Universite de Montreal
 FACULTE DE DROIT  Centre de recherche en droit public

[Home Page]  
[Previous menu] 


W.T. STANBURY UPS Foundation Professor of Regulation and Competition Policy
Faculty of Commerce and Business Administration
University of British Columbia

[Draft; Subject to Revision]


Although it is intimidating, I must begin with George Priest's (1986, p. 21) observation that "in the current state of knowledge, economists know almost nothing about the effects on social welfare of the patent system or of other systems of intellectual property." What does one do? Stop immediately or soldier on in the hope of refuting the proposition if only in a tiny degree? Since I deal only indirectly with this issue, I shall carry on.

1.1 Information Technologies and the Digital Age

The new communication and information technologies (CITs) in what is now being called the digital age[2] present both threats and opportunities for the federal government. The threats include the following:

There will be increased pressure for more direct democracy [3] (instead of representative democracy, the Westminster model of which gives the Cabinet extraordinary power).

New CITs facilitate the formation and functioning of more interest groups. Rauch (1994) argues that the result is "hyperpluralism" which is producing "demosclerosis" in the United States.

CITs also facilitate the internationalization of interest group activities and may, as a result, pose new threats to the autonomy of nation-states (see Stanbury & Vertinsky, 1994).

Computer networks ("webs") constitute an alternative source of news/information about policy issues and government actions which is unmediated both by the traditional news media and government sources (Stanbury & Vertinsky, forthcoming). (The federal government employs more public relations and media relations people than any other entity in Canada.)

The opportunities presented by the new CITs include the following:

The federal government can reduce its funding (direct and indirect) of interest groups -- a policy which has become the object of considerable criticism in recent years.[4]

When a large portion of the population is on the Internet it will be efficient and effective to distribute government-created information through the net. It will also create a new channel of communication from citizens to government whether it is to supply data, file a tax return, apply for particular benefits or write to an MP.

The government's huge stock, and continuing large flow, of intellectual property is a potential source of revenue in the form of fees charged to commercial enterprise that wish to reproduce/distribute that property. This opportunity is not without adverse consequences as well, of course.

1.2 Focus

The focus of this paper is to determine, in the digital age, what should be the policy of the federal government[5] with respect to its own intellectual property which is or could be protected by copyright.[6]

This paper has two main components. The first examines various issues which arise in the course of developing a policy for the federal government concerning its own intellectual property which may be the subject of copyright protection.[7] This is done in section 2 where it is assumed that in the digital age it is still practicable[8] to protect a creator's (or subsequent owner's) copyright.

The second component of the paper examines the question of whether it is, in fact, practicable to protect intellectual property subject to copyright. I do this by analyzing a provocative article by John Perry Barlow (1994) in section 3.


2.1 Multiple Roles

While I focus on a normative perspective of the Crown's policy with respect to its own intellectual property which is or could be protected by copyright, it is critical to indicate at the outset that the federal Crown is a unique owner of such material:

The federal Crown is a major (the largest?) producer (owner) of potentially copyright-protected materials. This includes the laws, regulations and judicial decisions which are at the core of its functions but these are only part of the copyright material it produces.

It is also a very large user of copyright-protected material in carrying out its wide range of activities. In other words, it is on both sides of the market regarding copyright materials.

The federal government creates the statutes and regulations relating to copyright and thus makes the rules of the game for itself and for others subject to the authoritative decisions of the courts.

2.2 Government Documents as the Subject of Copyright Protection

For many participants in the conference, the issue of Crown copyright focuses on policy relating to the "primary sources of law": statutes, regulations (more generally, subordinate legislation), and case law.[9]

But these documents do not actually make a comprehensive picture of any of the many regulatory regimes created by the federal government. A sophisticated understanding requires access to at least the following:

guidelines issued by government agencies (available to regulatees),

internal policy directives (usually not available to regulatees),

ministerial policy statements[10], and

speeches of regulators and others having decision-making power.

Presently, it appears that the federal government treats the case law differently from statutes and regulations. It relies primarily on private firms to print and distribute court judgments (reasons for decisions) although copies of the original typescript version of judgments are available from the courts for the cost of photocopying. Legal publishers such as Canada Law Book do not pay for the right to reproduce these judgments which are subject to Crown copyright. On the other hand, the federal government itself publishes the text of laws and regulations. For example, new regulations are published in the Canada Gazette. New statutes are available on an individual basis from what used to be called the Queen's Printer, and these are periodically collected in bound volumes in the Revised Statutes of Canada.

In summary, it has been federal policy to permit private or commercial reproduction of the text of statutes, regulations and judicial decisions without payment of fees to the Crown as holder of the copyright. Thus while the Crown claimed copyright protection, it was generous in allowing others to make use of its copyright material in the form of statutes regulations[11], and court judgments.

In the digital age, when it is said that these (and other materials subject to Crown copyright), can be made available in electronic form "radically cheaper and faster," should the Crown change its policy? I suggest that the answer, as in many other areas of public policy, depends on the objectives the policy is designed to achieve.

