Idate.fr

Intangible Economy and Financial Markets
Charles GOLDFINGER
■ Intangible Economy
That the economy is undergoing a far-reaching, rapid and ubiquitous change is a largely controvertible statement. What is considerably morecontroversial is the nature of change. Knowledge Economy, DigitalEconomy, Information Society, Experience Economy, names for the neweconomy proliferate to the point of becoming ubiquitous buzzwords. Yet,can we say that we really understand today’s economy? Do we agree on itsrationale and development path? The answer to those questions is clearlyno. Economists and statisticians, whose role is to explain the workings ofthe economy and to provide performance and value metrics, are perplexedand bewildered. Despite data sophistication and availability, substantivedeficiencies, concerning such key variables as productivity, foreign trade,investment and financial accounting measures, remain.
The need for a new conceptual framework for the modern economy remains paramount. Such a framework should build upon the contributionsof the service and information economy approaches but should be broaderto encompass other significant trends such as the financial marketsexplosion and development of entertainment industry.
I would like to suggest an alternative framework, based on a single defining trend: the shift from tangible to intangible. The economic landscapeof the present and future is no longer shaped by physical flows of materialgoods and products but by ethereal streams of data, images and symbols.
The well-known three stages theory of Colin or three waves vision of Alvin COMMUNICATIONS & STRATEGIES, no. 40, 4th quarter 2000, p. 59.
Toffler can thus be reformulated (CLARK, 1985) . At the core of theagricultural economy, there was a relationship between man, nature andnatural products. The core relationship of the industrial economy wasbetween man, machine and machine-created artificial objects. Theintangible economy is structured around relationships between man andideas and symbols. The source of economic value and wealth is no longerthe production of material goods but the creation and manipulation ofintangible content. We live in the intangible economy. The shift to the ethereal is general and long lasting. It affects all sectors and all aspects of economic life. According to Danny Quah, professor atLondon School of Economics, we live in an "increasingly weightlesseconomy", where greater and greater share of GDP resides in "economiccommodities that have little or no physical manifestations" (QUAH, 1997). Astriking example of this shift is the increase of value per unit of weight asshown in the table below.
Unit Price
(USD per lbs)
Source: G. Colvin. Fortune Magazine The intangible economy is too often seen as a purely technology-driven phenomenon. The new economy is thus seen as the triumph of bits overa t o m s (NEGROPONTE, 1995) . Some, such a Canadian consultant,D o n Ta p s c o t t, go as far to propose a notion of the Digital Economy(TAPSCOTT, 1996). This is a dangerous oversimplification. I do not seek todeny the key contribution of In-formation Technology (which we will discussfurther below) to the advent of the intan-gible economy but it is important toavoid a fallacy that the intangible economy is tech-nology-determined.
While the technological trend toward digitalisation is unmistakable, itseconomic and business impact remains unclear and the range of potentialoutcomes is wide open. Moreover, the development of the intangibleeconomy owes at least as much to basic trends in consumer behaviour andin business environment. Shift toward higher relative demand for leisure,information and knowledge is a strong and long-lasting trend in consumer behavior: for instance, the share of services in household con-sumption inFrance has increased from 42% in 1970 to 51% in 1990 (GADREY, 1992).
Business innova-tions such as brand-driven competition and unbundling ofcomputer hardware and soft-ware have been a major dematerialisationfactor. The intangible economy is non-deterministic and transcends theopposition between bits and atoms the same way that quantum physicstranscends the opposition between particles and waves. Three dimensions of intangible economy
The intangible economy is all around us. Yet, we have great difficulty to apprehend it: by definition, intangible phenomena are elusive. Not limitedby physical constraints, they defy attempts to fit them into standardeconomic categories and taxonomies. To understand the intangible economy, we propose to approach it from three different perspectives (GOLDFINGER, 1994): - Demand perspective: Intangible artefacts: final output for consumption.
- Supply perspective: Intangible assets, used by firms to establish andmaintain their competitive position and survival. They include: the brand,the intellectual prop-erty, the human capital, Research and developmentinformation and know-how.
- Economic system perspective: Logic of dematerialisation: aninterrelated set of trends and forces that affect all economic activities,changing the nature of eco-nomic transactions and market structures. Intangible artefacts include different forms of information and communication, high and low culture, audio-visual media, entertainmentand leisure, without forgetting finance, the ultimate intangible. All artefacts are joint products, combining intangible content with physical support: song with a magnetic tape for an audiocassette; historyand a building site for a classical monument. Traditionally, content andsupport were tightly linked, making them either unique or reproducible on asmall-scale only. The development of technologies of stor-age and contentreplication has loosened the links: the same content can now be easily andcheaply replicated and associated with various physical supports. Like adragon in a fable, artefacts with an identical content appear in variousdisguises: a song can be sung live, pressed on a CD or shown as a video- clip. A payment can be made in cash, by cheque, via card or wire transfer.
Not only is the cost of replication very low, and get-ting lower, but thereplication devices are widely available. The dissociation of content andsupport leads to the proliferation of intangible artefacts in two ways. First, itlifts capacity constraints limiting consumption of intangibles. Previously, atheatrical show or a sports game could be only watched by those who couldphysically attend the theatre or the stadium. Today, television can multiplythe number of spectators ad infinitum. One could argue that a stadiumattendance and a TV watching are two different arte-facts, with differentconsumption and pricing characteristics. That is precisely the sec-ondvector of proliferation: the same content provides a source for a family ofartefacts: a book can be offered as a hardcover, as a paperback, as a CD-ROM or on-line. The consumption of intangible artefacts displays specific and - it is joint (always consumed with other products, tangibles orintangibles);- iIt is non-destructive: the same artefact can be consumed repetitivelyeither by a same consumer or by a different one;- it is non-subtractive (or non-rival): one’s consumption does not reduceanyone else’s consumption. In other terms, the opportunity cost ofsharing is zero.
Intangibles such as information are often presented as a "public good," such as fresh air or national defense, whose consumption cannot be limitedto a single consumer and therefore is inherently collective (OLSON, 1973).
We prefer to use the term of a "shared good." In effect, although intangibleartefacts are often produced for the use of a specific consumer, thisexclusivity cannot be durably maintained. Sharing can be sequential orsimultane-ous. However, simultaneity in time does not mean simultaneity inspace: it is possible to consume the same artefact in several locations, asshown by television or on-line net-works. Intangible artefacts create theirown space-time that lifts the constraints of geography. On the other hand, sharing does not signify homogeneity. The same artefact can be consumed by very different groups, as anybody whoattended a football match can testify. The pervasiveness of sharing creates extensive externalities. One’s willingness to con-sume and to pay is strongly affected by consumption or non-consumption of others. Moreover, it is often very onerous for the ownerof artefact to limit its consumption and exclude those who want to consumewithout paying. Thus the traditional equality "purchase equalsconsumption," which is the cornerstone of consumer behavior ap-proach inthe market economy, is no longer universal. For intangible artefacts,purchase does not equal consumption (how many people read all the booksthey buy?) and consumption does not imply purchase: in newspapers or inbroadcast television, the number of "free riders" routinely exceeds that ofpaying consumers by a factor of three or four.
