Skip to content

Scratching my Head Over the SHIELD Act

I get that many people hate patent trolls. I get that many people would like to find a way to limit NPE activities in enforcing patents. But I’m scratching my head over the rhetoric and content SHIELD Act, which was reintroduced last week. For the uninitiated, the proposal would mandate fee shifting for all losing NPEs. Carved out of this group are initial assignees of a patent (that is, individuals and companies who obtained the patents themselves), universities, and companies that spend “substantial” resources making something. Further, NPEs must post a bond for the fees before they can even get into the courthouse.

I should start by saying that I don’t have a strong position on fee shifting. I think that mutual fee shifting might actually do some good to reduce litigation costs and force more reasonable licensing negotiations. I’m also in favor of all sorts of behavior based fee shifting: filing frivolous cases, demanding license fees that far exceed reasonableness, ridiculous claim construction arguments, frivolous discovery requests, unrealistic damages expert reports, etc. I think the threat of fee shifting is a stick that could be used to tame costs. But note that all of these proposals are based on behavior, not identity.

This leads to my concerns with the SHIELD Act, which leave me scratching my head. As a taste of my further thoughts, I’ll note that the EFFs poster child for the act, a podcasting NPE, is actually excluded from the act, and would not have to pay fees on loss.


Let’s start with the rhetoric. It seems like few people have actually examined the act in detail, and thus support for it sounds like a propagandistic echo chamber. I joined Twitter only recently, and originally attributed this to the 140 character limit. Get rid of trolls! $29 billion in costs! Meritless patents! However, it turns out that longer discussion also repeats the same claims without much analysis, albeit with more words. Indeed, very few of the reports actually link to the text of the act, which is here, but instead link to each other, all saying the same thing.

I have a problem with this rhetoric for two reasons. First, the facts are way, way more complicated than that (see Kesan & Schwartz, for example, for a critique of the $29 billion estimate, and see the PWC study on patent judgments since 1995 for discussion about how often and how much NPEs and practicing entities win).

Furthermore, there are many types of NPEs out there, from pre-adoption technology houses to the most abusive frivolous claim filers. The SHIELD Act and its proponents consider none of this nuance.

This leads to my real problem with the rhetoric: it is driven by large companies and presented as if it will benefit small companies. If we were so worried about small companies, why wouldn’t we make the act apply to all patent plaintiffs? After all, large companies routinely assert their patents against smaller, disruptive entities in order to stake out market position or even put them out of business. Just compare the Barnes & Noble submission to the FTC about Patent Assertion Entities with the Barnes & Noble complaints about Microsoft’s licensing practices. The complaint is the same, so it is unclear why the solution should not be the same.

I am bothered by any rhetoric purportedly intended to “protect innovation” that does not, well, protect innovation by its own terms. The EFF campaign focus on a patent that would not actually be subject to the act is just one example. And the fact that no one pushing the SHIELD Act — many of whom I respect even though I disagree — is asking these questions makes my knees jerk to oppose it.


My general point of view is that we should address the behavior, not the owner. Yes, it’s bad that an NPE has asserted what many think is a meritless podcasting claim against Adam Corolla. (I won’t address the merits here, but I will note that the initial patent application was filed in 1996, not last year).

But would we really feel better if it were a product company making the same claim? A meritless claim hinders innovation no matter who makes it. That’s about all I will say about the core question of whether NPEs should be required to post a bond just to get into the courthouse. I think they should not — nor should anyone — even if fee shifting were mutual. But defending that statement is more than I can do in this blog post. I’ve written at least two articles that address the topic and there are a multitude of other studies that address the role of NPEs in licensing as well as the parallels between NPE and product making patentee behavior.

Thus, my questions about the Act relate to the particulars and their unintended consequences. Note that the text of the act has been carefully written to carve out everyone who might object to it, except NPEs. This is the root of the problem, because “patent troll” means different things to different people.

