دادخواهی ثبت اختراع هنگامی که نوآوری بصورت گروهی است

This paper studies the effect of litigation as a way to enforce patents when firms hold private information. Patent protection granted by courts affect the entry, settlement and litigation decisions of future innovations. The model is broadly consistent with recent empirical evidence. We show that higher protection might be detrimental to the patentholder since it reduces entry of infringers that would otherwise license the patent. We argue that this is more likely to be the case for large improvements or large litigation costs. Finally, we compare the effects of Preemptive Injunctive Relief on innovation and litigation.

Patents are imperfect assignments of property rights on innovation. The uncertainty surrounding the innovation process usually does not allow a precise specification of the range of innovations protected, provoking extensive infringement of these rights. These unavoidable disputes have made of litigation an integral part of the patent system, with important consequences for the incentives to innovate and patent. The costs of patent enforcement are substantial. In the US, from October 1996 to September 1997, 1530 lawsuits involving patents were reported, although the number of disputes is much higher.1 The associated legal costs can represent as much as 25% of firms’ basic R&D expenses. For this reason recent debates have emphasized the observation that the supervision of patent applications is not very rigorous, and as a result, litigation is mainly used to obtain a judicial definition of the boundaries of the protection that firms obtain. An important dimension of a patent is its breadth or scope, understood as the range of competing products and processes that are covered by the patent. Although breadth is considered an essential component, the ambiguity of the concept has not allowed a general formulation. Most of the literature assumes that the infringement of a patent’s breadth is evident. 2 Such a setup makes the existence of litigation irrelevant. In reality, the boundaries of the protection associated with a patent are rather blurry, and the process through which courts assess whether an invention infringes an existing patent or not is subject to uncertainty. The existence of courts and their rulings can be interpreted as a way to aggregate the relevant features of a patent. In other words, the breadth of a patent is endogenously determined by the ‘litigation technology.’ When innovation is cumulative, Green and Scotchmer (1995), Chang (1995) and O’Donoghue et al. (1998) show that the optimal patent must provide claims over future research. Innovations represent permanent improvements over the stock of knowledge, but firms can only partially appropriate the value of the improvement through the sale of their products, since future producers will compete away the profits. By protecting the patentees against future innovation, the life of their monopoly power is extended. One of main concerns of those papers is that in general more protection benefits the patentholder, but it reduces the incentives for future research to be undertaken. Therefore, the design of the optimal patent must counterbalance these effects. These papers usually interpret patent breadth as the minimum size of the following invention that will be allowed and consider it a policy variable. One of the contributions of this paper is to show that the way courts enforce patents defines endogenously not only the decisions to enter of future inventors but also the licensing and litigation choices. In particular, the real protection that a patent grants to an innovator depends on the quality of the invention, and it is for bigger inventions that more patent breadth can reduce profits. Previous models of patent litigation, such as Meurer (1989) and Aoki and Hu (1999) deal with the case of a pure imitator. However, when the protection is related to the use that future innovators will make of the patented research, the definition of breadth becomes particularly difficult. While we do not attempt to provide a formal definition of patent breadth, we relate it to the probability that the patentholder succeeds in court. Following the legal arguments presented by Merges and Nelson (1990), this probability also depends on the size of future improvements. Consequently, patents provide less chances in court against better infringers, reducing the protection. With this caveat, we argue that patent breadth is related to the size of the future innovator. We assume that firms have private information about the quality of their inventions and courts might make mistakes when judging the conformity of an innovation with the existing patents. For this reason some patent disputes are settled in court. The legal procedure helps reveal some of the private information. As a result, there is some litigation despite it is costly and firms behave optimally.3 Moreover, the cost of litigation deters entry of some competitors that otherwise would try to innovate, while in other situations it gives incentives to reach a settlement between the patentholder and the potential infringers. We also show that there is an ambiguous relationship between our interpretation of patent protection and the probability of a patent being litigated, which seems to conform with empirical evidence provided by Lanjouw and Schankerman (1997) and Lerner (1994). In our framework, increasing the protection of the patentee does not necessarily imply an increase in the profits that he can obtain. The reason is that an important share of the revenue of the patentholder originates from future research. More protection implies that the patentee will have more chances in court, and so he can demand higher licensing fees. On the other hand, with more protection fewer firms are willing to build on that patent, reducing the licensing proceeds. When protection is small, the first effect dominates, but as protection increases and the number of potential licensees decreases, the second effect becomes dominant. This second effect is particularly important for innovators with a more valuable patent. The reason is that the value of the patent is related to the compensation that the infringer will be forced to pay if courts rule against her, and therefore, it becomes an additional barrier to entry, contributing to the real protection of the patentee. This excessive protection discourages more small improvements that would benefit the patentholder through licensing. An interesting policy implication of this setup is that valuable inventions should obtain weaker protection, allowing the patentholder to commit to license the patent to future entrants. It is important to emphasize that Green and Scotchmer (1995) also show that protection might hurt the patentholder when the negotiation with a future innovator is conducted before the investment occurs. The result in this paper does not rely on this assumption, but rather on the existence of patent litigation. We also show that this effect can occur when legal costs are large, since they are in fact a barrier to entry. Finally, we study the effects of a Preliminary Injunctive Relief. This is a legal motion that prevents the infringer to produce while the dispute is decided in court. One of the implications of this motion is the change in the decisions to litigate. We show that in this case patent protection is independent of the size of the original invention since the infringer is not required to compensate the patentholder for the foregone profits. Compared to the previous case, this legal motion seems to benefit relatively more innovators with more valuable patents, as opposed to the original intention of the law. The model is presented in 2 and 3. Section 4 compares the results with a setup without litigation. Section 5 considers the case of Preemptive Injunctive Relief and Section 6 concludes. All the results are proven in Appendix A.

