Intellectual property (IP) rights are, by definition, rights to exclude others from the use of that intellectual property. Society grants those rights in the hopes of incentivizing innovation. There is a trade-off between the long-term benefit of increased innovation and the short-term cost of allowing the innovator to (potentially) exercise market power. In order to obtain the benefits of IP rights, innovators must follow certain associated restrictions. For example, patents are granted for a limited time period and in exchange for the knowledge of the innovation being made public. Society must also beware misuse of the system—attempts to obtain market power through the IP system without complying with the relevant rules. Since intellectual property rights are rights to exclude others from the use of intellectual property, it is not surprising that antitrust claims related to exclusionary activity are often asserted in IP litigation.
For example, one class of antitrust claims named after a Supreme Court decision referred to as Walker Process claims involves patents obtained through fraud on the patent office. These claims involve showings of both fraud on the patent office and the standard elements of a Sherman Act Section 2 claim. Another class of claims—Handgards claims—involves sham or bad faith litigation of patents that the patent-holder knows to be invalid or unenforceable. Again, these claims involve showings of both knowledge that the patent is invalid or unenforceable as well as the standard elements of a Sherman Act Section 2 claim. However, the fact that such antitrust claims spring readily to mind because of the inherently exclusionary nature of IP rights does not necessarily mean there is a sound economic basis for them in any particular case. Indeed, some of the economics involved in these matters can be counterintuitive. This article walks through a variety of economic issues that may be mishandled or misunderstood in an IP antitrust claim.
There is sometimes a perception that an IP holder must be a monopolist in an antitrust sense because they have a legal right to prevent others from using that IP, e.g., a patented technology. That perception is incorrect for at least two reasons. First, the IP rights may not cover anything of economic importance. Second, even if they do, attempts to assert the rights may be unsuccessful (e.g., a patent is found invalid or not infringed).
For example, Prof. Shapiro explains that a patent is a:
bundle of uncertain and imperfect rights. . . . [that] are typically far less valuable than would be idealized “ironclad” patent rights. . . . Therefore, the patentholder is not “entitled” to negotiate a monopoly outcome, just because the patentholder asserts that its patent is valid and infringed by a particular rival. Rather, the patentholder’s rights are calibrated according to the likelihood that the patentholder would win the patent litigation, and the extent of exclusion that such a victory would permit.
In this example, the extent of (potential) exclusion is the patent scope and the likelihood a patentholder would win the patent litigation is the patent strength.
With respect to scope, patents (or other IP rights) may cover only very limited aspects of a product that do not correspond to an antitrust market. In fact, IP rights may not cover a product at all—they might only pertain to one particular way of making one particular part of a product. Even if the rights cover a product, it may be one of many similar competing products. Only a small percentage of patents have economic significance. As explained by the U.S. Department of Justice and Federal Trade Commission:
The Agencies will not presume that a patent, copyright, or trade secret necessarily confers market power upon its owner. Although the intellectual property right confers the power to exclude with respect to the specific product, process, or work in question, there will often be sufficient actual or potential close substitutes for such product, process, or work to prevent the exercise of market power.
With respect to IP strength, IP rights are probabilistic. They are complicated and open to interpretation and debate, and disagreements about the validity and extent of those rights are common. That means, for example, every patent granted by the Patent and Trademark Office is still subject to uncertainty as to what it actually covers and whether it is valid or not. Profs. Lemley and Shapiro explain that in the patent context:
The patent system does not grant an absolute right to inventors to exclude others from practicing their inventions[.] . . . Rather, the patent system gives the patent holder a right to try to exclude others by asserting its patent against them in court. The actual scope of a patent right, and even whether the right will withstand litigation at all, are uncertain and contingent questions. This uncertainty is not an accident or mistake. Rather, it is an inherent part of our patent system . . .
The uncertain nature of IP rights has a number of important economic implications.
One implication of the uncertain nature of IP rights is that invalidity and fraud are not the same thing. There may be a temptation to conclude that if a firm attempts to enforce an IP right and that right is subsequently held to be invalid, then the firm must have done something wrong (in the form of an illegal or fraudulent act) when obtaining the IP right initially. That is not the case. The uncertain nature of IP rights and the ability to test them in litigation are inherent to the IP system. A finding of invalidity, in whole or in part, does not mean that the IP holder necessarily did anything wrong. For example, a patent-holder is entitled to try and enforce their patent through litigation. The patent-holder may lose, but such a loss in and of itself does not retroactively make the patent-holder’s enforcement effort anticompetitive.
