01 August 2012

Patent Trolls: As American As Apple Pie?

Apple pie. Image credit: Wikimedia CommonsWhether they are called patent trolls, non-practicing entities, patent assertion entities, invention capitalists, defensive patent accumulators or patent licensing organisations, one thing is for certain – the only place in the world you will find them roaming the courts is the United States of America.

While some commentators have suggested that it is only a matter of time before similar business models are deployed in other jurisdictions, we are not convinced that this is an inevitable – or even very likely – outcome.

We would suggest that there are features of the US patent system in particular, and its legal system in general, which together make it a uniquely profitable environment for patent assertion entities (PAEs – we will use this relatively neutral term throughout this article to avoid the need to distinguish between different business models).  Conversely, the systems in other countries are configured so that turning a profit purely on licensing and litigation is a decidedly marginal business proposition.

But before anybody starts to get too excited about the prospect of slaying the monster trolls with a few well-placed legal reforms in the US, we need to point out that many of the PAE-friendly features of the US system are exactly the same features which many believe have fostered a historical culture of innovation and entrepreneurial spirit among individual inventors and the small businesses that are the backbone of the US economy.  In other words, the very reforms which might actually make a difference to the prevalence or PAE’s would also shift the balance of the system against smaller innovators and in favour of big companies and multinationals.

The major features of the US patent and judicial systems which we see as contributing to the PAE ‘problem’ are discussed in greater detail below.


In Suits at common law, where the value in controversy shall exceed twenty dollars, the right of trial by jury shall be preserved, and no fact tried by a jury, shall be otherwise re-examined in any Court of the United States, than according to the rules of the common law.
- Seventh Amendment to the United States Constitution

The inalienable right to a trial by jury in civil matters is unique to the US system.  It is not specific to patent cases, although the requirement that the value in controversy exceed a mere $20 essentially guarantees that the parties will be entitled to a jury in any patent matter worth litigating.  In practice, a jury trial is almost always requested by a patent holder commencing infringement proceedings.  It is less clear, however, that a jury is of any benefit to an alleged infringer.

Philosophically, the right to a trial by jury serves to protect defendants from the risk of government oppression.  A jury can provide an additional measure of fairness and credibility, ensuring that outcomes are not entirely in the hands of a legal elite.  The presence of a jury of ordinary citizens is also often claimed to encourage simplification of complex issues.  As a general social benefit, service on a jury provides members of the public with a practical education in civics.

However, some patent law practitioners (and observers) take a less positive view of jury fact-finding in patent cases.  Juries are sometimes criticised as being ill-equipped to understand the complex technical, scientific, and legal standards involved in many patent cases.  This may cause a jury to be influenced by tangential issues, such as conduct of the parties that is not strictly relevant to the more challenging issues which may be central to a case.

Some believe simply that patent juries are inherently biased in favour of patentees, since inexpert jurors are likely to defer to the expert examiners at the USPTO.  This is even more likely to be the case when the plaintiff is a small, unknown company or individual, able to present itself in court as a David before an accused Goliath, such as the Microsofts of the world.

So while trial by jury clearly benefits the genuine individual inventor, or small innovative company, it is by the same token likely to benefit many PAEs, particularly those of smaller stature and lesser fame.


Upon finding for the claimant the court shall award the claimant damages adequate to compensate for the infringement but in no event less than a reasonable royalty for the use made of the invention by the infringer, together with interest and costs as fixed by the court.
- 35 USC § 284

We are not aware of any other country in the world, and certainly of none with a common law heritage, in which a successful patent owner is entitled to guaranteed minimum damages.  The fact that these are based on a 'reasonable royalty', independent of the identity of the patent owner, is ideal for a PAE, which is placed in the same position, vis-a-vis damages, as if it were a direct competitor with the infringer.

Of course, this provision is intended to assist small innovators, who might lack a strong negotiating position, or the capability to compete, against a large prospective licensee.  The certainty that a loss in court will result in at least an award of reasonable royalties provides an added incentive for a large company to negotiate in good faith from the outset.

