The recent judgment delivered by Mr Justice Kenneth Parker on the Judicial Review Application brought by BT and TalkTalk against the Digital Economy Act 2010 (DEA) raises an interesting economic question: can one look at measures that require internet service providers to assist rights holders in combatting online infringement of copyright without considering at the same time whether current copyright strikes the right balance between the incentives for creators and the interests of consumers? Unfortunately, economics suggests that this is not the case. Copyright legislation and subsidiary provisions that merely deal with enforcement issues cannot be viewed in isolation from each other. The latter affect the trade-off at the heart of copyright law. Technologies that support low-cost distribution of information do not just raise an issue for enforcement of existing copyright protection, but should cause a re-examination of how far that protection should effectively go.
The DEA and the Judicial Review Application
The Digital Economy Act (DEA) was one of the last pieces of legislation passed by the previous Parliament, having received Royal Assent on 8 April 2010. It follows from the promise of legislation to help put Britain at the forefront of the move towards a digital knowledge economy (as set out in the Government’s 2009 ‘Digital Britain’ report).
A number of key provisions in the DEA are intended to address the problem of online copyright infringement, in particular the unauthorised downloading of copyrighted material through peer-to-peer file sharing. More specifically, the DEA imposes obligations on Internet Service Providers (ISPs) to provide assistance to rights holders in relation to the enforcement of their copyright by:
- processing so-called Copyright Infringement Reports (CIRs) from the rights holders containing information about IP addresses identified as having taken part in the sharing of copyrighted material;
- identifying and notifying the subscribers associated with these IP addresses;
- maintaining anonymised lists of subscribers who have been identified and notified in this manner, and providing this to rights holders upon request, which in turn would enable rights holders to seek a Court order requiring the ISP to disclose information about the subscriber so that legal action may be taken.
Should these so-called Initial Obligations not lead to the expected reduction in online infringement, the DEA contains further provisions that enable the introduction of technical measures that ISPs may be required to take against subscribers who have been found to be infringing copyright, such as for example, degrading their connection or disconnecting them completely from the internet.
BT and TalkTalk have applied for Judicial Review (JR) of these provisions on the basis that they are incompatible with provisions of EU law, including considerations of proportionality. The application was heard at the end of March this year, and the judgment was delivered on 20 April, rejecting the application with the exception of a claim in relation to administrative charges.
One of the grounds for the JR was that the measures introduced in the DEA were disproportionate, and that the impact assessment undertaken by the Government to evaluate whether benefits outweighed costs was flawed. On this point, the judge took the view that, in weighing competing interests, Parliament must enjoy a wide margin of discretion and that proportionality does not hinge upon the reasonableness of the assumptions made by Government when looking at the relative costs and benefits of the contested provisions.
A main component of the proportionality claim was that the impact assessment used the wrong framework for measuring costs and benefits: it considered the additional revenues to rights holders that might be expected to result from the contested provisions and the costs they would impose on ISPs and rights holders, whereas it should have looked at the effect of any potential increase in revenues on the incentives for the creation of audio-visual material on the benefit side, and included the welfare loss from limiting access to such material on the cost side. Dismissing this claim the judge held that “Parliament, through current copyright legislation, has already struck a balance between, on the one hand, the aim of providing incentives to actual and potential creators of audio-visual material, and on the other, the potential welfare loss to those consumers who would, in the absence of copyright protection, enjoy such material either free of charge or at substantially reduced prices but who, as a result of copyright restrictions, are either deprived of the material or are required to pay higher prices for it. Existing copyright legislation may strike that balance in a way that is controversial or open to criticism. However, in my view, Parliament, when considering measures such as the contested provisions, which could be expected to enhance copyright protection, is entitled to proceed on the basis that existing copyright law does strike a fair balance between the interests referred to.”1
From an economic perspective, there is unfortunately no such neat distinction between on the one hand, striking the right balance of creators’ and consumers’ interests, which can safely be left to one side and be assumed to have been dealt with through copyright legislation, and on the other hand, looking at the costs and benefits of measures on the basis that this balance is about right. As Judge Parker pointed out, the contested measures “enhance copyright protection”, and thus change the balance of conflicting interests. There is no obvious reason why changing the balance in this manner should be any different from changing it by, for example, extending or shortening the terms of protection. What is at stake is the effective level of copyright protection afforded to rights holders which has been eroded through technological developments that enable sharing of material on a much wider basis and at much lower costs than has been possible before, and which the contested provisions are intended to restore (albeit not necessarily to the full extent).2 It is therefore impossible not to look at the balance of conflicting interests when considering the effects of the contested provisions, and the belief that one can do so is as illusory as it may be comforting.
What is at stake with copyright?
The highly emotive language that is generally used by rights holders in the debate about file sharing and other forms of accessing copyrighted material hides some important economic truths. Branding file sharers as common thieves and merciless pirates, out for the wholesale destruction of the creative industries, ignores the fact that intellectual property is somewhat peculiar, and intellectual property rights are too.