2.3 Unique Creator/Owner?

Is the copyright material[12] produced by the federal government different from that produced in the private sector? Is the material itself different? Is the context in which it is produced different? The answer to both questions appears to be yes. More generally, is there a good case for the federal government to treat its own copyright material differently than would a private sector creator/owner?[13]

It appears that it is fairly easy to distinguish federal government-created copyright material from that produced in the private sector.

First, much of the federal government's copyright material is produced in the context of performing other important activities (creating the rules of the game for society through legislation and regulation and adjudicating disputes over various rights). The supply of such copyright material as statutes, regulations and court judgments does not depend-upon economic incentives created by the exclusive rights associated with copyright. [14] Rather,

Taxpayers are forced to pay money to cover the government's costs of hiring individuals part of whose output consists of a piece of copyright material.

They also pay for the distribution of such material within government where it is an input in the supply of other activities.

The government-financed producers of copyright material get paid (at competitive market rates) for their labour for certain, regardless of their output.[15] (However, it may be desirable for the government to offer the incentive of part of future copyright fees to employees/contractors to ensure greater efficiency in the production of copyright material.)

If the government seeks to maximize the economic rent from its copyright material, rather than licence its use for free, the result will not be an increase in the future effort to produce such material. Thus rent maximization by the Crown with respect to its own copyright in statutes, regulations and court judgments will not serve the social function associated with the creation and protection of the rights associated with copyright.

Second, an important part of government-created intellectual property consists of the text (or other form) of the laws, regulations, rules, guidelines, interpretations and so on that constitute the core of government itself. The point is that statutes, regulations and judgments are public goods in the technical sense: (1) consumption by one person does not reduce the amount available for others (non-rivalrousness inconsumption); (2) it is not practicable to deny the good to those who have/will not pay for it (subject to free-riders). While it is practicable to control access to the texts (in print or electronic form) of statutes, regulations and judgments, it is not possible to prevent people from gaining the benefit from the ideas (rules/reasons) they contain. Indeed, to try to do so would be inconsistent with the basic tenets of a democracy. Citizens delegate authority to their elected representatives to make laws/regulations for the common weal. Knowledge of those laws/regulations is essential for the system to work well (i.e., largely by means of voluntary compliance). The social benefits of the rule of law (effected by statutes, regulations and court judgments) requires the widest distribution of the content of the laws etc. This requires government subsidized production and distribution of the words that define the rule of law.

The fact that government could reap large sums of money from "marketing" this type of intellectual property points to a conflict among policy objectives. The federal government performs some unique functions which have no analogue in the private sector. In general, these tend to be collective in nature and which are also necessary for private actors to maximize their utility.[16] They include

the provision of national security (including the relations between nations),

the provision of domestic security (a justice system) which includes the police, judiciary and rules for the definition and protection of individual rights -- including property rights[17]

the protection of the environment,

control over the supply of money, and

the provision of infrastructure which has the characteristics of a public good in the technical sense.

Some public goods that the federal government produces consist of intellectual property (subject to copyright), e.g., the output of Statistics Canada. Allocative efficiency usually requires[18] that such goods be financed[19] by taxes levied by government.[20] It should be noted, however, that other types of federal copyright material has the characteristics of a private good. Unless these also serve some other social purpose, it seems reasonable for the government to adopt a rent-maximization strategy for such copyright material.

Third, government is one very of the very few entities which is (potentially) capable of seeking to maximize society's welfare by acting so as to closely align private costs with social costs and private benefits with social benefits. For example, it can tax a negative externally so that the producer's private cost is equal to its social cost of production (i.e., including the cost of the negative externally). I suggested that in defining its policy toward its own copyright material in the form of statutes, regulations and judicial decisions the federal government should focus on maximizing aggregate economic welfare, not economic rent.

2.4 Values/Criteria/Considerations in Setting Policy

Maximizing Aggregate Welfare: In establishing its policy regarding its own copyright material, the Crown should emphasize its role as a virtually unique social vehicle for focusing on aggregate social and economic welfare. It must recognize that there are competing desirable policy objectives, e.g., free flow of information, creation/protection of rights in intellectual property to encourage effort to create such (valuable) property, protecting the individuals right to control use of information about themselves which defines their persona (see Mackaay, 1992a).

The social value of a free flow of information seems high indeed. This is the idea that "once put in circulation, [information] may be used, copied, transmitted, and further circulated freely by anyone who comes by it legitimately" (Mackaay, 1992a, p. 57). In this vein, Samuelson (1994b, p. 26) argues that the Internet "has greatly facilitated and enhanced communication and learning of the very sort copyright law is supposed to promote...the Internet has promoted public access to information far more than it has promoted copyright infringements...the vast majority of Net users are law-abiding citizens who make no more than fair and reasonable uses of copyrighted works."

Government managers, pressed by budget cuts and ministers who want to find more non-tax revenues, might see visions of financial sugar plums when they consider how government copyright material could become a source of revenue. This is a way of being entrepreneurial -- one that Gaebler and Osborne (authors of Reinventing Government) would approve.