Antitrust claims involving IP rights often include claims of fraud in obtaining the IP right. Economic tensions can arise in claims about attempts to enforce fraudulently obtained IP rights. Committing fraud to obtain the IP right in the first place is a distinct act from enforcing that IP right to reduce competition (and, as discussed earlier, Walker Process and Handgards claims involve both showing a problem with the IP right itself as well as the standard elements of a Sherman Act Section 2 claim). Claims as to the first act can have implications with respect to the plausibility of claims as to the second act. Say that a firm commits fraud to obtain an IP right that it hopes to use to exclude rivals (or otherwise raise rivals’ costs). Then what? The IP right is only useful to the extent it can be enforced. But if the fraud is discovered, the IP holder may not only fail in its enforcement attempt but may be subject to penalties such as paying the other parties’ litigation costs and trebled antitrust damages. Any effort to enforce a fraudulently-obtained IP right runs the risk of the fraud being detected, which may result in significant costs to the IP holder.
Now, the IP holder may hope that the threat of litigation alone will be sufficient to achieve its goals. But as noted above, IP rights are inherently uncertain, and firms are well aware of this uncertainty. Any firms threatened by asserted IP rights have incentives to investigate the IP and test its strength, which will increase the chances of the IP holder’s fraud being discovered. The greater the economic stakes for the firms threatened, the greater the incentives to investigate the asserted IP. If a threatened firm calls the IP holder’s bluff and litigation commences (such as an enforcement effort by the IP holder or an Inter Partes Review initiated by the threatened firm), one would expect that the IP holder would be concerned about its fraud being exposed in the crucible of litigation, and so might be reluctant to press forward in that process.
Of course, IP fraud can and does occur, and the possibility should be taken seriously. However, consider an antitrust claim along the lines sketched above that includes purported evidence of clearcut fraud obtained through a Google search of key words from the patent claims. If evidence of fraud is so easily obtainable and the evidence is in fact clearcut, then why would the IP holder try to enforce the IP right at all since it would seem to be trivial to detect and reveal the fraud? That is, claims as to the existence of the fraud may be in tension with claims as to the plausibility of enforcing a fraudulent IP right. To understand the economic framework involved in an antitrust claim involving fraud, and whether that claim is economically plausible, it helps to ask why the IP holder might have expected to get away with the fraud, especially if the claim involves monopolizing a highly profitable industry where the other party potentially has a lot of money at stake, and why the IP holder might persist in litigation even after being confronted with evidence of the fraud.
If an IP right is the alleged tool through which anticompetitive harm was inflicted, one must consider whether it is economically plausible that the tool would be sufficient to achieve the claimed harm, such as monopolization of a market. Here, it is helpful to understand the size of the alleged harm relative to the likelihood of detection of fraud, the likelihood of the IP rights being upheld even if fraud is not detected, and the scope of the IP rights if they are upheld. Overall, is it economically plausible that the IP rights, in light of those three dimensions, could result in the claimed antitrust harm?
Beginning with the likelihood of detection of fraud, as noted in the prior section, the higher the economic stakes, the more likely a party against which the IP is asserted will dig into the IP and discover fraud. If billions of dollars are at stake, the other party would be incented to look really hard at the IP. The question is further complicated since evidence that can at least be characterized as indicating fraud presumably has been unearthed for an antitrust claim alleging fraud in the first place.
To be clear, plausible answers may exist for all these questions. Economically plausible reasons can explain how an IP holder might commit fraud and expect to get away with it. Gains to the IP holder might be large but potential gains to the party against which the IP is asserted are sufficiently small that it is simply not worth fighting over. Also, potential gains to excluded parties may be collectively large but diffused across many parties, such that no one party finds it worthwhile to devote resources sufficient for uncovering the fraud. It might also be the case that the fraud was well orchestrated such that the probability of finding fraud was small absent new information coming to light, e.g., a whistle-blower. Overall, IP fraud can certainly be used anticompetitively, but it is important to evaluate the internal economic consistency of claims of fraud and antitrust harm.
