The economic analysis of intellectual property rights in civil law ...
A. Property rights as a response to scarcity. B. Fences as a prerequisite of property. C. Common property. II. Common f...
The economics of intellectual property rights in civil law systems Ejan Mackaay Professor of Law at the Université de Montréal and member of the CRDP (Centre de recherché en droit public) at the same university
Forthcoming in Economic Analysis of Law - A European Perspective, Aristides N. Hatzis (ed.), Cheltenham, UK, Edward Elgar, 2008, (ISBN 1 84064 592 X)
‘If national patent laws did not exist, it would be difficult to make a conclusive case for introducing them; but the fact that they do exist shifts the burden of proof and it is equally difficult to make a really conclusive case for abolishing them Edith Penrose, The Economics of the International Patent System, Baltimore, Johns Hopkins University Press, 1951, 40
Contents I. Pedigree: property rights in tangibles 2 A. Property rights as a response to scarcity 2 B. Fences as a prerequisite of property 3 C. Common property 4 II. Common features of Intellectual property rights 5 A. Information as a special commodity 5 B. Encouraging creation 6 C. Public good qualities of information 7 D. The cumulative nature of creation 9 Conclusion 12 III. On specific features of some intellectual property rights 12 A. Restrictions on intellectual property rights 13 B. Moral rights 14 C. Trade-marks 15 IV. The extension of property rights to new objects 17 A. Rights vs rents - the puzzle 17 B. A discovery procedure 18 C. Discovery procedure and legitimacy 19 Conclusion 21 Suggested readings 22
Intellectual property is the term used for a variety of rights permitting in particular ways and for limited times to control some intangible assets that are human creations. Copyright, patent, trade-mark, trade secrets and newer varieties such as plant breeders’ rights, rights in trade-names, in industrial designs or in microchip topologies, and neighbouring rights are part of it. Economists understand rights and obligations through their incentive effects: how people modify their behaviour in response to changes in their legal situation. Acquiring an intellectual property right holds out the promise of reward for a creative effort and must therefore be expected to stimulate creation. This suggests treating intellectual property rights as a species of property rights generally. Property rights encourage good husbanding and creative use of scarce resources. This logic has been well explored for tangible goods (I). It carries over to intellectual property, allowing for the peculiarities of information as an object of property rights (II and III). This leaves to be explored how the property rights logic is extended to new objects which human ingenuity comes up with in technology, in artistic expression, in business practice. If law is more than legislation upon demand, this matter deserves some attention (IV).
I. Pedigree: property rights in tangibles
A. Property rights as a response to scarcity
Property rights in tangibles are a response to scarcity. Scarcity is opposed to abundance and arises where different individuals want to use a single object in ways which cannot all simultaneously be indulged in. It is hardly worth establishing property rights on abundant commodities: all can use them to their heart’s content without exhausting the supply. Emergent scarcity manifests itself in disputes over who can use what, when multiple uses are no longer simultaneously possible. One solution is to start a fight with the winner taking all. But this is not a recipe for lifting oneself out of the subsistence cycle. A more promising – as we have learnt – formula is to attribute the right to decide what shall be done with the newly scarce resource to one person or group, to the exclusion of others, and to attach to it the right to trade it to someone else: property rights.
The institution has been known from the earliest times, although at the outset it may have been used very restrictively: for personal clothing and weapons. Game theory can be used to explain how this institution cuts short potential conflicts that might otherwise develop. What calls for explanation is why this formula has remained so successful and indeed has been generalised in our economies. With property rights, the owner decides what shall be done with the property but also harvests the fruits of his decision or suffers the losses from sloth or misguided decisions. This provides immediate, automatic and decentralised feedback information on the quality of management decisions. Property rights give the incentive to manage wisely what one currently owns. Where the object can be traded, this effect is enhanced since people now have an interest in investing for future trade and in specialising. Trade gives rise to markets, to the invention of money and to competition, which in turn allow different uses of scarce commodities to be priced and hence more precisely compared. Commodities tend to be traded until they find themselves in the hands of those who value them most highly and this is a source of general wealth increase. To put it succinctly, property rights give incentives to enhance wealth and quell dispute. Law-and- economics scholarship has drawn attention to these qualities and explained how they work.
B. Fences as a prerequisite of property
Property rights are viable only in as much as use of the scarce object can effectively be reserved to the person or group designated as owners. For many objects this is unproblematic. You keep an eye on the food you just bought to eat; you keep your living quarters under lock. How serious you are about locking up depends on who might be tempted to take things from you; in remote areas where you know all your neighbours, you may not need to lock up at all. For some objects, it is difficult – with technology we now have – to reserve use to particular persons. Flowing water, clean air, fish in open sea are examples. For such objects property rights are not directly viable. If they are nonetheless scarce and left in open access, they tend to be overused (as in overfishing or pollution of the environment); if they must be produced by human effort, too little of it may be forthcoming. Of course, we can appeal to people’s sense of responsibility or more forcefully, adopt rules to curtail consumption, but the logic of the situation is such that each person, while paying lip service to the common weal, will be tempted secretly to pursue his or her private interest and consume more. Since all have the same incentives, we face collective ruin. This prospect is known, following a classical article by Garrett Hardin, as the tragedy of the commons. We need to be more precise about what is to be reserved to owners. The viability of property rights and the virtues referred to earlier depend on ways in which owners can effectively get their hands on the fruits flowing from the use of the scarce commodity. It will be helpful to use the term fences for a variety of devices and institutions used to accomplish this. Fences can be physical stops such as wooden fences, hedges or ditches. After the invention of barbed wire cattle could be bred in the American West on far smaller areas of land than before. This is a reminder of how profoundly the quality of the fencing technique applied may change the uses one can make of property. Fences can take many other forms as well: guard dogs and physical surveillance; tagging of animals in free roaming herds. Vending machines act as fences. The GPS system permitting instantaneously to locate cars contributes to fencing them in against theft. Doctors, lawyers and other professionals provide information only to paying customers (unless they work pro deo), thereby solving the fencing problem. The cost of the fencing technique is part of the cost of using the property. As the former go up in relation to the latter, it may no longer be interesting to use an existing fencing technique and one may have to resort to a simpler one. In cinemas, seats are no longer individually assigned; each viewer finds a seat on a first come, first serve basis. But for theatre, concerts and opera, individually assigned seats – and the ushers to guide you to them – are still viable. Salt and pepper, once very dear commodities, are no longer individually rationed in restaurants, but provided free with any meal one orders. This exemplifies a formula we will encounter often in intellectual property. Where it is not profitable to fence in an object by itself, the owner may yet succeed in getting paid for its use by tying it to a different commodity or combining it into a more expensive package – as in buffet style meals –which can be profitably fenced in. To put it differently, if a fencing technique no longer works so well, don’t condemn the property but change your business plan to collect revenue differently. The movie industry thus discovered that they could make more money from home viewing, initially regarded as piracy, than from cinema viewing, if they sold videos at $20 (rather than at $100, as they first tried). Fences are subject to obsolescence or may become too costly to maintain for what they are designed to fence in. The obsolescence of a fence does not make the scarcity of the fenced-in commodity disappear. As the owner sees the utility or profit the commodity was normally expected to procure dwindle, he or she has an incentive to put in place a cheaper or more effective fencing technique to reverse the trend. The new fence is viable if its cost (capital and running cost) is more than offset by the pilferage it avoids in comparison to the older fence. The incentive on the owner to put in place a better fence carries over to entrepreneurs who invent and market new fencing techniques. Fencing techniques may develop into an industry of their own, bringing considerable ingenuity to bear on them. Fences need not be fool proof; some pilferage can be tolerated so long as the owner can draw enough use or revenue from the fenced in object. House ownership does not become unviable because of the risk of a break-in; but it might, in areas exposed to repeated looting. In these examples, the responsibility for creating viable fences falls to the owner. This is properly so. The state’s role is generally limited to preventing outright violence and fraud. Further extension of its role opens the door to rent-seeking, that is the processes whereby citizens succeed in mobilising state power to get revenues or other advantages which they could not get in a market (rents). The owner’s responsibility for fences puts the incentive to erect the best fence which pays its way, or to abandon particular forms of exploitation of the property, on the person most closely affected by the decision. The law plays into this logic by accepting that persons may appropriate unowned objects if they can come up with a fence for them. Possession is the root of title and, conversely, protection is afforded only to people who have taken sufficient care to fence in their property, such as locking their car. In appropriating new objects, you may not interfere with already existing property rights.