However, maximizing the economic rent from the government's copyright material may not maximize social welfare, even aggregate economic welfare. Once the material exists, its costs are sunk. Raising the governments' return on its copyright material will not call forth more supply of intellectual effort unless government begins to produce copyright material as a business activity rather than as a by-product of its traditional activities (e.g., creating and enforcing laws, national defence, etc.).

Where a government activity is to be conducted in exactly the same fashion as if it was being conducted by a privately-owned firm, there is a serious question whether the activity ought to be conducted by government at all. What then is the point of government ownership? Pure profit maximization by a government entity violates Lincoln's advice that the state should do only those things that the people cannot do for themselves. (Presumably, Lincoln would reconsider where the state was substantially more efficient in conducting some activity.)

Failure to Maximize Rents Will Confer Benefits on Private Interests: It is certainly true that free use of government-financed copyright material (albeit subject to some constraints designed to ensure the integrity of the material[21]) may create substantial benefits for some individuals but not others. Thus, an electronic publisher (or even a current print publisher) may get rich distributing government copyright material.[22] So what -- provided that competition prevails in this market. [23] At the least, this will require open access to all potential users of Crown copyright material.

Better Access Via Electronic Distribution? If the Crown allows electronic publication and distribution of statutes and regulations will the public's access be made easier and less costly? Obviously, the potential users of this material vary in their sophistication and resources. The problem of access is not likely to arise for lawyers, administrators and policy analysts. All are, or soon will be, online. Electronic distribution is likely to reduce their total cost of getting and using this material. What about the casual user -- the average citizen? She is not online. Indeed, less than 40% of Canadian households have a computer at home. [24] As important, it appears that only a few per cent are online.[25] If the old method of distribution (print) is eliminated, they will likely be badly served. Will they be able to "look up the law" at the public library using a free computer terminal? Easy access to the materials that are at the core of a society governed by the rule of law has a number of dimensions: economic cost, physical access, and cognitive skills. Thus even in the digital age it may be necessary for the federal government to maintain the traditional method of print distribution for years while the switch-over to electronic communication occurs.

Differential Fees or Terms of Access? Should government change a copyright fee for electronic distribution by private enterprise when it does not do so for distribution of the same material in print form. Of course, to those who are online, the electronic form has a number of advantages: it can be diffused instantly; it can be addressable -- making searches easier; it can be stored in a more convenient fashion.

When making its copyright material available to profit-making firms (and other organizations) which propose to reproduce and distribute it in electronic (digital) form, I suggest that the Crown should

ensure non-discriminatory access to its copyright material,

use competition law to attack restraints of trade in markets for electronic distribution of its copyright material, and

retain the right to revoke the use of its copyright material when it is part of a scheme to restrict competition in violation of the Competition Act.26

Effective competition in the exploitation of the Crown's copyright in cyberspace is the best way to ensure that the social benefits of such material are maximized.

Importance of Innovation: In selecting among policy goals regarding its own copyright materials, it may be useful to remember from where increases in economic welfare (a proxy for which is real GDP per capita) come. There are two sources: (1) more inputs, and/or (2) more outputs per unit of input. The productivity improvements second source is called and these are based on innovation and technological change. These take two main forms: new (valuable) products[27], and new processes for production and distribution which produce previously existing outputs at less total cost to society. In a world worried about sustainability, the benefits of innovation and technological change seem more obvious than usual. To sum up, I suggest that public policy should be biased in favour of factors which create incentives to increase the rate of innovation and technological change. Obviously the definition and protection of rights in intellectual property[28] are rather important in this matter.

It is very difficult (impossible?) to specify in any practical terms the optimal rate of innovation in even a single society. In general terms, it depends upon

the expected costs and benefits of innovation in the narrow sense,

the expected costs and benefits in the far broader (social) sense,

the society's risk preferences including its attitudes toward the changes induced by innovation,

the rate of diffusion of innovation[29], and

the rule used to aggregate preferences in social decision making.

2.5 The Context for Policy Making

Public policy is not made in a vacuum. In particular, two key contextual factors are important: (1) the perceived consequences of the existing policy regime that has been in place, and (2) broader social, economic and political environment; policies must compete for attention at any time; what else is going on shapes the way new policy is created in any particular area.

Some obvious elements of the current content context include the following

There are huge government deficits and monstrous public debts.

There is obviously a strong need for revenue (unwilling to cut expenditures to balance the budget).

There appears to be a new level of opportunism -- grab revenues when you can (minimize squawk while plucking a live goose!) There is a great danger that government will act myopically.

The federal government has become a booster of the information super highway and see in it large benefits for the Canadian economy.

There are specific problems/complaints arising as copyright meets the digital age. Here lobbying is important because there is a real stakeholder pressing for action to deal with specific problems. (This is not to say that alarums are not raised in anticipation of harms.)

In this context, there will be strong pressure on and within the federal government to move toward a rent maximization approach to its own copyright material. As noted above, this seems inappropriate in the case of statutes, regulations and judicial decisions and perhaps other types of Crown copyright material.


