Intellectual Property and Competition Policy: Constructing the Interface
The relationship between intellectual property and competition policy has gone through different cycles at different times in different jurisdictions. At some periods the mere ownership of an intellectual property right has been equated with market power, and its actual or threatened exercise or non-exercise with abuse. At other times courts and regulators insist that intellectual property be treated no differently from other assets or business advantages. (Regulatory neutrality, when espoused, has generally been seen as a corrective to earlier periods of hostility to intellectual property rather than actively pursued as an objective in its own right.) The last phase in the cycle comes when intellectual property is privileged in various ways over other potential sources of market power whenever the activities of right holders come under regulatory scrutiny. Each of the five jurisdictions we have selected for analysis is currently at a different stage in this last cycle, with some edging back towards neutrality and others swinging definitively away from it.
Contemporaneously with these fluctuations in competition policy, but in no sense shadowing them, are developments taking place within intellectual property law itself. This time the movement is largely in one direction. Here there are no cycles of under- or over-protection to match the cycles of under- or over-regulation seen on the competition enforcement side of the fence. The history of intellectual property during most of the period under discussion is one of continuing and accelerating expansion both within and across jurisdictions. Courts and legislators leapfrog over each other to create new rights, or to extend the duration or protective reach of those that already exist. New technologies lock up information and ideas which previously formed part of the public domain. Technological locks then require to be backed by legislative measures that seek to prevent their unpicking. Multilateral and bilateral trade agreements ensure that expansionist intellectual property regimes are negotiated outwards, almost always permanently. Economic input into these expansionist jumps in protection has been minimal before the event and largely futile afterwards. Absent such input, intellectual property owners and their advocates have by and large been able to present lack of protection for new products and processes against various forms of imitation and free-riding as a problem needing to be solved. Often they have been able to persuade judges and legislators to agree with them, usually without any deep or detailed consideration of the economic arguments for or against extension in a particular case.1 (To be fair, infringement proceedings in most jurisdictions provide few opportunities for detailed expositions by economists. Legislators have no such excuse, however.) If we see intellectual property as a form of intervention in the market and non-intervention as the baseline, we confront (a touch ironically) a repeat of the type one versus type two error tug-of-war, but this time with a silent preference for type two in the form of rights extended far beyond any possible economic rationale for protection. When the property rhetoric is peeled away, the problem could be (but seldom is) seen as one of excess of regulatory zeal rather than its opposite.
An economic analysis of the effect of intellectual property rights on competitive markets would need to be able to explain and predict the likely reactions of particular market players to the actual and hypothetical expansion or contraction of legal protections for innovation and creativity in particular markets. Care must be taken, however, that in identifying potential market participants and describing and predicting their behaviour, one does not also identify with their interests. Competition policy is about protecting the process of competition itself, not persons.
When the process of artificial simplification called modelling is applied to intellectual property rights, it tends to focus on lone inventors or creators rather than on the team of which they are nowadays so often a part or the organisation that funds or employs them. No harm is done provided that, as with all models, it is clearly understood that the aim of the exercise is clearly to present that which it is sought to predict rather than make a prediction. Individual reactions to changes in the content of a right are after all easier to model than group or corporate reactions. This necessary stripping out of complexity should not be allowed, however, to slide over into the rhetorical privileging of the interests of actual or supposed first-movers in the process of innovation or creation just because they are first-movers. After all, authors and inventors usually remain owners for only the shortest of times, if indeed they become owners at all.
To intellectual property owners, competitors are potential infringers. That competitors may themselves be downstream innovators or creators should not be forgotten. Building on what has gone before is how market economies develop. Innovation is both sequential and cumulative. Extending intellectual property owners’ rights to control adaptive or derivative use can get in the way of this process. The tactical use of blocking patents can have a similar effect.2 Looking only at the interests of current right owners obscures a very important truth. If downstream innovation or creativity is blocked by the existence of an upstream intellectual property right, any resulting loss of innovative efficiency will have to offset whatever gains in innovative efficiency it is sought to attribute to the existence of the intellectual property right in the first place. Such netting out will not be easy given the relative ‘softness’ of innovative efficiency as a concept.3 Tackling one half of the problem does not make the other half go away.