Some points:

1. The act mandates a fee award to any party victorious on noninfringement or invalidity. This means that a company can sue for declaratory relief and force fee shifting, even though the patentee never filed suit. I realize that many people would say, “Great! No more demand letters!” But consider two things. First, there are many, many transactions — millions, if not billions of dollars worth — that are undertaken between technology licensing companies and product companies without the imprimatur of “trollishness.” This provision could disrupt that ecosphere in an effort to fix a different one. Second, this provision could have a perverse effect of multiplying the number of litigations. I don’t believe that patent holders should overclaim in order to seek higher licenses, but I also don’t believe that potential infringers should underclaim in order to force litigation.

2. The act excludes parties that have a “substantial investment” in the exploitation of a patent via production or sale. This is fraught with difficulty. First, what does “substantial” mean? Presumably it is intended to ensnare NPEs that decide to make things, but not make enough (or good enough) things. But why should a court get to decide what is enough? And what do we do about start-ups who have patents but have not yet commercialized their invention? Are they trolls, too?

And what is to stop NPEs from becoming distributors? They could buy competing products wholesale (perhaps products where they have secured licenses) and sell them. This only makes sense; by being resellers, they can claim that the competition of the infringing product is harming their sales business. This points to the general complexity of licensing in the first place: companies might license patents when their opponents do not in order to gain a competitive advantage, even if the patent is old.

3. This exploitation prong leads to other questions. What if a claim construction goes against the patentee, and the patent isn’t quite as broad as the owner thought. This happens all the time, to patentees of all types. Does this mean that the product selling patent holder is no longer exploiting the patent? The language seems to say so, and even the biggest producer would be ensnared.

4. The potential for producers to be liable doesn’t stop there. Consider Palm, which developed WebOS, and made stuff. Consider HP, which has spent billions of dollars in research and development. HP bought Palm, and made WebOS tablets. For various reasons, maybe in part due to patent claims from other tablet makers like Apple, HP decides to stop selling WebOS tablets. HP then decides to enforce Palm’s patents. Mind you, HP didn’t just buy the patents, it bought the company. And then it made stuff, it researched, it developed, and it has even licensed WebOS out to LG try to resurrect it for televisions. Is HP a troll now? It falls under the text of this act. I think that just cannot be right, and yet there it is, in black and white. By the way, HP sold some patents to HTC, who then asserted them against Apple. Soon after, HTC and Apple settled a long running, very costly litigation. Would the SHIELD Act change the dynamic of this dispute? I hope not, my family bought HTC phones.

5. Universities and technology transfer organizations are excluded. This offers a potential way for NPEs to avoid the law. They can become technology transfer organizations for small colleges that cannot afford them, or perhaps coordinate with each other to offer 1 year programs that satisfy the higher education requirements. This latter likelihood is a longshot, I think, but when money is at stake, you never know. This exception also ignores the role of universities in the patent ecosystem; do universities really never assert meritless claims? What happens when they start bringing more lawsuits? Maybe NPEs should team with tech transfer organizations and the tech transfer folks can bring suit instead; it would only take a few large university related intermediaries to make the entire bill fall apart.

6. Excluded are “original assignees,” which are assignees prior to patent issuance that appear on the face of the patent. This leads to two issues. First, what if the assignment is done late, even at a product company? It happens. I suppose they would rely on the “investment” prong, but the company might not be making the patented product.

Second, many NPE patents are assigned to the NPE as the initial assignee. NPEs often buy patents while they are in process and assign continuation applications to themselves; those continuations are often the broadest claims (and thus the least likely to be meritorious as a matter of logic). In other words, the patents we worry about most are the patents that are least likely to fall under the rule. Some of the most highly-litigated, most-litigated patents were originally assigned to the NPE enforcing them now. This brings us to the point I made above, Personal Audio, with its podcasting patent, is initially assigned to….Personal Audio. Sorry folks, this is not the SHIELD you are looking for.

Furthermore, to avoid this issue, NPEs can simply buy shell companies that continue to hold the patents, while the original owner, if still active, can spin off any business to a new entity. I think Acacia does a fair amount of this already, and will surely do more. The more I think about it, the more I think that this rule, intended to protect “real” inventors, will instead render the much of the act toothless.