In a widely-read article, John Perry Barlow (1994), co-founder of the Electronic Frontier Foundation, contends that in the emerging Virtual World, copyright protection is or soon will be impracticable. He asks, "If our property can be infinitely reproduced and instantaneously distributed all over the planet without cost, without our knowledge, without its even leaving our possession, how can we protect it? How are we going to get paid for the work we do with our minds?"

These questions and other statements in the paper contain a number of propositions about the digital world, which, if they are correct, pose a serious -- perhaps insoluble --problem for the protection of intellectual property. However, many of Barlow's propositions are open to serious challenge. Thus his conclusion may not hold. In any event, Barlow's analysis contains contradictions, as we shall see.

In summary terms, I argue that while it is true that all previous information can be digitized (encoded in strings of 0s and 1s on an addressable medium)[30], and can fairly easily be transmitted through computer networks, it does not follow that doing so make copyright impossible or impracticable. Minds capable of conceiving and implementing digitization and the Internet can be put to work finding efficient/effective ways of protecting a creator's/owner's intellectual property. By the weight of evidence he has presented, Barlow has not made the case that it is impossible, impractical or inefficient for individuals or the government to enforce copyright.

I shall now review the main contentions in Barlow's article.

3.1 Infinitely Reproduced?

Can digitized material (text, graphics, sound, video) be "infinitely reproduced" as Barlow suggests? In principle, yes, but not costlessly, as Barlow indicates.[31] This is an important point. Copies of digitized intellectual property require a medium on which to be copied (diskette, hard drive, optical disk, tape etc.). Software is needed to transfer the file from one place to another. In the case of CD-ROMs, for example, the equipment to record is several thousand dollars and the equipment to stamp them out is much more. It is probably true, however, that the short run marginal cost of each additional copy of a CD-ROM, diskette copy of a file, etc. is tiny.

It should be noted that digitized property which is transferred from one computer to another leaves traces -- although it may be expensive to identify them and identify who has misappropriated (stolen?) digital property. The threat of such misappropriation will itself call forth effort to find ways to make unauthorized copying/distribution more difficult /costly. [32] Providing security for one's property always involve costs. It is the price people have to pay to restrict unauthorized use (e.g., theft) of their property.

3.2 Instant Distribution Without Cost?

Barlow states that digital material can be "instantaneously distributed all over the planet without cost". This is not true -- for several reasons:

Considerable computer hardware and software is required at both ends. Even $1,000 for a computer and necessary software represents an insuperable barrier for many.

A broadband channel may be necessary for the last mile. This depends on the bandwidth requirements of the material being transmitted. Full motion video requires almost one-thousand times the bandwidth of voice and voice requires far more than text.[33] Of course, text, the staple of the Internet, can be handled by the traditional twisted copper pair over the last mile (see Stanbury, 1995).

Computer networks are learning-intensive things. Specialized human capital must be created and applied by the individuals in the distribution process envisioned by Barlow.

The Internet, while growing rapidly, still only serves under 40 million people, mostly in North America.[34] While the world's population exceeds five billion, there are only about 1.2 billion telephone terminals.[35]

In Barlow's analysis there is an implicit confusion between total and marginal cost. Even if the marginal cost of transmission is small (and it must include the probability that the message fails to get through), the total cost is much larger. Moreover, the marginal cost for a new user of the Internet who does not own a computer still constitutes a major barrier to the rate at which people are able to go online.

There is an even more important implicit misunderstanding -- that between private and social cost. While university students and faculty may pay nothing for use of the Internet, their university does. Further, what the university pays is not the full cost of providing the Internet, It has long been subsidized by the U.S. government, although the amount of the subsidy has declined sharply in recent years.[36] In general, however, it may well be true that even Barlow and other "net freaks" are being subsidized when they use the net. That fact helps to explain why the use of the Internet has expanded so rapidly.

It may also be useful to note that the cheap transmission of material over the Internet also reflects the fact that very important capital expenditures on the telephone network and on computers (both mainframe and PCs) had already been made. Currently, the net also benefits from the fact that competition in long distance (voice) telephony (beginning in late 1970s in the U.S. and 1992 in Canada) has resulted in a large amount of excess capacity on fibre optic cables which now handle most of the long distance traffic. There is also much excess capacity on trans-Atlantic undersea fibre optic cables.

In summary, the zero-cost of transmission of digitized intellectual property propounded by Barlow is a myth -- even at the margin. Further, low social marginal cost is largely the result of previous historical developments -- the vast investments in computer and telephone industries. Finally, most perceptions of cost associated with the digital world ignore the value of the time of the people involved. (Of course, for some, participating in cyberspace -- the process itself -- is not a cost but a benefit.)