Innovative and creative products need consumers just as much as any other goods and services. Consumers tend to get short shrift from both competition law and intellectual property. As we saw in chapter three, many economists would deny their interests any part in shaping competition policy. Their voice is not directly heard in infringement proceedings either. Even when their interests can be factored indirectly into judicial decision making (as they are occasionally in the course of legal argument made on behalf of others), consumers and users tend to be disadvantaged by the unequal status that comes from posing adverse consumer outcomes as defences rather than incorporating them in liability establishing rules. This matters greatly when it is only the defence that picks up competition values.4
One of the adverse consequences of personalising contributions to innovation and creativity in the form of a property right granted to a first owner is that it ignores the effect of more widely dispersed but nonetheless important stakeholders. It cannot always be assumed, for example, that it is only right owners or derivative users who are responsible for creating the net increase in social value attributable to the right. A properly inclusive economic analysis would recognise the contributions of:
a) the owners of the technology through which the protected work or product is disseminated;
b) the creative predecessors on whose effort the inventor or author builds (not all such predecessors will themselves be protected by intellectual property rights);
c) wealth creators in society as a whole, ie anyone who contributes to the consumer surplus which makes possible the purchase of the protected work or process.5
The obvious difficulty with this way of thinking is that the more widely the economic net is cast, the more fuzzily imprecise the analysis. On the other hand, if policy makers ignore these dispersed stakeholders entirely, they run the risk of unfairly tipping the balance of argument in the direction of right holders by leaving qualitatively important (if not always easily quantified) factors out of the equation.
One of the difficulties inherent in constructing or deconstructing the interface between competition policy and intellectual property is the use of emotionally-charged words and concepts. While terms such as ‘property’, ‘regulation’ and ‘monopoly’ have become terms of art in legal discourse, all too often their technical meaning is allowed to fall away so that the war of words is conducted at a purely rhetorical level. When this happens the words come freighted with approval or disapproval. ‘Property’ is often viewed as good, while the terms ‘regulation’ and ‘monopoly,’ if not seen as downright bad, certainly carry negative connotations. If the good can be captured by theorists and the bad or negative fended off, the battle is half won. It is therefore useful to try to penetrate some way into this linguistic thicket before embarking on an analysis of the law and economics of intellectual property. This necessary act of deconstruction is made easier by the fact that some of the rhetoric is contradictory. Thus the proposition that ‘property is sacred and intellectual property shares that sanctity’, and the proposition that ‘property is not special in competition terms but intellectual property is’6 are both arguments invoked against regulatory intervention but with radically different starting points. What they have in common, however, is that they both assume that property and regulation are inevitably and irredeemably antithetical concepts.
Property is a much-debated construct in all legal systems, and it is not our intention to traverse that debate in detail here. For the necessarily limited purposes of this book, we make the following assumptions and reservations about the nature and role of property rights:
a) That ‘property’ is a bundle of rights, not an object or a thing.7
b) That ‘property’ can have different meanings in different legal contexts, and may be used for different legal purposes. Thus while trade secrets may be granted property status when instituting a system of registrable personal property securities,8 this usually has little or no impact on judges’ propensity to act as though they were mere liability rules or obligations on other occasions.9
c) That all forms of ‘property’ require legal action for their recognition or creation,10 and that this reality is in no way diminished by the form, visibility or antiquity of that act of recognition or creation.
d) That while economists can tell us a great deal about the consequences of allocating property rights, both the act of allocation and the subsequent categorisation of property rights are tasks for lawyers. Distinctions that are important to one group may be irrelevant or distracting to the other. ‘Property’ does not equal ‘asset’, still less ‘competitive advantage’.11
e) That while the triad of assignability, exclusivity and severability may be taken as indicia of property, the absence of one or more of them is not necessarily fatal to the existence of a property right in a particular case.
What is the significance of these overlapping ambiguities for competition enforcement? Simply this: even if using property as a shield against regulatory intervention could be justified on economic grounds, property is too unstable a concept to provide a workable dividing line between immunity (actual or presumptive) and intervention.
Amalgamating all forms of property for regulatory purposes12 obscures some distinctions that do actually matter. The first is that while natural rights arguments are frequently mounted for tangible property, the case for intellectual property is usually made on more pragmatic or utilitarian grounds, in common law legal systems at least.13 Intellectual property is explicitly designed to bring about a particular result. Older forms of property do so only by accident and indirection. Secondly, justifications for property, whether they be Lockean or utilitarian, usually speak in terms of the need to conserve scarce resources. As one eminent writer on the subject has succinctly put it, property rights manage scarcity while intellectual property creates it.14 Thirdly, the boundaries of tangible property are usually easily recognisable (this being the true function of the physical object or parcel of land15), whereas the boundaries of a particular intellectual property right are by definition invisible to the external eye.16 Fourthly, while no property right is entirely inviolable, in the sense of being free from social constraints, those constraints are generally more overt and more onerous in the case of intellectual property. Thus the originality threshold and rules about fair dealing in copyright law, the independent discovery defence in trade secret law and public interest restrictions on patent acquisition all make greater inroads into the owner’s domain than is the case with the much more restricted social controls accompanying ownership of a parcel of land.17 Lastly, the subject matter of an intellectual property right is not diminished by being consumed and may be simultaneously enjoyed by more than one consumer (a characteristic which has a profound effect on the economic analysis of intellectual property, as we shall see).
