the entire continent.

Machlup estimated that, in the United States, the production and distribution of knowledge amounted to some 29 percent of the adjusted gross national product in 1958. This estimate was updated by Porat (1977) to 46 percent in 1967 and is most likely much higher today. In 1971, Machlup decided to continue and update his knowledge research by planning an eight volume work (approximately one book for each chapter of the 1962 book). Its first volume, published in 1980, again deals with knowledge and knowledge production; the second volume, published in 1982, surveys the branches of learning. The information sciences and the analysis of the economic notion of human capital were originally also planned for volume 2 but, under pressure of increasing material, were later scheduled as separate volumes. Volume 3, dealing with the economics of information and human capital, was completed at the beginning of january 1983 (yet published in 1984). But, three weeks later, Machlup died of a heart attack at age eighty-one. However, during the last years of his life, in preparation for further work, Machlup, together with his collaborator Una Mansfield, invited "39 information scientists to write a total of 56 essays on their various specialties so that," as Miller says, "he [Machlup], in his role as an editor, could go to school under the experts--could |see the stir of the great Babel, and not feel the crowd"'(16) (1983, p.x).

Machlup's approach is very broad and often reaches beyond the realm of economics. It illuminates the problems of knowledge and its creation from many sides and offers a wide panorama, ranging from philosophical reflections to semantical and economic analysis. His work covers various types of knowledge, knowledge production and knowledge industries, education, research and development, communication media and information services, the relation of knowledge production to the gross national product, to occupational structure, and many other topics.

The trend that Machlup established was continued (even with regard to developing countries) by several authors among whom Porat (1977), Mandi (1981), Lamberton (e.g., 1971, 1984, 1988), Jussawalla and Lamberton (1982), Jussawalla et al. (1988) are only a few that deserve mention. Knowledge and information are put into relation with organizations, markets and their efficiencies, government policies and institutions, business planning, monopolies and monopolistic competition, and so on.

But probably the widest response came from the economics of education. If this field is regarded as a branch of EKE, it certainly has its own pioneers. H. F. Clark (1928) published over sixty years ago a paper on "The Economic Effects of Education as Shown by Statements of Economists," and such authors as T W. Schultz (e.g., 1959), Nobel laureate Tinbergen (1960), and Blaug (e.g., 1968, 1969, 1970, 1978, 1987) published--from a relative early time onward--articles, bibliographies, anthologies, and books that pioneered and elaborated on this particular field. Indeed, since the 1960s, this area generated a great deal of interest and produced an enormous amount of literature (see the annotated bibliography by Blaug [1978b] which lists and describes well over a thousand publications). More recent publications--e.g., Wagner (1982), Worswick (1985), Psacharopoulos (1987), Siegel (1988), and Cohn and Geske (1990)--have kept the trend up to date.

Information, Knowledge, And Depreciation

The entire literature of information and knowledge economics pays little attention to the issue of depreciation. It was mentioned in the earlier section on "Information: Different Meanings and Common Ground" (first paragraph) that many economists are inclined to regard as information (or valuable information) only those data that possess the potential of changing one's expectation about the occurrence of a specific event. Hence, once this function has been fulfilled, and such a change in expectation has occurred, the general value of this information might quickly drop to zero (there exists plenty of literature dealing with the problem of public information (see the previous section "Information and the Market"). This implies that information (in the IE sense) may lose its value soon after it has been "exploited" just as a loaf of bread loses its value once it is consumed. This indicates that information--whether it serves intermediate (i.e., industrial) or final (i.e., household or governmental) consumption--is like a consumable good (to be written off at once) rather than as a durable good (to be depreciated over time).

The situation is somewhat different as far as knowledge is concerned. Indeed, the reference indexes of Machlup's books show a few entries under "depreciation." Further examination shows that they refer to the depreciation of equipment or, what Machlup occasionally called, instruments of knowledge creation. Those scarce and short entries are not likely to satisfy scholars from library science, even though they may consider books and learned journals as belonging to these kinds of instruments.