The goal of this paper was to study how the enforcement of Intellectual Property Rights affects the decisions to innovate. While other papers, such as Aoki and Hu (1999), have considered the effect of litigation on the value of patents, they usually study the trade-off that imitation constitutes between adding competition and the incentives to innovate. Nevertheless, patents also affect the decisions of future innovators. This paper tries to address this issue. We have presented a model of patent settlement that takes explicitly into account the consequences of litigation, or the threat of using it. The results show that with private information on the quality of new ideas there is in equilibrium a certain amount of litigation, together with the settlement of low quality innovations. We stress the fact that the protection that patents grant is intimately related to the way in which it is enforced by courts. The outcome of the legal procedure is used by the patentee as a threat to obtain better terms in the negotiation with the infringer. However, more protection has a downside. The patentee cannot commit beforehand to offer a particular license, and therefore, as courts become more favorable to patentees, only the best infringers will enter. Since an important part of the value of a patent stems from licensing to future innovators more protection does not necessarily mean more profits for the patentee. This effect is particularly important on inventions that have a high value, because they already deter more entry. By explicitly modeling the legal institutions, the model allows us to compare different legal systems. In particular, we study the consequences of a legal motion that has become popular in recent years, the Preemptive Injunctive Relief. Its use seems to benefit relatively more better inventors by increasing their licensing revenue. The answers provided in this paper are far from complete. We show that the effects of patent protection and litigation costs are more complex than what is usually considered. However, not much has been said about the socially optimal amount of protection. Some preliminary numerical results show that the Preemptive Injunctive Relief should be granted to innovations that have a higher market value, together with more protection to the corresponding patentees. Finally, in this paper we have abstracted from the dynamic environment associated to the negotiation among the parties, and for simplicity we have collapsed all the process in a take-it-or-leave-it offer. An interesting extension would include the study of the dynamic decisions undertaken by firms. In particular, the output market can be understood as a way to obtain noisy signals of the quality of the improvement. As a result, the decisions of the patentee to produce, the timing of the licensing agreements and trials, as well as the decisions taken by courts will depend on this signal extraction mechanism. The speed of the trials is also an interesting and unexplored policy variable that will affect the use and optimality of the Preemptive Injunctive Relief.