For all of the above reasons, even a purely juristic analysis of the nature of intellectual property would have to concede important differences between it and other forms of property. Juristic differences, however, do not and should not bind the regulator. The point we are seeking to make here is not that intellectual property can never be a fully credentialed member of the property club (in some jurisdictions this would be to fly in the face of statutory affirmation that copyright, designs and patents are indeed ‘property’18), nor that intellectual property can never share ownership features with other forms of property, but rather that neither similarity nor difference should be allowed to impede competition authorities. Perhaps the final word on this vexed subject can be left to a Canadian judge who (speaking of copyright, but the logic could be applied equally to patents or designs) pointed out that copyright law is neither tort law nor property law in classification but is statutory in origin, and it follows accordingly that copyright legislation simply creates rights and obligations upon the terms and under the circumstances set out in the statute.19
Property and regulation are generally treated as antithetical constructs. The first is a creature of private law. The second is as a regrettable public law intrusion into that otherwise inviolate private law world. This is something of an historical accident. Prior to the Statute of Monopolies 1623 and the Statute of Anne 1709, patents and copyright were badly- and barely-regulated monopolies, and visibly so. The passage of those particular pieces of legislation20 could be said to have privatised those monopolies in the pursuit of mainly utilitarian goals. Viewed in this way, the clash between intellectual property and competition law is a head-on collision between two systems of regulation with divergent processes but overlapping goals.21 In both cases there is a tension between private rights and public law objectives. This is no mere ‘path not taken’ whimsy. Courts and commentators in Australia22 and the United States23 not uncommonly compare a discussion of real or apparent conflicts between general competition principles and industry-specific utilities regulation with similar real (or sometimes in the court’s eyes apparent24) conflicts between the enforcement of intellectual property rights and competition law intervention. What these cases demonstrate is the futility of setting up a false polarity between ‘property’ (‘unqualified good’) and regulation (‘necessary evil’). Stripped of its rhetorical embellishments, intellectual property is a way of regulating innovation.25 What is at stake here is the clash of two regulatory regimes in no way more remarkable or different than the interface between competition law and the regulation of electricity transmission, telecommunications, capital markets26 or fishing rights. Just as one cannot ignore the adequacy of direct regulation in these cases, neither should one ignore the reach or depth of an intellectual property right when assessing the degree of market power wielded by the right holder and its potential for misuse.27
Intellectual property confers upon owners the legal right to do or authorise the doing of particular acts.28 These acts define the right, and by long usage are commonly referred to as monopolies. This, however, has nothing to do with monopolies as economists understand the term. The key to the distinction is, once again, substitutability.29 A patented product or process may have rivals that can compete with it without infringing. Copyright material can be out-competed by products that serve the same function or appeal to similar tastes, provided they do not cross the infringement line by appropriating form or expression. Trade secrets may be bypassed by information lawfully in the public domain.
The distinction between legal and economic monopolies pulls in two directions. The first (and most frequently invoked) is as a corrective to the wrong-headed notion once widely accepted by courts and regulators in the United States that intellectual property rights confer market power in themselves. The second and equally important (but less often referred to) iteration of the distinction is to reject the equally baseless assumption that any act done (or not done) in pursuance of a legal monopoly can never adversely affect the competitive process.
Thus far we have assumed a common set of justifications across all manifestations of intellectual property. This is not, however, how the law and economics debate about intellectual property rights has by and large proceeded. Courts, competition regulators and commentators have approached different rights differently. Sometimes this is a matter of the legal form in which the right is cast. On other occasions it reflects the different rationales underpinning each right. While both factors are relevant to the refusals to license debate, they can distort the analysis if they suggest either that particular rights are intrinsically more virtuous or more vicious in competition terms, or that some rights are inherently more vulnerable to over-regulation than others.