By now there is little doubt that information and knowledge are commodities that, in principle, could be depreciated or written off. Even accountants would have to agree, as they regard a commodity or asset as depreciable (in the broad sense of the word) if its value declines over time or suddenly. Since hardly anyone would question the fact that books, journals, and other library equipment (such as audio- and videotapes, etc.) can convey information and knowledge, these commodities are potential candidates for depreciation, and since their value usually declines (save for such exceptional cases as rare or antique books and documents), there seems to be justification for depreciating them.

But such a simple answer may not be satisfactory, and the further question arises, "In which sense is this library material (i.e., books, etc.) itself information or knowledge, and to what extent is it merely conveying those valuable commodities?" Just as signals may be considered to be the wrappings or boxes in which information is packaged, so books and journals, etc. are the containers and storehouses of knowledge. Thus, as far as depreciation is concerned, a major distinction ought to be made--whether one depreciates some library material because of its physical deterioration (e.g., due to the acid paper used in those books), or whether the knowledge contained in this material has become obsolete. In both cases, depreciation may be justified but for different reasons.

Yet, occasionally, it may be difficult to determine when or how much a specific kind of knowledge has declined in value or has even lost all its usefulness. Sometimes a book considered out of fashion suddenly bounces back into vogue, or the knowledge contained in such a book becomes interesting from a historical perspective. And as to journals, their reference value, no less, turns of ten from "actual" to "historical," thus extending considerably the material's useful life. This obviously makes it more difficult to determine the depreciable lifetime of such journals. Whether analogue cases of knowledge depreciation in industry, as occurring in the case of patents of scientific discoveries or technical inventions (cf. Sweet, 1990), could supply guidance for libraries is questionable.

A further question arising in this connection is the need for depreciation of library materials in public and university libraries. Where private libraries are concerned, the financial statements are likely to have relevance in determining the profitability of such enterprises. In the case of public libraries, on the other side, there seems to be little effort to measure the rate of return in dollars and cents; the aim is, at best, to estimate roughly the cultural or entertainment value of books and other holdings. Therefore, in public libraries, it is probably less the accounting question of how much to depreciate gradually every year than the question of "weeding" or eliminating material from the shelves and writing it of f. As pointed out above, accountants of ten make a differentiation between gradual depreciations versus sporadic write-offs. A sudden decline in readers' interest in specific material might justify such a write-off, though this need not be directly connected with the "knowledge value" of this material. To divine the reader's interest in some library material or even the knowledge value seems to be a particularly difficult problem.

Gupta (1990), for example, suggested (for the specific case of review articles in physics) to use the frequencies of material quoted in citation indexes as a means for evaluating readers' interest in such publications. But this suggestion concerns only specific articles and not entire journals; furthermore, there is the fact that books are not included in most citation indexes. Both of these obstacles probably render Gupta's approach of little use for the librarian's depreciation problem.

Finally, another aspect, hinted at before, has to be considered: the need to distinguish between the conceptual act of dollar depreciation (which is an accounting and financial matter) and the empirical erosion of this accumulated or stored knowledge, which accounting depreciation ought to reflect but often does not. Behind the notions of both, epistemic obsolescence as well as physical

deterioration, stand empirical phenomena. Yet the accountant's depreciation in the ledger is purely conceptual (apart from her or his physical endeavor of carrying out this work, which concerns a different segment of reality) (cf. the earlier section on "Mind and Matter versus Information and Energy?").

NOTES

(1) Financial support for this paper by the Social Sciences and Humanities Research Council of Canada is gratefully acknowledged. (2) "In spite of its universality, the general notion of energy as a basic concept of science is a relatively recent result of a long and intricate conceptual process" (Jammer, 1967, p. 511). (3) "Information theory" assumes that within a specific language there exists a probability distribution over the population of signals. Hence each signal (within a message) will be chosen with a certain probability [[p.sub.i]). The |expected' amount of information (measured in "bits"--see note 4) in the message is then:

H=-[[summation of].sub.i=1] [p.sub.i] [log.sub.2] [p.sub.i] Since each signal is weighted by its chance of occurring, one obtains an average measure of the uncertainty in the total message. As van Gigch (1991, pp. 191-93) points out, H is defined as the degree of uncertainty, or the amount of "entropy" that exists in a specific situation. Alternatively, H can be interpreted as the amount of information, required to remove the prevailing uncertainty. (4) A bit (abbreviation for "binary digit") is the unit of binary information (e.g., as used in digital computers), expressing the information content of a message or even an entire system. Popularly expressed, one bit is the answer to a single yes or no question. Sagan (1977, pp. 23-26), points out, for example, that the Viking Landers that landed on Mars in 1976 had programs of a few million bits, which is somewhat more than the DNA (deoxyribonucleic acid) containing the entire genetic material of a bacterium. (5) It should be noted that practically any representation (even the self-representation in genetic cloning, as the slight differences between two identical twins shows), is an approximation. (6) This definition might have to be supplemented by some related notions: the "usefulness" or "efficiency" of information which could be defined as the degree to which the action, intention or expectation (due to the information) is changed, and the "value of information" as the upper limit a rational person should be willing to pay for an information in a specific situation (for further reference to the "value of information," see section in text entitled "Information Economics as an Extension of Decision Theory"). Obviously, both of these supplementary definitions involve difficulties, particularly problems of measurement and further conceptualization.

van Gigch (1991, p. 190) too emphasizes the potential action or "counter action" (e.g., to counter "entropic" or disintegrating tendencies) which information entails. (7) This definition would have to be modified if semantics is meant to include "model theory" (i.e., the semantics of mathematics and logic) which only deals with the relationships between concepts without regard to empirical phenomena. (8) A similar attitude is taken by Wilson who assumes data to be neutral and defines "information to be data plus the meaning attributed to it" (1991, p. 89). (9) However, the fact that information in the social sciences is frequently of a conceptual nature, does not imply that social phenomena are conceptual, they are just as empirical as physical phenomena (cf. Mattessich, (1991a). (10) One of the most important systematic, nonverbal conceptual representations of prehistoric times (in this case, from ca. 8000 BC to 3000 BC) grew out of a configurative representation for the accounting of economic transactions (see Schmandt-Besserat, e.g., 1978, 1980, 1992, and Mattessich, e.g., 1987/90, 1991b). Cf. for example:

"Token accounting (by means of clay tokens or figurines) as well as cuneiform writing, and hieroglyphics, offer many examples of various steps by which morphological tokens (i.e., those with similarity to its referent) and pictographs (both of which seem to "show") developed into abstract tokens and ideographs (both of which seem "to say")....Thus the morphological tokens not only describe structures, they themselves are structures" (Mattessich, 1987, pp. 88-89). (11) Neither this entire area nor the present paper include the vast field of Management Information Systems and related disciplines. For literature that might be relevant to the current topic with regard to methodological issues, see Olle et al. (1988); with regard to quantifying the financial benefits of information, see Kleijnen (1980); and with regard to more philosophical reflections, see Wand and Weber (1990). (12) This is confirmed by Arrow (1984a) who points out in his Preface that "statistical method was an example for the acquisition of information," but warns that it is difficult to formulate a general theory of information because different kinds of information have so far "no common unit." This book also contains an excellent (though dated) survey article: Arrow (1984b), which is based on a presentation to businessmen, and thus relatively easy to comprehend, and offers insight into many aspects and implications of information economics. (13) Blackwell regards one information system as more informative than another--precisely speaking, it should be "at least as informative"--as another, if it is never less valuable (precisely, if there exists no other single-person decision situation in which it is less valuable). Cf. Feltham (1984, p. 182). (14) Phlips distinguishes the following three elements contained in the information structure: "the set of possible states of the world, the set of possible signals, and the probability that a signal is observed, given that state prevails" (1988, pp. 1-2). (15) The section entitled "Information Economics and Agency Theory" is partly based on Mattessich (1990, pp. 13-16). (16) The collection of these articles, edited by Machlup and Mansfield (1983), was published posthumously to Machlup's death. The present author had the honor of contributing three papers to this anthology--see Mattessich (1983a, 1983b, 1983c).