C. Common property
Not all commodities subject to property are individually owned. Some are owned by groups. It might be thought that this would lead to a tragedy of the commons, but in a classical study on Governing the Commons, Elinor Ostrom has shown that common ownership is practised in a wide variety of formulas in all regions of the world, for instance amongst farmers letting their cattle freely roam in the high Alps. Nearer to us, condominia are examples. One set of circumstances under which common property appears the preferable option is where fencing in property against other users within a community is too costly, but fencing the property of the entire community against outsiders is viable. Economists then speak of club goods. To avoid a slide into a tragedy of the commons for common property, a set of rules for the governance has to be in place. They determine under what circumstances community members may use the common resource. The simplest rule is equal access for all, but many other rules are possible. The limits of use will be set so as to avoid the exhaustion of what is scarce in the resource held in common. The rules will have to provide for supervision and for sanctions against those who transgress the rules, ranging from disapproval to exclusion to black listing, flaming and worse. Further rules will have to specify under what conditions persons will be admitted to the community and under which they can exit. Finally rules will have to be set for collective decision-making concerning a change of the rules or of the use of the common property. It should be stressed that common property is resorted to not only because individual fencing is too costly, but for other reasons as well. Sharing certain facilities may be a way of increasing the value of individually held objects, as it is in the case of condominia. Indeed, if one thinks of common roads as a form of common property, some common property may be an essential condition for the viability of individualised property. It is sometimes argued that sharing has other virtues as against individual property, in particular that it is more conducive to creativity and to living within one’s community.
II. Common features of Intellectual property rights
If intellectual property is akin to property in tangibles, these features should carry over to intellectual property rights. The thesis appears at first blush to be plausible enough. But intellectual property rights present a special twist because their objects are intangibles, more precisely information structures created by human ingenuity and effort. It is instructive first to look at information as a special commodity, to understand what adaptations the property rights logic requires to cope with the peculiarities of information (A) and then to examine some special features of intellectual property rights (B).
A. Information as a special commodity
In its broadest sense, information is the basic ingredient of all human decisions. You are informed if you learn something you did not already know; otherwise, what you learn is redundant. What you learn may help you make decisions more confidently or differently. Information is subjective: What informs you may be redundant to me and vice versa. This means that where decisions must be made affecting lots of individuals at once, be they what objects to manufacture or what services to provide for our collective welfare, arrangements must be put in place through which individuals can reliably reveal their preferences. Markets are such an arrangement; in various social contracts, procedures have to be invented through which individuals can voice their opposition to particular plans (or their support for them). Information is a very general phenomenon. We express our thought by means of information structures such as natural language and more specialised languages for our field of knowledge or trade; our culture is one complex set of information structures; so is scientific knowledge or news published in a variety of ways. Much of this information is generated almost unwittingly, as a by-product of daily activities undertaken for other purposes, and does not appear to require special encouragement to be undertaken. Whatever is generated automatically appears to be abundant in the sense used above and can be left to flow freely. But some information clearly would not be forthcoming without special effort by particular persons to create it and the corresponding incentives. One immediate conclusion is that for information, unlike tangible property, a single formula will not fit all forms. We should not be surprised to find that information requires a great diversity of rules corresponding to wide scala of contexts where it is used and exchanged
B. Encouraging creation
Innovators may earn a reward by being the first to bring an innovation to market (lead time giving a head start). But this may not be enough, in the eyes of some. Where the creation of information needs to be especially encouraged, various avenues have been tried in the course of history. Different techniques may be apposite depending on the type of innovations to be simulated: scientific and engineering inventions, musical and artistic creations, exploits of various kinds. Here is a list of such stimulative techniques: grants, subsidies or stipends; prizes and medals; lotteries; private employment, sponsorship or pensions for artists or engineers; monopolies for teaching trades or crafts; procurement contracts, in particular for military developments; and finally, intellectual property rights, tentatively from the late Middle Ages on, and systematically as from the 18th century, with international treaties consolidating the movement across national boundaries from the late 19th century. Why this last institution, now the dominant one, has been put in place so late in history is itself a fascinating question, which will not be addressed here. If we knew precisely what we wanted to develop and who could do it, we might well be indifferent amongst these various means of encouraging creativity. You contract with a builder to erect the house of your dreams precisely as you want it. If we had all that information, a central planner could bring about the right kind and amount of creativity for a society. But in fact we do not have that information. We did not, for instance, know that the system of easy communication amongst scientists had the potential of becoming the world-wide universal communication system that the Internet now is. We did not know that it would be profitable to put all communicable information in numerical form. Where creativity will take us is by and large uncertain, the fruit of entrepreneurial gambles. The question we then face, in comparing institutions for encouraging creativity, is which will get us, through a trial-and-error process, the most creativity that is appreciated by citizens at large. We cannot take for granted that central decision makers, even well-meaning, will know what is good for citizens (paternalism). History has shown all too often that they might confuse it with their personal interest and that societies relying of such a formula tend to fall behind others in the world economy. What we are looking for is an encouragement system granting rewards in a decentralised fashion and in direct proportion to the utility citizens draw from the creation. Formulated in this way, it is obvious why a form of property right would appear to have the advantage: it is available automatically (at least without discrimination or paternalism) and gives the rights holders a direct stake in convincing the community of the quality or usefulness of their creations; the better the creation, the higher the reward. This lesson acts prospectively to attract people into such activities. This suggests that we should transpose the property rights logic into the field of creation. The incentive and information effects of property rights, and their ability to quell conflict would appear to be desirable qualities in this field as well. But since only some creations require special encouragement, the property logic should apply selectively. Affording rewards to activities that would be forthcoming without them will encourage rent-seeking, which is entirely undesirable.