Sharing affects not only consumption but also production. Many intangible artefacts are produced through interaction between consumersand producers. Consumers not only often provide elements of content butthey create their own combination of content as well as a content-supportassociation. Facilitated by the low cost and the availability of replicationtechnology and the content-support dissociation, such interaction becomesthe prevailing mode of creating and consuming intangible artefacts.
The widespread sharing changes the concept of property, as the acquisition of an arte-fact does not preclude its ownership by others.
Hence, the emergence of the concept of intellectual property, whose valueis created by sharing. Intellectual property rules ap-plies to scientific andtechnological innovations, to artistic creation but also to the management ofbrands and other intangible assets.
Economic characteristics of intangible artefacts render conventional pricing and trans-action mechanisms largely inadequate to capture theireconomic value. The two stan-dard approaches are difficult to apply.
Production costs cannot be used as a pricing guide, as there is noproportionality between inputs and the output. Mass consumption does notimply mass production. Best-selling books, records or movies are createdby small creative teams and their revenues are not related to their costs.
Economies of scale for intangible artefacts are determined by consumptionnot by production. Yet, the other approach based on the willingness to payalso has serious pitfalls, given the ease of rep-lication and sharing andassociated externalities. Another problem, which particularly affectsinformational artefacts, is what Joseph Stiglitz called the "infinite regress": itis impossible to determine whether it is worthwhile to obtain a given pieceof information without having this information (STIGLITZ, 1985). Traditionally, the pricing of intangibles was a function of convenience and was based on the support rather than on the content. Thus, the price of a book was determined by its thickness and the printing quality. Theadvance of dissociation created opportunities for unbundling: content cannow be priced separately from the support. Price discrimination, based onthe estimated value of content, becomes more common. On-line services,for instance, differentiate between standard and premium services, whichare sold at higher prices. Yet, the bundling has its advantages. It facilitatespricing of composite artefacts, comprising several types of content(multimedia software or amusement parks). It also allows cross-subsidies.
In financial services for instance, equity research is bundled into brokeragecommissions. Thus, the range of intangibles pricing schemes is gettingbroader and more complex. Furthermore, depending on the supplier-consumer relationship, different pricing arrangements can apply toapparently similar artefacts. Computer software can be sold as a stand-alone product or it can bundled with hardware or be distributed as ashareware or freeware over a network (DYSON, 1992; VARIAN, 1994, 1995,1 9 9 9 ). Internet provides a fascinating laboratory of various approaches topricing through various combinations of selling, sharing and giving away.
The debate about the respective merits of those approaches is lively. Someargue that the development of technologies such as metering, whichmeasure the detailed use makes feasible fully variable usage-driven pricing(COX, 1994). Others plead in favour of a fixed access charge, independentof the actual use. Still another group considers that the ease of replicationmakes content practically free and therefore the only feasible approach is tocharge for ancillary services (DYSON, 1995) or for "eyeballs" via advertisingand other forms of third party pricing. These approaches should be seen ascomplementary rather than mutually exclusive. As pricing of intangibles focuses on content, it highlights its inherent volatility of valuation. Although physical goods show variation in price, theamplitude of changes is considerably larger for intangible artefacts. Theirvalue is highly time-sensitive and can change dramatically: the samefinancial information (about a company merger, for instance) can be worthmillions of dollars in the morning and nothing in the afternoon. Economistshave grappled with this issue and came up with theoretical solutions. How-ever, their implementation may be costly, due to the need for demand data,thus raising the issue of the trade-off between allocative efficiency andoperational cost-effectiveness (MITCHELL & VOGELSANG, 1991). The ascent of intangible artefacts is accompanied and stimulated on the supply side by the growing importance of intangible assets. Today, theseare considered more important to business performance and survival thanthe physical assets. For consumer goods companies, Coca-Cola, Nestlé or Danone, brand management is the top priority guiding all their strategies.
Brand is also essential for IT companies such as Intel and Compaq, whichare spending substantial sums to build it. Leading marketing specialists,such as David Aa k e r, consider brand an integral part of firm’s equity(AAKER, 1991). Acknowledgement of the importance of intangible assets is not limited to brands. Intellectual property - patents, trademark, technological know-how -is considered as a critical competitive weapon, particularly in software,electronics and biotechnology. The control of intellectual property rights isoften a matter of life and death for companies. It is through intellectualproperty litigation that AMD managed to preserve its foothold inmicroprocessors, despite Intel’s domination. In merger and acquisitiontransactions, the book value has become largely irrelevant to the companyvaluation, which is deter-mined primarily by intangible assets (PETERSENS& BJURSTRÖM, 1991) . Apparently extravagant amount paid for mediaassets, such as Hollywood studios or newspapers, can be explained by thecrucial role attributed to brands, contents and publishing rights in theemerging realm of infotain-ment, combining information and entertainment.
While managers live and die by intangible assets, many accountants refuse to include them in official accounts. Thus, the total amount ofintangible assets in the published 1992 accounts of Coca Cola was 300million dollars, while its brand was valued by Financial World at 35 billiondollars. The 1992 value of Intel brand, as measured by Financial World,was more than 200% higher than the total value of its 1992 balance sheet,8 billion dollars. Intel carries no intangibles on its balance sheet. Similarly,Microsoft considers software development, its core competence, as anexpense and writes it off in the year incurred. In its 1996 Annual Report,Reuters, the leading provider of electronic information, acknowledges thatits balance sheet does not include such strengths and resources as itsneutrality, its software and other intellectual property, its global databasesof financial information and integrated global organization including skilledworkforce. Is it therefore surprising that the average market value ofReuters represented in 1995 and 1996 some 600% of its book value?(Reuters Annual Report, 1996).
The main reason for the non-inclusion is the lack of agreement among experts on how to treat intangible assets. At the risk of stating the obvious,intangible assets are not like tangible assets. First, they are highlyheterogeneous, not only between categories but also within a givencategory: one hour of software programming does not equal another hour of programming. The revenue-generating capacity of an intangible asset ismuch more uncertain than that of a physical investment. When a plant addsanother machine, it can easily quantify the potential increase in output. Onthe other hand, when a com-puter department hires another programmer, itcannot predict with certainty either the quantity or, more importantly, thequality of his/her contribution. Because intangible assets are, by definition, non-physical, they do not follow the classi-cal progressive depreciation rules. Some assets depreciatevery rapidly, others, like a good wine, appreciate with age, stills othersfollow non-linear and often unpredictable life cycles.
Thus traditional ways of valuing assets cannot be applied. The historical cost of acquiring or creating an intangible asset is largely irrelevant.