REFERENCES

Akerlof, G. (1970). The market for "lemons": Quality uncertainty and the market mechanism. The Quarterly Journal of Economics, 84(3), 488-500. Alchian, A., & Demsetz, H. (1972). Production, information costs, and economic organization. The American Economic Review, 62(5), 777-795. Argote, L.; Beckman, S. L.; & Epple, D. (1990). The persistence and transfer of learning in industrial settings. Management Science, 36(2), 140-154. Arrow, K. J. (1979). The economics of information. In M. L. Dertouzos & J. Moses (Eds.), The computer age: A twenty year view (pp. 306-317). Cambridge, MA: MIT Press. Arrow, K. J. (1984a). The economics of information. Collected papers of Kenneth J. Arrow, vol. 4. Cambridge, MA: The Belknap Press of Harvard University Press. Arrow, K. J. (1984b). Information and economic behavior. A presentation to the Federation of Swedish Industries, Stockholm, 1973 (reprinted in Arrow, 1984a, pp. 136-152). Baiman, S. (1982). Agency research in managerial accounting: A survey. Journal of Accounting Literature, 1, 154-213 (reprinted in Mattessich, 1984, pp. 251-294). Baiman, S. (1990). Agency research in managerial accounting: A second look. Accounting, Organizations and Society, 15(4), 341-371 (reprinted in Mattessich, 1991c, pp. 221-256). Bamberg, G., & Spremann, K. (Eds.). (1989). Agency theory, information and incentives. New York: Springer-Verlag. Barnea, A.; Haugen, R. A.; & Senbet, L. W. (1985). Agency problems and financial contracting. Englewood Cliffs, NJ: Prentice-Hall. Black, F., & Scholes, M. (1973). The pricing of options and corporate liabilities. Journal of Political Economy, 81(3), 637-654. Blackwell, D. (1951). Comparison of experiments. Proceeding of the second Berkeley symposium on mathematical statistics and probability theory (pp. 93-102). Berkeley, CA: University of California Press. Blackwell, D. (1953). Equivalent comparisons of experiments. Annals of Mathematical Statistics, 24, 267-272. Blaug, M. (1978). Economics of education: A selected annotated bibliography. In G. Chandler (Ed.), International series in library and information science, 3d ed. Oxford, England: Pergamon Press. Blaug, M. (Ed.). (1968). Economics of education 1: Selected readings. Harmondsworth, Middlesex, England: Penguin Books Ltd. Blaug, M. (Ed.). (1969). Economics of education 2: Selected readings. Harmondsworth, Middlesex, England: Penguin Books Ltd. Blaug, M. (1970). An introduction to the economics of education. Harmondsworth, Middlesex, England: Penguin Books Ltd. Blaug, M. (1987). The economics of education and the education of an economist. New York: New York University Press. Brennan, M. J., & Schwartz, E. S. (1977). Convertible bonds: Valuation and optimal strategies for call and conversion. The Journal of Finance, 32(5), 1699-1715. Brennan, M. J., & Schwartz. E. S. (1978). Finite difference method and jump processes arising in the pricing of contingent claims: A synthesis. Journal of Financial and Quantitative Analysis, 13(3), 461-474. Brennan, M. J., & Schwartz, E. S. (1979). A continuous time approach to the pricing of bonds. Journal of Banking and Finance, 3(2), 133-155. Bunge, M., & Ardila, R. (1987). Philosophy of psychology. New York: Springer-Verlag. Butterworth, J. E. (1968). Accounting systems and management decision: An analysis of the role of information in the management decision process. Doctoral dissertation, University of California, Berkeley. Butterworth, J. E. (1972). The accounting system as an information function. Journal of Accounting Research, 10(1), 1-27. Butterworth, J. E.; Gibbins, M.; & King, R. D. (1982). The structure of accounting theory: Some basic conceptual and methodological issues. In S. Basu & J. A. Milburn (Eds.), Research to support standard setting in financial accounting: A Canadian perspective (Proceedings of the 1981 Clarkson Gordon Foundation Research Symposium held at Dalhousie University, Halifax, Nova Scotia, May 20-21, 1981) (pp. 