Copyright laws create a limited monopoly. As copyrighted works (e.g. software) play such an important role in our economy, dramatic changes to these laws present serious economic risks. We should take care to make such changes only where the benefits are clear. Several of the proposed changes to copyright law, notably restrictions on reverse engineering and on devices that can be used to break copy protection, carry with them greater risks than benefits.
Canada is revising its copyright laws because of new technologies. As a software developer and consultant, and as a consumer, this affects me directly. I am worried that in our haste to strengthen protections we will weaken the economy. Other issues, such as fair use and the rights of consumers also concern me, but I will concentrate on economic issues because I believe that if the right balance is struck there, consumers will also benefit.
Copyright grants an artificial limited monopoly to a holder of intellectual property (IP), allowing that holder a certain amount of control over how that property is disseminated and used. There are two related aims of copyright: first, to guarantee that the creator of a work is compensated for creating that work; and second, to thereby encourage the production of works for the benefit of businesses, consumers and the public at large.
The aim of copyright is not, and should not be, to make intellectual property equivalent to physical property. After all, these have different qualities. For example, all physical property is finite – there is only so much real estate or oil in the world, and the costs of production and the value of physical property increase in proportion to the quantity produced. This is not the case with IP – there is no limit to the number of books that could be written, or of songs that could be sung. With IP, creation costs are one-time, and duplication (at least in a digital environment) is essentially cost-free.
New technologies have changed the IP landscape. They make possible efficiencies in distribution that are not possible with physical property. This should lead to reduced costs and increased quality (e.g. from competition as a result of lower costs), quantity, and availability, benefiting everyone. But as distribution is the bottleneck that has allowed copyright’s limited monopoly to operate, this also creates a risk that the monopoly will be weakened to the point where creators will not be adequately compensated and therefore will produce fewer works.
Any changes to the law should aim to prevent this from happening. But there is a problem. In the past, distribution and marketing (e.g. printing books, advertising films) often comprised the lion’s share of costs. Therefore, the businesses that focus on distribution and marketing may feel they have the most to lose, as their share of overall IP market shrinks, and that strong copyright controls safeguard their industries. This can place them in direct opposition to consumers, who would benefit from lower prices, and to many creators, who would otherwise have a wider market for their work.
Witness the battle over MP3 music. Musicians are divided as to whether copying over the Internet is good or bad for them. Consumers clearly see a benefit. Publishers are terrified of being cut out of the loop. But the Internet does not make creativity easier; distribution is where market efficiencies are to be realised. If the share of distribution in the over all value chain does not decrease, then that is a sure sign that the market is being distorted and the economy is suffering. This hurts everyone – often even the distributors, because they could make more money with a smaller share of a larger market.
This is the same danger created by trade barriers and subsidies to unsustainable industries. If Canada makes copyright laws too strong, we will weaken our economy: IP costs will be artificially high, quality will suffer, and producers will actually be discouraged from creating more product because they are able to continue to rely on cash cows whose creation costs have already been paid.If we can strike the right balance, we will encourage growth in related industries, such as high tech and culture. Canada’s market will be more efficient and competitive. Witness the beneficiaries of the revolution in distribution, such as tremendous growth for telecommunications equipment makers (notwithstanding the recent collapse of the sector) and the service and consulting industries. We want the minimum regulation necessary to encourage a large, efficient market.
I realise that we are drafting changes partly in response to international treaties, but this does not mean that we must exactly match laws elsewhere. If we can find a better balance, then we will enjoy a competitive advantage. I believe that the best law will provide protections necessary to ensure the creation of works, and nothing more. The United States appears to have erred in its recent legislation (the Digital Millenium Copyright Act) by providing too much protection to its IP industries, and too little to industries and consumers who depend on IP. While this is unfortunate, it offers a window of opportunity for Canada to be more competitive. In the long run, protecting the interests of consumers will make us all better off.
Trends suggest that things may well sort themselves out. Many companies have shifted their emphasis from selling software to providing services; examples include market giants IBM (services accounted for 36.7% of their revenue in 2001) and SAP. Microsoft is moving in the same direction with its XP lines and its .Net and Hailstorm initiatives. I believe that the most expensive infringers of copyright – businesses – will end up in a services model, in which they rent software and support, rather than buying them outright. Individual consumers present a different problem, but in the past the market has found solutions. Television and radio had funding problems at one time, but they were resolved by a new business model. Advertising provided the revenue, and the markets exploded because of the low cost to consumers. Similarly, the VCR was seen as being a threat to Hollywood, but has proven a boon. While MP3s appear to have hurt music publishers, proposals for flat-rate subscription services might solve the problem efficiently. If, as with cable TV, virtually everyone subscribed, piracy would become a moot point. Current copy-protection schemes for other works, such as DVDs and video games, although technically possible to break, are already effective for the vast majority of consumers.