Opportunity costs are difficult to apply in light of asset heterogeneity. Amarket or transaction-based approach also has serious pitfalls. For mostintangible assets, markets are very narrow and extremely imperfect. Thisapproach also raises the issue of separability. Market transactions onlyrarely concern just one category of intangible assets. Usually what is beingbought or sold is a bundle of assets. Most frequently, the transactionconcerns the control of the firm, in which case it is extremely difficult toisolate contributions of brands, intellectual property or publishing rights asdistinct from the "pure" control premium paid by the acquirer. Finally,transaction-based values are subject to wide fluctuations.
Lat but not the last, intangible assets raise the issue of ownership.
Physical assets are fully and exclusively owned by the firm, which justifiestheir placement on the balance sheet. The situation of intangible assets isconsiderably more ambiguous. For instance, a firm owns its brands andother forms of intellectual property. But does it own its labour force, itshuman capital? And some intangible assets are completely outside thelegal perimeter of the firm. This is the case of customer base, whichconstitute the main asset of banks, telecom companies, retailers and manyother organisations. For Frederick Reichert, a strategy consultant, customerloyalty is the critical determinant of firms’ profitability (REICHERT, 1993).
Thus the issues of accounting treatment of intangible assets remain largely unsettled. Many experts feel that the traditional accountingapproaches are inadequate to handle these assets. Their specificity havespurred the emergence of alternative approaches (LEV, 1999). Dematerialisation logic
The impact of the intangible economy is pervasive and ubiquitous and affects all sectors and activities. Intangible economy does not eliminateagriculture or industry. However, its underlying logic affects all economicrelationships and profoundly transforms the ways firms and markets areorganized and transactions are carried out. The dematerialisation logic isunsettling to the extent that it runs squarely against some of the key tenetsof the conventional logic of economics. The conventional logic stressesequilibrium, the dematerialisation logic, disequilibrium. Phenomena such asobsolescence or redundancy, which had been perceived as peripheral andpernicious, are now pivotal and es-sential, moulding consumption patternsand guiding asset deployment. The three fundamental features ofdematerialisation logic are: abundance, interpenetration and inde-terminacy. Intangible economy is structurally abundant. Abundance, of course, is not a new phenomenon. The productive potential of industrial economy isenormous and clearly ex-ceeds the demand absorption capacity. However,physical goods are subject to physical decay and their consumption marksthe beginning of the end of their economic life. Intangible artefacts, on theother hand, are not only extremely cheap to replicate but fur-thermore arenot destroyed through consumption. The intangible economy superimposeson the abundance of production the abundance of accumulation. Contraryto a popular belief, intangible does not signify ephemeral. The lifecycle ofpopular intangible artefacts is considerably longer than that of materialgoods: we will forever read Balzac books, listen to Bach music or watchBergman movies. Financial systems generate too many transactions, Hollywood, too much entertainment, Internet, too much information. The gap between supply anddemand of intangible arte-facts is so huge that it has created an "infoglut":"the inability to absorb the torrential and continuously swelling flood of data,images, messages and transactions" (TETZELI, 1994) . The on-goingderegulation of markets for intangibles along with the technologicalevolution continue to aggravate the overload. For instance, the number oftelevision channels in the European Union has increased from 40 in 1980 to150 in 1994. Progress in transmission and distribution techniques makes itfeasible to increase the number of channels to 500 or even more(HEILEMANN, 1994). Moreover, the overload is self-perpetuating: to navigatethrough it we need catalogues, indexes, documentation, whose veryproliferation calls for more cross-references, hypertext links and so on.
Efficient infoglut management requires more rather than less information.
Information about information is a growing business. To cope with abundance, new modes of consumption have emerged: zapping, surfing or browsing. They are characterised by a short attentionspan, latency, high frequency of switching and capriciousness. They blurthe distinction between consumption and non-consumption, renderingpricing problems even more intractable. The expanded range of output makes consumer choice more difficult, by continuously raising the cost of acquiring information about the output. Tominimise this cost, the choice is increasingly determined by criteria otherthan product characteristics such as brand familiarity or mimicking andfashion (BIKCHANDANI, HIRSCHLEIFER & WELCH, 1993; VEBLEN, 1899).
These criteria are discretionary, generat-ing rapid and massive shifts indemand, which are hard to anticipate. The result is an uneasy coexistenceof stability, for products and artefacts associated with a strong brand, andvolatility for the others. The trend is clearly toward the latter. Product cycle is becoming shorter. Obsolescence is no longer an external constraint, it becomes an instrumental variable. In areas, such asmicrocomputers, obsolescence leads to cannibalisation: new products areintroduced to replace products that are still successful. Intel and Compaqare particularly skilful in the use of cannibalisation to keep their competitorsoff balance (CHREIKI, 1995; ALLEN, 1992). A crucial implication of supply abundance is the ubiquity of failure. Flops are the rule, successes, an exception. In Hollywood, one movie is made outof a hundred scenarios under development, and only one in six moviesreleased makes money. The flop rule is not limited to intangibles. In thepharmaceutical industry, only one in 4000 synthesised compounds evermakes it to market and only 30% of those recover their development costs.
In consumer goods industry, over 80% of new products launched in theUnited States fail within two years (MOORE, 1995; POWER, 1993). And yet,despite this dismal outlook, the pace of intro-duction of new products hasnot slackened. This has become a wager economy: higher and higherstakes against lower and lower odds.
The wager analogy helps to explain why most companies continue to generate new products at a rapid rate. As long as a player remains at thetable, he has a non-zero probability to recoup his losses. Only if he walksaway, his loss becomes final. Also, what really matters is not so much howmany times one plays but the overall magnitude of the gain (or loss). The seemingly irrational product development strategies reflect the need for brand preservation. New products can be considered as visible signalsof both brand continuity and renewal. Another factor is what can be called a"bookstore" effect. The best bookstore is the one that offers the widestchoice. Furthermore, a well-stocked bookstore stimulates browsing whichleads to greater book consumption. The bookstore effect explains forexample while Reuters maintains 20 000 pages of data in its on-linefinancial information services, while the overwhelming majority of its clientsuse only four or five. The value of its databases is derived not only fromparticular pieces of information but also from the total inventory of data. Structural abundance has also a major impact on the notion of capacity and the use of productive assets. Redundancy and excess capacitybecome the rule. For instance, over-all use of the capacity of traditionaltelecommunications network is of the order of 1 to 5%. While in theindustrial economy, excess capacity is synonymous with costly inefficiency,to be avoided, in the intangible economy it is widespread, functional andinexpensive. It is functional, even necessary, because it enables users andproducers to cope with demand volatility and to accommodate the newconsumption modes. Excess capacity is inexpensive because in theintangible economy the key flows are that of information rather of physicalgoods. The economics of adding additional capacity for information flowsare very different from that for physical goods handling. The latter is clearlysubject to diminishing returns and thus its marginal costs are high. In therealm of Information Technology, there might be diminishing returns atsome point but they are unlikely to be reached in the foreseeable future.