1-94). Toronto, Ontario: The Clarkson Gordon Foundation (reprinted in Mattessich, 1984, pp. 209-250). Butterworth, J. E., & Falk, H. (1986). Financial reporting--theory and application to the oil and gas industry in Canada. Hamilton, Ontario: Society of Management Accountants of Canada. Clark, H. F. (1928). The economic effects of education: As shown by statements of economists. Bulletin of the School of Education, Indiana University, 4(5), 1-39. Coase, R. H. (1937). The nature of the firm. Economica, 4(16), 386-405. Coase, R. H. (1960). The problem of social cost. Journal of Law & Economics, 3(October), 1-44. Cohn, E., & Geske, T G. (1990). The economics of education, 3d ed. Oxford, England: Pergamon Press. Cox, J. C., & Ross, S. A. (1976a). A survey of some new results in financial option pricing theory. Journal of Finance, 31(2), 383-402. Cox, J. C., & Ross, S. A. (1976b). The valuation of options for alternative stochastic processes. Journal of Financial Economics, 3(1/2), 145-166. Demski, J. S. (1970). Some decomposition results for information evaluation. journal of Accounting Research, 8(2), 178-198. Demski, J. S. (1972/80). Information analysis, 2d ed. Reading, MA: Addison-Wesley Publishing Co. Demski, J. S., & Feltham, G. A. (1972). Forecast evaluation. Accounting Review, 47(3), 533-548. Demski, J. S., & Feltham, G. A. (1976). Cost determination: A conceptual approach. Ames, IA: Iowa State University Press. Edwards, P. (Ed.). (1967). The encyclopedia of philosophy, 8 vols. New York: Macmillan Company and The Free Press. Eisenhardt, K. (1989). Agency theory: An assessment and review. Academy of Management Review, 14(1), 57-74. Feltham, G. A. (1967). A theoretical framework for evaluating changes in accounting information for managerial decisions. Doctoral dissertation, University of California, Berkeley. Feltham, G. A. (1968). The value of information. Accounting Review, 43(4), 684-696. Feltham, G. A. (1972). Information evaluation: Studies in accounting research, No. 5. Sarasota, FL: American Accounting Association. Feltham, G. A. (1984). Financial accounting research: Contributions of information economics and agency theory. In R. Mattessich (Ed.), Modern accounting research: History, survey, and guide (pp. 179-207). Vancouver, British Columbia: The Canadian Certified General Accountants' Research Foundation. Feltham, G. A.; Amershi, A. H.; & Ziemba, W. T. (Eds.). (1988). Economic analysis of information and contracts: Essays in honor of John E. Butterworth. Boston, MA: Kluwer Academic Publishers. Feltham, G. A., & Demski, J. S. (1970). The use of models in information evaluation. Accounting Review, 45(4), 623-640. Gupta, U. (1990). Obsolescence of physics literature: Exponential decrease of the density of citations to Physical Review articles with age. Journal of the American Society for Information Science, 41(4), 282-287. Hahn, F. (Ed.). (1989). The economics of missing markets, information, and games. Oxford, England: Clarendon Press. Hakansson, N. H.; Kunkel, J. G.; & Ohlson, J. A. (1982). Sufficient and necessary conditions for information to have social value in pure exchange. Journal of Finance, 37(5), 1169-1181. Hawking, S. W. (1988). A brief history of time: From the big bang to black holes. Toronto, Ontario: Bantam Books. Hirshleifer, J. (1971). The private and social value of information and the reward to inventive activity. American Economic Review, 61(4), 561-574. Holthausen, R. W, & Leftwich, R. W (1983). The economic consequences of accounting choice: Implications of costly contracting and monitoring. journal of Accounting and Economics, 5(2), 77-117. Hughes, J. S. (1984). A contract perspective on accounting valuation: Studies in accounting research, No. 20. Sarasota, FL: American Accounting Association. Jackson, M. C.; Mansell, G. J.; Flood, R. L.; Blackham, R. B.; & Probert, S. V. E. (Eds.). (1991). Systems thinking in Europe. New York: Plenum Press. Jammer, M. (1967). Energy. In