C. Public good qualities of information
For property rights to be viable, the prospective object must lend itself to being reserved to the owner. With information this is not to be taken for granted. The secret you share with a friend is out of your control and may become common knowledge shortly. After the Napster experience on the Internet, those who had any doubts will now be convinced that the copy of the music you ‘bought’ can readily become the source of many copies of equally good quality elsewhere. The intellectual content does not appear to be naturally scarce. And the cost of copying is falling as we discuss, and has become to all practical intents negligible. This creates the temptation for some to free ride on the creative efforts of others. How then will the latter be paid for their efforts? There appear to be problems with the exclusivity of information. A further peculiarity of information is that its use by one person does not preclude use by others. Unlike the apple you are now eating, the music you copied in MP3 format and now listen to is still available to the original holder to listen to as well. This is true for most forms of information, including culture, various skills, scientific knowledge. It does not work for stock market tips and other information where it is important to be the first to use it, even as the information will become publicly known subsequently. Trade secrets are useful precisely because competitors do not have access to the information. But these ‘early bird’ forms of information would appear to be the exceptions. Most information is what economists call non-rivalrous: it can be used without diminishing its utility for others. These features are important because of a debate in the 20th century economic literature dealing with the proper tasks of government. The initial conclusion was that some tasks in the economy must of necessity be assumed by government because they will not be undertaken by private actors (‘market failure’). One of these was the provision of so-called public goods, which have precisely the two characteristics of posing difficulties of exclusion and being non-rivalrous, while yet not being abundant, that is available without anyone’s effort and consumable at will without exhausting supply. This led some authors to conclude that the creation of particular forms of information requiring ingenuity and effort should be undertaken by government or with government help. The argument would extend to such forms as scientific information and consumer protection information. One form of help is the creation of rights on intangible objects that are not naturally scarce. The rights create something of a minor monopoly on the creation in question. How serious the monopolising effect is depends, of course, on what consumers will accept as substitutes for the work in question. The conclusion that the provision of information necessarily calls for government intervention may be premature. We should draw here a lesson of caution from the property rights literature and conclude that the problem of fencing – and hence the possibility of excluding non-paying users – falls to the owner and let the system run its course. One may expect much experimentation with new fencing techniques as the new technologies make their way into general usage. Admittedly, basic scientific research is widely supported through government grants, even though universities, where much of it takes place, also get monies for it through funding campaigns. But human ingenuity has come up with a variety of tricks to ensure a modicum of fencing for information. For instance, consumer protection groups sell their expertise as books or as part of a general subscription that also entitles one to emergency services when one’s car breaks down on the road. The software industry, having discovered that copy protection did not go over at all well with consumers – at least in North America – has changed tactics. Regular and lower-priced updates and tying arrangements like on-line assistance available only for legitimate clients of software are used to bring customers into the paying fold. The division between what is product, what advertising may thereby change. Many software companies make ‘lite’ versions freely available on the Internet, but require you to subscribe or pay otherwise to get the fully featured version and access to updates. They enforce their copyright on software – and collect payment, conceding flat fee site licences – against large companies who are solvent and can ill afford the bad publicity of being branded as pirates, leaving some private copying by individuals alone. Shareware becomes viable when firms such as Kagi develop methods to collect small fees economically and to encrypt software so that it can only be opened by paying users. Most recently watermarking and encryption have entered the scene as fencing methods. Contractual arrangements too may act as fences: you agree to give someone access to your technological know-how, specifying in the contract what measures he or she needs to take to keep it secret and how to return it to you, should either of you wish to end the relationship. You may do this on a reciprocal basis in a joint venture. The law, in the form of actions against violators of intellectual property, backs up these initiatives, but remains complementary to efforts by the interested persons themselves. Your business plan should direct you to collect revenue in places where you have a realistic hope of creating effective fences. Copyright distinguishes between artistic, dramatic, literary and musical works. The distinction may be explained as based on the different fencing techniques used in the earliest days of copyright for the creator’s control over the work.