Additionally, it is notoriously difficult to determine the value of a 'reasonable royalty'.  Reliable information about comparable licensing agreements is rarely available, because they tend to be confidential.  Rules of thumb, such as the so-called '25% rule' – which was deprecated by the Federal Circuit in Uniloc USA, Inc. v Microsoft Corp. (Fed. Cir. 2011) – typically produce excessive estimates in cases involving relatively small components of large, complex, software products.  This problem is compounded by the fact that the determination of damages is a question of fact, which is decided by the jury – and juries have consistently shown themselves willing to favour high damages payouts against wealthy corporations.


Section 284 of the US Patent Code also empowers a court to increase damages up to three times the amount of actual damages assessed. This provision is generally invoked as a form of punitive damages in case of wilful infringement, i.e. where the infringing activity of the defendant is continued after it becomes aware of a patent.

This is a strong disincentive for a wealthy infringer to ignore a smaller patent holder once placed on notice of the existence of a patent. Clearly this enhances the negotiating power of 'weaker' parties prior to the commencement of litigation, and is thus beneficial to small innovators.

It also helps PAEs. Once such an entity has put an accused infringer on notice of the patent, with the goal of securing a license agreement, the potential for an award of treble damages in the event of a successful infringement action is triggered.


In many common law jurisdictions, including Australia, it is usual for costs to ‘follow the event’.  That is, in the absence of some compelling reason to do otherwise, the losing party in civil proceedings will be ordered by the court to pay the reasonable costs (including attorney fees) of the successful party.  In practice,the amount awarded is typically around 50%-70% of the actual costs incurred. But in patent cases, these can run into millions of dollars.

The expectation that costs will be awarded acts as a disincentive to filing marginal suits.  When a loss means that the plaintiff will be expected to pay the defendant's costs on top of their own, the amount at risk is obviously much greater.

In the US system, the opposite rule tends to apply – attorney fees are only awarded against a losing party in exceptional circumstances.  This makes it much easier for a plaintiff to assess and mitigate risks when commencing litigation.  Since the patent owner determines the course of its own case, it can establish a reasonable expectation of budgeted costs in advance.

Once again, this is helpful to small innovators, who might be unwilling to enforce their rights against an infringer with much deeper pockets if they were at risk of being ordered to pay the larger party's costs.

It is also, however, helpful to PAEs, which can develop a business model based upon a portfolio of patents and lawsuits, conducting a risk/return assessment of each based on reasonably controllable costs.  The greater certainty that a PAE will carry through on a threat of infringement action also places additional pressure upon a target to acquiesce to a demand that it take a license to use a patented invention.


A patent shall be presumed valid. Each claim of a patent (whether in independent, dependent, or multiple dependent form) shall be presumed valid independently of the validity of other claims; dependent or multiple dependent claims shall be presumed valid even though dependent upon an invalid claim. 
- 35 USC § 282

As recently confirmed in Microsoft Corp. v. i4i Limited Partnership (Supreme Court 2011) , the statutory presumption of validity requires that a granted patent can only be invalidated by clear and convincing evidence.  It is not enough, as it would be in Australia and most (all?) other jurisdictions, that the general weight of evidence makes it seem more likely than not that the patent is invalid (i.e. the 'balance of probabilities' standard). 

The provision may be contrasted with section 20 of the Australian Patents Act 1990 (which is far more representative of the situation worldwide):

Nothing done under this Act or the PCT guarantees the granting of a patent, or that a patent is valid, in Australia or anywhere else.
The requirement for 'clear and convincing evidence' is far more akin to the 'beyond reasonable doubt' standard in criminal cases.  As is well-known, this higher standard is set in order reduce the likelihood that a person will be wrongly convicted of a crime – we give the accused the benefit of the doubt, and the presumption that they are innocent until proven guilty.

This is what 35 USC § 282 does for patent holders.  This is important to small innovators who – along with their investors – can have increased confidence in the strength and value of their granted patents.

But again, this is also great for PAEs, to which the law gives a great and powerful stick with which to threaten prospective licensees.


Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
- 35 USC § 101

As stated by the US Supreme Court in Diamond v Chakrabarty 447 US 303, section 101 defines in ‘expansive terms’ the categories of patent-eligible inventions.  The Supreme Court further indicated that, consistent with the intent of Thomas Jefferson that ‘ingenuity should receive a liberal encouragement,’ and that broad and flexible patent laws are integral to that goal,[t]he subject-matter provisions of the patent law have been cast in broad terms to fulfil the constitutional and statutory goal of promoting “the Progress of Science and the useful Arts” with all that means for the social and economic benefits envisioned by Jefferson.’