The long-term trend is for an expo-nential progression mode and for adramatic fall in unit processing and transmission costs. The ongoing shiftfrom 32 bit to 64 bit processors will increase the addressable memoryability by a factor of four billion. In telecommunications, fibre optic cable of-fers ten orders of magnitude greater bandwidth than copper wire with tenorders of magnitude lower bit-error rate (GILDER, 2000). The intangible economy undermines traditional frontiers and distinctions.
Sectoral boundaries are crumbling: previously separate activities oftelecommunications, informatics, electronics and audio-visual entertainmentare now overlapping. Time-honoured distinctions between work and leisure,home and work-place, intermediate good and final output, consumer andproducer, product and service, become blurred. Not only are theboundaries porous and overlaying, they are unstable.
This is not a one-off effect of transition to a new environment but a fundamental trend. The intangible economy does not follow the rules ofbinary logic of exclusivity but that of fuzzy logic of overlapping ( K O S K O ,1993). The interpenetration profoundly changes the nature of the firm and its relationships with the environment. Internal links, between firm and itsemployees, become weaker; external links, between firms and its suppliers,stronger. While employees are told to work at home, suppliers are invited towork on premises. Functions traditionally considered as central to the veryexistence of the firm are now subcontracted or outsourced. Nike, leader insport shoes, does not manufacture any shoes. Nor does Dell, a leadingsupplier of computers, own any production plant. In the semiconductorindustry, many leading firms are "fabless", concentrating on chip designand subcontracting their production (RAPAPORT & HALEVI, 1991) . Incomputer services, outsourcing is one of the highest growth sectors.
This development suggests that the traditional rationale for the existence of the firm, articulated by Ronald Coase as the minimisation of transactioncosts, is no longer universally valid (COASE, 1937; WILLIAMSON & WINTER,1 9 9 3 ). Not only has Information Technology dramatically reducedtransaction costs but the intangible economy has altered the nature of themarkets (see below) and enlarged the range of transaction mechanisms,blurring the well-known distinctions between markets, hierarchies andnetworks. An alternative and broader rationale for the firm needs to bedeveloped, which would stress the brand umbrella, the intellectual propertyrepository and the control of distribution channels as key cohesion factorsand functions of the firm. Dematerialisation logic modifies the market power balance and the value chain structure. In the industrial economy, the central position in the valuechain was that of final product assembly, while the position of subcontractorwas subordinate. Despite the fact that Michelin contributed more to thedevelopment of the automobile, by facilitating road travel with maps, signsand guides, than Renault or Citroen, the latter gained greater market power:few people ever buy their cars in function of the brand of its tires. In theintangible economy, a subcontractor often assumes a dominant position.
Thus, in personal computers, Intel and Microsoft, are in a considerablystronger position, and are more profitable, than IBM or Compaq, whichcontrol final assembly. Their dominance is due to their ability to establishintellectual property rights over key product components, in this instancemicroprocessor architecture and operating system software.
These changes in the value chain structure reflect a fundamental trend: the weight of the value chain is moving closer to the consumer. This trendhas led to the emergence of "power retailers" such as Wal Mart in the US orIkea in Sweden. They decide which products are put on the scarce realestate of store shelves. They also set prices and become increasinglyinvolved in product design. The transfer of market power does not stop at the check-out counter and frequently crosses the producer-consumer divide. In the personal computerindustry, between 1986 and 1991, customers captured 49% of value added,against 31% for software and services suppliers and 20% for equipmentmanufacturers (McKINSEY & Co, 1992). The intangible economy brings about a momentous change in the relationship between suppliers and consumers - the end of informationasymmetry. Today in many businesses, the customer knows as muchabout products and markets as the supplier. This entails not onlysubstantial end-user price falls, due to the loss of the market power ofsupplier, but also an unbundling of the production and assembly process,which becomes interactive. The unbundling is particularly apparent in theInformation Technology area. Software applications and corporate datanetworks are often designed and built by customers, using inputs fromdifferent suppliers. Of course, they can also be created by suppliers withinputs from customers. "Make-or-buy" decisions are becoming moreprevalent and more convoluted. The nature of competition changes: forcomputer services suppliers, such as IBM or EDS, their biggest competitorsare not the other suppliers but their clients. Markets for intangibles and intangible markets
Changes in the consumption mode, in the production function and in interfirm relation-ships necessarily entail - and also reflect - a change in thenature of the markets. The main purpose of markets is no longer to supportthe trading of physical goods but to facilitate exchanges of intangibles. Thisdoes not mean that markets for physical goods have disappeared orbecame irrelevant. They are alive, well and growing. However, markets forintangibles are growing considerably faster. Furthermore, the evolution ofphysical goods markets is heavily influenced by the logic of intangiblestrading. Is not a notion of markets for intangibles an oxymoron? In physical goods markets, it is easy to identify discrete transactions. Buyers and sellers areusually distinct. This is not universally true in markets for intangibles. There,transactions form a continuous process. For many artefacts, the distinctionbetween buyers and sellers is tenuous. Academics exchanging data overInternet or financial institutions on a trading network are in turn producersand consumers of information. We have seen above the difficulties ofestablishing appropriate pricing mechanisms for intangibles. Give-aways,subsidies, cross-subsidies, indirect (third-party) payments or bundled pricesare the rule rather than an exception. The peculiar characteristics ofintangibles lead many analysts to argue that they should not be tradedthrough traditional markets. Ronald Coase, Nobel Prize laureate, attackedthis argument and suggested that the market for ideas should be approachedin the same manner as the market for goods (COASE, 1974). We would like tosuggest a variation of this suggestion: markets for goods should be treatedas a special case of markets for intangibles. In any case, the distinction becomes more and more tenuous, all markets have become more and more intangible both in terms of underlying productstraded and in the way they operate. Take their most visible form, the financialmarkets. Over last thirty years, these have become enormous: the volume offoreign exchange transactions is close to 1 500 trillion dollars a day, whichmore than seventy times the daily volume of interna-tional trade of goods.
While the international trade is growing at a single digit rate and an annualincrease of 5 % is considered as a good performance, internationaltransaction grow at a double digit rate. Thus, according to the data gatheredby central banks and consolidated by the Bank for International Settlements,between April 1989 and April 1992, the value of foreign exchangetransactions has grown by 42% after doubling between 1986 and 1989.
Capital markets (equity and bonds) have become a principal conduit offunding of technological innovation, accelerating its diffusion and, in theprocess, radically changing traditional notions of economic hierarchy andcapital mobilization. This rapid growth would not have not been possible without a massive use of informa-tion technology and a resulting substitution of intangible datafor physical products. What changes hands in those markets are notbanknotes or stock certificates but book entries in computerised databases ofaccounts. Furthermore, the progress in financial economics theory led to thecreation of new markets that trade dematerialized derivates of traditionalproducts such as foreign exchange, interest rates or equity portfolios.