P. Edwards (Ed.), The encyclopedia of philosophy (vol. 2, pp. 511-517). New York: Macmillan Company and The Free Press. Jensen, M. C., & Meckling, W. H. (1976). Theory of the firm: Managerial behavior, agency costs and ownership structure. Journal of Financial Economics, 3(4), 305-360. Jussawalla, M., & Lamberton, D. M. (Eds.). (1982). Communication economics and development. Honolulu, HI: East-West Center. Jussawalla, M.; Lamberton, D. M.; & Karunaratne, N. D. (Eds.). (1988). The cost of thinking: Information economies in ten Pacific countries. Norwood, NJ: Ablex Publishing Corporation. Kleijnen, J. P. C. (1980). Computers and Profits: Quantifying financial benefits of information. Reading, MA: Addison-Wesley Publishing Co. Kunkel, J. G. (1982). Sufficient conditions for public information to have social value in a production and exchange economy. Journal of Finance, 37(4), 1005-1013. Laffont, J. J. (1989). The economics of uncertainty and information (J. P. Bonin & H. Bonin, trans.). Cambridge, MA: MIT Press. Laffont, J. J., & Moreaux, M. (Eds.). (1991). Dynamics, incomplete information and industrial economics (F. Laisney, trans.). Oxford, England: B. Blackwell. Lamberton, D. M. (Ed.). (1971). Economics of information and knowledge: Selected Readings. Harmondsworth, England: Penguin Books. Lamberton, D. M. (1984). Information age economics. In A. Goldsworthy (Ed.), Technological change: Impact of information technology (pp. 3-12). Canberra, Australia: National Information Technology Committee. Lamberton D. M. (1988). Preface: The cost of thinking. In M. Jussawalla, D. M. Lamberton, & N. D. Karunaratne (Eds.), The cost of thinking: Information economies of ten Pacific countries (pp. 1-5). Norwood, NJ: Ablex Publishing Corp. Lee, T. A. (Ed.). (1990). The closure of the accounting profession (vol. 1, pp. 246-266). New York: Garland Publishing, Inc. Machlup, F. (1962). The production and distribution of knowledge in the United States. Princeton, NJ: Princeton University Press. Machlup, F. (1980). Knowledge: Its creation, distribution, and economic significance (vol. 1, Knowledge and knowledge production). Princeton, NJ: Princeton University Press. Machlup, F. (1982). Knowledge, its creation, distribution, and economic significance (vol. 2, The branches of learning). Princeton, NJ: Princeton University Press. Machlup, F. (1984). Knowledge, its creation, distribution, and economic significance (vol. 3, The economics of information and human capital). Princeton, NJ: Princeton University Press. Machlup, F., & Mansfield, U. (Eds.). (1983). The study of information: Interdisciplinary messages. New York: John Wiley & Sons. Mandi, P. (1981). Education and national growth in developing countries. Wellingborough, England: Collets. Marschak, J. (1954). Towards an economic theory of organization and information. In R. M. Thrall, C. H. Coombs, & R. L. Davis (Eds.), Decision processes (pp. 187-220). New York: John Wiley & Sons, Inc. Marschak, J. (1964). Problems in information economics. In C. P. Bonini, R. K. Jaedicke, & H. M. Wagner (Eds.), Management controls: New directions in basic research (pp. 38-74). New York: McGraw-Hill Book Co. Marschak, J. (1974). Economic information, decision, and prediction: Selected essays (vols. 1 & 2, Theory and decision library: An international series in the philosophy and methodology of the social and behavioral sciences). Dordrecht, Netherlands: D. Reidel Publishing Co. Marschak, J., & Miyasawa, K. (1968). Economic comparability of information systems. International Economic Review, 9(2), 137-174. Marschak, J., & Radner, R. (1972). Economic theory of teams. New Haven, CT: Yale University Press. Mattessich, R. (1974). Informations- und Erkenntnisokonomik: Treffpunkt von Philosophie und Wirtschaftswissenschaft. Zeitschrift fur betriebswirtschaftliche Forschung, 26(12), 777-784. Mattessich, R. (1978). Instrumental reasoning and systems methodology: An epistemology of the applied