7. Finally, the original inventor can always sue. This leaves me scratching my head on both ends of the spectrum. On the large company end, I cannot comprehend why fee shifting is a bad thing if BigCo enforces patents, but a good thing if BigCo creates BigCoIPSubsidiary to enforce its patents. Are these two worlds really that different? Perhaps it is to those BigCo supporters of the bill that enforce their own patents. I should add that I’m told, but have not verified, that the tax code rewards companies that spin out their patents. The law gives with one hand and takes away with the other. Perhaps a better solution would be to get rid of the tax benefits, but I won’t hold my breath.

On the small side, I see the SHIELD Act as changing the way NPEs enforce patents. Original inventors aren’t covered, so perhaps the original inventor should be a nominal plaintiff and just get funded by the NPE. That’s a great incentive at a time when people are clamoring for more transparency in the system.

One more point: Individuals are a primary source of NPE patents. Small businesses have always held their own as patent plaintiffs in the system and continue to do so. But the number of patents represented by individual inventors is double with NPE participation. It’s easy to see why: it costs a lot to enforce a patent, or even to license it, and NPEs have skill and finances that individuals lack.

And so when we discuss the SHIELD Act, we should be crystal clear about what we are talking about: a system that favors big companies over individuals. I’m not making a value statement; there may be good reasons why we want fewer patents by individuals enforced. For example, individuals invent a lot of software, and a lot of software patents are really bad. But are individual software patents worse than everyone else’s? I’ve read thousands of software patents, and my belief is that big companies can write crappy patents just like everyone else. My point is that we should be truthful about what is at stake, what the impacts will be, and whether those are the results we want. Then, given that truth, we should look at some evidence to decide what to do.

A comment below notes that the podcasting patent might indeed fall under SHIELD because it was assigned by a family trust of one of the inventors instead of the inventor. This highlights another problem with the bill that I hadn’t even thought about. I noted issues with purchases of companies above.

But if we also say that a wholly owned and controlled intermediary cannot assign the patent on behalf of the original inventor, then SHIELD has the potential to throw a wrench into the works of corporate financing and restructuring. Every day of the week, companies reform through mergers or financing, often creating “NewCo” shell companies that will serve as intermediate owners of assets or will take on new names post financing. I have known lawyers who have such companies already formed so they can use them on a moments notice.

If we say that the intermediary trust excludes the next assignee from being the “original inventor,” then many restructured startup, merged, and financed companies may be in for a rude surprise when the attempt to enforce their patent portfolios. Now, if you dislike patents generally, this realization would lead to rejoicing. I doubt that’s how this act was intended, though.

Update 2
Colleen Chien has a helpful post on NPE litigation activity at Patently-O. The post notes two interesting points related to this post:
1. It confirms that individual plaintiff cases have decreased significantly as NPE cases have risen. This stands to reason NPEs are better at enforcing patents than individuals. But it also means that we always had individual patentee type patent litigation, and NPEs will figure out a way to go there again if they have to.

2. It argues that the SHIELD Act will be helpful because 9 out of the top 10 “high impact” trolls would fall under it. My response, assuming that this definition of high impact is accurate, is that:
a) These may be high impact today, but don’t know what tomorrow will bring. The high impact NPEs of today were not the most litigious of just 2 or 3 years ago. In the future, these patents will be assigned in a way that keeps them with their original owner.
b) The high impact NPEs are the ones that will be least deterred by the act. They are high impact because they have been adjudged valid multiple times, there is little prior art, and many seemingly infringe. They will have the funds and the motivation to post bonds and keep going, collecting winnings and settlements down the way.
c) I scratch my head even more at the scattershot approach of this act. Highly overinclusive, underinclusive, and really just targeted at a few high impact patents that happen to have been purchased in a particular way — a way that is used by many legitimate businesses. It feels like there’s got to be a better way.