D. The cumulative nature of creation
If this were the entire story of intellectual property rights, we would do well to extend them as far as possible, allowing only for some legal back-up to help with the policing problems just discussed. The extensions go in three directions: the conditions for getting a right in the first place, the time for which it is afforded and the types of information structure for which it is available. Rights in tangibles are available upon legitimate acquisition, without limitation in time and in virtually all objects (save those reserved for public use). For intangibles, the story is more complicated since virtually all creations build on earlier ones. Opening the door more widely for the protection of existing creations will make it more difficult to produce follow-on and altogether new creations, which nonetheless build on existing ones. In our desire to support current creators, we may foreclose openings for future creators and reduce overall creativity. Creativity requires direct encouragement but also a large public domain from which one can liberally draw ideas. This explains why, in intellectual property law, pure ideas and principles are left in the public domain, as are collections of facts. Where particular ideas can only be expressed in one or a few ways, that expression is open to free use by followers. American law uses the expression ‘scènes à faire’ for it. Similarly, the graphic user interface for micro-computers, once jealously guarded by Apple as its exclusive preserve for the Macintosh, is now considered part of common culture and no longer separately protected. The foreclosure may come about because current holders refuse permission to use their creation, cannot be traced or set their prices prohibitively high. ‘Information lock-up’ is one name invented for this phenomenon. To prevent it, intellectual property rights are granted only on objects for which creativity needs to be especially encouraged, in circumstances showing some form of creativity and for limited periods, deemed sufficient to produce the incentive for creation. One might further expect that rights would be granted and renewed only upon request. These background principles, but for the last one, appear to be borne out in intellectual property law. As to the latter, copyright is available upon the mere fact of creation, without a special request to get it; all other rights have to be applied for. With respect to the first three principles, all intellectual property rights, by varying duration against breadth of coverage and ease of obtaining the right, show an implicit trade- off between benefits in the form of encouragement for creation and costs of the monopolising effect on further creations. Patents are granted for a comparatively short period (20 years) and under demanding conditions of novelty, non-obviousness and practical usefulness (rather than mere theoretical interest), but give wide protection against products with the same or similar functions. Although 20 years may seem long in a world of rapidly changing technology, copyright, by contrast, is granted for a very much longer period and at relatively modest requirements of originality, but essentially protects against copying only and not against the borrowing of ideas underlying the creation. This suggests an optimality calculus such as economists are fond of designing. What models have been designed suggest considerable complexity in building in various forms of costs, depending on the nature of the industry and the type of innovation it calls for [Gallini & Scotchmer 2002]. Whatever the potential of this literature, we have currently no assurance that the particular forms in which intellectual property rights have now crystallised are necessarily optimal. What evidence we have of streams of revenue flowing from copyright acquired today suggests that for the vast majority of creators the streams will have dried up well within their lifetime. Scherer  observes that rewards for inventions and creations are extremely skewed: a few famous inventors and artists are big- time winners and draw huge revenues; for the vast majority the revenues are modest. The business of invention and creation is a highly risky lottery. But this very gamble on fame and revenue may be what is needed to attract creators into the creative activity. For large corporations, which prefer better calculable risks, this is not a desirable profile. They may prefer to sign up artists once they have reached some public recognition or, where it concerns industrial creations, buy out companies (start-ups) after they have reached some initial success. One may wonder what these observations teach us as regards the duration of intellectual property. For copyright, for instance, the term, in the United States, has been progressively increased from 14 years in the late 18th century to the author’s life plus 70 years currently. It is not at all obvious that these changes correspond to observed deficiencies in copyright to draw sufficient talent into the type of creation covered by that right. The recent extension of the copyright term from 50 to 70 years after the author’s death appears to be a clear instance of existing rights holders succeeding in extending a flow of revenue beyond what was necessary for those creations in the first place: a windfall for them; rent-seeking as regards society as a whole. Similarly we have no assurance that the extension of full length copyright to photographs, to derivative works (a film made out of a novel), to prohibit parody, to neighbouring rights, and indeed now to any activity on the Internet (since all of it requires one to make an electronic copy) is beneficial from the point of view of overall creativity or welfare. And similarly for the extension of patent to life forms developed as a result of bio-engineering, to software and to business methods. As regards high prices set by holders of intellectual property, it might be thought that compulsory licensing would provide the answer. Yet this solution goes against the very logic of property rights: use is granted on conditions set by or agreeable to the owner and the owner alone lives with the gains or losses resulting therefrom. Decentralised incentives come at that price. Competition is a better answer to high prices, but it works only in as much as creations are not so unique as to be virtual monopolies in their area. The cumulative character of knowledge suggests that the law should generally be accommodating to follow-on innovation and to reverse engineering to discover the working principles behind particular inventions. This appears generally to be borne out. Semiconductor protection legislation explicitly allows follow-on inventions and grants protection to them if they make an original contribution beyond the state of the art. One practice that may come in the way of building on to earlier knowledge is ‘blocking patents’. When the use of several patents, possibly in the hands of different owners, is required to build a complex object, any one patent holder is in a hold-out position and may block progress. Recent literature has invented the term ‘anti-commons’ for such situations. It is difficult to counteract the practice without undermining the logic of the intellectual property right itself, but in some instances, one might attack it through competition law or through doctrines such as misuse of rights. An answer to such reproach would be a showing that patents (or other intellectual property rights) have been placed in a pool from which they can be borrowed for a set fee or are made available through collective rights organisations. Merges [1996, 2001 and in earlier articles] has shown that such arrangements are quite common. Whilst the owners have to content themselves with the set fee, collecting money that way, by reducing transaction costs for prospective customers, may be a better business plan than attempting to collect individually and holding out for payment. The cumulative nature of knowledge and technical development is also evident in standards and in so-called network effects. Standards are norms which allow people to communicate and coordinate their plans, and to make equipment, services or software compatible or ‘interoperable’. Whoever has attempted to hook a portable onto the telephone net in different European countries realises the gains to be had from using a single plug system for telephones as well as for connection to electrical mains across all countries. The phenomenon has very widespread application in all walks of life: weights and measures, rail gauges, headlights and tyres in cars, the coding of television signals and recorded music etc. In most cases, persons concerned can arrive at agreement about the standard to be set, sometimes with the help of a national standards organisation, and the standard is then released into the public domain for everyone to use. In some cases, matters are more complicated. Where a complex product developed by one manufacturer and subject to an intellectual property right becomes the de facto standard, the manufacturer has to walk a delicate line combining the desire to make money on a product by exercising the exclusivity of his intellectual property right with the need to have it widely available as a standard. Practical problems arise with respect to such standards as the programming language Java, the operating systems Windows and Linux, and Sony’s game consoles. The dictates of the property logic come into conflict with the need to curtail monopolistic practices through competition law. Wide availability does not mean free availability, but it does mean full access to technical details for industrial players, be they competitors. Network externalities arise when the utility of a product to a user increases with the number of others who use it as well. In the early days of the telephone system, the more people subscribed to your telephone supplier the more use you had for your telephone (more people reachable) and, if you are a manufacturer of equipment, the more it pays you to develop equipment that can be hooked up to that system. But the supplier of the service would rather not make the service compatible for outsiders: it confers benefits of his efforts onto competitors. Network effects, driven to their extreme, lead to only one supplier for all customers, creating the appearance of a monopoly. This tends to lock customers into an existing technology, even where a better product is available. Everyone stays with the old system and watches the neighbours. But notice that no one is prevented from entering the market. If a change- over does come, it is likely to be quick and drastic: the network effects work as much for the newcomer as for the incumbent. Clearly, judging whether intellectual property is being abused and gauging the degree of competition in markets characterised by network effects requires considerably more subtlety than measuring market share.