Unlike ‘normal’ property rights, intellectual property rights are limited in time and scope. The exclusive rights granted to the creator of a copyrighted work – including the right to copy, distribute and adapt the work – have a limited term, and there are, for example, ‘fair use’ provisions that allow certain actions that would normally constitute an infringement to be carried out. That is very different from what we normally understand as property rights. An individual does not lose the ownership of his car after a given number of years, and there are no ‘fair use’ provisions that would somehow make it reasonable for others to drive the car for a certain number of miles without his consent. The line between what is permitted and what is not is often blurred, not least because rights holders themselves may sometimes decide that it is in their interest to allow some level of infringement. For example, so-called format shifting – such as recording tracks from a CD in order to be able to listen to music on a portable digital music player – is technically an infringement of copyright. However, it is generally tolerated by the industry – presumably because controlling extensively what customers do with the music they purchase is expected to undermine demand.3
Through these limitations, copyright strikes a balance between protecting the interests of the creator in receiving compensation for the effort involved in producing valuable content, and the interests of consumers who would benefit if that content were freely available. Copyright works by affording creators a limited monopoly that allows them to maintain the price of copyrighted material above the cost of making it available to additional users, with the aim that the associated profits will provide incentives for investing in the creation of further content.
The underlying reason for this is that audio-visual material – or creative content generally – can be used in a non-rival way. The difference between stealing someone’s computer and downloading some of the music tracks saved on their hard drive, for example, is that the latter does not in any way diminish their ability to listen to the tracks (at the same time as the person who has downloaded them), even if the tracks were downloaded without the consent and knowledge of the original purchaser. In other words, the purchaser would be no worse off. This would clearly be different in the case where her computer had been stolen.
This is not to say that sharing music files does not affect those who have composed, performed and recorded the music. Had the tracks not been downloaded, an additional CD may have been purchased instead (though not necessarily so), thus generating income for the record company and the artist. Therefore, those who have been involved in the production of the music that is being shared may be worse off.4 And because the copyright protection afforded to the creators of the music states that such downloading should not occur, it constitutes an infringement of copyright. But it does not amount to stealing from the creators any more than neighbours who agree to share a lawnmower can be said to be stealing from the lawnmower manufacturer.
A delicate balance
A simple example of the copyright balance
Assume that there are two potential works:
- Work A would cost 100 to produce, and would attract at most an audience of 200 it if were offered for free. 150 customers are prepared to pay 1 or more, 100 customers are prepared to pay 2 or more and 50 customers are prepared to pay 3.
- Work B would cost 70 to produce, and would attract at most an audience of 80 if offered for free. 60 customers would be prepared to pay 1 or more, 40 would be prepared to pay 2 or more and 20 would be prepared to pay 3.
Suppose that the marginal cost of serving customers with either type of work is zero.
The table below provides a summary of demand, profit and consumer surplus – measuring the difference between the actual price paid, and the maximum willingness to pay of those who purchase the work – for each of the two works at the different price levels.
Under these assumptions, the profit-maximising prices for either work would be 2, and at this price it would be profitable to create Work B. Work A, by contrast, would be created at a lower price as well.
Assume further that the level of copyright protection can be measured by the price that content creators can charge, given that it would be possible to make cheap copies of content that has been created – the more effective the copyright protection, the more the price of protected material can be increased above the marginal cost of making such copies available.
Assume that, without any effective protection, competition from illegal copies would set in at any positive price. Thus, neither of the two works would be created and overall welfare would be zero.
Now consider two levels of copyright protection – one that would allow creators to charge a price of 1, and another that would allow them to charge 2. Introducing the first level certainly improves welfare. It would provide sufficient incentives for the creation of Work A, producing consumer surplus of 150 and a total aggregate welfare (sum of consumer surplus and profits) of 200. Of course, some of those who would have enjoyed Work A at a price of 0 will now be priced out – but given that Work A would not have been created without copyright protection, that is not the relevant comparison. There would still be no incentive to create Work B.
Increasing the level of protection so that a price of 2 can be sustained means that both works would come into existence – but consumer surplus for Work A would fall by more than the gain to consumers from having work B. Aggregate profits would increase, as would total welfare. However, when looking only at consumers, extending the level of protection would be undesirable. Creators would be substantially better off, but consumers would lose out.
The effective scope of protection afforded by copyright is determined by a number of parameters: how long protection lasts, what counts as ‘fair use’, what penalties are meted out to infringers, and what measures rights holders can rely on in order to enforce their rights. The economic view is that these parameters are – or should be – set in such a way that the positive impact of stronger protection on the incentives for creators is balanced against the negative impact on the availability of protected material for users.5
In theory, the level of protection should be increased up to the point at which the consumer surplus created by the marginal work (i.e. the piece of music, book or film that would not be created without that level of protection) equals the loss of consumer surplus associated with higher prices or reduced availability of all the works that would have been created even at a lower level of protection (see box for a simple numerical example).