12 thoughts on “Scratching my Head Over the SHIELD Act”

  1. I am inclined to agree that the proposed Shield Act is unbalanced towards large companies. My own experience is that troll like behaviour by certain very large companies, including at least one that is a major sponsor of the Shield Act is very common … i.e., asserting very dubious patents out of a portfolio of thousands to force small companies to take licenses where the infringement analysis is weak at best and the patent fundamentally dubious. Indeed, that the shield law is designed to prevent very large companies from having to pay attorney’s fees to smaller ones caught in a patent assertion campaign is pretty ironic.

  2. Michael,

    I think those are good questions, but to the extent you are “scratching your head,” then I think you are overstating things a little. Some thoughts:

    1. A law that penalizes NPEs is not a law that is based on “identity” rather than “behavior,” at least not in the sense of a law against “Acacia Research Corp.” It targets specific entities based on a set of behavioral characteristics, or at least a set of omissions.

    2. What you really seem upset about is not the behavior versus identity thing but something slightly different, which is that that law relies on proxies. The law is targeting NPEs but its professed underlying rationale is to reduce frivolous litigation and excessive royalties. What you really seem to question is whether, and to what extent, the NPE characteristic is correlated with frivolous litigation and excessive royalties, and whether better tailored solutions are possible. I read almost all of your questions as basically directed to this intuition.

    3. To which my response is that I can certainly imagine better tailored solutions. But I can’t imagine better tailored solutions that are likely to pass Congress anytime soon. And in response to your “knee jerk” opposition you should understand there is an equal and opposite knee jerk reaction, which is that if we wait until we can answer all of your questions to your satisfaction, then all the legislative momentum will have dissipated and no law will be passed. Now if your baseline suspicion is that NPEs are no big deal and the whole proposal is cover for a welfare reducing handout to big companies anyway, then that should sit just fine with you. But if one’s baseline suspicion is that NPEs are a significant problem and a partial solution is better than no solution, not so much.

  3. The SHIELD act is indeed based on behavior, not identity. For an entity to acquire patents without having any capacity or intent to make the patented invention and to then use those patents solely to file suit suit against others in an effort to extract royalties is definitely “behavior.”

  4. TJ – always there to bust my chops. Thanks for reading. I think it is about identity, not of one, but of a group. You are right, of course, that it is a proxy. But I am still scratching my head, because the proxies are weak – terribly over and under inclusive of the behavior that is targeted. And we aren’t even sure what that behavior is – everyone says it is meritless claims, but that can’t be right. Or maybe a drag on the economy, but that can’t be right, either. The evidence that patent acquisition harms the economy is mixed, at best. If litigation costs are what we’re worried about, why not go after everyone? Practicing entities win only slightly more often than NPEs, after all.

    At least Rob Buyan is honest enough to point out what the real proxy is: those who buy patents with the purpose of enforcing them. Fair enough, but what do we make of HP buying Palm or Compaq? Or Google buying Motorola Mobility? Are they trolls if they don’t make a product for which they assert a patent? I would think not, but they would be swept in under the act.

  5. Professor Risch,

    I’m not so sure that the Personal Audio ‘504 patent would qualify under an exception in the SHIELD Act. The ‘504 patent is not “a patent filed by and awarded to an assignee of the original inventor or joint inventor.” If you look at the assignment history, Personal Audio was assigned its rights in the patent by the “James D. Logan and Kerry M. Logan Family Trust,” which is not an “original inventor or joint inventor.”

    In any event, your point still stands. The law leaves many, many “loopholes” (depending on your perspective) that have been discussed here and elsewhere.