The intellectual property pie may be sliced in various ways, using different legal devices to separate the pieces. The most obvious (and in competition terms, the least significant) is to distinguish between statutory and judge-made rules.30 Statutory rights may be further subdivided into those which depend on prior disclosure through a system of State registration and/or grant (for example, patents, registered trade marks and designs), and those (such as copyright and layout designs) which are self-constituting, coming into existence at the moment of creation or fixation. It is more useful (and arguably more principled) to try to draw a line between obligation and property.31 On one side lies the classic trinity of patent, copyright and registered trade marks, and their more recent sui generis attendants, plant variety protection and some forms of design right. On the other is that penumbra of liability rules which surrounds, sustains and extends these property rights: fair trading, passing off, unfair competition and breach of confidence. While this line largely corresponds with the statutory/common law divide, the fit is not perfect.32 Neither is the barrier impermeable. Commentators constantly urge that a particular right be pushed across the property/obligation line.33 Very occasionally some judge will oblige.34 Judicial resistance is as common, however.35
How relevant is juristic form to warding off (or encouraging) competition scrutiny? On the face of it such matters should be no more germane to the issue than any of the other subdivisions of private law. Their invocation in the refusals to license debate is usually for the rhetorical purposes as described in chapter one. Copyright, patents and trade marks being creatures of statute are not to be undone by a side blow from another statutory regime, in this case competition law. Trade secrets and business reputations not buttressed by a trade mark will fail this test. Conversely, if it is ‘property’ and ‘obligation’ that is to be the dividing line then commercially valuable information might make it to the protected zone, but only if one accepts the highly contentious notion that there can be property in closely-held facts.36 Put those same facts in a database for access to which a fee is charged, and any attempt to use the term ‘property’ to describe the various legal protections against unauthorised access is even more contentious.37 Given these uncertainties, ‘property’ seems a singularly inappropriate case breaker.
Again, while arguments can be (and have been)38 made for privileging rights that are registration based over self-constituting rights, it is not clear what the act of registration has to tell us in competition terms, unless it is perhaps that the right, since it has been ‘paid for’ by prior disclosure (in the case of patents), should be more immune from ‘confiscation’ or ‘suppression’ than one (as in the case of copyright or layout designs for integrated circuits) where no such ‘price’ is demanded. Whether this embellishment adds anything of substance to the ‘regulation equals confiscation’ argument is doubtful, unless perhaps it could be argued that by extracting disclosure while reneging on the promise of protection, the State stands convicted of double dealing and duplicity. Such attempts to introduce an ethical dimension into the debate are drawing a very long bow. Rights to extract minerals or petroleum also involve disclosure (sometimes public) but carry with them no corresponding immunity from regulatory oversight. No doubt all property owners would wish to hold their rights free of interference by competition authorities, but this is a hope not a right, and any push on their part against full exposure to regulatory oversight would in no way be strengthened because their right is guaranteed by a State agency in return for payment.
Another variation on the self-constituting theme, if this time telling against immunity or privilege for right holders, is the argument that where firms are able to adjust the content of the right by limiting or expanding the circulation of information inside or outside their own organisation (as is the case under most countries’ trade secret laws) then such ‘adjustable’ rights should be treated in the same way for competition purposes as contractual extensions of State-granted rights to activities not within the grantee’s exclusive rights under the statute. As an attempt to see beyond legal form to potential anti-competitive substance, this is to be welcomed, provided always that it is understood that not all manipulations of disclosure are anti-competitive any more than is every attempt to extend a right beyond the term of a statutory grant by contract.39 Market power and anti-competitive intent or outcome on the part of the right holder must be demonstrated. In other words, the argument must be firmly grounded in the rule of reason. No presumption for or against liability attaches to the mechanisms through which secrecy and non-disclosure are maintained.40
A better way of looking at the interaction of intellectual property rights and competition policy would be to ignore juristic form and focus on the activity that a particular rule seeks to encourage or enhance. This would make economic dissection of the intellectual property corpus considerably easier. Legal rules would then be split into the following families:
(a) Rules directly fostering innovation and creativity
It is these rules on which economists have lavished most attention. Copyright and patent are the two traditional prototypes. To them may now be added sui generis rights protecting layout designs for integrated circuits, plant varieties and novel but non-functional designs. While these rules have the common objective of encouraging various forms of innovative and creative endeavour, they proceed towards that end by two quite distinct means. Some (patents, plant varieties and registrable designs) create true legal (but not necessarily economic) monopolies which outlaw parallel creativity or invention. Potentially infringing processes, shapes or products do not cease to become so because they have been arrived at independently. By contrast, copyright and layout designs law forbid only imitative uses of the protected work, not independent creativity as such.41 While this tells us something about the permeability of the barriers to entry in each case, the reluctance of the courts to give much credence to claims of independent creation means that would-be entrants are unlikely to find it markedly easier to create around a copyright than to invent around a patent.