Next, even if actual fraud were not detected the assertion of the IP right still might not be upheld—it could be found invalid or inapplicable (e.g., not infringed), or both. If there were good reasons to expect that the IP right would be unlikely to be upheld on its own merits, it would be a less effective tool for anticompetitive purposes. IP challenges are often mounted specifically because the challengers believe they have good chances of prevailing. The practitioner can evaluate statements and actions of parties involved that might shed light on the perceived strength of the IP right in terms of likelihood of being upheld.
Finally, even if the IP right were upheld, is its scope sufficient for the job? That is, even if the IP holder managed to get an injunction concerning the IP right, would that be sufficient to result in significant antitrust harm? As previously discussed, there is no presumption that any given IP right is sufficient to provide market power. In order to demonstrate significant antitrust harm, an antitrust claimant would like to show that assertion of an IP right would be sufficient in scope to exclude rivals entirely, or at least to make them sufficiently less competitive so as to impact the overall market.
To summarize using a patent example, a narrowly defined patent widely perceived as both unlikely to be upheld and potentially fraudulently-obtained represents an unlikely tool for successfully monopolizing a market. Conversely, a broad patent that was widely feared in an industry until a whistle-blower later revealed it was fraudulently obtained might be a very plausible tool.
IP rights grant the right to exclude in a legal sense, but that does not indicate actual exclusion in an economic antitrust sense. If a firm has in fact been subject to an injunction or exclusionary order and believes it has a valid antitrust claim, then it can point to an actual exclusion of a competitor—itself (although that leaves open the question of whether such exclusion was sufficient to impact competition in a properly defined antitrust market). Similarly, a firm being sued over an allegedly fraudulent IP right may be able to point to prior instances where other firms exited after being challenged or faced significant competitive disadvantages.
On the other hand, if challenged firms continued to participate in the market either by settlements, by dropping challenged features, or by introducing non-infringing alternatives, and there was no obvious impact on market-wide output or price, then an antitrust claim should be prepared to explain why nonetheless there has been antitrust harm. Put another way, if no firm or product has ever actually been excluded, then an antitrust claim may have an uphill battle with respect to any claims of historical antitrust harm. Prospective claims need not worry about showing historical harm, but may also not have any actual antitrust harm to support damages claims.
In antitrust cases it is often useful to question whether actual outcomes have mapped to the theory of harm. For example, it is easier to claim a consummated merger was anticompetitive if it can be associated with increases in price and reductions in output in a properly defined antitrust market. The same is true with respect to IP-related antitrust claims. For example, exclusionary behavior is presumably motivated by the IP holder’s desire to better its own position. Is there evidence that the IP holder actually benefited from the alleged exclusionary behavior? Put another way, where is the “pot of gold” that the IP holder is presumably chasing? Is there some obvious change in the IP holder’s economic position that can be tied to the alleged exclusionary activity? If not, is there a plausible reason why one might not expect to see a change (e.g., was the IP holder preventing deterioration of an existing position rather than enhancing its position)?
In a similar vein, are there natural experiments that might shed light on the antitrust claim? For example, if a patent that is alleged to have been used to monopolize a market has expired, what has happened in the market since expiration? If we suddenly observe entry or the former IP holder’s profits or market position markedly decline following patent expiration, that evidence is consistent with a finding of monopolization. Conversely, if there is no observable change in the market or market position of the former IP holder after the allegedly monopolizing patent expires, that suggests the patent had no significant impact on the market.
Practitioners may also wish to consider the degree to which interpretation of market evidence depends on findings of liability. In particular, if an IP right is legitimate and economically significant, the IP holder may expect to earn significant economic returns and may be harmed if its IP is stolen. Enforcing those rights is a normal operation within the IP system. On the other hand, if the IP right was fraudulently obtained, then those same returns may be pointed to as evidence of anticompetitive harm. For example, one firm might claim another has stolen its trade secrets and entered its market, resulting in price erosion and lost profits. The allegedly misappropriating firm might counter that the first firm is trying to eliminate a significant competitor and harm competition. Both parties might point to the same economic evidence of market impact from entry, agreeing that entry significantly impacted the market, but interpreting that evidence through very different lenses.