It would, of course, be foolish to believe that the detailed rules involved in copyright legislation in practice can be set in such a way that their combined effect is precisely to achieve this optimum. Too many and too varied are the parameters that need to be taken into account to find the optimum balance. However, looking at the theoretical optimum still tells us what factors are important for balancing conflicting interests, and how changes in these factors should be reflected within changes in the scope of protection.
Technology, and technological change, are obvious candidates. Lower costs of copying protected material suggest both greater gains from reducing the level of protection, and greater risks to the incentives for creating new material in the first place.
If peer-to-peer file sharing reduces the cost of making audio-visual material available, this creates the scope for more consumers to gain access to such material, meaning that at any given level of protection the welfare loss from exclusion is greater than before. This would suggest that the level of protection might optimally be lowered, as the benefits of widening access to all the material that would still be created become greater. At the same time, without some additional measures it might be the case that the incentives for creation are substantially undermined. If this were the case, then many of the supposed benefits associated with wider access would not materialise and thus it might be necessary to strengthen protection.
Deciding whether the scope of protection ought to increase or decrease is a matter of looking at the impact of technological changes on the incentives of creators and the scope for lower-cost distribution of content. It is an empirical question – but one that needs to focus on the balance that is at the heart of copyright legislation. There is no way of avoiding the thorny issues by assuming that the welfare of those who engage in file sharing simply does not count because their gains are ill-gotten, and that one can simply look at the revenues earned by record companies and film studios. Technological changes of a sufficient magnitude should result in an adjustment of copyright legislation based on a balance of interests, and it would simply be wrong to dismiss some of those interests on the basis of a balance having been struck in the past that is no longer the right one. If the impact of peer-to-peer file sharing on the incentives for creators is small (which is different from the revenue impact), and the loss of welfare resulting from preventing customers from sharing files is large, then it might well be appropriate not to adopt any measures that would curb file sharing, cries of death and destruction from the camp of the rights holders notwithstanding.
In fairness, the need to grapple with these issues has been acknowledged in the Government’s impact assessment, which notes that there is a trade-off between the potentially welfare-enhancing effects of online copyright infringement and the rewarding and fostering of investment in content and creation. The acknowledgement of this trade-off, however, is not reflected at all in the actual impact assessment undertaken, which simply looks at the amount of revenues lost to online infringement that would be recovered and at the cost of implementing the contested provisions (to the extent that they could be quantified). Above and beyond all concerns about the reasonableness of the assumptions on which the impact assessment is based – and there have been many – it is this lack of proper consideration of the balance between the interests of creators and consumers that is the fundamental flaw. As the judgment makes clear, Parliament has wide discretion in looking at the effects of proposed legislation before it. This includes the right to pass bad legislation. With the DEA, Parliament would seem to have exercised this privilege, perhaps thinking that it was only tinkering with some aspects of enforcement without looking at the impact of the contested provisions on the balance of interests at the heart of copyright law.
-  EWHC 1021 (Admin), paragraph 249. [↩]
- There is no expectation that the contested provisions would eliminate all file sharing and reduce the level of infringement to what it was before peer-to-peer technology became widespread. The Government’s impact assessment assumed that only about half of the revenue lost as a result of file sharing would be restored through the measures in the DEA. [↩]
- Copyright practice is in many cases more forgiving than the letter of the law. … In the case of format shifting by the individual for personal use, the British Phonographic Industry (BPI) indicated to the Commons Select Committee for Culture, Media and Sport that the UK record industry had never taken action against an individual copying their CDs to their computer for the purpose of transferring those tracks to another device for their private and personal use only, and it added that the industry had no intention of doing so in the future.” (Intellectual Property Office and Department for Business, Innovation and Skills “© the way ahead – A Strategy for Copyright in the Digital Age”, October 2009, p 32). The Hargreaves review of the UK copyright laws recommended to change this situation and make this practice legal. At the same time, the attitude of rights holders may of course change as customers use so-called digital lockers, i.e. upload their music to remote servers in order to be able to stream it to multiple devices. Given that there seems to be a business opportunity there, rights holders are beginning to shift position, although their focus at the moment is on those businesses that offer digital lockers, such as Amazon (see “Raging bulls”, The Economist, 31st March 2011). [↩]
- Note that they may not be. For example, the infringer might not have bought the music, but as a result of listening to the track, he might decide to attend a live concert by the artist, or buy one of her other CDs. In this case, file sharing will have benefited the creator either by stimulating demand for complementary products, or through what is known as ‘sampling effect’. Whether file sharing has a positive or negative impact on creators is an empirical question, and a considerable amount of discussion has focused on the relative strength of the various effects. [↩]
- As an aside, even if one takes the view that there is more to copyright than a simple welfare balance calculation, and that the protection of the creative individual’s personality or moral rights are worth a longer term of protection than might be justified on the basis of strict economic considerations, the economic approach points out that there are welfare costs associated with that extended protection. [↩]