  6. Tony – going to the source documents! I preach it often enough, so I’m glad you did it. Thanks. I have three responses:
    1. There are two other individual inventors. Presumably, they assigned directly, though we obviously don’t know for sure.
    2. This highlights another problem with the bill – multiple inventor assignments. Must they all qualify?
    3. This one is important enough that I’m updating the main body

  7. Michael, I don’t think what you say in the response is inconsistent with what I say. The proxies are over- and under-inclusive (though I would contest “terribly”–within what framework of real life political solutions can we get a better way?). And I will say that the proxy is for economic welfare. You are correct that the evidence–by which I take it you mean actual empirical studies–is mixed. But that is why I’m saying that it all comes down to one’s prior intuitions. If your ex ante intuition is that NPEs are not a drag on the economy, then there is some empirical evidence to support that intuition, no conclusive empirical evidence to refute it, and so you stick with your original intuition and think there is no big deal. If one’s ex ante intuition is that NPEs are a drag on the economy, then there is some empirical evidence to support the intuition, no conclusive empirical evidence to refute it, and so one sticks to that original intuition and think that some solution to the problem is important. I’m less saying that your questions and doubts are objectively wrong than saying that there is nothing to head-scratch about, in that I took your head-scratching to mean that you couldn’t understand why anybody would not see things the way you see them.

  8. A law that penalizes “buying patents and then enforcing them without making anything” is a law that targets the identity of a group (i.e. NPEs) as much as a law penalizing murder targets the identity of a group (i.e. murderers). Every law targets identity in that sense. That is very different from what people usually mean when we say a law targets group identity (e.g. a law against African-Americans).

  9. TJ – true, but it all depends on your level of abstraction. If we had a law that said: you can go to court, but if you are African-American, it will cost you twice as much, we would say we are targeting the identity. If we had a law that said: you can get married, but only to someone of the same race, would you say we are targeting the behavior, or the identity? The result is the same: justice is closed for some and not others, based solely on fitting into some disfavored group attempting to do something that is otherwise legal.

    That may be OK for some. It might even be OK for me – that’s not really the basis for my objections. My primary objection is that the determination of identity/behavior in this case is haphazard and ineffectual at best, and harmful at worst.

    As for the head scratching, I didn’t mean that I couldn’t see why anyone else would see it differently. It’s obvious why people would see it differently, as your comment points out quite well. The head scratching is more a) why people think this is the solution and b) why no one is asking the hard questions, like will this work?

  10. So in one sense we should move past the identity/behavior thing, in that it is a fight about semantic labels. But the label matters here quite a bit, because there is a very strong and very disfavorable reaction in our legal and political culture to laws that target “identity” and not behavior, which you seem to be invoking. And I wanted to distinguish the types of laws that I think properly invoke that reaction from the law at issue here.

    As for will this work and why aren’t people asking questions, again it comes back to one’s priors. Again, lets stipulate that putting the bill into committee is a sure-fire way of killing the momentum and ensuring the status quo remains. So if your prior intuition is that the status quo is OK and the Act will be terrible, then you want to ask more questions. If one’s prior intuitions are that the status quo is terrible and the Act is unlikely to make things worse, then asking more questions seems like a delay tactic.

  11. Great post, and I’d like to offer a voice for some independent inventors, and why I agree that this is about identity rather than behavior. I am an independent inventor with several patents. One small portfolio is in a quick-moving area of technology. I tried to start a company (which I’ve done successfully in the past) focused on this technology that I invented and, unfortunately, wasn’t able to because of infringement by several large companies. I did not have the financial resources, legal expertise or credibility to pursue these infringers and so I turned to a NPE who, in essence, offered funding and expertise to enforce my rights and secure license fees. Again, I really wanted to build and commercialize this technology myself, and in fact we prototypes it, but this wasn’t going to happen in this instance — because of infringers.

    In my case, I believe that a NPE is acting properly to enforce my rights, and doing it in the right way, including not pursuing small infringers. This NPE is not a bad actor, as some are. So I agree with your distinction 100%. The law appears to focus on identity and not behavior.

    I’m not a lawyer, however I simply don’t see why property rights should be altered depending on who owns the property. If someone is abusing a patent, then that’s another issue entirely. I believe that’s what’s happening here, with improperly granted patents and overplayed claim interpretations on the one hand, and corporations who use the cost of litigation as a key strategy against any independent inventor with legitimate rights to enforce.

Comments are closed.