Later in the article, Barlow goes even further when he states that "it is now possible to convey ideas from one mind to another without even making them physical..." [37] Wrong -- unless he has invented mental telepathy. He has abstracted the Internet to an unrecognizable degree. He ignores the following: (1) the input computer (the tool on which the creator of the idea worked and initiated the process of distribution); (2) the telephone wires (or other method of transmission such as wireless and/or satellite) connecting the input computer to other computers; (3) the computer of the person(s) to whom the idea is being conveyed. The recipient or "user" must, at the very least, find the idea on another computer (bulletin board, web site etc.), transfer the file, and peruse the words, graphics, etc. on the screen or listen to the sounds of the idea; and (4) the software to make the bits dance in the desired fashion.

Each of these steps requires that a physical object be used. Each provides the possibility of creating a "trace" on subsequent reproduction and use of the intellectual property.[38] With a trace, it is possible to "get paid" for the use of this property.

3.3 Appropriation Without Knowledge?

Can one's digital intellectual property be reproduced and distributed "without our knowledge" as Barlow states? Possibly, but not legally, although there is a fairly wide domain of "fair use".[39] It is one thing to peruse another person's copyright material in digital form or even make a copy for personal use. It is quite another to reproduce the digitized material for commercial gain. As Samuelson (1994b) points out, the owner of a copyright protected book is not entitled to make thousands of copies of it for friends, but she may share their copy with friends.

Presumably, anyone who is concerned about this form of "taking" without compensation will make some effort to prevent unauthorized access. Given the skill of some "hackers," complex and expensive forms of security may be necessary to prevent theft. But this is true of other types of property. Think only of home security systems, and alarms on automobiles.

Under Barlow's scenario, where would the distributor get the "original" if the copyright material was put into digital form from inception? By invading a computer and stealing a file? As noted above there are ways to prevent this from happening. If the material is stolen in analogue form (e.g., hard copy of a text or graphics, long-playing record, painting, or sculpture) it must then be digitized with subsequent loss of quality. (Also, how can one digitize a sculpture[40]-- not just the two or even three dimensional picture of one?)

Barlow (1994) argues that the central distinction between information and physical property is that "information can be transferred without leaving the possession of the original owner." This point needs clarification. What is transferred to another person is perhaps the right to use the information in certain ways and/or under certain conditions. What did not leave the owner's possession is the bundle of property rights associated with ownership.[41] Thus a playwright licences X performance of his play by company Y in city Q, during a specified period of time and subject to other conditions as well. The director cannot -- legally at least -- take his copy of the script and produce additional performances in another city.[42]

3.4 Is the Information Really in Cyberspace?

Barlow (1994) contends that "digital technology is detaching information from the physical plane, where property law of all sorts has always found definition." Not so on at least two counts. First, the 0s and 1s must be recorded on some medium, e.g., computer floppy diskette, hard drive, optical storage device, tape or computer chip. Second, property consists of a bundle of intangible rights (and obligations) which define the meaning of ownership and/or use. In any event, the broadcasting has had to deal with a similar phenomenon for decades. Samuelson (1994b, p. 23) suggests that, "Broadcasters don't transmit physical objects they transmit intangible information the public can view with the aid of television and radio machines." I think it is better to say that broadcasters transmit electrical signals which require receivers to transform the signals into sound or sound and pictures intelligible to people.

Barlow's often fanciful words suggest that the term "cyberspace" may itself be misleading its users about what is going on when people use a computer network. All of what happens to digitized property is done by controlling electricity in various ways (e.g., including magnetism). Nothing happens "out there" in n-dimensional space. Everything happens in our ordinary three dimensional world -- no matter how magical it may seem. All the activity that is said to occur in cyberspace moves in accordance with well understood physical laws. All the bits dance in a physical medium right here on earth (or between the earth and satellites).

It is true that, by their very nature, ideas are not physical. They are mental constructions; they are abstractions. But their transmission does require some physical manifestation: text; audio/video tape; diskette; even talking. Certainly, oral communication is a physical manifestation upon which it would be very difficult to build an intellectual property regime.[43] On the other hand, talk (let alone grunts and sighs) is a poor method of transmitting many ideas. Other media are required -- capable of storing and/or playing back much more complex material (measured in millions of bits per second, for example).

3.5 Paying for Ideas?

Barlow (1994) asks, "How does society now pay for the distribution of ideas if not by charging for the ideas themselves?"[44] He forgets that computer networks are themselves the method of distribution and t hat distribution of digitized files requires a positive act. They are not as he implies, in the ether after all. The files are in a physical location and the files are transmitted over telephone lines (or a wireless substitute). In reality, as well as in principle, the several steps of getting a digital file from the Vatican Archives to me is or can be controlled. With control comes the capacity to collect tolls (the danegelt for brain workers).

Barlow (1994) argues that the hard disks of distant servers "have no macroscopically discrete or personally meaningful form." Nonsense! If the concept of cyberspace is sufficiently real to inflame the imagination of millions, how difficult can it be for people to come to terms with many types of information stored on a remote hard (optical?) disk? Recently, even the newspapers have been full of stories explaining how the movies for movies on demand will be stored in huge files attached to a massively parallel computer.