and social sciences (vol. 15, Theory and decision library: An international series in the philosophy and methodology of the social and behavioral sciences). Dordrecht, Netherlands: D. Reidel Publishing Co. Mattessich, R. (1983a). Cybernetics and systems theory: A search for identity. In F. Machlup & U. Mansfield (Eds.), The study of information: Interdisciplinary messages (pp. 445-452). New York: John Wiley & Sons. Mattessich, R. (1983b). Systems theory and information science. In F. Machlup & U. Mansfield (Eds.), The study of information: Interdisciplinary messages (pp. 555-559). New York: John Wiley & Sons. Mattessich, R. (1983c). Information economics, team theory and agency theory. In F. Machlup & U. Mansfield (Eds.), The study of information: Interdisciplinary messages (pp. 625-630). New York: John Wiley & Sons. Mattessich, R. (Ed.). (1984). Modern accounting research: History, survey, and guide. Vancouver, British Columbia: The Canadian Certified General Accountants' Research Foundation. Mattessich, R. (1987). Prehistoric accounting and the problem of representation: On recent archeological evidence of the Middle-East from 8000 B.C. to 3000 B.C. Accounting Historians Journal, 14(2), 71-91 (reprinted in Lee, (1990). Mattessich, R. (1990). Epistemological aspects of accounting. Keiri kenkyu (Accounting research, Tokyo), (34), 3-30. Mattessich, R. (1991a). Social reality and the measurement of its phenomena. Advances in Accounting, 9,3-17. Mattessich, R. (1991b). Counting, accounting, and the input-output principle: Recent archeological evidence revising our view on the evolution of early record keeping. In O. F. Graves (Ed.), The costing heritage: Studies in honor of S. Paul Garner (pp. 25-49). Harrisonburg, VA: The Academy of Accounting Historians. Mattessich, R. (Ed.). (1991c). Accounting research in the 1980s and its future relevance (Research Monograph No. 7). Vancouver, British Columbia: The Canadian Certified General Accountants' Research Foundation. Merton, R. C. (1973a). Theory of rational option pricing. Bell Journal of Economics, 4(1), 141-183. Merton, R. C. (1973b). An intertemporal capital asset pricing model. Econometrica, 41(5), 867-887. Merton, R. C. (1974). On the pricing of corporate debt: The risk structure of interest rates. Journal of Finance, 29(2), 449-470. Merton, R. C. (1976). Option pricing when underlying stock returns are discontinuous. Journal of Financial Economics, 3(1/2), 125-144. Miller, G. A. (1983). Foreword. In F. Machlup & U. Mansfield (Eds.), The study of information: Interdisciplinary messages (pp. ix-xi). New York: John Wiley & Sons. Mirrlees, J. A. (1971). An exploration in the theory of optimum income taxation. Review of Economic Studies, 38(2), 175-201. Mirrlees, J. A. (1976). The optimal structure of incentives and authority within an organization. Bell Journal of Economics, 7(l), 105-131. Mock, T J. (1969). Comparative values of information structures. Journal of Accounting Research, (supplement), 124-159. Mock, T J. (1971). Concepts of information value and accounting. Accounting Review, 46(4), 765-778. Modigliani, F., & Miller, M. H. (1958). The cost of capital, corporation finance and the theory of investment. American Economic Review, 48(3), 261-297. Ohlson, J. A. (1988). The social value of public information in production economies. In G. A. Feltham, A. H. Amershi, & W. T. Ziemba (Eds.), Economic analysis of information and contracts: Essays in honor of John E. Butterworth (pp. 95-119). Boston, MA: Kluwer Academic Publishers. Olle, T W; Hagelstein, J.; Macdonald, I. G.; Rolland, C.; Sol, H. G.; Van Assche, F. J. M.; & Verrijn-Stuart, A. A. (1988). Information systems methodologies: A framework for understanding. Wokingham, England: Addison-Wesley Publishing Co. Phlips, L. (1988). The economics of imperfect information. Cambridge, England: Cambridge University Press. Porat, M. U., with Rubin, M. R. (1977), The information economy, vols. 