26 October 2012

UK Patent Litigation Costs

The important 'Patent Litigation in the UK' (LSE Legal Studies Working Paper No. 12/2012) by Christian Helmers & Luke McDonagh reports on
a dataset that contains the complete set of patent cases filed at the courts in England and Wales during the period 2000-2008. The data cover all types of patent-related cases brought before the Patents County Court, the Patents Court at the High Court, the Court of Appeal, as well as the House of Lords/Supreme Court. We combine the detailed information on court cases with information on the patents in dispute as well as firm-level data for the litigating parties. We employ the dataset to analyze characteristics of the court cases, = litigating parties, as well as the contested patents. We also provide detailed discussion of the cases that were heard by the House of Lords/Supreme Court as well as of the costs involved in patent litigation before the courts in England and Wales. 
The authors comment that
Patent litigation has risen to dubious worldwide fame ever since the ‘global patent wars’ broke out between the giants in the information and communication technology (ICT) industry. Moreover, while the patent wars have clearly led to a surge in patent court cases in all major markets, it is also clear that the publicity surrounding these court battles has shed light on the fact that in some industries patent litigation is rapidly becoming a primary mechanism for moving issues of competition from the product market into the court room. Concern about the disruptive potential of patent litigation has also increased in recent years due to the increased participation in litigation of so-called ‘patent trolls’, sometimes referred to as ‘patent assertion entities’ (PAEs) or ‘non-producing entities’ (NPEs) (Bessen et al., 2011). The prominence of these cases in the eyes of the media has reinforced a broader public debate concerning innovation within the patent system and the role of litigation.
The functioning of the patent system fundamentally depends upon the ability of patentees to enforce their rights, if necessary with the help of courts. In particular, the role of patents in the dissemination of information by requiring the disclosure of detailed technical information relies on enforceable patent rights in order to provide firms with the incentive to make this information publicly accessible. Patents are also designed to encourage transactions via licensing and assignment; there is no doubt that enforceability is crucial for enabling these transactions. In this regard, Bessen and Meurer (2005) suggest that the value of patenting springs entirely from the threat of litigation, rather than the taking of actual legal proceedings. Empirical evidence by Lanjouw and Schankerman (2004) supports this hypothesis for the US. At the same time, the recent global patent wars within the ICT industry, including the increased involvement of PAEs, have also shown the potential for patent litigation to become a disruptive force with respect to product market competition and innovation.
While the state confers the legal right to exclude third parties from the use of intellectual property (IP), the state does not directly enforce this right itself. Instead, right-holders are required to enforce their rights through the institutions established by the state for this purpose. This implies that right-holders are taskedwith monitoring every potential infringement of their rights. Once a potential infringement has been identified, right-holders must decide whether to seek remediation. Consequently, full legal action/proceedings will not be taken with respect to every infringement of an existing IP right. A decision to take legal action inevitably depends on an assessment of the benefits and costs associated with monitoring infringement and taking legal action. Factors determining whether the owner of a potentially infringed patent takes legal action include (Weatherall et al., 2009):
(1) the amount of sales foregone due to the existence of IP infringement;
(2) the degree of competitive disadvantage which accrues when a comparison is made with those enterprises that are able to free-ride on the R&D and marketing expenses of the right-holding firm;
(3) the potential for loss of goodwill and prestige with regard to a brand, if counterfeits are freely available; and
(4) the expense of monitoring the market and instituting legal proceedings.
Obviously, all of these factors are difficult to measure, not least because of the absence of the counterfactual. For example, the issue of what a product’s market share might have been in the absence of infringement is a difficult factor to measure accurately. The overall costs of undertaking litigation are due to a host of factors including:
(a) the effort undertaken in order to observe and assess potential incidences of infringement,
(b) the ability to identify the infringer(s),
(c) an assessment of whether the IP right is likely to stand up in court e.g. the validity of a patent, and
(d) the (substantial) direct and indirect financial costs of the litigation itself.
The complex interplay between these factors may explain why existing research as well as anecdotal evidence suggests that only a tiny proportion of patent disputes eventually end up in court (Greenhalgh et al., 2010). Furthermore, infringement of patent rights often occurs when firms are unaware of existing rights which they are in fact infringing. Alternatively, infringement may be undertaken wilfully, either for strategic purposes or because firms learn about the infringed patent only when the sunk cost for a research project has already been paid (Bessen and Meurer, 2006). In fact, Lemley (2008) suggests that firms in the IT and biotech industry in the US ignore existing patent rights on purpose in order to avoid wilful infringement. Cockburn and Henderson (2003) provide survey evidence collected from company IP managers. This research shows that only a third of respondents conduct a prior art search before starting new R&D or product development.