Derivates markets, futures, options, swaps, etc. have dramatically expanded the notions of tradeability and risk management. They have been growingmore rapidly than cash markets in the underlying instruments(SCHOLTES, 1995).
■ Financial Markets Paradox
Financial markets, their rapid development and ubiquity, are at the core of intangible economy. Yet, their prominence constitutes a major paradox.
On the one hand, they represent as good example of well-functioningmarket mechanisms as we ever are going to get. On the other hand, theyare extremely controversial and many important and influential peopleargue that they are either parasitic or actually harmful. Ultimate triumph of the market paradigm?
For the economic profession, which has forever preached the virtue of markets, financial markets represent a dream come true, a real-lifeincarnation of the theoretical model of the perfect market, information-richand cost-effective. Financial markets are ubiquitous, both local and global;they are accessible twenty-four hors a day, seven days a week; they arehuge and diversified, covering a wide range of instruments and involving alarge number of participants; banks and other financial institutions, industrialcorpo-rations, service companies, individual and institutional investors. Moreimportantly, they offer an incredible wealth of information about the economy,both in the aggregate and in detail. Furthermore, financial markets are verycost-effective: transaction costs are low and falling. Widespread controversy
And yet, few people appear happy about the apparently irresistible surge of financial markets. This surge has generated a widespread controversyand dissatisfaction, not only among left-leaning politicians but also amongeconomic policymakers, market regulators and even market practitioners.
Dissatisfaction is based around three major accusations: - markets have become too powerful;- markets are too volatile and - markets send wrong signals about economic performance andeconomic value.
According to these accusations, we live in the era of the dictatorship of financial markets. Economic policy is driven by financial markets operators,who are oblivious to long term development perspectives and are onlyconcerned with short-term gains. Their increased power and influenceimply that the policy decisions and business strategies are increasinglyafflicted by short-termism, an excessive focus on short-term financialperformance.
The second major accusation is that the financial markets are unstable.
The volatility of financial prices is widespread, persistent and contagious.
No segment of financial markets is immune: foreign exchange marketshave been volatile since 1973, interest rates since 1979 in the UnitedStates and mid-1980s in Europe, equities have become more volatileduring the 1980s. Volatility in financial markets
Sources: Leuthold Group: Salomon Smith Barney Stability appears practically impossible to maintain in the long run. Thus, the foreign exchange stability between European Union countries, broughtabout the creation of at the euro in 1999, has been at least partially offsetby strong variations of parity between the euro and the dollar.
Volatility results not only in wide swings of value but also in a persistent gap between financial and economic value. In the foreign exchangemarkets, "overshooting" and "undershooting", a high degree of divergencebetween the exchange rate - as determined by the markets - and the ratecalculated on basis of purchasing power parities, are endemic. Even MiltonFriedman, an ardent defender of floating exchange rates, acknowledgedrecently that the mark was "extraordinarily overvalued", relative to thedollar (1). Similar divergence between financial value and economic valuecan also be seen in the equity markets and are often considered as a majorreason for reticence of many companies to be listed on public exchangemarkets. For many observers, these divergences are an evidence of the financial "bubble", the uncoupling of financial markets from the "real" economy.
Markets are no longer a reflection of the underlying economic reality; theyhave become autonomous. Thus, the value of foreign exchangetransactions is about seventy times higher than the value of physical tradeflows. Autonomous, markets are self-centered and incestuous: the bulk oftransactions in financial markets are carried out between financialintermediaries rather than between these and non-financial clients.
Bubbles cannot inflate indefinitely, they have to burst periodically and often brutally: hence the increasing frequency of financial crashes. Globalequity markets have crashed in 1987, in 1989 and again in 1998 and 2000,bond markets collapsed in 1987, 1994 and 1998 every time wipinghundreds of billions of dollars of market value. Losses caused by crashesare so large as to become immaterial. This immateriality is accentuated byan apparent lack of impact on the real economy. Thus, US economyexperienced excellent macro-economic performance in 1988 and 1994.
The accusations against financial markets need to be taken seriously.
Clearly, the evolution of financial markets has worrisome aspects. Theirrapid development accentuates the lag between the speed of the evolutionof economic systems and that of political and social structures, creatingstrong tensions and conflicts and perpetuating a specter of worldwide"meltdown ". National and international regulatory authorities such as cen-tral banks, Finance ministries and international organizations (IMF or theWorld Bank) appear unable to control market evolution, particularly as (1) Declaration to Reuter Press Agency, reported in L'Echo, 15-17 April 1995.
regards international transactions. They live in the mode of permanentcrisis management. A feeling of loss of control is widespread andpersistent. Numerous projects to remedy this situation have beenformulated, including many by high-ranking government officials. Yet, sincethe early 1980s, practically every ambitious project to reform internationalmonetary reform and to eliminate price volatility has failed.
Conventional explanations
One of the reasons for failure of reform attempts is the lack of understanding of fundamental dynamics of financial markets. Conventionalexplanations can be classified in two broad and radically opposed schoolsof thought.
The first school postulates the rationality of financial markets. It is built upon the classical economic theory, according to which a combination ofpure competition and abundant information lead to efficient pricing whichreflects economic value. The rationality view was used to justify the"floating" exchange rates and underlies the "random walk" theory ofsecurities markets. This theory postulates that in an efficient market, allavail-able information is incorporated in the quoted price of a security andthus at any given time, this is the optimal price (2). The second school postulates the irrationality of financial markets. This theory looks at the behavior of market participants and takes a pessimisticview of human nature. Financial markets are driven by speculation, by amix of greed and fear. Market participants indulge in strategic behavior:their action are determined not only or primarily by their views of the marketbut by assumptions about the attitude of other participants. Herd instinct isprevalent and explains the constant undershooting or overshooting.
According to the irrationality view, abundant information and low transactioncosts create excessive liquidity and overabundance of transactions, manyof them have no economic justification and only create "noise" whichreduces the signalling capacity of the markets. Holders of this view do notbelieve in self-correcting market mechanisms. Left to their own devices,financial markets will produce pernicious results. They need to be controlledor even restrained. For instance, one of the more prominent members of (2) The classic presentation of Random Walk Theory can be found in MALKIEL(1992).
the irrationality school, James Tobin, Nobel Prize in Economics, suggestedin 1978, that in order to reduce excessive liquidity, a tax on internationalfinancial transactions of speculative nature (TOBIN, 1978). This proposal hasbecome particularly popular in France, where it has been endorsed by JeanPeyrelevade, a leading French financier and present CEO of CréditLyonnais, and Le Monde Diplomatique. In June 1998, a group of Left-wingintellectuals founded an association, ATTAC, to support the Tobin tax ( 3 ).