The only distinct excluded categories of subject matter that the courts have defined are ‘… laws of nature, natural phenomena, and abstract ideas’ (Diamond v Diehr, 450 US 175, at 185) .

Such a liberal approach generally favours innovation in new fields of endeavour.  In recent decades, the US courts have struggled with the limits of patentability in technologies such as computer software, genetics, and business processes, each of which skirts one or more of the three excluded categories.

The difficulty with an expansive approach to patent-eligibility is that new technologies suffer from a lack of searchable prior art – which reduces the effectiveness of examination procedures.  Furthermore, techniques for clear claim-drafting can lack maturity and consistency while practitioners, examiners, and the courts come to terms with the characteristics of the new technology, and explore the boundaries with the three categories of excluded subject matter.

It is therefore no coincidence that 62% of PAE lawsuits feature ‘software patents’, according to James Besson and Michael Meurer, in their recent paper The Direct Costs from NPE Disputes (discussed and critiqued in A $29 Billion US Troll-Tax or Just Another Statistical Smokescreen? and Patentology Cited in Academic Paper on Cost of Patent Trolls). 

While there was much that we found to criticise in Besson and Meurer’s paper, perhaps the most surprising thing about the 62% figure is that it is not higher.  As they go on to explain:

[Software patents] are notoriously difficult to interpret.  Allison et al study patents litigated multiple times and usually asserted by NPEs; they find that software patents account for 94% of such lawsuits. The patents asserted in NPE lawsuits are often subject to lengthy prosecutions which delays public access to information about patent claims.  Rather than transferring technology and aiding R&D it appears that NPEs usually arrive on the scene after the targeted innovator has already commercialized some new technology. [Citations omitted.]

Besson and Meurer summarise these problems as ‘notice failure’, i.e. the difficulty for participants in a market to become aware of relevant patents in a timely manner, prior to investment in commercialisation.

Overall, then, the same expansive approach to eligibility which enables innovators to secure patent protection for their investments in new fields of technology also creates ticking time bombs which show a tendency to end up in the hands of PAEs.


When all of the above features of the US patent and court system are combined, they create a 'perfect storm' for PAEs to direct against prospective licensees.

The business model of the most aggressive PAEs is fairly straightforward – acquire patents; identify potential target licensees; pursue licenses under the threat of lawsuits; and, if the targets do not sign up, commence proceedings (often in the Eastern District of Texas).

This business model is only viable if it is possible to make a reasonable assessment of the cost of ‘investing’ in acquisitions and litigation, along with the expected average returns on the investment (taking into account the fact that not all assertions will be successful).  The features of the US system listed above all tend to support this model:
  1. jury trials, minimum damages, and treble damages all increase the likely returns in successful assertions, pushing the average up;
  2. the general rule in US courts that costs such as attorney fees are not awarded against a losing party reduces risk and maximum expenses;
  3. the presumption of validity increases certainty for the PAE, by reducing the chance that a patent will be found invalid; and
  4. operating in relatively immature fields of patent-eligible subject matter increases uncertainty for the target of an assertion, thus improving the prospects of an early settlement and license agreement.
There is no other jurisdiction which provides this unique combination of features, making countries other than the US decidedly less attractive – and quite possibly not even viable – for the PAE business model.

None of which is to say that the US system should necessarily be radically reformed.  As explained above, these PAE-friendly features are the very same factors that have made the US patent system such an effective engine of individual invention and innovation for nearly two and a half centuries.  According to NSF statistics, in 2006 there were around 6 million firms operating in the US, of which 99.7% were small businesses (fewer than 500 employees), accounting for about 50% of the nation’s employment.

A patent system which supports small, agile, innovative business enterprises therefore makes a substantial positive contribution to the US economy.  The fact that it is also PAE-friendly, and that some PAEs are ‘trolls’ which may represent a real cost to the economy, does not preclude the possibility – indeed the likelihood – that this system provides a net benefit.  At the present time, we lack any credible estimate of the deadweight cost of PAEs to the US economy, although it is almost certainly very much less than the $29 billion asserted by Besson and Meurer.

With better data, it might be possible to weigh up the cost of ‘trolls’ against the cost of change.  But such data will be very difficult to come by, as past attempts have shown.

No comments:

Post a Comment