3.6 Implications of a Borderless World

The problem of transborder transmission of digitized property obviously greatly complicates the ability to enforce intellectual property laws. That is also the case with transborder pollutants, the movement of people, and the movement of goods. The problem is one of degree, not of kind. It also is one of stage of development of the problem itself. International responses to transborder pollution are more advanced than for digitized property. As the problem of transborder transmission of intellectual property becomes more acute we can expect more effort to solve it. It is true, however, that cooperation among governments is likely to be slow to develop. However, the much maligned multi-nationals may be able to be more effective acting on their own or through cooperation with other multi-nationals.

3.7 Fundamental Nature of Wealth

I challenge whether the "fundamental nature of wealth is changing" at all, as Barlow (1994) asserts. Wealth is a stock of property rights which gives its owners the capacity to gain command over other things of value for consumption or investment. The bases of creating wealth have certainly changed -- compare the activities of Bill Gates, Commodore Vanderbilt and Sam Walton.

"Humanity now seems bent on creating a world economy primarily based on goods that take no material form," says Barlow. Again, this is a huge exaggeration. It is true that industrialized/developed countries have shifted from primary (resources) to secondary (manufacturing) to services over the past century. It is also true, that the knowledge-based activities are now a huge component of both the goods[45] and services industries. Barlow's gross generalization flies in the face of several key facts:

Most of the world's population lives in low-income, developing countries. For their citizens, the main problem is meeting their material needs.

Material things remain important in even the most advanced economies: food, shelter, water, capital infrastructure, industrial physical capital (buildings, plant equipment). Besides, consider the relative importance of things. Without water (or Perrier or other fluids) a person perishes in a few days. The absence of telephone service for the same period does not produce the same result.[46]

It is the activities of the material world that create the problems of environmental degradation and non-sustainable use of natural resources (including the air and water).[47]

Barlow (1994) suggests that "soon most information will be generated collaboratively by the cyber-tribal hunter-gatherers of cyberspace." This seems to be an exaggeration. Where does the information on the Internet come from? First, it comes from non-electronic sources (e.g., the text of the Canterbury Tales) outside the net. It is then put into digital form. Second, new information can be created from previously existing information on the net by selection, compilation, analysis, transformation and so on so as to produce what is perceived as new information by at least some people. [48] Third, information on the net can be created on the net itself, e.g., E-mail.[49] If these new hunter-gatherers simply appropriate and distribute the property of others, they are both uncreative and thieves --whether or not it is practicable to stop them or punish them.

3.8 Contradiction or Paradox?

There seems to be a fundamental contradiction in Barlow's article. About half way through it he switches from saying that in the digital age intellectual property can't be protected by copyright to saying that corporations and their lawyers are fencing in and deeding property rights over "what was previously considered a common human resource, distributed among the minds and libraries of the world, as well as the phenomena of nature herself..." With respect, Mr. Barlow is trying to suck and blow at the same time.

Or is Barlow pointing to a paradox? Is he saying that corporations and their lawyers are now "privatizing" the intellectual property which was previously embodied in early forms of technology (notably print) stored in libraries, archives and the like? In other words, that which was "free" is now subject to tolls. What Barlow ignores is the fact that it is important to distinguish actual from potential access to information because costs, broadly construed, may severely limit actual access. The information resources of the Library of Congress or the British Museum have existed for a long time and have been free to users. But their use required a personal visit and this might have required a lengthy journey at considerable expense -- not to mention the time spent at the library/museum to search the materials to find what one wanted. If the information in each repository is converted into a large electronic data base it becomes possible to gain remote access to the material over the information highway. The fees for access to these electronic files will have to be compared to the cost and time associated with the older method.

3.9 Efficient Enforcement of Copyright

Barlow (1994) seems not to understand what constitutes the efficient level of enforcement of intellectual property rights in the digital world he envisions. It may not be necessary to have anything close to perfect enforcement of such rights to generate sufficient economic rewards to call forth the optimal (efficient) amount of effort to produce various types of intellectual property that society values. Barlow must move from an implicit determinism to a stochastic universe. In it, the rational creator considers the expected net rewards of producing a piece of intellectual property [50]. The rewards are two types: pecuniary and non-pecuniary (ego, fame, satisfaction, utility of the creative process itself, etc.). The word "expected" refers to the fact that the rewards of both kinds are uncertain in amount and timing.[51] The word "net" refers to the need to deduct the costs of asserting and protecting ones rights in the intellectual property that is created. It recognizes that some piracy will occur and thus some users of others' intellectual property will ride for free. If the loss of potential revenue is expected to be so extensive that the net benefit is tiny, the effort will not be made (and the social benefits will not be created). However, even where "rip-offs" are widespread, the expected net rewards may be huge and thus justify considerable effort to create such intellectual property. An optimal (efficient) system reflects the real world of transactions costs [52] to police intellectual property rights. It also is a functional one: the rewards to individuals need not exceed the amount necessary to call forth the volume and quality of effort where the net social benefit[53] is maximized.