The broad features of information highlighted here – much of it is created anyway, but some needs special encouragement; it is often subjective and must be revealed through market purchases or voicing and other procedures; it is hard to keep exclusive and can often be shared without loss of utility; creation usually builds on earlier creation, which requires fairly easy access – appear to explain some general characteristics of intellectual property rights such as the absence of protection for ideas and principles, or for facts; the trade-off between duration and breadth of coverage and difficulty to obtain; considerable experimentation in the field with new fencing techniques such as encryption and watermarking, but also commercial and contracting practices.
III. On specific features of some intellectual property rights
Some features of intellectual property rights call for specific remarks. We discuss fair use and moral rights.
A. Restrictions on intellectual property rights
In all enactments of copyright throughout the world, some uses of copyright creation are free as being in some sense in the public interest. In American law, for instance, fair use, or fair dealing, as it is called in legislation of British inspiration, is a defence to copyright violation holding that the practice in question is freely allowable as an exception. A reviewer may quote short passages of the work reviewed; a software licensee may make a back-up copy; news may be freely reported. Most national legislation allows for a limited list of such acts, the scope of which may vary from country to country. The American Copyright Act of 1976 codified earlier case law in creating an exceptionally open-ended notion of fair use, which has been broadly interpreted by the courts directed by the act’s provisions to look at the nature of the work and purpose and amount of copying, and, in particular, at the effect the copying is likely to have on the copyright holder’s market. The discussion is consequential because, whereas in traditional media, copying was controlled but use of existing copies free, on the Internet any use implies copying. Economic analysis can suggest various rationales for these exceptions. The most banal one is that particular interest groups have got an exception in their favour. Traditional exemptions for playing music in churches and at fairs are of this sort. They amount to rent-seeking. The traditional economic rationale for fair dealing points to high transactions costs. Free use is allowed as fair where the user faces insurmountable costs to track the copyright holder for permission. Better then to allow the use: increases welfare for at least the user, without depriving the copyright holder of revenue that would not have been forthcoming anyway. Technical developments may change the persuasiveness of this argument. As collective rights organisations are set up and the Internet and database technology lend a hand to tracing authors and communicating with them or with agents acting for them there should be no obstacle to paying for use, if the work is granted copyright in the first place. An extension of the argument is to assume that, if contacted, the copyright holder would have granted permission anyway. This presumably justifies singing popular jingles around summer camp fires without paying royalties. A second rationale is that the particular use follows logically from the (licensing) relationship the copyright holder and the user already have. Permission to make back-up copies of software for which one has paid is in the order of obligations logically implied by the licence. In a similar way one could justify the user’s time- or space-shifting: you pay (in some form) for music or video and then record it for enjoyment at a different time or place. You may be willing to pay more for objects allowing you to do this and industry may capture this potential gain by subtle forms of price discrimination. A different interest is at stake in the freedom of reviewers and critics to cite what they review or criticise. Restricting this freedom opens the door to censorship or to complacent reviews, in which consumers could have no confidence. The producers of cultural products as a group as much as their customers have an interest in a non-partisan reviewing process, for the same reason that product labels and trade-marks are valuable: they facilitate the customer’s decisions about which good and services to acquire. Putting this right into legislation cuts short the temptation for individual producers to tilt the reviews in their favour, thereby creating a sort of tragedy of the commons. These concerns reach into the realm of freedom of expression as a fundamental right in most countries. Most recently, fair use has been justified as avoiding an anti-commons resulting from hold-out behaviour by copyright holders, even where transactions costs are low. Normally, competition would be the answer to this argument. A further response would be to point to the existence of collecting agencies holding the right in their portfolio. How serious the hold-out problem actually is is hard to gauge. At all events, the fair use exception for research and scholarship might be justified this way, considering the cumulative nature of knowledge discussed earlier. But this would argue merely for assured access, not necessarily for free access. Economists would point out that the use that is apparently free because fair, will in fact be paid for elsewhere, as North American consumers pay for ‘free’ television. Copyright holders can achieve this through price discrimination. Libraries pay a multiple of the price for individual readers for their subscriptions. The multiple covers the ‘free’ copying allowed within the confines of the library. Are the dimensions of fair use a matter of public order or policy, not to be set aside by contracts clauses in the licence through which users have access to copyright products? Should they be maintained in the face of the extensive control copyright holders are promised with the introduction of digital rights management information (DRM) on all products in numerical form? Defenders of DRM contend that the micropayments required for each use will only serve to tailor payments more closely to actual use, leading admittedly to price discrimination amongst consumers, but if anything to wider and easier access for them. As to other restrictions to fair use by licence clauses, they should be dealt with under the rationales sketched above.
B. Moral rights
Copyright holders have, besides their economic rights designed to procure them royalties directly, various moral rights which are thought to enable the author’s personal mark upon the creation to be preserved. Traditionally these rights encompass the right to publish (or not), the right to be recognised as the author (paternity right), the right to have the integrity of the work respected and to control its association with particular products, services, causes or institutions (integrity right); finally, the right of regret, leading to the withdrawal of the work from public circulation. Moral rights are considered to be akin to personality rights in one’s image and for that reason, are made non-transferable and non-waivable in countries such as France. The United States have been notoriously opposed to the introduction of moral rights, but it would be mistaken to consider them specific to civil-law systems. Canada, for in stance, recognises the right of integrity and of paternity in its copyright act. Moral rights, contrary to the rhetoric used for them, clearly have economic effects. The authors’ reputations act as quality signals to consumers interested in their works. For authors of some success this is valuable human capital. This human capital has commercial value in as much as it attracts consumers to buying (more of) the products or services of the author. The property rights logic suggests that the right to manage one’s name (commercial image) and the portfolio of works associated with it is best left with the author. This argument loses none of its strength were authors to use their moral rights in the pursuit of glory or in a desire to communicate particular ideas they hold dear. Values are subjective, as we saw above. For the right to paternity and integrity, this argument is straightforward. In common law jurisdiction protection may be afforded these interests by general law, such as the interpretation of contracts or by tort doctrines such as passing off. For the right of withdrawal, it is more complex where one wishes to exercise it after having given permission to publish. Clearly the publisher with whom one has contracted should be entitled to damages, lest the author exercise his moral right whimsically, creating a form of moral hazard. The right here takes on the form of a liability rule: you may exercise it, but subject to compensation as circumstances dictate. Why should these rights be inalienable (non-transferable) and non- waivable? It prevents authors from getting into slavery contracts, but also in ghostwriter agreements where they entirely eclipse themselves. This may not be as damaging as it sounds since one’s reputation amongst insiders may nonetheless be built up. But in fact the ghostwriter authors appear to insist on having their name recognised, even where legislation does not mandate it, where one sees Memoires by X ‘with Y’. For more industrial- style copyright products, such as software, the prohibition to waive may complicate transactions. This is presumably the reason why some national acts – such as the Canadian – prohibit sale of the paternity right, but allow it to be waived.