ATTAC has become one of the most active promoters of anti-globalisationmovement, which vehemently criticizes the World Bank, IMF and WTC. Both schools of thought appear somewhat unconvincing. They fail the reality check and lack explanatory power. In fact, markets aresimultaneously rational and irrational. It is true that due to the volatility,intrinsic values become difficult to establish, yet markets are often right.
Their predictive and corrective power is widely recognized. Thus in the caseof pound sterling devaluation in 1992, markets were right and the UKgovernment was wrong: the strong pound was strangling the Britisheconomy and the devaluation has had a positive impact on growth andemployment, without creating undue inflationary pressures. In case of securities markets, despite crashes and aberrations, markets do send useful signals about relative performance of companies andeconomic sectors. These markets have a critical role in the development oftechnologies, whose impact have been far reaching and long-lasting:personal computers, new media, biotechnology and Internet. One can applyto financial markets the Churchilian definition of democracy: it is the worstsolution, except for all the others. A more important shortcoming of conventional approaches is that they do not provide to key questions about financial markets: - Why is the volatility so pervasive and persistent?- What are the pernicious effects of financial markets? For instance,does the exchange rate volatility hamper international trade? - What are the growth drivers of international finance? Why financialtransactions grow more rapidly than underlying "physical" exchanges? (3) Association for Financial Taxation for the Aid of Citizens – www.attac.org ■ An Alternative Approach: Financial Markets as
Information Markets
In order to provide at least some elements of answer to those questions, we need a new approach. We would like to recommend such an approach,based on the view of financial markets as information markets.
The starting point for such approach is simple: we should not consider financial markets evolution as a temporary aberration. Markets are too largeto consider that they are driven by a small bunch of greed-crazedspeculators. Furthermore, market trends have been at work for over twentyyears. There is an underlying logic that explain the key features of financialmarkets.
Our key hypothesis that the information is not only the support for trading of physical goods and services. It has become the principal object oftrading. Financial markets are essentially, but not exclusively, informationmarkets. Financial markets are not only in-formation-driven, they areinformation-focused. Exchange of information, viewpoints, judgments andopinions has become their main function. This hypothesis is not entirely new. It has been first formulated by Walter Wriston, eminent American banker and Chairman of Citicorp between 1974to 1984. In a famous statement, Wriston said that "information about moneyhas become more valuable than money itself" (WRISTON, 1992) The information market hypothesis appears quite helpful in understanding of financial market dynamics. Thus, it provides anexplanation of market incestuousness, the pre-dominance of transactionsamong financial intermediaries. These transactions can be seen aselements of a search process, described by George Stigler in his classicalarticle on the Economics of Information. According to Stigler: "The larger of the fraction of repetitive (experienced) buyers in the market, the greater the effective amount of search (.)" (STIGLER, 1961).
Furthermore, Stigler shows that the amount of search (the number of transactions in our case) is proportional to the level of expenditures. In ourcase, this means that biggest players, large financial institutions, are alsothe most active players. It also means that an increased level of volatility(which increases the level of risk and uncertainty) entails a greater numberof transactions. We will come back later to this relationship.
In the information markets, the principal strength of participants is not a detention of financial assets but the mastery of price information. Suchmastery can only be acquired through extensive practice of numeroustransactions. The growing importance of price information also explainswhy the providers of this information such as Reuters or Bloomberg havebecome so powerful and profitable, considerably more than most of theirbanking clients.
The information market hypothesis offers a plausible explanation of the growth dynamics of financial markets. The globalization of the economyand the increasing variety of physical transactions have created greateruncertainty and thus generated a strong and continuous demand forinformation. Financial markets are an ideal conduit for displaying andexchanging such information. Higher level of risk and uncertainty alsocreate a strong demand for information about the future. Past can no longerbe considered as a reliable guide to the future. Derivative marketsrepresent an aggregation of collective views about the future. Since suchviews are more valuable to market participants than the information aboutthe past, the demand for former is relatively greater the demand for thelatter, hence the higher rate of growth.
Information as an economic good
If the exchange of information is the focus of financial market transactions, it is useful to inquire about the nature of information as aneconomic good. Using the framework, developed above, informationappears as an intangible good, to the extent that its value is determinedprimarily not by its physical support but by its immaterial content.
Consumption of information is non-subtractive. Moreover, their property is shared not only among many consumers but also between sellers andbuyers. Why a seller of a physical good looses its property, a seller ofinformation can continue to use it. Although information can be exclusiveand produced for a use of a specific individual of a group, such exclusivitycannot be durably maintained. This is because the marginal cost ofconsuming an intangible artifact is very low. A fundamental reason is thatintangible artifacts are fungible: separation into discrete units is difficult,arbitrary and unstable. They are simultaneously lumpy and infinitelydivisible. Like a witch in the enchanted forest, artifacts with identical contentcan appear in various disguises and shapes, in function of their physicalsupport: a same information can be printed, shown on television, spoken over a radio network, distributed via an on-line network and so on. Eachtime it is the same information and yet, each time, it is a different artifact. Traditionally, it was difficult if not impossible to dissociate content from its physical support. The growing dematerialisation of the economy meansthat content can now be more easily dissociated from the support;particularly if the former is in the digital form. Nevertheless, the property ofjointness has not disappeared. Accessing information through Internetrequires infrastructure and interfaces, which are distinct from the in-formation itself.
Information markets
Information is structurally abundant. Every economic activity produces more information than it consumes. Information supply and demand areself-sustaining and feed off each other in an ever-expanding growth spiral.
This interaction is further stimulated by the way the information is valuedand priced. The information value is content-based. But this value is largelydetermined, on the one hand, by the context and, on the other hand, by theability to associate and to relate a piece of information to other information.
The objective of search for information is not only to find the rightinformation but also to establish correlations and relationships betweendifferent data. A data on a price of a given commodity is more valuable tous if it can be compared with its past evolution (time series) or with prices ofrelated commodities. Information markets are associative. An associationand comparison of data can be created either by the supplier or by theuser. Thus the same information can be used as a final product or as a rawmaterial. This means that the value of information cannot be firmlyestablished prior to a buyer-seller transaction, because it changes as aresult of the transaction. In technical terms, information asymmetry betweenthe buyer and the seller evolves constantly through their interactions. Moreimportantly, one of the main purposes of financial market transactions is thereduction of information asymmetry.
Not only different users value the same information differently but, over time, the same information can be valued very differently by the same user.
Moreover, while the value of information changes continuously over time,this change is not linear. Financial markets provide numerous examples ofmajor price discontinuities. Prior to market opening, an inside informationabout a given company’s financial situation is worth millions or tens ofmillions of dollars. After the information has been published and for few minutes during which the market is absorbing it, it is still valuable. Once themarket has absorbed it becomes worthless. To make things even morecomplicated, one can observe in the information markets the coexistence ofcontinuities and discontinuities. For a given financial instrument, a bond oran equity, one can observe small variations over short term, large variationsover long-term and clear or even linear trends over long-term. Pricing difficulties
Valuation problems are further compounded by difficulties of establishing appropriate pricing mechanisms. Information cannot be sold asa discrete and homogeneous good. It is always sold as a joint product,content and support and often sold in bulk. Pricing mechanisms are oftendetermined by the supply considerations, the ease of capture of thephysical support or the ease of delivery. A price of a book is not a functionof its content but of the cost of its production. And yet, the value ofinformation bears little relationship to the cost of producing it. Willingness topay is very difficult to establish, not only because the value of informationchanges constantly over time but because of extensive externalities.