Antitrust harm is sometimes alleged to occur through sham litigation raising rivals’ costs. Such claims should be clear about the economic mechanism allegedly leading to raising rivals’ costs. In particular, how would litigation costs impact company operations and why would the rival be harmed relative to the IP holder given that they are both paying litigation costs?
First, economics is clear that marginal costs are the key determinants of prices, not fixed costs. Costs incurred for a discrete lawsuit are likely to be fixed costs, not part of a firm’s marginal costs of production. Therefore, as a matter of economic theory, one would not expect legal expenses unconnected to production decisions to cause a firm to increase its prices. The firm might be harmed by having to pay those costs, but harm to a competitor is not the same thing as harm to competition. However, in certain circumstances litigation costs might impact a firm’s competitive position. For example, sufficiently large costs might bankrupt a firm. An obvious question, then, is to evaluate the size of any legal costs relative to a firm’s operations. While litigation costs are expensive in absolute terms, they are often small relative to firms’ overall operations, and firms—particularly large firms—are often engaged in multiple litigations on a regular basis in the course of ordinary business.
Second, this sort of raising rivals’ costs argument can contain an internal economic inconsistency. Because the IP holder has to pay legal costs as well, under this sort of theory of harm there is no reason to assume that the rival is at a greater disadvantage than the IP holder. For example, it could be economically irrational for an IP holder to seek to raise its rival’s costs through anticompetitive litigation when the IP holder is exposed not only to equally high direct costs but the risk of even higher costs because of the possibilities of fee-shifting or trebled antitrust liability. The economic tension can be compounded when applied across multiple litigations. Sometimes IP holders are accused of using serial litigation to exclude rivals. But that would mean that while any individual rival to the IP holder has only had to pay the costs to defend a single lawsuit, the IP holder has had to pay the costs to prosecute all of the lawsuits. And if those lawsuits were all based on fraudulently obtained IP, the IP holder could face additional exposure through potential fee shifting and antitrust liability. Thus, if fixed legal costs did somehow make a firm a less effective competitor, then the IP holder might face higher total legal bills than any of its individual rivals, and so the rivals might actually have benefited as competitors relative to the IP holder.
Sometimes a firm against which an IP right is being asserted may claim that the IP holder has previously excluded other firms, gaining monopoly power in a market, and is now turning its attention to the firm bringing the antitrust claim. One counterintuitive phenomenon in this situation is that the firm making the antitrust claim actually might have benefitted from the alleged anticompetitive conduct. That is, if the firm making the claim has been participating in the market while other firms were excluded, that firm may actually have been better off than in a but-for world with no exclusion. This is a corollary of the antitrust axiom that harm to a particular competitor does not mean there is harm to competition—harm to competition does not necessarily mean harm to a particular competitor. In the antitrust merger context, the analogy is that a merger that reduces competition in an industry may harm consumers but benefit other firms in the industry.
To illustrate this issue in the IP context, assume that in the actual world three firms are competing. Further assume that in the but-for world with no exclusion five firms would have competed. Assume that the difference is due to exclusionary activity by one of the three firms. If there is less competition in the market due to the exclusionary activity, then, by definition, total unit sales would be lower in the actual world than in the but-for world (output is lower) but per unit prices would be higher (price is higher). The excluding firm would be expected to be better off or it would not engage in the activity. The two firms excluded from the market are presumably worse off. However, the remaining two firms that are still in the market despite the exclusionary activity actually may be better off. They earn higher prices on their sales, which is a benefit to them relative to the but-for world. Whether they are net better or worse off hinges largely on how their unit sales differ (and, if raising rivals’ costs claims are present, may also depend on how their costs are impacted). They might be better off even with lower unit sales if the increase in per unit price is high enough, but it is also possible that their unit sales could be higher than in the but-for world. That is, the exclusion of two firms will reduce industry output, but the unit sales of all three remaining firms individually may still be higher than in the but-for world.
Thinking through the economic incentives inherent in an IP antitrust case can help to focus the analyses to bring to bear on pertinent issues in a case. Not every economic point will be relevant in every case, but thinking about a case and working through the economic logic outlined in this article can help attorneys understand the claims more clearly, whatever the specifics.