(b) Reputational rules
These are rules which restrict attempts by the right holder’s rivals to confuse or mislead consumers as to the origin of goods or suggest a commercial connection with the right holder which does not exist. Trade mark law, the most developed of these rules, has never rested on an incentive justification.42 Its objective has always been the preservation of the mark owner’s goodwill by restricting rival traders’ ability to attach themselves to the business reputation represented by the mark. Similar objectives are served (if somewhat more loosely) by the tort of passing off and its legislative counterparts in various jurisdictions.43 Unlike patent and copyright law, reputational rules do not place restrictions on the kinds of goods or services a competitor may supply, but only on the ways in which these goods and services might be presented to the world. It is sometimes therefore suggested that such rules pose no threats to competition, or that they actively assist competition by facilitating the product recognition44 and differentiation on which competition often depends.45 This is in a limited sense true, if we confine our attention to traders inside a given market. It underestimates, however, the potential for reputation-based rules to inhibit entry to existing markets or the creation of new ones. (Such inhibitions may be economic or, in the case of parallel importing restrictions, legal.46)
(c) Fair dealing rules
This seemingly disparate group includes rules designed to protect information because it is secret47 and/or valuable.48 While they seem to lack any coherent unifying principle across jurisdictions,49 they could be said broadly to embody notions of fair dealing or commercial morality,50 principles too vague and formless to provide even the most indirect of links with innovative efficiency. If their application does, however, turn out to be innovatively efficient in particular cases, this is by accident not design. Equating them with patent law for competition law purposes absent tests of novelty and non-obviousness is too glib an elision. This would not matter if the exercise of all three types of right were subject to case-by-case competition scrutiny. It matters very much indeed when protection against that scrutiny is sought to be conferred on whole classes of right.
There are two aspects to this. Some jurisdictions may recognise intellectual property rights unknown in other jurisdictions (database rights in Europe, publicity rights in the United States51). Alternatively, different jurisdictions may draw the boundaries between rights in different ways (the lines between design and copyright, for example). One of two things will then happen. Competition rules that do manage to operate in similar ways across jurisdictions may have the unintended effect of evening out the economic effects of these differences between intellectual property regimes. Conversely, where competition principles themselves vary from jurisdiction to jurisdiction, they will magnify any problem caused by misaligned intellectual property rights.52
Signatories to the World Intellectual Property Organisation Copyright Treaty 1996 (WCT) are required to provide ‘paracopyright’ laws enabling copyright owners to block access to, and the copying of, their works in electronic form using technological protection measures (TPMs). In some jurisdictions (most notably the United States and the European Union) paracopyright protection extends copyright owners’ control over their works significantly beyond that which has traditionally obtained under copyright legislation,53 going well beyond the metaphor of a locked house54 that would simply see someone stopped from circumventing a TPM to access a work. United States paracopyright rules55 blithely state that they do not affect the normal fair use defence to copyright infringement, but this is little comfort to users prevented by a TPM from gaining access to a locked-up work to make fair use of it. Europe’s paracopyright provisions are more draconian, lacking this broad (if often useless) fair use exemption. They make no allowance for the common copyright exemptions for research or private study, criticism or news reporting, rendering European users liable for infringement unless they can come within one of a raft of 20 or so very narrow exemptions, all bar one of them only voluntary for Member States to implement into their national legislation.56 Paracopyright thus poses two dangers for the evolution of a coherent competition policy. The first is that courts and regulators allow the automatic extension of copyright’s already over-privileged status in many competition regimes to this new area. The second is that by allowing copyright owners to thwart or bypass fair dealing and fair use rules, paracopyright severely limits copyright’s ability to control anti-competitive behaviour through its own internal mechanisms.57
Competition lawyers are now used to asking whether and how competition rules should be adjusted to take account of the actual or supposed idiosyncrasies of intellectual property. The question after all has practical outcomes. Intellectual property lawyers seldom have occasion to put the question in reverse and ask whether the content of an intellectual property right should be adjusted to reflect the needs of competition policy.