While these solutions may not be the right ones for today, they suggest that the market is capable of finding ways to make IP profitable. The drive towards stronger protections appears to be an attempt to keep the landscape from changing, even though changes may well benefit all involved. It is premature to create burdensome new regulations when we aren’t even certain of the extent or existence of a real problem: the economic costs for making a mistake could be severe.
Regarding legislative measures to deter the circumvention of copy protection, I believe it is important that we tread carefully here. If copy protection is circumvented in order to violate copyright law, then of course it is illegal. But I wonder whether it is practical in the first place – if someone is breaking the law by committing piracy, then that is already a crime. Making the capability to do so illegal will not stop the law breakers. It will, however, prevent legitimate attempts to reverse engineer and analyse for the purposes of competing and for testing security measures (see below). We must also be careful to distinguish between breaking copy protection and violating copyright. There are many circumstance in it is perfectly legal to make a copy of something – to transfer an old LP to cassette or CD, for example, or to write a review of a work. A reverse engineering law could make analysis of security difficult or impossible, as independent reviewers would be unable to properly examine the system. This is somewhat akin to making it illegal for someone to test the strength of the locks on their own house, or to open their car when they’ve left the keys inside. Banning the tools to do the job carries the same problems. Furthermore, determining the intent of a tool can be difficult, as many devices that can be used to violate copyright have other uses (photocopiers and VCRs, for example).
As for forbidding tampering with rights management information, the same arguments apply. Breaking into a company registration database to grant yourself a license to a piece of software should clearly be illegal, but there is a very fuzzy line. What if the information is on your own computer? Do we want to make it illegal to lie to your own computer (e.g. about your age or identity), even if no human being will ever witness the untruth? Does a company have a right to control the contents of your own computer? Can a music company sell you an MP3 and then tell you that you are not allowed to replace the hard disk with a new one, moving the MP3 in the process?
These are both serious infringements on personal freedom, and should not be undertaken simply because they would make protecting IP easier. We would be legislating to enable a particular business model, one which might not even be efficient or appropriate. We might be locking in complexity and costs in distribution which would cancel out the gains technology offers. For example, should companies focus on copy protection for music, or would a subscription model in which consumers pay a flat fee for unlimited access make more sense? The second, if workable, entails far less overhead. The same model could well apply to software. I simply don’t see adequate or convincing evidence that the market would collapse without such legislation.
Finally, my understanding of the current liability of network intermediaries is that they are responsible only if they control the content of their network as a matter of course, or if they are made aware of infringement. This seems reasonable to me; I believe the recent Napster decisions in the U.S. drew the line in about the right place, by stating that while the Napster software could be used to pirate music, in itself it was legal so long as the company took action to halt the piracy of copyrighted works once it had been alerted to their presence on the system.
Disallowing reverse engineering is a bad idea. Reverse engineering is a costly process, and is usually only applied to very profitable works. The IBM PC market, for example, was created because the original IBM BIOS software was reverse engineered, allowing for competition. While this weakened IBM’s relative position, it grew the overall computer market tremendously and surely benefited IBM in the long run. If the United States had disallowed this, the high tech industry would probably be much smaller today, and it might not be centred in the United States.
Banning reverse engineering also carries serious risks in terms of security, because it makes it impossible to properly test critical software for flaws. With so much of the world’s network infrastructure reliant on a few pieces of software, it is essential that a process of analysis and peer review operates to minimise our vulnerability to hackers, viruses, and worms. Otherwise much of the wider economy would be at risk.
Copyright is an important concern, and we should take care that the laws are not left behind by new technologies. However, we must balance this with the risks of such changes. In my view, there is as yet no demonstrated need for a dramatic reform of copyright laws. In particular, banning reverse engineering and devices capable of breaking copyright is unwise. The market may well find efficient ways to take advantage of new technologies without the need for reforms that could create a burden for businesses and consumers. For the moment, we should hold off on such laws until we have seen how the economy adapts.
I wrote this document in September 2001 in response to a request by the Government of Canada for public input about future changes to copyright law to bring it in line with WIPO treaty requirements.