Opportunities for freeloading are numerous and their cost is low.
Conversely costs of controlling and policing the freeloaders are very highand often superior to the revenue generated. Furthermore, information isinherently incestuous; its producers are also its consumers. If we lookaround institutions seen as major producers of information, universities,medias, banks or insurance companies, is it not true that they consumeinternally the bulk of information ‘output’ they produce? This means simplythat the very large share of information generated are consumed free ofcharge. Some information suppliers recognize it explicitly: thus allnewspapers assert proudly that each copy of their paper is read by 4 or 5people: one paying customer for three or five "freeloaders"! Even whenpayment actually takes place, different pricing arrangements can apply tosimilar information artifacts. In the information market, give-aways,subsidies (through advertising), cross-subsidies, indirect (third-party)payments, composite or bundled prices are a rule rather than exception.
Thus, information markets are simultaneously highly efficient and liquid and riddled with externalities and imperfections, which explains theircomplexity and instability.
Let us conclude this summary overview of information as an economic good by noting a schizophrenic way in which economists approach the fundamental relationship between information and uncertainty. For theclassical economic theory, the relationship is simple: the role of informationis to reduce the uncertainty: more there is information, less there isuncertainty. Yet for the financial economists such as Charles Goodhart, therelationship is strikingly different: uncertainty determines the value ofinformation (GOODHART, 1975) . Higher the uncertainty, higher the value ofinformation. This view is consistent with the Communication theory ofClaude Shannon, which provides the theoretical underpinning for thedevelopment of modern telecommunications and informatics (SHANNON &WEAVER, 1947).
■ Some Implications
What implications can be drawn for this overview for this approach of Inherent and persistent instability
First of all, under the information market approach, there is no stable price equilibrium. In physical goods market, it is easy to identify discretetransactions and there is a "natural" balance between buyers and sellers.
Given the exclusive nature of the consumption of physical goods, buyersand sellers usually distinct and a connection between a given good and itsprice is direct and unequivocal. This is not true in markets for information.
There, transactions form a continuous process and the relationshipsbetween goods and prices are complex. Furthermore, due to the sharednature of information, there can be multiple sellers for a single buyer andvice versa. The rapid diffusion of information encourages a disequilibriumbetween buyers and sellers: it is very dangerous to go against marketconsensus, to be a single buyer when everybody else is selling or viceversa. Hence the well-known propensity of financial markets to herd. Thereare many striking examples of herding. For instance, in February 1994,there was a huge crash in US government bond market, with worlwidelosses totalling 1.5 trillion dollars. Following a Federal Reserve decision toincrease short-term interest, long-term interest rates rose as well, thuslowering the price of long-term bonds. Because this rise was unexpected,the fall in bond prices was dramatic. In the aftermath of the crash, FortuneMagazine carried a survey of portfolio managers to find out how many ofthem correctly anticipated the evolution of interest rates and position their portfolio accordingly. Out of some 1000 managers surveyed, they foundexactly four who have correctly anticipated the direction of interest ratemovement (EHRBAR, 1994). Thus, contrary to popular belief, the greateravailability of information does not necessarily lead to a greater dispersionof views and positions and it may actually lead to greater homogeneity,which can only breed instability. Anticipation
Instability is further stimulated by anticipatory nature of financial markets. Information about the past is widely known and certain and itsdirect value is falling. Thus the views about the future, less known andinherently uncertain, become main objects of trading. This explains aparadoxical behavior of financial markets. For instance, the value of somestocks fall on good news, while the value of others rise on bad news. Thereason that those news are compared to prior expectations. Thus, if marketanalysts anticipate that a given company will grow at 50% rate over thenext year, and the company only grows at 40% rate, its stock will fell whenthe actual growth rate is announced. And, if the market anticipates thatanother company will experience substantial losses, should the actuallosses be lower than the market anticipation, the stock price of the losingcompany would rise.
The search for information about the future has not only given existing markets a more anticipatory character but also led to the creation ofmarkets dedicated exclusively to trading on future information: forward,future and options markets. As information about the future isfundamentally and irreducibly uncertain, anticipatory markets are inherentlymore volatile than markets that aggregate past information. Thus, derivativemarkets are more volatile than the cash markets. Our analysis suggests that the volatility of financial markets is largely linked to their anticipatory nature. Higher the anticipatory content ofinformation, higher the volatility. As anticipatory markets are moreinformation-rich and efficient, higher the information content, higher itsvolatility. Thus the relationship between information and volatility mirrors therelationship between information and uncertainty. This explanation, whileradically different from that offered by traditional economic theory, appearsconsistent with observed market behavior. It explains for instance not onlythe persistence of vola-tility but also the coexistence of volatility andliquidity. Traditional analysis of markets postulates that volatility limits liquidity. While it is true that at some levels, volatility leads to marketbreaks, such as one seen in Chicago futures markets in October 1987, itcan be argued that in anticipatory markets volatility is a necessary conditiono f l i q u i d i t y , a f u n d a m e n t a l m e c h a n i s m t o c o p e w i t h i n f o r m a t i o ndisequilibrium. Strategic behavior
Another reason for fundamental instability of financial markets is their structure and the resulting attitudes of market participants.
We have noted the incestuousness of financial markets. Under information market approach, the main reason for this incestuousness isthat financial institutions are simultaneously largest producers andconsumers of financial information. This means that they aresimultaneously buyers and sellers. In technical terms, they are market-makers, displaying to the market two-way prices, the buy price and the sellprice, the former being preferably lower than the latter.
This two-way approach reflect a fundamental ambivalence of information: for any market participants, information is their biggest assetbut also their largest liability. They can make huge profits from theinformation trading but also huge, potentially fatal losses. Thus, theyconstantly grapple with major dilemmas: how much information to disclose?How active to be in the market? How to balance aggressive informationtrading and risk management? If they do not disclose their views andinformation they hold, nobody will trade with them. If they disclose toomuch, the size of position they hold in a given instrument, for instance,other participants will take advantage of it, causing large losses.
Furthermore, in an information-intensive market, any advantage fromexclusive access to information is short-lived and has to be capitalizedupon quickly.
In order to cope with this dilemma and associated risks, market participants adopt a range of strategic attitudes. Not only they constantlylook for new and more forward-looking information, they also seek toanticipate reactions of other participants. This goes well-beyond whatKeynes described as "beauty contest" approach: in order to choose themost beautiful girl one need to consider not only his own preference butalso that of other judges (KEYNES, 1936; ORLEAN, 1989 and THALER, 1991).