One of the more unfortunate consequences of treating property and regulation as polar opposites is that this way of thinking obscures the extent to which individual intellectual property regimes themselves reflect (or can be made to reflect) competition values. Very occasionally this is explicit, as in the recent enactment in Australia of provisions in the patent statute enabling the Federal Court to issue a compulsory licence58 and revoke a patent after that grant if the patentee is acting in contravention of that jurisdiction’s competition statute.59 More commonly, competition values lie silent within other provisions until teased out by courts during infringement proceedings.
(a) Compulsory licensing provisions in patent and design law
Despite a worldwide trend against compulsory licences,60 these can still be obtained under the patent61 and design62 laws of various countries where the right owner refuses to license at all, or will license only on unreasonable terms. Usually the applicant has to show that the design or patent is not being worked within the jurisdiction, and in some cases that the development of downstream markets is unfairly prejudiced by the market being supplied from outside the country on unreasonable terms. The licences are generally subject to an array of onerous conditions and qualifications to ensure compliance with TRIPS. The tests of reasonableness and abuse in these provisions tend to be restrictively interpreted.63
(b) Collective copyright licensing
Copyright licences and licences of performers’ rights are not uncommonly administered by collecting agencies on behalf of dispersed groups of right holders. In jurisdictions such as the United Kingdom and New Zealand such schemes are subject to confirmation and variation by a specialist tribunal at the behest of actual or would-be licensees.64 The tests applied by such bodies are those of reasonableness and non-discrimination, and they tend to focus on the interests of stakeholders represented before them rather than on the impact of those stakeholders’ actions on the competitive process itself. Whether and to what extent this form of mini-regulation is subject to competition principles is seldom made clear.
(c) Abuse of rights doctrines
In the United States the courts have extracted from intellectual property statutes an unarticulated first premise that these statutes not be interpreted in ways at odds with the statutes’ underlying objectives. At the forefront of these objectives, some courts have ruled, are competitive markets. None of this, it must be stressed, hangs on the words used in the individual intellectual property statutes and appears to have had purely remedial origins. Loosely gathered together under the ‘abuse of rights’ label, these principles of construction (for that is what they are) have drawn heavily on the forbiddingly named and now abandoned ‘Nine No-Nos’65 under which United States regulators sought to outlaw particular licensing practices outright. The fit with antitrust law is less than perfect, however, even after further legislative tinkering in the case of patents. The doctrine of patent abuse which we explore in the next chapter66 has been much criticised67 and likened to a court-run scheme of royalty-free licences for infringers.68 Copyright misuse tends to blend into wider copyright doctrines such as fair dealing69 or the idea–expression dichotomy. Defendants seem less likely to be successful in copyright abuse cases than in patent cases.
Commonwealth courts have been much slower to invoke abuse of rights theories, at least in cases involving intellectual property. In these jurisdictions their use appears to have been confined to two limited situations. The first involves mostly unsuccessful attempts to invoke the so-called ‘spare parts’ rule enunciated by the House of Lords in British Leyland v Armstrong, Patents Ltd.70 The second concerns isolated refusal to protect business confidences where potential breaches of competition law are said to be involved. The former has been outflanked, at least on its United Kingdom home ground, by the construction of a statutory bypass.71 Even in its heyday the Law Lords’ decision found few emulators among other Commonwealth judges,72 or indeed any application outside the narrow field of indirect copyright protection for mass-produced articles.73
The general public interest defence to breach of confidence might seem at first sight a more promising form of competition control. Certainly it has been applied in cases where the confidence sought to be protected involved alleged breaches of competition statutes,74 but its limits are otherwise unclear. It may apply only to intentional breaches, for example, and in some cases appears to operate as an element of the cause of action rather than a defence.
Whatever their future in particular jurisdictions, abuse of right doctrines are an undesirable development. The benefits they confer on one group of stakeholders (potential infringers) seem out of proportion to any proven harm to the competitive process. The existence of two sets of related, but on key issues divergent, rules on the American pattern is not reassuring. Judges who are called on to deal with this particular side wind in the course of ordinary infringement proceedings will be doing so without the procedural and evidentiary aids more usual in full-blown competition litigation. In particular, the opportunity for input from economists would appear to be limited, at least in jurisdictions outside the United States. Adjectival issues apart, the chief difficulty with the abuse of rights approach is conceptual. It is nothing more than the ‘right equals market’ fallacy writ large.