Trade-marks differ from the other intellectual property rights in that their object is not something in which consumers would be interested in its own right, but rather as a step in acquiring some other good. Trade-marks act as if they made invisible qualities of commodities visible and thereby facilitate decision making by consumers in the market. In a sense, the author’s reputation, controlled by moral rights, act like a trade-mark. For goods like apples, grapes or cherries, one can upon inspection or sampling (a grape) determine the main qualities of interest and decide on how much to buy. For many goods, by contrast, simple inspection will not tell the consumer enough. The relevant qualities will only reveal themselves upon experience. This will make the consumer hesitate before purchasing. In the economic literature this has been illustrated by Akerlof’s image of the market for lemons. On the consumer side of the market for used cars, customers expect, prudence commanding and in the absence of better indications, to run a serious risk of buying a lemon. This reduces the price they are willing to pay for the cars on display. To speed up transactions, suppliers of better quality cars therefore have an interest in convincing consumers of those qualities by signals that are indicative of them but would be hard for suppliers of lesser quality goods to imitate. One may offer warranties which would be too costly to honour for lesser quality cars; try-out periods or ‘lite’ versions (practiced for software); return privileges and so on. A complementary tack is to rely on repeat business: you convince consumers a first time and assure them that you maintain quality standards for future occasions. As consumers buy and try out your goods repeatedly and spread the word, this policy may be vindicated in their eyes and you create good-will, which translates into more transactions (lower search costs for consumers) for the future. If you can no longer know all your customers personally, it pays to associate your policies and practices with a sign which consumers readily recognise. You can now use the sign as shorthand for the qualities you have consistently offered to consumers. The trick is to avoid that this quality sign is watered down through misuse by competitors, lower quality or not, or becomes diluted into general usage. Use by competitors is a form of free riding and if left unchecked, would tend to undermine the entire system of conveying information through the marks. The law is generally severe with free-riding and here has combated it initially through actions in passing-off or ‘parasitical acts’ or more broadly by sanctioning unfair competition. As the institution proved its usefulness, it was formalised directly as a right in the trade-mark, actionable upon a mere showing of unauthorised use by a competitor. Trade-marks are interesting in that they show a way of selling information, which is hard to fence in. It is tied here to commodities, over which control is easily ensured, but the entire scheme must be ensured against pernicious deflation. The arrangement underlines the essential role of the law in preventing gross forms of free-riding which would deflate the information system to the detriment of trade. It is sometimes argued that the trade-mark system allows merchants to introduce artificial distinctions, for which consumers pay without getting real benefit (brand inflation). The answer to the argument is to look for situations where trade-marks are unavailable and gauge consumer reaction. Some time ago, super markets in North America offered ‘no-brand’ but cheaper products to consumers, side-by-side with brand name products. The experience appears not to have been very successful and in most cases, the no-brands were discontinued or occupy only a small part of the shelves. Brand names, one must conclude, usually offer consumers information of enough value for them not to want to go without it.
IV. The extension of property rights to new objects
For the objects of the physical world we worked out a well-defined property regime. Immoveable property has one regime, moveables have another, with specific variants for a number of registered goods such as cars, aircraft, ships. Even here new developments occur, as when people wish to own apartments within large buildings, or when the invention of aircraft and of electricity obliges us to specify how far into the sky the owner’s rights stretch. For intellectual property rights, the problem of how to extend the existing order to new objects is more acute because scientific advances show us all the time new objects or new uses for existing ones. When we learn how to transplant bodily organs, they become scarce in the economic sense and we are under pressure to come up with a property rights regime for them. When the mnemonic system developed to help us remember web site addresses turns out to have great appeal to commercial actors on the web, we are under pressure to come up with property rights to determine who can have which and avoid a pointless cybersquatting race.
A. Rights vs rents - the puzzle
It is tempting to think that this task falls properly to the legislator. Most existing intellectual property rights are codified in legislation. But the answer merely displaces the problem: why did we resort to legislation when these rights were first enacted; how did the legislature come by the information required to legislate in proper fashion? The legislature is the institution brokering compromises amongst opposing interest groups. It can act relatively quickly and create apparent certainty for a sizeable field at once. If you have great financial stakes where the law is fuzzy, as the amusement industry feels it has with regard to activities on the internet, a reduction of the uncertainty translates into substantial additional revenue, which justifies a lot of lobbying expenditure. On occasion, where a concentrated interest group is opposed to diffuse and unorganised groups, this may lead to skewed legislation favouring one group at the expense of the diffuse group. The economics of public choice has much to say about this phenomenon, which is a form of rent-seeking. The American Digital Millennium Copyright Act and the legislation extending the copyright protection term to seventy years beyond the author’s death are recent examples. Such legislation is short on legitimacy because it appears not to incorporate solutions with which all parties concerned have expressly accepted to live. One response to this difficulty is to invite the courts to invalidate as unconstitutional legislation that is so clearly one-sided as to amount to rent-seeking: concentrated benefits, widely dispersed costs. Whether courts will accept so political a role is a moot question. Another response is to turn to economic modelling to design substantive rules which are in some sense optimal and avoid the rent-seeking trap. Gallini and Scotchmer  survey the literature engaged in such efforts. Their survey does not show that this literature can as yet reliably deliver the design services hoped for. A further response, argued for here, is to turn from substance to procedure and to look for a decentralised discovery process whereby interested persons themselves work out the rules by which intellectual content they develop is to be governed. To guarantee its legitimacy, the process should be sheltered from the one-sidedness that characterises rent- seeking and not lead to a cartel-like conspiracy against the public.