From the point of view of right-holders, owners of infringed rights may possess one or more of a variety of motives for taking litigation. Most importantly, a right-holder can seek to use litigation to exclude its competitors from using a patented technology. Alternatively, a right-holder can force its competitors to acquire a licence for the patent right. A law suit also provides the possibility for the negotiation and collection of settlement payments, which could be used toweaken a competing firm or could even be used to push smaller firms out of the market. Litigation may also be used for more strategic motives, such as the case of a dominant incumbent firm which threatens to sue smaller, actual or potential competitors to prevent market entry.
The existing empirical literature on patent litigation overwhelmingly uses US court data, supposedly because US data are relatively easily accessible. Much less is known about patent litigation in other jurisdictions. Given the in part substantial differences in the enforcement systems across countries, empirical findings for the US may not carry over directly to other jurisdictions. The aim of this article is to provide empirical evidence on the enforcement of patents within the UK, and the legal system of the Patents Court (PHC) of England and Wales in particular as it is the venue where the majority of patent litigation in the UK takes place.
Courts in the UK are widely regarded as ‘thorough’, a description which has given the UK courts an ‘anti-patent’ reputation. This reputation has led potential claimants in infringement cases to seek to avoid litigation in the UK, whereas potential defendants and other parties regard the UK as a propitious jurisdiction for challenging the validity of patents and/or claiming non-infringement (Moss et al., 2010). Legal procedures in the UK are also reputed for their costs (Weatherall et al., 2009; Burdon, 2010; see also Section 3.3), but the UK courts are also considered to be relatively fast-working. However, so far there is little factual empirical evidence supporting these claims. This article provides factual empirical evidence, based on a study of all patent cases filed at the Patents Court (PHC) from 2000 to 2008 as well as the Patents County Court (PCC) for 2007 and 2008.
Our analysis shows that patent litigation is extremely rare: we only observe 255 patent cases at the PHC over the nine-year period studied. While the media coverage on the patent wars and PAEs may create the impression that patent litigation concerns only the ICT giants and patent trolls, our data show that patent cases in pharma/chemicals are far more frequent and cases involving PAEs accounted for less than 6% of all cases. The concentration of litigation at the PHC in pharma/chemicals is also reflected in the share of pharmaceutical and chemical companies in the population of companies involved in litigation before the PHC. Our data show that a third of all litigating companies are in the pharmaceutical and chemical industry. Our data also reveal the number of foreign litigants by far exceeds domestic parties. Most patents are aged between six and ten years at the moment they are litigated. A comparison of the set of litigated patents with a matched sample of control patents that have not been litigated shows that litigated patents are considerably more valuable as measured by standard patent-value metrics. They are also broader and contain more references to prior art including the non-patent literature. About 43% of all court cases are filed alleging the infringement of a patent whereas around 31% of cases seek therevocation of a patent. Only about half of all cases proceed to final judgment. Among those cases that ended with a judgment, we find that the by far the most likely outcome is the revocation of a patent – regardless of whether the case was filed as an infringement or revocation action. This confirms the widely shared view based on anecdotal evidence that the PHC is likely to invalidate patents in court. We also have some evidence on the costs of litigation. Our data indicate that most cases involve total costs for both claimant(s) and defendant(s) at the order of £1 million to £6 million, which also confirms the view that patent litigation before courts in England and Wales is extremely costly.
The remainder of the article is organised as follows. Section 2 briefly discusses the available empirical evidence for the UK. Section 3 provides a thorough description of the legal system for the enforcement of patents in the UK. Section 4 describes the data used in Section 5 for the analysis of patent litigation in the UK. Section 6 provides a brief summary and conclusion.
The paper could be read in conjunction with 'The Direct Costs from NPE Disputes' (Boston University School of Law, Law and Economics Research Paper No. 12-34) by James Bessen and Michael Meurer.

They comment that
 In the past, “non-practicing entities” (NPEs), popularly known as “patent trolls,” have helped small inventors profit from their inventions. Is this true today or, given the unprecedented levels of NPE litigation, do NPEs reduce innovation incentives? Using a survey of defendants and a database of litigation, this paper estimates the direct costs to defendants arising from NPE patent assertions. We estimate that firms accrued $29 billion of direct costs in 2011. Moreover, although large firms accrued over half of direct costs, most of the defendants were small or medium-sized firms, indicating that NPEs are not just a problem for large firms.