(d) Core concepts as competition enhancers
Every intellectual property right has a set of inbuilt constraints intended to limit its scope and potential application. Some are limited in time. Others, while potentially immortal, may be lost through disclosure, inaction or inappropriate use. In addition to temporal limitations, each set of rights has a core concept which restricts the subject matter to which the right may be applied. Patents and copyright were not originally designed to protect ideas as such. Reputational rights were not intended to be cut loose from the reputations they were created to defend. In this prelapsarian world, facts were thought to be free, and mere information was protected only in the service of some notion of fair dealing (however inarticulate) or because of the original form in which it was expressed. It was also thought important that the boundaries between the various forms of intellectual property be preserved so that core concepts did not bleed into each other in ways which defeated each right’s distinctive purpose. Copyright was not to be used to confer patent-like protection on functional aspects of works without the patent penalties of disclosure and registration. Know-how had to be kept secret, or be constrained by patent claims and specifications. Courts and legislators were content to leave the shape and get-up of goods to be protected by copyright or registered design law rather than have those things become reputational indicators in themselves. Judges were generally at pains to preserve these distinctions whenever it was obvious they were under threat (this was not always the case). And legislators generally did not disturb this happy state of affairs. Intellectual property statutes remained discrete and internally self-referencing.
None of the internal controls described above has any avowedly pro-competitive purpose. Indeed most of them long pre-date the existence of competition law in its modern form. They may, however, be given an economic rationale after the event, and most have been. Ideas are not protected, it is said, because in many cases no particular incentive is needed to produce them,75 or because of the massive deadweight loss which occurs when ideas are withdrawn from the public domain.76 Investment and effort are not protected as such, since this will often reward careless business decisions, discourage proper assessment of risk77 and pointlessly redistribute income from users to owners.78 Whatever the validity of these ex post facto justifications, there is no doubt that the internal controls described above do limit, however crudely and blindly, right owners’ ability to act in anti-competitive ways.
Absent an express statutory power to intervene in infringement proceedings or suspend the competition inquiry until the conclusion of these proceedings, enforcement agencies could be forgiven for ignoring intellectual property’s own self-correcting mechanisms. To do so, however, could distort the competition analysis by overstating the market power attributable to a particular intellectual property right. Equally, though, any erosion of intellectual property’s internal controls should give competition regulators, legislators and policy makers pause for thought. Elsewhere in this book we have argued strongly that intellectual property law cannot simply be assumed always to get it right in competition terms, but neither should we assume that it always gets it wrong. Intellectual property’s internal self-limiting mechanisms have to be part of the analysis.79
As seen in chapter one,80 the appropriateness or otherwise of coerced licensing is sometimes sought to be assessed in terms of a hierarchy of rights, with ‘strong’ rights being more entitled to protection against competition scrutiny than ‘weak’ rights. While we argue that this distinction, even if viable, is a rhetorical distraction in competition terms, the question also needs to be asked whether any such hierarchy exists in the first place. What do people mean when they classify rights as ‘strong’ or ‘weak’, and does the distinction make any sense in economic terms? In fact two quite different hierarchies are being put forward here. The first is predicated on the assumption that some types of intellectual property are not only ‘stronger’ but desirably so.81 This entirely normative polarity is said to operate not only between intellectual property regimes (trade secrets versus patents82), but also within them (‘basic’ versus ‘applied’ innovation).83 This is a matter of social policy and thus susceptible to economic analysis.84 The second hierarchy is predicated on being able to draw a distinction between claims for intellectual property protection which are clear-cut in terms of the applicable law and others where the application of that law to the activity or output under investigation is uncertain. Here the economists have to make sense of a construct that is not of their making and of no obvious social utility (unless it is sought to argue that the unclear right should never have existed in the first place). Since all intellectual property rights have their fuzzy edges somewhere, it is not quite clear where this argument is intended to lead, nor why competition policy should be the lever by which clarity is brought to a particular intellectual property regime.
In a perfect world the breadth and duration of intellectual property protection would exactly match the underlying justification. This is not, however, the way in which intellectual property works. Nor could it work that way. The amount of protection delivered cannot be adjusted on a case-by-case basis. Uniformly applicable rules will over-protect in some cases, under-protect in others. Pharmaceuticals may need 20 years of protection, fashion design five or fewer. Copyright in fictional works may need to last for 70 years plus the life of author to be effective. This is less obviously so in the case of copyright protection for computer programs which may need less. The resulting mismatch between protection and justification is said to be counterbalanced by the consequent savings in transaction costs. This unfortunately leaves two questions unanswered. The first is this: If some forms of intellectual property (notably copyright) do recognise different levels of protection for different kinds of activity, why could not similar multilevels be provided in the case of patents? There would still be a mismatch, but a much smaller one.