B. A discovery procedure
Such a process has been at work in France leading up to the enactment of a plant breeder’s right in 1970. Here is how it worked. From antiquity until early modern times, experimentation with plant breeding had been the work of monks and later of individual farmers and isolated amateurs; know- how passed on from father to son. By the end of the 19th century scientific knowledge in plant breeding had sufficiently progressed to make applied research an interesting prospect. What was then needed was a regime to ensure revenue for people working in the laboratories, preferably in proportion to the utility their inventions procured to persons in the field. In the era of international conventions regarding copyright and patent, the idea arose quite naturally to ask French parliament to enact a horticultural patent. The demand was formulated in 1904 and after much debate, rejected in 1920, on the ground that nature could no more be appropriated than the air, the sea and the water flowing in the rivers. This was not the end of the story, however. The interested farmers used associations already in place (such as marketing collectives) or created for the purpose to set up laboratories for applied research. The details are not important here; what matters is that contract and association rules were sufficient to accomplish this. The laboratories were paid for from membership fees and through usage fees collected from those who would use the plant varieties they developed. The farmers who were part of this scheme undertook to maintain quality standards (to maintain the quality of plant varieties); not to transfer any element permitting reproduction to outsiders; to destroy their plantation should they leave the regime; to allow inspection of all of the foregoing; and to pay royalties for the plant varieties they drew from the pool. Exchanges took place between the various regional associations. The whole system worked satisfactorily so that by 1970 it could be transformed into a formal plant breeder’s right. The interesting thing is to see what this means in terms of property rights. Property rights, we saw earlier, essentially require some measure of exclusive control for the rights holder and consist in exercising three prerogatives: to decide on the various uses of the object of one’s right; to collect revenues or absorb losses that flow therefrom; to transfer the entire bundle to another person or group. If one can ensure oneself the exclusive control, the three prerogatives can be ensured through contract, bilateral or in the form of rules of an association of which one is a member. In the plant breeder’s example, control of the seed stock is sufficient; all arrangements are worked out by means of contract. At the extremities, one needs some help of the courts in stopping forms of free riding, where a member secretly transfers seed stock to outsiders. It is hard to trace the leak. The courts can help the process along by sanctioning, upon being shown how the system works, the possession, in the hands of the outsider, of seeds which must necessarily have been leaked. Competition authorities, furthermore, must not attack the system as a cartel seeking to restrict output and raise prices. But these forms of intervention merely ensure the outward perimeter of the process and do not determine its substance. If all of this works, we have in place what functions like the beginning of a property right. The process allows experience to be accumulated with how it works, possibly leading to codifying it as a distinct property right in due course. One might summarise it as: control+freedom of contract+leakage control by courts = prototype right. The process is decentralised and under the control of interested persons. It presupposes the rule we encountered earlier : build your own fence. It is easy to imagine extensions whereby members of the group agree to contribute their inventions to the common pool and make them available to other members. The arrangements worked out amongst such players who are now inventors, now borrowers of others’ inventions give us a view of which borrowings should be paid for, which should be free. They are entered into as it were ‘under a veil of ignorance’ in which one does not know in advance on which side of the borrowing fence one will find oneself and hence would have an interest in developing ‘fair rules’. For these reasons, a discovery process of this kind has a strong claim to legitimacy and is a proper source of inspiration for legislation that may follow.
C. Discovery procedure and legitimacy
My contention is that this pattern has in fact been followed in a number of circumstances. Trade-marks developed historically from protected interests into fully-fledged rights through such a process, with court help in the form of ‘passing off’ decisions, streamlined in due course with simplifying presumptions and better tailoring of the appropriate sanctions (injunctions as against mere damages). The domain names appear to be going through a similar process. ICANN and other domain name granting organisations set the rules for obtaining the names; the rules ensure that conflicts with existing property rights are resolved (as the logic of ‘build your own fence’ requires if there is to be a coherent property order) and courts stop free-riding by sanctioning cybersquatting. In recent debates it has been contended that legislation should reserve a large place for the share economy, that is the free use of content subject to an intellectual property right. Sharing is thought to be essential for creativity. For this claim too, one may want to require legitimation through a discovery process such as the one sketched above. Here again we can point to experiences that appear to make that demonstration. GNU/Linux is a very successful operating system, which has been integrated into Apache server operating systems. It consists of 500 M lines of code and can draw on 400 K analysts world wide to correct its bugs. GNU/Linux is based on a sharing model, anchored in copyright. The copyright holder provides universal access through a general public licence (GPL), stating essentially that everyone may freely use and improve (providing the improvements are made available on the same terms, i.e. rendered to the pool), but may not appropriate and make it into an exclusive product. Versions of the program may be incorporated in packages sold commercially, but the program itself must be available free. How are creators rewarded here? The answer appears to lie in reputation as a programmer, either within the community of developers, or directly in contracts with clients. IBM and Oracle back the product on the consideration that its quality is so great as to enhance their own credibility in providing hardware and installation services. A different area in which the share economy makes its mark is in scientific exchange. Besides the trade circuit of the commercial publishers who offer costly journals and books, there is a very extensive system of papers available at cost or even free of charge, in particular on the Internet. In both cases, existing tools (here copyright and contract) are used to craft a usage regime which satisfies the persons concerned and has easy sharing qualities which the trade model does not directly offer. The experience is at once legitimising and a source of discovery of new ways of doing things. The discovery process appears here as a benchmark for proper law. It does, however, take time to unfold and players with high stakes may be tempted to shortcircuit it in an (illusive) quest for quick certainty in the law.