3.10 Protecting Copyright in the Digital Age

The experience with the Internet indicates that people want free expression and easy access to a cornucopia of information. However, they are not so naive as to believe that no one should be paid for the fruit of their intellectual labours. I suggest that convenient access to information includes a combination of modest (moderate) fees for use and a convenient method of gaining access and making payment. As an academic who often wants to use a variety of articles, book chapters and other materials in lieu or in conjunction with a textbook, I can testify to the high transaction costs of gaining the necessary permissions. In only a few cases are the fees themselves a deterrent. It's the hassle that deters one from gaining permission for much print material in a non-commercial use.

At the present and emerging state of development, the core issue is finding ways of collecting royalties on the use of digitized information stored on remote hard disks.[54] The components appear to be

1. identifying users,

2. defining/controlling legitimate use (preventing unauthorized reproduction and distribution), and

3. billing and collecting from users.

Probably #2 presents the greatest difficulty. Can we transmit a digital file which contains hidden codes which control use (e.g., only 12 playbacks of a song) and prevent reproduction or re-transmission? Yet later in the article Barlow explains this might be done.

Barlow argues that "we simply don't know how to assure reliable payment for mental works" in the case of non-physical transactions. To which I reply, not yet. Greater economic incentives tend to call forth greater effort (including the mental type). Perhaps the problem is not yet sufficiently acute to generate the necessary expected reward to have talented people apply their skills to solving it.

Billing costs matter. Consider the fact that the incremental cost of billing when a telephone company adopts local measured service (usage-sensitive pricing) can be quite significant relative to the price per minute of use. In some cases, the game is not worth the candle, i.e., the gains in efficiency (allocative and technical) are outweighed by the extra costs of measuring use and billing for it. Very recently, Sprint USA began to offer flat-rate long distance calls, i.e., the price per minute does not vary with distance. This more closely reflects its variable costs than the traditional distance-sensitive formula.

I suggest that a practical solution to the problem of protecting intellectual property rights in a digital world consists of three main elements: (1) creators/owners should set the lowest feasible price on their property (taking the price elasticity of demand into account); (2) they should make the method of payment cheap and convenient; and (3) they need to ensure that the method of payment is secure. As Barlow (1994) suggests, "If the payment process can be automated, as digital cash and signature will make possible, ...soft product creators will reap a much higher return from the bread they cast upon the waters of cyberspace."

Barlow (1994) believes that one possible solution to the problem of protecting intellectual property in cyberspace is to extend the concept of payment for real-time performance -- from soap operas to stock analysis. Another is to put creators of copyright material on a retainer (like lawyers). A third, is for there to be the direct interaction between users of copyright material and its creator/owners, e.g., the 1-800 "help line" for owners of a piece of software. More generally, encrypted transmissions [55] "live access" could contain hidden codes which recognize and prevent various unauthorized uses. However, encryption must not make computers even more difficult to use, Barlow emphasizes.

In the end, in a single line, Barlow gets it right when he says "While technology may undo law, technology offers methods for restoring creative [property] rights."


Anderson, Ross J. (1994) "Why Cryptosystems Fail," Communications of the ACM, Vol. 37, No. 11, pp. 32-40.

Barlow, John Perry (1994) "The Economy of Ideas: A framework for rethinking patents and copyrights in the Digital Age (Everything you know about intellectual property is wrong)," Wired, Vol.3.03, March, pp. ------.

Barzel, Yoram (1989) Economic Analysis of Property Rights (Cambridge: University Press, Political Economy of Institutions and Decisions).

Claffy, Kimberly et al. (1994) "Thackering the Long-Term Growth of the NSFNET," Communications of the ACM, Vol. 37, No. 8, pp. 34-45.

Cowen, Tyler ed. (1988) The Theory of Market Failure (Fairfax, VA: George Mason University Press).

De Jasay, Anthony (1989) Social Contract, Free Ride -- A Study of the Public Goods Problem (Oxford: Clarendon Press).

Demsetz, Harold (1968) Efficiency, Competition, and Policy (Oxford: Basil Blackwell).

Dommering, Egbert J. (1991) "An Introduction to Information Law -- Works of Fact at the Crossroads of Freedom and Protection," Protecting Works of Fact -- Copyright, Freedom of Expression and Information Law, Egbert J. Dommering and P. Bernt Hugenholtz (eds.) (Deventer: Kluwer Law and Taxation Publishers), pp. 1-40.

Hardin, Russell (1982) Collective Action (Baltimore, MD: The Johns Hopkins University Press).

Hillis, Daniel & James Bailey eds. (1992) A New Era in Computing (Cambridge, Mass: MIT Press).

International Telecommunication Union (1993) Yearbook of Common Carrier Telecommunication Statistics, 20th edition (Geneva: ITU).

Johnston, Jason Scott (1990) "Law, Economics, and Post-Realist Explanation," Law and Society Review, Vol.24, pp. 1217-1254.

Kay, John (1993) "The Economics of Intellectual Property Rights," International Review of Law and Economics, Vol. 13, pp. 337- .

Kling, Rob (1994) "Reading `All About' Computerization: How Genre Conventions Shape Non-Fiction Analyses," The Information Society, Vol. 10, pp. 147-172.