The second unanswered question is more significant: If over-protection is a regrettable necessity in competition terms, is that necessity justified only to the extent that intellectual property rights stay within their traditional bounds and their internal competition controls remain intact? If existing rules are a less than perfect fit with underlying economic rationales, it is not obvious where any further lowering of transaction costs or gains in innovative efficiencies are to come from if the area of protection is increased.
The debate between quantifiers and categorisers outlined in chapter three has particular resonance when an intellectual property right is introduced into the mix. Categorisers, for example, would say that if the right was lawfully acquired, it cannot be a barrier to entry because anyone could have acquired it. Nor can the right be a strategic barrier. Even accepting the categorisers’ logic, rights such as trade secrets are arguably self-constituting, or so the General Court accepted in its Microsoft decision.85
(a) The weak case for intellectual property as a barrier to entry
Attempts are sometimes made to argue that the costs of defeating or bypassing an intellectual property right are deterrents to entry by competitors. Such costs include litigation expenses, research and development of alternative products or processes, and the money expended to fight the branding implications of a trade mark or other reputational right.86 These are barriers to entry only in so far as they:
b) bring no collateral advantage beyond bypassing the intellectual property right; and
c) are quantifiably high enough to deter entry,88 or qualitatively different from the costs already borne by the incumbent.89
Only rarely will the costs of bypassing or contesting the right qualify on all three grounds. Most would-be entrants would spend something on product differentiation or research and development even if the incumbent held no intellectual property rights, and some such expenditure would be recoverable on exit from the market. Even litigation costs are a problematic barrier, since it cannot be assumed that competition with the incumbent necessarily means contesting the right. Potential competitors may be able to compete using non-infringing products or processes.90 The structural uncertainties associated with vague rules are perhaps more plausibly described as entry barriers. They certainly increase the right holder’s opportunities for gaming,91 but the cure cannot lie in clarifying the content of the right by over-extending it. The gaming would then be over, but only because the right holder had won the game.
(b) The strong case for intellectual property as a barrier to entry
Intellectual property is much more convincingly described as a barrier to entry not when it raises the costs of potential infringers per se, but when it raises them to levels which are quantitatively high enough (for the ‘quantifiers’) or qualitatively different enough (according to the ‘categorisers’) to cause real problems in real cases. This is most likely to occur when the right in question is used in such a way as to:
a) inhibit the creation of new markets by limiting derivative or developmental use of protected products or processes;
b) facilitate market segmentation by erecting geographic obstacles to product movement. Parallel importing restrictions, because they are built into intellectual property regimes themselves, can have the same effect as a purely contractual territorial restriction (the line between action and inaction is wobblier than usual here);
c) permit right holders to use the power conferred by the right to deter entry into markets not covered by the right, or requiring would-be entrants to enter all markets simultaneously or not at all.
In all of these cases, both the existence of the intellectual property right and its actual or threatened exercise have the potential to deter entry. Both are therefore fit subjects for competition scrutiny. Whether the barrier is structural or behavioural is also of little moment. Its effect is the same.92 Equally, the height of the barrier cannot be disconnected from the issue of over-protection. An intellectual property right that delivers to right holders more protection than innovative efficiency requires will by definition discourage more efficient competitors. The right holder’s motive in acquiring or enforcing the right is irrelevant, as is the fact that the right was acquired or exercised perfectly legally.93 Again, where a patent is obtained by fraud, it is not the fraud that constitutes the barrier to entry but the fact that the patent generates no efficiency gains of its own to offset any efficiency losses caused when competitors are tricked into staying out of the market. To focus on the fraud, as judges in the United States currently do,94 is to miss the point. Any patent found to be invalid post-grant will generate the same problems, no matter how pure the motive of the patentee may have been in taking out the patent. The same logic applies when patentees seek to see off or thwart competing technologies by threatening to bring patent infringement proceedings in relation to activities that are perfectly lawful in terms of the patent as granted. Competition law should be invoked not to punish the bluffer but to encourage an entry that might not otherwise have occurred. A mistaken but honest belief on the patentee’s part as to the true scope of the patent will still frighten off would-be entrants. The existence of anti-bluffing provisions in some jurisdictions’ patent law does not alter this dynamic.95
Economic explanations as to why something like an intellectual property regime had to exist were for a long time stable if never quite unanimous.96 In more recent years the expansion of intellectual property protection has led to a divergence of views among economists.
All economists agree that intellectual property rights increase price and reduce consumer choice, at least in the short term.97