Sophisticated players are looking for second and third order derivatives andseek manipulate the signalling aspects of their transactions.
In the financial market context, greater information transparency does not reduce the risks to the participants, to the contrary it may increase it. Thusmarkets players often seek to limit transparency, to make market moreviscous, by introducing zones and periods of opacity. The viscosity is aparticular concern of marketmakers, who want to hide the size of theirpositions. In foreign exchange markets, where transaction size is very large,all prices quoted on broadcasting networks such as Reuters or Telerate areindicative and have to be negotiated on a bilateral basis. In equity markets,such as London Stock Exchange, price quotes displayed on publicinformation networks are binding up to certain quantities but, until recently,market makers did not have to publish their large transactions. This ability tohide position has generated considerable controversy between partisans oftransparency and marketmakers who argue that the premature disclosurecreated an unacceptable risk level and therefore they would be unwilling toengage in large transactions (London Stock Exchange, 1995).
The risk of transparency and marketmaking explains the persistence of brokers in a market with abundant information flows, and therefore a priorisusceptible to disintermediation. Those brokers no longer intermediatebetween end-clients and financial institutions but between marketmakers(inter-dealer brokers). Market regulators are continuously trying to strike a right balance between transparency and opacity but such balance appears elusive as it isquite difficult to set hard and fast rules, applicable all the time to all markets.
Network of networks
The fact is that, in the world of global economy and geofinance, markets remain remarkably diverse and this diversity is growing. Financial marketsshould be seen as a loose set of interconnected markets rather than asingle tightly integrated market. They constitute a network of networks, afinancial equivalent of Internet. Not only, there are different markets fordifferent instruments, foreign exchange, bonds, equities but within singleinstrument, market structures can vary considerably. For instance, withinequity markets, it is customary to distinguish between order-driven markets,where all orders are transacted in a single market place, and quote-drivenmarkets, where transactions are carried out by competing market-makers.
A specific academic discipline, market microstructure analysis, has beencreated to deal with the issues of various structures of equity markets(COHEN, MALER, SCHWARTZ & WHITCOMB, 1986; GILLET & MINGUET, 1994).
The persistence of market diversity and fragmentation can be explained by the strategic behavior of market participants seeking to preserve theirparticular mix of transparency and opacity and the concerns of regulatoryauthorities, who want to maintain control of markets. But it is also areflection of fundamentally heterogeneous nature of information.
Information should not be seen as one large and undifferentiated flow ofbits. Rather it is comprised of thousand of distinct if interrelated streams ofdata, images and symbols. References
AAKER D. (1991): Managing Brand Equity, The Free Press, New York.
A L L E N M. (1992): "Development of a New Line of Low-Cost PCs Shakes UpCompaq", The Wall Street Journal Europe, June 16.
BIKCHANDANI S., HIRSCHLEIFER D. & WELCH I. (1993): "The Blind Leading theBlind: Social Influence, Fads and Information Cascades", UCLA Working Paper,October.
CHREIKI E. (1995): "Intel: des sauts de puce toujours plus rapprochés", Le NouvelEconomiste, December 8.
C L A R K C. (1985): Conditions of Economic Progress, Macmillan, London; Toffler,Alvin.
COASE R.:- (1974): "The Market for Goods and the Market for Ideas", American EconomicReview, May.
- (1937): "The Nature of the Firm", Economica, 4.
COHEN K.J., MALER S.F., SCHWARTZ R.A. & WHITCOMB D.K. (1986): T h eMicrostructure of Securities Markets, Prentice Hall, Englewood.
COX B. (1994): "Superdistribution", Wired, September.
DYSON E.:- (1992): "Who pays for data", Forbes, February 3.
- (1995): "Intellectual Value", Wired, July.
EHRBAR A. (1994): "Defying the Odds", Fortune, October 17.
GADREY J. (1992): L’économie des services, Decouverte, Paris.
GILLET R. & MINGUET A. (1994): Micro-structure et rénovation des marchésfinanciers en Europe, PUF, Paris.
G O L D F I N G E R C. (1994): L’utile et le futile, l’économie de l’immatériel, EditionsOdile Jacob, Paris.
GOODHART C.(1975): Money, Information and Uncertainty, Macmillan, London.
KEYNES J.M. (1936): The General Theory of Employment, Interest and Money.
KOSKO B. (1993): Fuzzy Thinking, Hyperion, New York.
London Stock Exchange (1995): Review of the market makers’ exemption fromdisclosing positions of three per cent or more, Consultative document, March.
MALKIEL B. (1992): A Random Walk Down Wall Street, Princeton University Press.
McKINSEY & Co. The 1992 Report on The Computer Industry.
MITCHELL B. & VOGELSANG I. (1991): Telecommunications Pricing, CambridgeUniversity Press.
MOORE S.(1995): "Glaxo Lab Initiates A High-Speed Chase in the Drugs Industry",The Wall Street Journal Europe, December 6.
NEGROPONTE N. (1995): Being Digital, Hodder and Stoughton, London.
OLSON M. (1973): "Information as a Public Good", in R.S. TAYLOR, Economics ofInformation Distribution, State University of New York, Syracuse.
ORLÉAN A. (1989): "Comportements mimétiques et diversité des opinions sur lesmarchés financiers", in H. BOURGIGNAT & P. ARTUS (ed.). Théorie économiqueet crises sur les marchés financiers, Economica, Paris.
PETERSENS F. & BJURSTRÖM J. (1991): "Identifying And Analyzing 'IntangibleAssets’", M&A Europe, September-October.
POWER C. (1993): "Flops", Business Week, August 16.
RAPAPORT A. & HALEVI S. (1991): "The Computerless Computer Company",Harvard Business Review, 3.
Reuters Annual Report, (1996).
SCHOLTES K. (1995): “La dynamique de la croissance des marchées de produitsdérivés”, Unpublished paper, Faculté des Sciences Economiques et Sociales,Namur, May 5.
SHANNON C. & WEAVER W. (1947): The Mathematical Theory of Communications,Illinois University Press, Urbana.
STIGLER G. (1961): "The Economics of Information", Journal of Political Economy,vol. 69.
STIGLITZ J. (1985): "Information and economic analysis: a perspective", EconomicJournal, 95.
TAPSCOTT D. (1996): The Digital Economy, McGrawHill, New York.
T H A L E R R.H. (1991): Quasi Rational Economics, Russell Sage Foundation, NewYork.
TOBIN J. (1978): "A Proposal for International Monetary Reform", E a s t e r nEconomic Journal, vol. 4.
W R I S T O N W. (1992): Twilight of Sovereignty, How the Information Revolution isTransforming the World, Scribners, New York.

Source: http://www.idate.fr/fic/revue_telech/68/goldfing.pdf