Kurzweil, R. (1990) The Age of Intelligent Machines (Cambridge, Mass: MIT Press).

Mackaay, Ejan (1990) "Economic Incentives in Markets for Information and Innovation", Harvard Journal of Law and Public Policy, Vol. 13, pp. 867-909.

Mackaay, Ejan (1992a) "An Economic View of Information Law," in Willem F. Korthals Altes et al. (eds.) Information Law Towards The 21st Century, (Deventer: Kluwer, pp. 43-65.

Mackaay, Ejan (1992b) " The Public's Right to Information," in Willem F. Korthals Altes et al. (eds.) Information Law Towards The 21st Century (Deventer: Kluwer), pp. 167-175.

Mackaay, Ejan (1994) "Legal Hybrids: Beyond Property and Monopoly?" Columbia Law Review, Vol. 94(8), pp. 2630-2643.

Macrae, Norman (1991) "A future history of privatization, 1992-2022," The Economist, December 1991, p. 3; January 1992, pp. 15-18.

Mitchell, W.C. & R.T. Simmons (1994) Beyond Politics: Markets, Welfare and the Failure of Bureaucracy (Boulder: Westview Press).

Negroponte, Nicholas (1995) Being Digital (New York: Alfred A. Knopf).

Olson, Mancur (1965) The Logic of Collective Action -- Public Goods and the Theory of Groups (Cambridge: Harvard University Press).

Palmer, Tom G. (1990) "Are Patents and Copyrights Morally Justified? The Philosophy of Property Rights and Ideal Objects," Harvard Journal of Law and Public Policy, Vol. 13, pp. 817-865.

Priest, George L. (1986) "What Economists can Tell Lawyers about Intellectual Property," in John Palmer (ed.) Research in Law and Economics, Vol. 8: The Economics of Patents and Copyrights (JAI Press), pp. 19-24.

Quaedvleig, Antoon A. (1992) "The Economic Analysis of Intellectual Property Law," in Willem F. Korthals Altes et al (eds.) Information Law Towards the 21st Century, (Devanter: Kluwer).

Rauch, Jonathan (1994) Demosclerosis: the Silent Killer of American Government (New York: Times Books/Random House).

Reichman, J.H. (1992) "Legal Hybrids Between Patent and Copyright Paradigms" in Willem F. Korthals Altes et al. (eds.) Information Law Towards the 21st Century (Devanter: Kluwer), pp. 325- .

Samuelson, Pamela (1994a) "Copyright's Fair Use Doctrine and Digital Data," Communications of the ACM, Vol. 37(1), January 1994, pp. 21-27.

Samuelson, Pamela (1994b) "The NII Intellectual Property Report," Communications of the ACM, Vol. 37, No. 12, pp. 21-27.

Schmidtz, David (1991) The Limits of Government -- An Essay on the Public Goods Argument (Boulder, CO: Westview Press).

Smith, Merritt Roe & Leo Marx eds. (1994) Does Technology Drive History?: The Dilemma of Technological Determinism (Cambridge, Mass: MIT Press).

Stanbury, W.T. (1995) "Redeeming the Promise of Confluence; Analysis of Issues Facing the CRTC," in Steven Globerman et al. (eds) The Future of Telecommunications Policy in Canada (Toronto: Institute for Policy Analysis & Vancouver: Bureau of Applied Research, University of British Columbia), Chap. 30.

Stanbury, W.T. & Ilan B. Vertinsky (1994) "Information Technologies and Transnational Interest Groups: The Challenge for Diplomacy," Canadian Foreign Policy, Vol. 2, No. 3, Winter 1994, pp. ------.

Stanbury, W.T. & Ilan B. Vertinsky (forthcoming) "Assessing the Impact of New Information Technologies on Interest Group Behaviour and Policy Making," Bell Canada Papers on Economics and Public Policy.

Staple, Gregory C. (1993) "Telegeography and the Explosion of Space," Telegeography Inc. 1993 (Washington, D.C.: Telegeography), pp. 49-56.

Sugden, Robert (1986) The Economics of Rights, Co-operation & Welfare (Oxford: Basil Blackwell), Chaps. 4 & 5.

Working Group on Intellectual Property Rights Information Infrastructure Task Force (1994) Intellectual Property and the National Information Infrastructure (Washington, D.C.: July).


Classifying the Various Types of Federal Copyright Material

Statutes, regulations, judicial decisions

essential components of the rule of law

product of the core function of a government

_ Aggregate statistical information

_ necessary for stabilization policy

_ public good characteristics

_ information necessary for an efficient economy

_ Copyright material generated in the course of carrying out other government activities (e.g., reports, studies, policy papers)

_ much of the material may be time-sensitive; some may never be released because of political sensitivities

_ the material suitable for release could probably be treated as if it was privately-owned (i.e., rent maximization)

_ Copyright material produced by commercial Crown corporations

_ material which can be made public should probably be treated as if it was privately-owned.

C.L. Mardi 6 juin 1995