It is one thing to understand the decentralised procedure by which we may discover rights to be enacted. But how to rid society of enacted rights that have outlived their utility? Rights, as other institutions, tend to have a stickiness to them. Once in place, they are hard to remove, in part because people adjust their investments and modes of dealing to them. They therefore stand to suffer instant losses when the institution is abolished. This prospect gives them a financial incentive to lobby against abolition. Theoretical argument alone is unlikely to carry the argument for abolition. In practice, competition here as elsewhere is a procedure through which to discover what arrangements have outlived their usefulness. Consider, by way of example, the evolution of the telephone system. In the 1950s, 1960s and 1970s, it was taken as axiomatic that it could only be run as a natural monopoly, which meant that telephone companies were granted total control of all equipment hooked onto the telephone network. By the late 1970s, the so-called interconnect market was opened up for competition, meaning that consumers could buy telephones, faxes and modems from third persons and hook them onto the telephone network. Merely lifting this small tip of the control network unleashed a wave of innovation and led to further restriction of the natural monopoly. One may wonder if the Internet and cellular telephony would have taken off as they have, had the natural monopoly structure of the telephone been maintained throughout the world. In the field of intellectual property, this logic would require freedom to contract around institutions that are thought to have outlived their usefulness. Consider for example, an institution usually mentioned in the same breath as moral rights, viz. the right for the creator of visual works to royalties in case of resale of his work, known in French law as the droit de suite. The resale right gives the creator a lottery ticket on his own potential success. It looks particularly rewarding for one who only achieves success late in life. But the coin has two sides. For the potential purchaser of work of unknown artists, who speculates on this success as well, it tends to make the work less attractive (reduced revenues) and hence it would restrict the range of unknown creators whose work the speculator will buy. Recall Scherer’s  observation on the skewness of (financial) success in creative endeavours. In practice, only a few countries recognise this right and so it is easy to evade it. In the countries where it is accepted, enforcement is spotty probably because of the transactions costs of tracing the creators. Yet easier communication over the Internet should reduce the transactions costs of enforcing this right. Is it worth doing so? The test might be to give artists the option of waiving such right and then after some time to measure what actually happens.
This paper has argued that intellectual property rights may be considered a species of ordinary property rights, but with a special twist because of their object, to wit information structures. Information is not naturally scarce and is frequently generated as a by-product of activities undertaken for other purposes and in no need of special encouragement. But where special encouragement is needed, we draw on the property rights logic to craft the requisite rights. They create an artificial scarcity (a mini- monopoly) allowing creators to earn a return on their creative investment in direct proportion to the utility they procure to other persons who are willing to pay for it. This exercise is not without its problems because of the ‘public goods’ nature of information, but here as in property logic generally it is wise counsel to let interested persons come up themselves with devices and arrangements ensuring control over what they want to claim as property, i.e. build their own fences. This logic must be carefully circumscribed because of the cumulative nature of knowledge: creations build on earlier creations; follow-on is essential. The wider the protection granted for already existing creations, the more the openings for further creations are thereby foreclosed. As intellectual property rights are enacted, the temptation is to think that all rights on new developments should be handled through legislation. But such a policy would have severe drawbacks because of the danger that interest groups will use it to seek protective legislation which can only be characterised as rent-seeking. To guard against this danger, the article drew attention to a decentralised process whereby the form of rights in new objects can be discovered. The process relies on the ability to ensure control over objects, freedom of contract and judicial help at the margin to curtail blatant free riding activities. Such a discovery process has in fact been followed in a number of instances and could serve as a benchmark for new rights in the trade world as well as for new arrangements in the share economy.
Boldrin, Michele et David K. Levine, Against Intellectual Monopoly, 2005 http://www.econ.umn.edu/%7Emboldrin/aim.html http://levine.sscnet.ucla.edu/general/intellectual/against.htm (State the case against intellectual property and point to a lot of evidence in support of it) Landes, William M. et Richard A. Posner, The Economic Structure of Intellectual Property Law, Cambridge, Mass., Belknap of Harvard University Press, 2003 (Summarises the ‘take’ of two of the founding fathers on intellectual property) Mackaay, Ejan, The Economics of Emergent Property Rights on the Internet, in: The Future of Copyright in a Digital Environment, P. Bernt Hugenholtz (ed.), The Hague, Kluwer Law International, 1996, pp. 13-25 (This article spells out the fencing logic in some detail.) Merges, Robert P., “Contracting Into Liability Rules: Intellectual Property Rights and Collective Rights Organizations”, (1996) 84 California Law Review 1293-1393 Merges, Robert P., Institutions for Intellectual Property Transactions: The Case of Patent Pools, in: Expanding the Boundaries of Intellectual Property : Innovation Policy for the Knowledge Society, Rochelle Cooper Dreyfuss, Diane Leenheer Zimmerman and Harry First (eds), Oxford, Oxford University Press, 2001, pp. 123-165 (Shows how widely patent pools are used in the USA, in ‘big’ as well as in ‘small’ industry.) Mokyr, Joel, The Lever of Riches - Technological Creativity and Economic Progress, Oxford, Oxford University Press, 1990 (a remarkable tale on how innovation has become the source of wealth in the West and how it differs therein from other regions of the world.) Ostrom, Elinor, Governing the Commons - The evolution of institutions for collective action, Cambridge, Cambridge University Press, 1990 (the locus classicus on common property.) Quaedvlieg, Antoon A., An Economic Analysis of Intellectual Property Law, in: Information Law Towards the 21st Century, Willem F. Korthals Altes, Egbert J. Dommering, P. Bernt Hugenholtz and Jan J.C. Kabel (eds), Deventer, Kluwer, 1992, pp. 379-393 (This is one of the earlier surveys in English by a European lawyer summarising the literature on the economics of intellectual property.) Rose, Carol M., “The Several Futures of Property: Of Cyberspace and Folk Tales, Emission Trades and Ecosystems”, (1998) 83 Minnesota Law Review 129-181 (This article, by an American property rights scholar, looks in a systematic way at how property rights principles carry over into cyberspace and ecology.) Scherer, F.M., The Innovation Lottery, in: Expanding the Boundaries of Intellectual Property : Innovation Policy for the Knowledge Society, Rochelle Cooper Dreyfuss, Diane Leenheer Zimmerman and Harry First (eds), Oxford, Oxford University Press, 2001, pp. 3- 21 (Scherer details for a number of innovative domains how the spoils of creativity are highly skewed: a few creators get most of the gains, while the majority are left with crumbs. This makes the chances of success hard to gauge in such industries as pharmaceuticals, biotechnology, information technology, music and film. The lottery quality may, however, be indispensable to attract creative individuals into inventive endeavours.) Teijl, R. and R.W. Holzhauer, De toenemende complexiteit van het intellectuele eigendomsrecht - Een rechtseconomische analyse, Arnhem, Gouda Quint BV, 1991 [The Increasing Complexity of Intellectual Property - A Law-and Economics Analysis] (This is to my knowledge the first book length exploration of the economics of intellectual property rights from an expressly civilian perspective. It examines in considerable detail the various ramifications of intellectual property, including its extensions such as the publishers’ rights and rights in developments of biotechnology. Unfortunately reserved to those who read Dutch.)