1-8. Washington, DC: U.S. Department of Commerce, Office of Telecommunications. Pratt, J. W, & Zeckhauser, R. J. (Eds.). (1985). Principals and agents: The structure of business. Boston, MA: Harvard Business School Press. Psacharopoulos, G. (Ed.). (1987). Economics of education: Research and studies. Oxford, England: Pergamon Press. Rothschild, M. (1974). Searching for the lowest price when the distribution of prices is unknown. Journal of Political Economy, 82(4), 689-711. Rubin, M. R. (1977). User's guide to the complete database (The information economy, vol. 9). Washington, DC: U.S. Department of Commerce, Office of Telecommunications. Sagan, C. (1978). The dragons of Eden: Speculations on the evolution of human intelligence. New York: Ballantine Books. Schmandt-Besserat, D. (1978). The earliest precursor of writing. Scientific American, 238(6), 50-59. Schmandt-Besserat, D. (1980). The envelopes that bear the first writing. Technology and Culture, 21(3), 357-385. Schmandt-Besserat, D. (1992). Before writing, vols. 1 & 2. Austin, TX: University of Texas Press. Schopenhauer, A. (1958). The world as will and representation (E. F. J. Payne, trans.). Indian Hills, CO: Falcon's Wing Press. Schultz, T. W. (1959). Investment in man: An economist's view. Social Service Review, 33(2), 109-117. Siegel, P. M. (1988). Education and economic growth: A legislator's guide. Denver, CO: National Conference of State Legislatures. Simon, H. A. (1951). A formal theory of the employment relationship. Econometrica, 19(3), 293-305. Spence, M., & Zeckhauser, R. (1971). Insurance, information and individual action. American Economic Review, 61(2), 380-387. Stigler, G. J. (1961). The economics of information. Journal of Political Economy, 69(3), 213-225. Stigler, G. J. (1962). Information in the labor market. Journal of Political Economy, 70(5), 94-105 (part 2: supplement). Stigler, G. J. (1988). Memoirs of an unregulated economist. New York: Basic Books. Sweet, P. J. (1990). Technology need not be patented to be depreciated, when its life is limited. Taxation for Accountants, 45(2), 110-112. Tinbergen, J. (1960). The place of education in the economy. Giornale degli economisti e annali di economia, 19(1/2), 3-12. van Gigch, J. P. (1991). System design modeling and metamodeling. New York: Plenum Press. von Neumann, J., & Morgenstern, O. (1944). Theory of games and economic behavior. Princeton, NJ: Princeton University Press. Wagner, L. (1982). The economics of educational media. London, England: Macmillan. Wand, Y., & Weber, R. (1990). Toward a theory of the deep structure of information systems. In J. I. DeGros, M. Alavi, & H. J. Oppelland (Eds.), Proceedings of the eleventh international conference on information systems (pp. 61-71). Copenhagen: Society for Information Management, TIMS, and IFIP. Watts, R. L., & Zimmerman, J. L. (1978). Towards a positive theory of the determination of accounting standards. Accounting Review, 53(l), 112-134 (reprinted in Mattessich, 1984, pp. 81-102). Watts, R. L., & Zimmerman, J. L. (1979). The demand for and supply of accounting theories: The market for excuses. Accounting Review, 54(2), 273-305 (reprinted in Mattessich, 1984, pp. 103-129). Wilson, B. (1991). Information management. In M. C. Jackson, G. J. Mansell, R. L. Flood, R. B. Blackham, & S. V. E. Probert (Eds.), Systems thinking in Europe (pp. 89-97). New York: Plenum Press. Wittgenstein, L. (1953). Philosophical investigations (G. E. M. Anscombe, trans.). Oxford, England: Blackwell. Worswick, G. D. N. (1985). Education and economic performance. Aldershot, Hants, England: Gower. Young, J. Z., & Margerison, T. (Eds.). (1969). From molecule to man: The explosion of science. New York: Crown Publishers, Inc.

Richard Mattessich, Commerce and Business Administration, University of British Columbia, Vancouver, BC Canada V6T 1Z2 LIBRARY TRENDS, Vol. 41, No. 4, Spring 1993, pp. 567-93 [Copyright] 1993 The Board of Trustees, University of Illinois